Glazed and Confused: Exposing the Mysteries of Glazed Architectural Terra Cotta

Download Glazed and Confused: Exposing the Mysteries of Glazed Architectural Terra Cotta

Post on 13-Nov-2014

1.578 views

Category:

Documents

5 download

DESCRIPTION

AbstractGlazed and Confused, is a compilation of two investigative studies. The first is a chronological annotated bibliography of physical property experimentations interspersed with key historical moments pertaining specifically to glazed architectural terracotta. The second is an experimental model created for architectural conservation, to be utilized prior to conservative treatments. This thesis began with wanting to test the commercially available repair and patching mortars employed during terracotta repair, hoping to discover which performed most similar to glazed terracotta. However, after speaking with several practicing conservators within the United States, it quickly became apparent there was a lack of knowledge surrounding the basic physical properties and behaviors of architectural terracotta. It is impossible to judge a repair system using unknowns as criterion, consequently the topic was shifted to discovery of previously established properties, the procedures used to procure this information, when theses experimentations were completed and why. Using the ascertained historical information a research program was developed specifically focused towards architectural conservation, with the goal of creating a program of examination with the findings aiding in the choice of conservation treatments. The model included five laboratories: visual analysis, petrographic analysis, cross-section inspection, water vapor transmission and inverted cup water vapor transmission. The historic documentation research exposed a wealth of material testing programs done and presented through The American Ceramic Society, The National Terra Cotta Society and The National Bureau of Standards, all of which dissipated by 1961, when the last document was produced. There are currently no required testing programs for manufactures of glazed architectural terracotta in the United States, and consequently no required specifications for material performance.The materials testing program was employed on six historic terracotta blocks fabricated between 1910 and 1921, revealing that any defect in the glaze fit, such as crawling and/or crazing results in an increased rate of water vapor transmission. The microscopic examinations divulged information pertaining to the varying vitrification levels of the glaze and body. The petrography and cross-section analysis proved that crazing of a glaze can continue into the body of the block and also that crawling exposes the body to exterior elements, in addition the cross sections showed that every specimen had a glaze layer thickness of 200 m, suggesting mechanical application methods. Full Title: Glazed and Confused: Exposing the Mysteries of Glazed Architectural Terra Cotta.Author: Xsusha Carlyann Flandro Graduation Date: May 2009 Thesis Advisor: Norman Weiss

TRANSCRIPT

FL A N 20 D R 09 OBy Xsusha Carlyann Flandro1- FLANDROGLAZED AND CONFUSED Exposing the Mysteries of Glazed Architectural Terra CottaSubmitted in partial fulfillment of the requirement for the degree Master of Science in Historic PreservationGraduate School of Architecture, Planning and Preservation Columbia University May 20092- FLANDROFL A N 20 D R 09 OPermission to Copy Copyright 2009, Xsusha Carlyann Flandro. For information about this work, please contact Xsusha Flandro, c/o Jennifer Flandro 277 Washington Avenue Apt. #2K, Brooklyn, NY, 11205. Permission is hereby granted to reproduce and distribute copies of this work for nonprofit educational purposes, provided that copies are distributed at or below cost, and that the author, source and copyright notice are included on each copy. This permission is in addition to rights of reproduction granted under Sections 107, 108, and other provisions of the U.S. Copyright Act. Before making any distribution of this work, please contact the copyright owner to ascertain whether you have the current version.FL A N 20 D R 09 O3- FLANDRO4- FLANDROFL A N 20 D R 09 OAcknowledgementsThere are many people who need to be thanked for their consistent pressure on me to do a good job and their ability to help me along by supplying me with their continual support, knowledge and expertise. Namely, Professors Norman Weiss and Dr. George Wheeler, whose enthusiasm for teaching and interest in material conservation continually amazes me. Judy Jacob, for offering her time to assist me in my research and serving as my thesis reader. Professor Richard Pieper who showed me that it is a gift to be obsessed with one material. And of course my family who have continually supported, mocked and humored me throughout my full-time ten-year college educational path.FL A N 20 D R 09 O5- FLANDRO6- FLANDROFL A N 20 D R 09 OAbstractGlazed and Confused, is a compilation of two investigative studies. The first is a chronological annotated bibliography of physical property experimentations interspersed with key historical moments pertaining specifically to glazed architectural terracotta. The second is an experimental model created for architectural conservation, to be utilized prior to conservative treatments. This thesis began with wanting to test the commercially available repair and patching mortars employed during terracotta repair, hoping to discover which performed most similar to glazed terracotta. However, after speaking with several practicing conservators within the United States, it quickly became apparent there was a lack of knowledge surrounding the basic physical properties and behaviors of architectural terracotta. It is impossible to judge a repair system using unknowns as criterion, consequently the topic was shifted to discovery of previously established properties, the procedures used to procure this information, when theses experimentations were completed and why. Using the ascertained historical information a research program was developed specifically focused towards architectural conservation, with the goal of creating a program of examination with the findings aiding in the choice of conservation treatments. The model included five laboratories: visual analysis, petrographic analysis, cross-section inspection, water vapor transmission and inverted cup water vapor transmission. The historic documentation research exposed a wealth of material testing programs done and presented through The American Ceramic Society, The National Terra Cotta Society and The National Bureau of Standards, all of which dissipated by 1961, when the last document was produced. There are currently no required testing programs for manufactures of glazed architectural terracotta in the United States, and consequently no required specifications for material performance. The materials testing program was employed on six historic terracotta blocks fabricated between 1910 and 1921, revealing that any defect in the glaze fit, such as crawling and/ or crazing results in an increased rate of water vapor transmission. The microscopic examinations divulged information pertaining to the varying vitrification levels of the glaze and body. The petrography and cross-section analysis proved that crazing of a glaze can continue into the body of the block and also that crawling exposes the body to exterior elements, in addition the cross sections showed that every specimen had a glaze layer thickness of 200 m, suggesting mechanical application methods. Full Title: Glazed and Confused: Exposing the Mysteries of Glazed Architectural Terra Cotta. Author: Xsusha Carlyann Flandro, May 2009 Advisor: Professor Norman WeissFL A N 20 D R 09 O7- FLANDRO8- FLANDROFL A N 20 D R 09 OPrefaceThere are literally hundreds of books and articles written about architectural ceramic uses in the United States. My goal here is not to rehash what has already been stated but rather to survey the descriptive written works of testing procedures used throughout history to study the physical properties of glazed architectural terra cotta. A brief historical introduction outlining the use of glazed terracotta in the United States has been included as a necessary means to the evaluation of terra cotta. My research had two main goals, first, to compile a history of glazed terracotta manufacturing and testing procedures completed in the United States and secondly, armed with the knowledge that there is currently no published data with regards to the rate of water vapor transmission of glazed terracotta experiencing crazing and crawling, to perform this relatively simple test and publish the findings. The hope is that through this test and very basic microscopic examination of the material the value in performing a few basic tests prior to restoration or replacement of terra cotta pieces will be exposed and become a best practice in the conservation field. In addition, credit has to be given to Harry G. Schurecht (1893-1968), a ceramic chemist who tested terracotta with such fervor and persistence, that I strongly believe without him architectural terracotta would have never have reached its peak.FL A N 20 D R 09 OMr. Harry G. Schurecht11As those have done before me the term terracotta has been hyphenated when used as an adjective, but not when used as a noun.Image from. Terra-Cotta Skyline by Susan Tunick, 1997, page 135.9- FLANDRO10- FLANDROFL A N 20 D R 09 OTABLE OF CONTENTSINTRODUCTION.....13 A Short History of Architectural Terra Cotta in the United States.....18 SECTION I: Previous Research and Key Historical Moments.24 SECTION II: Material Testing...............92 Testing Methodologies............95 Extracting Samples........95 