polymeric foams technology and developments in regulation process and products

61259_C000.indd i61259_C000.indd i 11/6/2008 4:39:29 PM11/6/2008 4:39:29 PM

series editor Shau-Tarng Lee

INCLUDED TITLES

Polymeric Foams: Mechanisms and MaterialsEdited by Shau-Tarng Lee and N.S. Ramesh

Thermoplastic Foam Processing: Principles and DevelopmentEdited by Richard Gendron

Polymeric Foams: Science and TechnologyShau-Tarng Lee, Chul B. Park, and N.S. Ramesh

Polymeric Foams: Technology and Developments in Regulation, Process, and Products

Edited by Shau-Tarng Lee and Dieter Scholz

POLYMERIC FOAMS

61259_C000.indd ii61259_C000.indd ii 11/6/2008 4:39:30 PM11/6/2008 4:39:30 PM

61259_C000.indd iii61259_C000.indd iii 11/6/2008 4:39:30 PM11/6/2008 4:39:30 PM

CRC PressTaylor & Francis Group6000 Broken Sound Parkway NW, Suite 300Boca Raton, FL 33487-2742

© 2009 by Taylor & Francis Group, LLC CRC Press is an imprint of Taylor & Francis Group, an Informa business

No claim to original U.S. Government worksPrinted in the United States of America on acid-free paper10 9 8 7 6 5 4 3 2 1

International Standard Book Number-13: 978-1-4200-6125-3 (Hardcover)

This book contains information obtained from authentic and highly regarded sources. Reasonable efforts have been made to publish reliable data and information, but the author and publisher cannot assume responsibility for the validity of all materials or the consequences of their use. The authors and publishers have attempted to trace the copyright holders of all material reproduced in this publication and apologize to copyright holders if permission to publish in this form has not been obtained. If any copyright material has not been acknowledged please write and let us know so we may rectify in any future reprint.

-

Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying, microfilming, and recording, or in any information storage or retrieval system, without written permission from the publishers.

For permission to photocopy or use material electronically from this work, please access www.copyright.com (http://www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-for-profit organization that pro-vides licenses and registration for a variety of users. For organizations that have been granted a photocopy license by the CCC, a separate system of payment has been arranged.

-

Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe.

Library of Congress Cataloging-in-Publication Data

Polymeric foams : technology and developments in regulation, process, and products / editors, Shau-Tarng Lee and Dieter P.K. Scholz.

p. cm. -- (Polymeric foams series ; 4)Includes bibliographical references and index.ISBN-13: 978-1-4200-6125-3 (alk. paper)ISBN-10: 1-4200-6125-9 (alk. paper)1. Plastic foams. I. Lee, S.-T. (Shau-Tarng), 1956- II. Scholz, Dieter P.K. III.

Title. IV. Series.

TP1183.F6P6482 2008668.4’93--dc22 2008042301

Visit the Taylor & Francis Web site athttp://www.taylorandfrancis.com

and the CRC Press Web site athttp://www.crcpress.com

61259_C000.indd iv61259_C000.indd iv 11/6/2008 4:39:30 PM11/6/2008 4:39:30 PM

To the Lord who said, “For everyone who has will be given more, and he will have an abundance. Whoever does not

have, even what he has will be taken from him.” (New Testament; Matthew Chap. 25 verse 29)

This simply summarizes the bubble nucleation and growth in polymeric foaming way before we understood them.

61259_C000.indd v61259_C000.indd v 11/6/2008 4:39:30 PM11/6/2008 4:39:30 PM

61259_C000.indd vi61259_C000.indd vi 11/6/2008 4:39:30 PM11/6/2008 4:39:30 PM

Contents

Preface ............................................................................................................. ixSeries Statement ............................................................................................ xiAcknowledgments ..................................................................................... xiiiEditors ............................................................................................................ xvContributors ............................................................................................... xvii

1. History and Trends of Polymeric Foams: From Process/Product to Performance/Regulation ................................... 1Shau-Tarng Lee

2. Development of Endothermic Chemical Foaming/Nucleation Agents and Its Processes ............................................... 41Dieter Scholz

3. Foam Extrusion Using Carbon Dioxide as a Blowing Agent ...................................................................................... 69Walter Michaeli, Dirk Kropp, Robert Heinz, and Holger Schumacher

4. Processes and Process Analysis of Foam Injection Molding with Physical Blowing Agents ....................................... 101Walter Michaeli, Axel Cramer, and Laura Flórez

5. Foaming Analysis of Poly(e-Caprolactone) and Poly(Lactic Acid) and Their Nanocomposites .............................. 143Ernesto Di Maio and Salvatore Iannace

61259_C000.indd vii61259_C000.indd vii 11/6/2008 4:39:30 PM11/6/2008 4:39:30 PM

viii Contents

6. Nanostructure Development and Foam Processing in Polymer/Layered Silicate Nanocomposites .................................. 175Masami Okamoto

7. New Material Developments from the Nitrogen Autoclave Process .............................................................................. 219Neil Witten

8. Polystyrene Foam and Its Improvement in Vacuum Insulated Panel Insulation .............................................................. 255Chang-Ming Wong

Author Index ............................................................................................... 291Subject Index ............................................................................................... 299

61259_C000.indd viii61259_C000.indd viii 11/6/2008 4:39:30 PM11/6/2008 4:39:30 PM

Preface

Since the 1960s, polymeric foams have grown into a solid industry that affects almost every aspect of our lives. As our style of living improved, foam consumption increased as well. Solid efforts have laid a good foun-dation in foam science and technology, which allowed it to weather the energy crisis in the ’70s, ozone issues in the ’80s, and recycle/reuse in the ’90s. Above all, innovations in Foam science and technology made the development of new applications and improvements in industry perfor-mance possible. It is no wonder this industry continues healthy growth into the 21st century.

As a result of the globalization of the manufacturing industries, emerg-ing nations benefi ted greatly from enhanced skills and capabilities, and economic prosperity quickly followed. With living standards rising across the world, the global polymeric foam industries grew in quantity as well as in quality to meet the new demands for consumer goods. Every year, new concepts, innovations, and developments continue to be showcased at inter-national foam conferences.

As published in earlier volumes in this Polymeric Foam series, mecha-nisms, processing, science, technology, and materials have been addressed. But the pace of development and the social climate are rapidly changing. As refl ected at the recent foam conferences, performance, sustainable resources, and energy security are becoming increasingly important. In 2005, we began dialogue about gathering a collection of renowned poly-meric foam publishings and assimilating them into a book so that readers could get a clear picture of foam development. After frequent communi-cations, certain commitments were secured, some unfortunately could not carry through, but there was enough to make the editing of this book possible. In general, it covers from early developments in blowing agent,

61259_C000.indd ix61259_C000.indd ix 11/6/2008 4:39:30 PM11/6/2008 4:39:30 PM

x Preface

process optimization, specialty foam development, and current trends. This book is divided into chapters written by world renowned authors, primarily from Europe, and are clearly practitioner oriented.

The fi rst chapter gives a historical perspective clearly showing a foam development trend from the ’50s into global production in the 21st cen-tury. The next chapter focuses on the blowing agent evolution and how this emerging technology was turned into an industry. Physical blowing agent in foam extrusion and injection moulding are addressed in Chapters 3 and 4, respectively. Chapter 5 illustrates interesting works in sustainable foam development. Nanocomposite foam is presented by a pioneer researcher from Japan in Chapter 6. Novel foam products and energy security foam are included in Chapters 7 and 8. This book is intended to present the development picture to benefi t the industrial practitioner, researcher, academic faculty, and graduate school student and hopefully direct more enthusiasm and dedication into the already active fl ow for an even more pros perous future.

61259_C000.indd x61259_C000.indd x 11/6/2008 4:39:30 PM11/6/2008 4:39:30 PM

Series Statement

The bubble is a wonderful creation: a perfect spherical shape, beautiful arch in various degrees of curvature, and minimum surface area per given volume. Without the bubble, both art and science would defi nitely have a narrower scope. The bubble consists of a weak phase surrounded by, and sustained in, a strong phase. It is like the traditional Chinese virtue, Qian

Xu ( ). Although foaming may be one of the more mysterious phe-nomena of the universe, it is fortunate that researchers and practitioners have been able to employ it to advantage with foamed products now com-monplace in our daily lives.

Foaming in polymers involves delicate scientifi c mechanisms, subtle processing techniques, unique morphology transformations, and structure formations. It combines material principles, engineering designs, process-ing methodologies, and property characterization. Polymeric foams are a 20th-century success story, which during the last quarter of the century, have evolved from laboratory-scale products to pilot-line samples and then onto commercial success. Today, it is viewed not only as a technique, but also as a well-established industry. Through challenges such as ozone depletion, recycling, and environmental regulation, in addition to upgrades, it has become a strong industry.

Since polymeric foams have encountered various upgrades—materials/technology, emissions/environmental concerns, properties/applications— it has been crucial to maintain the cohesiveness of polymeric foam by looking at it from various perspectives. This series aims to cover materials/ mechanisms, science/technologies, structures/properties, applications/post-usage, and so on. The reader will be given an overall view, together with some fascinating insights, of polymeric foam. It has to be admitted that foaming is still mysterious in quite a few areas. It is my hope that a

61259_C000.indd xi61259_C000.indd xi 11/6/2008 4:39:30 PM11/6/2008 4:39:30 PM

xii Series Statement

healthy and cohesive understanding will not only strengthen the struc-ture of the existing polymeric foam industry, but will also generate further developments to reveal more basic truths. Let us not forget that life and truth should go hand in hand.

Shau-Tarng Lee, Series editorSealed Air Corp., New Jersey

61259_C000.indd xii61259_C000.indd xii 11/6/2008 4:39:30 PM11/6/2008 4:39:30 PM

Acknowledgments

We appreciate the depth of the editorial responsibility in creating this volume, especially as it contains work from some of the foremost and dedicated exponents of polymeric foams. While it is now time to breathe a sigh of relief regarding the completion of this volume, polymer foam work is, nevertheless, an ongoing and challenging pursuit. We give thanks for inspiration and comfort throughout the editing process, especially when on the brink of giving up or at the point of going nowhere. We would like to convey our heartfelt appreciation to those who diligently met the dead-lines in an ever-increasing workload, sought approval for publication, and secured copyright permissions. In addition, the authors, who are all of diverse international origins, certainly deserve appreciation for their efforts to overcome the cultural differences involved with their contribu-tions to this book. Without their dedication, it would have been impossible for this volume to have been produced.

The principal editor would like to thank Richard Gendron of the Canadian National Research Council, Professor James Lee of The Ohio State University, Professor Tom Turng of University of Wisconsin at Madison, and Dr. Andrew Pacquet (Formerly of Dow) for offering insight-ful comments during the review process. Their valuable critiques have defi nitely helped to improve the quality of this work and in updating the references. Luis Costa and Slavek Kubicz of the Sealed Air Corporation helped prepare formatted drawings and reference searches for which their help is greatly appreciated.

Our special thanks go to our spouses, Friederike Scholz and Mjau-Lin Lee, for their perseverance and patience during our absence, when on nec-essary trips, and when tension and discord inevitably materialized in the

61259_C000.indd xiii61259_C000.indd xiii 11/6/2008 4:39:30 PM11/6/2008 4:39:30 PM

xiv Acknowledgments

family during the time-consuming and arduous editing process. We would also like to extend our thanks to Joseph Lee of Nyack Theological College and Matthew Lee, a Penn State graduate student, for their assis-tance with grammatical correctness. Above all, may this book be used by God to inspire more to explore the truth to benefi t future generations.

61259_C000.indd xiv61259_C000.indd xiv 11/6/2008 4:39:30 PM11/6/2008 4:39:30 PM

Editors

Shau-Tarng Lee was born and raised in Taipei, Taiwan, Republic of China. He received his Bachelor of Engineering degree from the National Tsing-Hua University. In 1980, he joined the Chemical Engineering department at Stevens Institute of Technology and was under Professor Joseph Biesenberger’s guidance in foam-enhanced devolatilization. In 1981 and 1982, he received summer internships from Farrel Company to investigate bubble phenomena in devolatilization. He also received a Stanley fellow-ship and a grant from the National Science Foundation (NSF) to support his research work at Stevens Institute of Technology. After receiving his Master’s Degree in Engineering and Doctor of Philosophy (PhD), he joined Sealed Air Corporation in 1986. Since then, he has specialized in foam extrusion research, development, and production support as a develop-ment engineer, assistant research director, and research director.

Dr. Lee has over 100 publications to his credit, including 26 US patents. He was elected a Fellow of the Society of Plastics Engineers in 2001 and was inducted into Sealed Air’s Inventor Hall of Fame in 2003. In July 2000, he was the editor for Foam Extrusion: Principles and Practice, published by CRC Press (now part of the Taylor & Francis Group). He is also the series editor for the Polymeric Foam Series (CRC Press) that began with Polymeric Foams: Mechanisms and Materials (edited by S. T. Lee and N. S. Ramesh) published in 2004 and was then followed by Thermoplastic Foam Processing: Principles and Development (edited by Richard Gendron). The third volume, Polymeric Foams: Science and Technology (edited by S. T. Lee, C. B. Park, and N. S. Ramesh), was published in 2007. In addition, he is also the co-editor-in-chief for the Journal of Cellular Plastics. He is married to Mjau-Lin Tsai and has three children, Joseph, Matthew, and Thomas. Currently, they reside in Oakland, New Jersey. Dr. Lee is a born again Christian and is actively involved in mission works in Asia.

61259_C000.indd xv61259_C000.indd xv 11/6/2008 4:39:30 PM11/6/2008 4:39:30 PM

xvi Editors

Dieter Scholz was born on March 1, 1938 in Germany and, in 1955 after fi nishing school, he started his career with an apprenticeship contract and education in chemistry. Organic chemistry and organic synthesis of chemicals (fi ne chemicals) for various applications (mainly intermediates for R&D, colorants, pharmaceuticals and other areas) were his main focus. After fi nishing the education program, he worked as a chemist for the same company until 1960 when he joined the Battelle Memorial Institute (Frankfurt) as a chemist working partly on organic synthesis again, but mainly on monomers and polymers (polyimids). In order to enhance his education in chemistry, he undertook evening courses in mechanics and process engineering and fi nished these engineering studies in 1965.

The following six years were spent working for Diversey, and then, in the summer of 1972, he joined Boehringer Ingelheim. In the initial years, he was responsible for the application of the chemicals produced. Those activities involved dealing with German and International Food Regulators (FDA and others) in order to negotiate approvals. From this experience, a wide overview of the many processes concerning chemicals used in the various industries was gained. One of them, the Hydrocerol product range, was the fi rst ready-to-use system for gas nucleation (distribution) in the production of polystyrene and polyethylene foams. Subsequently, it was discovered that this material had a use in chemical foaming as well and this was the start of ready-to-use endothermic blowing agents in the plastics processing industry.

After more than 25 years with Boehringer Ingelheim, Dieter Scholz elected for retirement in 1998. Since then, he has participated in smaller, time-limited projects related to foam and plastics processing and has helped in the setting up of international plastics foam conferences and in keeping contact with former and new “colleagues” from the relevant industries and R&D facilities.

61259_C000.indd xvi61259_C000.indd xvi 11/6/2008 4:39:30 PM11/6/2008 4:39:30 PM

Contributors

Axel Cramer Institute of Plastics Processing (IKV), RWTH-Aachen University, Aachen, Germany

Ernesto Di Maio Department of Materials and Production Engineering, Faculty of Engineering, University of Naples Federico II, Naples, Italy

Laura Flórez Institute of Plastics Processing (IKV), RWTH-Aachen University, Aachen, Germany

Robert Heinz Institute of Plastics Processing (IKV), RWTH-Aachen University, Aachen, Germany

Salvatore Iannace Institute of Composite Materials and Biomaterials, National Research Council, Portici, Italy

Dirk Kropp Institute of Plastics Processing (IKV), RWTH-Aachen University, Aachen, Germany

Shau-Tarng Lee Sealed Air Corporation, Saddle Brook, New Jersey

Walter Michaeli Institute of Plastics Processing (IKV), RWTH-Aachen University, Aachen, Germany

Masami Okamoto Advanced Polymeric Materials Engineering, Graduate School of Engineering, Toyota Technological Institute, Tempaku, Nagoya, Japan

61259_C000.indd xvii61259_C000.indd xvii 11/6/2008 4:39:31 PM11/6/2008 4:39:31 PM

xviii Contributors

Dieter Scholz (Retired from Boehringer Ingelheim) Gau-Algesheim, Germany

Holger Schumacher Institute of Plastics Processing (IKV), RWTH-Aachen University, Aachen, Germany

Neil Witten Zotefoams plc, Croydon, England, United Kingdom

Chang-Ming Wong Material and Chemical Research Laboratories, Industrial Technology Research Institute, Hsinchu, Taiwan, People’s Republic of China

61259_C000.indd xviii61259_C000.indd xviii 11/6/2008 4:39:31 PM11/6/2008 4:39:31 PM

1History and Trends of Polymeric Foams: From Process/Product to Performance/Regulation

Shau-Tarng Lee

CONTENTS

1.1 Introduction ...................................................................................... 11.2 Foundation: Science Lab/Pilot ....................................................... 71.3 Technology: Process/Product ...................................................... 131.4 Performance: Properties and Applications ................................. 20

1.4.1 Physical Properties ............................................................... 211.4.2 Mechanical Properties ......................................................... 221.4.3 Thermal Properties .............................................................. 251.4.4 Acoustic Properties .............................................................. 27

1.5 Regulation: Environmental and Regulatory .............................. 28References .............................................................................................. 38

1.1 Introduction

Every material that we come across has a natural origin. We humans are incapable of creating something out of nothing. No matter what we do, we can only create something out of something else, and are bound by the mass conservation law. As more scientifi c advancements are made,

61259_C001.indd 161259_C001.indd 1 10/25/2008 12:33:06 PM10/25/2008 12:33:06 PM

we are better able to harness natural resources and use them for our benefi t. Polymers, like wood and metal, are products made from natural sources that can be used for a variety of purposes. Indeed, producing and using polymers has greatly enhanced our standard of living and altered modern culture.1,2

The basic ingredients of polymers come from petroleum, plants or animals. After these ingredients are synthesized, polymers are imbued with the durability useful for many applications. This branch of science began to draw global attention in the nineteenth century and took off in the twentieth century, when polymer synthesis was established in the lab, followed by a scaling up to mass production. It became extremely popular after World War II (WWII) and within 30 years, its consumption exceeded both wood and metal as the most used material in the modern world. When nylon was introduced in the late 1930s by DuPont, it soon became a favorite in the textile and clothing industry.3 Nowadays, plastic bags and foam packaging are commonly used as everyday polymer products.

Polymers inherently possess unique structural arrangements that allow them to combine chemical intra-bond and van der Waals inter-bond forces to form distinct melting and crystallizing transitions, which offers a special property spectrum for applications. The processing cost and performance per unit weight for polymers became so favorable that capi-tal investments have continually expanded to meet demand. The profi ts were invested for further research and innovation to bring forth new technologies for a wide variety of applications, resulting in a positive investment cycle.

However, the life cycle of polymers, as illustrated in Figure 1.1, suggests time-magnitude issues that are the basis for life-cycle assessment (LCA). It is clear that petroleum formation from buried plants and organisms is a magnitude too large to be practical for regenerating resources. As a result, resources will continue to diminish. The second largest magnitude is the decomposition on polymeric products after use. No wonder then that landfi ll space has been, and continues to be, short on demand. It will continue to be so for as long as the degradation of (or alternatives to) polymers remains unresolved. It is worth noting that the decomposition of polymers will only resolve space or “presence” concerns,4 and will not necessarily help promote the cycle even if decomposed to the original ele-ments. For instance, when polyethylene is cracked into ethylene, which is lighter than air, it evaporates into the atmosphere and does not participate any further in the life cycle.

When the polymer industry began to take off in the 1930s, polymeric foam science was established to follow the expansion of the polymer industry. Its success in application (i.e. in WWII) encouraged technology development to start a very positive development–application reciprocal cycle. First, its lighter nature enlarged the already wide property spectrum of polymers by offering special cushioning, insulation, and absorption

2 Polymeric Foams

61259_C001.indd 261259_C001.indd 2 10/25/2008 12:33:06 PM10/25/2008 12:33:06 PM

properties. Its market shares continued to improve by extending into new applications and by replacing conventional business.

Before WWII, the chemistry required to generate gaseous bubbles was discovered in the polycondensation reaction.5 Polyurethane (PU) develop-ment was fi rst focused on rivaling polyamide (e.g. nylon) in the fi ber industry. However, when PU foam was developed, its properties soon enabled it to far outrun fi ber research. Its uses have rapidly extended into furniture, construction, and transportation since then. In the same period, the cryogenic industry found chlorofl uoro carbon (CFC) to be a superb cooling chemical, (which was also found as good in blowing polymeric foam.6) An important “pillar“ was the emergence of a technique for mak-ing thermoplastic polymers into which huge investment was poured. Polystyrene (PS) was adopted for foam-making and, instead of a batch process, a continuous extrusion process was developed to greatly enhance productivity. When resins and polymers were available, the fi nal “pillar” was the processing technology. Signifi cant improvements to machinery were made by Lavorazione Materie Plastiche (LMP) in the late 1930s7,8 and benefi ts were also derived from “pasta” processing. Post-WWII, more support to civilian manufacturing was secured which proved effective in upgrading processing technologies. The four “pillars” forming the foun-dation of polymeric foam technology are listed in Table 1.1.

103

102

101

Landfill,Incineration,Degradation

FeedPlants

CO2 +H2O

CustomerProcessing

(foam, film, etc.)

Cracking/Polymerization

Fossil

Tim

e (y

ears

)

1

10–1

10–2

Process

FIGURE 1.1 Organic life cycle.

History and Trends of Polymeric Foams 3

61259_C001.indd 361259_C001.indd 3 10/25/2008 12:33:06 PM10/25/2008 12:33:06 PM

From this foundation, laid in the 1940s and 1950s, two further advances resulted. One was thermoset foams such as polyurethane (PU); the other was thermoplastic foams such as polystyrene (PS), polyvinylchloride (PVC), and polyolefi n. Since foam appeared to be so useful, both foams were accepted from the outset, and that acceptance has been a solid driver for the foam industry ever since. During the 1950s to the 1970s, many new technologies were developed one after another and it would be fi tting to refer to this period as the “technology period”.

With regard to thermoset foams, two branches evolved: soft foam and rigid foam. It was amazing that the same chemistry applied to different raw materials in different morphological structures could result in com-pletely different properties ranging from as soft as a sponge to as hard as a rock. A varied range of equipment was developed to capture the reaction for desired products. Now, after 40 years, PU foam has become the domi-nant force in a variety of markets: furniture, automotive, construction, packaging, transportation, and recreation.

As for thermoplastic foam, extrusion of polystyrene was fi rst used by Dow for fl oating dock construction during WWII. The fl oating and insulation characteristics were soon recognized and subsequently, thermo-forming, packaging, and structural designs were developed. The technol-ogy was further strengthened by the development of polyolefi n foam extrusion, PS molded bead, injection molding, and cross-linked polyethyl-ene (X-PE) technologies. Thermoplastics would not only serve the public through mass production (e.g. cups, trays) but would also offer high qual-ity products (e.g. X-PE foam) for niche applications.

The 1980s can be termed “application period.” Foam found inroads into automotive interiors, transportation, aviation, recreation, construction, medical, military, oil drilling, and refl otation markets. More often than not, market specifi cation was built around foam products to direct development efforts. Nonetheless, more attributes were established such as insulation, shock absorption, acoustics, durable modulus, and light emission.9 Although the depletion of the ozone layer and the advent of microcellular foaming technology became interesting topics during the 1980s, application devel-opments maintained a steady momentum. The whole development is akin to a tree with its root, trunk, leaves, and fruits as shown in Figure 1.2.

TABLE 1.1

Four Pillars of Polymeric Foam Foundation

Process Attributes

1. Foaming chemistry Generate gas within polymeric matrix

2. Blowing agent Soluble and stable gas for polymer

3. Resin manufacture Material strength to hold foaming

4. Processing technology Melting, mixing, and cooling in one stage

4 Polymeric Foams

61259_C001.indd 461259_C001.indd 4 10/25/2008 12:33:06 PM10/25/2008 12:33:06 PM

Now, with the beginning of the twenty-fi rst century, social and environ-mental climates are dramatically changing. Global warming, degradable foam, energy security, and sustainable sources have become increasingly more important. Recycling including reuse and reprocessing became popular in the mid-1990s. In the US, since organizing a sorting system for the general public has not proved as convenient as in Japan, for instance, due to the residential distribution and collecting system, the results have not been so evident. Another factor concerns the petroleum price, which was stable for a quarter of a century. Then, in 2004, its price began to increase with the crude oil price increasing from $20 per barrel to over $100 per barrel in 2007 and continuing to climb in 2008. In addition to gasoline prices being forced to adjust, the raw material price for polymers was also heavily infl uenced. Therefore, the pressures for alternative resources to polymers have also increased signifi cantly.

Due to the discovery of the depletion of the ozone layer in the late 1980s, the Montreal Protocol was drafted to phase out CFCs fi rst then and hydrochlorofl uorocarbons (HCFCs) the two halogenated gases imple-mented in degrading the ozone layer.10 The threat of potential major climate change due to resulted global warming. Kyoto Protocol in 1996, which was drafted to maintain a neutral inorganic carbon dioxide fl ux in the low atmosphere. The Earth is mostly in a state of equilibrium with quite diverse parameters, all contained within a boundary. If one para-meter moves across the boundary, however, a chain reaction in the other

1960–now

1950–1970

1930–1950 Science

Technology

Applications

FIGURE 1.2 Foam development tree.


61259_C001.indd 561259_C001.indd 5 10/25/2008 12:33:06 PM10/25/2008 12:33:06 PM

parameters may ensure resulting in a global catastrophe, which may already have been set in motion, may thus be taking place. An increase in Earth’s temperature will contribute to an accelerated melting of the polar ice and, consequently, to sea level rises that will reduce the available land area and tighten living resources.

Since the emission of inorganic carbon dioxide is inevitable in any energy generation process, it has created a huge problem for industrial-ized countries. Technology, on the one hand, facilitates our living, but it gene rates carbon dioxide emission which may cause global warming to hurt our living quality. Since population and location play critical roles, a credible measurement to quantify the carbon footprint issue is necessary to the success of the Kyoto Protocol. Knowing that plants can convert inor-ganic carbon dioxide to organic carbon, which is a reversal of the energy process in generating inorganic carbon, and that some other processes can make a similar conversion too, it was therefore been proposed to generate “credit” for such systems to balance out the emissions of industrial systems in order to achieve a neutral carbon fl ux.

Carbon dioxide emission occurs in most conventional energy generation processes such as the burning of gasoline in vehicles for example. Polymer resin preparation, processing, converting, and even collecting a blowing agent to convert into atmospheric carbon, will all involve carbon dioxide emission. Growing plants to make polymers, which can turn inorganic carbon to organic carbon for the purposes of obtaining “credit” seems a viable use for polymer and polymeric foam. Table 1.2 lists some common ways of generating carbon dioxide.

More and more international communities are now focusing on poly-mers made from sustainable sources. In the early 1990s, Mitsui Toatsu Chemicals Inc. succeeded in making polylactide foam.11 The concept seems quite intriguing. At present, petroleum price rises alone may be enough of an incentive for the development of a sustainable foam. PLA development has gained momentum in innovation and technology recenty.12,13 Others include ethanol from sugar cane, which can be converted to ethylene as the basic feedstock for polyethylene. In short, the fourth generation is clearly defi ned by environmental and societal issues (the “global regulation” generation).

TABLE 1.2

Common Ways to Generate CO2

Action Consequence

1. Burning fossil fuels Organic carbon to inorganic carbon

2. Decomposing polymer Organic carbon to inorganic carbon

3. Human and animal breathing Inorganic oxygen to inorganic carbon

4. Chemical Reaction Inorganic carbon to inorganic carbon

6 Polymeric Foams

61259_C001.indd 661259_C001.indd 6 10/25/2008 12:33:06 PM10/25/2008 12:33:06 PM

Since the fi rst appearance of polymeric foam, four distinct generations have been outlined: science, technology, application, and regulation. Each of these generations will be further addressed in the subsequent sections.

1.2 Foundation: Science Lab/Pilot

Since most polymers are not tightly aligned, there are holes and micro-spaces that exist between polymer backbones that allow gas molecules to dwell.14 Gas is even present in the polymers. Foaming may not occur until enough gas molecules are clustered to reach the thermodynamic instabil-ity threshold. Unstable bubble nucleation and growth will then occur. The threshold is known as the “critical bubble radius,” defi ned as:

Rcr � 2s/(Pb � P) (1.1)

where s, Pb, and P represent surface tension, bubble pressure, and sur-rounding pressure, respectively. As mentioned before, during a gradual change of thermodynamic conditions, diffusion and/or vaporization may be suffi cient to restore the least energy state or equilibrium. Only when enough instability or disturbance occurs does the gas phase tend to follow

Bubble sizeVCV0

Wmin

Wor

k (W

)

FIGURE 1.3 Nucleation critical size.


61259_C001.indd 761259_C001.indd 7 10/25/2008 12:33:07 PM10/25/2008 12:33:07 PM

a rigorous foaming path to re-equilibrate. A spherical shape has the least surface area at a given volume in parallel with the least energy principle. As illustrated in Figure 1.3, after critical size is reached, growth favors energy dissipation until another equilibrium is reached, or there is enough resistance to stop the growth, which is very likely when the surrounding temperature continues to decrease.15

It is noteworthy that s is an inherent fl uid property, which suggests foaming is highly preferred in the molten state, or, at least, above its glass transition temperature, Tg. When temperature is reduced to solid state to hold the gaseous bubbles, surface tension literally disappears. The residual stress out of the entanglement of the polymer is formed, which may relax with time. Under certain circumstances, the bubble can grow to exceed the material limit causing cell rupture, which can lead to open cells as shown in Figure 1.4.

It can be imagined that when foaming (nucleation and growth) is completed, the polymeric foam is saturated with the blowing agent. When exposed to the atmospheric environment for most applications, the gas concentration gradient can induce counter-diffusion fl uxes. When the blowing agent exists, air comes in. After enough time, the concentration gradient diminishes to render a foam product fi lled with primarily air. In other words, the role of the blowing agent is like a catalyst. A simple illustration is presented in Figure 1.5.

FIGURE 1.4 Open cell structure.

8 Polymeric Foams

61259_C001.indd 861259_C001.indd 8 10/25/2008 12:33:07 PM10/25/2008 12:33:07 PM

Quite a few scientifi c breakthroughs have arisen by accident rather than by design. Foaming is no exception. Although cellular material readily exists on Earth, we have seldom succeeded in duplicating it under labora-tory conditions, especially in the fi rst half of the twentieth century. Around the 1930s, polycondensation was established as a viable way to synthesize polymers through functional group reaction and bonding. A chemical with a difunctional group, for instance, can self-react to build up the length of the backbone.16 A simple reaction is shown below:

A-dimer � B-dimer Æ (A � B)-dimer (1.2)

Polyester and polyamide (known as nylon) are typical examples. Poly-ethylene terephthalate (PET), a member of polyester family, is a popular polymer for foaming in food applications. Its esterifi cation reaction is:

HO–(CH2)n–OH � (C6H5)C–COOH Æ (C6H5)C–COO–(CH2)n–OH (1.3)

The condensation concept was easily extended into branched polymers by implementing multifunctional chemicals as catalysts or branching agents in order to modify the nature of the polymer’s into elastic characteristics,17,18 which is deemed necessary in extrusion foaming. Moreover, when multifunctional monomers were introduced for reaction, instead of a 2D linear structure, a 3D matrix was built up. This is known as a thermoset polymer, which is quite different from the thermoplastic polymers in heat processing. The latter possesses thermo-reversible mor-phology, whereas the former is no longer reprocessible once set. A simple comparison is presented in Table 1.3.

Solid Gas Homogenization

Expansion

Nucleation Foamed product

FIGURE 1.5 Blowing agent as “catalyst.”

TABLE 1.3

Thermoset and Thermoplastic Foam Comparison

Thermoset Thermoplastic

Gas formation Chemical reaction Chemical reaction

Physical blending Spinodal decomposition Nucleation and growth

Foaming path Permanent bond for structure van der Waal for inter-polymer

Polymer bond Open and closed cell Closed cell

Foam nature Very soft to very rigid Soft to rigid


61259_C001.indd 961259_C001.indd 9 10/25/2008 12:33:07 PM10/25/2008 12:33:07 PM

Sometimes, a gaseous byproduct can be generated during the condensa-tion reaction, which naturally becomes the blowing agent. As long as the backbone is providing enough strength to hold the gas in the polymer, polymeric foam will be formed. An exothermic reaction facilitates the activity of gas molecules for more expansion. An endothermic reaction goes the other way but reinforces the strength of the polymeric surround-ing. A summary polycondensation tree is illustrated in Figure 1.6.

Nonetheless, PU fi ts nicely into the foaming criteria. Its reaction is in the following:

R–N=C=O � H2O Æ RNHCOOH � R–NH2 � CO2 � heat (1.4)

R–N=C=O � R�–CH2–OH Æ R–NH–COO–CH2–R� (1.5)

R–N=C=O � R–NH2 Æ R–NH–CO–NH–R (1.6)

The number of functional groups, the chain length between functional groups, and different catalyst systems can alter the product from soft as a seat cushion to for hard as for an appliance door. The degree of reaction can be controlled to make an open-cell product, as in a sponge, or one with a solid skin surface. A variety of PU foam products is therefore possible using the condensation principle. Table 1.4 presents the foam chemistry of PU and its product attributes.

Foam

PUPIU

Blowing agent

Gas generated duringpolymerization

PETNylon

Polycondensation

Foam

FIGURE 1.6 Polycondensation foam architecture.

TABLE 1.4

PU Chemistry and Foam Properties

Foam Chemistry Attribute

Flexible PU Toluene diisocyanate (TDI) Cushion

Rigid PU diphenyl Methylene diisocyanate (MDI) Insulation/structure

Elastomeric PU Rubber cross-linked into TDI Repeated usage

10 Polymeric Foams

61259_C001.indd 1061259_C001.indd 10 10/25/2008 12:33:07 PM10/25/2008 12:33:07 PM

Dr. Otto Bayer and his research group were responsibe for the isocyanate polyaddition procedure, which led to the preparation of a variety of poly-urethanes.19 The success of polyamide (nylon), developed by DuPont, evi-dently encouraged research into PU. During WWII, the research program in the laboratories of Farbenfabriken Bayer was responsible for various PU products such as rigid foams, coatings, and adhesives.2,20–22 After WWII, Allied scientifi c teams studied German developments in PU, which subse-quently stimulated worldwide efforts. It then became a solid pillar in the polymeric foam industry.

When CFC-11 was developed in the 1930s, it was soon found to be very useful as an auxiliary blowing agent in PU foaming. In the exothermic urethane reaction, CFC-11 easily diffuses into cells for stable expansion.23 Its low conductivity opened the door for PU foam board for application in insulation markets.

As for thermoplastic foam development, this was heavily dependent upon the processing technique. Lack of injection and metering control made a friendly blowing agent necessary for overcoming laboratory- testing shortcomings. Polystyrene with pentane was developed and patented in 1935.24 Since then, the development of the refrigerant known as chlorofl uorocarbon (CFC) by Sir Thomas Midgley at General Motors became another powerful driving force for thermoplastic foam develop-ment, simply because of its solubility, stability, and non-fl ammability.6,25 Extruded PS foam was fi rst made available in the US by the Dow Chemical Company in 1943 and was used for fl oating docks during WWII. Not long ago, researchers became aware of just how critical the solubility and dif-fusivity of the gas/melt are, especially in foam extrusion. A high solubility can reduce the time to reach a homogeneous gas/melt solution, which dictates the size of the extruder. The solubility curves for common poly-mers with carbon dioxide are illustrated in Figure 1.7.26

Although extrusion is an inherently effi cient processing technology, it contributes heat during the melt process of the polymer. The removal of this excess heat is necessary to stabilize the melt solution for maximum foaming. From the solubility curve, it is clear that the lower the tempera-ture, the higher the solubility. Reducing the heat input/removal and higher solubility became logical. Foam extrusion can be avoided by sim-ply exposing solid PS to a blowing agent or the polymer soaking in the blowing agent reservoir until it becomes saturated. Reducing the pressure at an elevated temperature, normally above Tg and below Tm (melting temperature), will then allow foaming to occur followed by solidifi cation into a cellular morphology. Fortunately, PS possesses a very useful glass transition temperature, Tg, slightly over 100°C, which allows steam to be a readily available medium for heating the chamber for ideal foaming con-ditions. In the 1960s, PS beading blossomed into an interesting foaming method. From an energy conservation perspective, it avoids heat addition and the consequent removal that is inherent in the common extrusion and injection molding processes.


61259_C001.indd 1161259_C001.indd 11 10/25/2008 12:33:07 PM10/25/2008 12:33:07 PM

When there was an attempt to use polyolefi n to make foam, due to its huge volume in the market, there was a certain amount of doubt sur-rounding the resin characteristics when compared with polystyrene. It had neither an amorphous structure like foaming-grade PS, nor did it have a Tg like PS for strength and rigidity. In the early 1960s, Rubens et al., working for Dow Chemical, fi rst attempted cross-linking PE via ultra vio-let (UV) light by setting a window before the extruder exit.27 Foaming was achieved when UV light was not applied. This evidently enlarged the base material spectrum from amorphous to branched semi-crystallization for foaming. It led to the contined growth in PE foam extrusion. Japan Styrene Paper (JSP) Corporation experimented with PE foam extrusion by foam-ing PS with pentane on a single screw extruder and had a enough success to be able to invest in a continued growth in foam extrusion.28

Extrusion and molded bead development helped defi ne resin structural morphology for foaming, in which adequate material strength is key to holding the aggressive unstable expansion. Since heat is necessary to enhance expansion, the material strength at the foaming temperature is crucial for successful foaming. The necessary attributes are listed in Table 1.5.

Another technique is the X-PE foaming methodology. Instead of using extrusion system for foaming, extrusion can be used for compounding

0.10

0.08

0.06

Sol

ubili

ty (

g-ga

s/g-

poly

mer

)

0.04

0.02

0.000 5

Pressure (MPa)

10 15

FIGURE 1.7 Solubility chart for CO2. ®, Polypropylene; ○, Low-density polyethylene; Δ, high-

density polyethylene; •, ethylene-ethyl acrylate copolymer; ▲, Polystyrene; ——, Estimated

by Sanchez–Lacombe equation of state. (From Areerat, S., “Solubility, Diffusion Coeffi cient

and Viscosity in Polymer/CO2 system.” Research thesis, University, Kyoto, Japan, 2002.)

12 Polymeric Foams

61259_C001.indd 1261259_C001.indd 12 10/25/2008 12:33:07 PM10/25/2008 12:33:07 PM

to include a chemical blowing agent into the PE sheet, which is in turn subjected to a cross-linking reaction to build up extra material strength through chemical bonding. The chemistry is as follows:

R–H (heat, UV, electron beam) Æ R• � H• (1.7)

R• � R• Æ R–R (1.8)

where a chemical bond is formed with H2 as a byproduct. After that, it is sent to an oven to activate the chemical blowing agent to expand the cross-linked sheet. The result is a fi ne cell and soft-feel foamed product. The cross-linked structure confers extraordinary thermal strength for thermo-forming, which defi nitely extends the application spectrum.

Heat-induced gas generation through decomposition is another viable way to generate gas within a molten polymer. It has been uned as a chemi-cal blowing agent (CBA). Most suitable reactions occur at a temperature much higher than the foaming temperature. A typical sodium bicarbonate and citric acid reaction is:

C6H8O7 � 3NaHCO3 Æ (C6H5Na3O7)2H2O � 3CO2 � H2O (1.9)

Since the decomposition temperature is so high, only subsequent cooling or extra material strength can bring the foaming under control. Extrusion and cross-linking have become suitable for CBA foaming. A simple devel-opment chart is illustrated in Figure 1.8 with laboratory and pilot history included.

1.3 Technology: Process/Product

When the science of ploymer foaming was understood, processes were then developed to produce foamed products that were then sold to generate

TABLE 1.5

Polymeric Foam Scientifi c Attributes

Attributes Mechanisms

1. Enough gas in polymer Solubility: dissolution

Reaction: generate gas

2a. Nucleation and growth Thermodynamic instability

2b. Phase separation Spinodal decomposition

3. Product stabilization Curing: cooling, permeation


61259_C001.indd 1361259_C001.indd 13 10/25/2008 12:33:08 PM10/25/2008 12:33:08 PM

profi ts to support further developmental efforts. In essence, equipment design and scale-up have been the major challenges. However, the design challenge was very different in nature for thermoset foam in comparison with thermoplastic foam. The challenge for the former lay in the chemical reaction, and for the latter in the physical mixing. Both require high pres-sure, but this high pressure comes from gas generation in the thermoset reaction, and in thermoplastic foaming it is required to suppress the gas in the polymer.

The chemistry of PU foam is straightforward. Different reactant ratios and degrees of reaction will result in a wide variety of PU foam products, that can be tailored to suit various applications. In general, the necessary product characteristics dictate the formulation chemistry and processes are then developed to deliver the product. Processing and process design has become one of the key areas of the thermoset methodology. With the process design support, PU evolved into rigid, fl exible, and integral skin early on.29–31

A mixing chamber is necessary to keep chemicals thoroughly mixed until the reaction occurs. Foaming can take place in the constraints of a mold or in a free expansion format such as on a conveyer belt. After foam-ing and curing are completed, the product can undergo further fabrica-tion to suit various applications. A simple PU foaming process is presented in Figure 1.9.32

The foam-in-place, or foam-in-mold, was developed to fi ll irregular voids or molds with foam. After setting, the foam can either be ejected for usage or stay in place as a component of the assembly for absorption, fi lling, or insulation. This process can be employed for fl exible and rigid PU foams. Reactive injection molding (RIM), as illustrated in Figure 1.10,

1990

1970

1950 PIR

PU

PS, PE

PS molded bead

X-PE + CBA

PP foam extrusion

Microcellular

PET foam extrusion

PBA

CBA

1930

FIGURE 1.8 Foam science and process development. PiR, polyisocyanurate; PBA, physical

blowing agent.

14 Polymeric Foams

61259_C001.indd 1461259_C001.indd 14 10/25/2008 12:33:08 PM10/25/2008 12:33:08 PM

and foam to fi ll the void for packaging are typical examples. Rigid PU for appliance doors can also be made using this method.

When additional blowing agent was found to be effective in additional foaming promoted by the exothermic reaction between the isocyanate and hydroxyl reactants in order to reduce the thermoconductivity of PU foam, the insulation board process was established to meet construction codes. This process required PU streams to be poured on to the conveyer belt for free expansion and to set before cutting into the desired dimen-sions for shipping. Figure 1.11 demonstrates the process.

Another development was “spray in place,” as shown in Figure 1.12, in which atomization of the raw components was achieved by spraying onto the substrate to form a thin coating of PU foam. An expansion of up to �30 (e.g. about 2 lb/ft3 or 33 kg/m3) could be easily accomplished. The process is applicable to simple as well as complex designs in indoor as well as outdoor applications.

Process automation, mold design, and additives for performance were the main drivers in PU technology development. It also encouraged other thermoset foam developments such as isocyanurate foam for its fl ame retardancy33 and phenolic foam.34 In essence, fi tting for application has been the focal point, and success has driven technology, thereby enlarg-ing the application spectrum. It is little wonder that thermoset foam quickly became the dominant branch in the foam family from the outset.

With regard to thermoplastics, their development has been more diver-sifi ed especially from the plasticator design perspective. Table 1.6 shows the outline. In general, processes have been designed to fi t the polymeric

ChemicalA

Mixingchamber

Applications

Mold: Reactive/injection moldingOpen space: Foam in placeConveyor belt: Sheet or slab

nozzle

ChemicalB

FIGURE 1.9 PU foaming.


61259_C001.indd 1561259_C001.indd 15 10/25/2008 12:33:08 PM10/25/2008 12:33:08 PM

rheology and foaming chemistry for desired cellular products. Extrusion was fi rst adapted to accommodate the blowing agent into the polymeric melt for foaming into useful products.35 After WWII, some military exper-tise in machinery was made available to civilian industry and helped introduce precision processing into polymer machinery. Krobe and JSW (Japan Steel Works) have made solid contributions to extruders, single and twin screw, and injection molding systems. When blowing agent charac-teristics and their impact to polymer rheology were established, the extruder became more like a physical “reactor” for polymer and blowing agent to homo genize under pressure and temperature. When pressure is released at the exit, cellular structure will be developed. Unit operations such as melting, mixing, cooling, and foaming have improved tremen-dously in the last two decades.36–38

Polyol

Isocyanate

Pumps

5 Axes robotMix

head

Mold

FIGURE 1.10 Reactive injection molding process.

16 Polymeric Foams

61259_C001.indd 1661259_C001.indd 16 10/25/2008 12:33:08 PM10/25/2008 12:33:08 PM

FIGURE 1.11 PU board process.

Mixing head

PU foam slab

Conveyor belt

Kraft paper

B

A

FIGURE 1.12 PU spray process.

Chemical A

Chemical B

Pressurechamber

Spraynozzle

Air

Wall/Roof


61259_C001.indd 1761259_C001.indd 17 10/25/2008 12:33:09 PM10/25/2008 12:33:09 PM

Resin technology has also improved. Catalyst science has brought insight into morphological design to fi t foaming. Branched polymers showed favorable elastic characteristics in radial expansion control, which is critical in foam extrusion.39

Injection molding was developed for irregularly-shaped plastic prod-ucts. Although batch-like in nature (mold fi lling, setting, and opening), quite a few molds can be fi lled at the same time with sophisticated mold distribution design and increased injection tonnage. As a result, produc-tivity is no poorer than the continuous extrusion process. Foaming in the mold fi lling exhibited constrained foaming, expansion being constrained by the mold and surface structure by mold temperature. It became par-ticularly appealing for mass reduction and was especially attractive for expensive materials. Again, design and process have gone hand in hand to introduce new and intriguing products, with varying structure and shape, into the markets since the 1960s. When carbon dioxide was pro-posed as super critical blowing agent by Trexel, the benefi ts of cycle-time reduction were realised and embraced to improve product quality and productivity.40

As mentioned earlier, for gas/melt systems, a higher solubility at a lower temperature can save energy simply by loading polymer pellets with blow-ing agent at an elevated pressure and temperature and then allowing the expansion to complete below its melting temperature. Polystyrene can thus be saturated with pentane to expand in the mold into various shapes such as: cups, clamshells, and trays. Although it was possible to make thermo-forming polystyrene foam sheets in the same way, early economics seemed to favor extrusion sheeting. However, when energy costs continued to increase, the mold bead process gained acceptance in more applications.

In addition to energy input and subsequent removal in the extrusion process, the heat distribution and shear history posed concerns with regard to product performance. Poor heat distribution cannot only lower the

TABLE 1.6

Common Foams and Their Foaming Technologies

Product Technology

Flexible PU foam Reactive foaming in mold

Reactive foaming in place

Rigid PU foam Foam in place with auxiliary liquid

Foam in mold

PU foam with skin Foaming with mold temperature control

PS foam Extrusion, molded bead

PE and PP foam Extrusion, molded bead, oven foaming

PVC foam Mold foaming, extrusion

18 Polymeric Foams

61259_C001.indd 1861259_C001.indd 18 10/25/2008 12:33:10 PM10/25/2008 12:33:10 PM

material strength, which directly diminishes the foaming window, but also causes poor property distribution determined by the weakest point.

Another interesting technology was developed by shifting the process-ing domain as shown in Figure 1.13. Clearly, the higher the foaming tem-perature, the more material strength that is required. Extrusion basically characterizes the foaming process by pressure and temperature. For volatile inorganic blowing agents, it is barely possible to load enough for adequate foaming (e.g. 20 times expansion). Improvement of the material strength is necessary to sustain the volatile foaming of the inorganic blowing agent. However, when material strength is improved with more crystallinity and/or extra chemical bonding, the corresponding heat generation during processing and sharp crystallization kinetics can narrow down the processing window to improbably possible. A shear-free foaming process, cross-linked prior to foaming, was developed. The chemical blowing agent and cross-linking agent were compounded into polyolefi n and were passed through a staged oven to activate the cross-linking reaction fi rst, then chemical foaming. Cross-linked foam was thus made. Figure 1.14 shows the general cross-linking process, which is used for polyethylene. Its fi ne cell structure, resiliency, and thermal strength were quite unique and it became a separate brand technology.

Foam nucleation is a very complex, thermodynamically-driven, but kinetically-controlled, phenomenon. When it occurs in the crystallization process, two sets of phase transformation—liquid to gas and melt to solid—are taking place in a very short time period. The co-dependent foaming process still remains mysterious in certain ways. However, in the last decade, thermoplastic foam nucleation researchers have dedicated themselves to the correlation of process design, processing conditions, and foaming design.36,41,42 Recently, a more advanced design has made use

FIGURE 1.13 Extrusion versus cross-linking.

Extrusion Cross-linking

Foaming

Foaming

Processing

Tem

pera

ture S

hear

Compounding


61259_C001.indd 1961259_C001.indd 19 10/25/2008 12:33:10 PM10/25/2008 12:33:10 PM

of a volatile inorganic blowing agent in the foam extrusion process and is known as the super critical fl uid (SCF) blowing agent technique.43 Although not yet fully developed, the phase separation control in the foaming stage may enable it to attain large-scale success by upgrading polymeric foams into high-tech applications.

In summary, the last half-century has seen the number and quality of polymeric foam products grow and blossom. The products and technology continue to move forward hand-in-hand, so extending the polymer prop-erty spectrum. The on-going development has itself demonstrated not just survival of the fi ttest, but also adaptivity, fl exibility, and creativity.

1.4 Performance: Properties and Applications

When urethane science was established in the 1930s, the main focus was on fi bers in order to duplicate the success of nylon (polyamide). During WWII, a German company, I.G. Farbenindustrie, developed lightweight and high-strength rigid PU foams for aircraft components and insulation materials for submarines and tanks.44 Technical improvements were found to be necessary in residual components and there were performance defi -ciencies. After nearly 80 years of foam development, the properties and

FIGURE 1.14 Cross-linking process.

Compounding

Oven 1

Oven 2

Cooling

Foamroll

20 Polymeric Foams

61259_C001.indd 2061259_C001.indd 20 10/25/2008 12:33:10 PM10/25/2008 12:33:10 PM

applications are deemed as the necessary measurement for the technol-ogy development, and to defi ne the market, which success defi nitely brought reward to the scientifi c and technological efforts. In order to cover foam performance in full, it is necessary to begin with basics.

Polymers possess an interesting organic long chain structure and exhibit unique properties compared with inorganic materials. As with foamed inorganic materials, foamed organic materials also extend the polymer material spectrum. Their morphology and foam structure are the main components for their properties. The emphasis here is on properties and applications.

It is known that most thermodynamic properties are weight based—for example, heat capacity, enthalpy, and latent heat—and, since gas has negligent weight, most properties remain similar. However, volume-based properties and structurally-related properties appear quite different.

The four areas to be addressed are physical, mechanical, thermal, and acoustical properties.

1.4.1 Physical Properties

Polymers are characterized by their unique properties. In the presence of gaseous voids, gaseous cells are separated and supported by a polymeric skeleton. Since the skeleton is made of polymer, physical properties such as glass transition, melting, crystallization, and decomposition remain basically unchanged. It is worth noting that although the thermodynamic properties remain similar, kinetics may change dramatically. For instance, before reaching melting point, foam may soften much faster. This could be a serious thermally-related concern in some applications.

The volume-based property, however, shows dramatic change (e.g. density). Polymeric foam density is defi ned as:

rF � Wp � Wg ________ Vp � Vg

(1.10)

where r, W, and V denote density, weight, and volume, and subscripts F, p, and g represent foam, polymer, and gas, respectively. Since gases possess a large volume to weight ratio under normal circumstances, the more gas, the lower the density. How much gas can be dissolved into a polymer is dependent upon the solubility, whereas how soon the dissolution is complete is dependent upon the diffusivity.

When the foam density is less than 1, it will automatically fl oat on water. As long as water soaking is kept under control, the product can be used in fl otation devices such as life jackets and fl oating docks.

When cells are dispersed within a polymer, a mass reduction per unit volume is anticipated, which is important in the application of expensive engineered polymers. As long as performance needs are met, foaming is a form of mass conservation. How to meet the performance requirement


61259_C001.indd 2161259_C001.indd 21 10/25/2008 12:33:11 PM10/25/2008 12:33:11 PM

with the minimum weight of material, or the highest expansion, is of eco-nomic importance. Since quite a few properties are dependent upon the weakest point of the product, the challenge has been to improve the foam property distribution by improving thermal distribution in processing.

Another physical aspect that can deeply affect polymer properties is cell size and distribution.45 Quite a few applications are dependent upon the internal surface area; for example, diffusion-controlled phenomena such as devolatilization,46 and sound attenuation. Surface area can change greatly at a given expansion ratio when cell size changes. In measurement terms, when foam is sliced through, not all of the cell diameter is exposed. Some cells may be sliced at a certain height of the sphere. The American Society for Testing and Materials (ASTM) have proposed a method to calculate the cell size as illustrated in Figure 1.15.

This method is good for spherical cells and cells of equal size. Either or both are violated when expansion exceeds four times according to the packing proposals,47 which is very common for polymeric foam. The aver-age cell size calculation can become quite laborious, requiring various magnifi cation devices, and even scanning electron microscopy and 3D imaging are used when accuracy is important. Cell size and cell size dis-tribution contribute to foam structure and therefore foam properties.48,49

1.4.2 Mechanical Properties

In almost all applications, even for absorption or fi lling, some mechanical strength is needed. Material science has been so advanced that structure and property models can be borrowed for foam mechanical property calculation. Figure 1.16 demonstrates the skeleton model; in other words,

FIGURE 1.15 ASTM method for cell size measurement; r � (p/4)2 y � 0.617y.

22 Polymeric Foams

61259_C001.indd 2261259_C001.indd 22 10/25/2008 12:33:11 PM10/25/2008 12:33:11 PM

an open cell structure in which compressive or tensile modulus can be established:49

rc r

Ê ˆ= Á ˜Ë ¯

2

f f

p p

EE

(1.11)

where c, subscripts f and p denote shape factor, foam, and polymer, respec-tively. Basically, it follows the square of density reduction. When a closed cell is considered, the cell wall stretch under disturbance becomes quite unique in terms of dissipating the disturbance. The enclosed cell func-tions like an elastic balloon, in which expansion can absorb disturbance. This makes closed cell foam more effective in energy absorption with a smaller reduction in modulus than open cell foam at the same expansion. The modulus dependence of density reduction becomes:

rcr= -(1 )

f f

p p

EE

(1.12)

FIGURE 1.16 Strut—only open cell model.

Open cell face

Cell edge


61259_C001.indd 2361259_C001.indd 23 10/25/2008 12:33:11 PM10/25/2008 12:33:11 PM

Figure 1.17 presents the modulus distribution at various foam densities with closed cell content as a parameter. For repeated usage, either an elastic wall and strut open cell foam or closed cell foam with stretchable cell wall is necessary. Seat cushions and shoe soles are among the applications.

In general, compression and tensile strength can be deducted from the above equations. Foam rigidity and fl ex modulus are determined by the type of polymer and the expansion ratio, and, sometimes, the cell structure of the foam. Low mechanical strength is needed in packaging and fi lling types of applications. For seating, mattresses and trays, medium mechanical strength is required. For high-end structural appli-cations, engineered polymer foam is necessary for strength and insula-tion. When requirements become stringent, foam composite can serve the purpose.

A related area is shock absorption. Under sharp disturbance, certain foams tend to deform to absorb a good portion of the disturbance. This has proved effective in protecting a packed material from being damaged. When protection performance was improved, it became popular for ath-letic protection such as in helmet liners and knee and elbow pads.

FIGURE 1.17 Modulus.

1

10–1

10–2

10–3

10–4

Rel

ativ

e Yo

ung’

s m

odul

us, E

*/E

1

Relative density, r0 /r1

10–5

10–3 10–2 10–1 1

E *— =E1

r* 3

r1

c = 0.8

—

E *— = —E1

r*r1

24 Polymeric Foams

61259_C001.indd 2461259_C001.indd 24 10/25/2008 12:33:12 PM10/25/2008 12:33:12 PM

1.4.3 Thermal Properties

We are aware that thermal fl ow resistance is a material property. In steady-state composite conduction, as shown in Figure 1.18, the heat resistance for series composite and parallel composite can be expressed as:50–53

R � ÂRi � Â Li _____ Ali

Series (1.13)

l � Âuili Parallel (1.14)

where R, L, A, l, and u denote resistance, length of the component, surface area, conductivity, and volume fraction, respectively. The subscript i represents the number of components. Note that l � 1/R. Foam is seen as a combination of series and parallel composite and its conduction proper-ties were addressed in Reference 51.

l u u ll l

-1Ê ˆ

= + + -Á ˜Ë ¯Series Parallel

g pc g p p

g p

L L

(1.15)

where subscripts c, g, and p represent conduction, gas, and polymer, respectively. It is known that after a certain time, the majority of gas may be replaced by air which, in general, has a higher conductivity. However, when cell size becomes small enough, the retention of the gas over time may make a difference to thermal properties.

FIGURE 1.18 Heat conduction schematics: (a) parallel; (b) series; (c) combination of parallel

and series. [Redrawn from Leach, A. G., Journal of Physics D: Applied Physics 26 (1993):

733–739.]


61259_C001.indd 2561259_C001.indd 25 10/25/2008 12:33:12 PM10/25/2008 12:33:12 PM

In addition to conduction, there are convection and radiation terms for foam. It is basically a structure-based calculation and can be expressed as:

lF � lc � lv � lr (1.16)

where l represents the thermal conductivity and subscripts c, v, and r denote conduction, convection, and radiation, respectively. When cell size is in the micron range, the radiation term becomes non-negligible.54 In general, the higher the conductivity, the less the insulation capability. Note that conductivity increases for the foam compared to that of the polymer itself. But converting into amount of material, it turns out that much less material for foam is required to reach a given insulation value.

The insulation properties were easily adapted for the purposes of heat retention and opened up contained opportunities such as cups, cup sleeves and thermal chamber insulation. Some medical systems proved to be extremely sensitive to a thermal environment and necessitated the design of special foam packaging systems to meet their needs. Its use in building applications as insulation board colder was especially important in climates where energy loss is more critical. As energy prices continue to escalate, its role in energy security is becoming increasingly important. At present, polymeric foam is now widely used for appliance doors and panels, building insulation, and thermal chambers, as energy saving has an obvious economic value. As a consequence, less energy is consumed, gas emissions are reduced, and a contribution towards global warming is reduced. Figure 1.19 illustrates the comparison between polymeric foam and fi berglass board in insulation.55

FIGURE 1.19 Foam versus fi berglass as insulation.

5.0

4.5

4.0

3.5

F •

h •

fr2

Btu

R-v

alu

e p

er in

ch

3.0

2.5

2.00.0 0.5 1.0 1.5 2.0 2.5 3.0

Density (pcf)3.5 4.0 4.5 5.0 5.5 6.0

0.140

0.175

0.210Cellulose

Fiberglass

Polyurethane

0.245

0.280

0.315

0.350

m2 • kw

R-valu

e per cm

26 Polymeric Foams

61259_C001.indd 2661259_C001.indd 26 10/25/2008 12:33:13 PM10/25/2008 12:33:13 PM

Thermoforming is an extension of the thermal applications. Food trays made of polystyrene foam sheets and X-PE foam for utility cavities are typical examples of the utilization of the unique thermal strength while in a molten state. The heat-induced volume expansion and strength reduction to a soft state yet without sagging state allow it to be formed into various shapes. This property has allowed polymers to make inroads into automotive interiors such as dashboards, water shields, and liners.

1.4.4 Acoustic Properties

The sound wave is considered to be an energy wave or pressure transfer. The wave tends to dampen out after transfer through a medium, like a voice through air. It can also refl ect from a surface and penetrate through material by vibration. The refl ective index is a material property. However, the penetration can be controlled by the structure. An open cell structure and cell wall vibration are possible mechanisms for attenuating sound waves. Low frequency sound is diffi cult to attenuate due to its long wave-length. In general, a large open cell PU wedge is required to reduce low frequency sound. It is far easier to deal with high frequency sound, possibly due to its short wavelength, which can be cancelled out in each refl ection.

Foam, due to its cellular structure, has been shown to be effective in dealing with two kinds of sound waves: airborne and contact. Airborne refers to voice-like sounds; mid-frequency speech and high-frequency engine room noise are good examples. Contact sound refers to, for exam-ple, walking on a hard fl oor. Floor underlay is a growing market for foam as living space becomes more restricted in areas with a rising population. The American Society for Testing and Materials (ASTM) have proposed two standards for airborne sound absorption to the next chamber (E90/E413), and contact impact transfer to the room downstairs (E492/E989). ISO 140-8 (1997) also covered similar issues. Typical foams for fl ooring assembly sound absorption are presented in Figures 1.20 and 1.21. The higher numbers indicate more absorption. Table 1.7 shows the sound reduction coeffi cient for foams and fi berglass.56 It has become a very important construction code in relation to warehouses, gyms, studios, residences, and auditoriums. In populated areas, with living space becom-ing increasingly limited, privacy and the reduction of noise and distur-bance from neigh boring residents has become more desirable. As a result, sound absorption foam is increasingly used. A comparative chart (with fi berglass) is presented in Figure 1.22.

As the years have passed by, the evolution of polymeric foam applica-tions has continued, from fl otation to furniture, to food, recreation, and electronics. Table 1.8 provides a summary. The expansion in applications is better illustrated in Figure 1.23. The application trend continues to expand although the emphasis is shifting.


61259_C001.indd 2761259_C001.indd 27 10/25/2008 12:33:13 PM10/25/2008 12:33:13 PM

1.5 Regulation: Environmental and Regulatory

Up until recently, the foam development picture has been clear: from scientifi c foundation to established technology to performance products. As we moved into the twenty-fi rst century, increased regulation became more signifi cant.57,58 It is, in general, intended to protect living environ-mental life quality. Although some have not been fi rmly confi rmed, or never will be, but there is only one earth, it is not wise to take chances. A fi ne line has to be drawn between scientifi c evidence and political speculation.

The wide consumption of polymeric foam and its use in our daily lives has raised environmental, health, and safety concerns. Since the 1990s,

FIGURE 1.20 Transmission loss in acoustics.

Sound transmission loss

STC 54

Sound transmission loss

Sou

nd tr

ansm

issi

on lo

ss (

dB)

STC contour

80

70

60

50

40

30

20

10

0100 125 160 200 400

One-third octave band center frequency (Hz)

500 630 800 1000 1250 1600 2000 2500 3150 4000250 315

28 Polymeric Foams

61259_C001.indd 2861259_C001.indd 28 10/25/2008 12:33:13 PM10/25/2008 12:33:13 PM

TABLE 1.7

Noise Reduction Coeffi cient (NRC) for Several Foam Products at 1” Thick from Reference 56 with the Exception of 2” Fiberglass from Owens Corning Product Sheet

Material Density (lb/ft3) NRC

Polystyrene foam 2.5 0.18

Rigid polyurethane foam 2.0 0.32

Flexible polyurethane foam 1.9 0.6–0.7

Phenolic foam 2.0–4.0 0.5–0.75

Fiberglass board at 2” thick 1.0 1.0

FIGURE 1.21 Impact insulation coeffi cient.

80

70

60

50

40

30 80 Impa

ct in

sula

tion

clas

s

Sou

nd p

ress

ure

leve

l (dB

re:

0.0

002

mic

roba

r)

90

100

110

70

60

50

40

30Impact insulation class

IIC 60

20

10

0100 125 160 200 400

One-third octave band center frequency (Hz)

Impact sound pressure level IIC contour

500 630 800 1000 1250 1600 2000 2500 3150250 315


61259_C001.indd 2961259_C001.indd 29 10/25/2008 12:33:13 PM10/25/2008 12:33:13 PM

close attention has been paid to health and safety issues. Material safety data sheets (MSDS) became mandatory in the 1990s and a renewed safety and toxic list was published on the regular basis. On the environmental front, there are two main subjects. One is environmental impact; how pro-duction and decomposition impact the ecology cycle of Earth. The other is that at the present consumption rate, can Earth's resources meet demands in the long run? In short, emission, decomposition (or degradation), and resource issues must be considered.

In the 1980s, satellite pictures revealed ozone thinning and the forma-tion of holes over both Antarctic and Arctic region. There were immediate

TABLE 1.8

General Thermoplastic Foam Processes

Process Equipment Blowing Agent Features

Extrusion Screw extruder PBA and CBA Continuous and high speed for

simple shape

Injection molding Injection and

mold system

PBA and CBA Semi- and continuous for

specifi c shape

Molded bead Heat/cool mold PBA Energy saving, specifi c shape,

batch process

X-linking Oven/mold CBA Fine cell, smooth and fl at

product for PO and PVC

FIGURE 1.22 Fiberglass versus foam in sound absorption.

30 Polymeric Foams

61259_C001.indd 3061259_C001.indd 30 10/25/2008 12:33:13 PM10/25/2008 12:33:13 PM

concerns that a diminished ozone layer, which acts as a fi lter to reduce harmful radiation from the Sun, would result in UV-related damage including skin cancers, eye impairments, and crop reductions. Scientists claimed that the massive usage and emission of the very stable haloge-nated hydrocarbons (HC), especially chlorinated and brominated HC, had drifted into the higher atmosphere to hinder ozone formation. In 1987, the Montreal Protocol was established and signed to phase out these agents. Extra taxes were demanded for the banned chemicals to force them out production and consumption. Alternative blowing agents were formulated and introduced into the polymeric foam industry that not only helped it through the crisis without damage but also added to the technical strength of the industry.

The global cooperation in reducing CFC and HCFC usage seems to have been successful via the Montreal Protocol. As illustrated in Figure 1.24,59 as CFCs are phased out, HFC and HC have increased. Nonetheless, the concerns over global warming caused by gas emissions initiated con-certed efforts to regulate global warming gases. In 1997, the Kyoto Conference proposed global regulations in order to reduce the emission of greenhouse gases and so slow down global warming. Carbon dioxide became a focal point. On the one hand, it is integral to photosynthesis and the consequent production of oxygen. On the other hand, it is one of the fi nal gases after the decomposition of organic materials. Fossil burning emission is a common phenomenon among developing countries and gasoline emission among the developed countries. Now the global chal-lenge is how to slow down carbon dioxide emission without slowing down industrial progress. For polymeric foam, where carbon dioxide is used as a blowing agent, the direct implication of curbing carbon dioxide emis-sion has had little impact. Instead of generating carbon dioxide, it is only

FIGURE 1.23 Application tree.

1990

1970

1950

Flotation dockInsulation

Flotation mediumStructural void filler

PackagingCushioning

Appliance insulationStructural insulation

Shock-absorbent paddingCarpet backing

MattressesUpholstery

Food containersThermoforming

Food traysAutomotive liners

Wind bladeStructural reinforcement

Decorative materialGarment insulation

Electronic encapsulationElectrical insulationGaskets and filters

Acoustical insulationFloral displaysToys, novelties

Leather substituteMedical tape

Sporting goodsProtective pads


61259_C001.indd 3161259_C001.indd 31 10/25/2008 12:33:13 PM10/25/2008 12:33:13 PM

used to replace blowing agents which may possess a higher potential to cause the environment in themselves and from the production of them.

However, the use of hydrocarbons, as shown in Figure 1.24, are steadily increasing in blown polymeric foam and their emission has caused serious environmental concerns. Figure 1.25 illustrates a PE foam plank emission sketch. Even after incineration, carbon dioxide and carbon monoxide are emitted. Pentane, used in the molded bead foam process, has a similar scenario as shown in Figure 1.26.60 In brief, smog is a direct concern for hydrocarbon emission, and the corresponding emission of global warm-ing gases while hydrocarbon is destroyed in the incinerator.

The social climate is a double-edged sword. On the one hand, people like to protect forests by encouraging the use of plastic in order to replace

FIGURE 1.24 Blowing agent consumption chart. (From Ashford, P. “Impact of Regulatory

Developments on the Demand for Blowing Agents in Foams.” Paper presented at the Blowing

Agents and Foaming Processes Conference. Rapra, Stuttgart, 2005.)

0

50

100

150

200

250

300

350

1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016

Year

Con

sum

ptio

n (t

hous

ands

of t

onne

s)

Total CFCsTotal HCFCsTotal HFCsTotal HCsCombined total

FIGURE 1.25 PE foam plank emission estimate.

Emissions

Butanedestroyed

(90%)

Butaneloss

(10%)

Thermaloxidizer

NOxCO2

Smog

VOCs

Butanecollected

(90%)

Plankprocess

Resinstorage

Extrusion Foaming Handling* AgingShipping

tocustomer

1256

1 1 3 0.8

0.2>0.1>0.1>0.1>0.1Loss

32 Polymeric Foams

61259_C001.indd 3261259_C001.indd 32 10/25/2008 12:33:14 PM10/25/2008 12:33:14 PM

wood-based products (e.g. foamed PP binder to replace pulp binder). On the other hand, plants or bioplastics, also known as sustainably sourced polymer, are being promoted in order to reduce the use of petroleum-based plastics. In the past decade, more and more biotechnologies have been facilitated to produce bio-based “monomers” through fermentation and/or enzymatic treatment, which, in turn, are polymerized to make polymers.61 This reduces the dependence on petroleum, and the material can be reused, as demonstrated in the cycle period (raw material to product to raw material) shown in Figure 1.1.

A good example is ethanol from the fermentation of sugar cane, which can be converted to ethane. It can be used as a blowing agent for PE and PS.62 It can also be reduced to ethylene as a feedstock for polyethylene. Polylactic acid (PLA) is another example and is made from corn as shown in Figure 1.27.63,64 Fiber, fi lm, and foam can be derived from PLA. Other common bio-based polymers are listed in Table 1.9.

Another benefi t associated with bio-based polymers is the use of green plants as a starting material, which can convert inorganic carbon into organic carbon during the photosynthesis reaction. In other words, the growth of agricultural feedstock can contribute to inorganic carbon con-version to organic carbon, which is accredited to the success of bio-based polymers. For example, the production of PLA required 56 MJ/kg in 2003 (about twice as high as polyethylene, 29 MJ/kg).56,61 After improvements in PLA processing, its manufacture reached carbon neutral in 2006, which means inorganic carbon emission was equal to inorganic absorption by

FIGURE 1.26 PS molded bead process emission estimate. (From Kannah, K., “Pentane—

Environmental and Regulatory Considerations.” Paper presented at the Blowing Agents

and Foaming Processes Conference. Rapra, Frankfurt, 2007. With permission from RAPRA.)

Emissions

Pentanedestroyed(3.5-7.5%)

Pentanenot

collectedThermaloxidizer

Boiler

fuel

Low atmosphere

High atmosphere

NOxCO2

NOxCO2

smog

Ozone depleting

VOCs

Pentanecollected

Moldingprocess

Resinstorage

Pre-expander

Pre-storage

Operation Storage Finishedproduct

1.02.13.36.0 4.5

0.7-1.5 0.5-1.2 0.3-0.9 0.0-1.0

1.00.1-1.10.3-0.90.0-0.70.0-0.8Loss


61259_C001.indd 3361259_C001.indd 33 10/25/2008 12:33:14 PM10/25/2008 12:33:14 PM

TABLE 1.9

Foam Property and Application Attributes

Property Attributes Applications

Physical Density reduction Save material, reduce cycle time

Mechanical Compressive strength Floating device, toys, surfboards

High modulus Coring component

Rigidity and open cell Trays, liquid retention

Flexibility and strength Seat cushions, matrices, shoe soles, seals,

gaskets

Energy Soft and absorption Packaging

Rigid and absorption Car bumpers

Thermal High heat retention Insulation boards, trays, cups, thermo-

containers

Rigid and energy saving Appliance panels and doors

Thermoforming Trays, dashboards

Acoustics Rigid and sound blocking Sound barrier panels

Soft and sound absorption Studio panels, ear phones

Semi-rigid and absorption Floor assembly

FIGURE 1.27 PLA foaming.

Corn field

Fermentation

Polymerization

Foam processing

ThermoformingExpanded bead

BlowingAgents

FreshFeed

Fermeter

Effluent

BiomassSeparation

System

34 Polymeric Foams

61259_C001.indd 3461259_C001.indd 34 10/25/2008 12:33:14 PM10/25/2008 12:33:14 PM

the plants Polyethylene, however, has a manufacturing emission of 2 kg CO2 emission for every kg of polyethylene. The burden is still with energy consumption for PLA manufacturing and processing, but the emphasis has enlarged the horizon. It is no longer carbon emission, but carbon and energy together in the production of a bio-based polymer.

Polymer decomposition has two mechanisms as illustrated in Figure 1.28: moisture and oxidation. The relevant test methods are also included. Some, like PLA, decompose a lot faster under landfi ll conditions and can be categorized as degradable polymers. The ASTM established criteria for degradable and compostable materials. It is well known that esterifi cation is a reversible reaction. In the presence of moisture and heat, decomposi-tion can occur. Table 1.10 shows that different polyesters have different decomposition temperatures with moisture. In essence, some polyesters seem to have intriguing characteristics: made from plants, carbon neutral processing, and are compostable or degradable. Most environmental issues can be resolved using them.

TABLE 1.10

Common Polyesters in the Market

Polyester Tg (°C) Tm (°C) Moisture Sensitivity

PET 75 265 High, under 30 ppm for extrusion

PLA 55–60 130–160 Medium, under 300 ppm for extrusion

PCL �60 60

Copolyester* �30 110–115 Low, drying not necessary for extrusion

* Aliphatic–aromatic copolyester as Ecofl ex™ from BASF.

FIGURE 1.28 Polymer decomposition mechanisms and test methods.

PVOH

Moisture Energy (heat, UV, ...)

StarchPolyester

PET

PE

PPPS

DegradationMeasurement

ASTM D640068687021

EN 13432ISO 17088

PLAPCL


61259_C001.indd 3561259_C001.indd 35 10/25/2008 12:33:15 PM10/25/2008 12:33:15 PM

Starch-based foams and polyvinylalcohol (PVOH) foams are basically water soluble. As long as moisture can be properly maintained in process-ing as well as in usage, they can then disintegrate in water after disposal. Table 1.11 shows common degradable foam. It is worth noting that when foam is degraded or has disappeared from sight, it hastens the reduction to a basic unit process but does not solve all the problems. It could be a major concern to recycle, reuse, and ensure durable usage of the products. When dissolved, it may generate global warming byproducts. A summary for common disposal is presented in Table 1.12. It should be pointed out that the advances in the recycling and reuse of common polymers in the 1990s should be continued. The municipal sorting and segregation program is worth further support from every individual and local government.

A plausible route appears to be the use of green plants to produce the monomer, during which inorganic carbon can be converted to organic carbon. The next step is an energy effi cient process to make polymer and foam. The fi nal products can, for instance, save more energy or conserve material to become “less for more” components in the whole cycle. Foam has the potential, in terms of saving energy and improvement of performance/weight ratio, to change its role from commodity to environmental solution. The insulation benefi ts of foam are presented in Table 1.13. It shows a great energy security advantage. The ideal characteristics for environmental foam are summarized in Table 1.14.