Specimen labeling..........96 Visual Analysis.......................96 Petrographic Analysis............96 Cross Section Analysis...........97 Water Vapor Transmission Test..............98 Inverted Cup Water Vapor Transmission Test............99 Normalized Data......................................................................................100 Water Vapor Transmission Comparative Analysis...................................101 Terra-cotta Block #1 Test Results........102 Terra-cotta Block #2 Test Results....109 Terra-cotta Block #3 Test Results....118 Terra-cotta Block #4 Test Results........125 Terra-cotta Block #5 & #6 Test Results.......130 Water Vapor Transmissions Comparative Analysis.....137 Material Testing Conclusion............................141 RECOMMENDATIONS.145 CONCLUSION149 BIBLIOGRAPHY152 LIST OF FIGURES.170 CERAMIC NOMENCLATURE......................................................................................176 APPENDIX..185 Water Vapor Transmission Raw Data......186 Inverted Cup Water Vapor Transmission Raw Data.....189 Rate of Water Vapor Transmissions.........192 Comparative Analysis Water Vapor Transmission Rates.....194 Dixons Q-Test Mathematics............202FL A N 20 D R 09 O11- FLANDRO12- FLANDROFL A N 20 D R 09 OFL A N 20 D R 09 OINTRODUCTION13- FLANDROFL A N 20 D R 09 OGlazed architectural terracotta on 154-160 West 14th Street, New York City (ca 1912).14- FLANDROThe initial plan for this thesis was to evaluate patching materials used in the United States for the repair of damaged glazed architectural terracotta. With two major brands being marketed in the United States, the goal was to determine which one was the most effective and the pros and cons of each product, hoping that this information would lead to better repairs. Through a series of conversations and a search of historical references, it was found that little had been recently completed to determine the physical properties of terra cotta. It is hard to judge the quality of a repair when the properties of the original fabric are unknown. An exhaustive research project was undertaken to determine what experiments had been previously completed on glazed architectural terracotta to establish its physical and mechanical properties. From this research an experimental design was modeled specifically for glazed terracotta materials conservation and tried out on pieces of glazed terracotta taken from four historic buildings, in hopes that the information revealed from this experiment would be sufficient to make more informed decisions prior conservation treatments. In the United States there are two main companies that produce repair mortars for terra cotta, Cathedral Stone Products (also known by their German name, Jahn) and EdisonCoatings Incorporated. Each company has excellent and readily available representatives, as well as external and on-site training sessions (a three day training course is required for the purchase and use of Cathedral Stone Products). Cathedral Stone Products terracotta repair mortar is Jahn M100 to be used in conjunction with Benjamin Moores AURA line of paints to replicate glaze sheen (color is achieved by the matching of the mortar through the addition of synthetic pigments). For varying sheens AURA Low Lustre 634, Semi Gloss 632 and Flat 629 are used. During the time of this research and shortly before its publication, Cathedral Stones repair mortar, Thin-Set 125, used for the repair of shallow glaze spalls and their own system of glaze replication, TerraCoat Glaze Repair, was removed from the market due to reoccurring improper installation and subsequent failures. The claim the company makes about M100 is that it is completely water permeable and contains no latex or acrylic bonding agents or additives.1 Their goal is to match the physi-FL A N 20 D R 09 O1cal properties of the substrate to allow salts, water vapor and liquid water to penetrate the repair. While the patching mortar may be compatible with the substrate once the addition of AURA is applied this no longer maters. AURA paints are 100% Acrylic latex; thereby eliminating any water vapor transmission from the rear of the terracotta through the front. This fact was discussed with the laboratory personal at Cathedral Stone and it was revealed that they are aware of this and are unconcerned, as terracotta glazes are believed to be impermeable, so the paint mimics the impermeable characteristic of ceramic glaze. Edison Coatings has a variety of products available for glazed terracotta repairs, but the most common ones used are: Thin-Fill 55 for filling shallow losses (1/32 to 1/8 deep), Custom System 45 TC or Custom System 45 for deeper repairs, Elastowall 351 and Aquathane UA210 type NCL used for glaze replications and Aquathane UA210-E and AquaSpex 220 for glaze detail replication (such as speckles, smears and mottling inherent in some terra-cotta glazes). Edisons claims about their products are that they are latex-modified mortars, which allows them to be tough, durable, have excellent water vapor transmission rates, high tensile bond strengths, and low modulus of elasticity ensuring good adhesion. In addition the latex allows for achievement of incredibly smooth finishes (the repairs can be polished with 400 grit sand paper) and they are available in over 900 standard colors. Their goal is to allowCathedral Stone website, product literature for Jahn Repair Mortar M100. www.cathedralstone.com, 2009.15- FLANDROthe repair to move with the substrate, eliminating stresses in either material. The glaze replication products are waterborne polyurethane based coatings with varying degrees of hardness and permeability. They can be applied once the repair mortar has set. Once the differences between the two different repair mortars were determined, several architectural conservators throughout the United States were interviewed. The goal of these interviews was to reveal which product was used more and how the decision to go with a specific product was made, was it workability? Durability? Water vapor transmission rate? Color? Inquiries were made specifically about testing procedures used prior to deciding which product to purchase. The interviews revealed a national phenomenon among conservators as well as a lack of knowledge about the basic properties and mechanics of polychrome terracotta. Choices between the two products were being made based on best business practices, with little or no materials testing being done at all. The company that the conservators choose to supply the treatment materials was synonymous with the company that could provide the best warranty on their product (this also allowed for one supplier to be held libel in the event of premature failure). The thought arose that perhaps the physical properties of glazed terracotta had already been determined and therefore the material testing of historic fabric was not required prior to repair. Research was done for any publications stating the physical properties of glazed architectural terra cotta and any testing procedures completed to determine said properties, a particular focus was given to glaze fit tests. This research was compiled chronologically to display the evolution of glazed terracotta manufacturing and testing programs (who knew what and when). The most recent publication retreived was printed in 1961, by the American Institute of Architects and had a specific focus towards ceramic veneer rather than terracotta. It was also revealed that there are currently no required ASTM (American Society for Testing and Materials International) standard tests pertaining specifically to the physical properties and glaze fit on architectural terracotta and consequently no minimum material standards that have to be met in the current manufacturing of glazed architectural terracotta. In addition while the ASTM has standardized some nomenclature pertaining to ceramic materials, the document is not thorough, to help resolve this issue a new glossary of ceramic terms was compiled and can be found in the appendix. Using the information discovered in the historical research an experimental design was modeled for conservationists to determine what information about a terracotta block could be obtained with easy tests and minimal samples. This model was tested using four historicFL A N 20 D R 09 O16- FLANDROFL A N 20 D R 09 O17- FLANDROterra-cotta blocks. The model experiment included five procedures: Glazed surface visual inspection- to reveal the typical glazed surface conditions on a block. Cross section analysis- to determine the thickness of the glaze layer, the presence of a slip layer and if any glaze fit defects (crazing/crawling) expose or continue into the terracotta clay body, which it was hypothesized would cause a higher water vapor transmission rate. Petrography- to expose the level of vitrification the glaze and terracotta body achieved during firing. Water vapor transmission (and inverted cup water vapor transmission)- to determine if glazed pieces exhibiting signs of glaze fit defects (crazing and crawling) had higher, lower or similar rates of water vapor transmission to piece with no glaze fit defects. It was expected that any glaze fit defects and less vitrification of either the glaze or body would cause a higher rate of water vapor transmission.A Short History of Architectural Terra Cotta in the United States Ceramic production came to America with early settlers as a memory; it had to be recalled without written guidelines and instructions. America did not have the historic ceramic factories with well equipped laboratories like Europe did, and colonists had to improvise building kilns, mixing clays, and formulating glaze recipes, a daunting and difficult task. The frustrations caused by these conditions in the 1700s were exasperated by the constant comparison of colonial produced goods against those of the semi-industrialized Europe. In this commercial society, the survival of American potteries was determined by their ability to evolve the quality of their goods quickly and keep up with the products imported from Europe. American terra cotta must therefore bid for favor only in its most perfect form. We do not tolerate crooked lines or varying color in our buildings.2 In 1876 the UnitedFL A N 20 D R 09 O2States was finally recognized as being a substantial producer of pottery at the Philadelphia Centennial in Pennsylvania.3 The first large scale American producer of architectural terra cotta was the Tolman, Luther and Co. in Worcester, Massachusetts.4 Henry Tolman Jr. established the company in 1848; he was greatly influenced and worked closely with an architect from Worcester by the name of Elbridge Boyden. As quoted by Shockley and Tunick in the Winterthur Journal, Boyden claims in his diary I conceived the idea of using burnt clay for ornaments. I interested a potter in the notion, started some potteries, and had some ornaments made which proved a success.,5 (while he may have had a dream, terra cotta had already been used for centuries on architecture in other parts of the world). The Tolman, Luther and Company employed Boyden as an agent to sale their terra-cotta decorations, which at the time consisted of ready-made ornaments (cornices, moldings, column capitals etc.) rather than custom projects for an individual site. Their terra cotta has recently been found on the Copper Union Building in New York City, erected in 1853.6Large construction projects began to use American made terra cotta as early as 1851 (The Trinity Building by Richard Upjohn New York City, now demolished); however, the majority of buildings decorated with terra cotta are from the 1870s forward. The first smallThe Manufacture of Terra-Cotta in Chicago, The American Architect and Building News. (December 30, 1876. 3 Elaine Levin, The History of American Ceramics (New York, NY: Harry Abrams Inc., 1988), 10. 4 Jay Shockley and Susan Tunick, The Cooper Union Building and Architectural Terra Cotta, Winterthur Portfolio Volume 39 Number 4, (Winter 2004): 211. 5 Shockley and Tunick, 222. 6 Shockley and Tunick, 210.18- FLANDROwave of terra cotta use slowed in the early 1860s when the general acceptance of the fabric had not materialized among builders and architects. The second wave of terra-cotta factories had the additional challenge of competing with the stone and cast iron industries. Commensurate decorations could be produced in terracotta, stone or cast iron. In the 1870s when terracotta use surged again, it was the result of a few different social happenings. First, there were better machines, making it easier to produce large amounts of homogenous (both in shape and hardness) pieces. Second, the Chicago fire of 1871 caused a demand for fireproof construction materials and third, technology had advanced so that terra cotta could be properly set on buildings. Terra-cotta ornamentation boomed during the last three decades of the 19th century and it can be found extensively onFL A N 20 D R 09 Obuildings from this period.7 8The next development for the industry was the introduction of glazes. While glazing on tiles had been done for centuries, the use of them on exterior terracotta units in the United States was a slow development that began in the 1890s. There was both hesitation and excitement among architects to employ color on the exterior buildings, with one group rallying against its use stating it to be distracting from the pure architectural form and an imitative material and the other group being The lovers of color.7 Its peak use was between1915 and 1925. In addition to color, glazed terracotta was marketed as the self-cleaning8 building material, an important characteristic in the time of coal burning fuel. One of the earliest uses of polychrome terracotta was completed in 1891, when it was applied liberally to add color to the lobby of The Charlesgate, a residential hotel in Boston. This creation was actually the result of two industries coming together. Eugene Attwood and William Grueby (workers from the well known Low Art Tile Works company in Chelsea, Massachusetts) designed the glazes for terra-cotta forms made by the Boston Terra Cotta Company. Prior to this venture the use of polychrome glazes had been limited to tiles on floors and walls. The most widely publicized use of polychrome terra cotta on an exterior was in 1906 at McKim Mead and Whites Madison Square Garden Church (demolished in 1916), however earlier examples have been discovered. While the first building to use polychrome terracotta extensively on the exterior in all of the United States was a small floral shop (Figure 1) in St. Louis Missouri erected in early 1898, and the second the Kelly & McAlinden Hardware Store in Perth Amboy, New Jersey also in 1898. The earliest remaining example is located at 277 Broadway in New York City, erected in 1899, with pieces supplied by the Perth Amboy Terra Cotta Company (Figure 2).Howe, Samuel. Polychrome Terra Cotta, The American Architect. February 28, 1912. The Clay-Worker. Self-Cleaning Buildings, Volume 39/40, 1903 page 172.19- FLANDROFL A N 20 D R 09 OFigure 1: The Fred C. Weber Floral Shop, St. Louis, Missouri109Figure 2: Polychrome terra cotta on the Broadway-Chambers building, NYC.11These three examples were not completed during the peak use of the material, but rather several years before. The use of polychrome terra cotta in the United States was most popular during the Art Nouveau and the Art Deco periods (roughly 1910-1925). It was during this time that the majority of the scientific experimentations were done to determine the physical properties of terra cotta. Many of these studies were done in conjunction with the National Terra Cotta Society (members of which included the top twenty-five architectural terra cotta producers during the time) and the National Bureau of Standards. Even during the peak period architects were hesitant to use color on masonry buildings exteriors, as noted in a 1911 periodical. It is easy to imagine the decadence in the art of sculpture that would result from giving predominance to questions of color over those of form. 9 Efforts were made to combat the bias. The Brickbuilder magazine held a competition in 1911 for an a hotel with intelligence shown in the constructive use of architectural terra cotta.10 With the winning design going to completely glazed terra cotta plan by William La Zinsk and Dwight James Baum (the hotel was never erected). In addition to this, The National Terra Cotta Society, published a series of brochures starting in 1914 with images of suggested ways in which to use colored terracotta.11Hewlett, J. Monroe, Polychrome Terra Cotta in Exterior Architecture, The Brickbuilder. Volume 20 April 1911, 71. 10 The Brickbuilder, Extra edition to Volume XX, No.2, February 1911. 11 The National Terra Cotta Society, Architectural Terra Cotta Brochure Series: The School, The Theatre,20- FLANDROFL A N 20 D R 09 OFigure 3: Glaze applied with a pulsichromerter.21- FLANDROAnother part of the problem and consequent infrequent use of glazed terra cotta was its manufacture was a cost and time prohibited process. One of the manufacturing difficulties included achieving an even coat of glaze, in the tile industry polychrome glazes were hand painted, this method was inefficient and vulnerable to human error. Terra-cotta manufactures primarily used atomizers to spray their glazes on, however, this became problematic when multiple colors were desired on one piece. Areas would have to be masked off using plaster or other methods. In 1924 the pulsichrometer was introduced to combat the time consumption of using multiple colors. The pulsichrometer could apply several colors at once and some models included a conveyor belt system, lessening the amount of work need to spray one piece. It also helped to popularize the use of mottled glazes to imitate granite. Before this innovation the application of glazes to look like granite required a worker to evenly splash on slip (similar to a Jackson Pollock painting), Any ordinary intelligent workman can be taught in a very short time to obtain a good result, for the machine gauges the amount of enamel applied, and, at the same time, the desired effect is obtained by the regulation of air pressure, thereby making the general effect consistent and similar throughout.12 Pulsichrom-eter sprayed pieces have a very distinctive soft look and are easily identifiable.Figure 4: Glaze applied by splashing.The Store, The Bank, The Garage. 1914-1922. 