Nonetheless, polymeric foam is not a simple manufacturing and consum-ing issue any more. Since its volume is so signifi cant, it not only affects our living standard, but also impacts on the ecology cycle. A bigger picture is

TABLE 1.11

Common Bioplastics for Foam

Polymer Source

Polylactic acid (PLA) Corn

Starch Corn, potato, wheat

Polyvinylalcohol (PVOH) Vinyl acetate monomer (VAM)

Poly caprolacton (PCL) Crude oil

TABLE 1.12

Analysis of Common Disposal Methods

Methods Advantage Disadvantage

Incineration Generate energy Produce global warming gas

Landfi ll Low cost Polymer takes time to decompose,

need huge space

Water soluble Easy Expensive

Biodegradable Quick volume reduction Negate recycle and reuse

36 Polymeric Foams

61259_C001.indd 3661259_C001.indd 36 10/25/2008 12:33:15 PM10/25/2008 12:33:15 PM

necessary to defi ne its role. Few relevant questions are: what is the source material for polymer, how is it processed, what is the life time of the fi nal product and its impact index toward enriching our living by performance or by replacing other products, what is the potential in reuse and recycle, how to dispose it, how it degrades. As the fi rst step, the energy index and carbon status in making, consuming, and degrading into raw components should be established and compared with other materials, especially the materials replaced by foam with other materials on the same basis; by weight and by cost. We are all familiar with the second law of thermody-namics: the output energy is less than the input in all systems. The contin-ued challenge, not just for polymeric foam but for every industry, will be higher effi ciency and effective inorganic to organic carbon processing, as well as decomposition back to inorganic material to complete the life cycle.

The intriguing question is: what follows after the environmental regula-tion for polymeric foam? We know foam plays an important role in our civilization, making our lives much more convenient, although a lot of problems remain to be solved. It is clear that environmental conditions tend to have more impact on human living conditions. Issues such as the ozone layer, carbon concentration, earthquakes, and climate-induced fam-ine will not go away. How foam plays a more active role in those pending topics, yet continues to enrich human convenience, is a challenge to the global foam society.

TABLE 1.13

Insulation Foam Advantages

Direct Advantage Indirect Advantage

Heat retention Reduce CO2 and NOx emission by less boiler usage

Sound absorption Reduce heating fl uid usage to warm houses

Save energy or higher energy effi ciency Save the material

Less prone to mold and fungus growth Improve living quality

Rigidity as part of the appliance structure

TABLE 1.14

Ideal Environmental Polymeric Foam Characteristics

Items for Foam Environmental Benefi ts

Raw material From green plants to negate carbon credits

Processing Energy effi ciency (one step extrusion process)

Using environmental friendly blowing agent

Foam product Save energy in insulation

Favorable performance/weight ratio

Disposal Recycle and reuse

Water soluble and degradable


61259_C001.indd 3761259_C001.indd 37 10/25/2008 12:33:15 PM10/25/2008 12:33:15 PM

References

1. Meikle, J. L. American Plastic: A Cultural History. Rutgers University Press, New Brunswick, NJ, 1995.

2. Frisch, K. C. “Historical developments of polyurethanes.” In 60 Years of Poly-urethanes, ed. J. E. Kresta and E. W. Eldred. Technomic, Lancaster, PA, 1998.

3. Ndiaye, P. A. Nylon and Bombs: Dupont and the March of Modern America. Belin, Paris, 2007.

4. Franklin, W. E., Franklin, B. S., and Franklin, M. F. “Recycling of materials in municipal solid waste—A view to the year 2000.” Paper presented at the Society of Plastics Engineers Annual Technical Conference (ANTEC), preprint 3504-08. Society of Plastics Engineers, Boston, MA, 1995.

5. Saunders, J. H. and Frisch, K. C. Polyurethanes: Chemistry and Technology, Vols I and II. Interscience-Wiley, New York, 1962, 1964.

6. Midgley, T., Henne, A. L., and McNary, R. R. Manufacture of Aliphatic Fluoro Compounds. US Patent 1,930,129, 1933.

7. Colombo, R. Italian Patent 370,578, assigned to LMP, 1939. 8. Martelli, F. Twin-Screw Extruders. Van Nostrand Reinhold, New York, 1983. 9. Kojima, J., Takada, T., and Jinno, F. “Thin microcellular plastics sheet incor-

porating designed foaming patterns made by photochemical foaming tech-nology.” Journal of Cellular Plastics 43 (2007): 103–109.

10. Hampson, R. F., Kurylo, M. J., and Sander, S. P. “Reaction rate constants for selected HCFCs and HFCs with OH and O(1D).” In Scientifi c Assessment of Stratospheric Ozone: 1989 Volume II. World Meteorological Organization, Global Ozone Research and Monitoring Project, Report No. 20, 1990.

11. Morita, K., Uchiki, K., and Shinoda, H. High Polymer Network. US Patent 5,223,546, assigned to Mitsui Toatsu Chem., 1993.

12. Lee, S. T., Kareko, L., and Jun, J. “Study of thermoplastic PLA foam extru-sion.” Journal of Cellular Plastics 44 (2008): 293–305.

13. Reignier, J., Gendron, R., and Champagne, M. “Extrusion foaming of poly(lactic acid) blown with CO2.” Paper presented at the Society of Plastics Engineers Annual Technical Conference (ANTEC), spon. Society of Plastics Engineers, Charlotte, NC, 2006.

14. Chen, C., Cheng, M. L., Jean, Y. C., Lee, J., and Yang, J. “Effect of CO2 exposure on free volumes in polystyrene studied by positron annihilation spectroscopy.” Journal of Polymer Science B: Polymer Physics 46 (2008): 388–405.

15. Blander, M. and Katz, J. L. “Bubble nucleation in liquids.” American Institute of Chemical Engineers Journal 21 (1975): 833–848.

16. Flory, P. J. Principles of Polymesr Chemistry. Cornell University Press, Ithaca, NY, 1953.

17. Muschiatti, L. C. US Patents 5,229,432 and 5,391,582 assigned to E. I. DuPont de Nemours and Co., 1993, 1995.

18. Al Ghatta, H. A. K., Severini, T., and Astarita, L. US Patents 5,362,763 and 4,422,381, assigned to M. & G. Richerche S.p.A., 1994, 1995.

19. Bayer, O., Rinke, H., Siefken, W., Ortner, L., and Schild, H. German Patent 728,981, assigned to I. G. Farbenindustrie, 1942.

20. Bayer, O., Angew. Chem., A59 (1947): 257. 21. Benning, C. J. “Polyurethane foam.” In Plastics Foams I: Chemistry and Physics

of Foam Formation. Interscience-Wiley, New York, 1969.

38 Polymeric Foams

61259_C001.indd 3861259_C001.indd 38 10/25/2008 12:33:15 PM10/25/2008 12:33:15 PM

22. Hoff, G. P., Eicker, D. B., and Perlon, U. Polyurethanes at I. G. Farben, Boringen, Augsburg P. B. Report 1122, September 1945.

23. Margedant, J. A. Freon-Blown Rigid Foams. E. I. DuPont de Nemours & Inc., Elastomers Division Bulletin HR-31, July 1958 (CFC-11).

24. Munters, G. and Tandberg, J. G. US Patent 2,023,204, 1935. 25. Vachon, V. “Researches for alternative blowing agents.” In Thermoplastic

Foam Processing, ed. R. Gendron, Taylor and Francis, Boca Raton, FL, 2004.

26. Areerat, S., “Solubility, diffusion coeffi cient and viscosity in polymer/CO2 system.” Research thesis, Department of Chemical Engineering, Kyoto University, Kyoto, Japan, 2002.

27. Rubens, L. C., Griffi n, J. D., and Urchick, D. US Patent 3,067,147, assigned to Dow Chemical, 1962.

28. 40th Anniversary for Japan Styrene Paper, JSP, Tokyo, 2004. 29. Wirtz, H. Chapter 7.1. In Polyurethane Handbook, ed. G. Oertel, Hanser,

Munich, 1985. 30. Schlack, P. “Preparation process of polyesters containing high molecular

weight amide groups.” French Patent 869,243, 1940. 31. Cellular Plastics. National Academy of Science, Washington DC, 1967, p. 218

and 230. 32. Barito, R. W. and Eastman, W. O., “Plastics and elastomeric foams.” Chapter

7 of Handbook of Plastics and Elastomers, ed. C. A. Harper, McGraw-Hill, New York, 1975.

33. Ashida, K. “Polyisocyanurate foams.” In Polymeric foams, ed. D. Klempner and K. Frisch. Hanser, Munich, 1991.

34. Mao, J., Chang, J., Chen, Y., and Fang, D. “Review of phenolic foam.” Chemical Industry Engineering 15 (1998): 38–43.

35. Kennedy, R. N., “Extruded expanded polystyrene.” Sec. XII of Handbook of Foamed Plastics, ed. R. J. Bender, Lake, Libertyville, IL, 1965.

36. Collins, F. H. “Controlled density polystyrene foam extrusion.” Society of Petroleum Engineering Journal 705 (1960).

37. Chung, C. I. Extrusion of Polymers. Hanser, Munich, 2000. 38. Rauwendaal, C. Polymer Mixing. Hanser, Munich, 1998. 39. Walczak, K., Gupta, M., Koppi, K. A., Dooley, J., and Spalding, M. A.

“Elongational viscosity of LDPEs and polystyrenes using entrance loss data.” Polymer Engineering Science 48 (2008): 223–232.

40. Okamoto, K. T. “General Description of the MuCell Process.” Chapter 2 in Microcellular Processing. Hanser, Munich, 2003.

41. Lee, S. T., Park, C. B., and Ramesh, N. S. Polymeric Foams; Science and Technology. Taylor and Francis, Boca Raton, FL, 2006.

42. Lee, L. J., Wingert, M. J., Guo, Z., Shen, J., Han, X., Tomasko, D., and Koelling, K. W. “Foaming using a polystyrene/poly(methyl methacrylate) blend and nanocomposites.” Paper presented at the American Institute of Chemical Engineers Annual Meeting, 139d, San Francisco, 2006.

43. Suh, N. P. “Microcellular plastics.” In Innovation in Polymer Processing, ed. J. F. Stevenson. Hanser, Munich, 1996.

44. Reed, D. Urethane Technologies, March, 1987. 45. Bureau, M. N. “Relationship Between Morphology and Mechanical Properties

in Thermoplastic Foams.” Chapter 6 of Thermoplastic Foam Processing, ed. R. Gendron. Taylor and Francis, Boca Raton, FL, 2005.


61259_C001.indd 3961259_C001.indd 39 10/25/2008 12:33:15 PM10/25/2008 12:33:15 PM

46. Biesenberger, J. A. and Todd, D. “Section I: Fundamentals in devolatilization of polymers.” In Devolatilization of Polymers, ed. J. A. Biesenberger. Hanser, Munich, 1983.

47. Maron, S. H. and Lando, J. B. “The Solid State.” Chapter 2 of Fundamentals of Physical Chemistry. Macmillan, New York, 1974.

48. Hamza, R., Zhang, X., Macosko, C., Stevens, R., and Listemann, M. “Imaging open-cell polyurethane foam via confocal microscopy.” In Polymeric Foams, ed. K. C. Khemani. ACS Symposium Series 669, American Chemical Society, Washington DC, 1997.

49. Gibson, L. J. and Ashby, M. F. Cellular Solids. Pergamon, Oxford, London, 1988. 50. Welty, J. R., Wicks, C. E., and Wilson, R. E. “Fundamentals of heat transfer.”

Fundamentals of Momentum, Heat and Mass Transfer. John Wiley & Sons, New York, 1973.

51. Leach, A. G., “The thermal conductivity of foams I: Models for heat conduc-tion.” Journal of Physics D: Applied Physics 26 (1993): 733–739.

52. Glicksman, L. “Foams and cellular materials: Thermal and mechanical prop-erties.” Massachusetts Institute of Technology, summer course note, Cambridge, MA, 1992.

53. Ahern, A., Verbist, G., Weaire, D., Phelan, R., and Fleurent, H. “The conductiv-ity of foams: a generalization of the electrical and the thermal case.” Colloids And Surfaces A: Physicochemical and Engineering Aspects 263 (2005): 275–279.

54. Zhu, Zhengjin, “Modeling of foam expansion and collapse of extruded fi la-ment foam through cell-to-cell diffusion.” PhD thesis, Department of Mechanical and Industrial Engineering, University of Toronto, 2007.

55. Ned Nisson, J. D. The Fiber War: Loose Insulation for Houses. Cutter Information Corp., Arlington, MA, 1997.

56. “Plastic foams.” Material Design Engineering 236 (1966). 57. Narayan, R. “Fundamental principles and perspectives of biodegradable/

compostable plastics and bioplastics.” Paper presented at the International Symposium on Polymers and Environment: Emerging Technologies and Science. BioEnvironmental Polymer Society (BEPS), Vancouver, October 2007.

58. Ashford, P. “Impact of regulatory developments on the demand for blowing agents in foams.” Paper presented at the Blowing Agents and Foaming Processes Conference. Rapra, Stuttgart, 2005.

59. Ashford, P. “Climate change, energy security and thermal effi ciency—Are foams the answer?” Foams 2006. Society of Plastics Engineers, Chicago, 2006.

60. Kannah, K., “Pentane—Environmental and regulatory considerations.” Paper presented at the Blowing Agents and Foaming Processes Conference. Rapra, Frankfurt, 2007.

61. Huneault, M. A., “What does the bio really mean in bioplastics? A personal viewpoint on biobased plastics.” Paper presented at the Polymer Processing Society Annual Meeting, Salvador, Brazil, 2007.

62. Lee, S. T., “Expandable composition and process for extruded thermoplastic foams”. US Patent 5,462,974, assigned to Sealed Air Corp., 1995.

63. Bopp, R. C. and Whelan, J. “Method for producing semicrystalline polylactic acid articles.” US Patent Pub. 2003/0038405, 2003.

64. Ma, P. “Biomimetic materials for tissue engineering”, Advanced Drug Delivery Reviews 60 (2008): 184–198.

40 Polymeric Foams

61259_C001.indd 4061259_C001.indd 40 10/25/2008 12:33:15 PM10/25/2008 12:33:15 PM

2Development of Endothermic Chemical Foaming/Nucleation Agents and Its Processes

Dieter Scholz

CONTENTS

2.1 History ............................................................................................. 422.1.1 How Did Foam Come into Use? ........................................... 422.1.2 Nucleating and Foaming (Blowing) Agents ....................... 43

2.2 Physical Foaming and its Nucleation ............................................ 442.2.1 Direct Gassed Foam Extrusion ............................................. 45

2.2.1.1 Nucleation in PBA Extrusion .................................... 482.2.2 Injection Molding Direct Gassing........................................ 50

2.2.2.1 General Remarks ....................................................... 502.2.2.2 Problems and Solutions ............................................ 50

2.3 Chemical Blowing Agents (CBAs) ................................................. 512.3.1 Injection-Molded Foams ........................................................ 51

2.3.1.1 Methods ....................................................................... 522.3.1.2 Role of Endothermics versus Exothermics ............. 53

2.3.2 Extrusion ................................................................................. 572.3.2.1 Chemically Foamed Extrudates .............................. 572.3.2.2 General Aspects of Foam Extrusion ....................... 582.3.2.3 Extruder Overview: Screw, Melt Filters,

and Dies ...................................................................... 592.3.2.4 Coextrusion ................................................................ 61

61259_C002.indd 4161259_C002.indd 41 11/5/2008 4:26:12 PM11/5/2008 4:26:12 PM

2.3.3 Other Foam Processes ........................................................... 622.3.3.1 Foam Blow Molding with Endothermics .............. 622.3.3.2 Rotational Molding Aspects of Endothermic

Blowing Agents .......................................................... 622.4 Summary ........................................................................................... 632.5 Abbreviations .................................................................................... 64References ................................................................................................. 65

2.1 History

2.1.1 How Did Foam Come into Use?

Gases that have been chemically liberated under heat have been found useful in generating polymeric cellular structure through common processes such as extrusion and injection molding. In the past 40 years or so, quite a few technical hurdles were successfully overcome to make the technology known as chemical blowing agent (CBA) very popular in the global polymeric foam industry. It simply became indispensable in cross-linked PE and PP foam, PVC foam, structural and profi le foam, and woodplastic manufacturing. The application list continues to expand.

Foam history dates as far back as to when humans fi rst discovered techniques to make material lighter, softer and sponge-like for specifi c purposes. They came across natural foam structures like bones and porous minerals, and discovered fermentation to prepare food.

Foams are modifi ed substrates and belong to dispersed systems where non-dissolving properties are used with a mix of material properties in different aggregate states (gaseous, liquid, solid). Foams are dispersed systems that require gases and solid/liquid components for it to form. Foam can consist of open, closed, and combined cell structures. Gases can be introduced directly or released from other material by heat and/or chemical/biological decomposition or reaction.

The polymeric universe started more than 100 years ago with trials using gases like carbon dioxide mainly resulting from the decomposition of traditional kind of baking powders like sodium carbonates with acids (mainly fruit acids like tartaric acid or citric acid). It was easy to achieve and produce these materials. The baking powder industry had the advan-tage based on synthetic NaHCO3 provided by the Ernest Solvay Ammonia Process,1 which was fi rst developed in 1861/1885. With this process large scale production was possible, making an affordable source of carbonate/bicarbonate available.

In relevant plastics literature, very early gases like air, nitrogen, carbon dioxide and decomposing chemicals like carbonates (either by

42 Polymeric Foams

61259_C002.indd 4261259_C002.indd 42 11/5/2008 4:26:12 PM11/5/2008 4:26:12 PM

itself or in combination with mono- and multi-basic acids, some of which decompose), are named in a large number of publications, patents, and patent applications.

The real breakthrough into expanded or foamed polymeric substrates happened after the Second World War. Industrial production and use of foamed materials started in the 1950s in rubber and plastics processing.2–4 Various techniques came into use by utilizing gases and decomposable substances and most are still in foam production today as additives.

2.1.2 Nucleating and Foaming (Blowing) Agents

Though this group is rather small compared with the other additive groups, its growth surpasses the overall growth of the rest of the fi eld.5,6 There are two major groups:

1. physical blowing agents (PBA); mainly gases or liquids;

2. chemical blowing agents (CBA); release gas by decomposition reaction, and nucleants.

In general, nucleating agents are necessary in the polymeric foaming process for even dispersion in the gaseous phase, and in turn, bubble nucle-ation. The common nucleating agents are porous minerals (e.g. talc), whose surface becomes the residence for blowing agents. Cell size distribution improves signifi cantly. In some processes, such as foaming with volatile inorganic gas, adding nucleating agent is not necessary. Since most physical blowing agents are not as volatile as inorganic blowing agents, the latter can be used as nucleating agents. The inorganic can be added through an injection system for PBA or as CBA. The main benefi t is there is no residue left in the foam, and reuse or recycling has minimum impact on nucleation.

In conventional foam extrusion, PBA is generally favored over CBA due to expansion ratio and economics. However, CBA can generate fi ne cell structures by liberating inorganic gases. From an environmental and emission perspective, CBA is heavily favored. When hydrocarbon is used for low-density foam, safety devices and inventory for aging are vital, which again makes CBA attractive. Moreover, CBA can be added as a nucleating agent in PBA foaming, which, when compared with talc, leaves no residue, making reuse much easier in nucleation control. However, the main hurdle is the solubility of inorganic gas which is much less than that of the organic gas in common thermoplastic polymers, and when translated to economics, presents itself as CBA’s major challenge.

There are two kinds of CBA: endothermic and exothermic, depending upon its reaction thermogram. It is well known that polymeric material strength plays a critical role in holding the dynamic foam nucleation and growth, especially in volatile inorganic foaming agent. Although the heat liberated or absorbed is not so signifi cant when compared with

Development of Endothermic Chemical Foaming/Nucleation Agents 43

61259_C002.indd 4361259_C002.indd 43 11/5/2008 4:26:12 PM11/5/2008 4:26:12 PM

the processing heat, the distribution of the reaction makes it impossible to ignore in certain applications. The absorbed heat has the capacity to cool down the surrounding polymeric melt to stabilize it for better cell integrity. The quality benefi t can be easily transferred to performance benefi t. It is no wonder that endothermic CBA has carved itself a niche among applications.7

2.2 Physical Foaming and its Nucleation

It is understood that an aid can be used to improve gas distribution in the relevant substrate. It can be an inert substance such as talc or another material or a decomposable (reactive) ingredient such as a chemical foam-ing agent. Endothermal systems based on fruit acids (presented in Table 2.1) such as carboxylic, polycarboxylic and polyhydroxy carboxylic acids (pref-erably citric-based), and carbonates (preferably bicarbonates) tend to have the best performance.8

Many acids are applicable in endothermic foaming agents as well. The criteria of use are: effi ciency, pricing, decomposition products, reactivity (discoloration, physiological aspects, smell, remaining substances in the substrate), availability, process condition, and stability. Organic fruit/food

TABLE 2.1

Relevant Fruit/Food Acids

Acid/Acidic Salts Data (mp, decomp. etc.)/Comments

Tartaric acid mp 170°C

Monopotassium tartrate mp 250°C/decomp. 250–300°C

Citric acid mp 153°C/decomp. 153–170°C

Monosodium citrate mp/decomp. 180–210°C7

Acid citric esters 210–240°C (decomp.) liquid/pastes8,9

Gluconic acid mp 125–126°C

Gluconates No information

Malic acid mp c. 100°C

Fumaric acid mp ? 276–287 (300–303)°C

Succinic acid mp 185–190°C/anhydride: 235°C (?)

Oxalic acid mp 189°C sublimation/decomp./toxic!!

Ascorbic acid mp 199°C/decomp.

Glutaric acid mp 97°C

Lactic acid (pure) mp 18–26°C

Calcium lactate No information

mp, melting point; decomp., decomposition.

44 Polymeric Foams

61259_C002.indd 4461259_C002.indd 44 11/5/2008 4:26:12 PM11/5/2008 4:26:12 PM

acids, non-fruit inorganic acids, and carbonates have been found useful as nucleating agents in physical foaming systems. They can also be used as chemical blowing agents. Relevant food/fruit acids are given in Table 2.1. Relevant non-food/inorganic acids include:

boric acid;•

acidic phosphates (e.g. monocalcium phosphate, sodium acid • pyrophosphate, disodium pyrophosphate, sodium aluminum phosphate);

acidic sulfate (e.g. sodium aluminum sulfate).•

Relevant carbonates used as CBA include:•

– NH4HCO3

– Ca(HCO3)2

– KHCO3

– NaHCO3

– NaAl(OH)2CO3 10,11

– other alkali and earth-alkali bicarbonates or carbonates.12

Exothermals like azodicarbonamide (ADC) are used in non-food appli-cations like cable insulation together with nitrogen as blowing gas. Hansen and Martin13 reported about nucleation’s tendency to generate “hot spots” in the substrate when creating fi ne cell structure. ADC is used in cable applications today. It may be doubtful if the hot spots or decomposition products nucleate at all due to the fact that when energy is released, relative to the energy content of a polymer melt, it is extremely small in practical technical applications.

It is worth pointing out that exothermic ADC has a better foaming effi ciency (220 cm3/g), than the endothermic citric acid/sodium bicarbonate (120 cm3/g). The former liberated nitrogen-based gases and the latter carbon oxides. When the polymeric strength is strong enough, such as after cross-linking, ADC became a popular foaming agent in low-density foam.

2.2.1 Direct Gassed Foam Extrusion

Gases are introduced directly into the machines (extruder or similar) via pres-sure in the melt, distributed and mixed before leaving the machine exit (i.e. die, nozzle, etc.) as a blend. There are a number of gases found their way for industrial use. The newly introduced gas acts like a temporary “plasticizer” and reduces the melt viscosity drastically. This allows lower processing tem-peratures. Major groups include gases from the atmosphere like nitrogen, carbon dioxide, and even just air. Further gases from gasoline production—the alkanes like propane, butane, pentane, and others—or blends and halo-genated alkanes were used. They are the most debated materials today.


61259_C002.indd 4561259_C002.indd 45 11/5/2008 4:26:12 PM11/5/2008 4:26:12 PM

For improved distribution of the gas in substrates, nucleating agents came into use. These included any kind of fi ne minerals (talc, silica, calcium carbonate, pigments, etc.) used to chemically produce carbon dioxide, blends of carbonates (sodium, calcium, magnesium, etc.) with organic acids (citric, tartaric, numeric, etc.) or inorganic acids (boric acid, acidic phosphates like monocalcium phosphate, sodium acid pyrophosphate, disodium pyrophosphate, sodium aluminum phosphate, etc.). Nucleating agents found their way into various formulations, patents, and publica-tions.16 Desired properties of a gas are as follows14,15:

Environmentally acceptable•

Non-fl ammable•

Non-toxic•

Adequate solubility•

Non-reactive•

Low vapor thermal conductivity•

Low diffusion rate•

Appropriate latent/specifi c heat•

Low molecular weight•

Low cost.•

The idea was to get the acid and carbonates to react with each other and form tiny spots from the gas and other decomposition or reaction compo-nents to create an environment for better gas distribution and foaming, resulting in a fi ner cell structure and hopefully lower densities. Furthermore it was important to develop highly effi cient and affordable systems. Very soon the blends from sodium bicarbonate and citric acid components found their way into various processes. The major applications in the 1950s16–19 were extrusion processes of general purpose polystyrene (PS/GPPS) called direct gassing and the extrusion of polystyrene beads, which contained mainly butane already as described.18

Originally the direct gassing of PS was done by using halogenated hydrocarbons until most of them were banned or substituted by low- or non-ozone depletion materials or plain hydrocarbons. Later on, the direct gassing of low-density polyethylene (LDPE) proved to be a similar process to PS. Today, atmospheric gases have made their way into the relevant industry.20 For a list of those relevant gases refer to Table 2.2.

The major nucleants still in use today consist of special grades of talc and balanced endothermic systems based on the chemistry described previously. Due to the moisture pickup from the atmosphere, Boehringer Ingelheim has sold hydrophobic coated or a kind of encapsulated citric acid to the “foam” industry since about 1967.21 The fi rst ready-to-use endo thermic system for nucleation (and expansion) was developed by

46 Polymeric Foams

61259_C002.indd 4661259_C002.indd 46 11/5/2008 4:26:12 PM11/5/2008 4:26:12 PM

TAB

LE 2

.2

Rel

eva

nt

Ga

ses

Fo

am

ing

Ag

en

tF

orm

ula

Mo

lecu

lar

Weig

ht

(g/

mo

l)

Bo

ilin

g P

oin

t

(°C

)

Vap

ou

r

Pre

ssu

re a

t

20°C

(p

si)

Liq

uid

Sp

eci

fi c

Gra

vit

y

(Wate

r at

20°C

� 1

.0)

Fla

mm

ab

ilit

y

LE

L–U

EL

(Vo

l. %

in

Air

)O

DP

GW

P

Nit

rog

enN

228.0

1�

195.8

1—

0.8

07

No

ne

——

Carb

on

dio

xid

eC

O2

44.0

1�

78

829.6

21.5

2N

on

e—

1

Wate

rH

2O

18.0

2100

0.3

39

1N

on

e—

—

Eth

an

ol

C2H

5O

H46.0

778.3

50.8

49

0.7

93.3

–19

——

Pro

pan

eC

3H

844.1

0�

42.0

5121.4

95

0.5

07

2.1

–9.5

——

n-b

uta

ne

C4H

10

58.1

2�

0.1

530.0

98

0.6

1.9

–8.5

——

i-b

uta

ne

C4H

10

58.1

2�

11.1

543.6

28

0.5

62

1.8

–9.5

——

n-p

enta

ne

C5H

12

72.1

536.0

58.2

02

0.6

26

1.4

–7.8

——

i-p

enta

ne

C5H

12

72.1

527.9

511

.122

0.6

21.3

–8.3

——

Cy

clo

pen

tan

eC

5H

10

70.1

349.2

55.0

20

0.7

45

1.1

–8.7

——

HC

FC

-22

CH

F2C

l86.4

7�

40.7

8132.0

93

1.1

94

No

ne

0.0

51810

HC

FC

-142b

CF

2C

lCH

3100.4

9�

9.9

542.0

61.1

26.0

–18.0

0.0

72310

HF

C-1

52a

CH

F2C

H3

66.0

5�

24.7

84.6

97

0.9

5.1

–17.1

—140

HF

C-1

34a

CH

2F

CF

3102.0

3�

26.5

582.5

31.2

1N

on

e—

1300

HF

C-2

45fa

CH

F2C

H2C

F3

134.0

515.3

517.9

81.3

5N

on

e—

950

HF

C-3

65m

fcC

F3C

H2C

F2C

H3

148.0

840.2

6.2

81.2

73.8

–13.3

—890

CF

C-1

1C

FC

l 3137.3

723.7

512.8

60

1.4

6N

on

e1

4750

CF

C-1

2C

F2C

l 2120.9

1�

30.1

582.2

41.2

9N

on

e1

10890

CF

C-1

13

CF

2C

lCF

2C

l187.3

747.8

55.4

91.5

5N

on

e1

6130

CF

C-1

14

CF

Cl 2

CF

2C

l170.9

23.7

526.6

65

1.4

4N

on

e1

10040

Met

hy

lal

CH

3O

CH

2O

CH

376.0

941.8

56.2

67

0.8

669

2.2

–13.8

——

OD

P, o

zo

ne

dep

leti

on

po

ten

tial

wit

h O

DP

fo

r C

FC

-11 d

efi n

ed a

s 1.0

.; G

WP,

glo

bal

warm

ing

po

ten

tial

wit

h G

WP

fo

r C

O2 d

efi n

ed a

s 1.0

; L

EL

, lo

wer

exp

losi

ve

lim

it; U

EL

, u

pp

er e

xp

losi

ve

lim

it.


61259_C002.indd 4761259_C002.indd 47 11/5/2008 4:26:12 PM11/5/2008 4:26:12 PM

C.H. Boehringer Sohn in Ingelheim, Germany and came onto the market end of 1972 under the name Hydrocerol® (ideas and realization: Elmar Bisle PhD. and D. Scholz). Even having other formulations like acid and bicarbonates together in one blend, Boehringer could guarantee a shelf life of one year. The fi rst ready-to-use active nucleant and endothermic blowing agent was born. Hydrocerol was registered in July 1972. Clariant bought the technology in 1999.

Later some “reinventions” came onto the market in England, Japan and in the 1980s/1990s in North America and other areas. Today this concept is a well-known and widely accepted commodity with growing markets and additional modifi cations.22,23

2.2.1.1 Nucleation in PBA Extrusion

The presence of nucleating agent evidently makes the foaming more het-erogeneous than homogeneous in nature. As a result, a higher nucleation rate occurs, and cells become much fi ner. Nucleation examples with active (decomposable/reactive) materials can be found in various references.6,14,24 The LDPE profi le made with HCFC-142b and butane are shown in Figures 2.1 and 2.2.

In the past, the fastest and easiest information regarding general behavior was for GPPS and gases like CFC-11/CFC-12 (see Table 2.2) or alkanes. Here, drastic results were obtained by the various changes; for example, with citric acid and sodium bicarbonate.17 Parameter investiga-tion helped process improvement signifi cantly. A vast amount of research

FIGURE 2.1 LDPE profi le made with HCFC-142b, with and without CBA nucleating agent.

48 Polymeric Foams

61259_C002.indd 4861259_C002.indd 48 11/5/2008 4:26:13 PM11/5/2008 4:26:13 PM

was dedicated to the formula optimization. Obscure formulations were reported in the literature, such as talc, silica/silicates, solvents like acetone and ethanol with doubtful fi nal results. The “equimolar” citric acid and sodium bicarbonate was proved to be effective in generating fi ne cell structure as illustrated in Figures 2.1 and 2.2.

As soon as the equimolar ratio is disturbed, the cell size increases as suggested in Figure 2.3. Tests were carried out where only the molar ratio was varied while the processing temperature, nucleant, and other addi-tives such as colorants, were kept constant. After the test was completed, the foam product was kept in storage for a few days before the cell size

FIGURE 2.3 Equimolar cell optimization results, the oval area represents the fi nest cell size.

General aspects to formulateactive nucleants

Lowest possible cellsize under constant

Optimizing formulation for nucleation

Citric comp100%mm

NaHCO3100%mm

Area of equimolar

0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

0.125 0.250 0.300 0.500 0.750 1.000

Loading (%)

Cel

l siz

e (m

m)

GPPSLDPE

FIGURE 2.2 Cell size effects at different loadings of CBA for LDPE and PS foam extrusion

with butane blowing agent.


61259_C002.indd 4961259_C002.indd 49 11/5/2008 4:26:13 PM11/5/2008 4:26:13 PM

and structure were measured. It was found that the reproducibility in cell structure and its variation during storage were very consistent.

Another aspect is the combined use of equimolar systems and talc. There is a synergism observed and published.24 In GPPS foam extrusion, fi ner cells, lower density and faster outgassing were reported. Another advantage is the reduced storage time between production (extrusion) and thermoforming, since the foam sheet seemed “stiffer.”

2.2.2 Injection Molding Direct Gassing

2.2.2.1 General Remarks

Foam injection molding is often referred to as structural foam molding25–27 due to the formation of an integral skin. Injection molding machines are built using two main concepts. Today’s standard is a screw which also acts like a piston for pushing. The classical concept was using a piston, as in rubber processing, but feeding with an extruder into an accumulator and shooting with a piston-operated system into the mold. This helps introduce gases in the molten polymer through the barrel of the extruder to make foamed parts as shown in Figure 2.4.26 This process was known as UCC (Union Carbide Corporation process).

2.2.2.2 Problems and Solutions

When nucleants started to be used in extrusion, the question became: Why not introduce nucleants into the injection molding of structural foams? Early results in testing this concept in the late 1970s in Europe and early 1980s in North America gave very promising results and the endo-thermic materials naturally found their way into this application with some advantages:

1. The gas was more uniformly distributed in the melt and less gas loss through the hopper was observed. Therefore, the gas pres-sure could be lowered to receive the same results.

2. The skin appeared better and somewhat smoother, and the cell structure became more uniform with less dense (weight) parts. Thus, faster cycles could be realized.

3. Because this application used aluminum-based molds that were used in lower pressure applications, it gave longer running/life times for the molds and, in general, generated more parts per shift.

4. When either the nucleant was overdosed or the wrong formula-tion was used, or both, corrosion of the aluminum mold would occur after some time. Today these problems have been solved and endothermic nucleants generally arouse no corrosion concerns.

50 Polymeric Foams

61259_C002.indd 5061259_C002.indd 50 11/5/2008 4:26:13 PM11/5/2008 4:26:13 PM

This process allows the production of rather huge parts with high weight at reasonable costs and a good lifespan of the molds due to the fact that the lower injection pressure is aluminum based and not steel. The active endonucleants have proven useful in this process. Today, common machine manufacturers include Wilmington, Johnson Controls, and Battenfeld.

2.3 Chemical Blowing Agents (CBAs)

The CBAs were and are widely distributed all over major processing technologies, they can be used for all thermoplastic resins, and play a major role in the foamed plastics industry.2,18

The endothermals found a good place in practice due to certain advan-tages over the classical ADC (azodicarbonamide). There are applications like cross-linked polyethylene and EVA (ethylene-vinyl acetate) which still is the major market for ADC. On the other hand there is a synergistic effect of both systems in various areas to improve properties and pro-cesses.4–6,16,28,29 Major groups of foaming/blowing agents are given in Table 2.3.

2.3.1 Injection-Molded Foams

When the endothermics were introduced by Boehringer Ingelheim in the early 1970s, the standard foaming or blowing agent was azodicarbonamide (ADC) or other exothermics with similar chemistry such as sulfohydraz-ides, OBSH, 5-PT, and so on (see Table 2.3). They were well established due to the availability, good gas yield, ease of blending, decomposition range,

FIGURE 2.4 Union Carbide Corporation process.


61259_C002.indd 5161259_C002.indd 51 11/5/2008 4:26:13 PM11/5/2008 4:26:13 PM

and handling. The disadvantages of exothermics in most cases were smell, discoloration, corrosion, and non-food grades.

The endothermics were based on raw materials which matched the requirements for food or food additives (according to most regulations in Europe, America, and Asia). The endothermics fi lled a gap for food pack-aging, toys, and pharmaceutical and cosmetics applications.31 There was no smell, no discoloration, and in most substrates an even fi ner cell struc-ture and smoother surface could be obtained. Painting or decoration could be done “in-line,” without long outgassing or storage time. The total liberated gas is much less than that of ADC, but the distribution is excellent and smaller gas losses owing to slower decomposition can be achieved, making the item of very practical use. Today, endothermics are commodities with many “reinventions” and well established in the global market. They are used almost in every kind of injection-molding pro cesses.26,27,32

The advantage of lower pressures can be applied in the molding process. For this reason, lower clamping force and less wear at the molds can be expected. Shut-off nozzles at machines are necessary to produce good and reliable results.

2.3.1.1 Methods

2.3.1.1.1 Standard (conventional) injection molding

The reduction of sink marks and chemical foaming of polymers for weight reduction can help achieve certain properties such as a matt surface, good fl oating, better fl ow and mold fi lling, and faster “outgassing” than ADCs. In addition, CBA powder can be easily fed into hopper of an injection-molding machine.

TABLE 2.3

Major Properties of Chemical Blowing Agents

Product

Decomp. Range

(°C)

Appr. Gas Evolution

(ml/g) Main Gases

ADC (ADCA) 200–215 220 CO, CO2, NH3

ADC (ADCA) activated 140– 215 130–220 N2, CO, CO2, NH3

DNPT 190–200 190–200 N2, NH3, CH2O

THT 245–285 180–210 N2, NH3

TSH 105–110 115 N2, H2O

OBSH 155–165 110–125 N2, H2O

TSSC 225–235 120–140 N2, CO2, NH3

5-PT 240–250 190–210 N2

NaHCO3 110–150 160–190 CO2, H2O

NaHCO3/citric comp. 130–230 110–180 CO2, H2O

52 Polymeric Foams

61259_C002.indd 5261259_C002.indd 52 11/5/2008 4:26:13 PM11/5/2008 4:26:13 PM

2.3.1.1.2 Direct gassing method

An accumulator is fi lled through an extruder,26 which is gassed primarily with nitrogen under the use of nucleants for improved foam structure, a smoother surface and less gas losses. A piston or plunger injects the melt into the mold via a single or multi-nozzle system (e.g. Figure 2.4).

2.3.1.1.3 Gas counter process

Here a good surface area is wanted. Compared with ADC, the endother-mics give better results. In a rather thin wall molding, ADC may give better fl ow and less sinks. Also modern methods like injection of “supercritical” fl uids in the melt may achieve better results due to the higher and con-trolled gas pressure of the introduced gases. Compared with the standard process, here the mold is sealed and preloaded with gas (nitrogen) and the injection takes place against this pressure buffer to obtain a good surface; the pressure is released during the shot and the mold fi lled with the help of the gas pressure inside the melt.

2.3.1.1.4 Co-injection process

This process (sandwich process, 2K process, etc.) allows the use of foaming agents in the inner, outer or inside and outside, depending on what fi nal results and applications are required. When using the foaming agent inside, the surface from the outside layer is like compact molding. The inner gas/blowing agent-containing material avoids sink marks and helps for weight reduction and excellent mold fi lling.

2.3.1.1.5 Special processes

In processes with expandable or “opening” molds after and/or during the shot, the endothermics help to achieve good surface, low densities, no or low warpage and good cell distribution with good controllable foam-ing.26,27 Today, there are combinations of various processes where the blowing agents are applied accordingly.

In summary, the main benefi ts of using foaming agents in injection molding are weight reduction, less resin consumption, no sinks, less clamping force, and a faster cycles. The kind and level of CBA affects the degree of the benefi ts.

2.3.1.2 Role of Endothermics versus Exothermics10,11,33

Table 2.4 shows a comparative summary between endothermals and exothermals, and their blends in injection molding process.

2.3.1.2.1 Surface roughness

With the increase in exothermic ADC in resins such as ABS, PPE/PPO, and PS, the increase of the surface roughness is quite signifi cant14 compared with endothermics.


61259_C002.indd 5361259_C002.indd 53 11/5/2008 4:26:13 PM11/5/2008 4:26:13 PM

Let us take ABS injection molding as an example. Conditions were con-stant, values taking on parts with the same weight with the measurement of the roughness in the centre between gating and outside rim. Comparison allows producing the same part and the same settings. The roughness data are as follows (plotted in Figure 2.5):

Non-foamed: 100•

Standard endothermic: 208•

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5

No

Endo

Enexo

ADC

Blo

win

g ag

ent

Value (microns)

Ry max

Ra min

FIGURE 2.5 Surface roughness comparison for ABS without and with various CBAs in

injection molding.

TABLE 2.4

Comparative Summary between Exothermals, Endothermals, and Their Blends in the Injection Molding Process

Properties Exothermals Enexothermals* Endothermals

Weight/density reduction Excellent Very good Good

Surface roughness Poor Reasonable Excellent

Fine cell structure Coarse Reasonable Excellent

Sink mark reduction Good to excellent Good to excellent Excellent

Cycle time reduction 0% 10–30% 20–40%

Food grade status Limited Limited No problem

Color Yellow White to yellowish White

Discoloration tendency Yes Reasonable No

Smell Pungent (NH3) Reasonable Little

Gas evolution Nitrogen Nitrogen, CO2 CO2

Gas pressure High Medium Medium/low

Environmental aspect Limited impact Limited impact Little/no impact

*Blends between exo- and endothermic foaming agents.

54 Polymeric Foams

61259_C002.indd 5461259_C002.indd 54 11/5/2008 4:26:13 PM11/5/2008 4:26:13 PM

90.0

80.0

70.0

60.0

50.0

40.0

30.0

20.0

10.0

0.0300 330 360 390 420 450

Temperature/K

Pre

ssur

e/ba

r

480 510 540 570 600

1

2 3

5

Pressure Development; Autoclave-Testing(conditions see above)

TEST PRODUCTS: 1 Azodicarbonamide2 Blend Endo/Exo 1:13 Endo Standard like Hydrocerol-Compound/CF4 Sodium Bicarbonate5 Blend Endo/Exo 4:1 like Exocerol 232

4

FIGURE 2.6 Pressure profi les for various CBAs at 0.5 K/min heating rate with nominal

content of 0.11 g/1 cm3.

Enexo (50 : 50 Endo/Exo blend): 292•

ADC: 340•

2.3.1.2.2 Gas pressure development

The pressure exposure during the decomposition is particularly important in regards to the cooling time, the tendency of uncontrolled post-expansion of the molded parts, and the cell structure. This can also infl uence the physical properties of the parts, depending on the grade of expansion rate. It is quite obvious that CBAs that liberate nitrogen have the highest gas pressure development, which is illustrated in the pressure development graph (Figure 2.6). The differences of the remaining pressure from above test under room temperature are quite drastic, as demonstrated in Table 2.5.

2.3.1.2.3 Cooling phenomena

The most signifi cant difference in application is the cooling times in com-parison with the calculation of the needed CBA. In this case the cooling time under the test conditions with a somewhat low loading of CBA will be constant. Similar results and details have been reported.9,32 The blowing agent percentage for the cooling time test was taken from gas evolution testing in Figure 2.6 and compared in Figure 2.7.

It is fair to conclude that endo- or exothermal CBAs have similar cooling time results, owing to the enormous energy (heat) content in the melt. At normal loading in injection molding the pressure of the gas during the


61259_C002.indd 5561259_C002.indd 55 11/5/2008 4:26:14 PM11/5/2008 4:26:14 PM

process and the remaining pressure are responsible for weight reduction and cooling time.

2.3.1.2.4 Weight reduction and outlook

The weight reduction in foam molding is drastically infl uenced by the wall thickness of the parts as demonstrated in these tests (Figure 2.8) just by comparing the 5 mm and 8 mm parts (same shape, different thickness). The blowing agent was a standard endothermic CBA.

In injection molding, the differences in the nature of various blowing agents are rather simple to demonstrate and the conditions can be easily repeated, and to a certain extent be translated into extrusion or similar processes where the test series are not easily attainable. It should be noted

0

10

20

30

40

50

70

60

80

90

0 0.4 0.51 0.82

Blowing agent loading (%)

Cooling time sec.

NON

ADC Enexoth. Endoth.

FIGURE 2.7 Cooling time comparison for various CBAs.

TABLE 2.5

Maximum Gas Pressure at Decomposition and Remaining Pressure in Room Temperature for Various Chemical Blowing Agents

Substance Max. Pr. (Bar)** Rem. Pr. (Bar)***

ADC 85.4 30.3

5–PT* 40.6 21.1

ADC/endo 1 : 1 42.2 15.8

Stand. endo 41.6 11.9

ADC/bic. 1 : 4 28.1 11.2

Bicarbonate 24.2 11.0

*See Reference 4.**At 495 K.***At 300 K.

56 Polymeric Foams

61259_C002.indd 5661259_C002.indd 56 11/5/2008 4:26:14 PM11/5/2008 4:26:14 PM

that the major aspect in any case is not the thermal nature of the foaming agent, but the gas pressure in the melt and the fi nal results for the suitable foaming agent.

2.3.2 Extrusion (See also Section 2.2)

2.3.2.1 Chemically Foamed Extrudates

Made with suitable chemical blowing agents (Table 2.3), the decomposi-tion must yield suffi cient gases to obtain the cellular structure. The gas formation has to take place at a temperature range close to the processing temperature. The gas should be easily dispersible within the polymer melt. The received decomposition products should be compatible with the resin and not have any negative infl uence on properties, color, and plate out, corrosion, toxic, and environmental impact.35

Generally, there is no perfect blowing agent in the market, but certain substances and blends have signifi cant market shares. Most import materials are either inorganic (like sodium bicarbonate, sodium borohy-dride, etc.) or organic (azodicarbonamide, hydrazide/sulfonylhydrazides, organic acids, semicarbazides and tetrazoles). There are exothermic and endothermic decomposition characteristics depending on which group of CBAs are preferred. Mixtures of both groups are called “enexother-mics,” and showed interesting improvements in properties.

In extrusion, the endothermals found inroads in new products, where fi ne cell structure, high throughput, lack discoloration, lack of smell, and in many cases food grade classifi cation are needed. Due to slower decom-position reactions, the distribution in the substrates is rather easy.

20

30

40

50

0

% Density reduction

0.2 0.4 0.80.6

8 mm

5 mm

1.0

27.2

35.4

41.2

22.629.2

32.7

Blowing agent loading (%)

FIGURE 2.8 PMMA density reduction at different CBA percentages for two-part thick-

ness: 5 mm and 8 mm.


61259_C002.indd 5761259_C002.indd 57 11/5/2008 4:26:14 PM11/5/2008 4:26:14 PM

In some traditional applications like rigid PVC foams, both “enexother-mal” and plain ADC have been used. The typical examples of chemically foamed extrudates are:

PVC sheets, boards, profi les, pipes

PS sheets, profi les, picture frames

PP decorative ribbons, strapping material, cable wrapping, mono- fi laments, sheets, sheets for thermoformed food packaging and microwave heating, packaging material, carpet backing

LDPE cable insulation like coaxial cables, sealants from sheet mate-rial, etc.

HDPE cable insulation, shopping bags, foam netting, marine piles, etc.

PET sheets for food packaging etc.

PLA rather new foamed material for food packaging.

2.3.2.2 General Aspects of Foam Extrusion15,34–36

For the production of foamed extrusion products, thermoplastic polymers are extruded with chemical foaming agents (Table 2.3). The foaming agent decomposes in the melted polymer and the resulting gas dissolves and disperses in the polymer melt. All common extruders can be used for foaming if the following requirements are fulfi lled:

1. The melt temperature must be high enough to guarantee a total decomposition of the foaming agent.

2. The pressure of the melt must be kept high enough to keep the gas generated by the decomposition of the foaming agent dis-solved in the polymer melt until the melt exits the extrusion die.

3. The pressure is controlling the gas escape through the hopper, which is undesirable. Pressure or gas losses in the feeding section lead to uneven and irregular results.

If the melt temperature is too low, the decomposition of the foaming agent will be incomplete, which is uneconomical, and the non-decomposed foaming agent particles can form agglomerates to clog the melt fi lter or cause undesirable pressure increase. As a result, voids, irregular cell struc-ture, or poor surface appearance are obtained.

Generally, an unusually low pressure profi le leads to what is called prefoaming; even with a subsequent pressure increase, the gas cannot be “redissolved.” This results in a large and irregular cell structure, and broken, collapsed cells. The coarse foam produced this way leads to holes in fl at and or rupture in blown fi lms, while profi les and sheets get a rough and uneven surface (shark skin).

58 Polymeric Foams

61259_C002.indd 5861259_C002.indd 58 11/5/2008 4:26:14 PM11/5/2008 4:26:14 PM

Formingzone

Metering zone Groovedfeeding zone

FIGURE 2.9 Grooved barrel extruder.

2.3.2.3 Extruder Overview: Screw, Melt Filters, and Dies8,15,34 (See also Figures 2.9–2.12)

After decades of foam extrusion, it became clear that the right understand-ing of foam extrusion process is essential. Boehringer Ingelheim published an extrusion guide for foam producers to avoid a rough start-up.34

Most single screw extruders are suitable for chemical foam extrusion. The L : D (length to diameter) ratio should be at least 24 : 1 and screws with an L : D ratio of 30 : 1 are more common. The temperature in the feeding zone should be lower than the initial decomposition temperature of the used foaming agent.

The use of a grooved barrel leads to a relatively quick pressure increase in the extruder. This is very advantageous when using foaming agent batches with low melting temperatures. As the foaming agent masterbatch melts in an early stage, and reaches the decomposition temperature earlier, the resulting gas will dissolve the melt owing to the high pressures present at the beginning and can be very well distributed in a short period of time.

When using smooth barrels, a suffi cient melt pressure is reached more slowly. Foaming agent batches with a low melting point can melt too early at the barrel wall, and the resulting gas can partially or completely escape through the hopper. In this case, the temperature of the feeding zone should be adjusted to a lower temperature, to prevent the premature decomposition of the foaming agent.

2.3.2.3.1 Screw geometry

All common screws can be used for foam extrusion, as long as there is no large pressure decrease in the single zones of the screw, which leads to undesirable prefoaming in the melt. Established screws for processing are three-zone screws (feeding, compression, metering/mixing). Good results have also been achieved with degassing screws (PS), as long as you con-sider the conditions mentioned at the beginning of this chapter.


61259_C002.indd 5961259_C002.indd 59 11/5/2008 4:26:14 PM11/5/2008 4:26:14 PM

Melting &compression

zone

Feedingzone

2. Meteringzone

Degassingzone

1. Meteringzone

FIGURE 2.12 Extruder with degassing zone.

Formingzone

Melting & compressionzone

Smooth feedingzone

Meteringzone

FIGURE 2.10 Smooth barrel extruder.

FIGURE 2.11 Pressure profi le.

60 Polymeric Foams

61259_C002.indd 6061259_C002.indd 60 11/5/2008 4:26:14 PM11/5/2008 4:26:14 PM

New problems were encountered while introducing the barrier screws in CBA foam extrusion. One of them was an increased tendency to pre-mature foaming, caused by a high pressure gradient between the barrier fl ight and the driving pitch. When using short barrier segments, the pres-sure decrease can be compensated for with higher screw speeds.

2.3.2.3.2 Melt fi lters

The use of melt strains, such as screen changers, is generally not necessary in foam extrusion. Owing to foam structure and impurities, gels and additive agglomerates are usually not visible and do not affect the foam product. However, when the impurity becomes too high to cause foaming concerns, screen pack is necessary to maintain quality products. Moreover, screen pack can improve thermal uniformity of the melt fl ow. If melt fi lters are used, it is important not to use screens that are too fi ne. This can cause a high pressure drop after the screens, possibly resulting in prefoaming.

2.3.2.3.3 Slot dies (T, fl at, and coathanger dies, etc.)

This type of die is used for the production of thin fi lms (with a chill-roll calendar stack), or thicker fi lms, and sheets (with a calendar), respectively. When using slot dies with a restrictor bar (for a better dispersion of the melt), the restrictor bar should not be closed too much. If the restrictor bar is closed too much, pressure reduction can occur right after the restrictor bar, which can result in prefoaming.

2.3.2.3.4 Profi le dies, tubular dies, etc.

Compared with the cross-section of a slot die, the cross-sections of profi le dies generally cannot be changed or adjusted. The product geometry seems to dictate the die design. Generally, the best practice in die design is to keep a short land length to maintain high melt pressure up to the die lip. This is also true for the production of foamed blown fi lm. It must also be noted that the pressure characteristics of a given die can be affected by many factors, such as type of resin, resin viscosity, temperature, desired density reduction, output rate, and actual product cross-section.

2.3.2.4 Coextrusion

In general, single layer systems can be translated to the coex design with minimum modifi cation. Many coextruded foam products are produced with a foamed inner layer and solid, non-foamed outer layers. In this case, it is very important, to select the right material. For the outer layer, a “softer” material is recommended, while a material that is somewhat harder is recommended for the inner foamed layer. This type of structure is suggested due to the fact that a foamable melt has better fl ow characteris tics (lower apparent viscosity) compared with a solid melt of the same resin. If the layers differ in viscosity, it can result in poor or damaged foam structure.


61259_C002.indd 6161259_C002.indd 61 11/5/2008 4:26:15 PM11/5/2008 4:26:15 PM

2.3.3 Other Foam Processes

Quite a few popular exothermic CBA foams can be found elsewhere;37,38 for example, PVC (plasticized), different processes in carpet backing, cross-linked foams, woodplastics, and the classic rubber materials.

2.3.3.1 Foam Blow Molding with Endothermics16,28,39

The fi rst trials for the use on an industrial scale where done at the end of 1980s and early 1990s. Some applications had already demonstrated the possibilities in packaging, transportation, automotives, and leisure. The endothermal systems worked out satisfactorily and gave surprisingly good results, due to the fi ne cell structure (self-nucleation) and slow gas release during the process (refer to graphs and pictures below). Foamed walls have higher resistance against UV infl uence, which is an advantage in containers and bottles. Figure 2.13 shows the bottle products, cell struc-ture, and foam density at various blow pressures.

2.3.3.2 Rotational Molding Aspects of Endothermic Blowing Agents

The use of blowing agents in rotational molding is not new, but only recently the market began to accept this alternative technology compared with other

Blowpressure

Gas pressureinside of thefoam cells

Blow pressure

0.7

0.75

0.8

0.85

0.9

0.95

0.8 1 1.2 1.5 1.8

Density (g/cm3)

Blow pressure (bar)

FIGURE 2.13 Blow molding bottles: products, cell structure, and density chart.

62 Polymeric Foams

61259_C002.indd 6261259_C002.indd 62 11/5/2008 4:26:15 PM11/5/2008 4:26:15 PM

molding techniques.11,40–42 The advantages are similar to the blow molding process: weight reduction, small gap between walls, higher stiffness or thicker walls. In reality, modifi ed ADC, endothermics, enexothermics and other exothermic blowing agent can be used. The traditional blowing agent for this application is mainly OBSH. This material is also found in ready-to-use compounds. Unfortunately OBSH is not food approved and the most recent endothermics gave undesirable results in the testing. It is very likely a new combination of food-approved endothermics (acid and carbonate compo-nents) could do the trick.

The temperature plays an important role in this process, owing to the fact that no “screw machines” with good mixing, shear and temperature control could be used. The best results are obtained by using precom-pounded materials. The basic results and problems are demonstrated in Figure 2.14. The test was a static test by using LLDPE with drum blended blowing agents under various temperature conditions and measuring later the obtained density as illustrated in Figure 2.14.

2.4 Summary

Today, the use of blowing or foaming agents is a well-established technol-ogy especially in the well-known additive fi eld. Many polymeric foam products contain CBA. More and more endothermic CBA has been used as blowing agent and nucleating agent. After decades of practice, most exothermics are produced in Asia. The traditional manufacturers such as Bayer gave up or transferred production to China or other countries. Also, the markets for ADC and other exothermics are very large in Asia. Up to

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

150 160 170 180 190 200 220 240Temperature (°C)

Endotherm. BEnexoth 3:1Stand. Endoth.ADC modif.OBSHADC

Density(g/cm3)

FIGURE 2.14 LLDPE rotational molding with various CBAs.


61259_C002.indd 6361259_C002.indd 63 11/5/2008 4:26:16 PM11/5/2008 4:26:16 PM

80% of the applications are in PVC, and cross-linked material (EVA, PE, rubber, etc.) primarily in Asia (India, China, Indonesia, and others). But there is still room for new developments, as the possibilities for endother-mics have yet to be fully realized. Foaming agents will continue to carve out their place in the global markets and in various applications with healthy, rising interest. After CBA foaming, polymer’s property spectrum seemed to expand: lower density, better insulation properties, energy absorption, shock dissipation, and many other production advantages with almost unlimited possibilities.

The success in microcellular processing with supercritical carbon dioxide, and cross-linked PE and PP into low-density foam with CBA certainly brings a strong message to the CBA foaming industry. The endothermic CBA has the unique feature of dispersed cooling to locally stabilize the polymeric melt as opposed to melt cooling devices in PBA foaming. Combining the above processes, low-density CBA foam processing could be developed when chemistry and mechanical developments are opti-mized. It can certainly open up a lot of application gates. A bright future can thus be anticipated.

2.5 Abbreviations

2K Two-component molding5-PT 5-Phenyl tetrazoleABS Acrylonitril butadiene styreneADC (ADCA) AzodicarbonamideCBAs Chemical blowing agentscm3 Cubic centimeterCitric comp. Citric acid componentDecomp. DecompositionDNPT Dinitroso pentamethylene tetramineEnexo Endo-/exothermic blendsEVA Ethylene vinyl acetateg GramGCP Gas counter-pressure methodGPPS General purpose polystyreneHDPE High density polyethyleneK Degrees KelvinLDPE Low-density polyethyleneLLDPE Linear low-density polyethyleneMax. pr. Maximum pressuremin Minute

64 Polymeric Foams

61259_C002.indd 6461259_C002.indd 64 11/5/2008 4:26:16 PM11/5/2008 4:26:16 PM

Min. pr. Minimum pressuremm Millimetermp Melting pointNaHCO3 Sodium bicarbonateNaHCO3/citric comp. Sodium bicarbonate/citric acid componentOBSH 4,4’-Oxybis(benzenesulfonyl hydrazide)PE PolyethylenePET Polyethylene terephthalatePLA Polylactic acidPP PolypropylenePPE/PPO Polyphenylene ether/polyphenylene oxidePS PolystyrenePS/GPPS Polystyrene/general purpose polystyrenePVC Polyvinyl chlorideR11/R12 CFC-11/CFC-12 � CFCl3/CF2Cl2

Ra min Lowest value in micronsRy max Highest value in micronsSBH Sodium borohydrideStand. endo Standard endothermal foaming agent

( hydrocerol-like compound)THT 2,4,6-Trihydrazino-1,3,5-triaazineTSH Toluene sulfonyl hydrazideTSSC p-Toluene sulfonyl semicarbazideUSM United Shoe Machinery Company

References

1. Neumüller, O.-A. “Römpps Chemie-Lexikon.” 7. Aufl ., Band (part) 5 W. Keller & Co., stuttgart, (1975): 3254.

2. Landrock, A. H. Handbook of Plastic Foams. Plastics Techn. Evaluation Centre, Dover, NJ, 1985.

3. Müller, E. et al. “Aliphatische Diazo- und Azoverbindungen in der Kunst-stoffchemie.” Angewandte Chemie 63 (1951): 18–20.

4. Lober, F. “Entwicklung und Bedeutung von Treibmittel bei der Herstellung von Schaumstoffen aus Kautschuk und Kunststoffen.” Angewandte Chemie 64 (1952): 65–76.

5. Kirkland, C. “Blowing agents—New alternatives on all fronts.” Plastics Technology (1986): 83–87.

6. Scholz, D. and Amecke B. “Foaming agents.” In Modern Plastics Encyclopaedia, Modern Plastics, New York, 1989, pp. 166–170.

7. Scholz, D. “Structural foams in technical resins using endothermal blowing agents.” In Proceedings of the Structural Plastics Conference, April 25–28 1993, San Francisco, pp. 95–98.


61259_C002.indd 6561259_C002.indd 65 11/5/2008 4:26:16 PM11/5/2008 4:26:16 PM

8. Scholz, D. “How to get the most out of endothermic blowing agents.” In RAPRA Conference Blowing Agent Systems; Formulations & Processes, 19 February 1998, Shrewsbury Shropshire, UK.

9. Kretzschmann, G. et al. “Citric acid esters.” USP 4,572,740. 10. Kosin, J. A. et al. “High temperature endothermic blowing agents composi-

tions and applications.” US patent 5,009,809. 11. Wason, S. K. “Endothermic blowing agents compositions and applications.”

US patent 5,009,810. 12. Garcia, R. A. “Endothermic blowing agents for strengthening weld lines in

molded thermoplastic resins and products.” US patent 5,037,580. 13. Hansen, R. H. and Martin, W. M. I&SC Product Research and Development

3 (1964): 137–141. 14. Pontiff, T. M. “Factors affecting foam cell nucleation in direct gassed foam

extrusion.” In Proceedings of the FOAMPLAS Conference, 4–5 November 1997, Mainz, pp. 251–261.

15. Scholz, D. “25 Years endothermic blowing or foaming agents—Is that enough?” In Proceedings of the Structural Plastics Conference, 27th annual design competition held by the Society of the Plastics Industry Inc.’s Structural Plastics Division, 18–21 April 1999, Boston, MA.

16. “Foamed Plastics.” In Ullmann’s Encyclopaedia of Industrial Chemistry, Gerhartz, W., Schulz, G., et al., Eds. 5th ed., Vol. A11, 1985–1996, Weinheim, Germany, 435–464.

17. Houston, J. C. et al. Koppers & Comp. (Composition Comprising Polystyrene. . .). US Patent 2,941,964, 1960.

18. Collins, F. H. “(Dow Chemical), Controlled density—Polystyrene foam.” SPE Journal 16 (1960): 705–709.

19. Hansen, R. H. “Production of fi ne cells in the extrusion of foams.” SPE Journal 18 (1962): 77–82.

20. Kropp, D. et al. “Foam extrusion of thermoplastic elastomers using CO2 as blowing agent.” In Annual Technical Conference Antec ’97 April 27–May 2, Toronto, Canada, pp. 3473–3478.

21. Fricke, H. “Coated citric acid/Zitronensäure Type H.” Technical Information, C.H. Boehringer Sohn, Ingelheim, April 1967.

22. Lübke, G. “Thin-walled components.” KU Kunststoffe Plast Europe 92 (2002): 12, 36–38/79–82.

23. Padareva, V. et al. “Modifi cation of blowing agent system based on sodium bicarbonate with activated natural zeolite.” Journal of Materials Science Letters 17 (1998): 107–109.

24. Kretzschmann, G. et al. “Nucleating agents (Pore Regulators) for the Preparation of Direct Gassed Thermoplastic Foams.” US patent 5,225,107.

25. Scholz, D. “The position of endothermic nucleating and blowing agents in thermoplastic foams.” Paper presented at Ausplas Conference, 12–17 October 1987, Melbourne, Australia.

26. “Engineering structural foam.” Celanese Plastics Company Brochure, Newark, NJ, 10, 1979/1980(?).

27. “Engineering structural foam.” GEP—General Electric Plastics Europe Brochure, 15–17, 1980/1981(?).

66 Polymeric Foams

61259_C002.indd 6661259_C002.indd 66 11/5/2008 4:26:16 PM11/5/2008 4:26:16 PM

28. Spiekermann, R. “New applications for chemical foaming agents in thermoplastics.” In Proceedings of the FOAMPLAS Conference, 1997, Mainz, pp. 263–275.

29. Hurnik, H. et al. “Chemische treibmittel.” Kunststoffe 86 (1996): 997–1001. 30. Lapierre, R. et al. “Chemical blowing agent compositions.” US patent

4,769,397. 31. Jacobs, P. M. “Endothermic blowing agents and structural foam quality.” In

Proceedings of 12th Annual Structural Foam Conference and Part Competition, The Society of the Plastic Industry, Inc., 7–9 May 1984, San Francisco, CA.

32. Scott, R. M. “Polymeric performance in thin wall structural foam.” Proceedings of 12th Annual Structural Foam Conference and Part Competition, 1984.

33. Scholz, D. “New aspects using endothermal foaming agents.” Paper pre-sented at Ausplas Conference, Stream 3, Commercial & Technical Sessions, 13 October 1993, Melbourne, Australia.

34. Boehringer Ingelheim, K.G. “Chemically blown thermoplast foams: Extrusion.” Technical Information/Processing Guides. Issues: 07/1991 and 03/1997.

35. Berghaus, U. et al. “Foaming of fi lm and sheet from polypropylene.” Paper presented at the Polypropylene ‘95 Conference, September 11–12, 1995, Zurich.

36. Hensen, F. Plastics Extrusion Technology. Hanser Publishers, Munich, New York, 1998, pp. 430–487.

37. Throne, J. L. Thermoplastic Foams. Sherwood, Hinckley, Ohio, 1996. 38. Lee, S. T., Park, C. B., and Ramesh, N. S. “Wood composite foams.” In Polymeric

Foams: Science and Technology. Taylor and Francis, Boca Raton, 2007. 39. Spiekermann, R. “Development of large and small applications for foamed

blow moulding.” Presentation at National Plastics Exhibition, Chicago 1997. 40. Scholz, D. “Blowing agents for rotational moulding.” Presented at the 16th

Annual ARM Australia Conference, March 13–15, 1994, Melbourne. 41. “Foaming of thermoplastics—Blowing agents for rotational moulding:

Plastics technology.” Hong Kong Plastics Technology Centre Ltd. 18 (1994): 30–32.

42. Büttner, H. “Foaming of thermoplastic resins by rotational moulding”, Presented at the ARM Australasia Conference, Bali, Indonesia, 1997.


61259_C002.indd 6761259_C002.indd 67 11/5/2008 4:26:16 PM11/5/2008 4:26:16 PM

61259_C002.indd 6861259_C002.indd 68 11/5/2008 4:26:16 PM11/5/2008 4:26:16 PM

3Foam Extrusion Using Carbon Dioxide as a Blowing Agent

Walter Michaeli, Dirk Kropp, Robert Heinz, and Holger Schumacher

CONTENTS

3.1 Introduction .................................................................................... 703.2 Effect of Carbon Dioxide on the Flow Behavior of Polymer Melts ................................................................................. 70

3.2.1 Measuring the Shear Viscosity of Blowing Agent Containing Melts .................................................................. 71

3.2.2 Calculation of the Viscosity of the Polymer/Blowing Agent Mixtures ..................................................................... 72

3.2.3 Flow Behavior of Carbon Dioxide Containing Polymer Melts ....................................................................... 75

3.3 Infl uence of the Flow Channel Geometry on the Foam Quality .................................................................................. 783.4 Infl uence of Spider Legs on the Thickness Distribution

of Foamed Sheets ........................................................................... 813.4.1 Infl uence of Melt Temperature ........................................... 843.4.2 Infl uence of Die Temperature ............................................. 863.4.3 Infl uence of Spider Geometry ............................................. 88

3.5 Infl uence of the Pressure Profi le in the Extrusion Die on the Foam Structure ................................................................... 90

3.6 Nomenclature ................................................................................. 97References .............................................................................................. 97

61259_C003.indd 6961259_C003.indd 69 10/25/2008 12:35:28 PM10/25/2008 12:35:28 PM

3.1 Introduction

Since chlorofl uorocarbons (CFCs) and some hydrochlorofl uorocarbons (HCFCs) were banned due to their ozone depletion potential and their contribution to the greenhouse effect, the use of inert gases as physical blowing agents in foam extrusion has gained interest. Among these, carbon dioxide (CO2) is one of the most promising blowing agents for foam extrusion. It is environmentally benign and, due to its high availability, economical reasonable. On the other hand, it has a lower solubility, a higher diffusivity, and a stronger plasticizing effect in most polymers compared to traditional blowing agents. Therefore, the foam extrusion process needs to be adjusted to the properties of CO2. Particularly in die design, the specifi c requirements for foam extrusion have to be taken into account when CO2 is used as a blowing agent. For instance, the plasticiz-ing effect of the blowing agent changes the fl ow behavior of the polymer melt that leads to a reduced pressure in the die. In combination with the lower solubility of CO2 in most polymer melts, the risk of premature foam-ing inside the die is increased. The shape of the fl ow channel and spider legs therefore also need special attention to prevent visible defects in the foam product. In particular, the pressure profi le at the die exit has a big infl uence on the foam structure.

3.2 Effect of Carbon Dioxide on the Flow Behavior

of Polymer Melts

The rheological behavior of polymer melts is one of the most important factors to describe processes in polymer processing by a theoretical approach. In particular, the fl ow behavior of polymer melts is an essential key factor in the design of extrusion dies. In order to predict the pressure profi le, the output rate, and the velocity distribution in the die, the viscos-ity of the polymer has to be known.

In foam extrusion with physical blowing agents, the blowing agent is injected into the extruder barrel and dissolved within the polymer melt. The dissolved blowing agent acts like a plasticizer and decreases the vis-cosity of the polymer melt. Depending on the blowing agent concentration and the polymer/blowing agent interaction, the pressure profi le can be affected noticeably. However, the pressure in the die has to be kept above the solubility pressure of the blowing agent in the polymer to prevent a premature foaming of the melt. Since the entire extrusion process is affected by the blowing agent, it is essential to gain knowledge of the impact of the blowing agent on the fl ow behavior.

70 Polymeric Foams

61259_C003.indd 7061259_C003.indd 70 10/25/2008 12:35:28 PM10/25/2008 12:35:28 PM

3.2.1 Measuring the Shear Viscosity of Blowing Agent Containing Melts

The standard rheometers that are commonly used for the measurement of shear viscosity of polymer melts are the capillary rheometer and the rotary rheometer. Both types of rheometer can be used in a wide shear rate range but it is not possible to introduce a defi ned amount of blowing agent into the melt. An alternative method is to charge polymer granules with blow-ing agent prior to the viscosity measurement, but the actual blowing agent concentration in the polymer melt during the measurement can hardly be determined. Since rotary rheometers are not capable of building up a pres-sure that is high enough to prevent foaming of the melt, even pre-charged polymer cannot be measured in this type of rheometer. By contrast, capillary rheometers can operate under high pressure, but during the melting of the polymer, blowing agent is lost through the capillary and the gap between cylinder and plunger. Furthermore, a premature foaming of the melt inside the capillary will lead to incorrect values. Consequently, standard rheometers are not capable of measuring the viscosity of polymer melts in dependence on the blowing agent concentration.

Han et al. used a capillary die to investigate the fl ow behavior of poly-ethylene (PE) and polystyrene (PS) melt containing chemical blowing agents1 and charged with fl uorocarbon blowing agents.2,3 The volume fl ow rate through the die was varied by changing the rotational speed of the extruder screw. Thus, extrusion conditions; that is, shear rate, mixing and residence time in the extruder were also changed. The shear rate in the die could only be varied in a range of approximately 100–300 s�1.

Gendron et al. used a commercial online rheometer for rheological mea-surements of mixtures of PS with fl uorocarbons.4,5 The online rheometer was positioned in a side stream of the extruder and the fl ow rate through the rheometer slit was controlled by two gear pumps. This guarantees constant process conditions in the rheometer isolated from process fl uctu-ations. The shear rates in the rheometer were varied between 1 and 100 s�1. However, an observation of the melt stream is not possible.

Kropp developed a special in-line rheometer die which enables the measurement of shear viscosity of blowing agent charged melts at process conditions.6,7 The rheometer is designed as a slit capillary rheometer and can be mounted directly on the extruder. Using different inserts, the height of the 200 mm long slit can be varied between 2, 3.5, and 5 mm. The width of the slit is 50 mm. Along the fl ow channel, the pressure is measured at three different points in an interval of 60 mm in order to determine the pressure gradient in the slit. The level of pressure in the in-line rheometer die can be adjusted by means of a throttle at the exit of the in-line rheometer die. Glass inserts at both sides of the rheometer allow the observation of the melt fl ow in the slit. Figure 3.1 shows a picture of this in-line rheometer die.

Foam Extrusion Using Carbon Dioxide as a Blowing Agent 71

61259_C003.indd 7161259_C003.indd 71 10/25/2008 12:35:28 PM10/25/2008 12:35:28 PM

During the operation of the in-line rheometer die, the mass fl ow rate through the slit capillary can be varied by a bypass upstream of the rheo-meter. By opening a valve, more material fl ows through the bypass and the fl ow rate through the rheometer decreases. Using this technique, the fl ow rate through the rheometer and thus the shear rate can be varied in a wide range without changing the process settings of the extruder.

A schematic of the in-line rheometer die and the corresponding pressure course is depicted in Figure 3.2. Assuming that no premature foaming occurs in the rheometer, a linear pressure drop rate in the slit is expected. By closing the throttle at the exit of the in-line rheometer die, the level of the pressure is increased as illustrated in the diagram (dotted lines).

Besides the viscosity measurement, the in-line rheometer die can also be used to investigate the bubble formation (Figure 3.3). Releasing the pressure in the rheometer by opening the throttle at the exit of the die, the initial point of bubble formation moves into the observable region in the rheometer. Since the pressure profi le in the slit is known, the critical pressure; that is, the pressure when fi rst bubbles are initiated, can be determined. This critical pressure is an essential value in the design of foam extrusion dies.

3.2.2 Calculation of the Viscosity of the Polymer/Blowing Agent Mixtures

The shear rate and the melt viscosity are derived using the concept of rep-resentative viscosity.8–10 This method assumes that certain points exist in every laminar fl ow where the shear rate of both Newtonian and shear-thinning fl uids are identical. These points are called representative points.9

FIGURE 3.1 In-line rheometer die.

72 Polymeric Foams

61259_C003.indd 7261259_C003.indd 72 10/25/2008 12:35:28 PM10/25/2008 12:35:28 PM

The true shear rate for shear thinning fl uids is derived from the Newtonian shear rate multiplied with a factor e considering the geometry of the fl ow channel (Equation 3.1).

g._ �

6 ¥ V.m ______

B ¥ H2 e� (3.1)

In principle, the geometric factor e depends on the fl ow characteristics of the material. However, for most polymer melts with fl ow exponents m in the range 2–4, it can be considered as constant for a given fl ow channel shape.11 Equation 3.2 shows the geometric factor for a rectangular slit geometry of the fl ow channel.

e� � 0.772 (3.2)

FIGURE 3.2 Pressure course in the in-line rheometer die with small throttle gap. [From

Kropp, D. Extrusion thermoplastischer Schäume mit alternativen Treibmitteln (Extrusion of thermoplastic foams using alternative blowing agents). PhD Thesis, RWTH Aachen, 1999.]

Pre

ssur

e p

(bar

)

Flow length L (mm)

p3

p2

p1

p1 p2 p3

Adjustablethrottle

Pressure transducer

In-line rheometer die

Decreasingthrottle gap


61259_C003.indd 7361259_C003.indd 73 10/25/2008 12:35:28 PM10/25/2008 12:35:28 PM

The true viscosity of the melt/blowing agent mixture can be calculated from the pressure loss �p at a certain fl ow channel length L with Equation 3.3:

h– � Dp ¥ B ¥ H3

__________ 12 ¥ L ¥ Vm

(3.3)

The mass fl ow rate m through the rheometer is determined by weighing the melt extruded in a certain period of time. The volume fl ow rate of the mixture Vm needed for the calculation of shear rate and viscosity is derived by the following equation:

Vm � m ___ rm (3.4)

FIGURE 3.3 Pressure course and bubble formation in the in-line rheometer die with

large throttle gap. [From Kropp, D. Extrusion thermoplastischer Schäume mit alternativen Treibmitteln (Extrusion of thermoplastic foams using alternative blowing agents). PhD Thesis,

RWTH Aachen, 1999.]

74 Polymeric Foams

61259_C003.indd 7461259_C003.indd 74 10/25/2008 12:35:28 PM10/25/2008 12:35:28 PM

The density of the polymer/blowing agent mixture rm can be derived using a linear mixing law. Since blowing agents are generally soluble in polymer melts, the density of the mixture can be considered as the density of a polymer solution with a certain weight fraction of blowing agent wS:

12

rm � rP � rs _____________________

(1 � wS) � rS � wS � rP (3.5)

The temperature-dependent density of the polymer rP(T) is given by the melt density at a reference temperature T0 and the expansion coeffi cient a:13

rP(T) � r(T0) _______________

1 � a (T � T0) (3.6)

In Table 3.1,13 the expansion coeffi cient and the density at a reference tem-perature are listed for some standard thermoplastics. Since the density of polymers is also dependent on the pressure, more accurate values can be taken of p–v–T plots.