12 Clark, Ernest, The Journal of the American Ceramic Society. The Use of Pulsichrome from a Manufactures Standpoint Volume 5, Issue 11, November 1922, 826-827.FL A N 20 D R 09 O13The last major development in the use of terracotta was created by the Gladding, McBean Company, located in Lincoln, California, invented during the 1920s it is known as terra cotta veneer.13 Ceramic veneer differs from architectural terracotta in that it is usually plain, flat and has been extruded rather than molded. This veneer was used in Art Deco buildings to stream line the facades, but was found to be no more cost or time efficient than using other materials such as glazed brick or tile. Terra cottas advantage was that it could carry a precise and high quality reproduction of fine architectural sculpture,14 and the veneer did not exploit this. As the head of the President of the Federal Seaboard Terra Cotta Company put it in 1932, The dream of the modern architect is to build houses entirely made out of metal, glass and cement. In this construction, brick, tile or terra cotta has no place,15 in 1934 the National Terra Cotta Society folded and by 1947 only seven terra cotta factories remained. In 1961 one of the remaining factories, Gladding McBean, combined efforts with the American Institute of Architects to create the most recent document pertaining to material testing procedures for architectural ceramics, The Public Work Specifications for Ceramic Veneer, being almost fifty years old is now outdated and difficult to locate with only a few copies in public circulation. The seven remaining factories after 1947, including Gladding, McBean, survived the next twenty years producing large quantities of terra-cotta sewer pipes and doing small terracotta decorative jobs. In the 1970s along with the establishment of the first Historic Preservation Masters Program in the United States at Columbia University, a renewed interest in the historic built world started, and this included curiosities about the historical uses of architectural terracotta. The majority of terra-cotta productions during the 1970s to the present time are for replacement pieces on historic buildings. In 1981 the non-profit organization Friends of Terra Cotta was formed which promotes the preservation of terracotta. Currently in architecture there has been a reintroduction of the use of alternative construction materials and this has lead to the upsurge of interest in terracotta. Because of its easy repetitiveness and inherent refinement, terra cotta can be a means of creating small scale articulation as a counter point to the big size of many of our buildings today.16 Terra cotta factories today welcome artists who can help develop new designs and have reestablished the connections between artists, history, chemistry andMartin E. Weaver with F.G. Matero, Conserving Buildings: Guides to Techniques and Materials Revised Edition (New Jersey: John Wiley and Sons, 1997), 110. 14 Weaver with Matero, 10. 15 Gary F. Kurutz, Architectural Terra Cotta by Gladding, McBean (California: Windgate Press, 1989), 10. 16 Kurutz, 11.22- FLANDROindustry. The architectural ceramic industry is often not given enough credit when it comes to their manufacturing processes. Because clay is often thought of as an art pottery product very little attention has been given to the chemical and technical achievements made by those working in the architectural terracotta industry. The pieces that are installed on buildings weight hundreds of pounds (still lighter than stone) and can be meters in length. The production and installation of such works is not a small task and should not be dismissed with lightly. In the following section a never before printed or compiled chronological bibliography of published experimentations and their key points have been outlined and interspersed with key moments in the architectural terracotta industry. The focus of this bibliography was limited to experimentations involving the application of glaze, the glaze bodies and glaze fit on terracotta. It begins shortly prior to the use of polychrome terracotta in the 1890s.FL A N 20 D R 09 O23- FLANDROFL A N 20 D R 09 O24- FLANDROSECTION I: PREVIOUS RESEARCH AND KEY HISTORICAL MOMENTSPre-18991830, May. Establishment of the Office of Standard Weights and Measures as a section of the Coast Survey, Department of the Treasury. The office was required to report on the various weights and measures in use at the U.S. customhouses. 1852/ 1853. Terra cotta is used to construct portions of the Cooper Union Building (now the oldest remaining use of terra cotta in NYC) and also the string courses and window trimmings on the Hotel St. Denis (on Broadway and 11th Street in New York Citynow stripped) as well as for the bracket pieces on a James Renwick building in New York.FL A N 20 D R 09 O25- FLANDRO1876, December. The American Architect and Building News, The Manufacture of Terra-Cotta in Chicago. The six-page article includes illustrations of how terracotta is made but most interestingly states, Recently, glazed terra-cotta has been employed for inlays in brick buildings with good effect; and semi-glazed cornices have proved effective and durable. The work has stood well wherever care has been taken to prevent moisture from invading the hollow cells, and freezing therein. Colored slips were created by a process of boiling clay to separate the fine clay particles from the grit. It also reveals the recipe for the glazes and its application. The glazes were lead glazes consisting of lead oxide ground in oil with pulverized calcined flint and applied as a wash before firing. 1893, January, February, March. The Brickbuilder, Architectural Terra-Cotta. A three part article on uses and manufacturing of architectural terracotta. There is only one mention of glaze and it is in reference to drying pieces that contain carbonic acid from rainwater. It was believed that as the carbonic acid dried it left a scum on the surface of the wet clay, which if was not removed would cause any coloring matter to peel off. While there buildings with glazed architectural terracotta built prior to 1900, there were no other written records found pertaining to the mechanisms used by the industry to ascertain the proper glaze fit on the terracotta body.1897. The Clay-Worker, Volume 28/30, Terra Cotta Construction. This short one page article gives written suggestions about the use of architectural terra cotta. These six suggestions include: 1. All terra cotta work should be filled thoroughly with solid brick and mortar 2. All lime mortar used for filling in and setting the terra cotta should be thoroughly slacked and should be made up two full weeks before its use.3. The beds of all terra cotta blocks should be at least 1/8th inch thick, and no blocks should be placed directly on top of another or on top of brick work without a bed of mortar. 4. All the rebates in the joints of the terra cotta should be filled up with mortar and not left hollow.FL A N 20 D R 09 O26- FLANDRO5. No wedges should be used, either in the beds or joints. The beds and joints should be made entirely from mortar 6. All of the terra cotta work should be washed down after the building is set, with a weak solution of oxalic acid and water, and also some recommended a coat of raw linseed oil (with no colorants in it).1897, January. The Clay-Worker, Clay Glazes- No. 1, written for The ClayWorker by professor Edward Orton Jr., this article goes through a basic history of glazes and also basic differences between a glaze and a glass. 1897, February. The 11th Annual Brick convention is held in Buffalo New York.The American Terra Cotta and Ceramic Company hosted an exhibition room where they displayed several different samples of glazed architectural terra cotta.1897, April. The Clay-Worker, Chicagos Latest Architectural Achievement. A buff colored glazed architectural terra cotta was used for the majority of the interior at the Van Buren Street Train Station in Chicago, IL.1898. The Clay-Worker, Volume 29/30, A New Architecture, the article is about the use of polychrome schemes in architecture, specifically employing terra cotta as the material. It mentions the Jewish Synagogue in New York, which was poorly received. 1898. The Clay-Worker, Volume 29/30, Terra Cotta Fronts, this article featuresa house with the entire front built from terra cotta supplied by the American Terra Cotta Company. It was located on West Adams Street in Chicago and designed by Jenney & Mundie and Handy and Cady Architects for Mr. Herzog. 1898. The Clay-Worker, Volume 29/30, Clay in Architecture. This article is about the use of a white terra cotta for the door surrounds on a small floral shop in St. Louis Missouri. This is the first documented use of polychrome terra cotta on a building in the United States.1898. The little Kelly & McAlinden Building (a small hardware shop) was redesigned in Perth Amboy New Jersey, and is the second recorded use of exterior polychrome terracotta in the United States. The pieces were supplied as experimentation in color by the Perth Amboy Company. 1898, February. Read at the 12th annual National Brick Manufactures Association convention in Pittsburgh, a young man named Elmer E. Gorton representing the American Terra Cotta & Ceramic Co. of Terra Cotta, Illinois, presented, Experimental Work, Wise and Otherwise, the subject of which was concerned with his struggles to make a glaze for terracotta. His paper was the first of such a scientific nature to be presented at a NBMA convention and it sparked a group of ceramic engineers, chemists and practitioners to get together and form the American Ceramic Society, which held its first meeting in 1899. What is referred to as Section Q was a secret society within the American Ceramic Society where attendees met without formality and would exchange knowledge, test results and strategies. 1898, March. The Clay-Worker, Volume 29/30, Terra Cotta in Architecture. This article was read before the Cincinnati Chapter of Architects on March 20 , 1898. It list the earliest uses of terra cotta in the United States for buildings (including the Cooper Union Building and also the State House in Springfield, IL)FL A N 20 D R 09 O1900-190927- FLANDRO1899, January. Transactions of the American Ceramic Society, Volume I. The Independence of Burned Clay as a Decorative Building Material, written by Herman C. Mueller, he brings up the point of why terra cotta is used as an imitative material when it has many characteristics that stone doesnt have, and also why these characteristics should be exploited. What is great about this article is in the discussion at the end where Mr. William D. Gates (founder of the American Terra Cotta & Ceramics Company) lets the participants of the meeting know that he has In the case of my own home, I tried a new idea. I used square blocks, but I did not try to imitate the color of any stone. I tool a mottled, speckled effect that I admired, that did not look like any particular stone.1901, March 3rd. Abolishment of the Office of Standard Weights and Measures and the creation of the National Bureau of Standards, which is in charge of determining methods of testing and also material standards for building materials, including terra cotta. 1901, March. Transactions of the American Ceramic Society, Volume III, The Useof Glazed Clay-Ware as an Architectural Decoration in Exterior vs. Interior Work. Otto Hensel compiled this article speaking specifically about the increased used of glazed terra cotta on buildings in Chicago. This article relates specifically to the use of color on buildings and where on buildings color should be used. 1901, May. The Clay-Worker Volume 35/36, Terra Cotta Buildings, in this short three paragraph piece it is stated that terra cotta with glaze is non-absorbent and can be cleaned by turning the hose on it. Both statements we now in 2009 know to be false. 1901, May. The Clay-Worker Volume 35/36, CrazingIts Cause and Remedy Wanted, published in the portion of this periodical that allows for letters of inquiry, a worker in a Brooklyn, NY factory states that every few months their glazes show crazing not immediately visible upon removal of specimens from the kiln. The crazing shows up either two or three days later or even three months later. He would like an answer to this question. 1901, June. The Clay-Worker Volume 35/36, Unequal Expansion of Body and Glaze, written in response to the query published in the May edition of the periodical there are four short correspondence about why crazing happens. The suggestions for its remedy includes making the clay body denser, increasing the amount of silica while decreasing the amount of heavy oxides in the glaze as well as rational analysis of the raw materials used to make the clay and glaze bodies.FL A N 20 D R 09 O28- FLANDRO1901, June. The Clay-Worker Volume 35/36, A House Built of Terra Cotta, The house of Mr. John R. True, president of the Northwestern Terra Cotta Company, in Chicago Illinois was built entirely out of glazed terra cotta. 1903. The Clay-Worker-, Volume 39/40. Self-Cleaning Building. This short little article is about the use of glazed terra cotta on buildings in cities. It suggests that glazed materials provide no surface for dirt to rest upon and therefore can be washed clean by the rain. 1903, May. Transactions of the American Ceramic Society, The Development of the Matt Glaze, written by Mr. Charles Binns, of Alfred NY, this is the first written reference found pertaining to the development of matte glazes. Binns mentions that Dead Surface came into high demand among architects and to achieve this surface manufactures tried the attack of fluoric vapors, abrasion by sand-blast, an immature glaze, all with limited success. Binns ran experiments using glazes that had a high acidity, low Alumina content and cooled slowly with no success. He outlines his procedure used to determinewhat the thresholds were for producing a matte glaze using this glaze formula. 1904. The Clay-Worker, Volume 43, Terra Cotta at the Worlds Fair. This article gives a brief description of the manufactures and their displays in the Clay Industry section of the worlds fair. Of particular interest in this article is the numerous references to the creamy white glaze that was applied on several pieces as well as speckled surfaces. In addition there is reference to a piece made by the Perth Amboy Terra Cotta Company that has a dead white or sand-blasted enamel surface, as well as pieces designed by Rookwood Pottery which had a matte glaze that had been made without the aid of sandblasting (replicas of pieces installed in the 23rd and 86th street NYC Subway stations)FL A N 20 D R 09 O29- FLANDRO1904, April-December. The St. Louis Worlds Fair was held in St. Louis Missouri. The fair included an exhibition of architectural terracotta wares. 1905, April. The Clay-Worker, Terra Cotta in Landscape Architecture. This article gives a reason why sand-blasting of regular glazes to achieve a matte finish was found unacceptable. It states that glazes that have been sand-blasted have a very early and rapid accumulation of soiling that cannot be removed. 1905, June. The Clay-Worker, The Manufacture of Architectural Terra Cotta in America. This article provides images of the working spaces at the Conkling-Armstrong Terra Cotta Company (includes images of the post firing inspection area, the modeling department, the molding department and part of the stock room). In addition this article references The Wanamaker building in New York City (Broadway and 8th Street), in which the top ten floors are done entirely of terra cotta, and the trimmings of the interior court are done in a matte white glaze. This would be the first reference of a matte glaze being used on a building. It would be very interesting to see if the matte glaze employed on this building is actually a matte glaze or a glaze that has been sand-blasted or etched with Hydrofluoric Acid. 1905, July. Transactions of the American Ceramic Society, Lessons From the St. Louis Exposition, 1904. Mr. Binns, writes this piece for the American Ceramic Society, he states: Body, glaze and color are, like the poor, always with us. 1906, April. The Architectural Record, Glazed and Colored Terra-Cotta. The article is submitted through their technical department. The word bright is used in reference to glossy glazes and it states that there is a new type of glaze that comes out of the kiln with a dull finish (prior to this matte glazes were produced by sandblasting glossy glazes, and the switch was not immediate as there are later references to the use of sandblasting toachieve a dull finish). The author also references that many experiments were undertaken (no further specifications) to get colors and glazes that were not susceptible to failure caused by the gases generated during the firing, so far good methods were found to create blue, yellow green and grey. 1906, August. Transactions of the American Ceramic Society, Kilns for Burning Architectural Terra Cotta. Written by I.L. Conkling of The Conkling Armstrong Terra Cotta Co., Philadelphia, he describes different types of kilns that can be used to fire architectural terra cotta and the pros and cons of each. This document also includes three drawings of kilns being used at the time of this publication. 1907. The Clay-Worker, Volume 47/48 Wants to Buy Bella Robbia, a man fromFL A N 20 D R 09 O30- FLANDROthe Spaniard Republic of Columbia writes to the Clay-Worker to purchase the material that is fired in a furnace called Bella Robbia as the shipping for such items is too expensive. 1907. The Clay-Worker, Volume 47/48, The Corning Brick and Terra Cotta Company, in this short article the author describes the manufacturing set up at the Corning Brick and Terra Cotta Company, including the placement of the chemical (glaze) laboratory directly above the spraying room, where the glazes are mixed and then conveyed directly to the spray room. 1907, March. Transactions of the American Ceramic Society, Polychrome Glaze Decoration in Architecture. It is stated by the author that glazes are applied not by painting but by the use of a small atomizer. In this article Mr. Plusch describes a brief history of polychrome ceramics used in architecture through out the world, including in his references the Tower of Nankin and Della Robbia.1907, April. Transactions of the American Ceramic Society, Note on Terra Cotta Glazes. Mr. C.W. Parmelee, a ceramic engineering professor at the University of Illinois, wrote this article about his experiments to develop a glaze that was suitable for architectural terra cotta. Bristol glazes were being used prior to this, however, Bristol glazes were designed for stoneware and to be painted directly on the clay body. Parmelee was looking for a variation of a Bristol glaze that could be sprayed over an underslip and also fired in a muffle kiln to cone 5 (higher temperature than a stoneware kiln).1908. The Clay-Worker, Volume 49/50, The Definition and Application of Terra Cotta, in this article a listing of the earliest uses of terracotta in the United States are given (including Cooper Union).1908, August. Real Estate Record and Builders Guide, Development of Polychromatic Exterior Glaze Decoration. References the first large use of polychromatic terracotta on an exterior at the Madison Square Garden Church in New York City, designed by McKim Mead and White in 1906. 1909. The Clay-Worker Volume 51, Clay Products of the United States. In a clever diagram, originally published in the Scientific American, the amount of terra cotta produced in the United States is placed in the bottom five of the clay products produced in the United States. 1909, April. Real Estate Record and Builders Guide, How Terra-Cotta is Made. The author states that terracotta was given a required coloring before the firing because of the variances in color that occurred due to uneven firing. He goes on to say that most terracotta has a hard spray (slip) or a matte glaze finish, which would prevent moisture from permeating into the body. The slip was sprayed on using a compressed air and then followed by a spraying of the final glaze.FL A N 20 D R 09 O191031- FLANDRO1910, February. Transactions of the American Ceramic Society, Notes on the Manufacture of Enamel Brick with Some Investigations on Enamel Brick Slips. Supplied by R.T. Stull from Urbana Illinois, this fifty-four page article goes through every step of enameled brick manufacturing. He conducted various tests on 478 bricks fired at three different temperatures and noted their conditions. His research was the first broad study published concerning both fired ceramic material, slips and glazes. 