Since most blowing agents are in a supercritical state at process condi-tions, they cannot be considered as an incompressible fl uid. In this state their density is dependent on both temperature and pressure. Table 3.2 lists the density of CO2 in a range relevant for foam extrusion.14

3.2.3 Flow Behavior of Carbon Dioxide Containing Polymer Melts

Using the in-line rheometer die, the viscosity of polymers charged with different concentrations of blowing agent were measured.15,16 Three poly-mer types that are commonly used in foam extrusion were chosen for these measurements: polystyrene (PS), low-density polyethylene (LDPE) and polypropylene (PP). The viscosity curves for the polymers without blowing agent were determined using a high-pressure capillary rheo-meter with a slit capillary having nearly the same height/width ratio like the in-line rheometer die. Viscosity measurements of PS using both the

TABLE 3.1

Parameters for the Calculation of Melt Density of Unfi lled Standard Polymers

Expansion Coeffi cient

a (1/K)

Reference Density

r0 (g/cm)

Reference Temperature

T0 (°C)

LDPE 0.69 ¥ 10�3 0.801 115

HDPE 0.69 ¥ 10�3 0.792 131

PP 0.61 ¥ 10�3 0.759 186

PS 0.56 ¥ 10�3 1.029 84

Source: Menges, G. “4. korrigierte und aktualisierte.” In Werkstoffkunde Kunststoffe. Aufl age,

Carl Hanser Verlag, München, 1998.


61259_C003.indd 7561259_C003.indd 75 10/25/2008 12:35:29 PM10/25/2008 12:35:29 PM

high-pressure capillary rheometer and the in-line rheometer die have proven that these methods lead to the same results.17 Since the viscosity measurement of uncharged melts at such low temperatures which are required in foam extrusion is impossible, the curves were determined by means of the time–temperature superposition principle developed by Williams et al.18

Figure 3.4 shows the viscosity curves of PS 158 K (BASF AG, Ludwigshafen, Germany) at 180°C with different CO2 concentrations. The decreasing viscosity with increasing concentration of CO2 shows the plas-ticizing effect of the blowing agent. However, this plasticizing effect seems to reach a certain limit: the higher the blowing agent concentration, the lower the degree of viscosity reduction. Furthermore, the curves of differ-ent blowing agent concentrations are approximately parallel; that is, The shear thinning behavior of PS is not affected by CO2.

The viscosity of LDPE (Stamylan 2102 TN 26, DSM,* Geleen, The Netherlands) was measured at temperatures of 120°C, 130°C, and 140°C, each with 0.5%, 1%, and 1.5% of CO2. As an example, the viscosity curves at 140°C are shown in Figure 3.5. Again, the viscosity reduction due to the

TABLE 3.2

Density of CO2 in kg/m3 Depending on Pressure and Temperature

Pressure

(Bar)

Temperature

50°C 100°C 150°C 200°C 250°C 300°C

10 17.05 14.52 12.69 11.29 10.17 9.27

20 35.61 29.77 25.76 22.78 20.46 18.59

30 56.05 45.82 39.23 34.48 30.86 27.97

40 78.90 62.75 53.11 46.40 41.37 37.41

50 104.96 80.68 67.43 58.54 52.00 46.91

60 135.47 99.69 82.19 70.89 62.73 56.46

70 172.50 119.90 97.39 83.45 73.57 66.06

80 219.78 141.41 113.02 96.18 84.49 75.69

90 285.40 164.31 129.08 109.09 95.48 85.36

100 388.06 188.67 145.56 122.16 106.54 95.04

150 699.55 332.65 233.90 189.47 162.43 143.52

200 784.96 481.18 327.52 258.89 218.76 191.71

250 834.50 590.49 415.35 326.65 274.16 238.99

300 870.59 663.91 492.46 389.63 326.92 284.47

350 899.48 716.05 556.52 447.00 376.35 327.64

400 923.72 756.74 608.01 497.72 422.11 368.40

Source: N. N. VDI-Wärmeatlas. VDI-Verlag, Düsseldorf, 1984.

* LDPE 2102 is now distributed by Saudi Basic Industries Corporation (SABIC).

76 Polymeric Foams

61259_C003.indd 7661259_C003.indd 76 10/25/2008 12:35:29 PM10/25/2008 12:35:29 PM

blowing agent can be seen. Contrary to the results with PS, the plasticiz-ing effect of CO2 on LDPE is signifi cantly lower.

Since the development of new PP grades with improved melt elasticity and high melt strength (HMS-PP), the use of PP in foam extrusion has been risen.19 The viscosity of a HMS-PP developed for foam extrusion (Pro-fax 814, Montell*) was measured at 170°C using the in-line rheometer

FIGURE 3.4 Viscosity curves of polystyrene PS 158 K at 180°C and different CO2 concentra-

tions. [From Kropp, D. Extrusion thermoplastischer Schäume mit alternativen Treibmitteln (Extrusion of thermoplastic foams using alternative blowing agents). PhD Thesis, RWTH Aachen, 1999.]

10 100 1000100

1000

10000

Vis

cosi

ty (

Pas

)

Shear rate (1/s)

0% CO2 1% CO2 2% CO2 3% CO2

FIGURE 3.5 Viscosity curves of LDPE stamylan 2102 at 140°C and different CO2 concentra-

tions. [From Kropp, D. Extrusion thermoplastischer Schäume mit alternativen Treibmitteln (Extrusion of thermoplastic foams using alternative blowing agents). PhD Thesis, RWTH Aachen, 1999.]

10 100 100

1000

10000

1% CO2 1.5% CO2

Shear rate (1/s)

Vis

cosi

ty (

Pas

)

0.5% CO2

* Profax PF814 is now distributed by Basell Polyolefi ns.


61259_C003.indd 7761259_C003.indd 77 10/25/2008 12:35:29 PM10/25/2008 12:35:29 PM

die (Figure 3.6). Similar to the results of LDPE, only a slight decrease in viscosity can be observed. Analog to the other tests, CO2 seems to have just about no effect on the shear thinning behavior of PP.

3.3 Influence of the Flow Channel Geometry on the

Foam Quality

Inert gases like CO2 and N2 possess a lower solubility in most polymer melts than hydrocarbons or fl uorocarbons. Therefore, a higher risk of pre-mature foaming of the polymer/blowing agent mixture in the die exists when these gases are used as blowing agents. Particularly at points of high shear rates, premature foaming in the die may occur even at pressures above the solubility pressure of the blowing agent. Such high shear rates appear primarily at variations in the cross-section of fl ow channels; for example, convergent or divergent fl ow channels, breaker plates, or spider legs in annular dies. Furthermore, a pressure decrease at dead spots can also lead to a supersaturation of the melt and thus to premature foaming.

In order to examine the effect of cross-sectional variations in fl ow channels on bubble formation, different inserts of dissimilar shape were positioned in the 3.5 mm slit of the in-line rheometer die.16,20 The glass windows at the side of the slit allowed the observation of the melt and thus the study of

FIGURE 3.6 Viscosity curves of polypropylene Profax PF 814 at 170°C and different

CO2 concentrations. [From Kropp, D. Extrusion thermoplastischer Schäume mit alternativen Treibmitteln (Extrusion of thermoplastic foams using alternative blowing agents). PhD Thesis,

RWTH Aachen, 1999.]

10 100 1000100

1000

10000

Shear rate (1/s)

Vis

cosi

ty (

Pa

s)

0% CO2 2% CO2 3% CO2 4% CO2

78 Polymeric Foams

61259_C003.indd 7861259_C003.indd 78 10/25/2008 12:35:30 PM10/25/2008 12:35:30 PM

the initiation of bubble formation. The inserts had a height of 2.5 mm and angles of 30°, 45°, 60°, and 90° at the downstream side, respectively. A combination of LDPE (Stamylan 2102 TN26) charged with 2% CO2 at 130°C was chosen as the polymer/blowing agent mixture for the tests. By adjusting the throttle at the die exit, the level of pressure at the down-stream edge of the inserts was kept constant during the tests.

The fl ow behavior of the CO2-charged melt at the inserts with a down-stream angle of 30° and 45° is compared in Figure 3.7. At the fl ow channel insert with a downstream angle of 30° (upper picture), no premature foaming occurred at the downstream edge. Bubbles were fi rst initiated in the dead space downstream of the insert. Subsequently, they grew due to blowing agent diffusing into the bubbles. In contrast, the bubbles were initiated immediately at the downstream edge of the insert with 45° even though the pressure at this point was higher than the solubility pressure of the blowing agent (Figure 3.7, lower picture). Again, the bubbles grew and caused visible defects in the foam structure of the extrudate.

Besides the practical examination of the bubble formation, the melt fl ow at the inserts was calculated with the fi nite-element method (FEM).21 Figure 3.8 shows the qualitative results of the isothermal, two-dimensional fl ow simulation at the insert with a downstream angle of 45°. Regarding the pressure distribution in the narrowed fl ow channel, the isobars run parallel as expected. Due to the fl ow defl ection, a high-pressure gradient is generated at the downstream edge of the insert. This is shown by the

1 mm

α = 45°

α = 30°Melt flowdirection

Melt flowdirection

FIGURE 3.7 Bubble formation at fl ow channel expansions. [From Heinz, R. Prozess-optimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD

Thesis at RWTH Aachen, 2002.]


61259_C003.indd 7961259_C003.indd 79 10/25/2008 12:35:30 PM10/25/2008 12:35:30 PM

isobars running very close to each other. Downstream of the fl ow channel expansion, the distance between the isobars becomes wider; that is, the pressure loss is lower due to the lower velocity of the melt.

The visualization of the melt fl ow shows a symmetrical velocity distribution in the narrowed fl ow channel and downstream of the fl ow channel expansion. The velocity at the walls is set to zero, which is repre-sented by the dark color. The maximum velocity is reached in the center of the narrowed fl ow channel. The shear rate in the melt/blowing agent mixture can be assessed by the distance between two isotachs (lines of equal velocity). The maximum shear rate occurs close to the walls of the narrowed fl ow channel.

Assuming that the bubbles are initiated solely by the pressure drop below the solubility pressure of the blowing agent, the foaming should occur along one of the isobars shown in Figure 3.8. This seems to be valid for the experiments with the 30° insert. However, when using the insert with 45° downstream angle, the foaming starts at the downstream edge reproducibly. Consequently, the bubble nucleation in the melt fl ow has to be infl uenced by further mechanisms.

Shear effects are believed to have an infl uence on bubble nucleation in foam extrusion. Lee stated in his modifi ed cavity model that it is not only the degree of supersaturation and the amount of nucleators that infl uence bubble nucleation in foam extrusion; shear forces also have an effect.22 Due to the symmetric velocity distribution in the fl ow channel, bubble nucleation should also be observable at the opposite side of the down-stream edge of the 45° insert if shear-induced nucleation is assumed. Furthermore, the shear rate distribution in the narrowed fl ow channel is

FIGURE 3.8 Simulated melt fl ow at a channel expansion. [From Heinz, R. Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD Thesis at RWTH

Aachen, 2002.]

Pressure distribution(isobars)

Velocity distribution(isotachs)

Melt flowdirection

Melt flowdirection

80 Polymeric Foams

61259_C003.indd 8061259_C003.indd 80 10/25/2008 12:35:30 PM10/25/2008 12:35:30 PM

the same in all experiments, but bubbles are only nucleated at the down-stream edge when the downstream angle of the insert exceeds 45°.

As illustrated in Figure 3.8, the local pressure gradient at the down-stream edge of the insert is very high. Heinz explains the premature foaming at the downstream edge of the inserts by the high pressure gradient at this point.20 Figure 3.9 shows the pressure gradients in the narrowed fl ow channel and at the downstream edge of the inserts with a downstream angle of 30°, 45°, and 60°, respectively. It is obvious that the pressure gradient at the downstream edge is signifi cantly higher than in the fl ow channel. Furthermore, it can be seen that the pressure gradient at the downstream edge increases with the downstream angle of the insert.

To prevent premature foaming in foam extrusion dies, Heinz proposed avoiding high local pressure gradients, particularly at variations of cross-sections and breaker plates. Therefore, variations in cross-section should be provided with small angles and adequate radii.20

3.4 Influence of Spider Legs on the Thickness Distribution

of Foamed Sheets

In extrusion of low-density foam sheets, annular dies are the most commonly used dies.23 The mandrel in these dies is supported either by one spider leg (single-spider die), two spider legs (dual-spider die), or

FIGURE 3.9 Pressure gradients at the downstream edge of the fl ow channel inserts. [From

Heinz, R. Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD Thesis at RWTH Aachen, 2002.]

30 45 60 0

100

200

30 45 60 0

10

20

Downstream angle (°)

Pre

ssur

e gr

adie

nt (

MP

a/s)

Pre

ssur

e gr

adie

nt (

MP

a/s)

Downstream angle (°)

Pressure gradientin the narrowed slit

Pressure gradient atthe downstream edge


61259_C003.indd 8161259_C003.indd 81 10/25/2008 12:35:30 PM10/25/2008 12:35:30 PM

a breaker plate (Figure 3.10). All these supports cause spider lines (streaks) in the extrudate. In foam extrusion, these spider lines are even more pronounced than they are in the extrusion of solid sheets or tubes.24 In the region of the spider lines, foamed sheets exhibit local thin sections and poor foam structure. If CO2 is used as a blowing agent, additionally, a higher risk of premature foaming at the spider legs exists due to the low solubility of CO2.

20

Based on the results of the experiments with different inserts in the fl ow channel of the in-line rheometer die, Heinz conducted experiments to optimize the spider leg geometry in annular dies.20 Since foam sheets are cut at the spider lines, the quality of the resulting foam sheets is almost unaffected. The objective of these experiments was not the minimization of the difference in thickness but more the reduction of the width of the spider lines in order to reduce edge trim waste. The infl uence of melt and die temperature on the spider lines were also examined.

Experiments were conducted using an annular die with an outlet diam-eter of 50 mm that was mounted on a 60 mm single screw foam extruder. The mandrel of the die was supported by one spider leg. At the opposite side of this spider leg, spider dummies of various geometries were installed. Figure 3.11 shows the cross-section of the die with the spider leg (bottom) and an installed spider dummy (top).

The geometries of the spider dummies are depicted in Figures 3.12 and 3.13. Spiders I and II have different angles at the downstream side of 20° and 45° towards the centerline, respectively (Figure 3.12). Upstream angle, length, and width of these two spiders are equal. Spiders III–V have the same angles at the upstream and downstream side (Figure 3.13), while featuring different sizes. Spider IV is shorter and Spider V is thicker than Spider III.

FIGURE 3.10 Different concepts of mandrel supports. [From Heinz, R. Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD Thesis at RWTH

Aachen, 2002.]

Single -spider Dual -spider Breaker plate

Mandrel

Spider leg

Housing

82 Polymeric Foams

61259_C003.indd 8261259_C003.indd 82 10/25/2008 12:35:31 PM10/25/2008 12:35:31 PM

The experiments were conducted with LDPE (Lupolen 1810 H, Basell, Hoofddorp, the Netherlands) and HMS-PP (Profax PF814, Basell). As a nucleating agent, a masterbatch-concentrate (Hydrocerol CF 20 E, Clariant Masterbatch, Lahnstein, Germany) based on a sodium carbonate/citric acid system was added at a concentration of 1.5%. When LDPE was used,

FIGURE 3.12 Spider geometries with different downstream angles. [From Heinz, R.

Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD


Spider I

Spider II

40°

40°

12,5

70

40°

90°

12,5

70

Melt flowdirection

Melt flowdirection

FIGURE 3.11 Mandrel with spider leg and spider dummy. [From Heinz, R. Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD Thesis at RWTH

Aachen, 2002.]


61259_C003.indd 8361259_C003.indd 83 10/25/2008 12:35:31 PM10/25/2008 12:35:31 PM

a cell stabilizer (Activex CT 325, Clariant Masterbatch) and an ionomer (Surlyn 9910, DuPont de Nemours International, Geneva, Switzerland) were added.

The thickness of the foam sheets was measured at several points using a hall-type thickness gauge (Magnamike 8000, Panametrics, Hofheim, Germany) with a 3/16” ball. The data were plotted against the position defi ned by the angle. An angle of 0° corresponds to the position of the cutter; that is, the edge of the slit foam sheet. The whole die was rotated by 45° so that the spider leg of the die was at 315° and the spider dummy at 135°. As illustrated in Figure 3.14, both the spider leg and the spider dummy generated pronounced local thin sections.

3.4.1 Influence of Melt Temperature

At fi rst, the infl uence of the melt temperature on the thickness profi le of the LDPE foam sheets was examined. Therefore, the melt temperature was varied between 110°C and 115°C whereas the die temperature was kept constant at 105°C. Figure 3.15 shows the thickness profi le of the foam sheets at different melt temperatures in the region of the spider dummy for Spider IV exemplarily. The thickness of the sheets increases with decreasing melt temperature over the entire circumference. Simultaneously, the foam density decreases from 212 kg/m3 at 115°C to 166 kg/m3 at 111°C melt temperature. This might be caused by the lower

Spider IV

Spider III

Spider V

40°

40°

12,5

70

40°

40°

12,5

56

40°

40°

70

20

Melt flow

direction

Melt flow

direction

Melt flow

direction

FIGURE 3.13 Spider geometries with different length and width. [From Heinz, R.



84 Polymeric Foams

61259_C003.indd 8461259_C003.indd 84 10/25/2008 12:35:31 PM10/25/2008 12:35:31 PM

diffusivity of the blowing agent at lower temperatures. Thus, the loss of blowing agent through the surface of the foam sheets is reduced and more blowing agent is available for density reduction.

The thickness of the spider line at the position of the spider dummy increases, too. In contrast, the thickness of the sheets at the position of the spider leg does not change with the melt temperature. This indicates that

FIGURE 3.14 Thickness profi le of the foam sheet. [From Heinz, R. Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD Thesis at RWTH

Aachen, 2002.]

0 45 90 135 180 225 270 315 3600.20.4

0.60.8

1.01.2

1.41.6

1.82.0

2.22.4

Spider

Spider dummy

LDPE PP

Circumference of the foam tube (°)

Thi

ckne

ss (

mm

)

45 90 135 180 2250.0

0.5

1.0

1.5

2.0

2.5

Spider IVLDPEDie temperature 105°C

Melt temperature 111°C 113°C 115°C


Thi

ckne

ss (

mm

)

FIGURE 3.15 Thickness profi le for different melt temperatures (LDPE). [From Heinz, R.




61259_C003.indd 8561259_C003.indd 85 10/25/2008 12:35:32 PM10/25/2008 12:35:32 PM

the spider dummy is thermally capsuled from the die. Thus, the tempera-ture of the spider dummy adapts to the melt temperature.

The thickness profi le of the PP foam sheets at various temperatures between 175°C and 180°C is illustrated in Figure 3.16. The thickness of the sheets increases with decreasing melt temperature. In contrast to the LDPE sheets, the thickness of the spider line after the spider dummy does not increase with falling melt temperature. At 175°C melt temperature, a noticeable spider line is observable independent of the spider geometry. This could be due to the low viscosity of the PP melt at 175°C, reducing the intermixing of the melt streams behind the spider. The spider lines of the spider leg are not infl uenced by the melt temperature for PP.

3.4.2 Influence of Die Temperature

In order to examine the infl uence of the die temperature on the thickness distribution of LDPE foam sheets, the temperatures of die housing, mandrel and die lips were varied from 105°C to 135°C whereas the melt tempera-ture was kept constant at 115°C. Figure 3.17 shows a distinct infl uence of the die temperature on the spider lines. At a die temperature of 105°C, the thickness profi le is only slightly affected by the spider leg, whereas a pro-nounced thin section emerges when the die temperature exceeds 120°C. The infl uence of the die temperature on the spider line of the spider dummy is rather low. The thickness distribution of the sheets at 105°C and 120°C die temperature is nearly the same. At a die temperature of 135°C, the thickness of the foam sheet is slightly lower. This is probably caused by an increase of the melt temperature due to the higher die temperature.

FIGURE 3.16 Thickness profi le for different melt temperatures (PP). [From Heinz, R.



90 135 1800.8

1.0

1.2

1.4

1.6

1.8

2.0

2.2

Spider IVPPDie temperature 170°C



Thi

ckne

ss (

mm

)

86 Polymeric Foams

61259_C003.indd 8661259_C003.indd 86 10/25/2008 12:35:32 PM10/25/2008 12:35:32 PM

Because of the narrow processing window, the die temperature can only be varied in a range of 165°C to 175°C in the experiments with PP. As Figure 3.18 depicts, the spider lines in the PP foam sheets are not affected signifi cantly by the die temperature. Compared to the melt temperature of 175°C, the considered temperature range is probably too small to have a signifi cant effect on the thickness distribution of the foam sheets.

FIGURE 3.17 Infl uence of temperature settings on the thickness profi le (LDPE). [From


125 130 135 140 1450.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

300 305 310 315 3200.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4Die temperature

105°C 120°C 135°C

Melt temperature 115°C

Spider legLDPE


Thi

ckne

ss (

mm

)

Spider IVLDPE

Thi

ckne

ss (

mm

)


FIGURE 3.18 Infl uence of temperature settings on the thickness profi le at the spider (PP).

[From Heinz, R. Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD Thesis at RWTH Aachen, 2002.]

305 310 315 320 325305 310 315 320 3250.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0


Die temperature170°C

Circumference of foam tube (°)

Thi

ckne

ss (

mm

)

Spider legPP

Die temperature 165°C 170°C 175°C

Melt temperature175°C

Thi

ckne

ss (

mm

)

Circumference of foam tube (°)


61259_C003.indd 8761259_C003.indd 87 10/25/2008 12:35:32 PM10/25/2008 12:35:32 PM

3.4.3 Influence of Spider Geometry

In the fi rst experiments on the spider geometry, the infl uence of the downstream angle of the spiders was examined. As the experiments with the inserts in the fl ow channel of the in-line rheometer die have shown, a too steep downstream angle leads to a premature foaming in the die. Figure 3.19 shows the thickness profi le of foam sheets of LDPE and PP produced with the Spiders I and II (cp. Figure 3.12). Neither the thickness distribution of the sheets nor the form of the spider lines exhibit signifi -cant differences for both spider geometries. This is true for both materials and all considered melt temperatures.

The downstream angle of the spider seems to have no effect on the quality of the produced foam sheets at the examined conditions. However, the risk of premature foaming at the relatively low melt temperatures and blowing agent contents of 0.5–1.5% is rather low.

In further experiments, the length and the width of the spiders were varied. In Figure 3.20, the thickness distribution of LDPE sheets at a melt temperature of 111°C is plotted for spider geometries III–V. The wider spider geometry (Spider V) leads to a wider thin section in the foam sheet than the standard geometry (Spider III). However, the thickness of the foam sheet at the spider line is higher when the wider spider geometry is used (Figure 3.20, right side). The shorter spider geometry (Spider IV) produces foam sheets with a smaller spider line that is as thin as that of Spider III. The other melt temperatures show similar results.

FIGURE 3.19 Infl uence of the downstream angle on the thickness profi le. [From Heinz, R.



0 45 90 135 180 225 270 315 360

0.8

1.0

1.2

1.4

1.6

1.8

2.0

2.2

2.4

0 45 90 135 180 225 270 315 360

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

Geometry I Geometry II

LDPE


Thi

ckne

ss (

mm

)

PP

Thi

ckne

ss (

mm

)


88 Polymeric Foams

61259_C003.indd 8861259_C003.indd 88 10/25/2008 12:35:33 PM10/25/2008 12:35:33 PM

The experiments with PP were conducted at a melt temperature of 179°C. The results are depicted in Figure 3.21. In total, the thickness distribution of the PP foam sheets is only slightly affected by the spider geometry. Again, the melt temperature shows no signifi cant effect on the thickness distribution.

FIGURE 3.20 Infl uence of the spider geometry on the thickness profi le (LDPE). [From


125 130 135 1400.4

0.6

0.8

1.0

1.2

1.4

1.6

45 90 135 180 2250.20.40.60.81.01.21.41.61.82.02.22.42.6

Geometry III Geometry IV (short) Geometry V (wide)


Thi

ckne

ss (

mm

)

LDPEMelt temperature 111°C Die temperature 105°C

Thi

ckne

ss (

mm

)


FIGURE 3.21 Infl uence of the spider geometry on the thickness profi le (PP). [From Heinz, R.



120 125 130 135 1400.8

1.0

1.2

1.4

1.6

1.8

45 90 135 180 225

1.0

1.2

1.4

1.6

1.8

2.0

2.2

2.4 Geometry III Geometry IV (short) Geometry V (wide)


Thi

ckne

ss (

mm

)

PPMelt temperature 179°CDie temperature 170°C

Thi

ckne

ss (

mm

)



61259_C003.indd 8961259_C003.indd 89 10/25/2008 12:35:33 PM10/25/2008 12:35:33 PM

3.5 Influence of the Pressure Profile in the Extrusion

Die on the Foam Structure

According to the classical nucleation theory (Equations 3.7 and 3.8), the heterogeneous nucleation rate Nhet is dependent on the difference between the solubility pressure pS of the blowing agent and the surrounding pres-sure p0.

25 Since the pressure is not released abruptly in the foam extrusion process, some cells are nucleated before the melt has exited the die; that is, before the pressure of the polymer/blowing agent mixture has dropped down to ambient pressure. This means that some cells begin to grow whereas others have not been nucleated yet. Consequently, nucleation, bubble growth, and thus the resulting foam structure depend on the rate of pressure release.

Nhet � f � c � e -DGhet/k � T (3.7)

DGhet � 16 � p � g 3

____________ 3 � (pS � p0)

2 � g(q) (3.8)

Park et al. validated this assumption in microcellular foaming experi-ments using different nozzles with dissimilar length and diameter.26 In their tests with high impact polystyrene (HIPS) charged with 10% of CO2, they found that the cell density of the foams increases signifi cantly with increasing pressure gradient in the nozzle.

Heinz conducted experiments to study the infl uence of pressure gradient at the die exit on the foam structure of LDPE foam sheets.20,27,28 Therefore, an annular die with exchangeable die lips is used. Three pairs of die lips with different die gaps and lengths are constructed in order to vary the pressure gradient. The total pressure loss of the die is equal for every pair of die lips, so that all extruding conditions can be kept constant during the tests. Figure 3.22 shows the die exit with the different lip geometries schematically.

The shape of the lips is derived using computer software for fl ow calcu-lations in axially symmetric fl ow channels developed at the Institute of Plastics Processing (IKV, Aachen, Germany). The calculations are based on viscosity data of LDPE charged with 1% CO2 at a temperature of 130°C. The results of these calculations in Figure 3.23 show the pressure course over the fl ow length in the die for three different pairs of die lips.

Upstream of the die lips, the pressure loss is rather low in order to prevent a premature foaming in the die. The pressure is then released very quickly to initiate cell nucleation. The different pressure gradients are represented by the different slopes of the plots at the end of the die (270–290 mm).

Due to the simplifi cations and assumptions in the simulation, the pressure loss encountered in the die during preliminary foam extrusion

90 Polymeric Foams

61259_C003.indd 9061259_C003.indd 90 10/25/2008 12:35:34 PM10/25/2008 12:35:34 PM

tests differs from the calculated values. Therefore the die lips are reworked by adjusting the length and the die gap in order to obtain the same total pressure loss in the die with all three pairs of lips. The geometries of the reworked die lips are listed in Table 3.3.20

The pressure gradients of the reworked die lips are calculated via the pressure loss divided by the residence time in the die lip section. As shown in Table 3.3, they differ by almost two orders of magnitude. Higher- pressure gradients at the die exit would lead to corrugation of the foam

FIGURE 3.22 Die lip geometries. [From Heinz, R. Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD Thesis at RWTH Aachen, 2002.]

FIGURE 3.23 Simulated pressure course for different die lips. [From Heinz, R.



160 180 200 220 240 260 280 300

0

10

20

30

40

50

Die lips no. 1 Die lips no. 2 Die lips no. 3

18 kg/h LDPE1% CO2

ϑM = 130°C

Flow length (mm)

Pre

ssur

e (b

ar)


61259_C003.indd 9161259_C003.indd 91 10/25/2008 12:35:34 PM10/25/2008 12:35:34 PM

sheets that could not be handled. The residence time of polymer in the die lips is between 39 and 847 ms.

The experiments with different pressure gradients are conducted with Lupolen 1810 H (Basell, Hoofddorp, the Netherlands) with 0.8% CO2 at a melt temperature of 116°C. The mass throughput rate is 20 kg/m3. As a nucleating agent, either a talc masterbatch concentrate (Hydrocerol CT 316, Clariant Masterbatch, Lahnstein, Germany) or a chemical blowing agent masterbatch concentrate (Hydrocerol CF 20 E, Clariant Masterbatch) are added in different concentrations.

The cell structure of the foams is analyzed using micrographs of the foamed sheets. The mean cell diameter and the cell density; that is, the number of cells per volume, are measured to evaluate the foam quality.

Figure 3.24 shows the cell density and the cell diameter in dependence on the nucleating agent content (talc) for three different pressure gradi-ents. For example, at a constant nucleating agent content of 0.6%, an increase of the pressure gradient from 5 to 24 MPa/s leads to an increase in cell density from 3.4 � 103 to 1.5 � 104 cells/cm3. Simultaneously, the mean cell diameter decreases from 1.15 mm to 0.67 mm. When the pres-sure gradient is further increased up to 103 MPa/s, the cell density rises up to 2.8 � 104 cells/cm3 whereas the mean cell diameter decreases to 0.59 mm. Within the considered range of nucleating agent content, this effect can be observed in all tests.

In total, the experiments show three main effects20,28:

The infl uence of the pressure gradient on the foam structure • decreases with increasing pressure gradient. Although the difference between 103 and 24 MPa/s is higher, the increase in cell density is higher when the pressure gradient rises from 5 to 24 MPa/s.

The infl uence of talc content on the foam structure decreases with • increasing pressure gradient. This becomes obvious when the dependence of cell diameter on the nucleating agent content is compared between 5 and 103 MPa/s.

TABLE 3.3

Dimensions of the Reworked Die Lips

Die Lips

Length

(mm)

Die Gap

(mm)

Pressure Loss

(Bar)

Residence

Time (s)

Pressure

Gradient (MPa/s)

1 1.9 0.4 40 0.039 103

2 3.4 0.6 40 0.170 24

3 5.1 1.1 40 0.847 5

Source: Heinz, R. Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD Thesis at RWTH Aachen, 2002.

92 Polymeric Foams

61259_C003.indd 9261259_C003.indd 92 10/25/2008 12:35:34 PM10/25/2008 12:35:34 PM

At high talc contents, the cell density and the cell diameter seem • to reach a limit asymptotically, which is not crossed even if the nucleating agent content is further increased.

The micrographs of foamed sheets produced with the different die lips in Figure 3.25 illustrate the effect of pressure gradient on foam structure. At constant process conditions and a nucleating agent content of 1.8% talc, only the pressure gradient at the die exit was varied in the range from 5 to 103 MPa/s.

In contrast to this, when Hydrocerol CF 20 E is used as nucleating agent, the cell densities of the foam sheets increase by one order of magnitude while the cell diameters decrease tremendously below 250 μm. However, the infl uence of pressure gradient on the foam structure is weaker when Hydrocerol is used instead of talc. Figure 3.26 presents the cell density and the cell diameter of the foam sheets produced with Hydrocerol CF 20 E at different pressure gradients. As observed with talc, the cell density increases with the pressure gradient at constant nucleating agent content, but the courses of 24 MPa/s and 103 MPa/s do not differ very much. Also the cell diameter decreases when the pressure gradient is raised from 5 to 24 MPa/s. A further increase in pressure gradient to 103 MPa/s does not show any effect. However, at high nucleating agent contents of 0.6% Hydrocerol, the cell diameter of the foams produced with 103 MPa/s at the die exit increases again. Heinz explains this phenomenon by collaps-ing cells.20 Figure 3.27 shows micrographs of foams produced with 0.3%

FIGURE 3.24 Infl uence of the pressure gradient on cell density and cell size (talc). [From


0.4

0.6

0.8

1.0

1.2

0.6 0.8 1.0 1.2 1.4 1.6 1.8 0.6 0.8 1.0 1.2 1.4 1.6 1.8103

104

105

106

107

Δp/Δt = 103 MPa/sΔp/Δt = 24 MPa/sΔp/Δt = 5 MPa/s

Nucleating agent content (%)

Cel

l den

sity

(1/

cm3 )

Ave

rage

cel

l dia

met

er (

mm

)



61259_C003.indd 9361259_C003.indd 93 10/25/2008 12:35:34 PM10/25/2008 12:35:34 PM

Hydrocerol at various pressure gradients. According to the diagrams, the difference in cell size and cell density at various pressure gradients is lower than with talc.

The pressure gradient at the die exit also infl uences the foam density. Figure 3.28 shows the density of the foams produced with talc in depen-dence on the nucleating agent content for three different pressure gradients. All three plots of foam density exhibit a pronounced minimum. The higher the pressure gradient, the lower the nucleating agent content at which the density becomes a minimum.

FIGURE 3.25 Infl uence of the pressure gradient on the foam structure (1.8% talc). [From


FIGURE 3.26 Infl uence of the pressure gradient on cell density and cell size (Hydrocerol).


0.10

0.15

0.20

0.25

0.30

0.40.2 0.60.40.2 0.6105

106

107



Cel

l den

sity

(1/

cm3 )

Ave

rage

cel

l dia

met

er (

mm

)


94 Polymeric Foams

61259_C003.indd 9461259_C003.indd 94 10/25/2008 12:35:35 PM10/25/2008 12:35:35 PM

Similar results can be observed when little amounts of chemical blowing agents (Hydrocerol) are used as a nucleator. Figure 3.29 shows the course of density over nucleating agent content for the three pressure gradients. Again, a minimum can be detected in the plots but it is reached at lower nucleating agent contents. Furthermore, the lowest densities are achieved in a small range of nucleating agent content; that is, the shift of minimal density in dependence on pressure gradient is less pronounced than it is with talc. The rise in density with increasing nucleating agent content is very high when the chemical blowing agent is used as a nucleating agent. This is caused by emerging cell collapse at high nucleating agent contents.

FIGURE 3.27 Infl uence of the pressure gradient on the foam structure (0.3% Hydrocerol).


1.00.5 1.5 2.00.20

0.22

0.24

0.26

0.28

0.30

0.32

0.34



Den

sity

(g/

cm3 )

FIGURE 3.28 Infl uence of the pressure gradient on foam density (talc). [From Heinz, R.




61259_C003.indd 9561259_C003.indd 95 10/25/2008 12:35:36 PM10/25/2008 12:35:36 PM

For both nucleating agents, the content that produces foam with the lowest density is dependent on the pressure gradient. If a low foam density is desired, the most effi cient nucleating agent content can be defi ned as that at which the lowest density is reached. Figure 3.30 shows this most effi cient content in dependence on the pressure gradient for the consid-ered nucleating agents.

FIGURE 3.29 Infl uence of the pressure gradient on foam density (Hydrocerol). [From

Heinz, R. Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD. Thesis at RWTH Aachen, 2002.]

0.1 0.2 0.3 0.4 0.5 0.6 0.7

0.20

0.25

0.30

0.35



Den

sity

(g/

cm3 )

FIGURE 3.30 Infl uence of the pressure gradient on the most effi cient nucleating agent

content. [From Heinz, R. Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD Thesis at RWTH Aachen, 2002.]

0 20 40 60 80 100 120

0.5

1.0

1.5

2.0

Talc Hydrocerol

Pressure gradient (MPa/s)

Nuc

leat

ing

agen

t con

tent

(%

)

96 Polymeric Foams

61259_C003.indd 9661259_C003.indd 96 10/25/2008 12:35:36 PM10/25/2008 12:35:36 PM

3.6 Nomenclature

B (m) Width of fl ow channelC — Concentration of heterogeneous nucleation sitese� — Geometric factor for rectangular slitsF (s�1) Frequency factorG — Geometric factorH (m) Height of fl ow channelK (J/K) Boltzman’s constantL (m) Lengthm (kg/s) Mass fl ow rateNhet [1/(m3 · s)] Heterogeneous cell nucleation ratep0 (Pa) Ambient pressurepS (Pa) Solubility pressureT (K) TemperatureT0 (K) Reference temperatureVm (m3/s) Volume fl ow of melt/blowing agent mixture�Ghet ( J) Change in Gibbs free energy�p (Pa) Change in pressurea (1/K) Expansion coeffi cients (N/m) Interfacial tensiong– (s�1) Representative shear rateh– (Pa · s) Viscosityrm (kg/m3) Density of melt/blowing agent mixturerP (kg/m3) Density of polymer meltrS (kg/m3) Density of solvent (blowing agent)q (°) Wetting anglewS — Weight fraction of solvent (blowing agent)

References

1. Han, C. D. and Villamizar, C. A. “Studies on structural foam processing: I. The rheology of foam extrusion.” Polymer Engineering and Science 18 (1978): 687–698.

2. Han, C. D. and Ma, C. Y. “Rheological properties of mixtures of molten polymer and fl uorocarbon blowing agent: I. Mixtures of low-density polyethylene and fl uorocarbon blowing agent.” Journal of Applied Polymer Science 28 (1983): 831–850.

3. Han, C. D. and Ma, C. Y. “Rheological properties of mixtures of molten polymer and fl uorocarbon blowing agent: II. Mixtures of polystyrene and fl uorocarbon blowing agent.” Journal of Applied Polymer Science 28 (1983): 851–860.


61259_C003.indd 9761259_C003.indd 97 10/25/2008 12:35:37 PM10/25/2008 12:35:37 PM

4. Gendron, R., Daigneault, L. E., Dumoulin, M. M., Dufour, J., and Caron, L.-M. “Rheological measurements of PS/blowing agent mixtures.” In Proceedings of the 10th Annual Meeting, Polymer Processing Society (PPS), Akron, OH, 1994: 415–416.

5. Gendron, R., Daigneault, L. E., and Caron, L.-M. “Rheological behavior of polystyrene/blowing agent mixtures.” In Proceedings of the Annual Technical Conference (ANTEC), Society of Plastics Engineers (SPE), Indianapolis, IN, 1996: 1118–1122.

6. Kropp, D. and Michaeli, W. “Rheological melt behaviour during foam extru-sion using carbon dioxide as blowing agent.” In Proceedings of the Polymer Processing Society European Meeting, Stuttgart, 1995.

7. Michaeli, W., Kropp, D., Pfannschmidt, O., Rogalla, A., and Seibt, S. “Kohlendioxid als Verarbeitungshilfs- und Treibmittel beim Spritzgießen und Extrudieren von Thermoplasten.” Gummi Fasern Kunststoffe (GAK) 49 (1996): 652–661.

8. Chmiel, H. and Schümmer, P. “Eine neue Methode zur Auswertung von Rohrrheometer-Daten.” Chemie–Ingenieur–Technik (CIT) 43 (1971): 1257–1259.

9. Schümmer, P. and Worthoff, R. H. “An elementary method for the evaluation of a fl ow curve.” Chemical Engineering Science 33 (1978): 759–763.

10. Michaeli, W. Extrusion Dies for Plastics and Rubber: Design and Engineering Computations, 2nd revised edition. Carl Hanser Verlag, München, 1992.

11. Wortberg, J. “Werkzeugauslegung für die Ein- und Mehrschichtextrusion.” PhD Thesis, RWTH Aachen, 1978.

12. Gundert, F. and Wolf, B. A. “Polymer-Solvent Interaction Parameters.” In Polymer Handbook, ed. J. Brandrup and E. H. Immergut. Wiley, New York, 1989.

13. Menges, G. “4. korrigierte und aktualisierte.” In Werkstoffkunde Kunststoffe. Aufl age, Carl Hanser Verlag, München, 1998.

14. N. N. VDI-Wärmeatlas. VDI-Verlag, Düsseldorf, 1984. 15. Kropp, D. Extrusion von amorphen thermoplastischen Schäumen niedriger Dichte

mit CO2 als Treibmittel. Report of research project, Aachen, 1998. 16. Kropp, D. Extrusion thermoplastischer Schäume mit alternativen Treibmitteln

(Extrusion of thermoplastic foams using alternative blowing agents). PhD Thesis, RWTH Aachen, 1999.

17. Niggemann, M. Untersuchung des Fließverhaltens bei der Schaumextrusion von Polystyrol mit CO2 als Treibmittel. Unpublished Thesis at RWTH Aachen, 1995.

18. Williams, M. L., Landel, R. F., and Ferry, J. D. “The temperature dependence of relaxation mechanisms in amorphous polymers and other glass-forming liquids.” Journal of the American Chemical Society 77 (1955): 3701–3707.

19. Stadlbauer, M. “Polypropylen Schaum.” In Proceedings of Thermoplastische Schaumstoffe—Verarbeitungstechnik und Möglichkeiten der Prozessanalyse, Institut für Kunststoffverarbeitung an der RWTH Aachen, Aachen, 2004.

20. Heinz, R. Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD Thesis at RWTH Aachen, 2002.

21. Herrmann, T., Analyse der scherinduzierten Blasenbildung bei der thermo-plastischen Schaumextrusion. Thesis at RWTH Aachen, 2000.

98 Polymeric Foams

61259_C003.indd 9861259_C003.indd 98 10/25/2008 12:35:37 PM10/25/2008 12:35:37 PM

22. Lee, S. T. “Shear effects on thermoplastic foam nucleation.” Polymer Engineering and Science 33 (1993): 418–422.

23. Throne, J. L. Thermoplastic Foam Extrusion: An Introduction. Carl Hanser Verlag, München, 2004.

24. Lauterberg, W. “Dickentoleranzen bei Polyethylenschaumfolien.” Plaste und Kautschuk 25 (1978): 294–295.

25. Colton, J. S. and Suh, N. P. “The nucleation of microcellular thermoplastic foam with additives: Part I: theoretical considerations.” Polymer Engineering and Science 27 (1987): 485–492.

26. Park, C. B., Baldwin, D. F., and Suh, N. P. “Effect of the pressure drop rate on cell nucleation in continuous processing of microcellular polymers.” Polymer Engineering and Science 35 (1995): 432–440.

27. Michaeli, W. and Heinz, R. “Extrusion of thermoplastic foams using CO2 as blowing agent.” In Proceedings of the Blowing Agents and Foaming Processes Conference, Rapra Technology Ltd., Heidelberg, 2002.

28. Michaeli, W. and Heinz, R. Analyse des Einfl usses der Werkzeuggeometrie und des Nukleierungsmittels auf die Schaumqualität bei der thermoplastischen Schaumextrusion mit CO2 als Treibmittel. Report of Research Project, Aachen, 2002.


61259_C003.indd 9961259_C003.indd 99 10/25/2008 12:35:37 PM10/25/2008 12:35:37 PM

61259_C003.indd 10061259_C003.indd 100 10/25/2008 12:35:37 PM10/25/2008 12:35:37 PM

4Processes and Process Analysis of Foam Injection Molding with Physical Blowing Agents

Walter Michaeli, Axel Cramer, and Laura Flórez

CONTENTS

4.1 Characteristics of Injection Molded Thermoplastic Foams ....... 1024.1.1 Benefi ts and Drawbacks of the Foam

Injection Molding Process ................................................. 1024.1.2 Chemical and Physical Blowing Agents ......................... 1044.1.3 Properties of the Polymer Matrix ..................................... 1074.1.4 Mechanisms of Nucleation and Bubble Growth ............ 108

4.2 Process Concepts for Foam Injection Molding (FIM) .............. 1144.2.1 Fundamentals of the Foam Injection

Molding Process ................................................................. 1144.2.2 Injection of the Blowing Agent into an Extruder ........... 1144.2.3 Injection of the Blowing Agent into

the Plastifi cation Unit .......................................................... 1154.2.4 Injection of the Blowing Agent into

a Special Gassing Unit ........................................................ 1164.2.5 Processing of Pellets Pre-Charged with

Blowing Agent ..................................................................... 1184.2.6 Injection of the Blowing Agent with Aid

of a Special Nozzle .............................................................. 1184.3 Experimental Analysis of the Processing

Parameters in Foam Injection Molding .................................... 120

61259_C004.indd 10161259_C004.indd 101 10/25/2008 12:36:12 PM10/25/2008 12:36:12 PM

4.3.1 Experiments with the IKV Blowing Agent Injection Nozzle .................................................................. 1214.3.1.1 Infl uence of the Process Parameters

on the Foamed Part Characteristics .................... 1214.3.1.2 Investigations on Cycle Time Reduction ............ 123

4.3.2 Experiments on Pre-Loaded Polycarbonate ................... 1254.3.2.1 Morphological Characterization Methods ........... 1264.3.2.2 Infl uence of the Process Parameters on the

Foamed Part Characteristics ................................ 1284.4 Optimization of the Surface Quality of Foamed

Injection Molded Parts ................................................................ 1314.4.1 Occurrence of Silver Streaks ............................................. 1324.4.2 Possibilities of Increasing the Surface Quality

Through Mold and Process Technology ......................... 1344.4.3 Investigation of “Breathing” Molds and Gas

Counter-Pressure ................................................................ 1374.5 Summary ....................................................................................... 139Acknowledgments .............................................................................. 140References ............................................................................................. 140

4.1 Characteristics of Injection Molded Thermoplastic Foams

4.1.1 Benefi ts and Drawbacks of the Foam Injection Molding Process

Injection-molded foams are counted among the most promising develop-ments to lower material consumption and reduce both cycle time and molding tonnage in the injection molding of plastic parts. Their convenient specifi c material properties, besides their processing and performance advantages, make them the ideal choice for the production of different kinds of structural parts.

These types of materials, also known as structural foams, have a typical “sandwich” morphology, characterized by a foamed core surrounded by a compact skin. In comparison to a compact part, which exhibits an almost constant density profi le through its cross-section, structural foams have a density close to that of the compact material in the surrounding skin, where the largest concentration of material is present, and a lower density in the part’s middle, where the closed-cell foamed core is found. This particular confi guration is responsible for some signifi cant advantages during the processing and the service life of the molded parts.

The most obvious effect of foam injection molding is the density reduction leading both to raw material savings and to a decrease in the part’s weight, offering not only economical but also design benefi ts. By keeping the same weight, a foamed part will exhibit higher specifi c stiffness, specifi cally when

102 Polymeric Foams

61259_C004.indd 10261259_C004.indd 102 10/25/2008 12:36:12 PM10/25/2008 12:36:12 PM

bending forces are applied. This corresponds to a larger geometric moment of inertia in comparison to a compact molded part of the same weight, because of the accumulation of material away from the part’s neutral axis.1

Due to the lower pressure level during injection, machines with lower clamping forces can be used in comparison to conventional injection molding.2 The holding pressure is often not needed, and using unchanged processing temperatures a reduction of the internal mold pressure of up to 50% is possible.3 If the viscosity-reducing effect of the blowing agent is used for the reduction of the processing temperature, a decrease in cool-ing time is also possible. Furthermore, the reduction of internal stresses, the reduction of warpage, and the decrease of sink marks enhance the dimensional accuracy of ribs, openings, or dimensional changes in the part’s cross-section.

However, if a part is foamed without changing its dimensions, its mechanical properties drop as a function of the weight reduction. It has been proven that the reduction in the elastic modulus and the part’s resis-tance can be correlated with the density reduction:

Xf

___ Xc � ( rf

__ rc )

n

(4.1)

where X is either the tensile or fl exural stress or elastic modulus, � is the density, and the subscripts f and c represent respectively the foamed and the compact polymer. The exponent of this correlation, n, is determined experimentally, and commonly takes values between 1.0 and 2.0.4 This correlation makes clear that the properties of the foam drop at a steeper rate than the density reduction. Current studies show that the drop in mechanical properties can be counteracted through an optimization of the foam morphology.

Another disadvantage is the achievable surface quality of foamed parts, which is rather poor in comparison to compact parts.5 By further process developments and the combination of already existing mold concepts (gas counter-pressure, breathable molds, etc.) these drawbacks are subjected to prospective optimizations.

Some of the earliest applications of structural foams were made in the fi elds of appliances and furniture, where foaming was achieved with chemical blowing agents. In these cases the objective was to improve the dimensional stability while reducing the part’s weight. Housings for elec-tric appliances have also traditionally made use of the advantages of foam injection molding, to eliminate the sink marks and the warpage typical of their ribbed geometries. Larger parts, such as front-ends and door panels for automotive applications, are currently being developed, and an inter-esting trend arises in the combination of the foam injection molding with other injection processes, such as multi-component or back molding, to achieve an optimized properties profi le in the fi nal part. A typical appli-cation is presented in Figure 4.1.

Processes and Process Analysis of Foam Injection Molding 103

61259_C004.indd 10361259_C004.indd 103 10/25/2008 12:36:12 PM10/25/2008 12:36:12 PM

The achievable weight reduction depends strongly on the thickness of the part and the type of foaming agent used, but for typical applications in injection molding, with wall thicknesses between 1.0 and 3.0 mm, weight reductions between 10% and 20% are common. For thicker applications weight reductions of up to 60% are possible.

Due to their broad application potential, injection molded thermoplastic foams have gained a well-established position during the last few years. The process of foam injection molding is growing in importance, triggered by the recent developments in blowing agents and processing technologies. The associated reduction in consumption of raw material is gaining rele-vance due to the increase in the price of resin, and in the automotive indus-try, the ever-tightening regulations that head for the reduction of vehicle’s weight drive a struggle for the production of reliable foamed parts.

4.1.2 Chemical and Physical Blowing Agents

The production of thermoplastic foams is achieved with the aid of foam-ing agents, either physical or chemical, that can be dosed into the polymer in different ways. Chemical blowing agents have existed in many foaming applications for a long time. However, the straight development of different process technologies for the production of physically blown foams is responsible for the rising demand on foamed parts. Whereas the develop-ment of marketable techniques for the extrusion of physically blown foams had already begun at the end of the 1970s, research activities in the injection molding area has intensifi ed since the middle of the 1990s.1

The chemical blowing agents are added to the polymer pellets in solid form and are activated through addition of heat, releasing a fl uid, mostly nitrogen, carbon dioxide, or water.6 However, the appearance of residual products is a disadvantage, given the fact that they can represent up to 70% of the fi nal composition of the agent.1,2 Their decomposition can lead to a degradation of the polymer matrix, to a decrease in mechanical properties, to coloration of the part and to corrosion and contamination of the mold. For these reasons only a defi ned amount of foaming agent ought to be incorporated into the polymer melt when using chemical blowing agents. A list of chemical blowing agents is provided in Table 4.1.

FIGURE 4.1 Application produced through foam injection molding.

104 Polymeric Foams

61259_C004.indd 10461259_C004.indd 104 10/25/2008 12:36:12 PM10/25/2008 12:36:12 PM

Foaming agents like carbon dioxide, nitrogen, hydrocarbonates like pentane, and even water,1 are dosed directly into the polymer melt and are referred to as physical foaming agents. Compared with the chemical agents these foaming agents do not evoke residual products, but their dosage is usually more complex from the technological point of view, due to higher demands on the control and the transient fl uid incorporation in the injection molding process. A list of physical blowing agents can be found in Table 4.2.

TABLE 4.1

Commonly Used Chemical Blowing Agents

Chemical

Denomination Abbreviation

Decomposition

Temperature (°C)

Rate of Gas

Yield (ml/g)

Main Gas

Components

Azodicarbonamide ADC 200–220 250–320 N2, CO, CO2

(NH3)

Modifi ed ADC 155–220 150–300 N2, CO, CO2

(NH3)

4,4�-Oxybis (benzene-

sulfonylhydrazide)

OBSH 140–165 120–150 N2, H2O

5-Phenyltetrazole 5-PT 240–250 190–210 N2

p-Toluenesulfonyl-

semicarbazide

TSS 215–235 120–140 N2, CO2

p-Toluenesulfonyl-

hydrazide

TSH 110–140 120–140 N2, H2O

Sodium carbonate Bicab 120–150 120–170 CO2, H2O

Citric acid 200–220 90–120 CO2, H2O

TABLE 4.2

Commonly Used Physical Blowing Agents

Blowing

Agent

Chemical

Formula

Molar

Weight

(g/mol)

Boiling

Temperature

(°C) Flammable ODP GWP

Isobutane C4H10 58.1 �11.7 Yes — —

Cyclopentane C5H10 70.1 49.3 Yes — 0.00275

Isopentane C5H12 72.1 29.0 Yes — —

CFC-11 CFCl3 137.4 23.8 No 1.0 1.0

HCFC-22 CHF2Cl 86.5 �40.8 No 0.05 0.35

HCFC-142b CF2ClCH3 100.5 �9.2 Yes 0.05 0.38

HFC-134a CH2FCF3 102.0 �26.5 No — 0.27

Nitrogen N2 28.0 �195.7 No — —

Carbon

dioxide

CO2 44.0 �56.5 No — 0.00025


61259_C004.indd 10561259_C004.indd 105 10/25/2008 12:36:12 PM10/25/2008 12:36:12 PM

Thanks to the developments achieved in control and valve technology, physical foaming agents, particularly the inert gases, are increasingly important.

The type and amount of foaming agent added to the thermoplastics will determine the achievable foam density as well as the required engineering system for its production. The possible density reductions achievable with each type of foaming agent are illustrated in Figure 4.2, taking as an example the process of fi lm foam extrusion. As we can see, when higher foaming grades are to be reached, foaming should be achieved through physical blowing agents because of the great extent of decomposition prod-ucts associated with the use of large amounts of chemical blowing agents.

One of the factors to be considered concerning the selection of the appro-priate foaming agent in a specifi c foam application is its solubility in the selected thermoplastic material. The cell nucleation is originated by a pressure drop as a consequence of a thermodynamic instability in the mixture polymer and blowing agent. A higher pressure drop rate implies a higher nucleation rate and the formation of a larger number of cells. If the diffusion rate of the blowing agent through the polymer is slow, a larger number of cells is formed due to the longer time needed for the blowing agent to diffuse into existing cells.

Another issue is, of course, the cost of the blowing agent and last but not least its environmental and toxicological impact. Due to their compliance of this particular requirement, nitrogen and carbon dioxide are broaden-ing their acceptance as physical foaming agents.3

On a volumetric basis, carbon dioxide can be up to 18 times more soluble than nitrogen in semi-crystalline polymers, such as LDPE, at 25°C and a pressure of 1 bar. Under the same conditions, in amorphous polymers such as PVC, the solubility of carbon dioxide is 20 times larger than that of nitrogen.5 To guarantee the reproducibility of the dosage of carbon dioxide, it is advisable to bring its condition to a supercritical state (above

FIGURE 4.2 Achievable densities in foam extrusion with chemical and physical blowing

agents.

1000

Chemicallyfoamed

Physicallyfoamed800

600

400

Den

sity

(kg

/m3 )

200

0

106 Polymeric Foams

61259_C004.indd 10661259_C004.indd 106 10/25/2008 12:36:12 PM10/25/2008 12:36:12 PM

critical values of temperature and pressure of 31°C and 73.83 bar respec-tively). Once this value is reached, neither more condensation nor phase changes will take place, independently of the pressure and the tempera-ture induced. The states of nitrogen and carbon dioxide at storage and at injection are shown in Figure 4.3.

The properties of the supercritical fl uids are in between those of the liquid and the gas phase.6,7 In this area these agents exhibit a low viscosity, low surface tension, comparatively good diffusion properties and a similar fl uid molar volume (Table 4.3).

These properties are responsible for an excellent dissolving power as well as for fast loading response. As a consequence, higher nucleation densities are achievable, and therefore a fi ner cell structure of the foam. The processing of plastics with supercritical fl uids is a technology that has been around for years in industry, and used, for example, in gas-assisted injection molding (GAIM).

4.1.3 Properties of the Polymer Matrix

In principle, all thermoplastic materials can be foamed. However, the effort required to foam a thermoplastic material with suffi cient quality

FIGURE 4.3 Phase diagram of carbon dioxide and nitrogen.

Solid Fluid

p/pc

1

1

Gaseous

Triple point

Critical point:

CO2

CO2

CO2

N2

N2

Temp.: 31.04 °C

Temp.: –146.9 °C

Pressure: 73.83 bar

Pressure: 33.98 bar

State at storage

State at injection

J/Jc

N2

Supercriticalstate

TABLE 4.3

Physical Properties of Carbon Dioxide

Parameter Unit Gas Fluid Super Critical

Density g/cm3 10�3 1 0.6

Viscosity g/(cm·s) 10�4 10�2 10�4


61259_C004.indd 10761259_C004.indd 107 10/25/2008 12:36:12 PM10/25/2008 12:36:12 PM

specifi cations varies widely depending on its type. Independent of the process, a material will require less effort if it exhibits a broad processing window. This processing window is strongly subjected to the dependence of the viscosity on the melt temperature. For the formation of cells a low viscosity is advantageous, but if the viscosity is too low after cell growth, the foam collapses. Due to their low viscosity, semi-crystalline polymers should be processed near to their crystallization temperature to stabilize the foam structure before collapsing. With branched polyolefi ns such as LDPE, because they exhibit strain hardening in their elongational viscosity, this feature enables the stabilization over a broader range of temperature before crystallization takes place.

A larger processing window will be given if the viscosity increases slowly with declining temperature. For this reason amorphous polymers that are processed at temperatures closer to their Tg have a larger process-ing window. Also, the plasticization induced by the physical blowing agent has an impact on the processing of amorphous polymers.

The maximum viscosity at which a polymer may be foamed is deter-mined by the maximum admissible viscosity in the process and by the processing technology. Above this viscosity it may occur, for example, that the fl ow resistance in the mold is too high to be overcome. On the other hand, the type and the amount of the foaming agent will establish the lower limit for the viscosity. If the viscosity of the melt is too low and the pressure in the forming cells too high, a collapse of the growing cells may result.

In comparison to semi-crystalline thermoplastics, the amorphous ones are easier to process. Their processing window defi ned by the melt temperature is broader. As consequence of a slowly and homogeneously decreasing melt viscosity a more stable foam structure can be achieved. Whereas the limits for processing will be set through the fl ow behavior and therefore through the shear viscosity of the melt, the limit for cell growing will essentially be determined by the elongational viscosity.8 Figure 4.4 illustrates the dependence of the shear viscosity of semi- crystalline and amorphous thermoplastics as a function of temperature.9

4.1.4 Mechanisms of Nucleation and Bubble Growth

A homogeneous foam structure can be reached through the charging of the polymer with a high amount of foaming agent, thereby achieving high levels of sorption concentration. The charging phase is characterized by mechanisms of sorption and diffusion. While the sorption capacity of a polymer describes its ability to achieve a maximum degree of foaming agent concentration, the diffusion capacity determines the velocity at which the fl uid transport will occur. These mechanisms connect the formation of the nucleation sites and the subsequent stage of bubble growing. These

108 Polymeric Foams

61259_C004.indd 10861259_C004.indd 108 10/25/2008 12:36:13 PM10/25/2008 12:36:13 PM

separated mechanisms have consequences for process control. However, fi rst of all a model developed by Pfannschmidt8 will be presented, demon-strating each mechanism separately.

To understand the behavior of a polymer/foaming agent system the dependence of each phase on the temperature will be outlined fi rst. Figure 4.5 shows the behavior of a polymer under variations of pressure and temperature. For example, if the temperature is increased in the system, the velocity of the molecular movement and the kinetic energy will be increased as well. As a consequence of this energy increase, the number and frequency of collisions between molecules will rise, resulting in an increase of the residence volume of the molecules. On the other hand, the volume of the system will decline due to its compressibility if the pressure is increased in the fl uid phase.

FIGURE 4.4 Viscosity as a function of temperature for amorphous and semi-crystalline

thermoplastics.

AmorphousS

hear

vis

cosi

ty h

Semi-crystalline

Processing limit

DJa: Processing range of amorphous thermoplastics

DJa DJt

DJt: Processing range of semi-crystalline thermoplastics

Temperature J

Limit by cell collapse

FIGURE 4.5 Impact of temperature and pressure on the blowing agent.

Fluid-molecule

T0, p0, V0 T1 > T0, p1 = p0, V1 > V0 T2 = T1, p2 > p0, V2 > V1

Residence volume

0

1

2


61259_C004.indd 10961259_C004.indd 109 10/25/2008 12:36:13 PM10/25/2008 12:36:13 PM

Figure 4.6 illustrates the behavior of a polymer under variations of temperature. In comparison to a gas, a polymer will exhibit a quasi-incompressible behavior, particularly within the range of low temperatures.

The separated macromolecules of the considered system are not fi xed to a position, but oscillate due to the molecular movements inside a deter-mined volume, that will be referred to as “residence volume.” When the temperature is increased the macromolecules will oscillate on a larger scale, increasing the residence volume.

Once the process of absorption takes place, the molecules of the blowing agent surpass the surface of the polymer and fi ll the spaces in between the molecular chains of the polymer matrix. The incorporation of blowing agent molecules depends on the pressure and temperature. If the pressure of the system is increased, a higher concentration of the blowing agent will be dissolved. Figure 4.7 shows schematically the infl uence of pressure on the number of incorporated blowing agent molecules.

A mathematical formula to describe the correlation between the concen-tration of the dissolved material in the system, denoted below as c, and the partial pressure of the blowing agent phase, p, is given through Henry’s equation, which proposes a linear dependence:

c � S � ρ (4.2)

The S factor is referred as solubility, and is a substance dependent value.The blowing agent molecules located in between the polymer chains

reduce the intermolecular forces. Thus the distance between the polymer chains can be increased independently of temperature, which in turn leads to the embedding of more fl uid molecules. The separation of molecules will lead to an increased plastifi cation effect, as well as to an enlargement of the volume of the polymer.

FIGURE 4.6 Impact of temperature on the polymeric phase.

T0, p0, V0

01

Polymer chain Residence volume

T1 > T0, p1 = p0, V1 > V0

110 Polymeric Foams

61259_C004.indd 11061259_C004.indd 110 10/25/2008 12:36:13 PM10/25/2008 12:36:13 PM

A mathematical formula that describes the solubility of polymers above the glass or melt temperature is the Flory–Huggins theory. In this theory the real system is replaced by a model where the relative amount of polymer and blowing agent is represented with a proportional number of elements. The model is able to represent the exchange between all the elements present in the system, and can thus deliver some thermodynamic param-eters relevant for the sorption process, such as the free energy or the chemical potential of the system. For a deeper insight into this theory the reader is referred to References 8 and 10.

The infl uence of temperature on the solubility of the system polymer/blowing agent can be seen in Figure 4.8, for the case where the fl uid mol-ecules have a small diameter.

FIGURE 4.7 Infl uence of pressure on the number of incorporated fl uid molecules.

T0, p0, V0 T1 = T0, p1 > p0, V1 ≈ V0

0 1

FIGURE 4.8 Infl uence of temperature on the number of incorporated fl uid molecules.

T0, p0, V0

01

T1 > T0, p1 = p0, V1 > V0


61259_C004.indd 11161259_C004.indd 111 10/25/2008 12:36:14 PM10/25/2008 12:36:14 PM

If the blowing agent is absorbed, it will only fi t in the free spaces between the polymer chains. The residence volume of each polymer chain will be increased by a temperature rise, thus leaving less space for the blowing agent molecules. Therefore, it can be expected that the solubility of the blowing agent in a polymer will be reduced.

While the term “sorption” denotes the solubility of a fl uid in a polymer, the term “diffusion” describes the transport of material driven by concen-tration gradients. The mathematical theory related to diffusion in isotro-pic media says that the amount of transported material is proportional to the concentration gradient (Fick’s fi rst law):11,12

F � –D � grad c (4.3)

The fl ow density F describes the number of particles that cross a deter-mined sectional area per unit of time. The concentration of the diffusing material is denoted by c, and D is the diffusion coeffi cient. Attending to the equation of continuity, a differential equation for the diffusion phe-nomena can be stated as follows (Fick’s second law):11,12

∂c __ ∂t � div(D � grad c) (4.4)

The diffusion of a low molecular fl uid in a polymer is based on the move-ment capability of the molecules (Brownian molecular movement) and is therefore dependent on the temperature. To achieve a homogeneous distribution of the concentration, the fl uid molecules are induced to penetrate the polymer chain bonds. As the temperature arises, the move-ment freedom of the macromolecular chains is also increased as well as that of the fl uid molecules. The increasing movement of the small fl uid molecules facilitates a faster crossing of the entangled macromolecules. The stronger oscillation of the polymer chains leads to a further separa-tion between the molecules, which in turn results in an increase in residence volume. When the concentration increases, so will also the velocity of diffusion.

The foaming is initiated through the appearance of nucleation sites. If these nucleation sites are stable, they will lead to bubble growth. The apparition of such sites is dependent on the surface tension,13–15 the satura-tion pressure, and the pressure of the fl uid in the polymer phase. According to the traditional nucleation theory, there are three possible mechanisms for the occurrence of this phenomenon:

Homogeneous nucleation•

Heterogeneous nucleation•

Combination of these two mechanisms.•

With homogeneous nucleation the nucleation sites develop as a conse-quence of density fl uctuations and intermolecular movements. By contrast,

112 Polymeric Foams

61259_C004.indd 11261259_C004.indd 112 10/25/2008 12:36:15 PM10/25/2008 12:36:15 PM

with heterogeneous nucleation the nucleation sites appear in the inter-faces of a given phase (which may be impurities in the melt, nucleation agents, etc.). For heterogeneous nucleation a lower super-saturation grade will be required to initiate the bubble growth in comparison with homo-geneous nucleation.16–19

The super-saturation of the polymer, and therefore the nucleation, can either appear through a sudden drop of pressure or through an increase in temperature. In processes of industrial application, like injection mold-ing or extrusion, the foam expansion is initiated following a sudden pres-sure drop. While in injection molding this effect occurs when the gas-loaded melt reaches the mold, in foam extrusion the pressure is maintained through control systems or gear pumps, for example, to secure a certain pressure drop in the extrusion die. The velocity of the pressure drop can be controlled through an appropriate geometry.

To understand the infl uence of the pressure drop velocity, the concurrent mechanisms of cell nucleation and cell growth have to be taken into account. During the pressure drop and the resulting instable thermodynamic state, stable nucleation sites will be created initiating the fi rst growing cells. The dissolved blowing agent diffuses towards these cells to diminish the free energy of the system. Around the nucleated cells, regions of smaller blow-ing agent concentration remain, in which the appearance of nucleation sites is less probable. A further decrease in saturation pressure leads to an expansion of the existing cells and, eventually, a later cell nucleation. The appearance of a later cell nucleation depends on the low concentration regions of the blowing agents. If these regions overlap, there will be no further bubble development because the blowing agent will prefer to fl ow into the already formed cells.

The infl uence of pressure drop velocity can be explained by Figure 4.9. Here, pressure versus time is presented for two systems. Due to the higher pressure gradient the nucleation velocity of system A is higher than that of system B. At the time t1 the pressure drop is larger for the system A than the pressure drop for the system B. At t2 the nucleation points grow. This leads to a diminishing of the blowing agent concentration available for further bubble growth. In the regions in which the blowing agent is not used, the velocity of the nucleation site’s formation grows exponentially for both systems, but always faster for system A, due to the high pressure difference. As the blowing agent is used for system A for a shorter time, the cells will coincide with each other sooner. Due to the fact that the cells in system B have more time to grow, the zones of low blowing agent con-centration have more time to increase the size of the already formed cells in such a way that the concentration of blowing agent serves the size increase of the existing cells instead of the formation of new ones. Following this observation it can be stated that an increase in pressure drop velocity leads to the appearance of a larger number of nucleated cells. On the other hand, for slower pressure drops the diameter but not the number of fi rst-formed cells will increase.


61259_C004.indd 11361259_C004.indd 113 10/25/2008 12:36:15 PM10/25/2008 12:36:15 PM

4.2 Process Concepts for Foam Injection Molding (FIM)

4.2.1 Fundamentals of the Foam Injection Molding Process

In a similar way to the conventional injection molding process, in the production of injection molded thermoplastic foams a melt charged with blowing agent is processed by means of a plastifi cation unit and after-wards injected at high velocity into a mold. Triggered by the abrupt drop of pressure during fl ow, the cells start to grow through expansion of the foaming agent. The fi nal foamed part is composed of a solid outer skin surrounding a core of closed-cellular foamed material.

Whereas no special modifi cations have to be made to the injection molding machine when chemical blowing agents are used (except the use of a shut-off nozzle and ideally a position control of the screw), physical blowing agents require a specially engineered system.

Different process concepts for the incorporation of physical blowing agents compete with one another in foam injection molding. The differ-ences in these concepts involve mainly the way the blowing agent is incor-porated into the melt and, to a lesser degree, the processing itself. These concepts are presented in the following chapter.

4.2.2 Injection of the Blowing Agent into an Extruder

One of the oldest processes for the injection molding of physically blown foams was developed in the 1970s in North America. Using this technology it was possible to transfer the manufacturing of foamed parts from the extrusion to injection molding technology for the fi rst time. In this case,

FIGURE 4.9 Infl uence of pressure gradient on the formation of nucleation sites.

Pre

ssur

e p

Time tt1 t2 t1 t2

Polymer/fluid-solutionPressuregradient

A B

B

B

A

A

Fluid bubble>

114 Polymeric Foams

61259_C004.indd 11461259_C004.indd 114 10/25/2008 12:36:15 PM10/25/2008 12:36:15 PM

the plastifi cation of the polymer and the injection of the melt into the cavity are separated from each other. The blowing agent is injected into a continuously working extruder fi rst of all. The polymer/blowing agent mixture is then homogenized and subsequently delivered to the space in front of the injecting piston, as shown in Figure 4.10. Afterwards, the dosed melt is injected with high speed into the cavity.4 Due to the approxi-mately constant counter-pressure in the extrusion unit the blowing agent can be incorporated by pressure control. The leakage remains small even at high injection velocities to ensure high precision regarding the mass of the part because of the application of a piston injection unit. A decisive disadvantage of this concept is the fact that a special machine is necessary for the foam injection molding. The engineering of the process is advanta-geous, on the other hand, because the system design is relatively simple. In any case, the fi rst physically blown foams could be produced by this concept. Their quality and the reproducibility were not very high, because the technology was not yet fully developed. Nevertheless, it worked satis-factorily over a long period and represented the counterpart to chemically blown foams, which were used predominantly in Europe.

4.2.3 Injection of the Blowing Agent into the Plastification Unit

In the middle of the 1990s a new process was developed in the US, which incorporated the blowing agent into the melt over one or several injectors in the second half of the plastifi cation cylinder. The screw equipped with mixing and shear elements provided the homogenization of the polymer/blowing agent mixture. This process is relatively complex, although it is a fl exibly applicable system technology. Apart from the installation of one or several injectors in the plastifi cation cylinder and a suitable, precisely working, gas dosing station, a special screw (25–28D) with mixing and

Pressure release valve

Injection piston

Injection unit

Shut-off valve

Nitrogen inlet

Extruder

FIGURE 4.10 Injection of blowing agent in an extruder.


61259_C004.indd 11561259_C004.indd 115 10/25/2008 12:36:15 PM10/25/2008 12:36:15 PM

shear elements is essential. This special screw requires the extension of the plastifi cation unit. Another diffi culty is the varying effective mixing length of the screw, as the screw moves relative to the injector during plas-tifi cation. To compensate for this effect, several injectors can be piloted in a cascade, which requires an accordingly sophisticated control technique. Furthermore, the position of the gassing port determines the minimum dosing volume. Therefore, the application scope of this special plastifi ca-tion unit is limited compared with conventional aggregates.

This technology is closely linked with the term of the MuCell® technology20 of the company Trexel (Woburn, US). The concept is shown in Figure 4.11. After introduction onto the market, this process variant drew large atten-tion. Because of the advantages related to a reduction in the molding pres-sure and in the cycle time, as well as the diminution of internal stresses and the increase in dimensional accuracy of the molded part, the foam injection molding process achieved a lot of interest again. Foam injection molding was rediscovered by the rising interest in the market.

4.2.4 Injection of the Blowing Agent into a Special Gassing Unit

Another new concept for the dosing of physical blowing agents into the melt was fi rstly presented on the K-2001 show under the trade name ErgoCell® by the company DEMAG Ergotech GmbH, Schwaig, Germany.21 For the loading of the melt an additional equipment component is inserted between the plastifi cation cylinder and the injection nozzle of a modifi ed injection molding machine. In the original design this special gassing unit consisted of a blowing agent inlet zone, a mixing zone, and an attached piston injecting unit, in which the homogenized polymer/blowing agent is stored under pressure until the injection phase. The supply of the blowing agent was achieved with a high-pressure piston pump.

Controller valve Blowing fluid duct

Shut-off nozzle

Mixing- andshear-elements

Blowing fluid dosingstation

FIGURE 4.11 Injection of blowing agent in the plastifi cation unit (MuCell®).

116 Polymeric Foams

61259_C004.indd 11661259_C004.indd 116 10/25/2008 12:36:15 PM10/25/2008 12:36:15 PM

An advanced concept uses an axial arrangement of the different compo-nents. The function of the injection piston is accomplished as in the conventional injection molding by the screw. Another function of the screw is to drive a dynamic mixer, which consists both of fi xed and mobile elements (stator/rotor principle). Figure 4.12 shows the principle of this process. The layout of the gassing unit is shown in Figure 4.13.

While the original concept requires a special injection molding machine owing to both an additional hydraulic system and the size of the system, the new variant only needs a conventional system design with an appropriately dimensioned gassing unit. Thus, in principle, this system is retrofi ttable.

The advantage of the original design was the separation of the different process steps of polymer plastifi cation, incorporation of the blowing agent,

Gas duct

Ergocell-unit

Shut-off nozzleBlowing fluid dosing

station

FIGURE 4.12 ErgoCell® process engineering scheme.

Single phase (melt/gas) Gas injection nozzle

Melt

Spline Mixer

Screw

ErgoCell-Module Plastification cylinder

[Demag Ergotech]

Injection piston

FIGURE 4.13 ErgoCell® module scheme.


61259_C004.indd 11761259_C004.indd 117 10/25/2008 12:36:16 PM10/25/2008 12:36:16 PM

and injection of the polymer/blowing agent system. In the new design the incorporation of the blowing agent is coupled to the dosing parameters of the plastifi cation unit. In theory this is of no disadvantage, but it offers less freedom to the operator during the optimization of the process.

4.2.5 Processing of Pellets Pre-Charged with Blowing Agent

Not only can the blowing agent be injected into the melt, the pellets can also be charged with blowing agent in advance. Whereas the charging takes place in an autoclave the pellets are processed on conventional injec-tion molding machines. Using this concept, the concentration in the poly-mer can be varied by adjusting the pressure, temperature and charging pressure within the autoclave. After an adequate charging time, the system pressure is reduced to ambient pressure and the pellets are removed. The subsequent desorption process lowers the concentration of blowing fl uid in the polymer. After an initial sharp decline of blowing fl uid content, the concentration gradient falls with an increase in desorp-tion time. The charged polymer pellets need a long, defi ned desorption time under ambient conditions before the foam injection molding process becomes reproducible. Subsequently they are processed on conventional injection molding machines. By an adequate course of the process, micro-cellular foams can be produced with this method.

A major advantage of this design is that no modifi cations to the injec-tion molding machine are necessary. Only a pressure vessel is required for charging the pellets with blowing fl uid. Within this vessel the material to be molded has to be charged up to 2 days before processing. In this context the blowing fl uid concentration in the polymer is not only determined by the parameters during the charging in the blowing fl uid, but also by the desorption time after charging. As there is a continuous loss of blow-ing agent and no possibility of varying the concentration at short notice the fabrication of mass-produced articles is rather complicated and even unrealistic.

4.2.6 Injection of the Blowing Agent with Aid of a Special Nozzle

A further process variant of the foam injection molding doses the blowing agent into the melt through a special injection nozzle, which is installed between the shut-off nozzle and the plastifi cation unit. During the injec-tion phase the melt fl ows through the injection nozzle and is loaded with the blowing agent there. This process, developed at the Institute of Plastics Processing (IKV) in Aachen22 and fi rst presented at the IKV-Colloquium in 2000, is licensed to the company Sulzer Chemtech AG, Winterthur (Switzerland), which distributes it under the name Optifoam™.