1910, February. Transactions of the American Ceramic Society, Note on the Viscosity of Clay Slips as Determined by the Clark Apparatus. Written by renowned ceramic chemist of the time A.V. Beleininger initiated the concept of testing machinery to determine different mechanical and physical properties of refractory materials. His research initiated studies into the physical properties of terracotta, which would be transformed into standards required by the United States Bureau of Standards in the 1920s. The Clark machine measured the viscosity of a combination of materials. Its measurements are based on the degree to which the viscosity of water is changed by the addition of soluble salts, granular grains and mechanical motion required to produce movement of the suspension. The machine consisted of a paddle attached to a vertical axis. The speed of rotation of the paddle through the suspension was used to determine the amount of torquethat was necessary for the paddle to revolve.FL A N 20 D R 09 OFigure 5: The Clark apparatus.119111911, JanuaryApril. The Brickbuilder magazine publishes several articles on how ceramic chemists are being employed at terracotta manufacturing sites. They are employed to provide recipes for glazes, many being trained at schools in England or Germany, initially their knowledge was not immediately applicable to the field of architecture, but success was soon to come. Through their experiments advance became more rapid, and it was not very long before a fairly satisfactory bright white glaze was evolved.2 Through this we have established that experiments were being done. Published in the same issue in two different articles are references to the practice of sandblasting glossy glazes on terracotta to give it a matte look.3 The Brickbuilder also states that on the Madison Square Garden Church (1906) the polychrome glazes were painted onto the terracotta body. 4 This signifies that a spraying mechanism had yet to be developed for the application of multiple colors (although an atomizerImage from the article Note on the Viscosity of Clay Slips as Determined by the Clark Apparatus, Transactions of the American Ceramic Society, 1910. 2 Durand, A. The Brickbuilder. Architectural Terra Cotta, Volume 20, 1911. 3 Durand, A. The Brickbuilder. Architectural Terra Cotta, and The Ceramic Chemical Development of Architectural Terra Cotta by Herman A. Plusch. Volume 20, 1911. 4 The Brickbuilder. Architectural Terra Cotta, Architectural Terra Cotta: Its Physical and Structural Properties, Polychrome Terra Cotta in the Masonic Temple of Brooklyn, Polychrome Terra Cotta in the Exterior Architecture, and The Ceramic Chemical Development of Architectural Terra Cotta. All articles were published in Volume 20, 1911.132- FLANDROwas used for single color applications). 1911, December. The National Terra Cotta Society was set up, consisting of fifteen companies representatives and initiated by Walter Greer of the New York Architectural Terra Cotta Company.FL A N 20 D R 09 O2. Boil them for one and half hours. water has converted to steam. 7. Inspect the following morning for crazing and cracks.33- FLANDRO19121912, February. The American Architect, Enameled Brick. This is the first article that was discovered with testing information and steps. The author of this article was the owner of an extruded brick factory in Chicago. William P. Varney describes the two different systems for glazing a brick or terracotta; the one burn process or the two burn process. In the one burn process the wet clay brick is hand dipped into a slip, allowed to dry and then hand dipped into a glaze, then fired to maturation. In the two-burn process the brick is first fired to a bisque temperature, cooled down, hand dipped into a slip, allowed to dry, hand dipped into a glaze and then fired again to mature the glaze. He performed the following tests to determine if the one step process or the two step processes supplied better bricks. He determined that the two step-process was better. To determine if the body of the brick and the glaze had similar thermal coefficients of expansion he used the following procedures:1. Take the bricks that are to be tested and wire them together (bricks he tested were of the English size, approximately 3x4x1). 3. Remove from the water and immediately bake for one hour until or until all the 4. Remove from the oven and place them in a furnace (kiln) and heat them around 1200-1400 degrees for an hour and a half until the bricks were red hot. 5. Remove from the kiln and plunge into cold water, then gradually heat them until they are boiling. 6. Remove from water and place them outside to freeze when the thermometer reads ten degrees above zero.Freeze Thaw Test:1. Immerse brick or other materials into water for twelve or twenty-four hours 2. Without allowing them to dry out freeze them as hard as possibleFL A N 20 D R 09 OHeat Test: 1. Wire several samples together so that fair treatment is applied. 2. Heat to various degrees starting at 250 degrees 3. Plunge into cold water 4. Inspect specimens for crazing and cracking Ink Test for Discoloration:34- FLANDRO3. Repeat this test several times and inspect the specimens for crazing after the first freeze.1. A small hole should be bored entirely through the enamel until the bisque is exposed 2. Pour black or red ink into this hole, allowing it to soak in behind the enamel 3. If the ink stains show through the enamel it will tell you that the brick will discolor if used for exterior work as it will soak up dust or soot through the rear of the wall or joints.1912, March. Transactions of the American Ceramic Society, The Production of Black Spots Upon Terra Cotta Glazes, Matte Glazes, A Theory for the cause of Matteness in Glazes, The Microscopic Examination of Twelve Matte Glazes, Note on Terra Cotta Glazes. In this issue of the transactions of the American Ceramic Society the listed articles were all printed together, there had been no prior discussions about the production of matte glazes for terra cotta. It was determined that matteness was formed in two different ways the first being similar to chemical precipitation. The precipitate could either be a calcium alumnia compound or a zinc compound. The second way in which mattes are formed is caused by the presence of materials that enter into solution slowly or with difficulty. A matte glaze designed by Mr. Leroy of the Perth Amboy Terracotta Company was exam-FL A N 20 D R 09 O1. Press the terra-cotta clay in a plaster mold.Figure 6: Trial piece5ined using a microscope. Professor W.J. McCaughey of the Department of Mineralogy at Ohio State University was employed to help use the microscope and analyze the information. Both cross sections and thin sections were looked at. The surface of the matte glaze was found to be rough and granular, no crystal development was found and only 35-40% of the sample was determined to be glass. The remaining 65% was determined to be a heterogeneous mass showing aggregate polarization, resulting from immaturity of the glaze. An experiment was run on a specifically shaped piece of terracotta, using varying Bristol glaze recipes. The piece was specifically made with ridges and beveled corners to observe how well a glaze would cover raised ornamentation. The processes was as follows:2. Once the specimen is removed from the mold it is sprayed with a coat of slip, followed by a spray coat of glaze.3. The trials were then set on saggers and fired in a laboratory kiln to cone 6 in 18 hours. Once cone 6 had been achieved it was held at this temperature for a period of 3 hours and then slowly cooled. Trials were also fired in commercial kilns, both types of kiln firings produced the same results, confirming that the laboratory kiln firing could be used to determine results similar to the commercial kilns.19131913, May. The Pacific Reporter, Volume 169, published in 1918, Permanent Edition. This was a monthly publication that held all of the legal cases that had been heard and in which rehearings had been denied. Civil court case #1732 was heard on November 12, 1917, between the plaintiff Mary Neugebauer and the defendant Gladding McBean & Company. In the court report it was revealed that her husband Frank Neugebauer had been working for Gladding McBean & Co. in their architectural terracotta manufacturing depart5Image from Transactions of the American Ceramic Society, Note on Terra Cotta Glazes, 1912.35- FLANDROment starting in 1907, where prior to May 1913 he used sandblasting for the purpose of dulling the surface of the terracotta which came from kilns too bright or lustrous to match other ware ordered for certain buildings. After May 23, 1913 the company switched to painting on a diluted solution of hydrofluoric acid with a brush to accomplish the same effect. (Mr. Neugebauer died from repeated exposures to HF fumes).FL A N 20 D R 09 O1917ings, porous coatings619141914, May. The National Terra Cotta Society published their first book of standards for terracotta, by the name of Architectural Terra Cotta Standard Construction. The publication was mainly concerned with the setting of terracotta blocks. The book has no standards for glazed materials, or for testing glazed materials. It does, however, make suggestions for appropriate glaze colors, such as Field White and Iron Cream.61917. It is known that The National Terra Cotta Society begins testing terra cotta for information pertaining to its material properties.1917. Transactions of the American Ceramic Society, Volume 19. Causes for the Failure of Terra Cotta in the Wall, and Note on