As the melt fl ows through the blowing fl uid injection nozzle, it is trans-ferred from a pipe fl ow over a torpedo into an annular slit fl ow. Here the

118 Polymeric Foams

61259_C004.indd 11861259_C004.indd 118 10/25/2008 12:36:17 PM10/25/2008 12:36:17 PM

blowing agent fl ows across permeable sinter metal sleeves and fi nally reaches the polymer melt. Thus, the agent is incorporated into the moving melt over the entire surface of the sinter metal sleeves. After the enrich-ment with blowing agent, the melt fl ows through downstream static mixing elements, before it is injected into the cavity through the shut-off nozzle.23 The principle of the blowing agent injection nozzle is shown in Figures 4.14 and 4.15.

The possibility of using conventional injection molding machines with-out modifi cations of screw or plastifi cation aggregate is of great advan-tage. This reduces substantially the investment costs and improves the fl exibility concerning the application scope. In this way, conventional injection molding machines can be retrofi tted for foam injection molding applications without substantial change. Figure 4.16 shows the system

Gas duct

Blowing fluid injection nozzle

Blowing fluid dosingstation

Static mixer

Shut-off nozzle

FIGURE 4.14 Fluid injection into a special nozzle.

Sinter metal sleeves

Melt flow

Phys. blowing fluid

FIGURE 4.15 Optifoam™ process scheme.


61259_C004.indd 11961259_C004.indd 119 10/25/2008 12:36:17 PM10/25/2008 12:36:17 PM

mounted on an injection molding machine as presented by the IKV on the K-2004 show in Düsseldorf, Germany. As the melt loading takes place before the head of the screw, the complete dosing volume of the injection unit can be used. However, depending on the design of the injection mold-ing machine the amount of space in front of the plastifi cation aggregate can be rather small.

4.3 Experimental Analysis of the Processing Parameters

in Foam Injection Molding

In the production of injection molded foamed parts, the material proper-ties and the process parameters have a decisive infl uence on the quality of the fi nal part. The foam injection molding process involves more parame-ters than conventional injection molding, which are not independent but correlate with each other. A thorough understanding of their interactions is thus required to control the resulting foam structure with their corre-sponding mechanical properties.

The following section will present two studies conducted at the IKV, in which the infl uence of process parameter variations on the foam part characteristics was analyzed. In the fi rst instance, a set of experiments performed with the IKV blowing fl uid injection nozzle will be presented, with the aim of evaluating the potential of cycle time reduction and ana-lyzing the effects of the process on the resulting foam morphology. After-wards the results of a second set of experiments will be presented, describing the effects of the process parameter variations on the density distribution and foam morphology. For these investigations, foamed parts

Shut-off nozzle Static mixer Blowing fluid injection nozzle

FIGURE 4.16 Optifoam™ system mounted on an injection molding machine.

120 Polymeric Foams

61259_C004.indd 12061259_C004.indd 120 10/25/2008 12:36:17 PM10/25/2008 12:36:17 PM

molded with pre-loaded polycarbonate were considered. In this case a regular injection molding machine was used without modifi cations of the plastifi cation unit. The prior process of charging the polymer with blow-ing agent was accomplished using an autoclave.

4.3.1 Experiments with the IKV Blowing Agent Injection Nozzle

In the foaming process with physical blowing agents the following param-eters were identifi ed as the most relevant:

Injection velocity•

Melt temperature•

Blowing agent content related to the gas mass fl ow.•

Further parameters like the mold temperature, the back pressure, and the system pressure of the gas dosing station are of interest.

A set of experiments was conducted with molding plates of 20% talc-fi lled polypropylene with a wall thickness of 4 mm.

4.3.1.1 Infl uence of the Process Parameters on the Foamed Part Characteristics

First, the infl uence of variations in the injection velocity was analyzed. As can be observed in Figure 4.17, for the same temperature in the melt and the same rate of blowing agent dosage, two injection speeds were tested. A fi ner and more even foam structure was reached with the increase in injec-tion velocity. The reason is that with an increase in the injection velocity a

Tm: 240°Cvinj: 40 mm/smCO2

: 0.2 g/s

Δr: 31.5%


: 0.2 g/s

Δr: 35.3%

FIGURE 4.17 Infl uence of injection velocity on foam structure and density reduction.


61259_C004.indd 12161259_C004.indd 121 10/25/2008 12:36:18 PM10/25/2008 12:36:18 PM

steeper pressure gradient in the cavity—and specifi cally in the runner system—is achieved, increasing therefore the rate of nucleation.

Subsequently the infl uence of variations of melt temperature was investi-gated. As can be seen in Figure 4.18, for the same injection velocity and rate of blowing agent dosage, a more homogeneous foam structure was achieved by increasing the melt temperature. It is remarkable that a variation of the melt temperature also infl uences the density reduction, and lighter foams are achieved with higher temperatures. A thickness reduction of the outer skin is another effect of the melt temperature increase, because the melt freezes later and allows foaming of the polymer closer to the cavity walls. However, as can be seen in Figure 4.19, the foam structure is not only dependent on the process parameters, but also on the location in the part.


: 0.2 g/s

Δr: 26.5%


: 0.2 g/s

Δr: 35.3%

FIGURE 4.18 Infl uence of melt temperature on foam structure and density reduction.

L/D = 5 L/D = 14

Close to the gate

0.68

0.64

0.60r (g

/cm

3 )

Far from the gate

L/D = 23

FIGURE 4.19 Comparison of foam structures along the fl ow path.

122 Polymeric Foams

61259_C004.indd 12261259_C004.indd 122 10/25/2008 12:36:19 PM10/25/2008 12:36:19 PM

4.3.1.2 Investigations on Cycle Time Reduction

Previous investigations showed that the use of the process-specifi c char-acteristics of FIM offers the possibility of reducing cycle time, and thus achieve more economic processing. However, today only vague knowl-edge exists about the mechanisms giving origin to this phenomenon. Furthermore, it is unclear how far the part characteristics (for example the wall thickness), the foam structure, or the process parameters (e.g. blow-ing agent, pressure and/or concentration) have an effect on the potential of the cooling time reduction.

Possibilities for the reduction of cycle time arise because of the lack of a packing phase. Otherwise, a lower melt viscosity grants shorter injection times. Beside the fact that lower melt temperatures can be applied and less mass is injected per shot, a smaller amount of heat has to be removed from the melt, reducing thus the cooling time. The part will also experience a more intensive cooling due to the evenly distributed cavity pressure. Another contribution to the removal of heat from the melt might come from a phase transition of the blowing agent, which is regarded of minor infl uence.

In foam injection molding trials, the cycle time reduction potential was investigated using a design of experiments. A disk shaped part with a diameter of 160 mm and different thicknesses was selected as testing geom-etry (Figure 4.20). Results for three sample thicknesses of 3, 4, and 6 mm were analyzed. The used material was polypropylene with 20% of talc content; carbon dioxide was used as blowing agent. The temperature of the part was measured after demolding by means of IR thermography. The digital infrared camera allows the detection of the surface tempera-ture distribution as well as the recording of the cooling process for a defi ned time period in a fi lm sequence. As a reference, the cooling time of compact parts was measured.

Nomenclature

m 230 100 60

9075150

200

Tm (°C) vinj (mm/s) pCO2 (bar)

240250

cpp

2n + 1 (n = 3) ⇒ 8 + 1 sets of parameters

Geometry: disc (Δ = 160 mm, d = 3, 4 and 6 mm)

Gate system: bar gate

Material: Polypropylene (talcum filled – 20 wt%)d

∅

FIGURE 4.20 Design of experiments to research cooling time reduction.


61259_C004.indd 12361259_C004.indd 123 10/25/2008 12:36:21 PM10/25/2008 12:36:21 PM

Figure 4.21 shows the surface temperature distribution of two 3-mm-thick parts just after demolding. The left picture shows the recording of the compact part, which was manufactured with the settings of the cen-tral point (cp, melt temperature 240°C), and the picture on the right side shows a part foamed in the range of the high level of process parameters (ppp, melt temperature 250°C).

The foamed part is demolded with a clearly lower surface temperature than the compact one. In this case the difference in the averaged surface temperatures amounts to 17°C. This result is particularly remarkable, because it can be seen as a substantial potential for cooling time reduc-tion. The fact that a lower surface temperature can be achieved even at a higher melt temperature is evidence that, contrary to expectations, there is a relatively large processing window, and therefore a high potential for the optimization of the part characteristics exists.

Figure 4.22 shows the mean demolding temperatures of parts produced with different settings of parameters and their corresponding density reductions. This diagram demonstrates that a lower demolding tempera-ture can be achieved for the 3-mm-thick foamed parts, in comparison to the compact part, regardless of the variations in process parameter set-tings. As the demolding temperature holds a relatively constant value for dissimilar process conditions, the infl uence of melt temperature variations seems to be relatively small.

Following the comparative investigations for the determination of the demolding temperature, the possibility of reducing of the cooling time was examined. Some of the test points for which the lowest demolding temperatures were determined in the test plan were repeated, applying a gradual reduction of the cooling time. The cooling time was shortened until a demolding without additional expansion could not be avoided.

Figure 4.23 shows the results of these tests. It is interesting that for the wall thicknesses of 3 and 4 mm a cooling time reduction of over 20% is

Compact

95.0°C

25.0°C

80

60

40

Foamedsetting: ppp

Cooling time: tC = 28 s

FIGURE 4.21 Surface temperature distribution after demolding.

124 Polymeric Foams

61259_C004.indd 12461259_C004.indd 124 10/25/2008 12:36:22 PM10/25/2008 12:36:22 PM

possible. Thus, an enormous potential for cost savings exists for these molding/material combinations. With 6-mm wall thickness the cooling time reduction diminishes to less than 10%. In consideration of the wall thickness, this result is still remarkable; however, it is not directly trans-ferable to other polymers.

4.3.2 Experiments on Pre-Loaded Polycarbonate

In a second series pre-dried polycarbonate pellets were loaded for 18 hours with carbon dioxide in an autoclave, set at 20 bar and ambient tem-perature. After a certain desorption of gas the charged material was dosed

60

50

40

Dem

oldi

ng te

mpe

ratu

re (

°C)

Den

sity

red

uctio

n (%

)

30

20

10

70

0

d = 3 mm

density reduction

Compa

ct

mm

mm

mp

mpm m

pppm

mpm

p cppp

mpp

p

demolding temperature

tC = 28 s 30

25

20

15

10

5

35

0

FIGURE 4.22 Demolding temperature and resulting density reduction depending on the

process parameters (m: low level, p: high level, cp: central point—see Figure 4.21).

60 40

30

20

10

0

50

40

30

20

10

03 4

Wall thickness (mm)

Tim

e (s

)

Coo

ling

time

redu

ctio

n (%

)

Cooling time reduction

Compact

Foamed

6

FIGURE 4.23 Comparison of cooling times and attainable cooling time reductions.


61259_C004.indd 12561259_C004.indd 125 10/25/2008 12:36:22 PM10/25/2008 12:36:22 PM

in an injection molding machine, and processed within a window of 2.5% and 2.0% CO2 content. With this processing window and a fi xed dosing volume in the injection molding machine, an overall density reduction of about 20% was attained. Plates with a thickness of 3 mm, a length of 200 mm, and a width of 100 mm were molded through one gate of a hot runner system located in the center of the plate.

For the production of the plates, three process parameters were varied: mold temperature, melt temperature, and injection velocity. The plan followed is depicted in Table 4.4. For different combinations of these parameters the effects on the density distribution and cell morphology were analyzed.

4.3.2.1 Morphological Characterization Methods

The density distribution was analyzed in both macro- and microscopes. In the macroscope, the density distribution along the major dimensions of the part was determined through three different methods: X-rays, grid punching, and IR thermography. The density distribution and foam mor-phology across the transverse section of the part were observed with the aid of a microscope. Each method will be explained in detail below.

With the X-ray method an image of the cell distribution along each part was obtained, as can be seen in the left image of Figure 4.24. The “grid punching” method allowed a quantitative characterization of the density distribution: each part was divided into a grid, and discs were punched from each grid cell. As the size and weight of each disc were known, the local density could be determined. In the IR thermography method of density characterization a series of images of the part was made just after demolding. Thus the temperature distribution could be observed over a time window.

In IR thermography, the local density of the part can be correlated to the local heat; where there is more concentration of material and therefore less foam, the temperatures are higher. On the contrary, in the regions where the temperature is low, there are more cells and less material to cool, and therefore the part experiences a faster heat transfer.

TABLE 4.4

Design of Experiment (DOE) of Injection Molding Parameters

Parameter Dimension Levels

Variable Injection speed (cm3) 20 60 100

Mold temperature (°C) 80 100 120

Melt temperature (°C) 290 310 330

Constant Density reduction % 20

Plate thickness (mm) 3

126 Polymeric Foams

61259_C004.indd 12661259_C004.indd 126 10/25/2008 12:36:23 PM10/25/2008 12:36:23 PM

For the microscopic analysis small samples were cut at a location equi-distant from the injection point and the end of the part. Figure 4.24 shows the exact location where the foam structure observations were made. A typical microscopic image of the foam structure achieved within this set of experiments is included.

With the purpose of comparing the validity of the methods used to determine the macro distribution of the foam density, the results obtained with the grid-punching method and the IR thermography method were plotted in the same diagram, as shown in Figures 4.25 and 4.26. It can be seen that the correlation between both parameters is quite satisfactory.

Injection gate

Down Up

Material:PC Makrolon 2005

FIGURE 4.24 X-ray image of a foamed polycarbonate plate (left), magnifi cation of local

foam structure (right).

Temperature distribution Density distribution

T (°C)

120 1.0501.0361.0231.0090.9950.9820.9680.9550.9410.9270.914

117115112109106104101989593

Conditions: Tmelt = 310°C / vinj = 60 cm3 / Tmold = 100°C

r (g/cm3)

FIGURE 4.25 Surface temperature and density distribution.


61259_C004.indd 12761259_C004.indd 127 10/25/2008 12:36:23 PM10/25/2008 12:36:23 PM

4.3.2.2 Infl uence of the Process Parameters on the Foamed Part Characteristics

As an initial point for the analysis, the variation in the homogeneity of the density distribution was observed. In Figure 4.27, the gradient of tempera-ture distribution as mean values of m1 and m2 (cp. Figure 4.26), which as mentioned before correlates directly with the density distribution, is plot-ted against the variables of melt temperature, injection velocity, and mold temperature. As the gradient of temperature distribution increases, a less homogeneous distribution of the foam density is observed. The depen-dence of the density distribution on each of the injection parameters is explained below.

Figure 4.27 reveals that a more homogeneous foam structure is achieved along the fl ow path if the temperature of the melt is increased. This can be explained by the effect of temperature on the viscosity of the melt. When the temperature is increased, the viscosity and therefore the resistance to cell formation decreases, and thus a homogeneous cell structure will be likely to appear, even at locations distant from the injection gate.

The increase in melt temperature also has a positive effect on the cell structure. Once the maximum value is reached, a fi ner cell morphology is achieved, as can be seen in Figure 4.28. The number of nucleation sites will be increased at higher temperatures, and therefore the primary mecha-nism of cell forming, nucleation, will predominate over the secondary mechanism, the diffusion of blowing agent into primarily formed cells. Measurements made at the IKV also showed that at high melt temperatures the surface tension of the melt is reduced, thus promoting the nucleation of independent, newly formed cells with a fi ner structure cell morphology.

The effect of different injection velocities on the gradient of tempera-ture distribution is depicted in Figure 4.27. As expected, the temperature

–100 –50 0

1.0

0.9

0.8

0

m 1 m2

Sample length (mm)

Nor

mal

ized

dis

trib

utio

n (–

)

TemperatureDensity

50 100

FIGURE 4.26 Correlation between the surface temperature of the samples and the corre-

sponding density distribution: plot of normalized values for temperature and density

distributions.

128 Polymeric Foams

61259_C004.indd 12861259_C004.indd 128 10/25/2008 12:36:24 PM10/25/2008 12:36:24 PM

gradient of the demolded part will be less steep with a lower injection velocity. This is mainly an effect of the appearance of a thicker outer skin. A less abrupt pressure drop, coming from lower injection velocities, will result in the appearance of less nucleation sites, and therefore the foamed portion of the part will be smaller. A thicker compact border layer will be able to carry on a more effective heat transfer through the mold surface, and in turn will generate a cooler and more homogeneous surface temperature distribution.

Another interesting effect of the velocity variation can be seen in the orientation of the cell structure. Figure 4.29 shows the difference in foam structure resulting from variations in the injection velocity. As can be seen, at low velocities cells of relatively large dimensions and strong orientation are formed. This is a result of the simultaneous effect of cell

Tm = 290°C Tm = 310°C

Flow direction

Tm = 330°C

FIGURE 4.28 Dependence of foam structure on melt temperature. Samples taken in the

middle of the fl ow path.

0.4

0.3

0.2

0.1

Tem

pera

ture

gra

dien

t (°C

/mm

)

0.0290 300

Melt temp. (°C) Injection velocity (cm3/s) Mold temp. (°C)

310 320 330 40 60 80 80 9020 100 100 110 120

FIGURE 4.27 Infl uence of injection velocity, mold temperature, and melt temperature on

the homogeneity of the density distribution.


61259_C004.indd 12961259_C004.indd 129 10/25/2008 12:36:24 PM10/25/2008 12:36:24 PM

growth and melt displacement. On the other hand, if the injection velocity is increased, a much fi ner and homogeneous foam structure is revealed: a steeper pressure gradient grants more nucleation sites and a later bubble growth. Here the mechanism of primary cell nucleation predominates over that of gas diffusion in the process of foam formation.

The effect of the mold temperature on the foam structure is illustrated in Figure 4.30. At a low mold temperature, a foam structure with clear orientation and large cells can be attained. With an increasing tempera-ture difference between mold and melt the heat will be transferred faster, giving place to an earlier increase in viscosity and therefore to less nucle-ation sites. Another consequence of this effect is the increase of the melt surface tension. The growth mechanism of blowing agent diffusion into the primarily formed cells is responsible for most of the resulting foam

vi = 20 cm3/s vi = 60 cm3/s

Flow direction

vi = 100 cm3/s

FIGURE 4.29 Dependence of foam structure on injection velocity. Samples taken in the


Tm = 80°C Tm = 100°C

Flow direction

Tm = 120°C

FIGURE 4.30 Dependence of foam structure on mold temperature. Samples taken in the


130 Polymeric Foams

61259_C004.indd 13061259_C004.indd 130 10/25/2008 12:36:26 PM10/25/2008 12:36:26 PM

structure, and thus the cells exhibit larger mean diameters. On the other side, with a high mold temperature the heat is conducted through a less steep gradient, and the material is kept viscous enough to allow the apparition and growth of a more homogeneous foam structure. A com-prehensive summary of the effects of changes in the injection molding parameters on the cell size, cell number and compact skin thickness is given in Figure 4.31.

4.4 Optimization of the Surface Quality of Foamed

Injection Molded Parts

Besides the many advantages of foam injection molding (FIM), the achiev-able surface qualities are rather poor in many cases. Occurring silver streaks, melt eruptions, and cold-displaced polymer melt areas cause more uneven and non-uniform part surfaces in comparison to conventional injection molding. For this reason, foamed parts are often excluded as visually exposed parts. A comprehensive understanding of the effects arising during the fi lling phase establishes new possibilities for increasing

Mold temperature

Cel

l siz

e

Cel

l num

ber

Ski

nth

ickn

ess

Cel

l siz

e

Cel

l num

ber

Ski

nth

ickn

ess

Cel

l siz

e

Cel

l num

ber

Ski

nth

ickn

ess

Mold temperature Mold temperature

Melt temperature Melt temperature Melt temperature

Injection velocity Injection velocity Injection velocity

FIGURE 4.31 Summary of the infl uence of the process parameter on the resulting foam

structure.


61259_C004.indd 13161259_C004.indd 131 10/25/2008 12:36:28 PM10/25/2008 12:36:28 PM

the surface qualities in foam injection molding. New research shows that different process variants of FIM such as “breathing” molds, gas counter-pressure, and structured and coated cavity surfaces can increase the surface quality effectively.

4.4.1 Occurrence of Silver Streaks

If no particular measures are taken, for example, in mold technology, foamed parts show rather poor surfaces. In many cases low surface quali-ties are caused by bright silver streaks, which are oriented in the direction of the fl ow path. Semerdjev and Popov explained the development of these streaks with gas bubbles leaving the fl ow front during fi lling, which are sheared when contacting the cavity wall.24 A schematic description of their development is shown in Figure 4.32. The example shows the surface of a physically foamed part of polypropylene (PP-T20) fi lled with 20 wt% talc with a wall thickness of 6 mm, which was gated in the center. The surface was photographed with a fl atbed scanner and increased in con-trast to point out the occurring surface defects. This surface shows clearly a multitude of silver streaks in the form of bright shining lines oriented in direction of the fl ow path.

The development of these silver streaks can be analyzed in detail by looking at the surfaces of foamed polypropylene parts. Figure 4.33 shows two impinging light microscopy pictures of two cut-outs of a PP-surface. On the left picture, a 420 μm broad white band can be clearly seen, which can be detected as a fi ne white line or as a silver streak on the part surface. In the picture on the right, it can be noticed that the white band consists of many small white spots with an average diameter of 40–50 μm. This pic-ture leads to the assumption that a macroscopic gas bubble, grown during

FIGURE 4.32 Schematics of the development of silver streaks.

132 Polymeric Foams

61259_C004.indd 13261259_C004.indd 132 10/25/2008 12:36:28 PM10/25/2008 12:36:28 PM

the fi lling phase, is split into many small microscopic gas bubbles by shear strain on the cavity wall. Thereby, these small gas bubbles seem to modify the part surfaces resulting in different refl ection properties, which evoke the bright shining lines—called silver streaks—on a larger scale.

A possible modifi cation of the surface will be analyzed in further exam-inations. A white shining spot on the surface was investigated in detail using scanning electron microscopy, which is marked by a white dashed line (see Figure 4.34, top left). On closer inspection one can see a rather rough and uneven surface outside of the dashed range. This roughness is caused by shear strain of the gas loaded melt on the cavity wall and by the low cavity pressures in foam injection molding.25 Inside of the dashed area the surface seems rather even and sunk in comparison to the sur-rounding surface. A cut through the part in Figure 4.34 (top right) shows

FIGURE 4.33 Cut-out of a foamed part surface with silver streaks.

A

50 mu

B

Cut: A – B

Silver streak (even)

(Close to gate)Pressure p

Distance s

Part surface (uneven)

(Far from gate)Deepening/dentin part surface

500

FIGURE 4.34 Stretched microbubble as deepening on the part surface.


61259_C004.indd 13361259_C004.indd 133 10/25/2008 12:36:28 PM10/25/2008 12:36:28 PM

the geometry of the surface. This deepening gives rise to the supposition that the microscopic gas bubbles release a small amount of gas, which is enclosed between melt and cavity wall displacing the hot melt. The dimen-sion of the deepening, which can be regarded as a dent, should be depen-dent on the amount of gas, on the gas pressure, and the pressure in the melt. Following this idea the displaced melt volume results from the balance of forces between gas pressure and resistance of the melt to be displaced. Despite the low cavity pressures in foam injection molding in comparison to compact injection molding,26 the pressures within the dents are suffi cient to cause a deepening in the surface.

Regarding the surfaces of foamed parts along the fl ow path, it is well known that the surface qualities decrease with increasing distance to the gate. Following the above presented idea this is caused by the decreasing melt pressure along the fl ow path. Corresponding to the balance of forces between gas pressure and resistance of the melt to be displaced, the dents will be smaller closer to the gate than at the end of the fl ow path. A second reason for inferior surface qualities over the fl ow path is that the gas bubbles have more time to grow within the melt depending on the diffu-sion rate to deliver a bigger gas volume to be enclosed within the dents. Both effects lead to inferior surface qualities along the fl ow path.

Within the performed investigations a correlation between silver streaks and the surface deformations could be detected. However, the reason for the bright refl ections has not yet been clarifi ed. Possible explanations are on the one hand a different evenness on the surface and on the other hand a different crystallization behavior of the surfaces inside and outside of the dents. Considering a different evenness, the surface within the area of the deepenings is less rough than on the surrounding surfaces, which may be caused by less shearing and the surface tension of the melt. The smoother surface within the dents in turn has a signifi cant effect on the refl ection properties, as it leads to a more directed light refl ection in com-parison to a diffuse refl ection of the surrounding areas (Figure 4.35). The second explanation approach is based on a different crystallization behav-ior inside and outside the dents. Due to the gas volume encapsulated in the deepenings, the heat transfer between mold and polymer melt is reduced resulting in higher surface temperatures in the area of the dents. Thus, the hot polymer has more time for the crystallization process, which in turn has an effect on the degree of crystallization.27 For this reason an effect on the optical density is expected increasing the degree of refl ec-tion. Both approaches will be analyzed in future examinations.

4.4.2 Possibilities of Increasing the Surface Quality Through Mold and Process Technology

Owing to the examinations that the development of surface defects is strongly dependent on pressures and temperatures within the melt, the

134 Polymeric Foams

61259_C004.indd 13461259_C004.indd 134 10/25/2008 12:36:29 PM10/25/2008 12:36:29 PM

choice of the process parameters is of high importance for the surface qualities of foamed parts. However, the impact of a sole optimization of process parameters is limited.

In the past, poor surface qualities also led to many developments in the fi eld of mold and process technology in FIM. All process variants are based on conventional foam injection molding in which the mold is only partially fi lled during the machine injection phase depending on the weight reduction to be achieved. This standard process leads to low pres-sures in the cavity resulting in poor surface quality. For this reason the gas counter-pressure process, “breathing” molds, structured cavities, heat insulating cavity coatings, and the variotherm process were investigated. The fi rst four will be briefl y described in the following paragraphs.

Using the process variant “breathing” mold, which is shown in Figure 4.36, the mold cavity is completely fi lled with the polymer/blowing agent mixture during the machine injection phase.28 To achieve compact border

Directed light reflection

Distribution ofluminosity

Diffuse reflection

Distribution ofluminosity

Uneven surfaceEven surface

αα ββ

FIGURE 4.35 Light refl ection on plain and rough surfaces.

FIGURE 4.36 Scheme of the process variant “breathing” mold.


61259_C004.indd 13561259_C004.indd 135 10/25/2008 12:36:29 PM10/25/2008 12:36:29 PM

layers, a short packing phase is applied. Subsequently the cavity is enlarged initiating the foaming phase. By the fi lling of the cavity under high pres-sures and the following packing phase, occurring surface defects can be “repaired” after the fi lling phase. The holding pressure in the packing phase causes the reduction or avoidance of deepenings against the encapsulated gas pressure, so that the surface qualities can be increased signifi cantly.

By use of the process variant gas counter-pressure, as shown in Figure 4.37, a foaming of the melt/blowing agent-mixture can be avoided during the injection phase by application of a gas pressure inside the cavity of about 40 bars.28 By means of counter-pressure, the blowing fl uid is kept above the saturation pressure inside the melt, whereby the development and the growth of gas bubbles, which are sheared during the fi lling phase, are avoided. This process variant is often combined with a “breathing” mold.

A suited and effective mold technology for increasing the surface quality is the modifi cation of the cavity surface with textures. The functional prin-ciple is not based on the increase of the process pressures, but rather on a better venting of the escaping gas and concealment of possible surface defects. By the use of texturized cavities, the development of silver streaks can be avoided by reducing shear strain of the growing gas bubbles on the cavity wall, which can be reached by a mechanical anchoring of the poly-mer in the structured mold surface. Furthermore the structures superpose the rough and uneven surfaces of foamed parts, so that the appearance of silver streaks is reduced by a diffuse backscatter (Figure 4.38, top).

A further possibility is the use of coated surfaces. A heat insulating coating on the cavity surface enables an increase of the surface quality of foamed parts with only a minor increase in cycle time.29 Those coatings cause a reduction of the cooling rate of the outer part layers, causing a better fl ow with less orientation, less internal stress, and optical appealing

FIGURE 4.37 Scheme of the gas counter-pressure process.

136 Polymeric Foams

61259_C004.indd 13661259_C004.indd 136 10/25/2008 12:36:30 PM10/25/2008 12:36:30 PM

surfaces (Figure 4.38, bottom). The improvement of the part surface can be attributed to the fact that fl ow marks resulting from sheared gas bubbles can longer be deformed and corrected through a higher contact tem-perature on the part surface. For this reason the silver streaks, which can be seen as a multitude of dents, are minimized or completely avoided depending on the coating by the prevailing melt pressure. The encapsu-lated gas within the deepenings is either compacted or diffuses back into the polymer.

4.4.3 Investigation of “Breathing” Molds and Gas Counter-Pressure

The process variants “breathing” mold and gas counter-pressure have a signifi cant infl uence on the process pressures, which in turn have a strong impact on the surface qualities. Characteristic cavity pressure curves are presented in Figure 4.39. This fi gure shows that the pressures achieved with conventional foam injection molding are much lower than with com-pact injection molding, which is an important reason for the poor surface qualities. The process variants aim to increase the surface qualities by raising the process pressures. Using gas counter-pressure, the pressures are increased primarily during the injection phase avoiding the develop-ment of gas bubbles and melt explosions during the fi lling stage. The pro-cess variant “breathing” mold uses primarily the increase in pressure after the fi lling stage, whereby the characteristic surface defects are to be “repaired” afterwards.

The effect of both process variants on surface quality were studied. Therefore surfaces of black colored parts of polypropylene fi lled with 20 wt% talc, being produced with the respective processes, have been photo graphed and graphically analyzed. Figure 4.40 shows four pictures of these part surfaces in direct comparison. Owing to the fact that the

FIGURE 4.38 Use of structured and coated cavity surfaces.


61259_C004.indd 13761259_C004.indd 137 10/25/2008 12:36:30 PM10/25/2008 12:36:30 PM

human eye detects dark areas on white areas more easily than vice versa, all pictures are inverted with regard to color scale. Thus dark lines are bright shining silver streaks in reality. Regarding the surface pictures, multiple dark spots exist on the parts produced without gas counter- pressure. These spots can be explained as melt particles, which were blasted from the fl ow front and were surrounded afterwards by the melt again. As one can see, these surface defects can be avoided by gas counter- pressure. Beyond that the surfaces are more even and smooth in compari-son to conventional injection molding. Regarding the surfaces of the parts manufactured by the process “breathing” mold, one can see an articulate brighter coloration. This means that in reality the surfaces appear darker and exhibit fewer defects in the form of silver streaks. Both process variants

pCP pcompact

pFIM+GCP

pTSG+breathing

pFIM

TimeStart injection

FIGURE 4.39 Characteristic cavity pressure curves of different process variants in foam

injection molding.

FIGURE 4.40 Comparison of the surface qualities of foamed parts produced with differ-

ent process variants.

138 Polymeric Foams

61259_C004.indd 13861259_C004.indd 138 10/25/2008 12:36:30 PM10/25/2008 12:36:30 PM

show the capability of improving the surface qualities especially when used in combination.

4.5 Summary

Foam injection molding (FIM) is gaining interest due to the advantages associated with process and part performance. In comparison with conven-tional injection molding, new possibilities arise from the production point of view, with a reduction in cycle time and machine tonnage required, and regarding fi nal part performance from the reduction in warpage, material consumption, and dimensional accuracy. The specifi c mechanical perfor-mance, particularly under bending load, can also be improved. However, considerations associated with the complexity of the process technology, the low surface quality of the fi nal part, and an overall reduction of mechan-ical properties, require a deep understanding of the process.

The foaming is achieved by dosing of a blowing agent into a polymer melt. This dosing will be infl uenced by the temperature, the process, and the material. The mechanisms of cell nucleation and growth, which have a large infl uence in the fi nal foam structure, strongly depend on the pres-sure drop velocity. An increase in the pressure drop gradient will lead to the formation of a fi ner and more even foam structure.

The interest in physical blowing agents is growing parallel with the development of FIM technology, owing to the higher levels of foaming attainable as well as to the absence of residual products (as with chemical blowing agents). Their dosing into the polymer melt has required the development of special technologies, some of which are reviewed within this contribution. Each of them uses a different principle to incorporate the blowing agent into the melt.

More process parameters are involved in the production of foamed injection molded parts compared with the conventional injection molding process. These parameters are correlated and affect each other. Through design of different experiments the infl uence of parameter variations on the molded foamed parts was analyzed. It can be concluded that with increasing injection velocity, mold temperature, and melt temperature a fi ner and more homogeneous foam structure can be attained, and that the thickness of the outer skin was increased by lower mold temperatures. It was also demonstrated that there exists a potential for cycle time reduc-tion in foam injection molding, which does not show a strong dependence on the melt temperature.

The poor surface quality normally associated with foam injection mold-ing comes from the apparition of gas bubbles and their destruction on the surface. The induction of pressure in the mold cavity, either through


61259_C004.indd 13961259_C004.indd 139 10/25/2008 12:36:31 PM10/25/2008 12:36:31 PM

technologies such as “breathing” the mold or gas counter-pressure can counteract these negative effects. Other alternatives are the use of textures or coverings on the cavity surface, which allow the controlled migration of gas from the part surface.

Despite an already broad scope of application of foamed parts, the implementation of foam injection molding is to be examined specifi cally for each application. Even if economic advantages seem available through a reduced cooling time, the application of this process is not always possible, and the direct substitution of a compact part by a foamed one may not always satisfy the stated requirements. In many cases, the exclu-sion criterion is the low surface quality or an inadmissible loss in mechani-cal properties. However, these problems can be avoided by an integral approach considering the requirement of the foam injection molding pro-cess and the properties of thermoplastic foams in the early design phase of the product.

Acknowledgments

The investigations set out in this chapter received fi nancial support from the Federal Ministry of Economics (BMWi) within the “Arbeitsgemeinschaft Industrieller Forschungsvereinigungen “Otto von Guericke” e. V. (AiF) (Project No. 14377 N), to whom we extend our thanks. The material was provided by Ticona GmbH, Kelsterbach, Germany, Borealis GmbH, Linz, Austria, Bayer Material Science AG, Leverkusen, Germany and Clariant Masterbatch GmbH & Co. OHG; Ahrensburg, Germany. The machine tech-nology was provided by Sulzer Chemtech AG, Winterthur, Switzerland, and Demag Ergotech GmbH, Schwaig, Germany. We thank all of them for their support.

References

1. Lübke, G. “Jedem das Seine – Treibmittelsysteme und Nukleierungsmittel für thermoplastische Schaumstoffe.” In IKV-Seminar zur Kunststoffverarbeitung, Aachen, 26–27 June 2001.

2. Spiekermann, R. “Entwicklungstendenzen bei chemischen Treibmitteln für Thermoplastschäume.” In IKV-Seminar zur Kunststoffverarbeitung, Aachen, 23–24 October 1997.

3. Leppkes, R. “Polyurethane.” In Werkstoff mit vielen Gesichtern. Verlag Moderne Industrie AG, Landsberg, 1993.

140 Polymeric Foams

61259_C004.indd 14061259_C004.indd 140 10/25/2008 12:36:31 PM10/25/2008 12:36:31 PM

4. Shutov, F. A. Integral/Structural Polymer Foams. Springer Verlag, Berlin Heidelberg, 1986.

5. Van Krevelen, D. Properties of Polymers. Elsevier, Amsterdam, 1990. 6. Li, C. C. “Critical temperature estimation for simple mixtures.” Canadian

Journal Chemical Engineering 19 (1971): S.709. 7. Kreglewski, A. and Kay, W. B. “The critical constants of conformal mixtures.”

Journal of Physical Chemistry 73 (1969) 10: S.3359. 8. Pfannschmidt, L. O. Herstellung resorbierbarer Implantate mit mikrozellulärer

Schaumstruktur. Dissertation, Rheinisch-Westfälische Technische Hochschule Aachen, 2001.

9. Menges, G. Werkstoffkunde Kunststoffe. Carl Hanser Verlag, München, New York, 1990.

10. Flory, P. J. Principles of Polymer Chemistry. Cornell University, Ithaca, 1969. 11. Fischer, M. and Schmidt, R. “Diffusion und Permeation, Eigenschaften von

Polymeren.” In Chemie und Physik Band I. Georg Thieme Verlag, Stuttgart, New York, 1985.

12. Crank, J. The Mathematics of Diffusion. Clarendon Press, Oxford, 1975. 13. Wu, S. Polymer Interface and Adhesion. Marcel Dekker, New York, 1982. 14. Reid, R. C., Prausnitz, J. M., and Poling, B. E. The Properties of Gases & Liquids,

4th edition. McGraw Hill, New York, 1988. 15. Sander, B. Mikrozelluläre Schäume – Einfl uss des physikalischen Treibmittels und

des Polymers auf den Schäumprozess. Institut für Kunststoffverarbeitung, RWTH Aachen, unveröffentlichte Diplomarbeit. O. Pfannschmidt, Betreuer, 1999.

16. Colton, J. S. “Making microcellular foams from crystalline foams of crystal-line polymers.” Plastics Engineering (1988) 8: S.53–S.55.

17. Nicolay, A. Untersuchung zur Blasenbildung in Kunststoffen unter besonderer Berücksichtigung der Rissbildung. Dissertation, Rheinisch-Westfälische Technische Hochschule Aachen, 1976.

18. Colton, J. S. and Suh, N. P. “The nucleation of microcellular thermoplastic foam with additives. Part I: Theoretical Consideration.” Polymer Engineering Science 27 (1987) 7: S.485–S.492.

19. Han, J. H. and Han, C. D. “Bubble nucleation in polymeric liquids. Part II: Theoretical considerations.” Journal of Polymer Science Part B 28 (1990): S.743–S.761.

20. Pierick, D. E., Anderson, J. R., Cha, S. W., Stevenson, J. F., and Laing, D. E. Injection Molding of Microcellular Material. Europäische Patentanmeldung EP1264672A1, 2002.

21. Jaeger, A. “Schäumen beim Spritzgießen neu entdeck.” In: Tagungshandbuch “Präzisionsspritzguss heute.” KI Lüdenscheid, Lüdenscheid, February 2002.

22. Schröder, T. Entwicklung einer Eingasungsdüse zur Herstellung von Thermoplastschäumen im Spritzgießverfahren unter Verwendung eines physika-lischen Treibmittels. Student project work (supervisor: O. Pfannschmidt), Institut für Kunststoffverarbeitung, RWTH Aachen, 2001.

23. Habibi-Naini, S. Schaumspritzgießen – Untersuchungen zur Beladung der Schmelze mit physikalischen Treibmitteln während der Einspritzphase. Advisory group meeting, IKV, Aachen, 2000.

24. Semerdjiev, S. and Popov, N. “Probleme des Gasgegendruck-Spritzgießens von thermo-plastischen Strukturschaumteilen.” Kunststoffberater 4 (1978): 198–201.


61259_C004.indd 14161259_C004.indd 141 10/25/2008 12:36:31 PM10/25/2008 12:36:31 PM

25. Okamoto, K.T. Microcellular Processing. Carl Hanser Verlag, München, Wien, 2002.

26. Habibi-Naini, S. Neue Verfahren für das Thermoplastschaumspritzgießen. PhD thesis, RWTH Aachen University, 2004.

27. Menges, G. Werkstoffkunde Kunststoffe. Carl Hanser Verlag, München, Wien, 2002.

28. Semerdjiev, S. Thermoplastische Strukturschaumstoffe. VEB Deutscher Verlag für Grundstoffi ndustrie, Leipzig, 1980.

29. Horn, B., Mohren, P., and Wübken, G., Spritzgusswerkzeuge mit wärmedäm-menden Formnestbeschichtungen. Notes from the Institute of Plastics Processing (IKV) at the RWTH Aachen University, 1976.

142 Polymeric Foams

61259_C004.indd 14261259_C004.indd 142 10/25/2008 12:36:31 PM10/25/2008 12:36:31 PM

5Foaming Analysis of Poly(e-Caprolactone) and Poly(Lactic Acid) and Their Nanocomposites

Ernesto Di Maio and Salvatore Iannace

CONTENTS

5.1 Introduction .................................................................................. 1445.2 Foaming of PCL and PLA with CO2 and N2: Key Issues ....... 145

5.2.1 Foaming of PCL .................................................................. 1455.2.2 Foaming of PLA ................................................................. 149

5.3 Molecular Modifi cation of PCL and PLA ................................. 1515.3.1 Branching/Cross-Linking of PCL .................................... 1515.3.2 Chain Extended PLA ......................................................... 154

5.4 Nanocomposites ........................................................................... 1565.4.1 Nanocomposites from PCL and PLA: Rheology,

Sorption, Mass Transport, and Crystallization .............. 1575.4.2 Foaming of PCL and PLA-Based

Nanocomposites .................................................................. 1625.4.3 Bubble versus Crystal Nucleating Effect

of Nanoparticles .................................................................. 1635.5 Conclusions ................................................................................... 168References ............................................................................................ 169

61259_C005.indd 14361259_C005.indd 143 10/25/2008 12:37:29 PM10/25/2008 12:37:29 PM

5.1 Introduction

The large and increasing volume of foamed materials has stimulated interest in developing useful innovations related to environmentally friendly processes and materials that can be utilized in these high volume applications. Biodegradation can be considered as a possible approach to solve the disposal problem, especially when recycling is diffi cult or costly or when biodegradation is a functional requirement of the product.1

Biodegradable polymers are already utilized in many biomedical applica-tions such as biodegradable sutures, wound dressing, bio-resorbable implants, and drug delivery systems, applications where the high cost of the materials is justifi ed.2 However, their use in commodity applications, such as packag-ing or agriculture, is still limited either for economic reasons or for diffi culties related to their processing, often due to their poor thermal stability.

Biodegradable materials for industrial foaming applications must dis-play adequate properties, and their manufacturing process must be rela-tively simple and inexpensive. Among the most interesting biodegradable polymers that can be potentially employed for foaming are the polyesters such as poly(e-caprolactone) (PCL) and poly(lactic acid) (PLA) and their copolymers, polyhydroxyalkanoates (PHA), polyesteramide, polyure-thanes (PU), and biopolymers such as polysaccharides and proteins. Since the availability of a large volume of commercial raw material supplies is uncertain, the combination of these polymers in blends and/or com-posites is often the way to prepare polymeric systems with a lower cost and with properties that can be tailored for the preparation of foams. This chapter will focus on the foaming behavior of PCL and PLA named poly(e-caprolactone) and poly(lactic acid). They have received a great attention by researchers working in the biomedical fi eld and, more recently, an increasing interest for larger scale industrial applications. More information on other biodegradable foamed materials can be found in recent reviews.3,4

The main requirements for foamability are, for any thermoplastic poly-mer in general, the rheological characteristics of the melt,5,6 the blowing agent solubility and diffusivity, and the presence of adequate setting mechanisms. In particular, strain-induced hardening behavior is a funda-mental characteristic for the foaming process, since it allows withstand-ing of the stretching forces at the latter stage of the bubble growth. There are basically two possibilities to improve the elongational properties of the melt: (1) optimize the molecular weight and molecular weight distri-bution of the polymer and/or (2) the branching of the macromolecules.7

In principle, chemical modifi cation (i.e. chain extension and/or branch-ing) can be used to modify the rheological properties of biodegradable polyesters such as PCL and PLA. In particular, two different techniques are described in this chapter: (1) branching and cross-linking with peroxides

144 Polymeric Foams

61259_C005.indd 14461259_C005.indd 144 10/25/2008 12:37:29 PM10/25/2008 12:37:29 PM

for PCL; and (2) chemical modifi cations with chain extenders for PLA. The chemically modifi ed materials were batch foamed and the effects of the molecular modifi cations on the physico-chemical properties of the polymers as well as on the foaming process are discussed.

Another strategy for improving the foamability of thermoplastic bio-degradable polymers is the use of nanoparticles.8,9 Nanometric fi ller, in effect, determines extensive modifi cation of properties such as gas absorp-tion and diffusivity, local thermodynamic properties responsible for nucleation and growing phenomena, and rheological characteristics which affect the fi nal morphology of the foams. The effect of type and content of nanofi llers on the foaming process of PCL and PLA is reviewed here.

5.2 Foaming of PCL and PLA with CO2 and N2: Key Issues

In recent years, few research groups have focused their attention on foam-ing of PCL and PLA, the two biodegradable polyesters of interest in this chapter. In this paragraph, a comprehensive picture of the key aspects of the foaming process of these two polymers, with CO2 and/or N2 as blow-ing agent, will be given. As a general, primary comment, both PLA and PCL have proved to be diffi cult to foam, mainly for their poor rheological properties and small processing windows. As a signifi cant difference between the two polymers it is important to underline that the main limi-tations in the use of PCL are related to low processing temperatures while, in the case of PLA, to low crystallization kinetics, both of which reduce the possible setting mechanisms of the newly formed cellular structures.

5.2.1 Foaming of PCL

PCL is synthetic aliphatic biodegradable polyester with chemical formula -[-O-(CH2)5-CO-]n-, which has been foamed by different research groups in a narrow range of molecular weights (Mn from approximately 69,000 to 80,000 g/mol, from different producers). PCL has a glass transition tem-perature of �64.2°C and a melting point of 69.0°C.10,11 When crystallized from the melt at �10°C/min, the onset of crystallization is at approxi-mately 30°C.12 PCL crystallization is quite rapid: when cooled isothermally from the melt at 40°C, the crystallization half-time, t1/2, is 6.5 min, with an Avrami exponent close to 2 and a fi nal degree of crystallinity of 40%,13 while, when cooled isothermally from the melt at �1°C, the crystallization half-time, t1/2, is 140 ms.14

The described crystallization kinetics are adequate for the stabilization of the foam but, in industrial applications (e.g. extrusion foaming), it is often diffi cult to cool the extrudate to such low temperatures, resulting in

Foaming Analysis of PCL and PLA 145

61259_C005.indd 14561259_C005.indd 145 10/25/2008 12:37:29 PM10/25/2008 12:37:29 PM

a reduced effi ciency of the expansion process. In batch foaming, conversely, the better control of the processing variables allows a fi ner control of the morphology and the achievement of foams of lower densities. As a matter of fact, most of the experimental work on PCL foaming has been per-formed in batch.

Regarding the sorption properties of the blowing agents of concern here, solubility and diffusivity of carbon dioxide (CO2) and nitrogen (N2) in PCL have been measured at typical processing conditions. Equilibrium sorption concentrations of CO2 and N2 in PCL at different pressures and temperatures are reported in Figure 5.1 and denote a fairly higher solubil-ity of CO2 with respect to N2. Mutual diffusivities have also been evalu-ated for PCL-CO2 and PCL-N2 systems, with an expected higher mutual diffusivity for the system PCL-N2 with respect to PCL-CO2. These differ-ences in the sorption behavior and the associated differences in the plasti-cization effect of the two gases on the polymer, as reported in Reference 15, are responsible for the very different foaming behavior of the two different polymer-gas systems.

Figure 5.2a reports a micrograph of a PCL sample foamed at 35°C with a rather slow pressure drop rate, after being saturated with CO2 at 55 bar and 70°C. In these experimental conditions, the resulting foam showed a rather coarse morphology with mean cell size of 0.5mm and a density of 0.05g/cm3.15 Similar results were obtained by Jenkins et al.,16 Xu et al.,17 Reignier et al.,18 and by Cotugno et al.19 Figure 5.3 summarizes the morphologies achieved by the different research groups in foaming PCL with CO2.

The indicated temperature would let the reader think that the process-ing window for PCL foaming with CO2 is indeed wide (more than 40°C).

0.25

0.2

0.15

Gas

mas

s fr

actio

n

0.1

0.05

00 5 10 15

Pressure (MPa)

CO2, 70°C

CO2, 80°C

CO2, 90°C

N2, 75°C

20 25 30

FIGURE 5.1 Equilibrium concentrations of CO2 and N2 in PCL at different temperatures.

146 Polymeric Foams

61259_C005.indd 14661259_C005.indd 146 10/25/2008 12:37:29 PM10/25/2008 12:37:29 PM

800Xu et al. 40°C (17)

Jenkins et al. 65°C (16)

Di Maio et al. 35°C (15)

Cotugno et al. 30°C (19)



700

600

500

400

300

200

100

05 10 15 20

Saturation pressure (MPa)

Mea

n ce

ll di

amet

er (

μm)

25 30 35

FIGURE 5.3 Variation of the mean pore dimension of PCL foamed with CO2. See cited

references for experimental details.

FIGURE 5.2 SEM micrograph of PCL foam: (a) with CO2 at 35°C, 10 bar/s; (b) with CO2 at

35°C, 30 bar/s; (c) with N2 at 43.3°C; (d) with 80/20% volume mixture of N2 and CO2 at

43.3°C.


61259_C005.indd 14761259_C005.indd 147 10/25/2008 12:37:29 PM10/25/2008 12:37:29 PM

Actually, the experimental apparatuses used for the preparation of the foams by the different research groups are rather different, mostly in the control of temperature, the cooling protocols, and in the gas evacuation system. This results in very different temperature histories in the batch experiments that do not allow a direct comparison among all of the results published so far. For this reason we would say that, for each single appa-ratus, the useful processing window is restricted to less than 10°C.

As can be observed in Figure 5.3, with an increase of the saturation pressure foams with fi ner morphologies compared to that reported in Figure 5.2a can be obtained. As extensively reported in the literature, fi ner morphologies can also be obtained by increasing the pressure drop rate. As an example, Figure 5.2b reports a micrograph of a PCL sample foamed at 35°C with a faster pressure drop rate, after being saturated with CO2 at 55 bar and 70°C.

The specifi c design of the experimental apparatus represents, com-monly, the fi rst limitation to the increase of the pressure drop rate to fur-ther reduce the mean cell diameter of the foams. This is also the case for industrial equipment. In order to modify the foam morphology, therefore, it is possible to change the blowing agent, as reported in Reference 15. In effect, foams with cellular structures characterized by fi ner morphologies have been obtained with N2 (see Figure 5.2c) by using the same experi-mental apparatus used for the foams reported in Figures 5.2a and b. In this case, however, the lower solubility of N2 with respect to CO2 led to foams with a higher density. A better compromise between cell morphol-ogy and density has been obtained by using mixtures of CO2 and N2 as blowing agents. In this case, small amounts of CO2 supply gas to infl ate the high number of bubbles generated by N2, to achieve foams with a very fi ne morphology and low density at the same time (see Figure 5.2d).

A detailed analysis of the decoupled effect of the main processing vari-ables, the blowing agent concentration, the pressure drop rate, and the foaming temperature has been reported by Marrazzo et al.20 on PCL foam-ing with N2 in a fairly restricted processing range (Psat � 140–200 bar, Tfoam � 44.5–48.5°C and �P � 260–380 bar/s) by using a batch foaming apparatus designed to independently control the three process variables and defi ne a foaming protocol.

In contrast to the batch process, the continuous extrusion process is important to achieve industrial productivity. Until now, little interest has been given to biodegradable polyesters, mostly because they are not con-sidered to be foamable in general, due to poor properties of the melt and reduced setting mechanisms. In fact results of the batch foaming sug-gested that it is possible to overcome these problems by controlling and improving the whole foaming process. The extrusion foaming is charac-terized by the same intense collapsing, density, and morphology issues as in the case of batch foaming and the methods to be used to improve the quality of extruded foams are analogous.

148 Polymeric Foams

61259_C005.indd 14861259_C005.indd 148 10/25/2008 12:37:30 PM10/25/2008 12:37:30 PM

Experimental tests, performed on lab-scale foam extrusion equipment on a PCL/CO2 system led to foam characterized by a density of 0.05 g/cm3 and mean cell diameter of 200 μm, with the following temperature profi le from the hopper to the die: 90, 100, 100, 100, 70, and 40°C. Higher unifor-mity of cell structure but higher densities were achieved with N2 as foam-ing agent, by using Tdie� 45°C. The minimum mean cell diameter was 50 μm with a density of 0.3 g/cm3. As in the case of the batch process, the best results in term of morphology and density were achieved using as foaming agent in the extrusion a 20–80%wt CO2–N2 mixture: foams were obtained characterized at the same time by a low density (0.15 g/cm3) and a fi ne morphology (average cell diameter, d � 20 μm).15

5.2.2 Foaming of PLA

PLA is a bio-based biodegradable aliphatic polyester with chemical formula -[-O-(CHCH3)-CO-]n-. It is widely used in biomedical applica-tions and is now fi nding commercial use for disposable items. PLA is gen-erally produced by the ring-opening polymerization of lactide, a cyclic dimer prepared by the controlled depolymerization of lactic acid. Lactic acid can be manufactured by either a chemical synthesis or a carbohydrate fermentation from renewable natural resources such as corn starch, sugar cane, and sugar beet. Lactic acid polymers consist of mainly lactyl units, of only one stereo-isoform (PLLA and PDLA) or combinations of D and L lactyl units in various ratios (PDLLA). Commercial materials have, in gen-eral, low D content (up to 10%). The L-lactic acid based polymers (PLLA) may produce polymer which is a linear homopolymer of molecular size 70kDa.21 Although crystalline PLLA possesses many desirable proper-ties (physical and mechanical), crystallization rates are extremely slow. The fastest half-time of crystallization reported in the literature for pure PLLA is 1.9 min for a 101 kg/mol sample.22 As already mentioned, this aspect limits the use of PLLA in foaming as well as in the other thermo-plastic processing.

Sorption experiments on the system PLA/CO2 have been performed by the Ohshima group at 200°C and up to 14 MPa. The authors report a rather linear isotherm with an equilibrium CO2 concentration of 0.05% at 10 MPa and mutual diffusivities in the range 2–5 × 105 cm2/s.23

PLA foams have been prepared by different research groups with the temperature increase method and with the pressure quench method, in a wide range of processing conditions. The two methods differ in the way the polymer-blowing agent solution is brought out of equilibrium, to induce the formation of the new gas phase (bubbles). In fact, during foaming the polymer is fi rst saturated with the blowing agent at high temperature and pressure and then, to induce foaming, it is brought to super-saturation conditions, in which the gas prefers to run off the solution. Since, in general, the blowing agent solubility decreases with an increase in temperature


61259_C005.indd 14961259_C005.indd 149 10/25/2008 12:37:30 PM10/25/2008 12:37:30 PM

and/or a decrease in pressure, two methods can be used: the temperature increase (TI) and pressure quench (PQ). Generally, the former allows the achievement of ultra-fi ne morphologies: nanometric to micrometric celled PLA foams have been prepared in the temperature range 50–140°C with CO2. In particular, foams with a mean cell diameter of 200 nm have been obtained by heating CO2-saturated PLA (at 5.5 MPa and 25°C) to 60°C.24 At higher foaming temperatures, morphologies characterized by bigger pores have been obtained. For example, the Okamoto group obtained foams with a mean diameter of 230 μm at 140°C on CO2-saturated PLA (at 10 MPa and approximately 150°C).8 The temperature increase method, as with other thermoplastics, led to foams of relatively high density. The pressure quench method, on the contrary, allows the achievement of lower density foams, characterized by coarser structures. Mathieu et al. foamed PLA saturated with CO2 at pressures ranging from 15 to 21 MPa and at 195°C, obtaining foams with mean cell diameters ranging from 200 to 1000 μm. The selected foaming temperature was 195°C and pressure drop rates were selected in the range 0.2–1.2 MPa/s.25 Ema et al. foamed CO2-saturated PLA, obtaining foams with densities as low as 0.1 g/cm3 and mean cell diameter of 60 μm, at foaming temperature of 150°C.26 Di et al. foamed PLA at 110°C after saturating the polymer with an 80/20%vol N2/CO2 mixture, obtaining foams with a density of 0.125 g/cm3 and mean cell diameter of 230 μm.27 Figure 5.4 summarizes the described results.

As already observed in describing PCL foaming, from Figure 5.4 it seems that the processing window of PLA is rather large, with the possi-bility of achieving a wide range of morphologies and densities. This is, again, not true: for each single apparatus and processing technique, the

FIGURE 5.4 Variation of the mean pore dimension of PLA foams. See cited references for

experimental details; PQ: pressure quench method; TI: temperature increase method.

500

200

400

600

800

1000

100

Foaming temperature (°C)

Mea

n ce

ll di

amet

er (

μm)

150 200

Mathieu et al. PQ (25)Ema et al. PQ (26)Di et al. PQ (27)Fujimoto et al. TI (8)Liao et al. TI (24)

150 Polymeric Foams

61259_C005.indd 15061259_C005.indd 150 10/25/2008 12:37:30 PM10/25/2008 12:37:30 PM

useful processing window is very narrow and, with respect to PCL, PLA foaming is even more critical, for the poor rheological properties and the very slow crystallization kinetics.

In the following paragraphs, two methods utilized to enhance foam-ability of PCL and PLA will be described. In particular it will be shown how the design of the molecular architecture targeted to foaming and the addition of nucleating agents (micro- and nano-sized) may help foam stabilization, widen the processing windows, and reduce cell coalescence.

5.3 Molecular Modification of PCL and PLA

5.3.1 Branching/Cross-Linking of PCL

The rheology of melts is much affected by molecular weight, molecular weight distributions (MWD) and the presence of short (SCB) and long chain branching (LCB). In the case of polyolefi n melts, literature results mostly indicate that the effect of polydispersity on viscosity, elastic char-acter, and activation energy of fl ow could be very similar to that expected due to the presence of LCB. That notwithstanding, the effects of LCB seem to be stronger than those due to polydispersity for a given molecular weight. Different relaxation processes appear as a consequence of the presence of LCB: slower terminal relaxation behavior than that of linear counterparts, and a faster additional branch relaxation (gel-like behavior) at higher frequencies, clearly distinguishable from polydispersity effects. Moreover, the branch number contributes less to the rheological behavior with respect to the topology of the branched polymers. Thus, branch posi-tion and architecture along the main polymer chain are the main factors controlling the viscosity function, elastic character and activation energy of fl ow.28

Within the polyolefi n group, the polyethylene (PE) family is a good example of how it is possible to tailor the rheological properties, including those in extensional fi elds, by optimizing the molecular architecture, thanks to the evolution of the industrial polymerization of PE in line with market needs.

The molecular architecture of linear thermoplastic polymers may be converted to branched/cross-linked polymers by promoting free-radical reactions induced by incorporating peroxides or by solid-state irradiation techniques. Both methods yield carbon–carbon cross-links, where cross-linking by irradiation takes place in the solid state and peroxide cross-links occur in the molten state. Irradiation cross-linking is therefore a selective process that takes place mainly in the amorphous region, whereas in peroxide cross-linking the polymer is totally amorphous and thus structurally non-selective to the reaction with the peroxide radicals.29


61259_C005.indd 15161259_C005.indd 151 10/25/2008 12:37:30 PM10/25/2008 12:37:30 PM

Radiation effects in saturated linear polyesters depend on the specifi c chemical structure of the polymer. For example, biodegradable polyesters such as poly(glycolic acid) (PGA) and PLA are known to cleave and cross-link. PGA degrades and cross-links at about the same rate30,31 whereas PLA predominantly degrades.32

It has already been reported that PCL does not cross-link well by irradi-ation, due to the occurrence of chain scission during treatments, while it cross-links effi ciently with peroxide in the molten state.33 The viscoelastic properties of PCL can be modifi ed through the branching of the poly-meric matrix using dicumyl peroxide (DCP), which was reported to gen-erate polymeric radicals by hydrogen abstraction reactions.29 Polymeric radicals immediately react to generate branching and cross-linking and the resulting molecular modifi cation can be easily followed by measuring the dynamic rheological properties of the melt in isothermal experiments.

As reported in Reference 34, the polymer was fi rst blended with dicu-myl peroxide at a low temperature (80°C) in order to prevent premature peroxide decomposition. The peroxide modifi cation was then performed at different temperatures, from 110°C to 150°C. The reaction kinetics was followed by measuring the dynamical rheological properties of the melt in isothermal experiments by using a parallel plate rheometer. The evolu-tion of the macromolecular structure during the chemical reaction was followed by analyzing the time evolution of the real and imaginary com-ponent of the complex viscosity. As reported in Reference 35 the measure-ment of the rheological properties allows the analysis of the branching reaction extent, even at very low branching levels.

The degree of chemical modifi cation, analyzed by measuring the increase of storage modulus during dynamic rheological experiments in isothermal conditions (see Figure 5.5), can be controlled by: (a) DCP con-tent, (b) reaction time, and (c) temperature. As expected, the reaction rate of branching increases when the curing temperature increases. Moreover, the fi nal degree of branching is correlated to the value of storage modulus (G�) at the plateau, and increases with the amount of DCP.

The degree of molecular modifi cation and, in particular, the level of branching/cross-linking, resulted in a strong change of the levels of elas-tic and dissipative components and a strong variation of the dependence of these components upon frequency. Figure 5.6 reports the complex vis-cosity and tan d versus frequency of samples fully cured at 130°C and at different DCP concentrations. Even at very low DCP content, the complex viscosity increases rapidly, particularly at the lower frequencies, revealing a reduced Newtonian behavior.

The analysis of the dependence of tan d versus frequency allows assess-ing the occurrence of gelation phenomena, characterized by constant val-ues of tan d versus frequency.36 As shown in Figure 5.6 tan d decreases with frequency at concentration of DCP below 0.5% and it is constant or increases with frequency at higher concentrations of DCP. These results

152 Polymeric Foams

61259_C005.indd 15261259_C005.indd 152 10/25/2008 12:37:30 PM10/25/2008 12:37:30 PM

are in agreement with the literature data29 where it has been shown that some gel content is present only for concentrations higher than 0.5% and it increases with the amount of DCP. The authors have reported that a gel content of 35% can be obtained at 1% of DCP and around 60% at 2% of DCP (180°C in a Brabender).

The chemically modifi ed materials were foamed in a batch apparatus using the following procedure. PCL samples were saturated at 75°C with N2 at saturation pressure, Psat � 180 bar, for at least 6 hours. The vessel was then cooled to different foaming temperatures, Tfoam, in the range 40–55°C. The pressure drop rate was �P � 320 bar/s. A more detailed description of the experimental conditions has been provided in Reference 34. Table 5.1 reports the effect of the foaming temperature on the fi nal foam density for both neat PCL and peroxide-modifi ed PCL.

Results show that the density of neat PCL decreases with the increase of temperature from 40°C to 45°C and then it increases at higher temperatures.

FIGURE 5.5 Viscoelastic properties of PCL modifi ed with DCP: (�) cured at 140°C;

(�) cured at 130°C; (�) cured at 110°C. Solid lines: DCP content of 1%; dashed lines: DCP

content of 0.25%.

1.2 × 105

1 × 105

8 × 104

6 × 104

G′ (

Pa)

G′ (

Pa)

4 × 104

3 × 104

104

8 × 103

6 × 103

4 × 103

2 × 104

0 × 100

0 × 100 5 × 103 1 × 104

time (s)1.5 × 104 0 × 100 5 × 103 1 × 104

time (s)1.5 × 104

FIGURE 5.6 Viscoelastic properties of fully cured PCL–DCP samples; effect of DCP

content.

105

104

h* (

Pa·

s)

tan

d

10–1 100

w (rad/s)101 10–110–1

100

100

w (rad/s)101

101% DCP

2

1

0.5

0.25

0

% DCP

210.50.25

0


61259_C005.indd 15361259_C005.indd 153 10/25/2008 12:37:30 PM10/25/2008 12:37:30 PM

This behavior is due to several mechanisms taking place during foaming. At a high temperature, cell collapse occurs due to the decrease of the melt strength of the polymeric matter, leading to foams of higher density. At lower temperatures, higher densities are related to an increase of vis-cosity and/or to the occurrence of partial crystallization of the polymeric matrix. Both the increase of viscosity and the occurrence of crystallization reduce the deformability of the expanding matter and thus the foaming effi ciency.

When using the peroxide cured PCL, the increase of viscosity due to the molecular modifi cation led to foams with higher densities (see Table 5.1) at lower temperatures. As reported in Table 5.1, chemically modifi ed PCL can be foamed at higher temperatures and lower densities, compared to unmodifi ed PCL, can be obtained. Conversely, foams prepared with chemically modifi ed PCL at temperature of 55°C showed neither coales-cence nor cell wall rupture. Moreover, the morphologies of these foams are characterized by uniform cellular structure and fi ne cell dimensions. This is related to the enhancement of the viscoelastic properties that reduce the cell wall rupture and the collapse of the cellular structure.37 Neat PCL foams, conversely, at the same temperature have collapsed, as evidenced by the higher density and by the poor morphology.

5.3.2 Chain Extended PLA

PLA presents several limitations such as low thermal, oxidative, and hydrolytic stability that leads to a decrease of molecular weight during processing. The melt viscosity and, in general, the rheological properties of the melt in both shear and elongational fi elds are therefore often compromised by these degradation processes. To overcome such short-comings, considerable research efforts have been devoted to control the melt rheology by increasing the molecular weight to compensate for the molecular weight decrease caused by processing degradation. Several approaches were followed and among them we should mention the

TABLE 5.1

PCL and Modifi ed PCL Foam Densities

Foaming

Temperature (°C) Pure PCL (g/cm3) 0.5%wt DCP (g/cm3) 1%wt DCP (g/cm3)

40 0.0688 — —

41 0.0538 — —

45 0.0338 0.368 0.271

49 0.0372 0.0691 0.0791

53 0.0793 — —

55 0.181 0.0508 0.0576

154 Polymeric Foams

61259_C005.indd 15461259_C005.indd 154 10/25/2008 12:37:30 PM10/25/2008 12:37:30 PM

free-radical branching/linking with peroxide as radical initiators,38–41 the solid-state post-polymerization technique,42 the chain extension tech-nique.27,43–46 Moreover, among the most recent work on the use of chain extension reactions we should mention the works by Vollalobos et al. on the use of novel oligomeric and low molecular weight polymeric chain extenders based on epoxy-functional (meth)acrylic monomers and non-functional (meth)acrylic and/or styrenic momomers47,48 and on the use of isocyanates compounds patented by Unika Ltd.49

It has been recently reported that 1,4-butanediol (BD) and 1,4-butane diisocyanate (BDI) can be used to modify commercial PLA in two steps. In the fi rst step, BD was selected as the fi rst coupling agent and acid value reducer to link carboxyl groups of PLA and then, in the second step, BDI was added to let it react with hydroxyl end groups of PLA to achieve chain-extended PLA. The different ratios of two chain extenders were used to investigate their effect on the structure of modifi ed PLA samples which were then characterized and foamed in a batch foaming process. In particular, three modifi ed PLA samples were obtained by using differ-ent ratios of BD to BDI: modifi ed sample 1 (PLAM1): COOH/BD � 2 : 1, OH/BDI � 2 : 1, that is, equimolar amount of BD and BDI relative to end groups COOH and OH of PLA, where COOH and OH contents were calculated from the acid values determined by titration; sample 2 (PLAM2): COOH/BD � 2 : 1, OH/BDI � 1 : 1, that is, BDI amount was excessive compared to M1; sample 3 (PLAM3): COOH/BD � 1 : 1, OH/BDI � 2 : 1, that is, BD was excessive compared to PLAM1. The original PLA was also processed at the same conditions for comparison purposes and is named neat PLA hereafter. Details on the materials and chemical modifi cation procedures are reported in Reference 27.

Table 5.2 reports GPC results of plain and modifi ed PLA. It is clearly shown that molecular weight of samples PLAM1 and PLAM2 were increased compared to plain PLA because of chain extension. The excess of DBI in sample PLAM2 resulted in materials with the highest molecular weight (Mw) and larger molecular weight distribution (MwD) while the excess of BD (PLAM3) resulted in materials with lowest Mw and even larger MwD compared to plain PLA.

The increase of Mw and of polydispersity in samples PLAM1 and PLAM2 resulted in materials whose rheological properties were charac-terized by higher viscosity and higher elasticity.27,50 While PLA exhibited typical Newtonian behavior at low frequency and a small shear thinning starting at a frequency of 10 rad s�1, chemically modifi ed materials (both PLAM1 and PLAM2) showed non-Newtonian behavior even at very low frequency (see Figure 5.7). The increase of complex viscosity at lower fre-quency and the lower slope in G� in the same frequency range is the evidence of the presence of gel-like structures coming out from cross- linking reactions taking place during the chemical modifi cation of PLA. It is reasonable to expect that, due to the excess of BDI, the isocyanate groups


61259_C005.indd 15561259_C005.indd 155 10/25/2008 12:37:30 PM10/25/2008 12:37:30 PM

can react with carboxyl groups of PLA leading to amide bond that can further react with additional BDI and cause cross-linking in PLA.27 The improved viscous and elastic properties of modifi ed PLA samples resulted in foams of lower foam density and higher cell density, as reported in Table 5.2.

5.4 Nanocomposites

Nanometric additives have been reported to dramatically change proper-ties such as gas solubility and diffusivity, rheological characteristics, and

FIGURE 5.7 Viscoelastic properties of pure PLA and modifi ed PLA; closed symbols G¢; open symbols h*. (●) neat PLA; (■) M1; (▲) M2.

106

105

104

103

102

106

105

104

103

102

101

100

10–1

0.01 0.1 1

Frequency (rad/s)

h* (

Pa·

s)

G′ (

Pa)

10 100

TABLE 5.2

Characteristics of Pure and Modifi ed PLA

Sample Pure PLA M1 M2 M3

Mn/103 (g/mol) 57 84 107 48

Mw/103 (g/mol) 124 225 308 156

Mw/Mn 2.2 2.7 2.9 3.3

Tg (°C) 61.8 63.2 63.7 55.7

Tc (°C) 111.5 129.1 130.2 113.8

Tm (°C) 169.6 154.5 153.8 156.3

Average cell size (μm) 227 37 24 223

Cell number density (108/cm3) 0.008 1.9 6.7 0.008

Foam density (g/cm3) 0.125 0.067 0.092 0.179

156 Polymeric Foams

61259_C005.indd 15661259_C005.indd 156 10/25/2008 12:37:30 PM10/25/2008 12:37:30 PM

crystallization, which are extremely important during foaming. In the industrial practice of thermoplastic foaming, furthermore, it is of vital importance the use of nucleating agents, additives which induce the het-erogeneous nucleation of bubbles, with the aim of controlling the fi nal morphology of the foam.8,51 In this section, the effects of nanofi llers in modifying the relevant properties for foaming (mass transport phenomena, rheology, and crystallization) as well as their role in bubble nucleation is reviewed for PCL and PLA-based nanocomposites.

5.4.1 Nanocomposites from PCL and PLA: Rheology, Sorption, Mass Transport, and Crystallization

Several papers, reviews, and patents have been published recently in the scientifi c literature, following the pioneering work from the Toyota Central Research Group52 on the use of nanometric additives to improve certain properties of the polymer (mainly mechanical) at very low con-centration, with respect to conventional (micrometric) fi llers. These con-cepts were also applied in the area of biodegradable polymers and there are now numerous reviews on nanocomposites based on biodegradable polymers.53–55 Specifi c literature on PCL and PLA has also been presented, reporting on the preparation and properties of nanocomposites from different materials such as clays, organo-modifi ed clays, titanate, hydroxy-apatite, mica, saponite, and smectite, produced via melt mixing, solvent casting, or in-situ polymerization. Intercalated and exfoliated structures have been obtained with improved mechanical, rheological, transport, and thermal properties.56–65

In the analysis of rheological properties of these systems, the character-istic features of nanocomposites; for example, the deviation from terminal fl ow behavior and pseudo-solid-like behavior, have been observed by Lepoittevin et al. on PCL (of an average molecular mass of 49,000) melt mixed with a montmorillonite (MMT) modifi ed by methyl bis(2-hydroxy-ethyl) ammonium cation, MMT-(OH)2 and PCL melt mixed with a MMT modifi ed by dimethyl 2-ethylhexyl ammonium cation (MMT-Alk).66 Authors, to show the importance of organo-modifi cation for achieving intercalation and exfoliation, also reported the rheological properties of unmodifi ed MMT-Na, which resulted in micro-composites with unchanged rheological properties with respect to pure PCL. Di et al. melt mixed PCL (of an average molecular mass of 69,000) with MMT-(OH)2 and a MMT modifi ed by methyl dihydrogenated tallow ammonium (MMT-2HT), reporting exfoliated structures only with the former organo-modifi er.12 In order to give a picture of the extent of the rheological modifi cation induced by the nanometric particles, complex viscosities versus frequencies are reported in Figure 5.8 for PCL-MMT-(OH)2 nanocomposites. The h*�s of the pure molten polymer show only a small frequency dependence, reveal-ing a Newtonian plateau at low frequency. The h* of the nanocomposites


61259_C005.indd 15761259_C005.indd 157 10/25/2008 12:37:31 PM10/25/2008 12:37:31 PM

are higher overall than that of the neat polymer within the frequency range studied and h* increases with the organoclay content.

Figure 5.9 reports G� and G�� versus frequencies for PCL-based nanocom-posites. Typical patterns of exfoliated nanocomposite systems, with solid-like behavior and the presence of a yield stress can be observed.12,67–69

Similar results have been obtained on several nanocomposites based on PLA. Ray and Okamoto,70 for example, reported the rheological properties of PLA melt-mixed with MMT, organically modifi ed with octadecyl ammonium cation (C18-MMT) and observed non-terminal fl ow and pseudo-solid behavior at concentrations of 3–7 wt%. Similar results have been

FIGURE 5.8 Complex viscosity curves of PCL and PCL-MMT-(OH)2 nanocomposites.

1000

104

h* (

Pa·

s)

105

106

107

0.1 1 10 100

Pure PCL1%2%3.5%5%7%10%

w (rad/s)

90°C

FIGURE 5.9 Viscoelastic properties of neat PCL and PCL/clay nanocomposite, 80°C.

G¢ (open symbols), G≤ (closed symbols). (●) neat PCL; (▲) 2%wt clay; (■) 10%.

10

100

1000

Mod

uli (

Pa) 104

105

106

0.001 0.01 0.1 1

w (rad/s)

158 Polymeric Foams

61259_C005.indd 15861259_C005.indd 158 10/25/2008 12:37:31 PM10/25/2008 12:37:31 PM

obtained by Di et al.71 on PLA nanocomposites with 2–10 wt% MMT-(OH)2 and by Pluta72 on the same system at 3 wt%. Typical curves are reported in Figure 5.10.

Ray and Okamoto,70 also performed elongational rheology experiments and observed the rise of a strain-induced hardening for the nanocompos-ites with 5 wt% C18-MMT, despite the very low rheological properties of pure PLA. As will be seen in the following, these improvements are crucial for foaming.

Less numerous, but important in foaming, are the studies on mass transport and sorption thermodynamics of gases in nanocomposite sys-tems. In particular, gas barrier properties have been shown to improve dramatically upon exfoliation of clay platelet in a number of polymeric systems.73–78 In our systems, air permeability was reduced by 50% with the addition of 5 wt% MMT-(OH)2 to PCL,79 while O2 permeability is reduced by as much as 70% when modifi ed fl uorine mica was added to PLA.80 Figure 5.11 reports the mutual diffusivity of CO2 and PCL and its nano-composites with MMT-(OH)2, evidence of the decrease of diffusivity with the increase of clay content, as already observed. The mechanism of improvement is attributed to the increase in the tortuosity of the diffusive path for a penetrating molecule, this effect depending on the aspect ratio of the additives, their concentration, and dispersion. In fact, the observed reduction in diffusivity at 10 wt% loading was less pronounced than expected, being slightly below the diffusivity at 5 wt% loading. This behav-ior can be related to the poorer dispersion and the lower degree of exfolia-tion of clay platelets that is obtained at a concentration of 10% compared to 5%. In effect, when the concentration of clay is 10%, the peak of the organo-modifi ed clay in the X-ray diffractograms did not disappear, as in the case of

FIGURE 5.10 Complex viscosity curves of PLA and PLA/MMT-(OH)2 nanocomposites

with different weight fractions of MMT at 170°C.

106

10 wt%

5 wt%

2 wt%

0 wt%

105

104

h* (

Pa

s)

103

10–3 10–2 10–1

w (rad/s)100 101 102


61259_C005.indd 15961259_C005.indd 159 10/25/2008 12:37:31 PM10/25/2008 12:37:31 PM

5%, but shifted to a lower angle suggesting that the nanoparticles were intercalated and not fully exfoliated.12

Figure 5.12 reports the equilibrium concentration of CO2 in PCL and PCL-MMT-(OH)2 nanocomposites, with respect to the polymeric matrix, evidence of a slight increase in solubility in the presence of nanometric particles, which can be justifi ed by the introduction of defects and micro-scopic gas volumes close to the polymer/clay interfaces. In effect, it has

FIGURE 5.11 Mutual diffusivity as a function of CO2 concentration in pure PCL and its

nanocomposites with MMT-(OH)2.81 (From Cotugno, S. “Modeling of Thermodynamic and

Transport Properties of Polymer Melts and Solutions to be used in the Simulation of

Foaming Processes.” PhD thesis, University of Naples Federico II, Italy, 2003.)

8 × 10–5

7 × 10–5

6 × 10–5 T = 70°C

Pure PCLNano-PCL 5%Nano-PCL 10%

5 × 10–5

4 × 10–5

3 × 10–5

2 × 10–5

1 × 10–5

00 50 100

mgCO2/g polymer

Mut

ual d

iffus

ivity

(cm

2 /s)

150 200 300250

FIGURE 5.12 Sorption isotherms for pure PCL and its nanocomposites with MMT-(OH)2.

T = 70°C

Pure PCLnc 5%nc 10%

00

50

50

100

100

Pressure (bar)

mgC

O2/

g PC

L

150

150

200

200

300

300

350

250

250

160 Polymeric Foams

61259_C005.indd 16061259_C005.indd 160 10/25/2008 12:37:31 PM10/25/2008 12:37:31 PM

been observed that the concentration of the blowing agent might not be uniform throughout the polymer matrix but may be higher in some domains close to the solid particles.

Thermal properties are also important in foaming, as already observed, since they take part in the setting mechanism of a newly formed cellular structure. This is particularly true for semi-crystalline polymers such as PCL and PLA. It has been extensively reported that nanocomposites added to polymeric materials favor crystallization, causing signifi cant modifi -cation of the thermal behavior of the polymeric matrix. In Figure 5.13 we compare the crystallization process for neat PCL and PCL-based nanocomposites.

The presence of the nanometric fi ller results in a relevant increase of the crystallization temperature during cooling. The same effect has been observed with PLA-based nanocomposites, with a faster and more intense crystallization with respect to the pure polymer.71 This phenomenon has been classically described as a heterogeneous nucleation effect of the solid particles for the polymeric crystals. Heterogeneous nucleation was observed in several polymers such as polyethylene,82 polyamide 6,83 polyamide 6,6,84 polyamide 12,12,85 polypropylene,86,87 and syndiotactic polystyrene.88

In foaming, the easier crystallization will result in an easier stabilization of the cellular structure and an improved foamability. As a counterpart anyway, a possible increase in the foam density can occur due to prema-ture crystallization during foaming at low temperature.

FIGURE 5.13 Non-isothermal, melt crystallization experiments of pure PCL and PCL/clay

nanocomposites.

0

5

10

Hea

t flo

w (

mW

)

15

20

10 15 20 25 30 35 40

Pure PCL1%wt clay2%wt clay

10%wt clay

5%wt clay

T (°C)


61259_C005.indd 16161259_C005.indd 161 10/25/2008 12:37:32 PM10/25/2008 12:37:32 PM

5.4.2 Foaming of PCL and PLA-Based Nanocomposites

The improved elastic and viscous properties of nanocomposites have encouraged researchers to study their foaming process. Figure 5.14 reports, as an example, the SEM images of PCL-based nanocomposites with MMT-(OH)2 (neat PCL, 1 wt% and 0.4 wt%, 16a, b, and c, respectively, foamed at 40°C with an 80/20% volume mixture of N2 and CO2).

It is evident that the morphologies of the foams obtained from the pure polymer and its nanocomposites are rather different, with a fi ner mor-phology of the nanocomposites, characterized by smaller and uniformly distributed cells (almost a one-order-of-magnitude reduction in the mean cell diameter). The density of the 1% nanocomposites foam is higher than the one of the neat PCL (0.07 g/cm3 instead of 0.04 g/cm3 for the neat PCL). This was due to the increased viscoelasticity and crystallization rate of the nanocomposites with respect to the neat PCL. By comparing Figure 5.14c with Figure 5.14a and b, an interesting effect can be noticed: at 0.4 wt% of clay, an open-cell, very fi ne structure was achieved. This behavior has been ascribed to the crystalline phase nucleation and bubble nucleating effect of the fi ller. In fact, the very low clay concentration (0.4 wt%) has been reported to nucleate the crystalline phase, resulting in a steep increase of the crystallization kinetic, with respect to the pure PCL and to higher clay concentration. The occurrence of crystallization increases the strength while reducing the deformability of the expanding matter, hence an open-cell, relatively high density (0.12 g/cm3) structure was achieved.89

Similar results have been achieved with PLA. Figure 5.15 reports the SEM images of foamed PLA and PLA nanocomposites. We can see that, under the same foaming condition, fi nely dispersed cells were formed in foams of PLA and PLA nanocomposites.

Table 5.3 summarizes the characteristics of PLA-based foams. We note that the average cell size of pure PLA foam is very large, while it decreases signifi cantly with 1 wt% of clay, and then levels off at higher MMT-(OH)2 concentration. The cell densities increase steeply with the presence of organoclay [from 0.008 ¥ 108 cell/cm3 for pure PLA to 5.1 ¥ 108 cell/cm3 for PLA/5 wt% MMT-(OH)2]. The bulk foam densities were measured and found to be signifi cantly affected by the clay content.

FIGURE 5.14 SEM micrographs of (a) neat PCL; (b) nanoclay PCL (1%); and (c) nanoclay

PCL (0.4%).

162 Polymeric Foams

61259_C005.indd 16261259_C005.indd 162 10/25/2008 12:37:32 PM10/25/2008 12:37:32 PM

The outlined results are common to several nanocomposite systems, like in PP/clay, as reported in recent work from the Toyota Technical Institute, in which up to two orders of magnitude increase in the cell num-ber densities have been measured.90,91 The same group also reported suc-cessful results on a polycarbonate/fl uoroectorite system.51 Other examples include polystyrene,92–95 polyamides,96 polyurethanes,97 and high-density polyethylene.98

To summarize, the benefi cial effect of nanometric fi llers in thermoplastic foaming can be ascribed to (1) the enhancement of the strength of the poly-mer melt to prevent melt fracture and foam collapse during bubble growth; (2) the enhancement of the crystallization kinetics to facilitate foam setting immediately after the growth; and (3) the induction of the formation of numerous bubbles. This aspect of crystal nucleation induction and of bubble nucleation induction is further analyzed in the next paragraph.

5.4.3 Bubble versus Crystal Nucleating Effect of Nanoparticles

The effect of solid particles in foaming of a semi-crystalline polymer can be considered to be of a dual nature: from one side the particles can take

FIGURE 5.15 SEM micrographs for foams of (a) PLA and (b) PLA nanocomposite with

2 wt% MMT-(OH)2 foamed at 110°C with an 80/20% volume mixture of N2 and CO2.

TABLE 5.3

Average Cell Size, Cell Density and Densities of PLA Nanocomposite Foams with Different Weight Fractions of MMT-(OH)2. The Averages were Calculated from Experimental Observations on Five Different Samples Having Densities in the Reported Range

Sample

Average Cell

Size (μm)

Average Cell

Density (108/cm3) Foam Density (g/cm3)

PLA 230 0.008 0.10–0.14

PLA/1 wt% MMT-(OH)2 42 1.098 0.18–0.22

PLA/2 wt% MMT-(OH)2 32 2.38 0.23–0.35

PLA/5 wt% MMT-(OH)2 25 5.18 0.30–0.42


61259_C005.indd 16361259_C005.indd 163 10/25/2008 12:37:32 PM10/25/2008 12:37:32 PM

part in the nucleation of polymeric crystals and induce, increase or fasten crystallization; from the other side the particles can take part in the nucle-ation of bubbles and induce an improvement of the foam morphology. The two-nucleation phenomena of crystallization and bubbling should not be, in principle, correlated. In particular, the effi ciency of the different nucle-ating agents, which is the main interest in industry, in nucleating crystals or bubbles, can be different and selective nucleation towards crystals or bubbles can be observed. We have recently analyzed this problem and studied the effect of different micrometric and nanometric nucleants in the foaming behavior of PCL.99

Nanoparticles from different materials with different shapes and/or characteristic dimensions and a traditional micrometric nucleating agent (talc) have been melt mixed with PCL and subsequently batch foamed with nitrogen and carbon dioxide. The main characteristics of the nucleat-ing agents are reported in Table 5.4.

The isothermal crystallization of PCL nanocomposites analyzed by dif-ferential scanning calorimetry (Figure 5.16) show the different abilities in inducing the formation of the crystalline phase (crystal nucleation) by dif-ferent nucleating agents. All the nanocomposites investigated showed a faster crystallization rate compared to neat PCL. In particular, nanocom-posites based on carbon nanotubes showed the highest crystallization rate while those based on alumina spherical nanoparticles the smallest.

The immediate consequence of these changes in the crystallization rates is that the polymer can experience premature crystallization phenomena during the cooling from the temperature at which gas is solubilized to the temperature at which the material is foamed. To better explain this relationship between foaming and crystallization, it is possible to ideally

TABLE 5.4

List of the Nucleating Agents used in this Study, from Producer’s Datasheets

Trade

Name Typology Shape

Characteristic

Dimension

Aspect

Ratio

Surface

Area

(m2/g)

Talc Finntalc

M03

Talc Cubic 1 μm 1 13.5

Alumina Aeroxide

AluC

Al2O3 Spherical 13 nm 1 15–100

130F Hombitec

RM130F

TiO2 Spherical 15 nm 1 15–100

30B Cloisite

30B

Modifi ed

mont-

morillonite

Platelet Exfoliated 100 (width/

thickness)

750

cn Aldrich

636509-2G

multi-walled

carbon

nanotubes

Tube 15 nm 20–5000

(length/O.D.)

40–600

164 Polymeric Foams

61259_C005.indd 16461259_C005.indd 164 10/25/2008 12:37:32 PM10/25/2008 12:37:32 PM

superpose a cooling time window (that is the period of time during which the polymer–gas solution is cooled prior to foaming) to the crystallization curves, as schematically shown in Figure 5.16. When crystallization is fast and the DSC peak is anticipated with respect to the cooling time window (shadowed area in Figure 5.16), crystallization has already occurred before foaming which in fact cannot take place, as in the case of carbon nano-tubes. When the cooling time window precedes crystallization, foaming occurs on a molten polymer and foaming effi ciency is maximum (mini-mum density) as is the case of pure PCL and PCL with alumina. In between these two extremes, all of the possibilities of partial superposition of the time windows, with partial densifi cation effect, are possible. Of course, cooling rate and foaming temperatures can be increased to avoid crystal-lization and shift the shadowed area accordingly.

Figure 5.17, as an example, reports the effect of the nucleating agents on the fi nal densities of PCL-based foams produced with N2 at a foaming tem-perature of 46.5°C. As can be observed, the effect of alumina on foam densi-fi cation is negligible, while the other nanoparticles affect foaming to an extent that depends on the nucleating agent type and content, in particular on its ability to induce crystal nucleation. In the case of nanocomposites based on carbon nanotubes, the selected foaming temperature and/or cool-ing rate were too low to allow foaming before crystallization and, as a result, fi nal density is that of the unfoamed, bulk semi-crystalline polymer.

More complex, for the more numerous inter-dependencies, is the evalu-ation of nucleation effi ciency of bubbles by the different nucleating agents. In effect, despite the two nucleating phenomena being very different, the presence of newly formed crystals (already nucleated) may induce the bubble nucleation, adding a new possible bubble nucleation mecha-nism.18 Hence, in this case, bubble nucleation can be related to several

FIGURE 5.16 Isothermal melt-crystallization at 45°C for the 0.4 wt% composites.

0 10 20 30

Time (min)

40 50 60

Cooling time windowcn

130F

30B

Talc

Alumina

Neat PCLSpe

cific

hea

t flo

w (

a.u.

)


61259_C005.indd 16561259_C005.indd 165 10/25/2008 12:37:33 PM10/25/2008 12:37:33 PM

mechanisms: (1) the homogeneous bubble nucleation in the polymer; (2) the heterogeneous nucleation on the surface of the nucleating agents; and (3) the heterogeneous nucleation induced by the forming crystals. The relative importance of the two latter phenomena (nucleation of bubbles induced by the presence of the nucleating agents and by the presence of the polymeric crystals) depends again on the superposition of the two time frames, foaming and crystallization, which are strongly dependent on the foaming temperature.

Bubble nucleation is, moreover, strongly affected by the kind of nucleat-ing agent, the number of particles, their dispersion in the polymeric matrix and by the different surface interactions between the polymer and nanopar-ticles which can, in turn, eventually be affected by the chemical–physical properties of the blowing agent. This effect is reported in Figure 5.18, where cell number densities per unit initial volume of the composites are reported for selected samples foamed with carbon dioxide and nitrogen.

All of the nanocomposites foamed with nitrogen, except the one based on spherical TiO2 (130F), led to cellular structure characterized by higher cell densities compared with materials foamed with carbon dioxide. Nanometric Al2O3 did not result in any improvement of the cellular structure compared to neat PCL, while talc and nanoclay (30B) displayed a bubble nucleation effi ciency that seemed to be independent of the type of expanding gas. A singular behavior was observed for nanocomposites containing TiO2. The highest nucleation effi ciency obtained with carbon dioxide suggests that, for this type of nanoparticle, the bubble nucleation is not only based on a heterogeneous mechanism occurring at the surface of the particles but also on a more complex mechanism that may involve specifi c interactions

FIGURE 5.17 Effect of concentration of the nucleating agents on the foam densities of

PCL-based systems, foamed with N2 at 46.5°C.

00

0.2

0.4

0.6

0.8

1.0

1.2

Neat

TalcAlumina

130F

30B

cn

0.1Nucleating agent content (wt. %)

Foa

m d

ensi

ty (

g/cm

3 )

0.4 0.7 1.0

166 Polymeric Foams

61259_C005.indd 16661259_C005.indd 166 10/25/2008 12:37:33 PM10/25/2008 12:37:33 PM

between gas, polymer, and particles as well as mechanisms related to pre-mature crystallization of the polymer. All of these mechanisms are strongly affected by the foaming temperature. This is clearly showed in Figure 5.19, where cell number densities as a function of foaming temperature are reported for the different polymer/nucleating agent systems. It is evident that at lower foaming temperatures, where crystallization has already occurred, cell number density is high and, in particular, follows the same trend of the crystal nucleation effi ciency (compare with Figure 5.16, where the crystallization rates for 30B talc neat PCL). By increasing the tem-perature, crystallization is shifted to longer times with respect to foaming and the nucleation of bubbles induced by the presence of crystals is sup-pressed (bubble nucleation takes place before crystallization). This results in a decrease in cell number densities. At the highest temperature investigated, the cell number density increased again for neat PCL and for all the nucleat-ing agents except, surprisingly, nanocomposites with nanoclay (30B).

These data clearly show that the bubble nucleation mechanism induced by nanoparticles is still far from completely understood since there are many unknown aspects that could contribute to the nucleation effi ciency of these very small size nucleants: their size; geometry and aspect ratio; surface properties; the presence of compatibilizers (often low molecular weight compounds), which can modify the local gas solubility and diffu-sivity around the particles; the presence of a higher macromolecular order close the particle surface that can further modify local thermodynamic; and/or the molecular mobility of the molecular chains.

FIGURE 5.18 Effect of the nucleating agent (at 0.4 wt%) on the cell number densities;

comparison between N2 and CO2 as the blowing agent, foamed at 47.5°C and 30°C

respectively.

104

105

106

107

108

Nea

t

Talc

Alu

min

a

130F

30B

Cel

l den

sity

(#/

cm3 )

N2

CO2


61259_C005.indd 16761259_C005.indd 167 10/25/2008 12:37:33 PM10/25/2008 12:37:33 PM

5.5 Conclusions

Among the most interesting biodegradable polymers that can be potentially employed for foaming are polyesters such as PCL and PLA. However, they have several limitations due to their poor rheological properties ansd small processing window. In particular, the main limitations in the use of PCL are related to the low processing temperatures and, in the case of PLA, to the low crystallization kinetics and the poor extensional rheological properties.

The branching of polymer chains as well as the optimization of molecu-lar weight and molecular weight distribution are common methods used to improve the extensional viscosity of a polymer and to make it suitable for foam formation. In fact, peroxide modifi ed PCL can be foamed at higher temperatures compared with unmodifi ed PCL and the morpho logies of these foams are characterized by a uniform cellular structure and fi ne cell dimensions. Commercial PLA can be modifi ed by using low-molecular-weight chain extenders and improved melt viscosity and elasticity can be obtained. Such improvements result in foams with much reduced cell size, increased cell density, and lowered foam density compared with unmodi-fi ed PLA.

Nanometric additives have shown to dramatically change properties such as gas solubility and diffusivity, rheological characteristics, and crys-tallization behavior, properties that are extremely important for the foam formation and lead to cellular structures characterized by different foam density, cell size, and cell density. In particular, nanometric fi llers induce

FIGURE 5.19 Effect of foaming temperature on the foam densities of selected composites

containing 0.4 wt% of nucleating agent.

46 47 48 49 50 51 52105

106

107

108

109

NeatTalc

Alumina130F30B

Tfoaming

Cel

l den

sity

(#/

cm3 )

168 Polymeric Foams

61259_C005.indd 16861259_C005.indd 168 10/25/2008 12:37:33 PM10/25/2008 12:37:33 PM

both crystalline phase nucleation and bubble nucleation and these effects were found to depend on the type, dimension, shape, and surface func-tionality of the nucleating agent, on the blowing agent, and temperatures used for foaming.

References

1. Klempner, D. and Sendijarevic, V. Polymeric Foams and Foam Technology. 2nd edition. Hanser Publishers/Hanser Gardener Publications, Inc., Cincinnati, OH, 2004.

2. Boyce, S. T. “Skin repair with cultured cells and biopolymers.” In Human Biomaterials Applications, ed. D. L. Wise, The Humana Press, Inc., Totowa, NJ, 1996, pp. 347–377.

3. Cotugno, S., Di Maio, E., Mensitieri, G., Nicolais, L., and Iannace, S. “Biodegradable foams.” In Handbook of Biodegradable Polymeric Materials and their Applications, ed. B. Narasimhan and S. K. Mallapragada. American Scientifi c Publishers, 2006, pp. 55–106.

4. Lee, S. T., Park, C. B., and Ramesh, “Biodegradable foams.” In Polymeric Foams, Science and Technology. Taylor and Francis (CRC Press), Boca Raton, 2006, pp. 165–192.

5. Siripurapu, S., Gay, Y. J., Royer, J. L., and DeSimone, J. M. “Generation of microcellular foams of PVDF and its blends using supercritical carbon dioxide in a continuous process.” Polymer 43 (2002): 5511–5520.

6. Spitael, P. and Macosko, C. W. “Strain hardening in polypropylenes and its role in extrusion foaming.” Polymer Engineering Science 44 (2004): 2090–2100.

7. Xanthos, M., Young, M. W., Karayannidis, G. P., and Bikiaris, D. N. “Reactive modifi cation of polyethylene terephthalate with polyepoxides.” Polymer Engineering Science 41 (2001): 643–655.

8. Fujimoto, Y., Ray, S. S., Okamoto, M., Ogami, A., Yamada, K., and Ueda, K. “Well-controlled biodegradable nanocomposite foams: from microcellular to nanocellular.” Macromolecular Rapid Communications 24 (2003): 457–461.

9. Ray, S. S. and Okamoto, M. “New polylactide/layered silicate nanocomposites, 6.” Macromolecular Material Engineering 288 (2003): 936–944.

10. Gaur, U., Lau, S.-F., Wunderlich, B. B., and Wunderlich, B. “Heat capacity and other thermodynamic properties of linear macromolecules. VIII. Polyesters and polyamides.” B. Journal of Physical Chemistry 12 (1983): 65–89.

11. Lebedev, B. and Yevstropov, A. “Thermodynamic properties of polylactones.” Makromolekulare Chemie 185 (1984): 1235–1253.

12. Di, W., Iannace, S., Di Maio, E., and Nicolais, L. “Nanocomposites by melt intercalation based on polycaprolactone and organoclay.” Journal of Polymer Science Part B: Polymer Physics 41 (2003): 670–678.

13. Di Maio, E., Iannace, S., Sorrentino, L., and Nicolais, L. “Isothermal crystal-lization in PCL/clay nanocomposites investigated with thermal and rheo-metric methods.” Polymer 45 (2004): 8893–8900.

14. Adamovsky, S. and Schick, C. “Ultra-fast isothermal calorimetry using thin fi lm sensors.” Thermochimica Acta 415 (2004): 1–7.


61259_C005.indd 16961259_C005.indd 169 10/25/2008 12:37:33 PM10/25/2008 12:37:33 PM

15. Di Maio, E., Mensitieri, G., Iannace, S., Nicolais, L., Li, W., and Flumerfelt, R. W. “Structure optimization of polycaprolactone foams by using mixtures of CO2 and N2 as blowing agents.” Polymer Engineering Science 45 (2005): 432–441.

16. Jenkins, M. J., Harrison, K. L., Silva, M. M. C. G., Whitaker, M. J., Shakesheff, K. M., and Howdle S. M. “Characterisation of microcellular foams produced from semi-crystalline PCL using supercritical carbon dioxide.” European Polymer Journal 42 (2006): 3145–3151.

17. Xu, Q., Ren, X., Chang, Y., Wang, J., Yu, L., and Dean, K. “Generation of microcellular biodegradable polycaprolactone foams in supercritical carbon dioxide.” Journal of Applied Polymer Science 94 (2004): 593–597.

18. Reignier, J. Tatibouët, J., and Gendron, R. “Batch foaming of poly(e-caprolactone) using carbon dioxide: Impact of crystallization on cell nucle-ation as probed by ultrasonic measurements.” Polymer 47 (2006): 5012–5024.

19. Cotugno, S., Di Maio, E., Mensitieri, G., Iannace, S., Roberts, G. W., Carbonell, R. G., and Hopfenberg, H. B. “Characterization of microcellular biodegrad-able polymeric foams produced from supercritical carbon dioxide solutions.” Industrial and Engineering Chemical Research 44 (2005): 1795–1803.

20. Marrazzo, C., Di Maio, E., Iannace, S., and Nicolais, L. “Process-structure relationship in PCL foaming.” Journal of Cellular Plastics 44 (2008): 37–52.

21. Narayanan, N., Roychoudhury, P. K., and Srivastava, A. “L (+) lactic acid fermentation and its product polymerization.” Electronic Journal of Biotechnology 7 (2004): 167–178.

22. Iannace, S. and Nicolais, L. “Isothermal crystallization and chain mobility of poly(L-lactide).” Journal Applied Polymer Science 64 (1997): 911–919.

23. Ohshima, M., Shikuma, H., E., Kihara, S., and Taki, K. “Supercritical CO2 and polymer systems—Experiments and prediction of properties of polymer/CO2 mixtures.” PPS Americas Regional Meeting, Quebec, 2005.

24. Liao, X., Nawaby, A. V., Whitfi eld, P., Day, M., Champagne, M., and Denault, J. “Layered open pore poly(L-lactic acid) nanomorphology.” Biomacromolecules 7 (2006): 2937–2941.

25. Mathieu, L. M., Montjovent, M.-O., Bourban, P.-E., Pioletti, D. P., and Månson, J.-A. E. “Bioresorbable composites prepared by supercritical fl uid foaming.” Journal Biomedical Materials Research 75A (2005): 89–97.

26. Ema, Y., Ikeya, M., and Okamoto, M. “Foam processing and cellular struc-ture of polylactide-based nanocomposites.” Polymer 47 (2006): 5350–5359.

27. Di, Y., Iannace, S., Di Maio, E., and Nicolais, L. “Reactively modifi ed poly(lactic acid): properties and foam processing.” Macromolecular Materials Engineering 290 (2005): 1083–1090.

28. Vega, J. Aguillar, M., Peòn, J., Pastor, D., and Martìnez-Salazar, J. “Effect of long chain branching on linear-viscoelastic melt properties of polyolefi ns.” E-polymer 2000, no. 046.

29. Narkis, M. and Wallerstein, R. “Crosslinking of polycaprolactone with per-oxides.” Polymer Communications 27 (1986): 314–317.

30. Chu, C. C. and Campbell, N. D. “Scanning electron microscopic study of the hydrolytic degradation of poly(glycolic acid) suture.” Biomedical Material Research 16 (1982): 417–430.

31. Pittman, C. U., Iqbal, M., and Chen, C. Y. “Radiation degradation of poly(y-hydroxyisobutyric acid) and poly(glycolic ester).” Journal of Polymer Science, Polymer Chemistry Edition 16 (1978): 2721–2724.

170 Polymeric Foams

61259_C005.indd 17061259_C005.indd 170 10/25/2008 12:37:33 PM10/25/2008 12:37:33 PM

32. Gupta, M. C. and Deshmukh, V. G. “Radiation effects on poly (lactic acid).” Polymer 24 (1983): 827–830.

33. Gandhi, K., Kriz, D., Salovey, R., Narkis, M., and Wallerstein, R. “Crosslinking of polycaprolactone in the pre-gelation region.” Polymer Engineering Science 28 (1988): 1484–1490.

34. Di Maio, E., Iannace, S., Marrazzo, C., Narkis, M., and Nicolais, L. “Effect of molecular modifi cation on PCL foam formation and morphology of PCL.” Macromolecular Symposia 228 (2005): 219–227.

35. Wood-Adams, P. M. and Dealy, J. M. “Using rheological data to determine the branching level in metallocene polyethylenes.” Macromolecules 33 (2000): 7481–7488.

36. Winter, H. H. and Mours, M. “Rheology of polymers near liquid-solid transi-tions.” Advances in Polymer Science 134 (1997): 166–234.

37. Naguib, H. E., Park, C. B., Panzer, U., and Reichelt, N. “Strategies for achiev-ing ultra low-density polypropylene foams.” Polymer Engineering Science 42 (2002): 1481–1492.

38. Carlson, D., Dubois, P., Nie, L., and Narayan, R. “Free radical branching of poly-lactide by reactive extrusion.” Polymer Engineering Science 38 (1998): 311–321.

39. Carlson, D. Nie, L., Narayan, R., and Dubois, P. “Maleation of polylactide (PLA) by reactive extrusion.” Journal of Applied Polymer Science 72 (1999): 477–485.

40. Sodergard, A., Niemi, M., Selin, J. F., and Nasman, J. H. “Changes in peroxide melt-modifi ed poly(L-lactide).” Industrial and Engineering Chemical Research 34 (1995): 1203–1207.

41. Park, E., Cho, H., Kim, M., and Yoon, J. “Chain extension and mechanical properties of unsaturated aliphatic copolyesters based on poly(L-lactic acid).” Journal of Applied Polymer Science 90 (2003): 1802–1807.

42. Shinno, K., Miyamoto, M., and Rimura, Y. “Solid-state postpolymerization of L-lactide promoted by crystallization of product polymer: an effective method for reduction of remaining monomer.” Macromolecules 30 (1997): 6438–6444.

43. Kylma, J., Tuominen, J., Helminen, A., and Seppala, J. “Chain extending of lactic acid oligomers. Effect of 2,2′-bis(2-oxazoline) on 1,6-hexamethylene diisocyanate linking reaction.” Polymer 42 (2001): 3333–3343.

44. Zhong, W., Ge, J., Gu, Z., Li, W., Chen, X., Zang, Y., and Yang, Y. “Study on biodegradable polymer materials based on poly(lactic acid). I. Chain extend-ing of low molecular weight poly(lactic acid) with methylenediphenyl diiso-cyanate.” Journal of Applied Polymer Science 74 (1999): 2546–2551.

45. Hiltunen, K., Seppala, J. V., and Harkonen, M. “Lactic acid based poly(ester-urethanes): Use of hydroxyl terminated prepolymer in urethane synthesis.” Journal of Applied Polymer Science 63 (1997): 1091–1100.

46. Tuominen, J., Kylma, J., and Seppala, J. “Chain extending of lactic acid oli-gomers. 2. Increase of molecular weight with 1,6-hexamethylene diisocyanate and 2,2′-bis(2-oxazoline).” Polymer 43 (2002): 3–10.

47. Vollalobos, M. A., Awojulu, A., Greeley, T., Turco, G., and Deeter, G. “Oligomeric chain extenders for economic reprocessing and recycling of condensation plastics.” Energy 31 (2006): 3227–3234.

48. Blasius, W. G., Deeter, G. A., and Villalobos, M. A. WO Patent 2003 066704. 49. Ano, Y. T., Ueda, K., Matsuoka, F., and Matsumoto, T. US Patent 2006 276617. 50. Marrazzo, C., Di Maio, E., and Iannace, S. “Foaming of synthetic and natural

biodegradable polymers.” Journal Cellular Plastics 43 (2007): 123.


61259_C005.indd 17161259_C005.indd 171 10/25/2008 12:37:33 PM10/25/2008 12:37:33 PM

51. Mitsunaga, M., Ito, Y., Ray, S. S., Okamoto, M., and Hironaka, K. “Intercalated polycarbonate/clay nanocomposites: nanostructure control and foam processing” Macromolecular Materials and Engineering 288 (2003): 543–546.

52. Usuki, A., Kojima, Y., Kawasumi, M., Okada, A., Fukushima, Y., Kurauchi, T., and Kamigaito, O. “Synthesis of nylon 6-clay hybrid.” Journal of Materials Research 8 (1993): 1174–1184.

53. Sinha Ray, S. and Bousmina, M. “Biodegradable polymers and their layered silicate nanocomposites: In greening the 21st century materials world.” Progress in Materials Science 50 (2005): 962–1079.

54. Okamoto, M. “Biodegradable polymer/layered silicate nanocomposites: A review.” In Handbook of Biodegradable Polymeric Materials and their Applications, ed. B. Narasimhan and S. K. Mallapragada. American Scientifi c Publishers, Stevenson Ranch, 2005.

55. Sinha Ray, S. and Okamoto, M. “Polymer/layered silicate nanocomposites: a review from preparation to processing.” Progress in Polymer Science 28 (2003): 1539–1641.

56. Paul, M.-A., Alexandre, M., Degée, P., Calberg, C., Jérôme, R., and Dubois, P. “Exfoliated polylactide/clay nanocomposites by in-situ coordination-insertion polymerization.” Macromolecular Rapid Communications 24 (2003): 561–566.

57. Hiroi, R., Sinha Ray, S., Okamoto, M., and Shiroi, T. “Organically modifi ed layered titanate: a new nanofi ller to improve the performance of biodegrad-able polylactide.” Macromolecular Rapid Communications 25 (2004): 1359–1364.

58. Sinha Ray, S., Yamada, K., Okamoto, M., Ogami, A., and Ueda, K. “New polylactide/layered silicate nanocomposites. 5. Designing of materials with desired properties.” Polymer 44 (2003): 6633–6646.

59. Hao, J., Yuan, M., and Deng X. “Biodegradable and biocompatible nanocom-posites of poly(ε-caprolactone) with hydroxyapatite nanocrystals: thermal and mechanical properties.” Journal of Applied Polymer Science 86 (2002): 676–683.

60. Sinha Ray, S., Yamada, K., Ogami, A., Okamoto, M., and Ueda K. “New poly-lactide/layered silicate nanocomposite: nanoscale control over multiple properties.” Macromolecular Rapid Communications 23 (2002): 943–947.

61. Maiti, P., Yamada, K., Okamoto, M., Ueda, K., and Okamoto, K. “New poly-lactide/layered silicate nanocomposites: role of organoclays.” Chemistry of Materials 14 (2002): 4654–4661.

62. Jimenez, G., Ogata, N., Kawai, H., and Ogihara, T. “Structure and thermal/mechanical properties of poly (ε-caprolactone)–clay blend.” Journal of Applied Polymer Sciences 64 (1997): 2211–2220.

63. Ogata, N., Jimenez, G., Kawai, H., and Ogihara, T. “Structure and thermal/mechanical properties of poly(l-lactide)-clay blend.” Journal of Polymer Science B 35 (1997): 389–396.

64. Lepoittevin, B., Pantoustier, N., Devalckenaere, M., Alexandre, M., Kubies, D., Calberg, C., Jérôme, R., and Dubois, P. “Poly(ε-caprolactone)/clay nano-composites by in-situ intercalative polymerization catalyzed by dibutyltin dimethoxide.” Macromolecules 35 (2002): 8385–8390.

65. Krikorian, V. and Pochan, D. “Crystallization behavior of poly(L-lactic acid) nanocomposites: nucleation and growth probed by infrared spectroscopy.” Macromolecules 38 (2005): 6520–6527.

172 Polymeric Foams

61259_C005.indd 17261259_C005.indd 172 10/25/2008 12:37:33 PM10/25/2008 12:37:33 PM

66. Lepoittevin, B., Devalckenaere, M., Pantoustier, N., Alexandre, M., Kubies, D., Calberg, C., Jerome, R., and Dubois, P. “Poly(ε-caprolactone)/clay nanocomposites prepared by melt intercalation: mechanical, thermal and rheological properties.” Polymer 43 (2002): 4017–4023.

67. Krishnamoorti, R. and Giannelis, E. P. “Rheology of end-tethered polymer layered silicate nanocomposites.” Macromolecules 30 (1997): 4097–4102.

68. Hoffmann, B., Dietrich, C., Thomann, R., Friedrich, C., and Muelhaupt, R. “Morphology and rheology of polystyrene nanocomposites based upon organoclay.” Macromolecular Rapid Communications 21 (2000): 57–61.

69. Shenoy, A. V. In Rheology of Filled Polymer Systems. Kluwer Academic, New Dehli, 1999.

70. Sinha Ray, S. and Okamoto, M. “Biodegradable polylactide and its nanocom-posites: opening a new dimension for plastics and composites.” Macromolecular Rapid Communications 24 (2003): 815–840.

71. Di, Y., Iannace, S., Di Maio, E., and Nicolais, L. “Poly(lactic acid)/organoclay nanocomposites: thermal, rheological properties and foam processing.” Journal of Polymer Science Part B: Polymer Physics 43 (2005): 689–698.

72. Pluta, M. “Melt compounding of polylactide/organoclay: Structure and prop-erties of nanocomposites.” Journal of Polymer Science Part B: Polymer Physics 44 (2006): 3392–3405.

73. Kojima, Y. Usuki, A. Kawasumi, M. Okada, A., Fukushima, Y., Kurauchi, T. T., and Kamigaito, O. “Synthesis of nylon 6-clay hybrid.” Journal of Materials Research 8 (1993): 1179–1184.

74. Kojima, Y., Usuki, A., Kawasumi, M., Okada, A., Kurauchi, T. T., and Kamigaito, O. “Synthesis of nylon 6-clay hybrid by montmorillonite interca-lated with ε-caprolactam.” Journal of Polymer Science Part A: Polymer Chemistry 31 (1993): 983–986.

75. Usuki, A., Kawasumi, M., Kojima, Y., Okada, A., Kurauchi, T., and Kamigaito, O. “Mechanical properties of nylon 6-clay hybrid.” Journal of Materials Research 8 (1993): 1185–1189.

76. Yano, K., Usuki, A., Okada, A., Kurauchi, T., and Kamigaito, O. “Synthesis a properties of polyimide–clay hybrid.” Journal of Polymer Science Part A: Polymer Chemistry 31 (1993): 2493–2498.

77. Messersmith, P. B. and Giannelis, E. P. “Synthesis and characterization of lay-ered silicate-epoxy nanocomposites.” Chemistry of Materials 6 (1994): 1719–1725.

78. Xu, R. Manias, E., Snyder, A. J., and Runt, J. “New biomedical poly(urethane urea)-layered silicate nanocomposites.” Macromolecules 34 (2001): 337–339.

79. Di, Y., Iannace, S., Sanguigno, L., and Nicolais, L. “Barrier and mechanical pro-prerties of poly(caprolactone)/organoclay nanocomposites.” Macromolecular Symposia 228 (2005): 115–124.

80. Sinha Ray, S., Yamada, K., Okamoto, M., Ogami, A., and Ueda, K. “New polylactide/layered silicate nanocomposites. 3. High-performance biode-gradable materials.” Chemistry of Materials 15 (2003): 1456–1465.

81. Cotugno, S. “Modeling of thermodynamic and transport properties of poly-mer melts and solutions to be used in the simulation of foaming processes.” PhD thesis, University of Naples Federico II, Italy, 2003.

82. Gopakumar, T. G., Lee, J. A., Kontopoulou, M., and Parent, J. S. “Infl uence of clay exfoliation on the physical properties of montmorillonite/polyethylene composites.” Polymer 43 (2002): 5483–5491.


61259_C005.indd 17361259_C005.indd 173 10/25/2008 12:37:34 PM10/25/2008 12:37:34 PM

83. Liu, X. and Wu, Q. “Non-isothermal crystallization behaviors of polyamide 6/clay nanocomposites.” European Polymer Journal 38 (2002): 1383–1389.

84. Liu, X., Wu, Q., and Berglund, L. A. “Polymorphism in polyamide 66/clay nanocomposites.” Polymer 43 (2002): 4967–4972.

85. Wu, Z., Chixin, Z., and Zhu, N. “The nucleating effect of montmorillonite on crystallization of nylon 1212/montmorillonite nanocomposite.” Polymer Testing 21 (2002): 479–483.

86. Li, J., Zhou, C., and Gang, W. “Study on nonisothermal crystallization of maleic anhydride grafted polypropylene/montmorillonite nanocomposite.” Polymer Testing 22 (2003): 217–223.

87. Liu, X. and Wu, Q. “PP/clay nanocomposites prepared by grafting-melt intercalation.” Polymer 42 (2001): 10013–10019.

88. Tseng, C.-R., Wu, J.-Y., Lee, H.-Y., and Chang, F.-C. “Preparation and crystal-lization behavior of syndiotactic polystyrene–clay nanocomposites.” Polymer 42 (2001): 10063–10070.

89. Taki, K., Yanagimoto, T., Funami, E., Okamoto, M., and Ohshima, M. “Visual observation of CO2 foaming of polypropylene-clay nanocomposites.” Polymer Engineering Science 44 (2004): 1004–1011.

90. Okamoto, M., Nam, P. H., Maiti, P., Kotaka, T., Nakayama, T., Takada, M., Ohshima, M., Usuki, A., Hasegawa, N., and Okamoto, H. “Biaxial fl ow-in-duced alignment of silicate layers in polypropylene/clay nanocomposite foam.” Nano Letters 1 (2001): 503–505.

91. Nam, P. H., Maiti, P., Okamoto, M., Kotaka, T., Nakayama, T., Takada, M. Ohshima, M., Usuki, A., Hasegawa, Ns., and Okamoto, H. “Foam processing and cellular structure of polypropylene/clay nanocomposites.” Polymer Engineering Science 42 (2002): 1907–1918.

92. Han, X., Zeng, C., Lee, L. J., Koelling, K. W., and Tomasko, D. L. “Extrusion of polystyrene nanocomposite foams with supercritical CO2.” Polymer Engineering Science 43 (2003): 1261–1275.

93. Tatibouët, J., Gendron, R., Hamel, A., and Sahnoune, A. “Effect of different nucleating agents on the degassing conditions as measured by ultrasonic sensors.” Journal of Cellular Plastics 38 (2002): 203–218.

94. Shen, J., Zeng, C., and Lee, L. J. “Synthesis of polystyrene–carbon nanofi bers nanocomposite foams.” Polymer 46 (2005): 5218–5224.

95. Zeng, C., Han, X., Lee, L. J., Koelling, K. W., and Tomasko, D. L. “Polymer-clay nanocomposite foams prepared using carbon dioxide.” Advanced Materials 15 (2003): 1743–1747.

96. Karbas, H., Nelson, P., Yuan, M., Gong, S., and Turng, L.-S. “Effects of nano-fi llers and process conditions on the microstructure and mechanical proper-ties of microcellular injection molded polyamide nanocomposites.” Polymer Composites 24 (2003): 655–671.

97. Cao, X., Lee, L. J., Widya, T., and Macosko, C. “Polyurethane/clay nanocompos-ites foams: processing, structure and properties.” Polymer 46 (2005): 775–783.

98. Lee, Y. H., Park, C. B., Wang, K. H., and Lee, M. H. “HDPE-clay nanocompos-ite foams blown with supercritical CO2.” Journal of Cellular Plastics 41 (2005): 487–502.

99. Marrazzo, C., Di Maio, E., and Iannace, S. “Conventional and nanometric nucleating agents in PCL foaming: crystals vs. bubbles nucleation.” Polymer Engineering Science 48 (2008): 336–344.

174 Polymeric Foams

61259_C005.indd 17461259_C005.indd 174 10/25/2008 12:37:34 PM10/25/2008 12:37:34 PM

6Nanostructure Development and Foam Processing in Polymer/Layered Silicate Nanocomposites

Masami Okamoto

CONTENTS

6.1 Introduction .................................................................................... 1766.2 Nanostructure Development ...................................................... 177

6.2.1 Melt Intercalation ................................................................. 1776.2.2 Interlayer Structure of OMLFs and Intercalation .......... 177

6.2.2.1 Nanofi llers .............................................................. 1776.2.2.2 Molecular Dimensions and

Interlayer Structure ............................................... 1786.2.2.3 Correlation of Intercalant Structure

and Interlayer Opening ........................................ 1826.2.2.4 Nanocomposite Structure ..................................... 183

6.3 Flow-Induced Structure Development ...................................... 1886.3.1 Elongational Flow and Strain-Induced Hardening ....... 188

6.4 Foam Processing ........................................................................... 1906.4.1 Foam Processing of PP-Based Nanocomposites ............ 1906.4.2 In-Situ Observation of Foaming ....................................... 1966.4.3 PLA-Based Nanocomposite Foaming ............................. 2006.4.4 Foaming Temperature Dependence of

Cellular Structure ............................................................... 2016.4.5 CO2 Pressure Dependence ................................................ 2046.4.6 TEM Observation ............................................................... 208

61259_C006.indd 17561259_C006.indd 175 10/25/2008 12:38:06 PM10/25/2008 12:38:06 PM

6.4.7 Polycarbonate (PC)-Based Nanocomposite Foams ........ 2096.4.8 Mechanical Properties of Nanocomposite Foams ............ 2116.4.9 Porous Ceramic Materials via Nanocomposite ................. 214

6.5 Conclusions and Future Prospects .............................................. 215References ................................................................................................ 216

6.1 Introduction

A decade of research has shown that nanostructured materials have the potential to signifi cantly impact growth at every level of the world econ-omy in the twenty-fi rst century. This new class of materials is now being introduced in structural applications including gas barrier fi lms, fl ame retardant products, and other load-bearing applications.

Of particular interest are recently developed nanocomposites consisting of a polymer and layered silicate, which often exhibit remarkably improved properties1 when compared with polymer or conventional composites (both micro- and macro-composites). In polymer/layered silicate nanocompos-ites, a nylon 6/layered silicate hybrid2 reported by Toyota Central Research & Development Co. Inc. was successfully prepared by in-situ polymeriza-tion of ε-caprolactam in a dispersion of montmorillonite (MMT). The silicate can be dispersed in liquid monomer or a solution of monomer. It has also been possible to melt-mix polymers with layered silicates, avoiding the use of organic solvents. This method permits the use of conventional processing techniques such as injection molding and extrusion. The extensive literature on nanocomposite research are covered in recent reviews.1,3,4

Continued progress in nanoscale controlling, as well as an improved understanding of the physico-chemical phenomena at the nanometer scale, have contributed to the rapid development of novel nanocomposites. This chapter presents current research on polymer/layered silicate nano-composites (PLSNCs) with the primary focus on nanostructure development and foam processing operations.

Development of nanocomposite foams is one of the latest evolutionary technologies of the polymeric foam following a pioneering effort by Okamoto and his colleagues.5,6 They prepared polypropylene (PP)/layered silicate, poly(l-lactide) (PLA)/layered silicate and polycarbonate (PC)/lay-ered silicate nanocomposites foams in a batch process using supercritical CO2 as a physical foaming agent.6–8

To innovate on the material properties of nanocomposite foams, one needs to pin down the morphological correlation between the dispersed silicate particles with nanometer dimensions in the bulk and the formed closed-cellular structure after foaming. This chapter is devoted to the study of evaluation of the performance potential of PLSNCs in foam application.

176 Polymeric Foams

61259_C006.indd 17661259_C006.indd 176 10/25/2008 12:38:07 PM10/25/2008 12:38:07 PM

6.2 Nanostructure Development

6.2.1 Melt Intercalation

Since the possibility of direct melt intercalation was fi rst demonstrated,9 melt intercalation has become a preparation of the intercalated polymer/layered silicate nanocomposites (PLSNCs). This process involves annealing, stati-cally or under shear, a mixture of the polymer and organically modifi ed layered fi llers (OMLFs) above the softening point of the polymer. During annealing, the polymer chains diffuse from the bulk polymer melt into the nanogalleries between the layered fi llers.

In order to understand the thermodynamic issue associated with the nanocomposite formation, Vaia et al. have applied a mean-fi eld statistical lattice model and found conclusions based on the mean fi eld theory agreed with the experimental results.10,11 The entropy loss associated with con-fi nement of a polymer melt is not prohibited to nanocomposite formation because an entropy gain associated with the layer separation balances the entropy loss of polymer intercalation, resulting in a net entropy change near to zero. Thus, from the theoretical model, the outcome of nanocom-posite formation via polymer melt intercalation depends on energetic factors, which may be determined from the surface energies of the polymer and OMLF.

Nevertheless, we have often faced the problem where the nanocom-posite shows fi ne and homogeneous distribution of the nanoparticles in the polymer matrix [e.g. poly(l-lactide)] without a clear peak shift of the mean interlayer spacing of the (001) plane, as revealed by wide-angle X-ray diffraction (WAXD) analysis.12 Furthermore we sometimes encounter a decrease in interlayer spacing compared with that of pristine OMLF, despite very fi ne dispersion of the silicate particles. For this reason, infor-mation on the structure of the surfactant (intercalant)–polymer interface is necessary to understand the intercalation kinetics that can predict fi nal nanocomposite morphology and overall material properties.

6.2.2 Interlayer Structure of OMLFs and Intercalation

6.2.2.1 Nanofi llers

In characterizing layered silicate, including layered titanate (HTO), the surface charge density is particularly important because it determines the interlayer structure of intercalants as well as cation exchange capacity (CEC). Lagaly proposed a method consisting of total elemental analysis and the dimension of the unit cell:13

surface charge: e� ____

nm2 �

� __

ab (6.1)

Nanostructure Development and Foam Processing 177

61259_C006.indd 17761259_C006.indd 177 10/25/2008 12:38:07 PM10/25/2008 12:38:07 PM

where � is the layer charge [1.07 for HTO, 0.66 for synthetic fl uorine hectrite (syn-FH), and 0.33 for montmorillonite (MMT); a and b are cell parameters of HTO (a � 3.782 Å, b � 2.978 Å),14 syn-FH (a � 5.24 Å, b � 9.08 Å),15 and MMT (a � 5.18 Å, b � 9.00 Å)].16 For syn-FH, however, about 30% of the interlayer Na� ions are not replaced quantitatively by intercalants due to the non-activity for ion-exchange reactions.15 For HTO, only 27% of interlayer H� (H3O

�) is active for ion-exchange reactions.12 The remaining part is the non-active site in the HTO. Thus the incomplete replacement of the interlayer ions is ascribed to the intrinsic chemical reactivity. The characteristic parameters of three nanofi llers are summa-rized in Table 6.1.16 HTO has a high surface charge density of 1.26 e�/nm2 compared with those of syn-FH (0.971 e�/nm2) and MMT (0.780 e�/nm2). From these results, we can estimate the average distance between exchange sites, which is calculated to be 0.888 nm for HTO, 1.014 nm for syn-FH, and 1.188 nm for MMT. This estimation assumes that the cations are evenly distributed in a cubic array over the silicate surface and that half of the cations are located on one side of the platelet and the other half reside on the other side.

6.2.2.2 Molecular Dimensions and Interlayer Structure

The calculated models of the intercalant structures are presented in Figure 6.1. For octadecyl ammonium (C18H3N

�), obtained molecular length, thickness, and width are 2.466 nm, 0.301 nm, and 0.301 nm, respec-tively. Since the length of all alkyl units are more than 2 nm, these spacings (distance between exchange sites) of 0.888–1.188 nm do not allow parallel

TABLE 6.1

Characteristic Parameters of Nanofi llers

Parameters HTO syn-FH MMT

Chemical formula H1.07Ti1.73O3.95·0.5H2O Na0.66Mg2.6Si4O10(F)2 Na0.33(Al1.67Mg0.33)

Si4O10(OH)2

Particle size (nm) ~100–200 ~100–200 ~100–200

BET area (m2/g) ~2400 ~800 ~700

CEC* (meq/100 g) ~200 (660) ~120 (170) ~90 (90)

e� (charge/nm2) 1.26 0.971 0.708

Density (g/cm3) 2.40 2.50 2.50

Refractive index

(n20D)

2.3 1.55 1.55

pH 4–6 9–11 7.5–10

*Methylene blue adsorption method. The values in the parenthesis are calculated from

chemical formula of nanofi llers.

Source: From Yoshida, O. and Okamoto, M. Macromolecular Rapid Communications 27 (2006):

751–757. © 2006 Wiley-VCH. With permission.

178 Polymeric Foams

61259_C006.indd 17861259_C006.indd 178 10/25/2008 12:38:07 PM10/25/2008 12:38:07 PM

Intercalant C18H3N+ C18(CH3)3N+ 2C18(CH3)2N+ qC14(OH)

Length (nm) 2.466 2.601 4.766 2.090

Thickness (nm) 0.301 0.372 0.434 0.374

Width (nm) 0.301 0.372 0.318 0.881

(a)

(b)

(c)

(d)

FIGURE 6.1 Molecular dimensions of intercalans: (a) octadecyl ammonium [C18H3N�]; (b)

octadecyl trimethyl ammonium [C18(CH3)3N�]; (c) dioctadecyl dimethyl ammonium

[2C18(CH3)2N�]; and (d) N-(coco alkyl)-N,N-[bis(2-hydroxyethyl)]-N-methyl ammonium

[qC14(OH)] cations. [From: Yoshida, O. and Okamoto, M. Macromolecular Rapid Communications

27 (2006): 751–757. © 2006 Wiley-VCH. With permission.]

layer arrangement like fl at-lying chains13 in each gallery space of the nano-fi llers. All of the intercalants are oriented with some inclination to the host layer in the interlayer space to form an interdigitated layer. This is sug-gested by the paraffi n-type layer structure proposed by Lagaly, especially in the case of highly surface-charged clay minerals.13


61259_C006.indd 17961259_C006.indd 179 10/25/2008 12:38:07 PM10/25/2008 12:38:07 PM

Wide-angle X-ray diffraction (WAXD) patterns for three OMLF pow-ders are presented in Figure 6.2. The mean interlayer spacing of the (001) plane (d(001)) for the HTO intercalated with qC14(OH) [HTO-qC14(OH)] obtained by WAXD measurements is 2.264 nm (diffraction angle, 2Q � 3.90°). The appearances of small peaks observed at 2Q � 7.78°, 11.78°, and 15.74° confi rmed that these refl ections were due to (002) up to (004) plane of HTO-qC14(OH). HTO-qC14(OH) showed a surprisingly well-ordered suprastructure, as demonstrated by WAXD with diffraction maxima up to the fourth order, due to the high surface charge density of the HTO layers. On the other hand, syn-FH and MMT, which have low surface charge density compared with that of HTO, show a less-ordered interlayer structure; that is, the coherent order of the silicate layers is much lower in each syn-FH and MMT intercalated with surfactants.

0

2000

4000

6000

8000

0

2000

4000

6000

00 5 10 15 20

2000

4000

6000

2Θ (°)

Inte

nsity

(a.

u.)

d = 2.264 nm

(001)

d = 1.135 nm

(002)

(003) (004)

d = 0.751 nmd = 0.563 nm

d = 2.063 nm

(001)

d = 1.027 nm

(002)

d = 1.855 nm

(001)

HTO-qC14(OH)

syn-FH-qC14(OH)

MMT-qC14(OH)

d = 0.930 nm

(002)

FIGURE 6.2 WAXD patterns of HTO, syn-FH, and MMT intercalated with qC14(OH)+. [From:

Yoshida, O. and Okamoto, M. Macromolecular Rapid Communications 27 (2006): 751–757.

© 2006 Wiley-VCH. With permission.]

180 Polymeric Foams

61259_C006.indd 18061259_C006.indd 180 10/25/2008 12:38:07 PM10/25/2008 12:38:07 PM

From WAXD results, we can discuss the interlayer opening that is esti-mated after subtraction of the layer thickness value of 0.374 nm for HTO,14 0.98 nm for syn-FH15 and 0.96 nm for MMT.13 This is an important point for the following discussion of the interlayer structure. The illustration of a model of interlayer structure of the qC14(OH) in the gallery space of the HTO is shown in Figure 6.3. For nanofi llers with high surface charge density, the intercalants can adopt a confi guration with orientation where the alkyl chains are tilted under the effect of van der Waals forces, which decreases the chain–chain distance. For this reason, the angle α should be directly related to the packing density of the alkyl chains. The value of α decreases until close contact between the chains is attained, giving an increase in the degree of the crystallinity of the intercalants in the nano-galleries. To estimate the tilt angle α, we combined the molecular dimen-sion, interlayer spacing, and loading amount of intercalant in the layers, which was calculated from thermogravimetry analysis (TGA). The char-acteristic parameters are summarized in Tables 6.2 and 6.3. Note that HTO exhibits a large layer opening accompanied with large values of α and endothermic heat fl ow (�H) owing to the melting of the intercalants in the galleries when compared with those of syn-FH and MMT. This indicates that HTO leads to a highly interdigitated layer structure and the interlayer

0.888 nm

(0.794 nm2/charge)

Ti-O

Ti-O

1.889 nm

HTO-qC14(OH)

CH3N+

OH

– –

CH3N+

OH

CH3 N+

HO

CH3 N+

HOα

– –

FIGURE 6.3 Illustration of a model of interlayer structure of intercalant N-(coco alkyl)-

N,N-[bis(2-hydroxyethyl)]-N-methyl ammonium [qC14(OH)] cation in gallery space of layer

titanate (HTO). The average distance between exchange sites is 0.888 nm, calculated by

surface charge density of 1.26 e�/nm2. For qC14(OH), obtained molecular length, thickness,

and width are 2.09 nm, 0.881 nm, and 0.374 nm, respectively (see Figure 6.1). The tilt angle

� of the intercalants can be estimated by the combination of the interlayer spacing, molecu-

lar dimensions, and loading amount of intercalants when the alkyl chains adopt an all-trans

conformation. [From: Yoshida, O. and Okamoto, M. Macromolecular Rapid Communications



61259_C006.indd 18161259_C006.indd 181 10/25/2008 12:38:08 PM10/25/2008 12:38:08 PM

opening becomes more uniform compared with MMT and syn-FH (possessing lower surface charge density).

From this fact, we can observe well-defi ned diffraction peaks up to the (004) plane (see Figure 6.2). The entropic contribution of the intercalants, which leads to the entropy gain associated with the layer expansion after intercalation of the polymer chains, may not be signifi cant because of the interdigitated layer structure.

6.2.2.3 Correlation of Intercalant Structure and Interlayer Opening

For the interdigitated layer structure in MMT, alkyl chain length [i.e. C18H37, CH3 and (CH2)2OH in the amine structure] changes the interlayer opening. That is, when we compare different intercalants having the same long alkyl chain [i.e. C18H3N

� and C18(CH3)3N�], three methyl (CH3) sub-

stituents instead of hydrogen (H) disturb the contact with silicate surfaces. The value of α decreases until close contact between the ammonium cations

TABLE 6.2

Comparison of Characteristic Parameters between HTO, syn-FH, and MMT Prepared with qC14(OH)

HTO-qC14(OH) syn-FH-qC14(OH) MMT-qC14(OH)

Layer opening (nm) 1.889 1.083 0.895

Tilt angle � (°) 64.4 31.1 25.3

Organic content (wt%) 39.6 30.4 32.5

Tm* (°C) 108.3 111.3 97.7

�Ha (J/g) 214.5 141.2 138.6

*The melting and heat fl ow of qC14(OH)�Cl� are 35.8°C and 69.8 J/g, respectively.



TABLE 6.3

Comparison of Characteristic Parameters MMT-Based OMLF Prepared with C18H3N�, C18(CH3)3N�, and 2C18(CH3)2N�

C18H3N� C18(CH3)3N

� 2C18(CH3)2N�

Layer opening (nm) 1.350 1.011 1.540

Tilt angle � (°) 33.2 22.9 40.1

Organic content (wt%) 35.5 29.5 39.8

Tm* (°C) 69.9 69.5 44.0

�H* (J/g) 177.7 189.6 129.7

*The melting and heat fl ow of C18H3N�, C18(CH3)3N

�, and 2C18(CH3)2N� are 83.8°C and

95.6 J/g; 103.5°C and 161.2 J/g; and 37.0°C and 54.6 J/g, respectively.



182 Polymeric Foams

61259_C006.indd 18261259_C006.indd 182 10/25/2008 12:38:08 PM10/25/2008 12:38:08 PM

and silicate surfaces is attained, giving a decrease in the interlayer opening (� d(001)) (see Table 6.3 and Figure 6.2).

In cases where the intercalant has two long alkyl chains (i.e. 2C18(CH3)2N�),

the packing density of the alkyl chains is reduced and sterically limited in the nanogalleries. Consequently, MMT-2C18(CH3)2N

� exhibits large inter-layer opening accompanied by low crystallinity of the intercalant (�H ~ 130 J/g) compared with MMT-C18H3N

� and MMT-C18(CH3)3N�. Accordingly,

we observe a disordered diffraction peak of (001) plane of MMT-2C18(CH3)2N

� in the WAXD analysis (see fi gure 1 in Reference 17). We have to pay attention to the molecular size of the substituents instead of H attached to the nitrogen for the better understanding of the interdigitated layer structure and direct polymer melt intercalation. This feature has been observed in the results of OMLFs intercalated with various intercalants (such as octadecyl di-methyl benzyl ammonium, n-hexadecyl tri-n-butyl phosphonium, n-hexadecyl tri-phenyl phosphonium cations).18

6.2.2.4 Nanocomposite Structure

Figure 6.4 shows the results of TEM bright fi eld images of PLA-based nanocomposites, in which dark entities are the cross-section of interca-lated MMT layers. The organically modifi ed MMT content in all nano-composites was 4 wt%. From the TEM images, it becomes clear that there are some intercalated and stacked silicate layers in the nanocomposites. Yoshida et al. estimated the form factors obtained from TEM images; that is, average value of the particle length (L), of the dispersed particles, and the correlation length (�) between them.19 From the WAXD patterns, the crystallite size (D) of intercalated stacked silicate layers of each nanocom-posite was calculated using the Scherrer equation. The calculated value of D (� thickness of the dispersed particles) and other parameters for each nanocomposite are presented in Table 6.4.

For PLA/MMT-C18(CH3)3N�, L and D are in the range 200 ± 25 nm and

10.7 nm. On the other hand, PLA/MMT-C18H3N� exhibits a large value of

L (450 ± 200 nm) with a large level of stacking of the silicate layers (D~ 21 nm). The � value of the PLA/MMT-C18(CH3)3N

� (80 ± 20 nm) is lower than the value of PLA/MMT-C18H3N

� (260 ± 140 nm), suggesting that the interca-lated layers are more homogeneously and fi nely dispersed in the case of PLA/MMT-C18(CH3)3N

�. The number of the stacked individual silicate layers (� D/d(001) � 1) is 5 for PLA/MMT-C18(CH3)3N

� and � value of this nanocomposite is one order of magnitude lower compared with those of PLA/MMT-C18H3N

� and PLA/MMT-2C18(CH3)2N�, suggesting that inter-

calated silicate layers are more homogeneously and fi nely dispersed.Although the (initial) interlayer opening of MMT-C18(CH3)3N

� at 1.011 nm is smaller than MMT-C18H3N

� at 1.350 nm and MMT-2C18(CH3)2N� at

1.540 nm, the intercalation of the PLA in these different OMLFs gives almost the same basal spacing after preparation of the nanocomposites. Note that the existence of a sharp Bragg peak in PLA-based nanocomposites


61259_C006.indd 18361259_C006.indd 183 10/25/2008 12:38:08 PM10/25/2008 12:38:08 PM

500 nm

500 nm

(a)

(b)

(c)

300 nm

FIGURE 6.4 Bright fi led TEM images of PLA-based nanocomposites prepared with:

(a) MMT-C18H3N+; (b) MMT-C18(CH3)3N

+; and (c) MMT-2C18(CH3)2N+. The dark entities are

the cross-section and/or face of intercalated-and-stacked silicate layers and the bright areas

are the matrix. [From: Yoshida, O. and Okamoto, M. Macromolecular Rapid Communications


184 Polymeric Foams

61259_C006.indd 18461259_C006.indd 184 10/25/2008 12:38:08 PM10/25/2008 12:38:08 PM

after melt extrusion clearly indicates that the dispersed silicate layers still retain an ordered structure after melt extrusion.

In Table 6.4 they summarized the layer expansion after preparation (� � opening) of three nanocomposites, or after subtraction of the initial layer opening. For the same MMT with different intercalants [e.g. comparison between MMT-C18(CH3)3N

� and MMT-2C18(CH3)2N�], the layer expansion

of the former (0.879 nm) exhibits a large value compared with that of the latter (0.45 nm) in PLA-based nanocomposites. In other words, the smaller interlayer opening caused by the confi guration with a small tilt angle [� � 22.9° for C18(CH3)3N

�] promotes a large amount of intercalation of the polymer chains. Accordingly, PLA/MMT-C18(CH3)3N

� exhibits fi ner dis-persion of the nanofi llers compared with PLA/MMT-2C18(CH3)2N

� and PLA/MMT-C18H3N

� as discussed previously (see Figure 6.4).A more interesting feature is the absolute value of � opening. According

to the molecular modeling, the width and thickness of the PLA are 0.76 nm and 0.58 nm (see Figure 6.5). This may suggest that the polymer chains could not penetrate into galleries in the case of MMT-2C18(CH3)2N

� when we compare the apparent interlayer expansion (� � opening).

Now it is necessary to understand the meaning of the interlayer expan-sion in the intercalated nanocomposites. As discussed previously, we have to take the interdigitated layer structure into consideration. This structure may suggest that a different orientation angle could be adopted when the polymer chains penetrate into the galleries, giving a decrease in basal spacing after intercalation. At the same time, this structure apparently provides a balance between the polymer penetration and different orien-tation angle of the intercalants; that is, we have to pay attention to the polymer chain intercalation into the galleries from the result of the change of the basal spacing as revealed by WAXD. Presumably the penetration of the polymer chain is prevented or reduced by the steric limitation of the

TABLE 6.4

Form Factors of Three Nanocomposites Obtained from WAXD and TEM Observations

Nanocomposites

PLA/MMT-

C18H3N�

PLA/MMT-

C18(CH3)3N�

PLA/MMT-

2C18(CH3)2N�

d001 (nm) 3.03 2.85 2.95

� opening (nm) 0.72 0.879 0.45

Final layer opening (nm) 2.07 1.89 1.99

D (nm) 20.9 10.73 14.71

(D/d001) � 1 7.9 4.8 6.0

L (nm) 450 ± 200 200 ± 25 655 ± 121

� (nm) 260 ± 140 80 ± 20 300 ± 52




61259_C006.indd 18561259_C006.indd 185 10/25/2008 12:38:08 PM10/25/2008 12:38:08 PM

confi guration with a large value of [e.g. � � 40.1° for MMT-2C18(CH3)2N�].

Accordingly, we sometimes observe small interlayer expansion and encounter a decrease in the interlayer spacing after melt intercalation. As seen in Table 6.4, the initial interlayer opening depends on the interlayer expansion (� � opening) after melt intercalation. The smaller initial open-ing leads to the larger interlayer expansion, and gives almost same fi nal interlayer opening. This feature has been observed in the results of other nanocomposites prepared by different OMLFs intercalated with different surfactants.20 From this result, the entropic contribution of the intercal-ants, which leads to the entropy gain associated with layer expansion after intercalation of the small molecules and/or polymer chains, may not be signifi cant owing to the interdigitated layer structure. Presumably the penetration takes place by pressure drop within the nanogalleries, nano-capillary action, generated by the two platelets.

As reported in the literature,18 the pressure drop (�p) in the nanogalleries, which makes the polymer penetration more diffi cult, should be discussed. The estimated pressure difference (~24 MPa) is much larger than the shear stress (~0.1 MPa) during melt compounding.18 This suggests that shear stress has little effect on the delamination (exfoliation) of the layer. This reasoning is consistent with the intercalated structure reported by so many nanocomposite researchers, who can prepare only intercalated (not exfoli-ated) nanocomposites via the simple melt extrusion technique.1 A novel compounding process is currently in progress. Solid-state shear processing may be an innovative technique to delaminate the layered fi llers.21

Compared to OMLFs, the nanocomposite structure is diffi cult to model using atomic scale molecular dynamics (MD) because the intercalated polymer chain conformation is complex and is rarely in an equilibrium state. However, Pricl et al.22 explored and characterized the atomic scale

3.3763.933

Thickness 0.338 nm

PLA-oligomer

Width 0.393 nm

MD program(MM2 in Quantum CAChe)

FIGURE 6.5 Molecular dimensions of PLA-backbone using the molecular dynamics program

(MM2 in Quantum CAChe) in consideration of van der Waals radii into consideration. Optimi-

zation of structure is based on minimization of the total energy of the molecular system.

186 Polymeric Foams

61259_C006.indd 18661259_C006.indd 186 10/25/2008 12:38:08 PM10/25/2008 12:38:08 PM

structure to predict binding energies and basal spacing of PLSNCs based on polypropylene (PP) and maleated (MA) PP (PP-MA), MMT, and differ-ent alkyl ammonium ions as intercalants (see Figure 6.6). From a global interpretation of all of these MD simulation results, they concluded that intercalants with a smaller volume are more effective for clay modifi ca-tion as they improve the thermodynamics of the system by increasing the binding energy. On the other hand, intercalants with longer tails are more effective for intercalation and exfoliation processes, as they lead to higher

FIGURE 6.6 Three-component model used for basal spacing simulations, consisting of

two layers of MMT with K� cations (stick model), four molecules of trimethyl ammonium

cation (a) or dimethyl stearyl ammonium cation (b) (stick and ball model), and one molecule

of maleated PP (PP-MA) (ball model). [From Toth, R., Coslanicha, A., Ferronea, M., et al. Polymer 45 (2004): 8075–8083. © 2006 Elsevier Science. With permission.]


61259_C006.indd 18761259_C006.indd 187 10/25/2008 12:38:09 PM10/25/2008 12:38:09 PM

basal spacing. Additional information is necessary to predict a more reasonable nanostructure of PLSNCs. Some literature related to the confi ned polymer chains within the silicate galleries by using coarse-grained MD simulation has been published.23–26

6.3 Flow-Induced Structure Development

Rheological behavior, especially elongational and shear fl ow behavior in the molten state of PLSNCs, has not been well studied, although such knowledge should be indispensable in relation to their performance in processing operations. One objective of this chapter is to focus on a profound understanding of PLSNCs for their innovations in practical material production. For this purpose, it is indispensable to illuminate the nanostructure as well as rheological properties of PLSNCs to assess appropriate processing conditions for designing and controlling their hierarchical nanostructure, which must be closely related to their material performance.

6.3.1 Elongational Flow and Strain-Induced Hardening

Okamoto et al.27 fi rst conducted an elongation test of PP-based nanocom-posites (PPCN4) under molten state at constant Hencky strain rate, e.0 using an elongation fl ow opto-rheometry, and attempted to control the alignment of the dispersed MMT layers with nanometer dimensions of intercalated PPCNs under uniaxial elongational fl ow.

Figure 6.7 shows double logarithmic plots of transient elongational vis-cosity �E(�

.0; t) against time t observed for a nylon 6/OMLS system (N6CN3.7:

MMT � 3.7 wt%) and PPCN4 (MMT � 4 wt%) with different Hencky strain rates, �

.0, ranging from 0.001 s�1 to 1.0 s�1. The solid curve represents

time development of three-fold shear viscosity, 3�0(�.; t), at 225°C with a

constant shear rate �. � 0.001 s�1. In �E(�

.0; t) at any �

.0, N6CN3.7 melt shows

a weak tendency of strain-induced hardening compared with that of PPCN4 melt. A strong behavior of strain-induced hardening for PPCN4 melt was originated from the perpendicular alignment of the silicate layers to the stretching direction as reported by Okamoto et al.28

From TEM observation,29 the N6CN3.7 forms a fi ne dispersion of the sili-cate layers of about 100 nm in Lclay, 3 nm thickness in dclay and clay of about 20–30 nm between them. The clay value is one order of magnitude lower than the value of Lclay, suggesting the formation of a spatially linked-like structure of the dispersed clay particles in the nylon 6 matrix. For N6CN3.7 melt, the silicate layers are densely dispersed into the matrix and hence diffi cult to align under elongational fl ow. Under fl ow fi elds, the silicate

188 Polymeric Foams

61259_C006.indd 18861259_C006.indd 188 10/25/2008 12:38:10 PM10/25/2008 12:38:10 PM

layers might translationally move, but not rotationally in such a way that the loss energy becomes minimum. This tendency was also observed in PPCN7.5 melt having higher content of MMT (� 7.5 wt%).30

One can observe two features for the shear viscosity curve. First, the extended Trouton rule, 3�0(�

.; t) � �E(�

.0; t), does not hold for both N6CN3.7

and PPCN4 melts, as opposed to the melt of ordinary homo-polymers. The latter, �E(�

.0; t), is more than 10 times larger than the former, 3�0(�

.; t).

Second, again unlike ordinary polymer melts, 3�0(�.; t) of N6CN3.7 melt

increases continuously with t, never showing a tendency of reaching a steady state within the time span (600 s or longer) examined here. This time- dependent thickening behavior may be called anti-thixotropy or rheopexy. Under slow shear fl ow (�

. � 0.001 s�1), 3�0(�

.; t) of N6CN3.7 exhibits a

much stronger rheopexy behavior with almost two orders of magnitude higher than that of PPCN4. This refl ects a fact that the shear-induced

108

106

104

102

100

η (P

a s)

106

104

102

100

10–1 100 101

Time (s)

102 103

N6CN3.7

225°C

PPCN4

150°C

ε0 / s–1∗

1.00.50.1

3*η0

0.050.030.010.0050.001

(cone-plate; 0.001s–1

)(a)

(b)

FIGURE 6.7 Time variation of elongational viscosity �E(�.0; t) for: (a) N6CN3.7 melt at 225°C;

and (b) PPCN4 at 150°C. The solid line shows three times the shear viscosity, 3�E(�.; t), taken

at a low shear rate �.

� 0.001 s�1 on a cone-plate rheometer. [From Okamoto, M. “Polymer/

layered silicate nanocomposites.” Rapra Review Report No. 163, Rapra Technology Ltd,

London, 2003. 166 pp. © 2003 Rapra Technology Ltd. With permission.]


61259_C006.indd 18961259_C006.indd 189 10/25/2008 12:38:10 PM10/25/2008 12:38:10 PM

structural change involved a process with an extremely long relaxation time as well as for other PLSNCs having rheopexy behavior,31 especially under the weak shear fi eld.

In uniaxial elongational fl ow (converging low) for a PPCN4, the forma-tion of a ‘house-of-cards’ structure is found by TEM analysis.27 The per-pendicular (but not parallel) alignment of disk-like MMT clay particles with large anisotropy toward the fl ow direction might sound unlikely but this could be the case, especially under an elongational fl ow fi eld in which the extentional fl ow rate is the square of the converging fl ow rate along the thickness direction, if the assumption of affi ne deformation without volume change is valid. Obviously under such conditions, the energy dis-sipation rate due to viscous resistance between the disk surface and the matrix polymer is minimal when the disks are aligned perpendicular to the fl ow direction.

Some 20 years ago, van Olphen32 pointed out that the electrostatic attraction between the layers of natural clay in aqueous suspension arises from higher polar forces in the medium. The intriguing features such as yield stress thixotropy and/or rheopexy exhibited in aqueous suspensions of natural clay minerals may be taken as a reference to the present PLSNCs.

6.4 Foam Processing

Flow-induced internal structural change occurs in both shear and elonga-tional fl ow, but differs in each case, as noted from the above results on �E(�

.0; t) and 3�0(�

.; t) (see Figure 6.7). Thus, with the rheological features of

the PLSNCs and the characteristics of each processing operation, tactics should be selected accordingly for a particular nanocomposite for the enhancement of its mechanical properties.

For example, the strong strain-induced hardening in �E(�.0; t) is requisite

for withstanding the stretching force during the processing, while the rheopexy in 3�0(�

.; t) suggests that for such PLSNC a promising technol-

ogy is the processing in confi ned space (such as injection molding) where shear force is crucial.

6.4.1 Foam Processing of PP-Based Nanocomposites

PPCNs have already been shown to exhibit a tendency toward strong strain-induced hardening. On the basis of this result, the fi rst successful nanocomposite foam, processed by using supercritical (sc)-CO2 as a phys-ical foaming agent, appeared through a pioneering effort by Okamoto et al.5,6 A small amount of nanofi llers in the polymer matrix serve as

190 Polymeric Foams

61259_C006.indd 19061259_C006.indd 190 10/25/2008 12:38:10 PM10/25/2008 12:38:10 PM

nucleation sites to facilitate the bubble nucleation during foaming. Novel nanocomposite foams based on the combination of new nanofi llers and sc-CO2 led to a new class of materials. The process consists of four stages: (1) saturation of CO2 in the sample at desired temperature; (2) cell nucle-ation when the release of CO2 pressure started (supersaturated CO2); (3) cell growth to an equilibrium size during the release of CO2; and (4) sta-bilization of cell via cooling of the foamed sample. The autoclave setup used in their study is shown in Figure 6.8. Figure 6.9 represents the scan-ning electron micro scopy (SEM) images of PP-MA and various PPCNs foams conducted at various temperatures under a pressure of 10 MPa. From the SEM images it can be clearly observed that, apart from PPCN4 (MMT � 4 wt%) and PPCN7.5 foams prepared at 130.6°C, all exhibit neatly closed-cell structures with cells having 12- or 14-hedron shapes. The formed cells show their faces mostly in pentagons or hexagons, which express the most energetically stable state of polygon cells. They also calculated the distribution function of cell sizes from SEM images as shown in Figure 6.10. From the distribution curve it is clearly seen that PPCN7.5 exhibited a bimodal distribution of cell size, whereas the other samples neatly follow a Gaussian distribution. Another interesting obser-vation from Figure 6.10 is the width of the distribution peaks—the polydispersity of the cell size became narrower with the addition of clay into the matrix (PPCN2 and PPCN4). This behavior may be due to the

Autoclave

Bandheater

CO2 gas cylinderCooling water jacket

Sample

Pressure gauge

FIGURE 6.8 Schematic representation of autoclave set-up. [From Nam, P. H., Okamoto, M.,

Maiti, P., et al. Polymer Engineering Science 42 (2002): 1907–1918. © 2002 Society of Plastic

Engineers. With permission.]