Pinholing and Peeling on Terra Cotta. R.L. Clare, a ceramic engineer from Ohio State, lists the reasons for terracotta failures:1. Defective body mixture-Fire/cooling cracks, dry/shrinkage cracks, low absorption, high absorption, excess soluble salts, Iron pyrites, lime pebbles or other impurities, low bonding power 2. Defective glaze or slip- Crazing, shivering, peeling, separation of vitreous coat-3. Irregular burning conditions- shrinkage cracks, under or over fired materials, fire/cooling cracks4. Building stresses- Expansion/Contraction, deflection of steel frame, movement of reinforced concrete, all causing cracked, chipped or broken pieces. Settling of foundations, imperfect backing up or filling of the terracotta, imperfect pointing of the joints resulting in bursting due to freeze thaw.The National Terra Cotta Society. Architectural Terra Cotta Standard for Construction, May 1914, page 11.36- FLANDROFL A N 20 D R 09 O19181. Properties of Terra Cotta Clays: 2. Grog37- FLANDROHewitt Wilson described problems that were occurring with some architectural terra-cotta glazes, mainly pinholing and peeling of the glaze. These defects were allowing moisture to get into the body of the block. Application failures (such as trying to apply the glaze to a dusty piece of terracotta) were ruled out. The next step for him was to try and produce pinholing. He did this through a series of different controlled trials such as soaking the grog for the clay in a saturated solution of magnesium sulphate, dusting the surface of the terracotta with magnesium sulfate, using phenolphthalein to test the alkalinity of the terracotta clay, and brushing on barium chloride or barium fluoride to the surface of the unfired terracotta body. He determined that it was the barium chloride or barium fluoride which was causing the pinholing. The addition of Gum Arabic to the slip stopped all pinholing and caused the slip to hold better to the body, however, at the time Gum Arabic increased the cost of the slip from 9.9 cents per ton of glazed terracotta to 18.8 cents.1918, May 10th- February 24, 1922. Circular No. 1, Summary of the Printed Reports 1-8 on the Technical Work of the Society (see April 1, 1927). Printed by the National Terra Cotta Society and the National Bureau of Standards. Information in this circular relates to studies done in the earliest years of the National Terra Cotta Society. The summary was compiled from various society reports to make the studies done between 1918 and 1922 easily referable. It includes information relating to the following subjects:a. The average absorption of 20 terracotta clays was found to be 7.7%. b. The average modulus of rupture is 4,176 lbs. per square inch. c. The theoretical true strength of a clay may be calculated for the purpose of comparisons using the following formula. S(1-P)2/3, where S is the crushing or transverse strength, and P is the porosity expressed as a fraction of the external value.a. No.2 fire clays ordinarily used as in terra cotta bodies is a hard dense body and the introduction of refractory grogs open up the structure. While the use of broken saggers and other refractory materials is logical it is condemned, the use of sagger grog will decrease the strength. Theuse of grog consisting of a dense burning but still refractory clay, such as a plastic pot, crucible and bond clays is recommended. The material should not be of a siliceous character and should be properly sized. At one point it was reported that the use of stoneware grog instead of fireclay grog reduced the absorption from 2.75 to .45%. b. Screen analysis of grog from various companies was found to vary widely. In connection with other grog bodies a content of 25% of materials passing through the 100 mesh sieve resulted in the greatest strength. c. As part of the requirements for grog, a low firing clay will necessitate a grog more refractory than itself, and a high firing clay one of equal or lower refractory.FL A N 20 D R 09 O3. Terra Cotta Bodies a. Strengths: c. Expansion of Bodies38- FLANDROi. Compressive strength: To be obtained by crushing a cylinder 12 inches high, 6 inches in diameter and walls 1 inch thick is 13,500 lbs. per square inch, while the lowest is 3,440 lbs. Average is 7,873 lbs.ii. Tensile strength: Average obtained was 1,298 lbs. per square inch. With the lowest at 586 lbs. and the highest at 2,187 lbs.iii. Transverse strength: The maximum, using glazed bars with the glaze up is 2,928 lbs. per square inch, the minimum being 1,045 lbs. iv. Soaking terra cotta in water and allowing complete saturation of the piece decreases the strength by approximately 33%.b. Modules of Elasticity-to be obtained by testing a cylinder 12 inches high, 6 inches in diameter and with 1 inch thick walls. i. In a general way the modulus varied with the crushing strength and inversely as the absorption. 3,500,000 pounds per square inch was used as the modulus of terra cotta bodies in calculationsi. The average expansion of a body from 17-100C is 5.6x10-6, from 17-150C is 6.14x10-6, from 17-300C is 6.68x10-6, from 17-600C is 5.95x10-6. d. Absorption and Resistance to Freezing i. It is claimed that a limiting value as related to absorption cannot be stated, as one body may possess an absorption of 15% and be durable and another have an absorption of 12% and be inferior.FL A N 20 D R 09 Oii. The rate of absorption is of no value. iii. space for the expansion of ice. e. Sodium Sulphate Artificial Freezing Tests f. Water Freezing Test 4. Strains in Terra Cotta Bodies 5. Spalling on Terra Cotta 6. Pressing39- FLANDROWe must prevent or reduce absorption, but still allow sufficienti. This test was proven to produce results similar to rapid freezing but could be done at any time or place. However, later the results were deemed too severe.i. Water was found to expand in volume when frozen at 0C. Because pressure could not be released in pieces with glaze the results were severe.ii. Freezing tests should be done on 1x1x4 samplesa. Strains ordinarily found in glasses may be found in terra cotta.a. It is doubtful that it is caused by the sun as it has been observed in subwaysa. Care should be taken when pressing that pressers do not use their thumbs as it can cause pieces to be badly laminated. Lamination will show itself upon testing with the sodium sulphate test.7. Effect of Rate of Cooling and Quenching TestsFL A N 20 D R 09 O8. Expansion of Glazes 9. Expansion of Underslips. 10. Proposed specifications a. Transverse strength: 1000-1600 lbs. per square inch b. Absorption: Maximum of 11.5 (tentatively) 11. Some sources of defects on terra cotta b. Drying- cracks or strains a. Pressing- laminations and cavities40- FLANDROa. 3 inch cubes were used for the testing, they were fired to their maturing point and then cooled at different rates. Some at 100C per hour, some at 450C per hour and others at 20C per hour. They were then subjected to sudden heating (up to 400C) and cooling (quenching in room temperature water) tests until failures were noticed. It was found that some bodies were more resistant to failure when cooled slowly and other when cooled quickly. By conducting a test similar to this at manufacturing plants, manufactures could probably improve resistance to fire cracking.a. The average coefficient of expansion was found to be 2.0 to 7.0x10-6 from 17-100C, 2.6-7.2x10-6 from 17-150C and 3.6 to 7.5x10-6 from 17-300C.a. The average coefficient of expansion was found to be 3.6 to 5.9x10-6 from 17-100C, 4.4 to 6.0x10-6 from 17-150C and 4.9to 7.8x10-6 from 17-300C.c. Resistance to Sodium Sulphate Artificial Freezing Test (undefined at this point) d. Compressive Strength- A requirement of 2,500 lbs. per square inchc. Firing- cooling cracks (often traced back to drying cracks), excessive12. In Constructionquartz content, rapid smoking, insufficient oxygenation, glaze shivering or crazeFL A N 20 D R 09 O13. Movement in Buildings 14. Joints 15. Filling of Terra Cotta 16. Concrete Constructions 17. Rust proof metals for anchoring terra cotta 18. Flashing41- FLANDROa. Overloading/insufficient protection from moisture, failure to provide expansion joints. b. The use of sheathing, flashing, waterproof cements and insulation joints (the last point is especially important when concerned with projecting glazed terra cotta, as it is subjected to greater temperature fluctuations.a. At this point there is the question as to whether it would be satisfactory to make terracotta an integral part of a cement structure. However, there is not enough data at this time.a. Joints of inch are satisfactory. The use of asphalt in joints, tar, other bituminous materials were not sufficient. Of the three elastic cements Hydro-bar was used in the Woolworth building and failed after one year, Kuhls lasted a little longer and Horns proved the best of the three.a. Concrete as a filler is still under investigation, but may cause bursting.a. Failure of terra cotta on the Commerce building due to insufficient support by the concrete fin.a. Concrete may protect the steel from rusting in certain cases. Galvanized steel is recommended for some cases or the use of a special steel containing a small amount of copper are corroded less than the same steel without copper.19. Effects of crystallization of salts 20. Crackinga. Flashing is not necessary for work supported on a wall, but its use on large cornices may be helpful.FL A N 20 D R 09 O42- FLANDROa. Sulphates of lime, magnesium and alkalies may disintegrate terra cotta. The use of a cheap reagent such as barium carbonate, in mortar are suggested to overcome this difficulty.a. A large amount of cracking has been noticed in pilasters, one remedy is the use of a vertical joint.1918, May. Journal of the American Ceramic Society. Polychrome