61259_C006.indd 19161259_C006.indd 191 10/25/2008 12:38:11 PM10/25/2008 12:38:11 PM

heterogenous clay sites possibly acting for cell nucleation and their uniform dispersion in the matrix, which, if present, leads to the high homogeneity in cell size. On the other hand, the cell size of prepared foam gradually decreases with increasing clay content in the PPCNs. This behavior is due the intrinsically high viscosity of the materials with increasing clay loading, which were subjected to foam processing. In con-trast, the cell density of the foams behaved in the opposite way. The char-acteristic parameters of pre- and post-formed samples are listed in Table 6.5. The function for determining cell density (Nc ) in cells/cm3 is defi ned in the following equation:6

4

3

3[1 ( / )]10

4

f pcN

d

r rp

-=

(6.2)

On the other hand, the mean cell wall thickness (�) in �m was estimated by the following equation:6

d � d(1/ √ _________

1 - (rf/rp) - 1) (6.3)

Figure 6.11 shows the TEM images on the structure of the mono-cell wall (a) and the junction of three cell walls (b) for PPCN4 foamed at 134.7°C. In Figure 6.11a, the dispersed clay particles in the cell wall align along

7.5

Clay content (wt %)

4

2

0

Temperature (oC) 130.6°C 134.7°C 139.2°C 143.4°C

200 μm

FIGURE 6.9 SEM images for PP-MA and various PPCNs foamed at different temperature.

[From Nam, P. H., Okamoto, M., Maiti, P., et al. Polymer Engineering Science 42 (2002): 1907–

1918. © 2002 Society of Plastic Engineers. With permission.]

192 Polymeric Foams

61259_C006.indd 19261259_C006.indd 192 10/25/2008 12:38:11 PM10/25/2008 12:38:11 PM

PP-MA

Cell size (μm)

40

30

20

10

0

18

12

6

0

12

8

4

0

12

8

4

00 20 40 60 80 100 120 140 160 180 200

Fra

ctio

n (%

)

PPCN7.5134.7°C

PPCN4134.7°C

PPCN2134.7°C

PPCN2134.7°C

134.7°C

FIGURE 6.10 Typical example for cell size distribution of foamed PP-MA and PPCNs in

experiment at 134.7°C. Average values of d in �m and variances �2d in �m2 in the Gaussian fi t

through the data are: 122.1 and 12.1 for PP-MA foam; 95.1 and 9.8 for PPCN2 foam; and 64.4

and 3.1 for PPCN4 foam. [From Nam, P. H., Okamoto, M., Maiti, P., et al. Polymer Engineering Science 42 (2002): 1907–1918. © 2002 Society of Plastic Engineers. With permission.]

the interface between the solid and gas phase. In other words, the clay particles arrange along the boundary of cells. The orientation angle of the dispersed clay particles (versus cell boundary), calculated statistically from TEM photographs, is about 5 ± 3.6°, indicating that plane orientation of the dispersed clay particles to the cell boundary occurred. In a previous paper for PPCN4 melt,27 the perpendicular alignment of the clay particles to stretching or elongating direction was shown, which was the main reason for causing the strain-induced hardening in the uniaxial elonga-tional viscosity.

In this foam processing, apparently, a similar structure is formed, prob-ably by a different mechanism. Due to the biaxial fl ow of material during foam process, the clay particles probably either turned their face (marked


61259_C006.indd 19361259_C006.indd 193 10/25/2008 12:38:12 PM10/25/2008 12:38:12 PM

with the arrows (A) in Figure 6.11a or fi xed face orientation [marked with the arrows (B) in Figure 6.11a] and aligned along the fl ow direction of materials; that is, along the cell boundary. The interesting point here is that such aligning behavior of the clay particles may help cells to with-stand the stretching force from breaking the thin cell wall; in other words, to improve the strength of foam in mechanical properties. The clay parti-cles seem to act as a secondary cloth layer to protect the cells from being destroyed by external forces. How do such unique alignments represent an improvement in mechanical properties?

The compression modulus K� of the foams are shown in Table 6.6. The K� of the PPCN foams appears higher than that of PP-MA foam even though they have the same �f level. This may create the improvement of mechani-cal properties for polymeric foams through polymeric nanocomposites.

TABLE 6.5

Characteristic Parameters of Pre- and Post-Foamed PP and Various PPCNs

Sample

Tf

(°C)�

(g mL�1)

d

(�m)

Nc (cells

mL�1 107)�

( mm)

PP-MA 0.854

PP-MA foam 130.6 0.219 74.4 1.8 11.88

PP-MA foam 134.7 0.114 122.1 0.48 9.07

PP-MA foam 139.2 0.058 155.3 0.25 5.56

PP-MA foam 143.4 0.058 137.3 0.35 6.46

PPCN2 0.881

PPCN2 foam 130.6 0.213 72.5 1.99 10.76

PPCN2 foam 134.7 0.113 95.1 1.01 6.76

PPCN2 foam 139.2 0.058 133.3 0.39 4.62

PPCN2 foam 143.4 0.113 150.3 0.26 10.68

PPCN4 0.900

PPCN4 foam 130.6 0.423

PPCN4 foam 134.7 0.196 64.4 2.92 8.41

PPCN4 foam 139.2 0.193 93.4 0.96 11.98

PPCN4 foam 143.4 0.341 56.1 3.52 15.08

PPCN7.5 0.921

PPCN7.5 foam 130.6 0.473

PPCN7.5 foam 134.7 0.190 35.1 18.35 4.30

PPCN7.5 foam 139.2 0.131 33.9 22.00 2.70

PPCN7.5 foam 143.4 0.266 27.5 34.2 5.11

Source: From Nam, P. H., Okamoto, M., Maiti, P., et al. Polymer Engineering Science 42 (2002):

1907–1918. © 2002 Society of Plastic Engineers. With permission.

194 Polymeric Foams

61259_C006.indd 19461259_C006.indd 194 10/25/2008 12:38:12 PM10/25/2008 12:38:12 PM

In Figure 6.11b, besides the alignment of clay particles, we can observe a random dispersion of clay in the central area of the junction (marked with the arrow in Figure 6.11b). Such behavior of clay particles presumably refl ects the effect of stagnation fl ow region of material under the growth of three contacting cells.

Figure 6.12 shows the stress–strain curves and the strain recovery behav-ior of the PP-based nanocomposite (PPCN) foams28 in the compression mode at a constant strain rate of 5% min�1. The nanocomposite foams exhibit high modulus compared to neat PP-g-MA foam. The residual strain is 17%

(a)

(A)

(B)

(b)

Cell boundary

Cell boundary 500 nm200 nm

FIGURE 6.11 TEM micrographs for PPCN4 foamed at 134.7°C: (a) mono-cell wall and

(b) junction of three contacting cells. [From Okamoto, M., Nam, P. H., Maiti, M., et al. Nano Letters 1 (2001): 503–505. © 2001 American Chemical Society. With permission.]

TABLE 6.6

Morphological Parameters and Compression Modulus of PP and PPCN Foams

Foam

Samples�f

(g cm�3)

d

(�m)NC � 10�6 (cell

cm�3)

�

(�m)

K�*(MPa)

PP-MA 0.06 155.3 2.49 5.6 0.44

PPCN2 0.06 133.0 3.94 4.6 1.72

PPCN4 0.12 93.4 9.64 11.9 1.95

PPCN7.5 0.13 33.9 220 2.7 2.80

* At 25°C.

Source: From Okamoto, M., Nam, P. H., Maiti, M., et al. Nano Letters 1 (2001): 503–505.

© 2001 American Chemical Society. With permission.


61259_C006.indd 19561259_C006.indd 195 10/25/2008 12:38:12 PM10/25/2008 12:38:12 PM

for PPCN2 (MMT � 2 weight-percentage) as well as neat PP foam, provid-ing the excellent strain recovery and the energy dissipation mechanism, probably with the “house-of-cards” structure formation in the cell wall, which enhances the mechanical properties of the nanocomposites like a spruce wood which is close to right-handed helix (see Figure 6.13).33

6.4.2 In-Situ Observation of Foaming

To understand the complex mechanism of physical foaming, Taki et al. studied the dynamic behavior of bubble nucleation and growth in the batch foaming of PP-based nanocomposites.34 Employing image-processing techniques, the bubble nucleation and growth rate for different nano-composites are analyzed from the series micrographs. Together with the solubility and diffusivity of CO2 into the PP matrix, the mechanism of nanocomposite foaming is investigated.

Figure 6.14 shows a schematic diagram of the visual observation appa-ratus for batch physical foaming. It consists of a high-pressure cell, a gas supply line and a pump with a gas cylinder. The high-pressure cell is made of stainless steel and has two sapphire windows on the walls. The C-shape stainless steel is used for a spacer. A signal processing board (DITECT, Japan; HAS-PCI) is installed so as to record a series of micro-graphs onto an online computer.

Figure 6.15 shows the series of micrographs of PP-MA (upper) and PPCN7.5 (lower) foaming at 150°C under a pressure of 13 MPa. The

FIGURE 6.12 Stress–strain curves and strain recovery behavior of the PP-based nano-

composites (PPCNs). [From Okamoto, M. “Polymer/layered silicate nanocomposites.”

Rapra Review Report No. 163, Rapra Technology Ltd, London, 2003. 166 pp. © 2003 Rapra

Technology Ltd. With permission.]

1.0

0.8

0.6

0.4

Str

ess

(MP

a)

0.2

0

0 10 20

Strain (%)

30 40

Compression set (%)

PP-MA 18 PPCN2 17 PPCN4 29 PPCN7.5 41

50

196 Polymeric Foams

61259_C006.indd 19661259_C006.indd 196 10/25/2008 12:38:14 PM10/25/2008 12:38:14 PM

Z-HelixCellulose

fibrils

μ42 μm

FIGURE 6.13 X-ray microdiffraction experiment with a 2-μm-thick section of spruce wood

embedded in resin. Note the asymmetry of the patterns in the enlargement (far left) which

can be used to determine the local orientation of cellulose fi brils in the cell wall (arrows).

The arrows are plotted in the right image with the convention that they represent the pro-

jection of a vector parallel to the fi brils onto the plane of the cross-section. The picture

clearly shows that all cells are right-handed helices. [From Fratzl, P. Current Opinion in Colloid Interface Science 8 (2003): 32–39. © 2003 Elsevier Science. With permission.]

FIGURE 6.14 A schematic diagram of the visual observation apparatus for batch physical

foaming. [From Taki, K., Yanagimoto, T., Funami, E., Okamoto, M., and Ohshima, M. Polymer Engineering Science 44 (2004): 1004–1011. © 2004 Society of Plastic Engineers. With

permission.]


61259_C006.indd 19761259_C006.indd 197 10/25/2008 12:38:15 PM10/25/2008 12:38:15 PM

dynamic behavior of bubble nucleation and growth in the very early stages of foaming can be seen in Figure 6.16. The bubble nucleation rate and the fi nal density of bubbles were highest at PPCN7.5 foaming. Although a distinct difference in bubble nucleation rate as well as in the fi nal bubble density could not be observed between PPCN2 and PPMA foaming, the nucleation rate and the fi nal bubble density increased as the weight fraction of clay increased. Furthermore, the induction time became shorter as the clay content increased.

The bubble growth rate is quantifi ed by measuring temporal change in cross-sectional area of each bubble. Figure 6.17 shows the representative growth rate of the bubbles born at the designated time in PP-MA and nano-composite foaming. Since the change in cross-sectional area of bubbles can be approximated by a linear function of time as mentioned above, the bub-ble growth observed by micrographs is a mass transfer-controlled process. Therefore, it can be said that the clay content changes the mass transfer rate of CO2 from the matrix polymer to the bubbles. The clay particles decrease the diffusivity of CO2 while keeping the solubility of CO2 in the matrix polymer the same. Owing to the clay-induced diffusivity depression, the

PPCN7.5 (PP+clay7.5%)

0 s 1 s 2 s

0 s 1 s 2 s

PPMA (PP+clay0%)

1.2

mm

1.6 mm

1.2

mm

1.6 mm

FIGURE 6.15 Series of micrographs of foaming: PP-MA (upper), PPCN7.5 (lower). The

black dots are bubbles and white part is the polymer matrix. The color of the bubbles in the

micrograph appear black because the bubbles refl ect the light entering from the opposite-

side window of the high-pressure cell. [From Taki, K., Yanagimoto, T., Funami, E., Okamoto, M.,

and Ohshima, M. Polymer Engineering Science 44 (2004): 1004–1011. © 2004 Society of Plastic


198 Polymeric Foams

61259_C006.indd 19861259_C006.indd 198 10/25/2008 12:38:16 PM10/25/2008 12:38:16 PM

0

2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0

200

400

600

800

1000

1200

Num

ber

dens

ity o

f nuc

leat

ed b

ubbl

e (1

/mm

3 )

Time elapsed after the pressure release starts (s)

PPMA PPCN2 PPCN4 PPCN7.5

FIGURE 6.16 Time variation in number density of nucleated bubble of PP-MA and

nanocomposite foaming. [From Taki, K., Yanagimoto, T., Funami, E., Okamoto, M., and

Ohshima, M. Polymer Engineering Science 44 (2004): 1004–1011. © 2004 Society of Plastic


Rep

rese

ntat

ive

aver

age

grow

th r

ate

(μm

2 /s)

2.5 3.0 3.5 4.0 4.5 5.0 5.5

Time elapsed after the pressure release starts (s)

0

2000

4000

6000

8000

10000

12000

PPMA

PPCN2

PPCN4

PPCN7.5

FIGURE 6.17 Representative average growth rates for PP-MA and nanocomposite

foaming. [From Taki, K., Yanagimoto, T., Funami, E., Okamoto, M., and Ohshima, M. Polymer Engineering Science 44 (2004): 1004–1011. © 2004 Society of Plastic Engineers. With

permission.]


61259_C006.indd 19961259_C006.indd 199 10/25/2008 12:38:17 PM10/25/2008 12:38:17 PM

increase in clay content depresses the mass transfer of CO2 from the matrix polymer to the bubbles. As a result, the bubble growth rate is decreased.

6.4.3 PLA-Based Nanocomposite Foaming

Figure 6.18 shows the typical results of SEM images of the fracture surfaces of the PLA/MMT-ODA and neat PLA without clay foamed at a tempera-ture range of 100°C to 140°C under the different isobaric saturation condi-tions (14, 21, and 28 MPa).35 All foams exhibit the neat closed-cell structure. We noted here that homogeneous cells were formed in the case of nanocom-posite foams, while neat PLA foams show rather non-uniform cell structure having large cell size. The nanocomposite foams show smaller cell size (d) and larger cell density (Nc) compared with neat PLA foam, suggesting that the dispersed silicate particles act as nucleating sites for cell formation.5

For both foam systems, the calculated distribution function of cell size from SEM images are presented in Figure 6.19. The nanocomposite foams nicely obeyed the Gaussian distribution. In the case of PLA/ODA foamed at 150°C under high pressure of 24 MPa, we can see that the width of the distribution peaks, which indicates the dispersity for cell size, became narrow accompanied by fi ner dispersion of silicate particles.

Obviously, with decreasing saturation pressure condition (~140°C and 14 MPa), both foams exhibit large cell size due to the low supply of CO2

×7500

×1000

2 μm

20 μm

PC

O2 (

MP

a)

Tf (°C)

120100 140140

14

21

28

PLA/MMT-ODA PLA

FIGURE 6.18 Typical results of SEM images of the fracture surfaces of PLA/MMT-ODA

and neat PLA foamed at temperature range of 100°C to 140°C under different isobaric satu-

ration condition (14, 21, and 28 MPa). [Reprinted from Ema, Y., Ikeya, M., and Okamoto, M.

Polymer 47 (2006): 5350–5359. © 2006, Elsevier Science. With permission.]

200 Polymeric Foams

61259_C006.indd 20061259_C006.indd 200 10/25/2008 12:38:17 PM10/25/2008 12:38:17 PM

molecules, which can subsequently form a small population of cell nuclei upon depressurization. The incorporation of nanoclay (OMLS) induces heterogeneous nucleation because of a lower activation energy barrier compared with homogeneous nucleation.36 However, the competi-tion between homogeneous and heterogeneous nucleation is no longer discernible.

6.4.4 Foaming Temperature Dependence of Cellular Structure

The dependence of the foam density (�f) at the Tf under different CO2 pres-sures are shown in Figure 6.20. Throughout the whole CO2 pressure range, the mass density of PLA/MMT-ODA foams remains at a constant value at low foaming temperature (Tf) range and abruptly decreases beyond a certain Tf , and then attains a minimum constant value up to 150°C again. From the above results, it can be said that such behavior of mass density is due to the competition between cell nucleation and cell growth. At the low Tf range (~110°C), in which a large supply of CO2 molecules are pro-vided, the cell nucleation is dominant, while at the high Tf , (~140°C), cell growth and the coalescence of cells are prominent due to low viscosity of the systems compared with the low Tf range (~110°C). This behavior clearly appears in the plots of the cell size (� 2d), the cell density (Nc), and the mean cell wall thickness (d) versus Tf under various pressure condi-tions, respectively. As seen in Figure 6.21, with increasing Tf all nanocom-posite foams show an increasing tendency of 2d and/or �

. and attain a

00 10 20 30 40 50 60 70 80 90 100

10

20

30

40

50

PLAPLA/MMT-ODA

Fra

ctio

n (%

)

Cell size (μm)

FIGURE 6.19 Typical example for cell size distribution of foamed PLA/MMT-ODA and

neat PLA in experiments at 150°C under 24 MPa. Average values d in �m and variances �2d

in �m2 in the Gaussian fi t through the data are 24.2 and 19.1 for PLA/MMT-ODA foam, and

58.3 and 171.0 for PLA foam. [Reprinted from Ema, Y., Ikeya, M., and Okamoto,

M. Polymer 47 (2006): 5350–5359. © 2006, Elsevier Science. With permission.]


61259_C006.indd 20161259_C006.indd 201 10/25/2008 12:38:18 PM10/25/2008 12:38:18 PM

maximum. On the other hand, the temperature dependence of Nc shows opposite behavior compared with the tendency of 2d due to cell growth and coalescence. Both 2d and Nc affect the mass density of the foams.

Using Tg depressions (corresponding to �Tg), reconstructed plots of �f versus Tf � �Tg were drawn from the data of Figure 6.20. The results are shown in Figure 6.22. All of the data, including neat PLA and PLA/MMT-SBE, neatly conform to a reduced curve with �f ~ 1.0 ± 0.1 g/cm3 at Tf � �Tg � 140 ± 4°C (nanocellular region), whereas �f values approach around 0.3 ± 0.15 g/cm3 as reduced temperature (Tf � �Tg) increased well above 150°C (microcellular region). The critical temperature is thus 140 ± 4°C, above which cell growth prevails. Below the critical tempera-ture, cell nucleation dominates and cell growth is suppressed due to the high modulus and viscosity as revealed by the temperature dependence of stage, G�(�), and loss, G��(�), moduli (G� � 162 MPa and viscosity com-ponent G��/� � 2 MPa s at 140°C). Figure 6.23 shows temperature-reduced plots of 2d, Nc and � versus Tf � �Tg. All data nicely conform to a reduced curve such as in Figure 6.22. Interestingly, when both Tg and Tm depressions were used37 to conduct superposition, it was recognized that the reduced curve is neatly constructed but there is no signifi cant difference in compari-son with the case of Tf � �Tg. This indicates that Tg depression is important in optimizing foam processing conditions but Tm depression may be not a signifi cant factor for processing because the Tf range is still below Tm after CO2 saturation.

In Figure 6.24, the relationship between 2d and Nc, and � and 2d in this study are shown. The relationship neatly obeys Equations 6.2 and 6.3 but

0.2

0.4

0.6

0.8

1.0

1.2

90 100 110 120 130 140 150 160

142118242830

Tf (°C)

PCO2 (MPa)

r f (g

cm

–3)

FIGURE 6.20 Foaming temperature dependence of mass density for PLA/MMT-ODA

foamed under different CO2 pressure conditions. [From Ema, Y., Ikeya, M., and Okamoto,

M. Polymer 47 (2006): 5350–5359. © 2006, Elsevier Science. With permission.]

202 Polymeric Foams

61259_C006.indd 20261259_C006.indd 202 10/25/2008 12:38:18 PM10/25/2008 12:38:18 PM

the deviation occurs beyond the value of Nc ~ 1012 cell/cm�3 for panel (a) and below 2d ~ 1 �m for panel (b). The downward and upward deviations indicate that the heterogeneous cell distribution mechanism due to the rigid crystalline phases in the PLA matrix is caused by a high degree of crys-tallinity (~49 wt%) under the low foaming temperature range (~100°C). As seen in Figure 6.25, the PLACN foams exhibit a heterogeneous cell distri-bution. The PLA foam reduces the value of Nc accompanied by a large value of �

. compared with that of PLACN foams. In the case of PLACN foams,

the controlled structure of the PLACN foams is from microcellular (2d � 30 �m and Nc � 3.0 × 107 cells/cm3) to nanocellular (2d � 200 nm and Nc � 2.0 × 1013 cells/cm3).

141821242830

PCO2 (MPa)

90 100 110 120 130 140 150 160

10–1

100

101

102

1012

1010

108

106

101

100

10–1

10–2

2d (

μm)

Nc

(μm

)

Tf (°C)

(c)

(b)

(a)

d (m

m)

FIGURE 6.21 Foaming temperature dependence of: (a) cell size; (b) cell density; and (c)

mean cell wall thickness under different CO2 pressure conditions. [From Ema, Y., Ikeya, M.,

and Okamoto, M. Polymer 47 (2006): 5350–5359. © 2006 Elsevier Science. With permission.]


61259_C006.indd 20361259_C006.indd 203 10/25/2008 12:38:18 PM10/25/2008 12:38:18 PM

6.4.5 CO2 Pressure Dependence

At high pressure, both homogeneous and heterogeneous nucleation mechanisms may appear to be of comparable signifi cance. All systems demonstrate that Nc increases systematically with increasing CO2 pressure in the low Tf region (~100–120°C). For PLA/MMT-ODA foams, the system suggests that the heterogeneous nucleation is favored in high-pressure conditions. The cell nucleation in the heterogeneous nucleation system such as PLA/MMT-ODA foams took place in the boundary between the matrix and the dispersed nanoclay particles. Accordingly, the cell size decreased without individual cell coalescence for PLA/MMT-ODA and neat PLA systems, as seen in Figure 6.25. To clearly investigate whether the addition of internal surfaces of the dispersed nanoclay may hinder CO2 diffusion by creating a more tortuous diffusive pathway,19 character-ization of the interfacial tension between bubble and matrix was conducted using the modifi ed classical nucleation theory.36

According to the theory proposed by Suh and Colton, the rate of nucle-ation of cells per unit volume (N

.) can be written as

�2

3

2CO B

16 ( )~ exp

3( )

SN Cf

P k Tpg qÈ ˘-

Í ˙DÍ ˙Î ˚

(6.4)

0

0.2

0.4

0.6

0.8

1.0

1.2

120 130 140 150 160 170

PLAPLA/MMT-ODAPLA/MMT-SBE

r f (g

cm

–3)

Tf + ΔT (°C)

FIGURE 6.22 Plot of mass density for PLA/MMT-ODA, PLA/MMT-SBE and neat PLA

versus reduced foaming temperature (Tf � �Tg). The critical temperature (140 ± 4°C) is

shaded. [From Ema, Y., Ikeya, M., and Okamoto, M. Polymer 47 (2006): 5350–5359. © 2006,

Elsevier Science. With permission.]

204 Polymeric Foams

61259_C006.indd 20461259_C006.indd 204 10/25/2008 12:38:19 PM10/25/2008 12:38:19 PM

where C is the concentration of CO2 and/or the concentration of heteroge-neous nucleation sites, f is the collision frequency of CO2, � is the inter-facial tension between bubble and matrix, S(�) is the energy reduction factor for the heterogeneous nucleation (i.e. PLA/MMT-ODA), �PCO2

is the magnitude of the pressure quench during depressurization, kB is the Boltzmann constant, and T is absolute temperature.

The theoretical cell density is given by

�0

t

theorN N dt= Ú

(6.5)

PLAPLA/MMT–ODAPLA/MMT–SBE

120 130 140 150 160 170

10–1

100

101

102

1012

1010

108

106

101

100

10–1

10–2

Nc

(μm

)2d

(μm

)

Tf + ΔT (°C)

(a)

(b)

(c)

1014

104

103

d (μ

m)

FIGURE 6.23 Temperature-reduced plots of: (a) 2d; (b) Nc; and (c) � versus Tf � �Tg for

PLA/MMT-ODA, PLA/MMT-SBE and neat PLA. [From Ema, Y., Ikeya, M., and Okamoto, M.

Polymer 47 (2006): 5350–5359. © 2006 Elsevier Science. With permission.]


61259_C006.indd 20561259_C006.indd 205 10/25/2008 12:38:19 PM10/25/2008 12:38:19 PM

where t is the foaming time that takes approximately 3 s. Assuming no effect of the coalescence of cell on the value of Nc, we estimate the inter-facial tension of the systems calculated using Equations 6.4 and 6.5; that is, the slope of the plots (Nc versus 1/�PCO2

). The characteristic parameters of two systems are shown in Table 6.7. The interfacial tension of PLA/MMT-ODA and neat PLA are 6.65 mJ/m2 and 7.43 mJ/m2 at 110°C, respectively. These estimated �

. values are in good agreement with that of the other

poly(methyl methacrylate) (PMMA)-CO2 system (~10 mJ/m2).38 The PLA/MMT-ODA system has a low value compared to that of neat PLA. This trend refl ects the relative importance of heterogeneous nucleation, which

103(a)

102

101

1002d (

mm)

10–1

10–2

102(b)

101

100

d (m

m)

10–1

10–2

104 106 108 1010

Nc (cm–3)1012

PLA

PLA/MMT-ODA

PLA/MMT-SBE

PLA

PLA/MMT-ODA

–1/3

1

PLA/MMT-SBE

1014

10–2 10–1 100 101

2d (mm)102 103

FIGURE 6.24 Relationship between: (a) cell size versus cell density; and (b) cell wall thick-

ness versus cell size for all foams. [From Ema, Y., Ikeya, M., and Okamoto, M. Polymer

47 (2006): 5350–5359. © 2006 Elsevier Science. With permission.]

206 Polymeric Foams

61259_C006.indd 20661259_C006.indd 206 10/25/2008 12:38:19 PM10/25/2008 12:38:19 PM

FIGURE 6.25 SEM images of the fracture surfaces of: (a) neat PLA; (b) PLA/MMT-ODA;

and (c) PLA/MMT-SBE foamed at 100°C under 28 MPa. [From Ema, Y., Ikeya, M., and

Okamoto, M. Polymer 47 (2006): 5350–5359. © 2006 Elsevier Science. With permission.]


61259_C006.indd 20761259_C006.indd 207 10/25/2008 12:38:19 PM10/25/2008 12:38:19 PM

dominates over the homogeneous one in the event that the amount of CO2 available for bubble nucleation is limited because of a lower activation energy barrier, as mentioned previously. That is, in the heterogeneous nucleation (PLA/MMT-ODA), we have to take the reduction of the critical energy into consideration because of the inclusion of nucleants, which is a function of the PLA-gas-nanoclay contact angle (�) and the relative curvature (W) of the nucleant surface to the critical radius of the nucleated phase.39 In the case of W � 10, the energy reduction factor S(�) can be expressed:

S(�) � (1/4) (2 � cos �)(1 � cos �)2 (6.6)

In the case of homogeneous nucleation S(�) is unity (� � 180°). The obtained values of the contact angle are 107.3° at 110°C and 85.3° at 120°C. The estimated reduction factor [S(�) � 0.4–0.7] was not so small when we compared with the other nanofi llers [e.g. carbon nanofi bers, S(�) � 0.006].40 However, experimentally, nanoclay particles lead to an increase in Nc.

For PLA/MMT-SBE foams prepared under condition with low Tf (~100–110°C) and high pressure (~28 MPa), the nanocomposite foams exhibit no signifi cant difference in Nc compared with PLA/MMT-ODA foams. This reasoning is consistent with the large value of W in both systems.

6.4.6 TEM Observation

To confi rm the heterogeneous nucleation and the nanocellular features of foam processing, Okamoto et al. conducted TEM observation of the cell wall in the PLA/MMT-ODA foam.

Figure 6.26 shows a TEM micrograph for the structure of the cell wall foamed at 100°C under 28 MPa. Interestingly, the grown cells having a diameter of ~200 nm are localized along the dispersed nanoclay particles in the cell wall. In other words, the dispersed nanoclay particles act as nucleating sites for cell formation and the cell growth occurs on the sur-faces of the clays; that is, the cellular structure has an oval-faced morphol-ogy rather than spherical cellular structures for high Tf conditions (~140°C).

TABLE 6.7

Characteristic Interfacial Parameters of Two Systems

Tf (°C) �S()1/3 (mJ/m2) S() (°)

PLA/CO2 110 7.43

PLA/MMT-ODA/CO2 6.65 0.717 107.3

PLA/CO2 120 7.08

PLA/MMT-ODA/CO2 5.38 0.439 85.3

Source: From Ema, Y., Ikeya, M., and Okamoto, M. Polymer 47 (2006): 5350–5359.

© 2006 Elsevier Science. With permission.

208 Polymeric Foams

61259_C006.indd 20861259_C006.indd 208 10/25/2008 12:38:20 PM10/25/2008 12:38:20 PM

In Figure 6.26, in addition to the nanocellular structure formation, we can observe a lammellar pattern beside the nanoclay particles. This behavior appears to arise from the formation of the -phase of the PLA crystal in the presence of nanoclay particles.41 This is a unique observation of the epitaxial crystallization of PLA grown up from clay surfaces due to the nucleation effect of the dispersed nanoclays.

6.4.7 Polycarbonate (PC)-Based Nanocomposite Foams

Figure 6.27 shows the typical results of SEM images of the fracture sur-faces of the PC/SMA blend (matrix) and PC-based nanocomposites foamed at 140°C under different isobaric saturation condition (10, 18, and 24 MPa).42 PC/SMA foams exhibit polygon closed-cell structures having pentagonal and hexagonal faces, which express the most energetically stable state of polygon cells. Obviously, under low saturation CO2 pressure (~10 MPa), both PC/SMA/MAE1 and PC/SMA/MAE2.5 foams exhibit larger cell

FIGURE 6.26 TEM micrograph for the structure of PLA/MMT-ODA cell wall foamed

at 100°C under 28 MPa. [From Ema, Y., Ikeya, M., and Okamoto, M. Polymer 47 (2006):

5350–5359. © 2006 Elsevier Science. With permission.]

lamellar

nano-clay

nano-clay

nanocell

200 nm


61259_C006.indd 20961259_C006.indd 209 10/25/2008 12:38:20 PM10/25/2008 12:38:20 PM

size compared with PC/SMA, indicating that dispersed clay particles hinder CO2 diffusion by creating a maze or a more tortuous path.8 However, high CO2 pressure (~24 MPa) provides a large supply of CO2 molecules, which can subsequently form a large population of cell nuclei upon depressurization. The incorporation of nanoclay hinders CO2 diffu-sion and simultaneously induces heterogeneous nucleation because of a lower activation energy barrier compared with homogeneous nucleation.

In Figure 6.28, the relationship between d and Nc, and � and d are plotted. Equations 6.2 and 6.3 lead to these relations but some deviation occurs in each system. For example, PC/SMA/MAE2.5 (syn-FH-C18TM) (MMT � 1 wt%) exhibits a smaller value of Nc under the same d value when compared with PC/SMA and PC/SMA/MAE1. For the relationship between � and Nc, PC/SMA/MAE2.5 (MMT � 2.5 wt%) shows a large value of � compared with PC/SMA/MAE1. These deviations indicate that the heterogeneous cell distribution mechanism due to the rigid matrix phases in PC/SMA is caused by high MAE loading (MMT � 2.5–5.0 wt%) as seen in Figure 6.28. As well as PLA-based nanocomposite foams, in Table 6.8, the interfacial tension and energy reduction factor of systems are summarized.42

The interfacial tension of PC/SMA/MAE1 (including energy reduction factor S(�)) and PC/SMA (S(�) � 1) are 9.7 mJ/m2 and 10.9 mJ/m2 at 14 MPa, respectively. These estimated values of � are of the same order of magnitude compared with the PLA system (5–7 mJ/m2). The interfacial tension slightly decreases with an increase in clay content under the same CO2 pressure conditions, indicating heterogeneous cell nucleation occurs

FIGURE 6.27 Typical SEM images of the fracture surfaces of the PC/SMA blend (matrix)

and PC-based nanocomposites foamed at 140°C under different isobaric condition (10, 18,

and 24 MPa). [From Ito, Y., Yamashita, M., Okamoto, M. Macromolecular Materials Engineering


210 Polymeric Foams

61259_C006.indd 21061259_C006.indd 210 10/25/2008 12:38:21 PM10/25/2008 12:38:21 PM

easily with increasing clay content. The estimated reduction factor [S(�) � 0.3–0.8] is the same order compared with the foaming of PLA-based nanocomposites.35 The TEM micrograph of the structure of the cell wall foamed at 160°C is shown in Figure 6.29. The grown cells are local-ized along the dispersed nanoclay particles in the cell wall.

6.4.8 Mechanical Properties of Nanocomposite Foams

Figure 6.30 shows the relationship of relative modulus (Kf/Kp) against rela-tive density (�f/�p) of neat PLA and nanocomposite foams, taken in the parallel (a) and perpendicular (b) directions to the fl ow, respectively.

1000(a)

100

10

d (

mm)

1

0.1

1000(b)

100

10

1

d (m

m)

1

0.1

105 107

–1/3

109 1011

Nc cells (cm–3)1013 1015

0.1 1 10 100d (mm)

1000

PC/SMAPC/SMA/MAE1PC/SMA/MAE2.5PC/SMA/MAE5

PC/SMAPC/SMA/MAE1PC/SMA/MAE2.5PC/SMA/MAE5

FIGURE 6.28 (a) Cell size versus cell density; and (b) cell wall thickness versus cell size for

PC/SMA and PC based nanocomposite foams. [From Ito, Y., Yamashita, M., and Okamoto, M.

Macromolecular Materials Engineering 291 (2006): 773–783. © 2006 Wiley-VCH. With permission.]


61259_C006.indd 21161259_C006.indd 211 10/25/2008 12:38:21 PM10/25/2008 12:38:21 PM

FIGURE 6.29 TEM micrograph of the structure of the cell wall foamed at 160°C. [From

Ito, Y., Yamashita, M., and Okamoto, M. Macromolecular Materials Engineering 291 (2006):

773–783. © 2006 Wiley-VCH. With permission.]

TABLE 6.8

Interfacial Tension [�S(�)1/3] Including Energy Reduction Factor [S(�)] of Systems

Systems �S()1/3 (mJ/m2) S()

PC/SMA-CO2 10.7 1.0

PC/SMA/MAE1-CO2 PCO2 � 10 MPa 8.6 0.53

PC/SMA/MAE2.5-CO2 8.0 0.42

PC/SMA-CO2 10.9 1.0


PC/SMA/MAE2.5-CO2 9.9 0.77

PC/SMA-CO2 13.6 1.0


PC/SMA/MAE2.5-CO2 9.2 0.30

PC/SMA-CO2 11.3 1.0

PC/SMA/MAE1-CO2 PCO2 � 22 MPa 12.4 —

PC/SMA/MAE2.5-CO2 8.0 0.36

Source: From Ito, Y., Yamashita, M., and Okamoto, M. Macromolecular Materials Engineering

291 (2006): 773–783. © 2006 Wiley-VCH. With permission.

212 Polymeric Foams

61259_C006.indd 21261259_C006.indd 212 10/25/2008 12:38:22 PM10/25/2008 12:38:22 PM

To clarify whether the modulus enhancement of the nanocomposite foams was reasonable, we applied the following Equation 6.7 proposed previously by Kumar43 to estimate relative moduli with various foam densities:

Kf

___ Kp � (

rf __ rp )

4

� ( rf

__ rp )

4

� ( rf

__ rp )

4 (6.7)

where Kp and Kf are the modulus of pre-foamed and post-foamed samples, respectively. The solid line in the fi gure represents the fi t with Equation 6.7.

2.5(a)

2.0

1.5

1.0

Kf/K

p

0.5

0

2.5(b)

2.0

1.5

1.0

Kf/K

p

0.5

0

0 0.2 0.4 0.6 0.8

PLAPLA/MMT-ODAPLA/MMT-SBE

1

1

0 0.2 0.4 0.6rf /rp

rf /rp

0.8 1

FIGURE 6.30 The relation of relative modulus (Kf/Kp) against relative density (�f/�p) of neat

PLA and PLA-based nanocomposite foams, taken in directions parallel (a) and perpendicu-

lar (b) to the fl ow.


61259_C006.indd 21361259_C006.indd 213 10/25/2008 12:38:22 PM10/25/2008 12:38:22 PM

The neat PLA foams do not show any difference between the two moduli (a) and (b). On the other hand, for PLACN foams, the relative moduli exhibit a large value compared with the theoretical one. The dispersed clay particles in the cell wall align along the thickness direction of the sample. In other word, the clay particles arrange owing to the biaxial fl ow of material during foaming. The clay particles seem to act as a secondary cloth layer to protect the cells from being destroyed by external forces. In the directions perpendicular to the fl ow, the relative modulus of PLA/MMT-ODA and PLA/MMT-SBE foams appear higher than the predicted value even at the same relative mass density in the range 0.7–0.85 (see Figure 6.30a). This upward deviation suggests that the small cell size with large cell density enhances the material property as predicted by Weaire.44 This may create the improvement of mechanical properties for polymeric foams through polymeric nanocomposites. More detailed surveys on various types of nanocomposite foaming can be also be found in the literatures.45–48

6.4.9 Porous Ceramic Materials via Nanocomposite

A new route for the preparation of porous ceramic material from thermo-setting epoxy/clay nanocomposite was fi rst demonstrated by Brown et al.49 This route offers attractive potential for diversifi cation and applica-tion of the PLFNCs. Okamoto and coworkers have reported the results on the novel porous ceramic material via burning of the PLA/MMT system (PLACN).50 The PLACN contained 3.0 wt% inorganic clay. The SEM image of the fracture surface of porous ceramic material prepared from simple burning of the PLACN in a furnace of up to 950°C is shown in Figure 6.31. After complete burning, as seen in the fi gure, the PLACN becomes a white mass with a porous structure. The bright lines in the SEM image corre-spond to the edge of the stacked silicate layers. In the porous ceramic material, the silicate layers form a house-of-cards structure, which consist of the large plates having a length of ~1000 nm and thickness of ~30–60 nm. This implies that the further stacked platelet structure is formed dur-ing burning. The material exhibits the open-cell type structure having a 100–1000 nm diameter void, a BET surface area of 31 m2 g�1 and a low den-sity of porous material of 0.187 g ml�1 estimated by the buoyancy method. The BET surface area value of MMT (780 m2/g) and that of the porous ceramic material (31 m2/g), suggests about 25 MMT plates stacked together. When MMT is heated above 700°C (but below 960°C) all OH groups are fi rst eliminated from the structure and thus MMT is decom-posed into that of a non-hydrated aluminosilicate. This transformation radically disturbs the crystalline network of the MMT, and the resulting diffraction pattern is indeed often typical of an amorphous (or non-crys-talline) phase. The estimated rough value of the compression modulus (K) is of the order of ~1.2 MPa, which is fi ve orders of magnitude lower than the bulk modulus of MMT (~102 GPa).28 In the stress–strain curve, the

214 Polymeric Foams

61259_C006.indd 21461259_C006.indd 214 10/25/2008 12:38:23 PM10/25/2008 12:38:23 PM

linear deformation behavior is nicely described in the early stage of the deformation; that is, the deformation of the material closely resembles that of ordinary polymeric foams.51 This open-cell type porous ceramic material consisting of the house-of-cards structure is expected to provide strain recovery and an excellent energy dissipation mechanism after unloading in the elastic region up to 8% strain, probably each plate bend like leaf spring. This porous ceramic material is a new material possessing elastic feature and is very lightweight. This new route for the preparation of porous ceramic material via burning of nanocomposites can be expected to pave the way for a much broader range of applications of the PLSNCs. This porous ceramic material provides an excellent insulator fl ame retar-dant property for PLFNCs.28 The fl ame behavior must be derived from the morphological control of the shielding properties of the graphitic/clay created during polymer ablation.

6.5 Conclusions and Future Prospects

Development of nanocomposite foams is one of the latest evolutionary technologies of polymeric foams. The nanocomposite foams offer attrac-tive potential for diversifi cation and application of conventional polymeric materials. Some of them are already commercially available and applied in

FIGURE 6.31 SEM image of porous ceramic material after coating with platinum layer

(~10 nm thickness). [From Sinha Ray, S., Okamoto, K., Yamada, K., and Okamoto, M. Nano Letters 2 (2002): 423–425. © 2002 American Chemical Society. With permission.]


61259_C006.indd 21561259_C006.indd 215 10/25/2008 12:38:23 PM10/25/2008 12:38:23 PM

industrial products through Unitika Ltd., Japan, and open a new dimen-sion for plastics and composite foams. The major impact will be at least a decade away.

References

1. Sinha R., S. and Okamoto, M. “Polymer/layered silicate nanocomposites: A review from preparation to processing.” Progress in Polymer Science 28 (2003): 1539–1641.

2. Usuki, A., Kojima, Y., Okada, A., Fukushima, Y., Kurauchi, T., and Kamigaito, O. “Swelling behavior of montmorillonite cation exchanged for ω-Amino acids by ε-Caprolactam.” Journal of Material Research 8 (1993): 1174–1178.

3. Gao, F. “Clay/polymer composites: The story.” Materials Today 7 (2004): 50–55.

4. Usuki, A., Hasegawa, N., and Kato, M. “Polymer–Clay nanocomposites.” Advances in Polymer Science 179 (2005): 135–195.

5. Okamoto, M., Nam, P. H., Maiti, M., et al. “Biaxial fl ow-induced alignment of silicate layers in polypropylene/clay nanocomposite foam.” Nano Letters 1 (2001): 503–505.

6. Nam, P. H., Okamoto, M., Maiti, P., et al. “Foam processing and cellular structure of polypropylene/clay nanocomposites.” Polymer Engineering Science 42 (2002): 1907–1918.

7. Fujimoto, Y., Sinha Ray, S., Okamoto, M., Ogami, A., and Ueda, K. “Well-Controlled biodegradable nanocomposite foams: from microcellular to nanocellular.” Macromolecular Rapid Communications 24 (2003): 457–461.

8. Mitsunaga, M., Ito, Y., Sinha Ray, S., Okamoto, M., and Hironaka, K. “Inter-calated polycarbonate/clay nanocomposites: nanostructure control and foam processing.” Macromolecular Materials Engineering 288 (2003): 543–548.

9. Vaia, R. A., Ishii, H., and Giannelis, E. P. “Synthesis and properties of two-dimensional nanostructures by direct intercalation of polymer melts in layered silicates.” Chemistry of Materials 5 (1993): 1694–1696.

10. Vaia, R. A. and Giannelis, E. P. “Lattice model of polymer melt intercalation in organically-modifi ed layered silicates.” Macromolecules 30 (1997): 7990–7999.

11. Vaia, R. A. and Giannelis, E. P. “Polymer melt intercalation in organically-modifi ed layered silicates: Model predications and experiment.” Macromolecules 30 (1997): 8000–8009.

12. Hiroi, R., Sinha Ray, S., Okamoto, M., and Shiroi, T. “Organically modifi ed layered titanate: A new nanofi ller to improve the performance of biodegrad-able polylactide.” Macromolecular Rapid Communications 25 (2004): 1359.

13. Lagaly, G. Clay Minerals 16 (1970): 1. 14. Nakano, S., Sasaki, T., Takemura, K., and Watanabe, M. “Pressure-Induced

intercalation of alcohol molecules into a layered titanate.” Chemistry of Materials 10 (1998) 2044–2046.

15. Tateyama, H., Nishimura, S., Tsunematsu, K., Jinnai, K., Adachi, Y., and Kimura, M. “Synthesis of expandable fl uorine mica from talc.” Clay Minerals 40 (1992): 180–185.

216 Polymeric Foams

61259_C006.indd 21661259_C006.indd 216 10/25/2008 12:38:23 PM10/25/2008 12:38:23 PM

16. Yoshida, O. and Okamoto, M. “Direct melt intercalation of polylactide chains into Nano-Galleries: Interlayer expansion and nanocomposite structure.” Macromolecular Rapid Communications 27 (2006): 751–757.

17. Sinha Ray, S., Yamada, K., Okamoto, M., and Ueda, K. “Biodegradable polylactide/montmorillonite nanocomposites.” Journal of Nanoscience and Nanotechnology 3 (2003): 503–510.

18. Yoshida, O. and Okamoto, M. “Direct intercalation of polymer chains into Nano-Galleries: Interdigitated layer structure and interlayer expansion.” Journal of Polymer Engineering 26 (2006): 919–940.

19. Sinha Ray, S., Yamada, K., Okamoto, M., Ogami, A., and Ueda, K. “New polylactide/layered silicate nanocomposites.3. High-Performance biode-gradable materials.” Chemistry of Materials 15 (2003): 1456–1465.

20. Saito, T., Okamoto, M., Hiroi, R., Yamamoto, M., and Shiroi, T. “Intercalation and interlayer expansion in the mixtures of organically modifi ed layered fi llers and Poly(p-Phenylenesulfi de).” Macromolecular Materials Engineering 291 (2006): 1367–1374.

21. Saito T., Okamoto M., Hiroi R., Yamamoto M., and Shiroi T. “Delamination of organically modifi ed layered fi ller via solid-state processing” Macromolecular Rapid Communications 27 (2006): 1472–1475.

22. Toth, R., Coslanicha, A., Ferronea, M., et al. “Computer simulation of polypro-pylene/organoclay nanocomposites: Characterization of atomic scale struc-ture and prediction of binding energy.” Polymer 45 (2004): 8075–8083.

23. Sinsawat, A., Anderson, K. L., Vaia, R. A., and Farmer, B. L. “Infl uence of polymer matrix composition and architecture on polymer nanocomposite formation: Coarse-Grained molecular dynamics simulation.” Journal of Polymer Science Part B: Polymer Physics 41 (2003): 3272–3284.

24. Kuppa, V., Menakanit, S., Krishnamoorti, R., and Manias, E. J. “Simulation insights on the structure of nanoscopically confi ned Poly(ethylene oxide).” Journal of Polymer Science Part B: Polymer Physics 41 (2003): 3285–3298.

25. Zeng, Q. H., Yu, A. B., Lu, G. Q., and Standish, R. K. “Molecular dynamics simulation of organic-inorganic nanocomposites: Layering behavior and interlayer structure of organoclays.” Chemistry of Materials 15 (2003): 4732–4738.

26. Sheng, N., Boyce, M. C., Parks, D. M., Rutledge, G. C., Abes, J. I., and Cohen, R. E. “Multiscale micromechanical modeling of polymer/clay nanocompos-ites and the effective clay particle.” Polymer 45 (2004): 487–506.

27. Okamoto, M., Nam, P. H., Maiti, P., Kotaka, T., Hasegawa, N., and Usuki, A. “A house of cards structure in polypropylene/clay nanocomposites under elongational fl ow.” Nano Letters 1 (2001): 295–298.

28. Okamoto, M. “Polymer/layered silicate nanocomposites.” “Polymer/layered silicate nanocomposites.” Rapra Review Report No. 163, Rapra Technology Ltd, London, 2003. 166 pp.

29. Maiti, P. and Okamoto, M. “Crystallization control via silicate surface in nylon 6-clay nanocomposites.” Macromolecular Materials Engineering 288 (2003): 440–445.

30. Nam, P. H. The Structure and Properties of Intercalated Polypropylene/Clay Nanocomposite, MSc thesis, Toyota Technological Institute, Nagoya 2001.

31. Sinha Ray, S., Okamoto, K., and Okamoto, M. “Structure-Property relation-ship in biodegradable poly(butylenes succinate)/layered silicate nanocom-posites.” Macromolecules 36 (2003): 2355–2367.


61259_C006.indd 21761259_C006.indd 217 10/25/2008 12:38:23 PM10/25/2008 12:38:23 PM

32. van Olphen, H. An Introduction to Clay Colloid Chemistry. Wiley, New York, 1977.

33. Fratzl, P. “Cellulose and collagen: From fi bres to tissues.” Current Opinion in Colloid Interface Science 8 (2003): 32–39.

34. Taki, K., Yanagimoto, T., Funami, E., Okamoto, M., and Ohshima, M. “Visual observation of CO2 Foaming of Polypropylene–Clay nanocomposites.” Polymer Engineering Science 44 (2004): 1004–1011.

35. Ema, Y., Ikeya, M., and Okamoto, M. “Foam processing and cellular structure of polylactide-based nanocomposites.” Polymer 47 (2006): 5350–5359.

36. Colton, J. S. and Suh, N. P. “The nucleation of microcellular thermoplastic foam with additives. 1. Theoretical considerations.” Polymer Engineering Science 27 (1987): 485–492.

37. Takada, M. Crystallization Control and Foam Processing of Semi-crystalline Polymers via Supercritical CO2. PhD thesis, Kyoto University, 2004.

38. Goel, S. K. and Beckman, E. J. “Generation of microcellular polymeric foams using supercritical carbon-dioxide.1. Effect of pressure and temperature on nucleation.” Polymer Engineering Science 34 (1994): 1137–1147.

39. Fletcher, N. H. “Size effect in heterogeneous nucleation.” Journal of Chemical Physics 29 (1958): 572–576.

40. Shen, J., Zeng, C., and Lee, L. J. “Synthesis of Polystyrene–Carbon nanofi bers nanocomposite foams.” Polymer 46 (2005): 5218–5224.

41. Nam, J. Y., Sinha, S. R., and Okamoto, M. “Crystallization behavior and morphology of biodegradable polylactide/layered silicate nanocomposite.” Macromolecules 36 (2003): 7126–7131.

42. Ito, Y., Yamashita, M., and Okamoto, M. “Foam processing and cellular structure of polycarbonate-based nanocomposites.” Macromolecular Materials Engineering 291 (2006): 773–783.

43. Kumar, V. and Weller, J. E. The 49th Annual Technical Conference (ANTEC) (1991): 1401.

44. Weaire, D. and Fu, T. L. “The mechanical-behavior of foams and emulsions.” Journal of Rheology 32 (1988): 271.

45. Lee, L. J., Zeng, C., Cao, X., Han, X., Shen, J., and Xu, G. “Polymer nanocom-posite foams.” Composites Science and Technology 65 (2005): 2344–2363.

46. Cao, X., Lee, L. J., Widya, T., and Macosko, C. “Polyurethane/clay nano-compo sites foams: Processing, structure and properties.” Polymer 46 (2005): 775–783.

47. Chandra, A., Gong, S., Turng, L. S., Gramann, P., and Cordes, H. “Microstructure and crystallography in microcellular injection-molded polyamide-6 nano-composite and neat resin.” Polymer Engineering Science 45 (2005): 52–61.

48. Strauss, W. and D’Souza, N. A. “Supercritical CO2 processed polystyrene nanocomposite foams.” Journal of Cellular Plastics 40 (2004): 229–241.

49. Brown, J. M., Curliss, D. B., and Vaia, R. A. Proceedings of the Polymer Materials Science Engineering Spring Meeting, San Francisco, 2000, p. 278.

50. Sinha Ray, S., Okamoto, K., Yamada, K., and Okamoto, M. “Novel porous ceramic material via burning of polylactide/layered silicate nanocomposite.” Nano Letters 2 (2002): 423–425.

51. Gibson, L. J. and Ashby, M. F. (eds) Cellular Solids: Structures and Properties Pergamon Press, New York, 1988, p. 8.

218 Polymeric Foams

61259_C006.indd 21861259_C006.indd 218 10/25/2008 12:38:23 PM10/25/2008 12:38:23 PM

7New Material Developments from the Nitrogen Autoclave Process

Neil Witten

CONTENTS

7.1 Introduction ................................................................................... 2207.2 Brief History .................................................................................. 2207.3 Nitrogen Autoclave Technology ................................................ 221

7.3.1 Raw Materials ...................................................................... 2217.3.2 Process Overview ............................................................... 2237.3.3 First Stage—Sheet Extrusion ............................................. 2247.3.4 Second Stage—High Pressure Autoclave ........................ 2257.3.5 Third Stage—Low Pressure Autoclave ........................... 2277.3.6 Quality .................................................................................... 2287.3.7 Alternative Cross-Linked Polyolefi n

Foam Technologies ............................................................. 2297.4 Block Processes ............................................................................... 2297.5 Semi-Continuous (Roll) Processes ............................................. 2327.6 Nitrogen Autoclave Cross-Linked Polyolefi n Foam Products ...... 233

7.6.1 Product Range ..................................................................... 2337.6.2 Cross-Linked High-Density Polyethylene Foams .......... 2367.6.3 Cross-Linked Metallocene Polyethylene Foams ............ 2367.6.4 Markets and Applications ................................................. 237

7.6.4.1 Packaging ................................................................. 2377.6.4.2 Sports and Leisure .................................................. 2387.6.4.3 Medical/Health Care ............................................. 2387.6.4.4 Other ......................................................................... 238

61259_C007.indd 21961259_C007.indd 219 10/25/2008 12:39:00 PM10/25/2008 12:39:00 PM

7.7 New Material Developments ...................................................... 2397.7.1 ZOTEK F Polyvinylidene Fluoride (PVDF) Foams ......... 239

7.7.1.1 Flammability Characteristics ................................. 2407.7.1.2 Environmental and Chemical Performance ........ 2417.7.1.3 Basic Thermal Properties ....................................... 2437.7.1.4 Other Performance Attributes ............................... 2447.7.1.5 Markets and Applications ...................................... 245

7.7.2 ZOTEK N Polyamide (PA 6) Foams .................................. 2467.7.2.1 Thermal Properties ................................................. 2477.7.2.2 High Temperature Modulus and Moisture ......... 2497.7.2.3 Other Performance Attributes .............................. 2517.7.2.4 Markets and Applications ...................................... 251

7.8 Conclusions ................................................................................... 251References ............................................................................................ 252

7.1 Introduction

Zotefoams plc is the leading manufacturer of low-density, closed-cell, cross-linked block foams produced using a unique, proprietary high-pressure gas technology which yields signifi cant product advantages over competitive technologies.

In recent years the company has been involved in the development of a new generation of low-density foams based on materials where foaming has traditionally been an insurmountable technical challenge or limitations have existed as to the reduction in density possible. These novel foam products are based on fl uoropolymers and polyamides. The characteristics of the foams, not surprisingly, refl ect the general properties of these classes of polymer.

This chapter will review the nitrogen autoclave technology, historically employed in the manufacture of cross-linked polyolefi n foams, and con-sider the other major technologies used to manufacture such materials. Finally a review of the key performance attributes of the new, unique, cellular materials described above is provided.

7.2 Brief History

The process technology referred to variously as the “nitrogen autoclave,” “autoclave batch,” “BXL” (a previous company name), or simply the “autoclave” process is a unique process technology within the foam industry. The process is proprietary to Zotefoams plc (headquartered in Croydon, UK) and is one that has seen much development over the past

220 Polymeric Foams

61259_C007.indd 22061259_C007.indd 220 10/25/2008 12:39:00 PM10/25/2008 12:39:00 PM

100 years or so. These developments in both process technology and mate-rials continue apace and the process now offers an extremely wide range of foam materials for an equally wide end-user base.

Before describing the technology and the more recent material develop-ments, it is an interesting diversion to look briefl y at the early process his-tory, which is a story of individual technical adventure and entrepreneurial development dating back to the late nineteenth/early twentieth century. The origins of the process begin with the Austrian, Mr. Robert Pfl eummer, and his three sons, Hans, Fritz, and Herman. The Pfl eummer family often cycled and all three regularly suffered punctures. To alleviate this problem and following a particularly embarrassing failure in a cycle race, the father challenged his three sons to develop a puncture-proof tire. The father also recognized at the time that such a product could have applications in the developing automotive industry.

The sons were scientifi cally minded and looked at a number of options before developing the process with rubber. Figure 7.1 shows a laboratory set-up of plant and equipment in London, circa 1908. Figure 7.2 shows Fritz Pfl eummer with an expanded rubber cylinder, circa 1912. In the background are the gas cylinders, compressors, gauges, and so on neces-sary for the process. Figure 7.3 shows the rather disappointing result of a test run by one of the other sons, Hans Pfl eummer, using his Ford Buick, dated around 1912. Having had the tires of his car fi lled with expanded rubber, the foamed elastomer is clearly seen to have burst through the outer sheath of the tire during the test.

Although ultimately not successful in this particular application, the process technology was suffi ciently interesting for a number of individu-als to attempt to commercialize the products. It took a further 20 years or so however before a successful business was formed (with many failing in between) and many decades later before the technology was developed to large-scale production. In the early 1930s the process moved from the original base in Crystal Palace to the current location in Croydon, England where it has remained to this date. The company has passed through several different owners since these early stages. For those interested in industrial history, the development of the business from the earliest days through to 1957 was suitably documented in a book1 written by the then Technical Director, Mr. A. Cooper, which was published on the twenty-fi rst anniversary of the company.

7.3 Nitrogen Autoclave Technology

7.3.1 Raw Materials

One of the great fl exibilities of the autoclave technology is that it has proven applicable to expanding a great many synthetic materials. In the

New Material Developments from the Nitrogen Autoclave Process 221

61259_C007.indd 22161259_C007.indd 221 10/25/2008 12:39:00 PM10/25/2008 12:39:00 PM

simplest representation a sheet of the polymer is exposed to a gas envi-ronment (normally nitrogen) at a certain pressure and temperature for a time long enough to ensure complete saturation with gas. Ignoring for a moment the additional thermal and rheological characteristics of the material necessary for expansion to be successful, it can be seen that this fundamental process of gas dissolution can be easily applied to a wide range of polymeric materials.

Although the above indicates that the process could apply to any mate-rial, there are of course some limitations. Primary restrictions on material choice2 are that fi rst, the material must have a reasonable level of gas solubility and diffusivity at practical pressures and temperatures; and second, the material must have some degree of melt strength which may be further enhanced or controlled via cross-linking or other means.

FIGURE 7.1 Laboratory set-up of plant and equipment in London, ca. 1908.

222 Polymeric Foams

61259_C007.indd 22261259_C007.indd 222 10/25/2008 12:39:00 PM10/25/2008 12:39:00 PM

7.3.2 Process Overview

The process has three fundamental stages, although there are some varia-tions to this dependent on product and material type. For the most part the descriptions which follow refer to the basic production of cross-linked low-density polyethylene (LDPE) and ethylene-vinyl acetate (EVA) foams.

The process begins with the extrusion and cross-linking of a continu-ous sheet of material from which a “precursor” solid sheet is cut. This

FIGURE 7.2 Fritz Pfl eummer, with an expanded rubber cylinder, ca. 1912.

FIGURE 7.3 Result of test by Hans Pfl eummer, in his Ford Buick, ca. 1912.


61259_C007.indd 22361259_C007.indd 223 10/25/2008 12:39:00 PM10/25/2008 12:39:00 PM

precursor sheet is then transferred to a second stage where it is placed in a large, high-pressure autoclave and exposed to nitrogen gas at high tem-perature and pressure. The gas is absorbed into the polymer structure under these conditions until saturation is achieved at which point a ther-modynamic instability is introduced, via a rapid pressure reduction, to nucleate the cell structure. For control reasons, the sheet is not fully expanded at this stage and instead the product is cooled and removed from the autoclave in nucleated form. Finally the nucleated sheet from the previous stage is transferred to a further, lower-pressure autoclave for full expansion. In this fi nal stage the material is subjected to a mild gas pres-sure and reheated until the entire sheet is at a uniform temperature. Once at thermal equilibrium, the pressure is reduced and the polymer, being soft and extensible at this higher temperature, is fully expanded.

7.3.3 First Stage—Sheet Extrusion

The initial sheet extrusion is perhaps the most critical stage of the process since it is in this stage where the batch-to-batch variability of the raw mate-rials interacts with the day-to-day process and hardware variables along with having greater operator input than in the later process stages. This process step is shown schematically in Figure 7.4.

Any additives are mixed with the base polymer either via preblending or by gravimetric feed into the barrel of the extruder. In simplest form the formulation comprises a base polymer (e.g. LDPE) together with a small fraction of an organic peroxide to effect cross-linking. Obviously other additives such as pigments, fl ame retardants, and so on, can also be added at this point as necessary, but it is worth noting that the process is typi-cally operated without the use of nucleating agents.

Typically the melting/mixing takes place in single screw extruders, although twin screw extruders are also employed. The material is mixed and pumped through a heavy duty sheet die to produce a thick sheet. The target thickness of the sheet can range between 5 and 25 mm, with typical thicknesses being 10–12 mm. The sheet is extruded onto a moving con-veyor and is then passed through a series of oven units.

It is in the oven units where the cross-linking reaction occurs, which takes around 30–40 minutes at the temperatures employed. Once fully

FIGURE 7.4 Schematic of the sheet extrusion process step (stage 1).

224 Polymeric Foams

61259_C007.indd 22461259_C007.indd 224 10/25/2008 12:39:00 PM10/25/2008 12:39:00 PM

cross-linked and after subsequent cooling, the sheet is then cut to specifi c dimensions which are in direct proportion to the required dimensions of the fi nal foamed sheet. For each formulation, there is then an optimum degree of cross-linking to give suffi cient melt strength, while also limiting over-expansion of the sheet. The infl uence of the melt rheology throughout the process is paramount and is tightly controlled.

It is important to note that the result of this stage is no more than a sheet of cross-linked polymer, of high purity and homogeneity. Typically the poly-olefi n polymers used are of high molecular weight and low melt index, thereby maximizing the fi nal foam properties; however, the use of such materials also requires that quite low levels of cross-linking agent are employed, relative to alternative technologies.

Although perhaps more directly “critical to quality,” the extrusion stage utilizes, for the most part, standard hardware and the extrusion of thick sheets is by no means a new technology. That said, there are signifi cant diffi culties which need to be overcome to extrude a very thick solid sheet of the necessary internal quality. The reader should consider that at some later point the entire volume of the sheet will be expanded by up to 65 times; any minor fl aws or inhomogeneities introduced via the extrusion process will also be expanded by that same factor!

7.3.4 Second Stage—High Pressure Autoclave

It is in this second stage that the process technology becomes more obvi-ously unique. This process step is shown schematically in Figure 7.5. As noted above, the extruded sheets are completely free of blowing agent and nucleating agents. To introduce the blowing agent, the sheet is simply exposed to a gas at extremely high pressure and temperature.

The sheet stock from the extrusion stage is loaded into the autoclave using specially designed carriage equipment. The entire carriage is then transferred into the autoclave where the physical blowing agent is intro-duced. The physical blowing agent of choice is nitrogen. As nitrogen is supercritical at temperatures above �147°C and pressures above 34 bar, then it is quite correct to describe the nitrogen autoclave technology as a supercritical fl uid (SCF) technology.

Once in the vessel, the gas and polymer are heated and pressurized. Pressures and temperatures up to 670 bar and 250°C respectively are employed. At 250°C the temperature is well above the melting point of a

FIGURE 7.5 Schematic of the gas absorption process step (stage 2).


61259_C007.indd 22561259_C007.indd 225 10/25/2008 12:39:01 PM10/25/2008 12:39:01 PM

typical LDPE polymer. While at such pressures and temperatures, the cross-linking previously undertaken enhances the polymer sheet dimensional stability thereby preventing fl ow. Additionally the quite low solubility of nitrogen in the polymer is overcome such that suffi cient gas dissolves to allow signifi cant expansion of the sheet in the third stage. At the conditions mentioned above, cross-linked LDPE foam of 15 kg/m3 (i.e. greater than 60 times volume expansion) can be produced commercially.

It is well known that the solubility of gas in the polymer will be affected by both pressure and temperature.3,4 Varying both of these parameters in the process allows control over the amount of gas to be absorbed by the sheet of polymer. Operationally this means a wide range of fi nal expansion ratios are possible from the same starting formulation, as gas pressure and temperature are the key controls of product density in the process.

Pressure and temperature are the main drivers of gas dissolution but the third variable of importance is the sheet thickness. It is well known that the mass of gas absorbed within a given time, t, into a plane sheet is related to the square of the thickness of that sheet.5 Therefore in the nitro-gen autoclave process increasing the thickness of the cross-linked sheet has a t2 effect on the process cycle time in this second stage. To emphasize the importance of this point one needs to consider fi rst the costs associ-ated with operating at the very high pressures and temperatures involved and second that typical process cycle times for products of, for example, 10 mm (solid) thickness are of the order of 6–8 hours.

Once saturation has been assured, then the next step is to nucleate the cell structure. This is achieved via a rapid depressurization, whereby the gas pressure may be reduced to a small fraction of the saturation pressure in a matter of a few seconds. The introduction of this thermodynamic instability renders the sheet supersaturated with gas, causing the nitrogen gas to come out of solution to form cell nuclei. This process stage is again tightly controlled and the expansion limited, to allow nucleation but restrict bubble growth.

This step not only nucleates the structure of the fi nal foam but at the same time control of this depressurization, more specifi cally the rate of depressurization, leads to the ability to control fi nal foam cell structures. More rapid rates of pressure reduction result in a relatively high nucle-ation density and therefore fi ne cell structures in the foam, whereas slower rates of pressure drop tend to reduce nucleation density and therefore render more coarse cellular structures in the fi nal foam. The effect of employing thermodynamic instability in the nucleation of foam structures has been studied more recently in relation to the development of micro-cellular foam processes.6,7

After the rapid depressurization and once pressure has again stabilized, the sheets are then cooled and removed from the autoclave. Although technically feasible to expand fully in a single stage, as noted earlier, other considerations dictate that full expansion is undertaken in a separate step.

226 Polymeric Foams

61259_C007.indd 22661259_C007.indd 226 10/25/2008 12:39:01 PM10/25/2008 12:39:01 PM

The density of the polymer sheet at this point following nucleation may be reduced by around 30% or more—in the case of LDPE this equates to a density of around 600–700 kg/m3. The structure of the material is typically microcellular (cell diameters � 100 μm) and structurally remarkably con-sistent. Figure 7.6 shows an SEM of the structure of this intermediate nucle-ated material in LDPE, with cells of the order of 30–50 μm in diameter.

The description above demonstrates some of the important operational controls on the nitrogen autoclave technology. Both density and cell size are largely determined by the process conditions and much less so by the material formulation—a single formulation therefore can give rise to a very wide range of products (sheet size, foam density, and cell dimen-sions). It is these points combined that have led some academics8 to study the materials to gain a better understanding on the structure/property effects in cellular polymers, without the complications introduced as a result of formulation changes.

7.3.5 Third Stage—Low Pressure Autoclave

It is here that the fi nal foam expansion is achieved. The material from the previous stage is loaded onto an open tray carriage system and transferred into a further autoclave where the gas-laden nucleated sheet is reheated. Typically the temperature used for LDPE is around 150°C and the pres-sure is of the order of 15–20 bar. The gas environment within the autoclave may be either air or nitrogen.

Pressure is initially applied while the sheet is heating. Once the sheet has been allowed to equilibrate at the required temperature and pressure

FIGURE 7.6 SEM of the structure of the intermediate, nucleated LDPE sheet.


61259_C007.indd 22761259_C007.indd 227 10/25/2008 12:39:01 PM10/25/2008 12:39:01 PM

for the appropriate time, the pressure is reduced. As the temperature is some 40°C above the melting point of a typical LDPE polymer during this pressure reduction, the polymer sheet expands as the gas pressure is reduced. This fi nal stage of the process is shown schematically in Figure 7.7.

An obvious simplifi cation of this stage could be envisaged where the nucle-ated sheet is expanded in a simple air-circulating oven. While this is possi-ble, such an approach is diffi cult to control due to the uneven heating effects that occur in thick, low thermal conductivity sheets of gas-laden polymer.

The fi nal foam sheets, sometimes referred to as blocks or buns, are typically 2000 mm long, 1000 mm wide, and 30–50 mm in thickness; how-ever, the ability to quickly and easily manipulate sheet sizes indepen-dently of density/structure was indicated earlier and offers great fl exibility to the process.

Finally, as the expansion process is unrestrained, the properties of the foam material produced are largely isotropic. This is a signifi cant benefi t to end users in certain markets where the consistency of density and prop-erties throughout the foam sheet is a key attribute, such as in cushion packaging or thermoforming.

7.3.6 Quality

One key operational advantage of the process is the ability, at each dis-crete stage of the foaming process, to carry out quality assurance checks prior to further processing. From the extrusion stage through cross- linking, gas absorption, and then foam expansion, each intermediate product may be quality assured prior to the next process step.

Sheet weight, sheet dimensions, sheet profi le (thickness), and degree of cross-linking are all monitored on a continual and routine basis. The latter is monitored to ensure that the sheet expansion and properties will be as expected, assuming that all key parameters in the later stages are fi xed (most notably gas pressure and temperature). By way of example, Table 7.1 shows the effect that varying cross-link level can have on the density and performance of standard ethylene-vinyl acetate (EVA) foam.9 As would be predicted, the effect of increasing the cross-link level is to increase density by restricting expansion/bubble growth. In performance terms the effect of increasing cross-linking is to lead to less ductile behavior in tension

FIGURE 7.7 Schematic of the fi nal expansion process step (stage 3).

228 Polymeric Foams

61259_C007.indd 22861259_C007.indd 228 10/25/2008 12:39:01 PM10/25/2008 12:39:01 PM

(elongation at break falls as tensile strength increases), while the material becomes more elastic in compression.

7.3.7 Alternative Cross-Linked Polyolefin Foam Technologies

The nitrogen autoclave process as described above is one of only a small number of technologies that have been developed for foaming cross-linked polyolefi ns. With the exception of autoclave technology, the other major technologies in use today were all developed in Japan in the 1960s. Since then these technologies have been further enhanced and success-fully licensed around the globe.

In practical terms the processes can be categorized fi rstly according to the basic product form, either block or roll, and then further by the details of the processing methods employed. Figure 7.8 shows one such categorization.

7.4 Block Processes

The only major alternative to the nitrogen autoclave process for the foaming of discrete blocks (often also referred to as “sheets” or “buns”) of cross-linked polyolefi n foam is the press forming technique. This tech-nology was developed out of processes traditionally employed to foam natural and synthetic elastomers but has been adapted for the needs of polyolefi n materials over the past four decades such that commercial products now cover a broad density range, from 24 kg/m3 up to around 300 kg/m3. At the higher density end of this spectrum, the materials are expanded in a single-step however to achieve the very high expansions (between 15 and 40 times volume expansion) then a dual-step process is employed, achieving greater utilization of the available gas.

TABLE 7.1

Effect of Varying Cross-Link Level on the Density and Basic Properties of Standard Ethylene-Vinyl Acetate (EVA) Foam

Peroxide Level

(100% = Standard Foam Formulation Level)

Property 80 100 140

Nominal density (kg/m3) 47.8 48.6 52.0

Compression stress (e � 50%) (kPa) 102 115 129

Compression set (20%/48 h/0.5 h rec) 7.7 6.7 5.1

Tensile strength (kPa) 616 896 926

Elongation at break (%) 304 197 171


61259_C007.indd 22961259_C007.indd 229 10/25/2008 12:39:01 PM10/25/2008 12:39:01 PM

The basic process technology involves the mixing and homogenization of the polymer, chemical blowing agent, normally azodicarbonamide (ADCN) of specifi c particle size and a peroxide cross-linking agent in a batch mixer. The output from the mixer is then loaded into a mold and press formed. During this stage the cross-linking agent is reacted with the polymer to enhance the polymer melt strength and the blowing agent decomposition occurs. In the single-step process, rapid opening of the press platens leads to foam expansion. In the dual-step process, of which a number of variations exist, the sheets from the fi rst stage undergo a con-trolled and limited expansion before being transferred to a mold where they are steam heated and expanded fully.

One of the drawbacks of this process is the very high pressure that is generated in the polymer which leads to high-pressure hydraulic press equipment being required. A more recent novel variation of this is described in US patent 5955015,10,11 which uses annular cylindrical molds to overcome the very high clamping pressures that would otherwise be needed in man-ufacturing low-density block and thereby reducing capital cost also.

Relative to the physically blown products from the nitrogen autoclave process, the chemically blown block products typically suffer from a larger variation in density within a single sheet due to the diffi culties in control-ling the exothermic reaction of the blowing agent. At the center of the expanding chemically blown block the blowing agent typically sees a higher temperature for a longer time than the outer regions and as the gas yield is related to the time and temperature conditions during decomposition this

FIGURE 7.8 Categorization of alternative technologies for the production of low-density

cross-linked polyolefi n foams based on product form and processing methods.

Low density,crosslinked

polyolefin foamtechnologies

Discrete blockprocess

technologies(sometimes referred to as ‘sheet’

‘bun’ or ‘batch’ processes)

Semi-continuoussheet processtechnologies

(sometimes referred to as ‘roll’foam’ processe)

Physicallyexpanded

Nitrogen autoclaveprocess

Press formingprocess

Sekisuiprocess

Torayprocess

Horizontal(salt bath)

Vertical(oven)

Furukawaprocess

Hitachiprocess

(Physical blowing agent)

Chemicallyexpanded

Physicallycrosslinked

Chemicallycrosslinked

(Chemical blowing agent)

230 Polymeric Foams

61259_C007.indd 23061259_C007.indd 230 10/25/2008 12:39:01 PM10/25/2008 12:39:01 PM

leads to a greater variation in density through the thickness of the block than is seen in the physically blown autoclave products. This difference is measurable by cutting sections through the thickness of the product and calculating density of each section to develop a density profi le for the sheet.

Table 7.2 offers a comparison of the properties of nominal 33 kg/m3 foams physically blown using the nitrogen autoclave process and those of commercially available chemically blown block products. The notable differences are that the products from the nitrogen autoclave process typically exhibit higher hardness, greater stiffness in compression, and a more discernible yield point in the compression curve. These characteris-tics are partly attributable to the larger cell size generated in the nitrogen autoclave products but are also a result of the more regular structure that is achieved in the foams as a result of the combination of the use of physical blowing agent and the homogeneous nucleation step.

The improved structure of the products also tends to lead to improved compression set properties as gas escape and diffusion through the cell

TABLE 7.2

Properties of Physically Blown Foam from the Nitrogen Autoclave Process Compared to Three Commercially Available Chemically Blown Block Products (All Nominally 30 kg/m3 Cross-Linked Polyethylene)

Property Method Unit

Nitrogen

Autoclave

Product

Chemically

Blown

Product A

Chemically

Blown

Product B

Chemically

Blown

Product C

Density ISO 845 kg/m3 31 32 27 39

Cell size Internal mm 0.35 0.19 0.25 0.16

Shore hardness ISO 868 ‘00’ 60 57 54 57

Compressive

stress

ISO 7214

@ 10% strain kPa 54 43 37 46

@ 25% strain kPa 72 62 55 66

@ 40% strain kPa 105 95 87 99

@ 50% strain kPa 140 129 119 135

Compression

set

ISO 7214

22 h/50%/0.5 h

recovery

% 22.4 25.8 24.1 23.9

22 h/50%/24 h

recovery

% 13 16.2 15.6 13.4

Tensile

strength

ISO 7214 kPa 528 322 310 403

Elongation

at break

ISO 7214 % 121 140 88 178

Tear strength ISO 8067 N/m 590 728 857 518


61259_C007.indd 23161259_C007.indd 231 10/25/2008 12:39:01 PM10/25/2008 12:39:01 PM

walls is more restricted. Comparing the results in Table 7.2, product C can be seen to be similar in compression set, however the density of the samples should also be noted with product C being some 25% higher in density, which naturally improves the compression set behavior.

Finally, the nitrogen autoclave products tend to give higher strength values in tension while the elongation at break and tear strength values are generally lower than those of chemically blown products. In these latter properties the raw materials used, the degree of cross-linking and the cell size of the product all have a more signifi cant bearing than they would be expected to have on the compressive properties and so direct comparisons become more challenging.

7.5 Semi-Continuous (Roll) Processes

These processes are capable of producing a thin sheet on a reel which is perhaps a few hundred meters in length and a meter or more in width. The product thicknesses are from less than 1 mm to a maximum of 16 mm, which is a restriction that applies as a result of both the processing require-ments as well as the practical necessity to store and ship the material in practical lengths on a reel.

Roll foam processes are defi ned by the means of cross-linking, either chemical or physical. The chemically cross-linked processes were developed independently by the Furukawa Electric Co. and the Hitachi Co. and have been licensed successfully. Both processes employ horizontal ovens through which the material is cross-linked and expanded. In the Hitachi process these steps are slightly separated by means of a sheet preheating stage where cross-linking is initiated prior to expansion. In the Furukawa process the cross-linking and expansion processes overlap and progress simultane-ously in the foaming oven. This difference means the Hitachi process is per-haps a little more controllable and fl exible in material formulation choices. Other differences exist in the details of the design of the expansion ovens and the material handling equipment during the expansion step.

The second group of processes, developed separately by Sekisui Chemical Co. and Toray Plastics rely on the physical cross-linking of the extruded sheet. The sheet containing polymer and blowing agent (typically azodi-carbonamide) is extruded onto a reel. This reel is then physically cross-linked via electron beam irradiation at some later point. The thin section of the sheet means that there are no penetration depth issues leading to variable cross-linking and nor is a very high power beam necessary. Once cross-linked the sheet may then be expanded.

In the Sekisui process expansion takes place in a vertical oven. This expansion technology has the advantage of being more energy effi cient

232 Polymeric Foams

61259_C007.indd 23261259_C007.indd 232 10/25/2008 12:39:01 PM10/25/2008 12:39:01 PM

but means that the material is naturally drawn in the length direction as the sheet stretches under its own weight which leads to directionality of the properties of the fi nished sheet. This is not necessarily a negative point, however, as the elongated cell structure in these materials can also give a more drapeable, compliant, and ‘soft feeling’ material. Sekisui have continued to work with the process to develop ultra-thin foams down to 200 μm in thickness.

In the Toray process, the sheet travels horizontally during expansion and is fl oated on a molten salt bath, which in tandem with infrared heat-ing leads to expansion of the sheet. As a result of the horizontal movement during foaming, the expansion in all three dimensions is much more controlled and there is less tension in the sheet to lead to directionality.

In both of the physically cross-linked roll foam processes described, the sheet surface quality is a distinguishing feature. The thin foam sheets have very high quality surfaces relative to the chemically cross-linked materials; however, they are generally of restricted thickness with the thickest products being of the order of 8 mm.

Some of the advantages and disadvantages of the processes described above are found in Table 7.3. For further details of cross-linked polyolefi n foam processes the interested reader is initially directed to References 12–15, which contain more comprehensive descriptions, information, and comparisons as well as further references.

7.6 Nitrogen Autoclave Cross-Linked Polyolefin

Foam Products

7.6.1 Product Range

The current commercial product range (under the AZOTE brand) covers effectively the entire spectrum of polyolefi n materials. At one extreme are foams produced from co-polymers of ethylene such as ethylene-methyl acrylate (EMA) and ethylene-vinyl acetate (EVA) materials. At the other extreme are sheet foams produced from high-density polyethylene (HDPE) and polypropylene (PP). In the middle of this range there exist low- density polyethylene (LDPE) foams as well as metallocene polyethylene (mPE) foams and medium density polyethylene (MDPE) foams.

In some cases the basic process described earlier must be modifi ed to allow processing; for example in the case of the HDPE foams, the materials are cross-linked via an irradiation process since the extrusion temperatures for HDPE preclude the use of organic peroxides.

The polyolefi n products produced via the nitrogen autoclave technology range from soft, ultra-fl exible materials through to semi-rigid and rigid,


61259_C007.indd 23361259_C007.indd 233 10/25/2008 12:39:02 PM10/25/2008 12:39:02 PM

TAB

LE 7

.3

Co

mp

ari

son

of

the

Majo

r T

ech

no

log

ies

Use

d t

o P

rod

uce

Lo

w-D

ensi

ty C

ross

-Lin

ked

Po

lyo

lefi

n F

oa

ms

Cro

ss L

ink

ed

Fo

am

Pro

cess

es

Blo

win

g

Ag

en

t

Cro

ss-L

ink

ing

Ag

en

t

Exp

an

sio

n

Hard

ware

Pro

du

ct

Th

ick

ness

(mm

)A

dv

an

tag

es

Dis

ad

van

tag

es

Blo

ck F

oam

Pro

cess

es

Nit

rog

en

Au

tocl

av

e

Tec

hn

olo

gy

Ph

ysi

cal

Ch

emic

al

or

ph

ysi

cal

Au

tocl

av

e�

60

No

ch

emic

al

blo

win

g a

gen

t (“

clea

n”

pro

du

ct),

ap

pli

cab

le t

o w

ide

ran

ge

of

mate

rials

in

clu

din

g e

ng

inee

rin

g

pla

stic

s, s

hee

t si

ze

fl ex

ibil

ity,

den

sity

an

d c

ell

size

con

tro

l v

ia

pro

cess

, v

ery

hig

h v

olu

me

exp

an

sio

ns

po

ssib

le, h

igh

qu

ali

ty

stru

ctu

re, fr

ee e

xp

an

sio

n (

iso

tro

pic

shee

t), k

ey p

roce

ss s

tag

es a

re

dis

cret

e fr

om

on

e an

oth

er.

Hig

h c

ap

ital

cost

, re

gu

lati

on

of

hig

h

pre

ssu

re e

qu

ipm

ent,

les

s ec

on

om

ic

for

hig

h d

ensi

ty f

oam

, p

rod

uct

thic

kn

ess

cap

ab

ilit

y i

s li

mit

ed.

Pre

ss M

old

ing

Tec

hn

olo

gy

Ch

emic

al

Ch

emic

al

Hy

dra

uli

c

pre

ss�

110

Lo

w c

ap

ital

cost

, th

ick

pro

du

ct

(see

n a

s a y

ield

ad

van

tag

e),

fi n

e ce

ll.

Pro

du

ct d

ensi

ty v

ari

ati

on

, o

do

r,

pre

ssu

res

ver

y h

igh

fo

r lo

w d

ensi

ty

pro

du

ct (

pre

ss s

pec

ifi c

ati

on

), m

old

an

d d

ensi

ty v

ari

ati

on

im

part

pro

per

ty v

ari

ati

on

s th

rou

gh

ou

t

shee

t.

234 Polymeric Foams

61259_C007.indd 23461259_C007.indd 234 10/25/2008 12:39:02 PM10/25/2008 12:39:02 PM

Rol

l Foa

m P

roce

sses

Sek

isu

i

Tec

hn

olo

gy

Ch

emic

al

Ph

ysi

cal

Ver

tica

l

ov

en�

1–8

Ver

y t

hin

sh

eet

po

ssib

le, en

erg

y

effi

cien

cy (

fro

m v

erti

cal

ov

en),

exce

llen

t su

rface

qu

ali

ty, v

erti

cal

exp

an

sio

n e

nh

an

ces

soft

to

uch

cap

ab

ilit

ies

of

the

mate

rial

(elo

ng

ate

d s

tru

ctu

re i

s d

evel

op

ed),

wid

er r

an

ge

of

mate

rials

po

ssib

le

(in

c. P

P).

Hig

her

cap

ital

cost

an

d p

rod

uct

ion

cost

, th

ick

nes

s li

mit

ati

on

s to

pro

du

ct, an

iso

tro

pic

(m

ach

ine

an

d

cro

ss-m

ach

ine

pro

per

ties

dif

fer)

.

Fu

ruk

aw

a

Tec

hn

olo

gy

Ch

emic

al

Ch

emic

al

Ho

rizo

nta

l

ov

en�

5–16

Rel

ati

vel

y l

ow

cap

ital

cost

, sl

igh

tly

thic

ker

pro

du

ct p

oss

ible

,

pro

du

ctio

n c

ost

lo

wer

at

low

er

vo

lum

e.

Su

rface

qu

ali

ty c

an

be

po

or.

Po

or

cell

str

uct

ure

. C

ho

ice

of

mate

rials

is m

ore

dif

fi cu

lt. T

emp

eratu

re

con

tro

l is

cru

cial

to p

rod

uce

bes

t

qu

ali

ty.

Hit

ach

i

Tec

hn

olo

gy

Ch

emic

al

Ch

emic

al

Ho

rizo

nta

l

ov

en�

5–16

Rel

ati

vel

y l

ow

cap

ital

cost

, sl

igh

tly

thic

ker

pro

du

ct p

oss

ible

,

pro

du

ctio

n c

ost

lo

wer

at

low

er

vo

lum

e.

Su

rface

qu

ali

ty c

an

be

po

or.

Po

or

cell

str

uct

ure

. C

ho

ice

of

mate

rials

is m

ore

dif

fi cu

lt. T

emp

eratu

re

con

tro

l is

cru

cial

to p

rod

uce

bes

t

qu

ali

ty.

To

ray

Tec

hn

olo

gy

Ch

emic

al

Ph

ysi

cal

Mo

lten

salt

bath

� 1

–7

Th

in s

hee

t, e

xce

llen

t su

rface

qu

ali

ty,

wid

er r

an

ge

of

mate

rials

po

ssib

le

(in

c. P

P).

Th

ick

nes

s li

mit

ati

on

s o

f p

rod

uct

,

salt

bath

safe

ty c

on

cern

s.


61259_C007.indd 23561259_C007.indd 235 10/25/2008 12:39:02 PM10/25/2008 12:39:02 PM

energy absorbent foams. In all cases the range of density produced is between 15 kg/m3 and 120 kg/m3. Of note are the following product groups.

7.6.2 Cross-Linked High-Density Polyethylene Foams

Traditional high-density polyethylene (HDPE) resins are a group of polyethylene resins which can be defi ned as being in the density range 950–970 kg/m3 produced using low-pressure polymerization processes. The resins typically have higher levels of crystallinity leading to improved mechanical properties. Also, the materials will typically have higher melt viscosities and extrusion processing temperatures in excess of 200°C are necessary. HDPE bead foams are well known but in cross-linked sheet form the material is less familiar due to the problems associated with the control of cross-linking and availability of suitable blowing agents to produce low-density foams.

Table 7.4 shows a set of typical mechanical properties for the HDPE foams produced via the nitrogen autoclave technology. These low-density HDPE foams fi nd applications in a number of areas where the structural rigidity and high-energy absorption properties are of benefi t.

7.6.3 Cross-Linked Metallocene Polyethylene Foams

In the early to mid-1990s a series of polyolefi n materials were commercial-ized based upon metallocene catalyst technology. These methods of polymerization had been in research for many years. Foam converters, like many other polymer converters, were quick to evaluate the benefi ts of these materials. The products produced via the nitrogen autoclave pro-cess show a series of characteristics such as enhanced toughness, strength and durability, improved thermoformability, and reduced cell sizes. The impact of reduced cell sizes is mainly an aesthetic benefi t, leading to a soft feel to the material.

TABLE 7.4

Typical Mechanical Properties for the Cross-Linked Plastazote HDPE Foams

Property Method

Plastazote

HD30

Plastazote

HD60

Plastazote

HD80

Plastazote

HD115

Nominal density (kg/m3) ISO 845 30 60 80 115

Tensile strength (MPa) ISO 7214 1.01 1.72 2.09 2.39

Compression stress

(� � 25%) (kPa)

ISO 7214 158 326 523 791

Compression stress

(� � 50%) (kPa)

ISO 7214 222 398 593 897

Flexural modulus (MPa) BS 4370 3.6 11.0 15.0 23.0

Tear strength (N/m) ISO 8067 1415 3525 4970 8310

236 Polymeric Foams

61259_C007.indd 23661259_C007.indd 236 10/25/2008 12:39:02 PM10/25/2008 12:39:02 PM

These materials have found application in thermoformed parts where high draw ratios and excellent defi nition are required and also in returnable packaging (dunnage) applications where enhanced toughness and dura-bility are particularly valued.

7.6.4 Markets and Applications

Cross-linked polyolefi n foams from nitrogen autoclave technology fi nd a variety of uses, covering all of the major market segments. Most notably, the highly consistent nature of the foams produced, in property, density and purity terms, means that they are the material of choice in high speci-fi cation packaging. Briefl y examining some of the major markets:

7.6.4.1 Packaging

Typical examples of use would be for case inserts, display packaging, and highly specifi ed cushion packaging. In a number of packaging applications, the primary reason for using the foam may be simply cost or aesthetics; however, in the case of cushion packaging the material must perform a much more serious purpose.

Cushion curves (see typical example in Figure 7.9) are the basis of the traditional method of packaging design to transport delicate, high value, or fragile goods. The curves describe the ability of the foam to manage the deceleration of an article based upon the loading, the height from which the article is dropped (or in the case of design the height from which the arti-cle is likely to be dropped) and the thickness of the foam used to protect it.

Plastazote LD45 cushion curvesFirst impact / Drop height = 1300 mm

0

20

40

60

80

100

120

140

160

180

200

4 6 8 10 12 14 16 18 20Static stress (kPa)

Pea

k de

cele

ratio

n (G

)

Foam thickness = 25 mm




20

FIGURE 7.9 Typical cushion curves used in cushion packaging design (here showing

curves for various thicknesses of a 45 kg/m3 cross-linked PE foam).


61259_C007.indd 23761259_C007.indd 237 10/25/2008 12:39:02 PM10/25/2008 12:39:02 PM

In closed-cell foams, the preferred structural response of the foam to such an impact is plastic deformation of the cell walls. To achieve this, semi-rigid foams tend to perform optimally. In more rigid foams, the cushioning perfor-mance is reduced. Although often the shock absorption behavior of such materials is good, in a shock impact situation the deceleration is not man-aged and the packaged article can suffer damage from a deceleration which is too rapid. Additional to this, more rigid foams also tend to suffer from higher compression set, which worsens performance in second or multiple drop situations. More fl exible foams tend to absorb energy through pneu-matic compression of the gas in the cells and often a large proportion of this absorbed energy is returned to the packaged article via rebound—in this case the deceleration/acceleration combination can also yield poor performance.

Finally it is worth noting that a large volume of non-cross-linked foam, for instance EPS and LDPE, is also used in some high-specifi cation cushion packaging applications. However, while this may be so, the more durable cross-linked LDPE foams are generally found to outperform non-cross-linked foams in the critical area of multi-impact performance.

7.6.4.2 Sports and Leisure

This is a large market sector for cross-linked polyolefi n foams and the entire range of these types of material are used, from soft through to rigid foams. Typical applications include body protection (for example in hockey), swimming aids and fl oats, camping mats, sports mats and the like. Generally the materials are valued for their low weight, durability, buoy-ancy, and impact absorption properties.

7.6.4.3 Medical/Health Care

One of the major attributes of the products of the nitrogen autoclave tech-nology is the ability to maintain a relatively pure foam product. As noted earlier, in the basic foam materials, following the cross-linking reaction, there is little else in the foam other than LDPE and nitrogen/air. This high purity level offers the materials a certain niche in the medical and health care sectors where the foams are used for cervical collars, orthotic support, diabetic shoe insoles, and so on. In many cases for splinting, the foam may be thermoformed directly to the body to produce a perfect fi tting support.

7.6.4.4 Other

In addition to the key market sectors noted above, there are very few industries that do not utilize low-density cross-linked polyolefi n foams for some reason. Other examples include:

Marine industry for fenders, life jackets, and fl oating hoses•

Building and construction industry for backer rods, expansion • joints, eaves fi llers, and underfl oor thermal/acoustic insulation

238 Polymeric Foams

61259_C007.indd 23861259_C007.indd 238 10/25/2008 12:39:02 PM10/25/2008 12:39:02 PM

Highly fl ame retardant grades are used in the aerospace industry • for sealing, fl otation aids, and duct insulation

Automotive industry for water barriers, gaskets/seals, vibration • pads, and impact protection

Electronics and semiconductor industries, where electrically • conductive and static dissipative foams are used for pin inser-tion packs, faraday cage shielding, tote box liners, and work-station mats.

7.7 New Material Developments

From a material development perspective, efforts have diversifi ed over recent years,16 away from developments focused upon polyolefi n materi-als, to the study of the feasibility of foaming other polymers, typically those classed as “engineering polymers.” These material developments have again demonstrated the unique position and capability of the nitrogen autoclave technology within the industry.

The commercial products that have been produced as a result are being marketed under the ZOTEK® brand, giving a range of high-performance foams based upon fl uoropolymers, engineering plastics, and speciality elastomers.

ZOTEK F is a patented17 range of low-density, closed-cell foams based upon the polymer polyvinylidene fl uoride (PVDF). This is a remarkable material in that it offers some exceptional performance attributes along with ease of processability. Key attributes include temperature resistance, outstanding chemical and weathering resistance and exceptional fl amma-bility/smoke generation performance.

ZOTEK N is a patented18 range of low-density closed-cell foams based on the engineering thermoplastic, polyamide. Initial products are based upon polyamide 6, also known as nylon 6 or PA6. The key attributes of these materials are the resistance to hydrocarbon fuels and oils allied to the high temperature performance and rigidity of the resins.

In the following sections, these material types and some of the key attributes are examined in more detail.

7.7.1 ZOTEK F Polyvinylidene Fluoride Foams

The current commercial grades of ZOTEK F are shown in Table 7.5 with a brief description. These materials are all based on the polymer PVDF and specifi cally Kynar® PVDF grades from Arkema Inc., with whom Zotefoams plc have a worldwide agreement to develop the products both technically and commercially.


61259_C007.indd 23961259_C007.indd 239 10/25/2008 12:39:02 PM10/25/2008 12:39:02 PM

PVDF is in reality a large family of materials, similar to polyolefi ns, covering homopolymer grades and co-polymer grades which offer the customer a broad range of temperature stability and tailored properties. PVDF homopolymer is a rigid, crystalline polymer, melting at 171°C. Less relevant here but nonetheless interesting, the material has received some attention for the piezoelectric and pyroelectric behavior it exhibits; under certain circumstances the polymer can be made substantially more piezo-electric than crystalline quartz.

By introducing comonomers such as hexafl uoropropylene (HFP), tetra-fl uoroethylene (TFE), and chlorotrifl uoroethylene (CTFE), a wide range of copolymers and terpolymers of vinylidene fl uoride (VDF) have also been made possible, mostly yielding more elastomeric materials. The additional possibility of employing functional additives gives rise to an even wider range of material and performance options.

The key properties of the PVDF polymer of interest to foam markets are those typical of the fl uoropolymer family, namely excellent fl ame resis-tance, extremely low heat release and smoke generation, outstanding UV resistance, and broad resistance to chemical attack.

Finally in addition to the above are the attributes generated from foam-ing the material such as buoyancy, impact absorption, and low thermal conductivity, yielding products with markedly different behavior to those of foams currently available.

7.7.1.1 Flammability Characteristics

The ZOTEK F foams have recently been tested to some of the most strin-gent large-scale fl ammability specifi cation tests that industry applies, covering markets such as aerospace, construction, semiconductor clean-room, and at a laboratory level more generally applicable specifi cations such as UL94 V-0.

The example of the ANSI/UL 723 (ASTM E 84-01) test standard is a good example to demonstrate foam performance. This standard is often referred to as the ‘Steiner Tunnel’ test and is used to classify the surface burning characteristics of building materials. The test results are reported as Flame Spread Index (FSI) and Smoke Developed Index (SDI) (see Table 7.6).

TABLE 7.5

Current Commercial Grades of ZOTEK F PVDF Foam

Product Nominal Density Attributes

ZOTEK F 30 30 kg/m3 (1.9 lb/ft3) Low density, fl exible, closed-cell foam

ZOTEK F 38HT 38 kg/m3 (2.4 lb/ft3) Higher temperature, more rigid, closed-cell

foam

ZOTEK F 74HT 74 kg/m3 (4.6 lb/ft3) High density, higher temperature, more rigid

ZOTEK F 42HT LS 40 kg/m3 (2.5 lb/ft3) Higher temperature, more rigid,

closed-cell foam, low smoke

240 Polymeric Foams

61259_C007.indd 24061259_C007.indd 240 10/25/2008 12:39:02 PM10/25/2008 12:39:02 PM

Values of less than 25/50, respectively, indicate that the material under test has very limited combustibility and offers the highest level of fi re perfor-mance other than a non-combustible (non-organic) material.

In order to achieve the lowest smoke release requirements, the LS (low smoke) foams were developed, which incorporate a small amount of a proprietary smoke suppressant. This material has been tested to the requirements of FM 4910 “Test Standard for FM Approvals Cleanroom Materials Flammability Test Protocol”19 and is the fi rst polymeric foam to be specifi cation tested, listed by FM Global, against this standard.

In aerospace, regulatory changes to test protocols for the evaluation of materials for use as thermal/acoustic insulation in commercial aircraft have been introduced in recent years. These new regulations and the applicable test methods are described in the Code of Federal Regulations FAR §25.856.20 The ZOTEK F foams comply with the requirements of FAR 25.856(a), the so-called “radiant panel” test, and also demonstrate low heat release and smoke generation. In the aerospace industry, the combi-nation of these FST (fi re, smoke, and toxicity) properties along with low weight, fl exibility, closed-cell structure (low moisture absorption) and ease of fabrication have led to a great deal of interest and development activity in this sector. Finally, although not offi cially certifi ed, materials tested at laboratory level have indicated a UL94 V-0 rating at 13 mm, which is quite unusual for very low-density organic foams.

7.7.1.2 Environmental and Chemical Performance

The PVDF raw material has traditionally been used as an additive in paints and coatings for exterior use because of the exceptional UV/weathering resistance of the polymer as well as the broad range of chemical resistance it possesses. These important properties have been maintained in the foam and as an example of this the ZOTEK F 30 foam has been subjected21 to several thousand hours of exposure in three separate accelerated UV/weathering instruments. The conditions were set to simulate:

Indoor: temperature • � 55°C, 55% RH, wavelength 300–400 nm (glass fi lter) and very low power (36 W/m2).

TABLE 7.6

Results of Flame Spread Index (FSI) and Smoke Developed Index (SDI) for the ZOTEK F PVDF Foam Tested to ANSI/UL 723 (ASTM E-84) “Steiner Tunnel” Test

Foam Thickness FSI/SDI Class

3.2 mm (1/8��) 0/0 A

25.4 mm (1��) 5/0 A


61259_C007.indd 24161259_C007.indd 241 10/25/2008 12:39:03 PM10/25/2008 12:39:03 PM

Outdoor: temperature • � 60°C, moisture cycling, wavelength 300–400 nm and low power (63 W/m2).

Severe/dry: temperature • � 60°C, dry, wavelength 300–400 nm and high power (400 W/m2).

The foam was evaluated by a combination of mechanical (tensile), visual (microscopy) and color measurements (Yellowness Index). The results of the color measurements (see Figure 7.10) gave some mild bleaching of the foam surface within the fi rst 200–300 hours of exposure followed by stabilization. The scale of the Yellowness Index should be noted with the color change barely perceptible to the naked eye. Samples were also evaluated in tensile tests and these demonstrated no signifi cant change in tensile or elongation behavior. Finally, microscopy showed no evidence of crazing or cracking after the exposure time [6000 hours (250 days)], even after fl exing of the samples.

From a chemical resistance point of view the ZOTEK F foams were immersed in two separate “standard” fuels used in standard automotive test protocols and the results compared with those of cross-linked poly-ethylene (XL-PE) foam of equivalent density.6 Polyethylene foams are known to have poor resistance to diesel fuel and hydrocarbon oils/fuels so the results are perhaps not surprising for this material. The ZOTEK F 30 foam however was immersed for upward of 120 days and the equilibrium weight change after this period was of the order of 1–2% (see Figure 7.11).

0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 5500 6000 70006500

–2

–1

0

1

2

3

4

5

6

7

8

9

10 CI 3000

Xenotest 1200SEPAP 12-24

Yel

low

ness

Inde

x

Exposure time (hours)

FIGURE 7.10 Plot of Yellowness Index versus exposure time for the ZOTEK F PVDF foam

exposed using three different accelerated weathering test methods.

242 Polymeric Foams

61259_C007.indd 24261259_C007.indd 242 10/25/2008 12:39:03 PM10/25/2008 12:39:03 PM

Finally, under aerospace specifi cation RTCA DO-160D, Cat. F, Section II, the materials were exposed to around 16 different types of fl uids and fuels used in that industry. The specifi cation is primarily an exposure test for component or functional systems that are to be used in aircraft. The tests undertaken on the foam involved exposure of the surface of the foam to the fl uid (at temperatures ranging between 40°C and 150°C) for 24 hours. This third party testing confi rmed that the ZOTEK F 30 and F 38HT foams comply with the requirements of this specifi cation, again with little more than minor surface discoloration.

7.7.1.3 Basic Thermal Properties

In addition to the outstanding FST properties and as a result of the highly consistent, closed-cell foam structure the thermal conductivity of the foams is low, of the order of 0.030–0.040 W/m K making them ideally suited for insulating applications. The ZOTEK F HT foams are also char-acterized by a relatively high melting point and therefore the foams are very dimensionally stable up to temperatures of around 150°C. Given the exceptional FST properties, high operating temperature and the low thermal conductivity, the ZOTEK F foams are not surprisingly fi nding increased usage for the insulation of air conditioning ducts, pipework, and other process industry installations.

FIGURE 7.11 Results for the ZOTEK F PVDF foam showing equilibrium weight change

after immersion for >120 days in standard test fuels (comparison with XLPE foam of similar

density).

0 10 20 30 40 50 60 70 80 90 100 110 120 1300

2

4

6

8

10

12

14

16

18

20

22

24

26

28

30ZOTEK F30 / DIESEL FUEL 2D 30 kg/m3 XL-PE / DIESEL FUEL 2D

ZOTEK F30 / FUEL C 30 kg/m3 XL-PE / FUEL C

Wei

ght c

hang

e (%

)

Immersion time (days)


61259_C007.indd 24361259_C007.indd 243 10/25/2008 12:39:03 PM10/25/2008 12:39:03 PM

7.7.1.4 Other Performance Attributes

As demonstrated above, the ZOTEK F foams have outstanding FST, ther-mal, chemical and environmental resistance. Allied to this the foams have a closed-cell structure (Figure 7.12 shows a micrograph of the expanded foam structure) and are therefore buoyant, opening up potential uses as “sink-proof” fl otation devices for use in harsh environments. The closed-cell nature and ease of moldability also offers end-users a single material solution for insulation and moisture barrier that can be easily shaped and manipulated in situ.

Finally the PVDF raw polymer is known to be of extremely high purity, giving rise to the widespread usage of the material in piping systems for high-purity water systems (typically used in the semiconductor manufacturing industry). The ZOTEK F foams, being blown with nitrogen gas, can likewise be considered to be high purity foams. As a means of demonstrating the high purity, the ZOTEK F foams have been tested in accordance with the requirements of ISO 10993: Biological Evaluation of Medical Devices. The materials were shown to pass the requirements of Pharmacopoeia Monograph USP 661 testing. In extraction tests conducted at the same time, using isopropanol and water as the extraction media, total maximum extractables were found to be of the order of 30 ppm, none of which were considered harmful in any way. The conclusion from this work is that the high-purity PVDF polymer gives rise to a high-purity PVDF foam, one that is suitable for use in medical devices for use in skin contact as well as those in contact with mucosal membranes or breached and compromised surfaces for limited or prolonged exposure (up to 30 days).

FIGURE 7.12 SEM of the expanded ZOTEK F PVDF foam structure.

244 Polymeric Foams

61259_C007.indd 24461259_C007.indd 244 10/25/2008 12:39:03 PM10/25/2008 12:39:03 PM

7.7.1.5 Markets and Applications

As a result of the low density, inherent fl ame resistance, and low smoke properties of the product, the material has quickly been accepted by the mass transport industry. The ease of moldability, closed-cell nature (mois-ture barrier), and therefore the single material solution are of obvious benefi t in thermal/acoustic insulation applications.

A further very recent application development with the ZOTEK F grade foams is as insulation in cleanroom environments.22 This is being led by a product developed by UFP Technologies Inc. in the US. The products have been branded T-Tubes®.23 The foam is easily moldable and the product is typically formed into a clamshell confi guration. The shaped foam is then simply fi xed around pipework to both insulate and to protect personnel who often have to operate or maintain hot pipework and equipment within space limitations. Examples of the molded T-Tubes themselves are shown in Figure 7.13. Figure 7.14 then shows part of an installation with the T-Tubes fi tted. The advantages of the T-Tubes approach are many and apart from material performance which is outstanding, they are extremely fl exible, very clean and simple to clean once installed, easy to handle and easily formed/cut to shape on site. The latter is an important operational aspect which relates to speed and ease of installation without the produc-tion of dust, swarf, or other residues which could contaminate otherwise clean environments in the pharmaceutical, semiconductor, food, and other sensitive process industries.

The T-Tubes themselves have been tested to both the FM 4910 test stan-dard mentioned earlier as well as a further FM specifi cation related to the

FIGURE 7.13 Examples of the molded T-Tubes for cleanroom insulation.


61259_C007.indd 24561259_C007.indd 245 10/25/2008 12:39:04 PM10/25/2008 12:39:04 PM

fl ammability of pipe insulation, namely FM 492424 and is proven compli-ant. It is believed that T-Tubes will be the only closed-cell foam insulation on the market supported by both approvals—truly a new industry stan-dard for cleanroom insulation.

7.7.2 ZOTEK N Polyamide (PA 6) Foams

The fi rst ZOTEK N foam was launched commercially in late 2006. These materials remain in the market and application development phase of their lifecycle. The initial commercial grade, ZOTEK N B50, is a relatively stiff grade of foam based on the nylon 6 polymer. The solid has a density of around 1100 kg/m3 and the foam has a density of around 50 kg/m3 equating to a volume expansion of 22 times. Like PVDF, polyamides rep-resent a very large family of resins and have also been widely exploited in blends and alloys with other polymers to modify specifi c properties.

Nylon 6 is a relatively old polymer having been developed originally in the 1940s; however, it remains a polymer of signifi cant importance in both fi ber production and injection molding. The material is a rigid, crystalline polymer. Two important characteristics dictate the fi nal performance of the material. The fi rst is the level of crystallinity, which is highly depen-dent on the cooling regime used during processing and can vary consid-erably. The second is the level of moisture. Polyamides are hygroscopic in nature and equilibrium levels of moisture of 4–5% by weight are typical.

FIGURE 7.14 Cleanroom installation showing the T-Tubes fi tted in position.

246 Polymeric Foams

61259_C007.indd 24661259_C007.indd 246 10/25/2008 12:39:04 PM10/25/2008 12:39:04 PM

The plasticization of the polymer with such levels of moisture can lead to profound property differences when compared to the “dry” product.

Additives also provide a means to further tailor performance to particu-lar application requirements. The key end-use attributes of nylon 6 are temperature performance, hydrocarbon oil/fuel resistance, toughness, and a good cost/performance balance.

7.7.2.1 Thermal Properties

The temperature stability and continuous use temperature ranges of low-density foams are normally assessed via high temperature dimensional stability testing. The reason for this is related to the cross-linking/increas-ing molecular weight process step in most foaming processes. In the majority of such processes, the cross-linking or molecular weight increase must take place prior to foam expansion to ensure that there is adequate melt strength during expansion, thereby enabling a wide enough foam processing window.

In the results presented in Figure 7.15, 100 ¥ 100 ¥ 25 mm samples of the foam were simply exposed to a range of temperatures in an air circulating oven for a 24-hour period and the linear dimensional change was assessed for each sample after a further period of stabilization at room temperature.

FIGURE 7.15 Linear shrinkage results for the ZOTEK N polyamide foam exposed to a

range of temperatures (comparison with XL-PE foam of similar density).

0 25 50 75 100 125 150 175 200 225 2500

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80ZOTEK N B50XL-PE (45 kg/m3)

Line

ar s

hrin

kage

(%

)

Temperature (°C)


61259_C007.indd 24761259_C007.indd 247 10/25/2008 12:39:04 PM10/25/2008 12:39:04 PM

In the XL-PE foam market, the widely accepted defi nition of the maxi-mum operating temperature is the temperature at which 5% linear shrink-age is produced after the 24-hour exposure period. For the ZOTEK N B50 foam this point is around 210°C, some 100°C higher than most commer-cially available foams, including the equivalent density XL-PE foam used here for comparison.

Insulation is one of the largest general applications of foam materials, making good use of the typically very low thermal conductivity (k) of such materials. The ZOTEK N B50 foam is no different in this respect and the thermal conductivity has been measured over a wide range of temper-atures using the method described in ISO 8301. The results shown in Figure 7.16 show the change in thermal conductivity with mean test tem-perature. Of note is the reasonably fl at response over the temperature range of 0 to 170°C. At ambient temperatures the k value is in the region of 0.038 W/m K, while at 170°C the material still has an extremely low k value of 0.051 W/m K.

Clearly where high temperature insulation is required then this is an attractive proposition and due to the fl at response over this temperature range one can envisage the displacement of other higher density or less robust insulation systems with the ZOTEK N foam.

FIGURE 7.16 Plot showing effect of mean test temperature on the thermal conductivity

(ISO 8301) for the ZOTEK N polyamide foam.

–20 0 20 40 60 80 100 120 140 160 180 200

0.000

0.010

0.020

0.030

0.040

0.050

0.060

The

rmal

con

duct

ivity

, k (

W/m

K)

Mean test temperature (°C)

248 Polymeric Foams

61259_C007.indd 24861259_C007.indd 248 10/25/2008 12:39:05 PM10/25/2008 12:39:05 PM

7.7.2.2 High Temperature Modulus and Moisture

Thermal stability tests such as those described above are useful to a limited degree as they show only whether the material is relatively dimensionally stable but give little indication of the mechanical integrity at such tempera-tures. To shed more light on this aspect, dynamic mechanical thermal anal-ysis (DMTA) testing was undertaken in fl exural mode on the ZOTEK N B50 foam to evaluate the modulus over a range of temperatures in both “dry” and “conditioned” (equilibrium moisture) states. These results are shown in Figures 7.17 and 7.18. As noted previously, polyamides are hygro-scopic and it is usual for these materials to be used in the conditioned state where properties may be considered relatively stable and at equilibrium.

The dry and conditioned foam samples both show an expected glass transition temperature (Tg) of around 65°C, which is roughly unchanged by moisture level, although more pronounced in the dry material. In the range �10 to �10°C the effect of moisture is more clear—the dry sample yields almost no peak and very low tan d whereas in the case of the condi-tioned material, the peak is well defi ned and the peak position is shifted to the left on the axis.

Of more academic interest, the dry samples show, as would be expected, a marked stiffening at ambient temperatures (relative to the conditioned samples) and it is not until a temperature of more than 75°C or so that the moduli of the dry and conditioned samples begin to converge. The lower tan d and less pronounced peak in the dry sample would suggest more

FIGURE 7.17 DMTA plots for the ZOTEK N polyamide foam in the dry state (curves show

results of tests at frequencies of 1 Hz and 10 Hz).

–50 –25 0 25 50 75 100 125 150 175 200 2250.0

1.0×106

2.0×106

3.0×106

4.0×106

5.0×106

6.0×106

7.0×106

8.0×106

9.0×106

1.0×107

1.1×107

1.2×107

'Dry' ZOTEK N B50

Mod

ulus

(P

a)

Temperature (°C)

0.000

0.020

0.040

0.060

0.080

0.100

0.120

0.140

0.160

0.180

0.200

0.220

0.240

tan

d


61259_C007.indd 24961259_C007.indd 249 10/25/2008 12:39:05 PM10/25/2008 12:39:05 PM

brittle behavior may be expected at or around ambient temperatures, a fact borne out by the basic mechanical results presented in Table 7.7.

At around room temperature the modulus of the material using DMTA is 1.8 MPa. Applying a typical approach of 50% loss of modulus as an arbitrary limit for continuous use of the material, the plots defi ne a useful upper temperature in the range 170–175°C. Few low-density foam materi-als in commercial use today are able to retain load-bearing capabilities at such temperatures.

The results above have implications both for the material (i.e. use of the dry material at room temperature should be avoided or discouraged) and positive benefi t in applications requiring load-bearing capability allied to light weight at high temperatures.

TABLE 7.7

Basic Mechanical Test Results for the ZOTEK N Polyamide Foam Showing Effect of Absorbed Moisture

Property Method Dry Conditioned

Compression stress @ 25% strain (kPa) ISO 7214 510 275

Compression stress @ 50% strain (kPa) ISO 7214 605 345

Tensile strength (MPa) ISO 1798 1.54 1.25

Elongation at break (%) ISO 1798 60 75

Tear strength (N/m) ISO 8067 1840 3000

FIGURE 7.18 DMTA plots for the ZOTEK N polyamide foam in the conditioned state

(curves show results of tests at frequencies of 1 Hz and 10 Hz).

'Conditioned' ZOTEK N B50

–50 –25 0 25 50 75 100 125 150 175 200 2250.0

1.0×106

2.0×106

3.0×106

4.0×106

5.0×106

6.0×106

7.0×106

8.0×106

9.0×106

1.0×107

1.1×107

1.2×107

Mod

ulus

(P

a)

Temperature (°C)

0.000

0.020

0.040

0.060

0.080

0.100

0.120

0.140

0.160

0.180

0.200

0.220

0.240

tan

d

250 Polymeric Foams

61259_C007.indd 25061259_C007.indd 250 10/25/2008 12:39:06 PM10/25/2008 12:39:06 PM

The range of polyamide resin materials and alloys/blends available is very large. A key target for Zotefoams is to develop a more fl exible polyamide foam product without signifi cantly compromising the temper-ature resistance properties. Other functional properties such as recovery/hysteresis are also being investigated. This work is progressing well and a family of polyamide foams under the ZOTEK N brand will certainly develop to cover the specifi c needs of various end-user markets.

7.7.2.3 Other Performance Attributes

The descriptions above cover the main attributes that are derived from the polyamide polymer with regard to how they infl uence the behavior of the foamed product. Other characteristics, however, come solely from the use of the nitrogen autoclave process itself as a result of using a nitrogen pneumatogen. The foams tend to be low VOC (volatile organic com-pounds), low fogging, and low odor. These characteristics are of specifi c interest to the automotive sector for car interior applications. Finally, the impact performance of the ZOTEK N foams is yet to be fully evaluated but the rigid feel of the material is indicative of a material with high energy absorbing characteristics.

7.7.2.4 Markets and Applications

Low-density polyamide foams are a new material for the design commu-nity to handle and to understand and for that reason much work is currently focused on application development and technical support activity. Key areas of ongoing effort are in the automotive sector where the use of the material in such applications as under-bonnet insulation and noise control is being evaluated. In the same industry the material is also being assessed in press-formed parts for spacers and seals (including wire and cable management) close to or within the engine compartment. Outside the automotive industry the ZOTEK N material has found interest in the composite-manufacturing sector as a relatively low-cost/high- temperature core material.

7.8 Conclusions

The nitrogen autoclave process is a well-established process which is widely recognized within the industry as being a superior process tech-nology. The technology has been producing the highest quality and most consistent cross-linked polyolefi n foam products for four decades. The use of the technology continues to develop with the recent introduction of


61259_C007.indd 25161259_C007.indd 251 10/25/2008 12:39:06 PM10/25/2008 12:39:06 PM

the ZOTEK brand foams moving outside of the polyolefi n family of mate-rials for the fi rst time commercially.

These new and unique products offer the market and design commu-nity high performance through combinations of:

Outstanding fl ame, smoke, and toxicity performance•

Excellent UV resistance•

Broad chemical resistance•

Very low thermal conductivity•

Extremely high purity•

Outstanding high temperature dimensional stability•

Low thermal conductivity retained over wide temperature range•

High-temperature load-bearing capability.•

References

1. Cooper, A. The Story of Expanded Rubber. Expanded Rubber Company Ltd., Croydon, U.K., 1957.

2. Eaves, D. E. “New foams from the nitrogen autoclave process.” Paper pre-sented at the 2nd International Cellular Polymers Conference, Heriot Watt University, Edinburgh, 23–25 March 1993.

3. Lee, J. G. and Flumerfelt R. W. “Nitrogen solubilities in low-density polyethylene at high temperatures and high pressures.” Journal of Applied Polymer Science 58 (1995): 2213–2219.

4. Sato, Y. Fujiwara, K., Sumarno, T. S., and Masuoka, H. “Solubility of carbon dioxide and nitrogen in polyolefi ns and polystyrene under high pressures and temperatures.” Presented at the 5th Meeting on Supercritical Fluids and Natural Products Processing, Nice, France, 1998.

5. Crank, J. The Mathematics of Diffusion, 2nd edition. Clarendon Press, Oxford, 1975.

6. Park, C. B., Baldwin, D. F., and Suh, N. P. “Effect of pressure drop rate on cell nucleation in continuous processing of microcellular polymers.” Polymer Engineering and Science 35 (1995): 432–440.

7. Kim, S. G., Lee, J. W. S., Park, C. B., and Sain, M. “Strategies for enhancing cell nucleation of thermoplastic polyolefi n (TPO) foam.” In Proceedings of the Annual Technical Conference (ANTEC), Society of Plastics Engineers (SPE), Cincinnatti, OH, May 2007: 2099–2104.

8. Rodriguez-Perez, M. A., Gonzalez-Pena, J. I., Witten, N., and de Saja, J. A. “The effect of cell size on the physical properties of crosslinked closed cell polyethylene foams produced by a high pressure nitrogen solution process.” Cellular Polymers 21 (2002): 165.

252 Polymeric Foams

61259_C007.indd 25261259_C007.indd 252 10/25/2008 12:39:06 PM10/25/2008 12:39:06 PM

9. Eaves, D. E. “The properties of crosslinked foams produced from metal-locene polyolefi ns.” Presented at the Metallocene Technology Seminar, RAPRA Technology Ltd., Shawbury, U.K., 2 September 1997.

10. US 59955015 [Evolution Foam Mouldings Ltd.] 11. Moore, S. “Crosslinking technique is low-cost and effi cient.” Modern Plastics

(2002): 42. 12. Klempner, D. and Frisch, K. C. Handbook of Polymeric Foams and Foam

Technology. Hanser, Munich, Germany, 1991. 13. Eaves, D. E. Handbook of Polymer Foams. RAPRA Technology Ltd., Shawbury,

U.K., 2004. 14. Trageser, D. A. “Crosslinked polyethylene foam processes.” Radiation Physics

and Chemistry 9 (1977): 261–270. 15. Puri, R. R. and Collington, K. T. “The production of cellular crosslinked

polyolefi ns: Part 2—The injection moulding and press moulding techniques.” Cellular Polymers 7 (1988): 219–231.

16. Witten, N. “Extending the conventional boundaries of crosslinked polyolefi n foam production.” Presnted at FOAMPLAS, Teaneck, NJ, 19–20 May 1998.

17. WO 2005/105907 (Zotefoams plc). 18. WO 2006/077395 (Zotefoams plc). 19. See http://www.fmglobal.com/assets/pdf/fmapprovals/4910.pdf. 20. See http://www.gpoaccess.gov/index.html. 21. Van der Weide, I., and Werth, M. High Purity Plus Outstanding Resistance—

PVDF Foams. Kunstoffe, Munich, Germany, 2006, p. 54. 22. Partridge, R. “Fungal, chemical, and fi re resistance of PVDF foams and

polymers.” Controlled Environments (2007): 14–18. Available at: http://www.cemag.us/articles.asp?pid=743.

23. See http://www.t-tubes.com. 24. See http://www.fmglobal.com/assets/pdf/fmapprovals/4924.pdf.


61259_C007.indd 25361259_C007.indd 253 10/25/2008 12:39:06 PM10/25/2008 12:39:06 PM

61259_C007.indd 25461259_C007.indd 254 10/25/2008 12:39:06 PM10/25/2008 12:39:06 PM

8Polystyrene Foam and Its Improvement in Vacuum Insulated Panel Insulation

Chang-Ming Wong

CONTENTS

8.1 Introduction of Vacuum Insulation Panel ................................. 2558.2 PS Characteristics ......................................................................... 2598.3 PS Foaming ................................................................................... 2608.4 Heat Transfer in Plastic Foams ................................................... 2728.5 Thermal Conductivity of VIPs Using PS

Foams as Core Materials ............................................................. 2798.6 Conclusions ................................................................................... 2858.7 Abbreviations ................................................................................ 2858.8 Nomenclature ............................................................................... 286References ............................................................................................. 287

8.1 Introduction of Vacuum Insulation Panel

Plastics containing many cells or bubbles are called foamed plastics or plastic foams. Plastic foams have many unique characteristics such as a light weight, shock absorption, and good insulation. Generally, the cell structures of plastic foams can be classifi ed into either a closed-cell struc-ture or an open-cell (porous) structure. A closed-cell structure, where a

61259_C008.indd 25561259_C008.indd 255 10/25/2008 12:39:54 PM10/25/2008 12:39:54 PM

large proportion of the cells are independent of each other, is easy to produce during foaming. However, an open-cell structure, where a large propor-tion of the cells are connected to other cells by open passage, can be more diffi cult to achieve during foaming.

Vacuum insulation panels (VIPs) are a new generation product for thermal insulation and have only been produced in the past ten years. With the price of crude oil increasing, energy saving is an important issue nowa-days. VIPs mainly consist of an open-cell material and multilayer barrier fi lms, the open-cell material as a core material being encapsulated by multi-layer fi lms. Figure 8.l schematically illustrates VIP structure. The outward appearance of VIP looks like a pouch, as shown in Figure 8.1a. The pouch is then evacuated to a vacuum ranging from 1.33 Pa (0.01 torr) to 133.3 Pa (1 torr) and sealed afterwards to form a VIP, as illustrated in Figure 8.1b.

An addition to a core material and multilayer fi lm, getter(s) and/or desiccant(s) are also placed in the VIP to ensure the performance of the VIP during the projected lifetime. The purpose of a getter is to absorb the diffusing gases, while the desiccant is for absorbing moisture from the core material to maintain required vacuum levels in the VIP.1 Several porous materials such as open-cell polystyrene (PS) foam,2–14 open-cell polyurethane (PU) foam,15–17 aerogel,18,19 or silica powder20–22 can be used as core materials in VIPs. Basically, these porous materials have a very high content (more than 95%) of open-cells or pores.

Multilayer fi lms mainly include a protective layer (external layer), bar-rier layer (core layer), and sealing layer (inside layer) and are made by lamination or coextrusion.23–25 The protective layer is responsible not only for impact resistance, scratch resistance, and chemical resistance, but also for optical properties. The barrier layer provides multilayer fi lms with a low diffusion for gas or moisture from the atmosphere. The sealing layer is primarily responsible for sealing performance in multilayer fi lms.

Multilayer fi lms used in VIPs can be classifi ed into three types:

1. Those containing an aluminum (Al) foil as a barrier layer, an outer-most fi lm as a protective layer such as nylon (NY) or polyethylene terephthalate (PET) or metallized PET fi lm and an innermost fi lm as a sealing layer such as high-density polyethylene (HDPE) fi lm;

Vacuum

Multi layer-films

Core materialDesiccant(s)/getters

(a) (b)

FIGURE 8.1 Schematic representation of a VIP.

256 Polymeric Foams

61259_C008.indd 25661259_C008.indd 256 10/25/2008 12:39:55 PM10/25/2008 12:39:55 PM

for example, NY (15 μm)/PET (12 μm)/Al foil (5–25 μm)/HDPE (15 μm) shown in Figure 8.2a.

2. Those containing only polymeric fi lms, and ethylene vinyl alco-hol (EVOH) as a barrier material, such as NY/PET/EVOH/HDPE described in Figure 8.2b.

3. Those containing metallized fi lms as a barrier layer without Al foil; for instance, metallized polyester fi lm/sealant fi lm illustrated in Figure 8.2c.

The thermal conductivity of PS foams at the condition of one atmosphere is approximately 0.03 W/m.k.26 When the internal pressure in VIP reduces to 13.3 Pa (0.1 torr), the thermal conductivity of VIP can be lowered to between 0.009 W/m k (R � 16) and 0.006 W/m k (R � 24). However, the plastic foams used in VIP must not only have a high content of open cells, but also have a good mechanical strength. Figure 8.3 displays the photo-graph of a poor VIP. The porous PS foam is unable to maintain the original dimension under vacuum packing due to a poor mechanical strength of the porous PS foam. Figure 8.4 presents the photograph of a normal VIP. The porous PS foam has a good mechanical strength to keep the original size under vacuum packaging.

Protect layer

Sealing layer

(a)

(c)

(b)

Nylon filmAdhesive

Adhesive

Adhesive

PET film

HDPE film

Aluminum foil

Protect layer

Sealing layer

Adhesive

AdhesiveHDPE film

Polyester

Polyester

Metallization

Metallization

Protect layer

Sealing layer

Nylon filmAdhesive

Adhesive

Adhesive

PET film

HDPE film

EVOH foil

FIGURE 8.2 Schematic representation of multilayer fi lms.

Polystyrene Foam and Its Improvement 257

61259_C008.indd 25761259_C008.indd 257 10/25/2008 12:39:55 PM10/25/2008 12:39:55 PM

Because multilayer fi lms used in VIPs are unable to maintain the per-formance in a high-temperature environment, VIPs are mainly introduced into refrigerating applications such as shipping containers, reefers, and refrigerators for housing, boat, vehicles, and hotels. VIPs are generally three to seven times better in thermal insulation at an equivalent thick-ness than conventional products such as PU foam (R ~ 5.5), PS foam (R ~ 4), and fi berglass batting (R ~ 3.3).

Commercial goods using vacuum insulation use less thickness than conventional insulation under the same thermal insulation. Therefore, refrigerators using VIP can increase internal storage capacity in restricted spaces. VIPs can also reduce electrical power consumption and decrease

FIGURE 8.3 Poor mechanical property of a porous PS foam as a core material for a VIP.

FIGURE 8.4 Good mechanical property of a porous PS foam as a core material for a VIP.

258 Polymeric Foams

61259_C008.indd 25861259_C008.indd 258 10/25/2008 12:39:55 PM10/25/2008 12:39:55 PM

coolant requirements. PU foam is a basic material for thermal insulation in refrigerators. When 40% PU foam in the wall of a refrigerator is replaced with PS/VIPs, the energy factor (L/KWhr/month) for the refrigerator can be enhanced by about 30%.27 Many companies currently apply vacuum insulation technology for manufacturing refrigerators or shipping con-tainers. Examples include Sharp (Japan), Sanyo (Japan), Matsushtia (Japan), AEG-Electrolux (Europe), Glacier Bay (USA), and Acutemp (USA).

8.2 PS Characteristics

The structure of styrene (phenyl ethylene or vinyl benzene) is CH2�CH2C6H5. Styrene monomer can polymerize to form a homopolymer or polymerize with other monomers such as acrylonitrile, butadiene, and alpha-methyl styrene to become copolymers or terpolymers. The homo-polymer, copolymers, and terpolymers are called styrene-based polymers. Generally, there are two popular types of PS available in the market, one is general-purpose PS (GPPS)28,29 and the other is high-impact PS (HIPS).28,30,31 GPPS is formed by a styrene monomer under free-radical polymerization. PS produced by free-radical polymerization has an atactic confi guration that leads to an amorphous polymer with a glass transition temperature (Tg) of approximately 100°C. Moreover, styrene polymer synthesized by metallocene catalysts can yield syndiotactic PS (sPS), which is a crystalline polymer with a melting temperature around 270°C and a Tg of approximately 100°C.32,33

GPPS basically has three commercial grades: easy fl ow, medium fl ow, and high heat.

Easy-fl ow resins have the lowest molecular weight than the other • two grades and contain 3–4% mineral oil in resin to reduce melt viscosity. In addition to the application of injection molding, the resins can be used for coextruded packaging applications.

Medium-fl ow resins generally contain 1–2% mineral oil and have • melt fl ow properties intermediate between the other grades. Applications for medium-fl ow resins also include injection-molded products and extruded or coextruded food packaging.

High-heat resins are the highest molecular weight resins and • contain no mineral oil and low concentrations of additives such as mold-release and extrusion aids. The major applications for the resins are foam, sheet, and fi lm extrusion.

As a matter of fact, GPPS is brittle in nature and the impact strength is poor, if unoriented. In order to improve the defect, styrene is grafted with


61259_C008.indd 25961259_C008.indd 259 10/25/2008 12:39:55 PM10/25/2008 12:39:55 PM

a rubber such as polybutadiene to become HIPS. HIPS resins are much softer and have better impact strength than GPPS resins. This is due to the existence of rubber particles dispersed in a continuous matrix of polysty-rene. At a constant rubber level, increasing rubber particle sizes from average diameter 0.6 to 3.5 �m in HIPS can enhance Izod impact strength of HIPS from 48 to 100 J/m.34 However, if the rubber particles have diam-eters much larger than 10 �m in HIPS, the fi nished product has a low gloss surface and toughness decreases.

Table 8.1 shows the material characteristics of four PS resins including three GPPS and one HIPS (PS-4). The Tg is measured by a differential scan-ning calorimeter (DSC). Generally, a higher weight average molecular weight (Mw) for GPPS has a higher heat distortion temperature (HDT), tensile strength, and impact strength. However, impact strength is poor for GPPS and is not dramatically infl uenced by Mw. The material properties of HIPS do not have any relationship with Mw and the impact strength is much better than that of GPPS.

8.3 PS Foaming

PS can be foamed by chemical foaming agents (CFAs) such as azodicar-bonamide derivative and physical foaming agents (PFAs) such as carbon dioxide (CO2), nitrogen (N2), hydrocarbons, and hydrofl uorocarbons. Batch and continuous process are general methods to manufacture PS foam. When PS is foamed by CFAs, the foam density of PS foams normally ranges from 600 kg/m3 (0.6 g/cm3) to 800 kg/m3 (0.8 g/cm3) and a closed-cell structure exists in PS foams. PFAs used in foaming process can create PS foams with density less than 100 kg/m3 (0.1 g/cm3). The cell structures

TABLE 8.1

Characteristics of Polystyrene Resins

Properties (Units)

GPPS HIPS

PS-1 PS-2 PS-3 PS-4

Melt fl ow index (g/10 min) 2.2 5.0 8.0 3.0

Specifi c gravity 1.05 1.05 1.05 1.05

Tg (DSC) (°C) 107 97 95 103

HDT (°C) 86 83 78 83

Tensile strength (kg/cm2) 540 480 440 240

Tensile elongation (%) 2.0 2.0 2.0 45

Izod impact strength (kg-cm/cm)

(notched, 1/4”)

1.7 1.6 1.5 10.8

Mw (�103) 280 264 253 225

260 Polymeric Foams

61259_C008.indd 26061259_C008.indd 260 10/25/2008 12:39:55 PM10/25/2008 12:39:55 PM

in PS foam made by PFAs are either a closed-cell structure or an open-cell (porous) structure.

There are many theoretical and experimental works associated with PS foamed by CO2 under supercritical conditions in either a batch or a continuous process.35–43 There are two types of batch process used to make microcellular foam. One is that a resin is soaked with foaming agents at room temperature or a little higher than room temperature and high pres-sure, and then the resin is subjected to a high-temperature environment to form a microcellular foam; the other is that a resin is saturated with foam-ing agents at high temperatures and high pressure, and then the resin is foamed at a high pressure drop rate to obtain a microcellular foam. The fi rst method has a much longer saturation time than the second method. Figures 8.5 and 8.6 schematically describe the batch and continuous foam-ing system respectively. PS foam with a cell size of less than 10 �m can be obtained by a batch process. It is diffi cult to obtain a PS foam with cell size smaller than 10 μm in a continuous process because the pressure drop rate for a continuous process is much smaller than that for a batch process, and the uniform foaming temperature in foaming materials for a continuous process is more diffi cult to control.40–43 Basically the cells in most neat microcellular PS foams produced by CO2 have a closed-cell structure.

Since a foaming system using PFAs is better at making open-cell PS foam, several PFAs are generally applied to foaming systems, for example, aliphatic hydrocarbon with four to six carbon atoms2 or a mixture of gas such as CO2, ethyl chloride (EtCl), and partially halogenated aliphatic hydrocarbons including chlorodifl uoromethane (HCFC-22), 1-chloro-1,1-difl uoroethane (HCFC-142b), 1,1,1,2-tetrafl uoroethane (HFC-134a), or 1,1-difl uoroethane (HFC-152a).3–11

P1

15

3

7

6 4

2

P2

P3

1,2 Tank

3,4 Stable pressure unit

5,6 Injection unit

Lower heater

Lower die

Upper die

Upper heater

FIGURE 8.5 Schematic representation of the batch foaming system.


61259_C008.indd 26161259_C008.indd 261 10/25/2008 12:39:55 PM10/25/2008 12:39:55 PM

Blending of PS and other materials such as styrene/ethylene butylene/styrene (SEBS),8 ethylene/octane copolymer (EO),8 and ethylene/styrene interpolymer (ESI),9–11 is also an important method of producing open-cell PS foams because SEBS, EO, and ESI can act as cell-opening agents during PS foaming. A theoretical study indicates that the development of an open-cell structure includes two stages, namely bubble growth to impingement and then cell wall thinning to rupture.44

Figure 8.7 presents the foam density of four neat PS foams at various foaming temperatures and a foaming pressure of 20.7 MPa (3000 psi). The four neat PS sheets are foamed by CO2 as the foaming agent at foaming temperatures ranging from 110°C to 140°C in a batch process. The material characteristics of four heat PS resins are shown in Table 8.1. Generally, the four neat PS foams have a high foam density at lower foaming tempera-tures. The PS-1 foam always displays a higher foam density than PS-2, PS-3, and PS-4 foams. PS-4, HIPS resin has a lower Mw than PS-2 and PS-3 resins, but the foam density of the PS-4 resin is higher than that of the PS-2 and PS-3 resins. When the foaming temperature ranges between 110°C and 125°C, the foam density of the PS-1 resin is much higher than

Motor &gear drive

Polymer meltingGas/polymer mixing

cooling

(a)

(b)

Gas/polymer cooling

Foamingshaping

DieAdaptor

Gas inletExtruderHopper

Motor &gear drive

Motor &gear drive

Polymer meltingGas/polymer

mixing

Foamingshaping

DieAdaptor

Gas inletExtruder

Extruder

Hopper

FIGURE 8.6 Schematic representation of the continuous foaming system: (a) one extruder;

and (b) two extruders.

262 Polymeric Foams

61259_C008.indd 26261259_C008.indd 262 10/25/2008 12:39:56 PM10/25/2008 12:39:56 PM

that of the other three PS resins. As the foaming temperature rises above 130°C, the difference of foam density for four PS resins becomes narrow. Lower foam density for all four of the PS resins is observed in the range 130°C to 140°C.

Figure 8.8 illustrates the foam density of four PS resins blended with 2 phr CaCO3 (two parts CaCO3 per hundred parts of PS resin by weight) at various foaming temperatures. A higher foam density at lower foaming temperatures is observed for the four PS/2 phr CaCO3 foams. The foam density strikingly increases at the foaming temperature of 140°C. In com-parison with the foam density of the four neat PS resins in Figure 8.7, the occurrence of a low foam density for the four PS/2 phr CaCO3 compounds shifts to low foaming temperatures ranging from 125°C to 135°C. The situ-ation indicates that CaCO3 is able to change the viscoelastic property of PS resins typically seen with inorganic fi llers.45,46

The foam density of four PS resins blended with 4 phr CaCO3 and 6 phr CaCO3 are exhibited in Figures 8.9 and 8.10, respectively. The low foam

120

100

80

60

Den

sity

(kg

/m3 )

40

20

0100 110 120

Temperature (°C)130

PS

PS-1PS-2PS-3PS-4

140 150

FIGURE 8.7 Dependence of PS foam density on foaming temperatures.

250

PS/2 phr CaCO3200

150

Den

sity

(kg

/m3 )

100

50

0100 110 120

Temperature (°C)

130

PS-1PS-2PS-3PS-4

140 150

FIGURE 8.8 Dependence of PS/2 phr CaCO3 foam density on foaming temperatures.


61259_C008.indd 26361259_C008.indd 263 10/25/2008 12:39:56 PM10/25/2008 12:39:56 PM

density for the four PS/4 phr CaCO3 foams and most PS/6 phr CaCO3 foams occurs at foaming temperatures of between 120°C and 130°C. Moreover, the foaming temperatures having a low foam density for PS-3/6 phr CaCO3 foam become narrow, but are still near 120°C. This indicates more CaCO3 in PS resin cannot further decrease the foaming temperatures at which a low foam density of PS resin occurs. The foam density at the foaming temperature of 140°C for the four PS/6 phr CaCO3 foams is much higher than that for the four PS/4 phr CaCO3 and PS/2 phr CaCO3 foams. Generally, the more CaCO3 is added to the PS resin, the higher PS foam density is obtained during PS foaming.

The cells of the four neat PS foams and PS/2 phr CaCO3 foams at the foaming temperature of 125°C are shown in Figures 8.11 and 8.12, respec-tively. Figure 8.11a–d shows PS-1, PS-2, PS-3, and PS-4 foams, respectively, and Figure 8.12a–d shows PS-1/2 phr CaCO3, PS-2/2 phr CaCO3, PS-3/2

PS/6 phr CaCO3

200

300

400

500

600

700

Den

sity

(kg

/m3 )

100

0100 110 120

Temperature (°C)

130

PS-1PS-2PS-3PS-4

140 150



PS/4 phr CaCO3

100

150

200

250

300

350

400

Den

sity

(kg

/m3 )

50

0100 110 120

Temperature (°C)

130

PS-1PS-2PS-3PS-4

140 150

264 Polymeric Foams

61259_C008.indd 26461259_C008.indd 264 10/25/2008 12:39:56 PM10/25/2008 12:39:56 PM

phr CaCO3, and PS-4/2 phr CaCO3 foams, respectively. Few open cells are generally observed in either the four PS or four PS/2 phr CaCO3 foams at the foaming temperature of 125°C. Figures 8.13 and 8.14 describe the cells of four neat PS foams and PS/2 phr CaCO3 foams at the foaming temperature of 135°C, respectively. Figure 8.13a–d represents PS-1, PS-2, PS-3, and PS-4 foams, respectively. Only a few pores exist in each of the four foams. The content of open cells for PS-3 foam is slightly higher than that of the other foams and the porous structure is scarcely observed in the PS-4 foam. PS-1/2 phr CaCO3, PS-2/2 phr CaCO3, PS-3/2 phr CaCO3, and PS-4/2 phr CaCO3 foams are indicated in Figure 8.14a–d, respectively. The content of pores in the PS/2 phr CaCO3 foams is signifi cantly high in comparison with that in the neat PS foams. In particular, PS-2/2 phr CaCO3 and PS-3/2 phr CaCO3 foams have a very high content of open cells. Therefore, CaCO3 can act as a cell-opening agent during PS foaming.47

Figures 8.15 and 8.16 show the average cell sizes of four neat PS and PS/2 phr CaCO3 foams at the foaming temperatures of 125°C and 135°C, respec-tively. In general, the average cell sizes of the three neat GPPS at the foam-ing temperature of 135°C are slightly larger than that at the foaming temperature of 125°C. However, the average cell size of neat HIPS foam at

FIGURE 8.11 Cells in PS foams at the foaming temperature of 125°C: (a) PS-1; (b) PS-2;

(c) PS-3; and (d) PS-4.


61259_C008.indd 26561259_C008.indd 265 10/25/2008 12:39:57 PM10/25/2008 12:39:57 PM

the foaming temperature of 135°C is much larger than that at the foaming temperature of 125°C. The average cell sizes of three GPPS/2 phr CaCO3 foams at the foaming temperature of 135°C are, on the contrary, smaller than that at the foaming temperature of 125°C, but the average cell sizes of HIPS/2 phr CaCO3 foam are very close at the two foaming temperatures of 125°C and 135°C. The three GPPS/2 phr CaCO3 foams have much smaller average cell sizes than the three neat GPPS foams.

The foam density of four PS/2 phr LDPE, PS/5 phr LDPE, and PS/7 phr LDPE foams are presented in Figures 8.17, 8.18, and 8.19, respectively. The trend of foam density for PS blended with LDPE at the foaming tempera-tures from 110°C to 140°C is similar to the trend of foam density for PS blended with CaCO3 indicated in Figures 8.8–8.10. A PS/LDPE foam has a higher foam density at the same foaming temperature than a neat PS foam. A striking rise in foam density for the four PS/LDPE foams is observed at the foam temperature of 140°C, the occurrence of a low foam density for the four PS/LDPE foams shifts to low foaming temperatures ranging from 120°C to 130°C. The results indicate both LDPE and CaCO3

FIGURE 8.12 Cells in PS/2 phr CaCO3 foams at the foaming temperature of 125°C: (a) PS-1;

(b) PS-2; (c) PS-3; and (d) PS-4.

266 Polymeric Foams

61259_C008.indd 26661259_C008.indd 266 10/25/2008 12:39:57 PM10/25/2008 12:39:57 PM

can cause different grades in the shrinkage of PS foams and change the viscoelastic property of PS resins.

The cells of four PS blended with 2 phr LDPE and 7 phr LDPE foams at the foaming temperature of 125°C are described in Figures 8.20 and 8.21, respectively. Figure 8.20a–d represents PS-1/2 phr LDPE, PS-2/2 phr LDPE, PS-3/2 phr LDPE, and PS-4/2 phr LDPE foams, respectively. Small cell sizes for four PS/2 phr LDPE foams are observed and an open-cell structure exists in some PS/2 phr LDPE foams. Figure 8.21a–d illustrates PS-1/7 phr LDPE, PS-2/7 phr LDPE, PS-3/2 phr LDPE, and PS-4/2 phr LDPE foams, respectively. The cell sizes for the four PS/7 phr LDPE foams are close to that for the four PS/2 phr LDPE foams. However, open-cell structure slightly increases for PS/7 phr LDPE foams.

Figure 8.22 shows the comparison of average cell sizes for four neat PS and PS/LDPE foams produced at the foaming temperature of 125°C. The average cell sizes of four PS blended with 2 phr, 5 phr, and 7 phr LDPE foams are around 10 �m and are much smaller than that of four neat PS foams. The results indicate that PS blended with LDPE can lead to the reduction of average cell size of PS foam during PS foaming.

FIGURE 8.13 Cells in PS foams at the foaming temperature of 135°C: (a) PS-1; (b)PS-2;

(c) PS-3; and (d) PS-4.


61259_C008.indd 26761259_C008.indd 267 10/25/2008 12:39:58 PM10/25/2008 12:39:58 PM

PS

40

60

80

Cel

l siz

es (

mm)

20

0

Materials

PS-1 PS-2 PS-3 PS-4

125°C135°C

FIGURE 8.15 Average cell sizes of four PS foams at foaming temperatures of 125°C

and 135°C.

FIGURE 8.14 Cells in PS/2 phr CaCO3 foams at the foaming temperature of 135°C:

(a) PS-1; (b) PS-2; (c) PS-3; and (d) PS-4.

268 Polymeric Foams

61259_C008.indd 26861259_C008.indd 268 10/25/2008 12:39:58 PM10/25/2008 12:39:58 PM

PS/2 phr CaCO3

40

30

60

50C

ell s

izes

(mm

)

20

10

0

Materials

PS-1/CaCO3 PS-2/CaCO3 PS-3/CaCO3 PS-4/CaCO3

125°C135°C

FIGURE 8.16 Average cell sizes of four PS/2 phr CaCO3 at foaming temperatures of 125°C

and 135°C.

FIGURE 8.18 Dependence of PS/5 phr LDPE foam density on foaming temperatures.

PS/5 phr PE

100

200

300

400

500

600

Den

sity

(kg

/m3 )

0100 110 120

Temperature (°C)

130

PS-1PS-2PS-3PS-4

140 150


PS/2 phr PE

100

150

200

250

300

350

400

Den

sity

(kg

/m3 )

50

0100 110 120

Temperature (°C)

130

PS-1PS-2PS-3PS-4

140 150


61259_C008.indd 26961259_C008.indd 269 10/25/2008 12:39:58 PM10/25/2008 12:39:58 PM


PS/7 phr PE

100

200

300

400

500

Den

sity

(kg

/m3 )

0100 110 120

Temperature (°C)

130

PS-1PS-2PS-3PS-4

140 150

FIGURE 8.20 Cells in PS/2 phr LDPE foams at the foaming temperature of 125°C: (a) PS-1;

(b) PS-2; (c) PS-3; and (d) PS-4.

270 Polymeric Foams

61259_C008.indd 27061259_C008.indd 270 10/25/2008 12:39:59 PM10/25/2008 12:39:59 PM

FIGURE 8.22 Average cell sizes of four neat PS and PS/LDPE foams at the foaming

temperature of 125°C.

125°C foam

PS

PS-4PS-3PS-2Material

PS-10

10

20

30

Cel

l siz

es (

mm)

40

50

60

70

PS/2 phr LDPEPS/5 phr LDPEPS/7 phr LDPE

FIGURE 8.21 Cells in PS/7 phr LDPE foams at the foaming temperature of 125°C: (a) PS-1;

(b) PS-2; (c) PS-3; and (d) PS-4.


61259_C008.indd 27161259_C008.indd 271 10/25/2008 12:39:59 PM10/25/2008 12:39:59 PM

8.4 Heat Transfer in Plastic Foams

Plastic foams have a lower thermal conductivity than neat plastics. Plastic foams are used in industrial or daily goods are for their heat insulation. Therefore, the insulation property is an important characteristics of plastic foams. Heat transfer in plastic foams has been widely studied.

The total thermal conductivity, l, of plastic foams is the sum of the thermal conductivity contributed from solid, λs, the thermal conductivity contributed from gas, λg, the thermal conductivity contributed from radiations, λr, and the thermal conductivity contributed from convection, λc. The formula equation can be written as

λ � λs � λg � λr � λc (8.1)

The thermal conductivity through solids depends on the cell structure. However, it is diffi cult to describe the exact cell structure of the plastic foam because the cell structure is diverse from cell to cell. The cell structure is not spherically shaped, but the pentagonal dodecahedron, as depicted in Figure 8.23, is generally the most common shape for cells. The cells consist of cell walls formed by thin polymeric membranes and struts formed by the impingement of several walls at the inter secting position.48

Figure 8.24 shows the cell structure of a PS foam with foam density around 50 kg/m3 . The thickness of cell walls is less than 10 �m, but struts are much thicker. Based on a plastic foam with isotropic structure, References 49 and 50 propose the equation for the solid conductivity (λs) as follows:

λs � 2 __ 3 �

fs __ 3 (1 � � )λo (8.2)

FIGURE 8.23 Schematic of dodecahedral structure for a foam cell.

272 Polymeric Foams

61259_C008.indd 27261259_C008.indd 272 10/25/2008 12:40:00 PM10/25/2008 12:40:00 PM

� � ro - ra - rf

__________ ro - rg (8.3)

and Reference 51 simplifi es Equation 8.2 to become:

λs � ( 2 __ 3 �

fs __ 3 ) ( �f

__ �o ) λo (8.4)

where fs is the mass fraction of the struts in the foam; � is the void fraction or porosity of the foam; λo is the solid conductivity of the bulk material; �f is the foam density; �o is the density of the bulk material; �a is the air density; �g is the gas density inside the cell.

If a plastic foam is a non-isotropic structure; that is, cells with different structures in parallel and perpendicular directions to heat fl ux, Equation 8.2 can be expressed as:

λs � ( 2 __ 3 � fs __

3 ) (1 � �)�λo (8.5)

where � is a non-isotropic factor.49

Typically fs and � can be deduced from scanning electron microscopy (SEM) pictures of foams and fs in plastic foams is around 0.6–0.9.51 Therefore, the thermal conductivity through solid for plastic foams relates the void fraction of the foam or foam density, the mass fraction of struts, and the solid conductivity of the bulk material. The void fraction of the foam or foam density is dominant factor for solid conductivity. When foam density increases or the void fraction of the foam decreases, the solid

FIGURE 8.24 Cell structure of PS foam.


61259_C008.indd 27361259_C008.indd 273 10/25/2008 12:40:00 PM10/25/2008 12:40:00 PM

conductivity rises. The solid conductivity is not a constant and decreases with the reduction of temperature. However, the change in the solid conductivity with temperature can be considered small.52

The thermal conductivity through gas depends on the trapped gas.50,53–55 The cells of a plastic foam are initially full of a foaming agent (gas) or a mixture of foaming agents. The composition of gas in the cells then changes as time elapses due to the infusion of air from the atmosphere and the effusion of the foaming agents. The process can lead to an overall increase in the thermal conductivity of the plastic foam, since air has a greater thermal conductivity, as indicated in Figure 8.25. This phenome-non is often referred as aging.55–59 Aging is faster for plastic foams under a high thermal gradient environment than under isothermal conditions.57,58

Reference 55 suggests the mass fl ow rate (Jm) of one gas species when it diffuses across the cell walls of the plastic foam is

Jm � Pe __________

dw(P2 � P1) (8.6)

Jm � De __________

dw(C2 � C1) (8.7)

where Pe and De are the permeability coeffi cient and diffusion coeffi cient; dw is the thickness of a cell wall; P2 and P1 are the high and low partial pressure imposed on the two surfaces, respectively; and P2 and P1 can be associated with the concentration, C2 and C1.

Therefore, a basic parameter governing the gas conductivity (λg) in the plastic foam is the gas transport properties; that is, the diffusion of the foaming agent as well as air.26,55,56,59

FIGURE 8.25 Thermal conductivity of gases.

274 Polymeric Foams

61259_C008.indd 27461259_C008.indd 274 10/25/2008 12:40:01 PM10/25/2008 12:40:01 PM

Besides the mass fl ow rates of gases in the plastic foam, the thermal conductivity through gas can also be interpreted as

λg � A � �g � Cv � V � (L � Lv)/(Lv � L) (8.8)

where A is a constant; �g is the gas density inside the cell; Cv is the specifi c heat of gas at a constant volume; V is the velocity of gas molecules; L is the mean free path of gas molecules—the average distance of a gas molecule traveling before hitting another gas molecule, and Lv is the distance between cavities (cell size in the direction of thermal conduction).17

The increase in the gas density, specifi c heat of gas, velocity of gas mole-cules, and cell size can result in the enhancement of gas conductivity in plastic foams.17 Thermal conductivity through gas also depends on tem-perature. It decreases with the reduction of temperature because the decrease in gas molecular motion leads to reduced transport of thermal energy.52,57,58

Reference 26 concludes that the plastic foam having a foaming agent with low diffusivity, a low volume expansion ratio, a high cell density (small cell size), and a large foam thickness can attenuate the decay of the foaming agent in the plastic foam and show better thermal insulation capability. Heat transfer through gas in the plastic foams accounts for approximate 40–50% of the overall heat transfer.49,56,57 Moreover, Reference 16 estimates heat transfer through gas around 70–80% of the total heat transfer.

Eventually the thermal conduction for plastic foams can occur through bulk material itself and through gas. The volume of bulk material present in the low-density foam is of the order of 2–3% so that gases act as the main conducting medium.60

The thermal conductivity through radiation can be expressed as:

λr � 16�T 3/3Kr (8.9)

where � is Stefan–Boltzmann constant (5.669 � 10�8 W/m2 k4); T is the abso-lute temperature of the local material and Kr is the extinction coeffi cient.

For radiant energy traversing a material in a given solid angle with wavelengths, the decrease in intensity of radiant energy can be associated with the extinction coeffi cient. When radiant energy interacts with the plastic foam, part of the radiant energy is scattered by struts, part is refl ected by cell walls, part is absorbed by struts as well as cell walls, and part is transmitted by struts and cell walls.26 The extinction coeffi cient includes the absorption coeffi cient and the scattering coeffi cient. Therefore, radiation heat transfer is directly affected by the extinction coeffi cient of the plastic foam.

Reference 61 considers that the struts are thick enough to be opaque; the strut cross-section is constant and occupies two-thirds of the area


61259_C008.indd 27561259_C008.indd 275 10/25/2008 12:40:01 PM10/25/2008 12:40:01 PM

of an equilateral triangle (�ABC) formed at the vertices depicted in Figure 8.23. The triangle cross-sections of struts are converted into circular cross-sections with the same area indicated in Figure 8.23 and strut diam-eter (fs) is larger than wall thickness, or at most equal.51,62 The assumption is used in theoretical study for the extinction coeffi cient. The extinction coeffi cient in the plastic foam is directly proportional to the foam density and inversely proportional to the cell diameter (cell size).50,61–63 References 64 and 65 use more complicated methods to describe strut junctures and the struts used in the model to study the extinction coeffi cient. Basically the struts are not constant dimensions in the cross-section.

The solid in the plastic foam contains approximately 10–20% cell walls and 80–90% struts.66,67 Struts strongly infl uence the absorption and scat-tering properties of the plastic foam. The extinction coeffi cient of the plastic foam can be experimentally determined using a Fourier transform infrared spectrometer (FTIR).49,61,62,64,67

Radiative conductivity (λr) for the plastic foam can decrease by reducing cell diameter or by increasing foam density and strut diameter for a given mean cell diameter.17,50,52,61–63,66,68 The effect on the reduction of radiative conductivity occurs because struts become more opaque to radiation. Radiation for a plastic foam accounts for 2.5–34% of the overall heat trans-fer. This fraction of radiation may be decreased through reducing the cell size in the foaming process, lowering transmitted radiant energy and re-emitted radiant energy from the absorbed energy by additives or pig-ments in the formulation, as well as using a medium making boundary surfaces with low emissivity values.53

When the cell sizes in plastic foams are less than 4 mm, the effect of convective heat transfer is negligible.69 The closed-cell structure in plastic foams consists of struts and intact cell walls and cells are fi lled with gases. Therefore, the total thermal conductivity for closed-cell plastic foams can be expressed as

λ � λs � λg � λr (8.10)

Figure 8.26 presents the theoretical and experimental results for the thermal conductivity of PS foam boards blown with HFC-134a. The total thermal conductivity in theoretical study is based on the sum of the ther-mal conductivity through gas (λg) and the thermal conductivity through radiation (λr). The theoretical results for long-term thermal insulation per-formance of PS foam boards are shown in Figures 8.27, 8.28, and 8.29. Figure 8.27 indicates the total thermal conductivity infl uenced by blowing agent type. Foams produced by both HFC-134a and HCFC-142b have good long-term thermal insulation properties, but foam made by HFC-152a does not have. Figure 8.28 is the effect of cell density on the total thermal conductivity. A high cell density in the foam can lead to a low total ther-mal conductivity of the foam. Figure 8.29 displays the dependence of the

276 Polymeric Foams

61259_C008.indd 27661259_C008.indd 276 10/25/2008 12:40:01 PM10/25/2008 12:40:01 PM

0 2 4 6 8 100

10

20

30

The

rmal

con

duct

ivity

(m

W/m

/K)

Time (year)

ExperimentalNumerical

FIGURE 8.26 Relationship between thermal conductivity and time. [From Zhu, Z., Zong,

J., and Park, C.B. In Proceedings of the Society of Plastics Engineers Annual Technical Conference (ANTEC) 63 (2007): 1494–1498. With permission.]

0 10 20 300

10

20

30

40

Tot

al th

erm

al c

ondu

ctiv

ity(m

W/m

/K)

Time (year)

HFC 152aHFC 134aHCFC 142b

FIGURE 8.27 Effect of blowing agent type on the thermal conductivity. [From Zhu, Z.,

Zong, J., and Park, C.B. In Proceedings of the Society of Plastics Engineers Annual Technical Conference (ANTEC) 63 (2007): 1494–1498. With permission.]

0 10 20 30

20

30

40

Tot

al th

erm

al c

ondu

ctiv

ity(m

W/m

/K)

Time (year)

Cell density = 5.75 × 105#/ccCell density = 2.50 × 106#/ccCell density = 1.00 × 107#/cc

FIGURE 8.28 Effect of cell density on thermal conductivity. [From Zhu, Z., Zong, J., and

Park, C.B. In Proceedings of the Society of Plastics Engineers Annual Technical Conference (ANTEC) 63 (2007): 1494–1498. With permission.]


61259_C008.indd 27761259_C008.indd 277 10/25/2008 12:40:01 PM10/25/2008 12:40:01 PM

total thermal conductivity on volume expansion. A low total thermal con-ductivity can be obtained by a low volume expansion ratio for the foam.

The open-cell structure in plastic foams mainly comprises struts and a few cell walls. The total thermal conductivity for open-cell plastic foam is the same as Equation 8.10. However, when a porous material is used as a core material in VIPs, the thermal conductivity of the VIP decreases with the reduction of the internal pressure described in Figure 8.30. Three types of core material, open-cell PS foam, open-cell PU foam, and Nanogel (branded aerogel), are used in VIPs. The pore sizes of open-cell PS foam and open-cell

0 10 20 30

20

30

40

Tot

al th

erm

al c

ondu

ctiv

ity(m

W/m

/K)

Time (year)

VER = 40VER = 35VER = 30

FIGURE 8.29 Effect of volume expansion ratio on the thermal conductivity. [From Zhu, Z.,

Zong, J., and Park, C.B. In Proceedings of the Society of Plastics Engineers Annual Technical Conference (ANTEC) 63 (2007): 1494–1498. With permission.]

0.025

0.020

Open-cell psOpen-cell puNanogel

0.015

0.010K (

W/m

k)

0.005

00.01 0.1 1 10

Pressure (Pa)

100 1000 10000

FIGURE 8.30 Thermal conductivity of VIPs at various internal pressures.

278 Polymeric Foams

61259_C008.indd 27861259_C008.indd 278 10/25/2008 12:40:02 PM10/25/2008 12:40:02 PM

PU foam are microscale but Nanogel has nanoscale pores. The data of thermal conductivity for Nanogel are obtained from Reference 70.

The decrease in the internal pressure for the three VIPs to a certain level can lead to an approximate constant for thermal conductivity of the three VIPs. The VIP using a Nanogel as a core material has much lower thermal conductivity than that using an open-cell PS foam and open-cell PU foam at the same internal pressure. The VIP with an open-cell PS foam as a core material shows a similar thermal conductivity to that of an open-cell PU foam at an internal pressure of less than 13.3 Pa (0.1 torr). However, the VIP having an open-cell PU foam as a core material exhibits a higher thermal conductivity than that having an open-cell PS foam at a higher internal pressure.

As a result, the thermal conductivity of VIP is performed by thermal conductivity through solid and radiation and is no longer affected by thermal conductivity through gas. However, when the foam density of plastic foams increases at a constant temperature, the solid conductivity increases; radiative conductivity decreases; and gas conductivity is con-stant. An optimal foam density exists to have the lowest total thermal conductivity for the plastic foams with either open-cell or closed-cell structure.62 Pore sizes reaching nanoscale in the core material greatly help the thermal performance of VIP.

Therefore, the requirements for an open-cell plastic foam as a core material used in VIPs include

Plastics having low thermal conductivity•

High content of open cells (more than 95% open cells)•

Small cell size (less than 100 • �m)

Light weight and good mechanical properties to resist external • pressure from the atmosphere.

8.5 Thermal Conductivity of VIPs Using PS

Foams as Core Materials

The thermal conductivity of VIPs using different porous PS foams as core materials is presented in Figure 8.31. PS with fi llers such as CaCO3 and PS with LDPE and fi llers are foamed by a gas mixture of CO2 and fl uoro-carbon at a constant pressure. When porous PS foams used in VIPs are made at a low foaming temperature, for instance 123°C, VIPs exhibit a high thermal conductivity. VIPs using porous PS foams produced at the foaming temperature of 127°C have the lowest thermal conductivity of the three foaming temperatures, and at the foaming temperature of 125°C VIPs exhibit an in-between thermal conductivity.


61259_C008.indd 27961259_C008.indd 279 10/25/2008 12:40:02 PM10/25/2008 12:40:02 PM

Normally, a high thermal conductivity of VIPs indicates that the porous foam has a low content of open cells and vice versa. As the content of LDPE in the porous PS foam increases, the thermal conductivity of VIPs using porous foams made at the foaming temperature of 123°C shows an initially dramatic decrease and then levels off; however, at the foaming temperatures of 125°C and 127°C do not have that trend. The results indi-cate that the content of open cells in porous PS foams made at the foaming temperature of 123°C is more infl uenced by LDPE than that made at the foaming temperatures of 125°C and 127°C.

Figure 8.32 illustrates the thermal conductivity of VIPs using different porous foams as core materials manufactured under a gas mixture of CO2 and fl uorocarbon, three foaming pressures, and a constant foaming tem-perature of 125°C. VIP foam using PS with fi llers made at the foaming pressure of 6.88 MPa (1000 psi) has the highest thermal conductivity while the foaming pressure of 8.95 MPa (1300 psi) displays the lowest thermal conductivity. VIPs have an intermediate thermal conductivity when PS foam with fi llers is produced at the foaming pressure of 11.02 MPa (1600 psi). As the content of LDPE in porous PS foams increases, the thermal conductivity of VIPs decreases fi rst and then increases gradually for the three foaming pressures. However, the lowest thermal conductivity of VIPs is always observed at the foaming pressure of 8.95 MPa (1300 psi).

Therefore, an optimal foaming pressure exists to manufacture porous PS foam with a high content of open cells. Because PS foam with fi llers or PS/LDPE foam with fi llers are produced at the foaming pressure of 11.02 MPa (1600 psi), large-cell sizes scatter in porous PS foams and porous PS foams are slightly distorted. When they are foamed at the foaming pressure of 6.88 MPa (1000 psi), the content of open cells is lower than that

FIGURE 8.31 Thermal conductivity of VIPs using PS/LDPE foams with fi llers as core

materials at various foaming temperatures and CO2/fl uorocarbon as a foaming agent.

0.030

0

0.005

0.010

0.015

K (

W/m

k) 0.020

0.025

0 2 4 6

LDPE (%)

8 10 12

VIP (0.01 torr)

T: 123°C

T: 125°C

T: 127°C

280 Polymeric Foams

61259_C008.indd 28061259_C008.indd 280 10/25/2008 12:40:02 PM10/25/2008 12:40:02 PM

in porous PS foams made at the foaming pressure of 8.95 MPa (1300 psi) and average cell sizes are larger. How to balance the cell size, the content of open cells, and the mechanical strength for porous PS foams is an important issue. These factors can infl uence the thermal conductivity of PS foams.

Figure 8.33 shows the thermal conductivity of VIPs using different porous PS foams as core materials produced at four foaming tempera-tures of 121°C, 123°C, 125°C, and 127°C and a constant pressure of 8.95 MPa (1300 psi). The foaming agent is a gas mixture of CO2 and nitrogen (N2). The trend of thermal conductivity for VIPs using porous PS foams made by a gas mixture of CO2 and N2 is similar to that for VIPs using porous PS foams made by a gas mixture of CO2 and fl uorocarbon. However, porous PS foams made by a gas mixture of CO2 and N2 have a higher ther-mal conductivity and a lower content of open cells than those made by a gas mixture of CO2 and fl uorocarbon at the same temperature, as indi-cated in Figure 8.31. Generally speaking, porous PS foams with 7% LDPE and fi llers made by a gas mixture of CO2 and N2 generate a better result in thermal conductivity such as a narrow difference of thermal conductivity and low thermal conductivity for the four foaming temperatures than other PS foams with LDPE and fi llers.

The thermal conductivity of VIPs using different porous foams as core materials produced under a gas mixture of CO2 and N2 as a foaming agent, three foaming pressures, and a constant foaming temperature of 125°C is shown in Figure 8.34. The sequence from high to low thermal conductivity for PS foams with fi llers is the foam made at the foaming pressure of 11.02 MPa (1600 psi), then at the foaming pressure of 8.95 MPa (1300 psi), then at the foaming pressure of 6.88 MPa (1000 psi). The thermal conductivity for


materials at various foaming pressures and CO2/fl uorocarbon as a foaming agent.

0

0.005

0.010

0.015

K (

W/m

k)

0.020

0.025

0 2 4 6

LDPE (%)

8 10 12

VIP (0.01 torr)

P: 6.88 MPaP: 8.95 MPaP: 11.02 MPa


61259_C008.indd 28161259_C008.indd 281 10/25/2008 12:40:03 PM10/25/2008 12:40:03 PM

PS foams with fi llers produced by the gas mixture of CO2 and N2 is higher than that by the gas mixture of CO2 and fl uorocarbon indicated in Figure 8.32, and a different sequence of thermal conductivity is also observed due to different foaming agents. The thermal conductivity of VIPs for PS foams with 2% LDPE and fi llers at three foaming pressures is very close and is much lower than that of VIPs for PS foams with fi llers at three foaming pres-sures. VIPs for PS foams with 5% LDPE and fi llers made at the foaming


materials at various foaming temperatures and CO2/N2 as a foaming agent.

0

0.005

0.010

0.015K (

W/m

k)

0.020

0.025

0.030

0.035

0 2 4 6LDPE (%)

8 10 12

VIP (0.01 torr)

T: 121°CT: 123°CT: 125°CT: 127°C


materials at various foaming pressures and CO2/N2 as a foaming agent.

0

0.005

0.010

0.015

K (

W/m

k)

0.020

0.025

0 2 4 6

LDPE (%)

8 10 12

VIP (0.01 torr)

P: 6.88 MPaP: 8.95 MPaP: 11.02 MPa

282 Polymeric Foams

61259_C008.indd 28261259_C008.indd 282 10/25/2008 12:40:03 PM10/25/2008 12:40:03 PM

pressure of 8.95 MPa (1300 psi) show the lowest thermal conductivity in the study. However, the increase of LDPE in PS can lead to the enhancement of thermal conductivity for PS foams with LDPE and fi llers.

Figure 8.35a–d represents the cell structures of PS foams for VIPs having different thermal conductivity of 0.027 W/m k, 0.02 W/m k, 0.01 W/m k, and 0.0065 W/m k, respectively. The content of open cells is less than 50% in the PS foam having the thermal conductivity of 0.027 W/m k; in other words, this foam can be considered a closed-cell foam. When ther-mal conductivity decreases, the content of open-cells increases in the PS foam; for example, the PS foam with the thermal conductivity of 0.02 W/m k and approximately 70% open cells in the PS foam, but many closed-cells are still observed in the PS foam. As the thermal conductivity of the PS foam reaches 0.01 W/m k and the content of open cells in the PS foam is approximately 90%, many open cells exist and intact cell structures, such as pentagonal and hexagonal structures, gradually disappear. The thermal conductivity of the PS foam is 0.0065 W/m k. It is diffi cult for the PS foam to observe an intact cell structure. The PS foam contains approximately

FIGURE 8.35 Cell structures of PS/LDPE foams with fi llers as core materials for

VIPs having thermal conductivity: (a) 0.027 W/m k; (b) 0.02 W/m k; (c) 0.01 W/m k; and

(d) 0.0065 W/m k.


61259_C008.indd 28361259_C008.indd 283 10/25/2008 12:40:03 PM10/25/2008 12:40:03 PM

98% open cells. In other words, this is a PS foam with very high content of open cells. The open-cell content of rigid cellular plastics can be measured and analyzed with an air pycnometer using a standard test method.71

The relationship between thermal conductivity of three VIPs having PS foams as core materials and time is described in Figure 8.36. A getter is used within VIP(0) but not in the other two VIPs, VIP(1) and VIP(2). A low thermal conductivity at the beginning stage is observed for each VIP; the thermal conductivity of each VIP then gradually increases and then levels off.

A similar thermal conductivity for VIP(0) and VIP(1) is obtained ini-tially, but VIP(2) shows a higher thermal conductivity under the same internal pressure. Generally, a lower thermal conductivity for VIP indi-cates a higher content of open cells in the PS foam as described in Figure 8.35. Therefore, more closed-cells represent more gases trapped in the PS foam. When the PS foam is placed in a low-pressure environment, gases are released from the PS foam. The released gases can dramatically change the internal pressure and thermal conductivity of VIP. It takes approxi-mately 120 days for VIP(0), 240 days for VIP(1), and 360 days for VIP(2) to achieve a stable thermal conductivity. When a getter is placed in VIP(0), the stable thermal conductivity for VIP(0) is similar to the initial thermal conductivity. However, the stable thermal conductivity for VIP(2) is much higher than the initial thermal conductivity. VIP(0) and VIP(1) apparently have a lower gas release since they have lower initial conductivities and change very little over time compared with VIP(2).The results also indicate that the PS foam with a high content of open cells and a good mechanical strength is a good core material for a VIP.

0.023Getter in VIP (0)

No getter in VIP (1)No getter in VIP (2)

0.021

0.019

0.017

0.015

0.013

K (

W/m

k)

0.011

0.009

0.007

0.0050 50 100 150 250200

Time (days)

350300 450400 500

FIGURE 8.36 Relationship between thermal conductivity and time.

284 Polymeric Foams

61259_C008.indd 28461259_C008.indd 284 10/25/2008 12:40:03 PM10/25/2008 12:40:03 PM

8.6 Conclusions

Neat PS and PS with CaCO3 or PS with LDPE are foamed by CO2 as a foaming agent under supercritical conditions. The low foam density for neat PS occurs at foaming temperatures of between 130°C and 140°C. When PS blended with CaCO3 or LDPE are foamed, the occurrence of low foam density for PS/CaCO3 or PS/LDPE foams shifts to low foaming tempera-tures. Low foam density and the foaming temperatures which generate low foam density for PS/CaCO3 or PS/LDPE foams depend on the amounts of CaCO3 or LDPE used. Average cell sizes and cell structures such as open-cell or closed-cell for PS/CaCO3 or PS/LDPE foams are also infl u-enced by the content of CaCO3 or LDPE. In comparison with the average cell sizes of neat PS foams, CaCO3 or LDPE in PS foams can lower the aver-age cell sizes of PS/CaCO3 or PS/LDPE foams.

VIPs achieve a lower thermal conductivity when PS/LDPE foams with fi llers as core materials are made by both foaming agents, a mixture of CO2/fl uorocarbon and CO2/N2 at a higher foaming temperature and a constant foaming pressure of 8.95 MPa (1300 psi). However, PS/LDPE foams with fi llers are produced at various foaming pressures, 6.88 MPa (1000 psi), 8.95 MPa (1300 psi), and 11.02 MPa (1600 psi), and a constant foaming temperature of 125°C. An optimal foaming pressure exists to manufacture PS/LDPE foams with fi llers having a high content of open cells. Generally, a porous PS foam with a higher content of open-cells as a core material used in the VIP can result in a lower thermal conductivity and a more stable thermal conductivity during its lifetime.

8.7 Abbreviations

Al AluminumCaCO3 Calcium carbonateCFA Chemical foaming agentCO2 Carbon dioxideDSC Differential scanning calorimeterEtCl Ethyl chlorideEO Ethylene/octane copolymerESI Ethylene/styrene interpolymerEVOH Ethylene vinyl alcoholFTIR Fourier transform infrared spectrometerGPPS General-purpose polystyreneHCFC-22 Chlorodifl uoro methaneHCFC-142b 1-chloro-1,1-difl uoro ethane


61259_C008.indd 28561259_C008.indd 285 10/25/2008 12:40:04 PM10/25/2008 12:40:04 PM

HDT Heat distortion temperatureHDPE High-density polyethyleneHFC-134a 1,1,1,2-tetrafl uoroethaneHFC-152a 1,1-difl uoroethaneHIPS High-impact polystyreneLDPE Low density polyethyleneMw Weight average molecular weightN2 NitrogenNy NylonPET Polyethylene terephthalatePFA Physical foaming agentPS PolystyrenePU PolyurethaneSEBS Styrene/ethylene butylenes/styreneSEM Scanning electron microscopysPS Syndiotactic polystyreneVIP Vacuum insulation panel

8.8 Nomenclature

� Non-isotropic factor� Void fraction or porosity of the foamλ Total thermal conductivityλc Thermal conductivity contributed from convectionλg Thermal conductivity contributed from gasλo Solid conductivity of the bulk materialλr Thermal conductivity contributed from radiationsλs Thermal conductivity contributed from solid�a Air density�f Foam density�g Gas density inside the cell�o Density of the bulk material� Stefan-Boltzmann constant (5.669 × 10�8 W/m2 k4)A ConstantC1 Low partial concentrationC2 High partial concentrationCv Specifi c heat of gas at a constant volumeDe Diffusion coeffi cientdw Thickness of a cell wallfs Mass fraction of the struts in the foamJm Mass fl ow rateKr Extinction coeffi cient

286 Polymeric Foams

61259_C008.indd 28661259_C008.indd 286 10/25/2008 12:40:04 PM10/25/2008 12:40:04 PM

L Mean free path of gas molecules—the average distance of a gas molecule traveling before hitting another gas molecule

Lv Distance between cavities (cell size in the direction of thermal conduction)

P1 Low partial pressureP2 High partial pressurePe Permeability coeffi cientT Absolute temperature of the local materialTg Glass transition temperatureV Velocity of gas molecules

References

1. Kullberg, R. C., Manini, P., Corberi, F., and Rizzi, E. “Getters and desiccants for lifetime and performance maintenance in vacuum insulation panels.” In Proceedings of the International SAMPE Symposium and Exhibition 46 (2001): 1239–1250.

2. Lanzani, F. and Mauri, R. European Patent 0849309A1(1998). 3. Imeokparia, D. D., Shmidt, C. D. and Suh, K. W. US Patent 5,693,687, 1997. 4. Shmidt, C. D., Imeokparia, D. D., Suh, K. W., Malone, B. A., and Franklin, K.

“Extruded Open-Cell microcellular foams, and their preparation process.” WO 96/34038, 1996.

5. Shmidt, C. D., Imeokparia, D. D., Suh, K. W., Malone, B. A., and Franklin, K. US Patent 5,780,521, 1998.

6. Shmidt, C. D., Imeokparia, D. D., Suh, K. W., Malone, B. A., and Franklin, K. US Patent 5,863,960, 1999.

7. Shmidt, C. D., Suh, K. W., Imeokparia, D. D., Malone, B. A., and Franklin, K., US Patent 5,869,544, 1999.

8. Bland, D. G., Stobby, W. G., Mork, S. W., Staples, T. G., and McCann, G. D. “Absorbent, extruded thermoplastic foams.” WO 98/56430, 1998.

9. Park, C. P., Imeokparia, D. D., and Chaudhary, B. I. “Open-Cell foam and method of making.” WO 99/47592, 1999.

10. Park, C. P., Chaudhary, B. I., and Imeokparia, D. D. US Patent 6,093,752, 2000.

11. Park, C. P., Chaudhary, B. I., and Imeokparia, D. D. US Patent 6,174,471, 2001. 12. Suh, K. W., Park, C. P., Maurer, M. I., Tusim, M. H., De Genova, R., Broos, R.,

and Sophiea, D. P. “Lightweight cellular plastics.” Advanced Materials 12 (2000): 1779–1789.

13. Wong, C. M., Liang, W. C., Ying, C. H., and Hung, M. L. “Porous polystyrene foam produced at supercritical conditions.” Journal of Cellular Plastics 40 (2004): 421–429.

14. Wong, C. M., Tsai, S. J., Ying, C. H., and Hung, M. L. “Effect of low density polyethylene on polystyrene foam.” Journal of Cellular Plastics 42 (2006): 153–163.


61259_C008.indd 28761259_C008.indd 287 10/25/2008 12:40:04 PM10/25/2008 12:40:04 PM

15. Tao, W. H., Sung, W. F., and Lin, J. Y. “Development of vacuum insulation panel system.” Journal of Cellular Plastics 33 (1997): 545–555.

16. Wu, J. W., Sung, W. F., and Chu, H. S. “Thermal conductivity of polyurethane foams.” International Journal of Heat and Mass Transfer 42 (1999): 2211–2217.

17. Kodama, K., Yuge, K., Masuda, Y., and Tanimoto, Y. “Development of micro cellular open cell rigid polyurethane foams.” In Proceedings of Polyurethanes World Congress, October 10–13, (1993): 140–145.

18. Smith, D. M., Maskara, A., and Boes, U. “Aerogel-Based thermal insulation.” Journal of Non-Crystalline Solids 225 (1998): 254–259.

19. Rouanet, S., Floess, J., and Eberhardt, H. F. “Aerogels for optimized perfor-mance in vacuum insulation panels.” In Proceedings of the International Society for Advancement of Materials and Process Engineering Symposium and Exhibition 46 (2001): 1263–1270.

20. Wakili, K. G., Bundi, R., and Binder, B. “Effective thermal conductivity of vacuum insulation panels.” Building Research and Information 32 (2004): 293–299.

21. Schwab, H., Heinemann, U., Beck, A., Ebert, H. P., and Fricke, J. “Dependence of thermal conductivity on water content in vacuum insulation panels with fumed silica kernels.” Journal of Thermal Envelope and Building Science 28 (2005): 319–326.

22. Schwab, H., Heinemann, U., Beck, A., Ebert, H. P., and Fricke, J. “Prediction of service life for vacuum insulation panels with fumed silica kernel and foil cover.” Journal of Thermal Envelope and Building Science 28 (2005): 357–374.

23. Toyo Aluminum K.K. Brochure, Tokyo, Japan, 2002. 24. DuPont Teijin Films. Brochure, Hopewell, VA, 2002. 25. Jacobsen, S. “Barrier fi lms—ways to improve functionality.” Presented at

VIA 2002 Symposium, Atlanta, May 22–23, 2002. 26. Zhu, Z., Zong, J., and Park, C. B. In “Long-Term thermal insulation perfor-

mance of polystyrene foam boards with decay of blowing agent.” In Proceedings of the Society of Plastics Engineers Annual Technical Conference 63 (2007): 1494–1498.

27. Chang, W. R., Chang, C. H., Chung H. D., and Lin, J. Y. “Energy effi cient poten-tial for household refrigerators/freezers with high effi ciency compressor and vacuum insulation panels.” Presented at VIA 2002 Symposium, Atlanta, May 22–23, 2002.

28. Sosa, J. M. and Blackmon, K. P. Polymeric Materials Encyclopedia, Vol. 10, ed. T. C. Salamone. CRC Press, Boca Ration, 1996. p. 8032.

29. Traugott, T. D. Encyclopedia of Polymer Science and Engineering, Vol. 16, ed. J. I. Kroschwitz. John Wiley & Sons, New York, 1986. p. 63

30. Soderquist, M. E. and Dion, R. P. Encyclopedia of Polymer Science and Engineering, Vol. 16, ed. J. I. Kroschwitz. John Wiley & Sons, New York, 1986. p. 88.

31. Nanasawa, A. Polymeric Materials Encyclopedia, Vol. 9, T. C. Salamone, ed. CRC Press, Boca Raton, 1996. p. 6799.

32. Dow Plastics. “Restyling styrene.” European Plastics News 24(5) (1997): 24. 33. Dow Plastics. “Questra enters the Real World.” European Plastics News 26(11)

(1999): 30. 34. Lavengood, R. E. US Patent 4,214,056, 1980. 35. Martini, J., Waldman, F., and Suh, N. P. “The production and analysis of

microcellular thermoplastic foams.” In Proceedings of the Society of Plastics Engineers Annual Technical Conference 28 (1982): 674–676.

288 Polymeric Foams

61259_C008.indd 28861259_C008.indd 288 10/25/2008 12:40:04 PM10/25/2008 12:40:04 PM

36. Colton J. S. and Suh, N. P. “The nucleation of microcellular thermoplastic foam with additives part II: Experimental results and discussion.” Polymer Engineering and Science 27 (1987): 493–499.

37. Goel S. K. and Beckman, E. J. “Generation of microcellular polymeric foams using supercritical carbon dioxide I: Effects of pressure and temperature on nucleation.” Polymer Engineering and Science 34 (1994): 1137–1147.

38. Arora, K. A., Lesser, A. J., and McCarthy, T. J. “Preparation and characteri-zation of microcellular polystyrene foams processed in supercritical carbon dioxide.” Macromolecules 31 (1998): 4614–4620.

39. Stafford, C. M., Russell, T. P., and McCarthy T. J. “Expansion of polystyrene using supercritical carbon dioxide: Effects of molecular weight, polydisper-sity, and low molecular weight components.” Macromolecules 32 (1999): 7610–7616.

40. Park, C. B., Behravesh, A. H., and Venter, R. D. “Low-Density microcellular foam processing in extrusion using CO2.” Polymer Engineering and Science 38 (1998): 1812–1823.

41. Baldwin, D. F., Park, C. B., and Suh, N. P. “An extrusion system for the pro-cessing of microcellular polymer sheets: Shaping and cell growth control.” Polymer Engineering and Science 36 (1996): 1425–1435.

42. Han, X., Koelling, K. W., Tomasko, D. L., and Lee, L. J. “Continuous microcel-lular polystyrene foam extrusion with supercritical CO2.” Polymer Engineering and Science 42 (2002): 2094–2106.

43. Xu, X., Park, C. B., Xu, D., and Pop-iliev, R. “Effects of die geometry on cell nucleation of PS foams blown with CO2.” Polymer Engineering and Science 43 (2003): 1378–1390.

44. Rodeheaver, B. A. and Colton, J. S. “Open-celled microcellular thermoplastic foam.” Polymer Engineering and Science 41 (2001): 380–400.

45. Suetsugu, Y. and White, J. L. “The infl uence of particle size and surface coating calcium carbonate on the rheological properties of its suspension in molten polystyrene.” Journal of Applied Polymer Science 28 (1983): 1481–1501.

46. Chiu, F. C., Lai, S. M., Wong, C. M., and Chang, C. H. “Properties of calcium carbonate fi lled and unfi lled polystyrene foams prepared using supercritical carbon dioxide.” Journal of Applied Polymer Science 102 (2006): 2276–2284.

47. Lee, P. C., Li, G., Lee, J. W. S., and Park, C. B. “Improvement of cell opening by maintaining a high temperature difference in the surface and core of a foam extrudate.” Journal of Cellular Plastics 43 (2007): 431–444.

48. Klempner, D. and Frisch, K. C. Handbook of Polymeric Foams and Foam Technology. Hanser, New York, 1991.

49. Torpey, M. R. “A study of radiative heat transfer through foam insulation.” M.Sc. thesis, Department of mechanical engineering, massachusetts institute of technology, 1987.

50. Schuetz M. A. and Glicksman, L. R. Journal of Cellular Plastics 20 (1984): 114–121.

51. Kuhn, J., Ebert, H. P., Arduini-Schuster, M. C., Büttner, D., and Fricke, J. “Thermal transport in polystyrene and polyurethane foam insulations.” International Journal of Heat Mass Transfer 35 (1992): 1795–1801.

52. Bhattacharjee, D., King, J. A., and Whitehead, K. N. “Thermal Conductivity of PU/PIR foams as a function of mean temperature.” Journal of Cellular Plastics 27 (1991): 240–251.


61259_C008.indd 28961259_C008.indd 289 10/25/2008 12:40:04 PM10/25/2008 12:40:04 PM

53. Williams, R. J. J. and Aldao, C. M. “Thermal conductivity of plastic foams.” Polymer Engineering and Science 23 (1983): 293–298.

54. Prociak, A., Pielichowski, J., and Sterzynski, T. “Thermal diffusivity of rigid polyurethane foams blown with different hydrocarbons.” Polymer Testing 19 (2000): 705–712.

55. Ostrogorsky, A. G., Glicksman, L. R., and Reitz, D. W. “Aging of polyure-athane foams.” International Journal of Heat Mass Transfer 29 (1986): 1169–1176.

56. Albouy, A., Roux, J. D., Mouton, D., and Wu, J. “Development of HFC blowing agents. Part II: Expanded polystyrene insulating boards.” Cellular Polymers 17 (1998): 163–176.

57. Zarr, R. R. and Nguyen, T. J. “Effect of humidity and elevated temperature on the density and thermal conductivity of a rigid polyisocyanurate foam Co-Blown with CCl3 and CO2.” Thermal Insulation and Building Envelopes 17 (1994): 330–350.

58. Bomberg, M. J. “Predicting fi eld thermal performance of a modifi ed resol foam from laboratory data.” Thermal Insulation and Building Envelopes 17 (1993): 78–88.

59. Vo, C. V. and Paquet, A. N. “An evaluation of the thermal conductivity of extruded polystyrene foam blown with HFC-134a or HCFC-142b.” Journal of Cellular Plastics 40 (2004): 205–228.

60. Doherty, D. J., Hurd, R., and Lester, G. R. “The physical properties of rigid polyurethane foams.” Chemistry and Industry (1962): 1340–1356.

61. Glicksman L. R. and Torpey, M. R. “The infl uence of cell size and foam density on the thermal conductivity of foam insulation.” In Proceedings of Polyurethanes World Congress, September 29–October 2 (1987): 80–84.

62. Placido, E., Arduini-Schuster, M. C., and Kuhn, J. “Thermal properties predictive model for insulation foams.” Infrared Physics and Technology 46 (2005): 219–231.

63. Koo, M. S., Chung, K., and Youn, J. R. “Reaction injection molding of polyure-thane foam for improved thermal insulation.” 41 Polymer Engineering and Science 41 (2001): 1177–1186.

64. Baillis, D., Raynaud, M., and Sacadura, J. F. “Spectral radiative properties of Open-Cell foam insulation.” Journal of Thermophysics and Heat Transfer 13 (1999): 292–298.

65. Doermann, D. and Sacadura, J. F. “Heat transfer in Open-Cell foam insula-tion.” Journal of Heat Transfer 18 (1996): 88–93.

66. Reitz, D. W., Schuetz, M. A., and Glicksman, L. R. “A basic study of aging of foam insulation.” Journal of Cellular Plastics 20 (1984): 104–113.

67. Glicksman, L., Schuetz, M., and Sinofsky, M. “Radiation heat transfer in foam insulation.” International Journal of Heat Mass Transfer 30 (1987): 187–197.

68. Biedermann, A., Kudoke, C., Merten, A., Minogue, E., Rotermund, U., Ebert, H. P., Heinemann, U., and Fricke, J. “Analysis of heat transfer mechanisms in polyurethane rigid foam.” Journal of Cellular Plastics 37 (2001): 467–483.

69. Skochdopole, R. E. “The thermal conductivity of foamed plastics.” Chemical Engineering Progress 57 (1961): 55–59.

70. NanoPore Incorporated. Brochure, Albuquerque, NM, 2003. 71. American Society for the Testing of Materials. “Standard test method for

Open-Cell content of rigid cellular plastics by the air.” D2856, Volume 8(2) Plastics (II) (1998): 143–148.

290 Polymeric Foams

61259_C008.indd 29061259_C008.indd 290 10/25/2008 12:40:04 PM10/25/2008 12:40:04 PM

Author Index

Abes, J. I., 188Adachi, Y., 178, 181Adamovsky, S., 145Aguillar, M., 151Ahern, A., 25Al Ghatta, H. A. K., 9Albouy, A., 274, 275Aldao, C. M., 274, 276Alexandre, M., 157Amecke, B., 43, 48, 51Anderson, J. R., 116Anderson, K. L., 188Ano, Y. T., 155Arduini-Schuester, M. C., 273, 276, 279Areerat, S., 11, 12Arora, K. A., 261Ashby, M. F., 22, 215Ashford, P., 28, 31Ashida, K., 15Astarita, L., 9Awojulu, A., 155

Baillis, D., 276Baldwin, D. F., 90, 226, 261Barito, R. W., 14Bayer, O., 11Beck, A., 256Beckman, E. J., 206, 261Behravesh, A. H., 261Benning, C. J., 11

Berghaus, U., 57, 58Berglund, L. A., 161Bhattacharjee, D., 274, 275, 276Biedermann, A., 276Biesenberger, J. A., 22Bikiaris, D. N., 144Binder, B., 256Blackmon, K. P., 259Bland, D. G., 256, 261, 262Blander, M., 8Blasius, W. G., 155Boehringer Ingelheim, K. G., 58, 59Boes, U., 256Bomberg, M. J., 274, 275Bopp, R. C., 33Bourban, P.-E., 150Bousmina, M., 157Boyce, M. C., 188Boyce, S. T., 144Broos, R., 256Brown, J. M., 214Büttner, D., 273, 276Büttner, H., 63Bundi, R., 256

Calberg, C., 157Campbell, N. D., 152Cao, X., 163, 214Carbonell, R. G., 146Carlson, D., 155

61259_C009.indd 29161259_C009.indd 291 10/25/2008 12:40:30 PM10/25/2008 12:40:30 PM

Caron, L.-M., 71Cha, S. W., 116Champagne, M., 6, 150Chandra, A., 214Chang, C. H., 259, 263Chang, F.-C., 161Chang, J., 15Chang, W. R., 259Chang, Y., 146Chaudhary, B. I., 256, 261, 262Chen, C., 7Chen, C. Y., 152Chen, X., 155Chen, Y., 15Cheng, M. L., 7Chiu, F. C., 263Chixin, Z., 161Chmiel, H., 72Cho, H., 155Chu, C. C., 152Chu, H. S., 256Chung, C. I., 16Chung, H. D., 259Chung, K., 276Cohen, R. E., 188Collington, K. T., 233Collins, F. H., 16, 19, 46, 51Colton, J. S., 90, 113, 201, 204, 261, 262Columbo, R., 3Cooper, A., 221Corberi, F., 256Cordes, H., 214Coslanicha, A., 186Cotugno, S., 144, 146, 160Crank, J., 112, 226Curliss, D. B., 214

Daigneault, L. E., 71Day, M., 150De Genova, R., 256de Saja, J. A., 227Dealy, J. M., 152Dean, K., 146Deeter, G. A., 155Degée, P., 157Denault, J., 150Deng, X., 157Deshmukh, V. G., 152DeSimone, J. M., 144

Devalckenaere, M., 157Di Maio, E., 144, 145, 146, 148, 152,

155–161, 164Di, W., 145, 157, 158, 160Di, Y., 150, 155, 156, 159, 161Dietrich, C., 158Dion, R. P., 259Doelling, K. W., 19Doermann, D., 276Doherty, D. J., 275Dooley, J., 18D’Souza, N. A., 214Dubois, P., 155, 157Dufour, J., 71Dumoulin, M. M., 71

Eastman, W. O., 14Eaves, D. E., 222, 228, 233Eberhardt, H. F., 256Ebert, H. P., 256, 273, 276Eicker, D. B., 11Ema, Y., 150, 200

Fang, D., 15Farmer, B. L., 188Ferronea, M., 186Ferry, J. D., 76Fischer, M., 112Fletcher, N. H., 208Fleurent, H., 25Floess, J., 256Flory, P. J., 9, 111Flumerfelt, R. W., 146, 149, 226Franklin, K., 256, 261Franklin, W. E., 2Fratzl, P., 196Fricke, H., 46Fricke, J., 256, 273, 276Friedrich, C., 158Frisch, K.C., 2, 11, 3, 233, 272Fu, T. L., 214Fujimoto, Y., 145, 157, 176Fujiwara, K., 226Fukushima, Y., 157, 159, 176Funami, E., 162, 196

Gandhi, K., 152Gang, W., 161Gao, F., 176

292 Author Index

61259_C009.indd 29261259_C009.indd 292 10/25/2008 12:40:31 PM10/25/2008 12:40:31 PM

Garcia, R. A., 45Gaur, U., 145Gay, Y. J., 144Ge, J., 155Gendron, R., 6, 71, 146, 163Gerhartz, W., 46, 51, 62Giannelis, E. P., 158, 159, 177Gibson, L. J., 22, 215Glicksman, L. R., 25, 272, 274–276Goel, S. K., 206, 261Gong, S., 163, 214Gonzalez-Pena, J. I., 227Gopakumar, T. G., 161Gramann, P., 214Greeley, T., 155Griffi n, J. D., 12Gu, Z., 155Gundert, F., 75Guo, Z., 19Gupta, M., 18Gupta, M. C., 152

Habibi-Naini, S., 119, 134Hamel, A., 163Hampson, R. F., 5Hamza, R., 22Han, C. D., 71, 113Han, J. H., 113Han, X., 19, 163, 214, 261Hansen, R. H., 45, 46Hao, J., 157Harkonen, M., 155Harrison, K. L., 146Hasegawa, N., 163, 176, 188, 190, 193Heinemann, U., 256, 276Heinz, R., 78, 81, 82, 100, 101, 102Helminen, A., 155Henne, A. L., 3, 11Hensen, F., 58Herrmann, T., 79Hiltunen, K., 155Hiroi, R., 157, 177, 178, 186Hironaka, K., 157, 163, 176, 210Hoff, G. P., 11Hoffmann, B., 158Hopfenberg, H. B., 146Horn, B., 136Houston, J. C., 46, 48Howdle, S. M., 146

Huneault, M. A., 33Hung, M. L., 256Hurd, R., 275Hurnik, H., 51

Iannace, S., 144–146, 148–150, 152, 155–161, 164

Ikeya, M., 150, 200Ilto, Y., 157, 163Imeokparia, D. D., 256, 261, 262Iqbal, M., 152Ishii, H., 177Ito, Y., 176, 209, 210

Jacobs, P. M., 52Jacobsen, S., 256Jaeger, A., 116Jean, Y. C., 7Jenkins, M. J., 146Jérôme, R., 157Jimenez, G., 157Jinnai, K., 178, 181Jinno, F., 4Jun, J., 6

Kamigaito, O., 157, 159, 176Kannah, K., 32Karayannidis, G. P., 144Karbas, H., 163Kareko, L., 6Kato, M., 176Katz, J.L., 8Kawai, H., 157Kawasumi, M., 157, 159Kay, W. B., 107Kennedy, R. N., 16Kihara, S., 149Kim, M., 155Kim, S. G., 226Kimura, M., 178, 181King, J. A., 274, 275, 276Kirkland, C., 43, 51Klempner, D., 144, 233, 272Kodama, K., 256, 275, 276Koelling, K. W., 163, 261Kojima, J., 4Kojima, Y., 157, 159, 176Kontopoulou, M., 161Koo, M. S., 276

Author Index 293

61259_C009.indd 29361259_C009.indd 293 10/25/2008 12:40:31 PM10/25/2008 12:40:31 PM

Koppi, K. A., 18Kosin, J. A., 45Kotaka, T., 163, 188, 190, 193Kreglewski, A., 107Kretzschmann, G., 48, 50Krikorian, V., 157Krishnamoorti, R., 158, 188Kriz, D., 152Kropp, D., 71, 74, 75, 78, 157Kropp, D., 46Kudoke, C., 276Kuhn, J., 273, 276, 279Kullberg, R. C., 256Kumar, V., 213Kuppa, V., 188Kurauchi, T., 157, 159, 176Kurylo, M. J., 5Kylma, J., 155

Lagaly, G., 177, 179, 181Lai, S. M., 263Laing, D. D., 116Landel, R. F., 76Lando, J. B., 22Landrock, A. H., 43, 51Lanzani, F., 256, 261Lau, S.-F., 145Lauterberg, W., 82Lavengood, R. E., 260Leach, A. G., 25Lebedev, B., 145Lee, H.-Y., 161Lee, J., 7Lee, J. A., 161Lee, J. G., 226Lee, J. W. S., 226, 265Lee, L. J., 19, 163, 208, 214, 261Lee, M. H., 163Lee, P. C., 265Lee, S. T., xii, 1, 6, 19, 33, 62, 80, 144Lee, Y. H., 163Lepoittevin, B., 157Leppkes, R., 103, 106Lesser, A. J., 261Lester, G. R., 275Li, C. C., 107Li, G., 265Li, J., 161Li, W., 146, 149, 155

Liang, W. C., 256Liao, X., 150Lin, J. Y., 256, 259Listemann, M., 22Liu, X., 161Lober, F., 43, 51Lu, G. Q., 188Lübke, G., 48, 103, 104, 105

Ma, C. Y., 71Ma, P., 33Macosko, C. W., 22, 144, 163, 214Maiti, M., 176, 190Maiti, P., 157, 163, 170, 188, 190, 192, 193Malone, B. A., 256, 261Manias, E., 159Manias, E. J., 188Manini, P., 256Månson, J.-A. E., 150Mao, J., 15Margedant, J. A., 11Maron, S. H., 22Marrazzo, C., 148, 152, 155, 164Martelli, F., 3Martin, W. M., 45Martìnez-Salazar, J., 151Martini, J., 261Maskara, A., 256Masuda, Y., 256, 275, 276Masuoka, H., 226Mathieu, L. M., 150Matsumoto, T., 155Matsuoka, F., 155Maurer, M. I., 256Mauri, R., 256, 261McCann, G. D., 256, 261, 262McCarthy, T. J., 261McNary, R. R., 3, 11Meikle, J. L., 2Menakanit, S., 188Menges, G., 75, 108, 134Mensitieri, G., 144, 146, 149Merten, A., 276Messersmith, P. B., 159Michaeli, W., 71, 100, 102Midgley, T., 3, 11Minogue, E., 276Mitsunaga, M., 157, 163, 176, 210Miyamoto, M., 155

294 Author Index

61259_C009.indd 29461259_C009.indd 294 10/25/2008 12:40:31 PM10/25/2008 12:40:31 PM

Mohren, P., 136Montjovent, M.-O., 150Moore, S., 230Morita, K., 6Mork, S. W., 256, 261, 262Mours, M., 152Mouton, D., 274, 275Muelhaupt, R., 158Müller, E., 43Munters, G.,Muschiatti, L. C., 9

Naguib, H. E., 154Nakano, S., 178, 181Nakayama, T., 163Nam, J. Y., 209Nam, P. H., 163, 176, 188, 189, 190,

192, 193Nanasawa, A., 259Narayan, R., 28, 155Narayanan, N., 149Narkis, M., 151, 152, 153Nasman, J. H., 155Nawaby, A. V., 150Ndiaye, P.A., 2Ned Nisson, J. D., 26Nelson, P., 163Neumüller, O.-A., 42Nguyen, T. J., 274, 275Nicolais, L., 144–146, 148–150,

155–161Nicolay, A., 113Nie, L., 155Niemi, M., 155Niggemann, M., 76Nishimura, S., 178, 181

Ogami, A., 145, 157, 159, 176, 183, 204Ogata, N., 157Ogihara, T., 157Ohshima, M., 149, 162, 163, 196Okada, A., 157, 159, 176Okamoto, H., 163Okamoto, K., 190, 214Okamoto, K. T., 18, 133, 157Okamoto, M., 145, 150, 157–159, 162,

163, 176–178, 183, 186, 188, 190, 192, 193, 195, 196, 200, 209, 210, 214, 215

Ortner, L., 11Ostrogorsky, A. G., 274

Padareva, V., 48Pantoustier, N., 157Panzer, U., 154Paquet, A. N., 274Parent, J. S., 161Park, C. B., 19, 62, 90, 144, 154, 163, 226,

257, 261, 265, 274, 275Park, C. P., 256, 261, 262Park, E., 155Parks, D. M., 188Partridge, R., 245Pastor, D., 151Paul, M.-A., 157Peòn, J., 151Perlon, U., 11Pfannschmidt, L. O., 71, 108, 109, 111Phelan, R., 25Pielichowski, J., 274Pierick, D. E., 116Pioletti, D. P., 150Pittman, C. U., 152Placido, E., 276, 279Pluta, M., 159Pochan, D., 157Poling, B. E., 112Pop-iliev, R., 261Popov, N., 132Prausnitz, J. M., 112Prociak, A., 274Puri, R. R., 233

Ramesh, N. S., 19, 62, 144Rauwendaal, C., 16Ray, S. S., 145, 157, 163Raynaud, M., 276Reed, D., 20Reichelt, N., 154Reid, R. C., 112Reignier, J., 6, 146Reitz, D. W., 274, 276Ren, X., 146Rimura, Y., 155Rinke, H., 11Rizzi, E., 256Roberts, G. W., 146Rodeheaver, B. A., 262

Author Index 295

61259_C009.indd 29561259_C009.indd 295 10/25/2008 12:40:31 PM10/25/2008 12:40:31 PM

Rodriguez-Perez, M. A., 227Rogalla, A., 71Rotermund, U., 276Rouanet, S., 256Roux, J. D., 274, 275Roychoudhury, P. K., 149Royer, J. L., 144Rubens, L. C., 12Runt, J., 159Russell, T. P., 261Rutledge, G. C., 188

Sacadura, J. F., 276Sahnoune, A., 163Sain, M., 226Saito, T., 186Salovey, R., 152Sander, B., 112Sander, S. P., 5Sanguigno, L., 159Sasaki, T., 178, 181Sato, Y., 226Saunders, J. H., 3Schick, C., 145Schild, H., 11Schlack, P., 14Schmidt, R., 112Scholz, D., 43, 44, 48, 50, 51, 59, 63Schröder, T., 118Schümmer, P., 72Schuetz, M. A., 272, 274, 276Schulz, G., 46, 51, 62Schwab, H., 256Scott, R. M., 52, 55Seibt, S., 71Selin, J. F., 155Semerdjiev, S., 132, 135, 136Sendijarevic, V., 144Seppala, J., 155Severini, T., 9Shakesheff, K. M., 146Shen, J., 19, 163, 208, 214Sheng, N., 188Shenoy, A. V., 158Shikuma, H. E., 149Shinno, K., 155Shinoda, H., 6Shiroi, T., 157, 177, 178, 186

Shmidt, C. D., 256, 261Shutov, F. A., 103, 115Siefken, W., 11Silva, M. M. C. G., 146Sinha Ray, S., 157–159, 176–178, 183,

186, 204, 209, 210, 214Sinofsky, M., 276Sinsawat, A., 188Siripurapu, S., 144Skochdopole, R. E., 276Smith, D. M., 256Snyder, A. J., 159Sodergard, A., 155Soderquist, M. E., 259Sophiea, D. P., 256Sorrentino, L., 145Sosa, J. M., 259Spalding, M. A., 18Spiekermann, R., 51, 62, 103, 104Spitael, P., 144Srivastava, A., 149Stadlbauer, M., 77Stafford, C. M., 261Standish, R. K, 188Staples, T. G., 256, 261, 262Sterzynski, T., 274Stevens, R., 22Stevenson, J. F., 116Stobby, W. G., 256, 261, 262Strauss, W., 214Suetsugu, Y., 263Suh, K. W., 256, 261Suh, N. P., 20, 90, 113, 201, 204,

226, 261Sumarno, T. S., 226Sung, W. F., 256

Takada, M., 163, 202Takada, T., 4Takemura, K., 178, 181Taki, K., 149, 162, 196Tandberg, J. G., 11Tanimoto, Y., 256, 275, 276Tao, W. H., 256Tateyama, H., 178, 181Tatibouët, J., 146, 163Thomann, R., 158Throne, J. L., 62, 81

296 Author Index

61259_C009.indd 29661259_C009.indd 296 10/25/2008 12:40:31 PM10/25/2008 12:40:31 PM

Todd, D., 22Tomasko, D. L., 163, 261Topey, M. R., 272, 273, 275, 276Toth, R., 186Trageser, D. A., 233Traugott, T. D., 259Tsai, S. J., 256Tseng, C.-R., 161Tsunematsu, K., 178, 181Tuominen, J., 155Turco, G., 155Turng, L. S., 163, 214Tusim, M. H., 256

Uchiki, K., 6Ueda, K., 145, 155, 157, 159, 176,

183, 204Urchick, D., 12Usuki, A., 157, 159, 163, 176, 188,

190, 193

Vachon, V., 11Vaia, R. A., 177, 188, 214Vaillamizar, C. A., 71Van der Weide, I., 241Van Krevelen, D., 103, 106van Olphen, H., 190Vega, J., 151Venter, R. D., 261Verbist, G., 25Villalobos, M. A., 155Vo, C. V., 274Vollalobos, M. A., 155

Wakili, K. G., 256Walczak, K., 18Waldman, F., 261Wallerstein, R., 151, 152, 153Wang, J., 146Wang, K. H., 163Wason, S. K., 45, 63Watanabe, M., 178, 181Weaire, D., 25, 214Weller, J. E., 213Welty, J. R., 25Werth, M., 241Whelan, J., 33Whitaker, M. J., 146

White, J. L., 263Whitehead, K. N., 274, 275, 276Whitfi eld, P., 150Wicks, C. E., 25Widya, T., 163, 214Williams, M. L., 76Williams, R. J. J., 274, 276Wilson, R. E., 25, 72Wingert, M. J., 19Winter, H. H., 152Wirtz, H., 14Witten, N., 227, 239Wolf, B. A., 75Wong, C. M., 256, 263Wood-Adams, P. M., 152Wortberg, J., 73Worthoff, R. H., 72Wu, J., 274, 275Wu, J. W., 256Wu, J.-Y., 161Wu, Q., 161Wu, S., 112Wu, Z., 161Wübken, G., 136Wunderlich, B. B., 145

Xanthos, M., 144Xu, D., 261Xu, G., 214Xu, Q., 146Xu, R., 159Xu, X., 261

Yamada, K., 145, 157, 159, 183, 204, 214

Yamamoto, M., 186, 209Yanagimoto, T., 162, 196Yang, J., 7Yang, Y., 155Yano, K., 159Yevstropov, A., 145Ying, C. H., 256Yoon, J., 155Yoshida, O., 178, 183, 186Youn, J. R., 276Young, M. W., 144Yu, A. B., 188Yu, L., 146

Author Index 297

61259_C009.indd 29761259_C009.indd 297 10/25/2008 12:40:31 PM10/25/2008 12:40:31 PM

Yuan, M., 157, 163Yuge, K., 256, 275, 276

Zang, Y., 155Zarr, R. R., 274, 275Zeng, C., 163, 208, 214Zeng, Q. H., 188

Zhang, X., 22Zhong, W., 155Zhou, C., 161Zhu, N., 161Zhu, Z., 257, 274, 275Zhu, Zhengjin, 26Zong, J., 257, 274, 275

298 Author Index

61259_C009.indd 29861259_C009.indd 298 10/25/2008 12:40:31 PM10/25/2008 12:40:31 PM

Subject Index

A

Absorption, 2Accumulator, 51, 53Acid, 44

ascorbic, 44citric, 46citric esters, 44fumaric, 44gluconic, 44glutaric, 44lactic, 44, 149numeric, 46oxalic, 46succinic, 44tartaric, 46

Acidic salts, 44Acoustic, 27, 34, 238, 245Acrylonitrile, 259Acrylonitrile-butadiene-styrene

(ABS), 53, 54Adhesive, 257Aging, 274Air, 42, 45, 238, 274Alpha-methyl styrene, 259Alumina (Al2O3), 164–166, 168Aluminum (AL) foil, 256Aluminum mold, 50American National Standard

Institute (ANSI), 240

Ammonium cation, 198Amorphous, 12, 108, 109, 259Anti-thixotropy, 189Areospace, 239, 243Ascorbic acid, 44ASTM, 22

D-2872, 27, 35D-6400, 35D-6868, 35D-7021, 35E-84-01, 240E-90, 27E-413, 27E-492, 27E-989, 27

Autoclave, 219–221, 231, 234high pressure, 225low pressure, 227

Automotive, 4, D8462, 239, 242Azodicarbonamide (ADC), 30, 45,

51, 52, 55, 56, 57, 58, 63, 105, 230, 232

B

Barrel, 59grooved, 59, 60smooth, 60

Biodegradable, 36, 144, 145, 149

61259_C010.indd 29961259_C010.indd 299 10/25/2008 12:40:45 PM10/25/2008 12:40:45 PM

Biomedical, 145Bioplastics, 36Block process, 229Blow molding, 62Blowing Agents, 8, 19, 54, 55, 69, 70,

90, 109, 114–116, 118, 121, 109, 111, 113, 116, 123, 130, 169, 225, 277

chemical, 13, 30, 45, 49, 50, 51, 54, 55, 56, 230, 234

physical, 30, 101, 104, 105, 106, 230Boiling point, 47Boric acid, 45, 46Branching, 144, 151Breaker plate, 82breathing mold, 137, 138Brownian movement, 112Bubble formation, 79Bubble nucleation, 169, 198Bubble pressure, 7Bun, 229Butadiene, 259Butane (n-butane), 32, 471,4 butanediol (BD), 1551,4 butane diisocyanate (BDI),

155, 156

C

Cable insulation, 58Calcium, 46Calcium carbonate (CaCO3), 46,

263–266, 268, 269, 279, 285Calcium lactate, 44Capillary rheometer, 71, 75Carbon Dioxide, 6, 11, 12, 31, 32, 35, 37,

42, 45–47, 54, 69, 70, 76–80, 82, 90, 95, 105–107, 145–148, 159, 164, 167, 191, 196, 198, 200, 203, 204, 206, 208–210, 212, 260, 261, 274, 279–282, 285

solubiliity, 12Carbon Monoxide, 32Carbon nanofi bers, 208Carbon nanotubes (cn), 164–166Carboxylic acid, 44Carpet backing, 58Catalyst, 8, 10, 236Cation exchange capacity (CEC),

177, 178

Cavity wall, 132Cell

density, 93, 167, 168, 192, 203, 205, 211, 277

distribution, 13, 203formation, 200, 208morphology, 128nucleation, 201size, 22, 93, 131, 163, 168, 191, 203,

211, 268, 271, 279, 280structure, 54, 62, 129, 226, 273wall, 196, 201, 203, 211, 212,

275, 278Cellulose, 26Ceremic, 214Chemical blowing agents (CBAs),

(See Blowing agents)Chemical foaming agent

(CFA), 260Chemical nucleating agents, 31Chemical potential, 111Chlorofl uorocarbons (CFCs), 3, 5,

11, 32, 70CFC-11, 47, 48, 105CFC-12, 47, 48, 274CFC-113, 47CFC-114, 47

Chlorotrifl uoroethylene (CTFE), 240

Citric Acid, 13, 44, 46, 49, 83, 105Cleanroom insulation, 245Closed cell, 238, 243, 261, 276, 284

structure, 191, 209, 226, 279n-(coco alkyl)N, N-[bis(2-

hydroxyethyl)]-N-methyl ammonium, 179, 181

Co-extrusion, 61Co-injection, 53Color measurement, 242Compounding, 19Construction, 4, 238Convection, 272Cooling, 262Cooling time, 123–125Corn, 36Corn fi eld, 34Cosmetics application, 52Critical bubble radius, 7Critical pressure, 74

300 Subject Index

61259_C010.indd 30061259_C010.indd 300 10/25/2008 12:40:45 PM10/25/2008 12:40:45 PM

Cross-linked, 42high-density PE, 236low desnity PE, 223, 231Metallocene, 236PE, 42polyolefi n, 220, 229process, 19, 232, 238, 251

Cross-linking, 144, 151, 222, 224, 247Crystallization, 134, 145, 157, 161,

164, 165, 167Curing, 13Cushioning curve, 2, 237Cycle time, 54, 120, 123, 136, 139Cyclopentane, 47, 105

D

Degassing, 60Degradation, 3, 35Demolding, 124, 125Density, 107, 273

of CO2, 76reduction, 54, 103, 121, 122, 126

Depressurization, 226Design of experimentation (DOE),

123, 126Devolatilization, 22Dicumyl peroxide (DCP), 152–154Die, 45, 61, 70, 262

annular, 82coathanger, 61dual-spider, 81, 82fl at, 61gap, 91, 92in-line rheometer, 72, 78lip geometry, 91profi le, 61single-spider, 81, 82temperature, 86, 87, 89tubular, 61

Differential Scanning Calorimeter (DSC), 165, 260

Diffusion, 7, 46, 108, 112, 130coeffi cient, 112, 274

Diffusivity, 156, 168, 196, 198Dimer, 9dimethyl 2-ethylhexyl ammonium

cation (MMT-Alk), 157Dinitroso pentamethylene

tetramine (DNPT), 52

dioctadecyl dimethyl ammonium, 2C18(CH3)2N

+, 179, 182–186diphenyl methylene diisocyanate

(MDI), 10Direct gassing , 53Discoloration, 44, 52, 54, 57Disodium pyrophosphate, 45, 46Dynamic mechanical thermal

analysis (DMTA), 249, 250

E

Electron beam, 13Electronics, 239Electrostatic, 190Elongational viscosity, 193Emission, 32, 33EN 13432, 35Endothermic (Endo), 44, D14854,

55, 62, 63Energy absorption, 23Enexothermal (Enexo), 54, 55, 57EPS (See polystyrene)Equimolar, 49ErgoCell, 116, 117Ethanol, 6, 47Ethyl chloride (EtCl), 261Ethylene ethyl acrylate, 12Ethylene methyl acrylate (EMA), 233Ethylene octane (EO) copolymer, 262Ethylene styrene interpolymer

(ESI), 262Ethylene vinyl acetate (EVA), 51, 64,

223, 228, 229, 233Exfoliate, 160, 164, 186Exothermic (Exo), 54, 55, 63Expanded rubber cylinder, 222, 223Expansion coeffi cient, 75Expansion joints, 238Extinction coeffi cient, 275, 276Extruder, 45, 59, 60, 115, 262

single screw, 59Extrusion, 11, 18, 19, 30, 33, 50, 57, 58,

59, 70, 82, 113, 224, 225, 259die, 90

F

FAR 25.856, 241Fermentation, 33, 34

Subject Index 301

61259_C010.indd 30161259_C010.indd 301 10/25/2008 12:40:45 PM10/25/2008 12:40:45 PM

Fiberglass, 27, 29, 30, 258Fick’s law, 112Fillers, 281, 285Finite element method (FEM), 79Flame retardancy, 15Flame spread index (FSI), 240Floating, 238Floating devices, 244Flory-Huggin, 111Fluorocarbon, 71, 280, 281, 285Flroroectorite, 163Fluoropolymer, 239FM 4910, 241, 245FM 4924, 246Foam, 1, 3, 4

Blow molding, 62Cross-linked, 62Density, 21, 161, 168, 263, 269,

270, 285Extrusion, 14, 58, 61, 69, 70, 75

PET, 14PP, 14

Injection molding (FIM), 101, 102, 104, 114, 120, 131, 139

Isocyanurate, 15low-density, 81netting, 58Phenolic, 15polyethylene (See Polyethylene)polylactide, 6polyolefi n (See Polyolefi n)Polypropylene (See Polypropylene)Polystyrene (See Polystyrene)Polyurethane (See Polyurethane)phenolic, 15process, 226quality, 78structure, 70, 108, 122, 127–129, 139thermoplastic, 9, 19thermoset, 9

Foam-in-mold, 14Foam-in-place, 14, 15Foaming, 8, 16, 19, 33, 262

effi ciency, 165extrusion, 9oven, 232path, 9Polyurethane, 14technologies, 18

temperature, 201–203, 264–266, 269–271

x-linked PE, 4, 12Food packaging, 52, 58, D235259Forming zone, 59, 60Forurier Transform Infrared

Spectrometer (FTIR), 276Fossil, 3Free energy, 111Fruit acid, 44Fumaric acid, 44

G

Gas-assisted injection molding (GAIM), 107

Gas conductivity, 274Gas counter, 53

process, 135–138Gas evolution, 54Gas pressure, 54, 55Gaussian distribution, 200Gear pump, 71General purpose polystyrene

(GPPS), 46, 48, 49, 50Getter, 284Glass transition temp. Tg, 8, 11, 202,

249, 259, 260depression, 202

Global warming, 5, 36Global warming potential

(GWP), 47Gluconates, 44Gluconic acid, 44Glutaric acid, 44Grooved feeding, 59

H

Half-time, 145Health care, 219Heat distortion temp. (HDT), 260Heat resistance, 25Henry’s equation, 110Heterogeneous nucleation, 112, 113,

161, 201, 204, 205, 210n-hexadecyl tri-n-butyl

phosphonium, 183

302 Subject Index

61259_C010.indd 30261259_C010.indd 302 10/25/2008 12:40:45 PM10/25/2008 12:40:45 PM

n-hexadecyl tri-phenyl phosphonium, 183

Hexafl uoropropylene (HFP), 240Hexagonal, 283HMS-PP (high melt strength PP), 77Homogeneous nucleation, 112, 201Horizontal oven, 235Hydrocarbons (HC), 31, 32, 247, 260Hydrocerol, 48, 83, 92, 94, 95Hydrochlorofl uorocarbons (HCFCs),

5, 31, 32, 70HCFC-22, 47, 105, 261, 274HCFC-142b, 47, 48, 105, 261, 276, 277

Hydrofl uorocarbon (HFC), 260HFC-134a, 47, 105, 261, 274, 276, 277HFC-152a, 47, 261, 276, 277HFC-245fa, 47HFC-365mfc, 47

Hydraulic press, 51, D348234Hygroscopic, 246

I

Ionomer, 84Impact insulation, 29Impact Insulation Coeffi cient (IIC), 29Impact sound pressure, 29Incineration, 3, 36Infrared (IR), 123, 126, 127Injection molding, 4, 30, 50, 54, 101,

102, 104, 120, 126, 131, 139gas-assisted, 111

Injection nozzle, 117, 119, 120Injection velocity, 121, 126, 128–131, 139In-line rheometer die, 72, 78Insulation, 2, 26, 34, 238, 243, 245,

248, 258Interlayer opening, 182Irradiation, 151ISO

845 (desnity), 2361408, 271798 (tensile/elongation), 236, 2507214 (compression), 236, 2508067 (tear), 236, 25017088, 35

Isobutane, 47, 105, 274Isocyanurate foam, 15Isopentane, (I-pentane), 47, 105, 274

Isothermal, 274Izod, 260

L

L-lactic (L-lactide), 149, 176, 177Lactic acid, 44Lamellar, 209Landfi ll, 3, 36Lavorazione Materie Plastiche

(LMP), 3Layered silicate, 176, 177Layered titanate (HTO), 177, 178,

181, 182Leisure, 219Length to diamer, L:D, 59Life cycle, 3, 37Life Cyle Assessment, 2LLDPE (See [PE])Long chain branching (LCB), 151Low density polyamide foam, 251Low smoke (LS), 241

M

Malic acid, 44Mandrel support, 82Material safety data sheets

(MSDS), 30Medical, 219, 244Melt fi lter, 58, 59, 61Melt fl ow index (MI), 260Melt temperature, 84–86, 89, 122,

126, 129, 131, 139Metering zone, 59, 60methyl bis(2-hydroxy-ethyl)

ammonium cation (MMT-(OH)2), 157, 159, 162, 163

methyl dihydrogenated tallow ammonium (MMT-2HT), 157

Methylal, 47Microcellular, 14, 202, 226Mineral oil, 259Mixing zone, 116Modulus, 195, 213, 249Mold, 51, 53Mold foaming, 18

Subject Index 303

61259_C010.indd 30361259_C010.indd 303 10/25/2008 12:40:46 PM10/25/2008 12:40:46 PM

Mold temperature, 126, 128, 130, 131, 139

Molded bead, 14, 18, 30, 32, 33Molecular dynamics (MD), 186, 188Monocalcium phosphate, 45Monopotassium tartrate, 44Monosodium citrate, 44Montmorillonite (MMT), 157, 159, 162,

163, 176, 178, 180, 182, 183, 187–190, 196, 200, 201, 204, 206, 208, 209, 214

Morphology, 9, 148, 154, 208MuCell, 116

N

Nanoclay, 201, 209, 211, 212Nanocomposite, 143, 156, 157, 164,

175, 183, 184, 190, 194, 200, 209, 211, 214

Nanofi ller, 178, 190Nanogel, 278Newtonian, 72Nitrogen, 42, 45, 47, 105, 107, 145–148,

150, 167, 219–221, 231, 260, 281, 282, 285

autoclave, 221, 225, 229, 234Noise reduction coeffi cient (NRC), 29Nucleating agent, 93–96, 166, 167, 169Nucleation, 7, 8, 9, 19, 48, 80, 113, 128,

164, 196, 201, 226heterogeneous, 90

Nylon (NY) A403 (see Polyamide)

O

OBSH, 52Octadecyl ammonium, C18H3N

+, 178, 179, 182–185

Octadecyl di-methyl benzyl ammonium, 183

Octadecyl trimethyl ammonium, C18(CH3)3N

+, 179, 182–185Open-cell, 256, 265, 267, 278, 284, 285

structure, 8, 23modulus, 23

Optifoam, 118–120Order (See smell)

Organically modifi ed layered fi ller (OMLF), 175, 177, 180, 182, 183, 186

Outgassing, 52Oxalic acid, 444,4-Oxybis(benzenesulfonyl

hydrazide), OBSH, 51, 52, 105Oxygen permeability, 159Ozone, 33Ozoe depletion potential (ODP)

P

Packaging, 4, 31, 34, 62, 237, 259PBAs (See Blowing agents [PBAs])PE (See Polyethylene [PE])Pentane (n-pentane), 11, 33, 47, 274Permeability, 274Peroxide, 144, 151, 224Petroleum, 2, 5Pharmaceutical, 52Pharmacopoeia Monograph

USP 661, 244Phase separation, 13Phenolic, 15, 295-Phenyl tetrazole (5-PT), 51, 52, 105Physical blowing agents (PBA), 30,

101, 104–106Pipe insulation, 246Plasticizing, 77PMMA (See Polymethyle

methacrylate)PO (See Polyolefi n [PO])Polyamide, 3, 9, 11, 20, 163, 239, 256, 257

Polyamide 6, 161, 239, 246, 250, 251Polyamide 6,6, 161Polyamide 12,12, 161

Polycaprolactone (PCL), 36, 143, 144, 146–151, 153, 154, 157–165, 167

Polycarbonate (PC), 125, 127, 163, 176, 209–212

Polycarboxylic , 44Polycondensation, 10Polyester, 9, 149Polyethylene (PE), 6, 12, 33, 46, 64, 71,

151, 161high-density, 12, 58, 75, 163, 233, 256linear low density (LLDPE), 63

304 Subject Index

61259_C010.indd 30461259_C010.indd 304 10/25/2008 12:40:46 PM10/25/2008 12:40:46 PM

low-density, 12, 46, 48, 49, 58, 75, 77, 79, 83, 84, 89–91, 106, 108, 223, 224, 226, 227, 233, 238, 266, 267, 269–271, 280–282, 285

medium density (MDPE), 233metallocene (mPE), 233

Polyethylene terephthalate (PET), 9, 58, 256, 257

Polyhydroxyalkanoates (PHA), 144Polyhydroxy carboxylic acid, 44Polyisocyanurate foams, 6, 10Polylactic acid (PLA), 6, 33, 35, 58,

143, 144, 149–151, 154–157, 159, 162, 163, 168, 176, 183, 185, 186, 200, 201, 203, 204, 206, 208, 209, 214

Polymethyl methacrylate (PMMA), 57, 206

Polymer/layered silicate nanocomposite (PLSNC), 176, 177, 190, 215

Polyolefi n, 4Polyphenylene ether (PPE), 53Polyphenylene oxide (PPO), 53Polypropylene (PP), 12, 42, 58, 75, 77,

78, 89, 121, 132, 176, 187, 233binder, 33HMS-, 77, 83maleated (PP-MA), 187, 192–196,

198, 199Polystyrene (PS), 3, 4, 11, 12, 18, 29,

33, 46, 49, 58, 71, 75, 163, 238, 256, 257, 260–264, 266–273, 279, 282–285

bead, 11, 14molded bead, 4open cell, 278syndiotactic, 161

Polyurethane (PU), 3, 4, 10, 19, 20, 27, 256, 259

board process, 17fl exible, 10, 14, 18, 29elastomeric, 10integral skin, 14open cell, 278, 279rigid, 10, 14, 18, 29spray process, 17wedge, 27

Polyvinyl alcohol (PVOH), 35, 36

Polyvinylchloride (PVC), 4, 42, 58, 106plasticized, 62

Polyvinylidene fl uoride (PVDF), 239–244

PP (See Polypropylene [PP])PP-based nanocomposite (PPCN),

188, 189, 191–196, 198, 199Porous ceramic, 214Prefoaming, 58, 61Premature foaming, 78Pressure distribution, 80Pressure drop, 113, 186

rate, 148Pressure gradient, 81, 91, 92, 94, 95, 113Pressure quench method, 149Pressure release valve, 115Propane, 47PS (See Polystyrene [PS])PU (See Polyurethane [PU])p-v-T (pressure-volume-temp.) plot, 75

R

R-Value (See Thermal resistance)Radiation, 274Reactive foaming, 18Reactive foaming, injection molding

(RIM), 14, 16Recreation, 4Recycle, 36Refrigerating application, 258Reuse, 36Rheology, 70, 154, 157Rheometer (Also see in-line

rheometer), 189Rheopexy, 189, 190Rotational molding, 62, 63Rubber, 10, 64Rupture, 262

S

Salt bath, 230, 233, 237Sandwich process, 53Sapphire window, 197Scanning electron microscopy (SEM),

162, 163, 191, 200, 207, 209, 210, 215, 227, 244, 273

Subject Index 305

61259_C010.indd 30561259_C010.indd 305 10/25/2008 12:40:46 PM10/25/2008 12:40:46 PM

Seals, 239Semicarbazide, 57Semi-continuous process, 30, 232Semi-crystallization, 12Shape factor, 23Shark skin, 58Shear-thinning, 72, 78Shear viscosity, 188, 189Short chain branching, 151Shut-off nozzle, 116, 117, 119Silica, 46Silver streaks, 132, 133, 136Sink mark, 54Sinter metal, 119Smell, 44, 54, 57Smog, 32,33Smoke developed index (SDI), 240Sodium acid pyrophosphate, 45, 46Sodium aluminum phosphate, 45, 46Sodium aluminum sulfate, 45Sodium bicarbonate, 13, 49, 52, 57Sodium borohydride, 57Sodium carbonate, 42, 83, 105Solid state shear processing, 186Solubility, 11, 21, 46, 110, 148, 156, 160,

168, 196sound absorption, 34, 37Sound transmission loss, 28Spider, 82–84, 88

geometry, 88, 89legs, 70, 78, 81, 83, 87

Spinoidal decomposition, 13Sports, 219Starch, 36Static mixer, 119, 120Steinen Tunnel, 240Strain-induced hardening, 188, 190Stress-strain curve, 196Styrene, 259Styrene/ethylene butylene/

styrene (SEBS), 262Succinic acid, 44Sulfohydrazide, 51, 57Supercritical

fl uids, 107, 225state, 106, 107

Supersaturation, 113, 191Surface temperature, 123, 124, 127Surface tension, 7, 112

T

Talc, 46, 92, 93, 95, 164–166, 168Tartaric acid, 44Tensile, 242, 260tetrafl uoroethylene (TFE), 240Tetrazoles, 57Thermal conduction, 275Thermal conductivity, 26, 46, 243,

248, 252, 272, 274, 277, 281, 282, 284, 285

Thermal energy, 275Thermal insulation, 238, 245Thermal oxidizer, 32, 33Thermal resistance, 26Thermodynamic instability, 111

instability, 13Thermoforming, 27, 34Thermoplastic foam, 9

thermoset vs., 9Thermoset foam, 9Titanium dioxide (TiO2), 164,

166, 168Toluene diisocyanate (TDI), 10Toluene sulfonyl hydrazide (TSH),

52, 105Toluene sulfonyl semicarbazide

(TSSC), 52, 105Transportation, 4, D45862, 2452,4,6-Trihydrazino-1,3,5-triaazine

(THT), 52T-Tubes, 245

U

UL 94 V-O, 240, 241UL 723, 240Ultraviolet light (UV), 12, 13, 62,

240, 252Underfl oor (or underlay), 238Union Carbide Corp. Process (UCC),

50, 51

V

Vacuum Insulation Panel (VIP), 255–258, 280–285

Vacuum packaging, 257van der Waals +A230, 2, 9

306 Subject Index

61259_C010.indd 30661259_C010.indd 306 10/25/2008 12:40:46 PM10/25/2008 12:40:46 PM

Velocity distribution, 80Vertical oven, 230, 232, 235Vinyl acetate monomer (VAM), 36Vinyl alcohol (EVOH), 257Vinylidene fl uoride (VDF), 240Viscosity, 70, 72, 74, 76, 77, 78,

107, 108, 123shear, 69, 71, 108

Volatile organic chemical (VOC), 251

Volume expansion ratio (VER), 278

W

Water, 47, 105Water soluble, 36, 37

Wide angle X-ray diffraction (WAXD), 177, 180, 181, 183, 185

Wood plastics, 62

X

X-linked, 30X-PE (or XL-PE), 4, 14, 247, 248X-ray, 126, 127, 197Xenotest, 242

Y

Yellowness index, 242

Z

Zotefoams, 220

Subject Index 307

61259_C010.indd 30761259_C010.indd 307 10/25/2008 12:40:46 PM10/25/2008 12:40:46 PM

61259_C010.indd 30861259_C010.indd 308 10/25/2008 12:40:46 PM10/25/2008 12:40:46 PM

polymeric foams technology and developments in regulation process and products

Documents