Decoration of Terra Cotta with Soluble Metallic Salts. This journal entry provides different ways in which glazes are applied to terracotta and the problems encountered. First, painting on Bristol glazes with a brush. Bristol glazes are glaze mixtures that contain feldspar, flint, clay, whiting and zinc oxide. This method is slow and creates uneven coatings, the thicker areas of application crack and do not heal during the firing. Second, spraying with a mask (masking off the portions that are not be of the color currently being sprayed). The placing and removing of the masks would often damage the freshly sprayed area, and would leave uneven and broken lines between adjacent colors. Third, overglaze spraying, in this method a single glaze is sprayed over the whole piece and the overglaze colorant is sprayed onto the desired area and brushed off of the others. This worked well when only working with one color, additional colors required the use of masks. Forth, the use of soluble metallic salts. The entire piece is spray glazed with a clear glaze and allowed to dry. Then soluble metallic salts are dissolved in glycerine (glycerine was used in place of water because the water soaked into the clear glaze too fast and left brush marks) and painted on top of the clear glaze in the desired areas, the dry clear glaze will absorb any moisture from the glycerine and the ware is then fired to maturation in a single firing. Because the soluble salts were often colorless, the glycerine solution could be dyed so the applier could see the areas that had been covered. The problems with this application were caused by some of the chlorides and sulfates attacking the carbonates in the clear glaze, causing bubbles and blisters. To prevent this reaction from occurring, a metallic oxide had to be added to the solution to neutralize it. This method of color application was quicker than the previously mentioned applications but the soluble metallic salt solution was time consuming, had to be very precise and it was difficult to get even coloring across a sur-face.19191919, June. The Annual Report of the Director Bureau of Standards to the Secretary of Commerce, Terra Cotta Bodies and Clays. In this report it is noted that a study was currently underway to determine the porosity, volume changes, mechanical strengths (using the transverse bending test) and resistance to crystallization of sodium sulfate in various clays fired to various temperatures. The hope was to determine the minimum firing temperature that had to be achieved in order for pieces to obtain the required strength for construction. Furthermore another study specifically focused on the properties of architectural terracotta was also underway. In this study glazed terracotta that was in service on buildings was investigated and two failure patterns were pointed out. The first being glaze spalls (determined to be from insufficient brick and concrete back up allowing moisture penetration from the rear) and second the crushing of terracotta veneer due to improper methods of attachment. The laboratory testing that was completed consisted of making test pieces for adsorption, artificial freezing and transverse testing from clays used by the majority of the manufactures in the country. The specimens were fired at seven different temperatures. A report was to be compiled and published by 1921.FL A N 20 D R 09 O192043- FLANDRO1920. The Story of Terra Cotta, written by Walter Geer is published. This threehundred page book includes an invaluable history of terra-cotta production in the United States. It includes descriptions of the manufactures (separated by state and or geographic location), advantages of using terra cotta, the origins of The National Terra Cotta Society, former manufactures, and bibliographical sketches of prominent men in the business. 1920, January. The Journal of the American Ceramic Society, Some Data on the Development of Terra Cotta Glazes. E.C. Hill did a study on the addition of tin oxide, barium oxide and magnesium carbonate to Bristol glazes, these additives were commonly used in the glazing of terracotta. He made four samples of each additive and sprayed them onto commercially supplied terracotta. Two samples from each set were fired at cone 7 and the others at cone 5.5. He also did testing on just the glaze mixtures by allowing the mix-ture to dry enough to be molded into cones. These were fired in a laboratory test kiln. Using the findings from the study completed in 1917, when it was determined that barium was causing pinholing, he was hoping to find a substitute additive for Gum Arabic which was costly. He determined that Tin Oxide worked well, however its cost was also considerably higher than what was needed. 1920, June. The Journal of the American Ceramic Society, The Composition of Kiln Glasses and Their Effect on Terra Cotta Glazes and Colors. F.B. Ortman, a chemist for the Northwestern Terra Cotta Company, makes it apparent here that trying to standardize manufacturing processes of architectural terracotta is troublesome. The reason being that there are so many factors to consider when producing glazed terracotta, one way of firing may have produced good results, but when tried again it became unsuccessful. Ortman believed that gasses created in the kiln during firing caused the changes. To determine this he conducted an experiment that involved using several types of kilns (round, up draft, down draft, muffle) over a period of several months. He fired a number of terracotta samples with a sulfur free glaze to just under their maturity point, when they were removed from the kiln he would scrape off the glaze and analyze it for the presence of sulfur. Sulfur in the glaze could cause gas blisters, displace boric acid from glazes containing boric acid, or it could react with coloring oxides changing the final color of the glaze. He found that there was sulfur found in the glazes so the next step was to determine the source of the sulfur. The clay being used contain a certain amount of sulfur, most of the SO2 was distilled off of the clay at red heat and the final amounts aroundFL A N 20 D R 09 O44- FLANDRO900C. The grog also contained sulfur. The amount of sulfur in the grog was dependent on how long the grog had been sitting out and exposed to weather and how porous it was (grog being previously fired materials, over fired specimens overs, saggers, old fire brick etc.). It gained sulfur from SO3 in the rainwater. Coal also was a source of sulfur, even in kilns where combustion gases were kept separate from the kiln chamber it was almost impossible to keep all the gasses out.per five gallons of gas. A comparison of amounts of SO3 present in each type of kiln was done. The quantity was always far less in gas fired kilns. He hoped to be able to form a standard for kiln ventilation in terracotta production, which would improve the stability ofOrtman determined the amount of sulfur in the kiln atmosphere by bromine absorption. Five gallons of gasses from both the flue and the kiln chamber were drawn simultaneously, the bubble of gas was allowed to pass over a series of flasks containing bromine water forming H2SO4, which was later precipitated as BaSO4 and calculated to milligrams of SO3the glazes.19211921, January. Journal of the American Ceramic Society, The Effect of Glaze Composition on the Crazing of Terra Cotta. E.C. Hill, of the Conkling-Armstrong Terra Cotta Company in Philadelphia, PA used the following procedures to determine the relation between glaze formula and crazing: 1. He composed two different terracotta clay bodies using two different types of clay and 35% grog for each and three different glazes formulas were made (one being the control recipe).FL A N 20 D R 09 O6. Inspect for crazing.2. The testing was done on 8x8x2 pieces of terracotta with egg and dart ornamentation on the side, pressed in a plaster mold. 4. Once dry the specimens were fired in a small 6ft. test kiln to cone 5.5 in 40 hours and cooled in the same. 5. Two firings were done and three trials of each glaze were fired in each burn.3. When dry the test pieces were sprayed with the slip regularly used, and over this a heavy coat of glaze was sprayed.7. Knowing that crazing is dependent on the composition of the clay body, the glaze and the speed at which it is cooled he determined that increasing the amount of clay in the glaze formula was the best way to overcome crazing, the addition of feldspars to the glaze were less effective.19221922, March. Sprechsaal Volume 54. The Determination of the Resistance of Ceramic Products toward Solutions Especially Acid, by O. Kallauner and R. Barta (printed in English in the Journal of the American Ceramic Society May 1928). The following testing procedure is used to test the solubility of bodies in sulphuric acid: The body (do not include the glaze!) is ground to 80-100 mesh size. The sample is then washed thoroughly with distilled water to remove the dust that sticks to the particles (the45- FLANDROFL A N 20 D R 09 O[{(V1-Z)-V2}/(V1-Z)} X 10046- FLANDROfines). Any iron remaining in the sample is removed by the means of a magnet. The samples are then dried at 120C. One gram of the sample (V1) is placed in an Erlenmeyer flask and then boiled with 22cc. of concentrated H2SO4 (specific gravity 1.84) for one hour, after which the glass is air cooled for five minutes and the solutions poured off. The flask is then further cooled by dipping it in water, followed by the addition of 50cc.. The solution is then filtered and the residues in the flask washed three times with water. Fifty cubic centimeters of %5 sodium carbonated solution is added to the flask and heated on a water bath for 15 minutes, stirring every 5. The sodium carbonate solutions are then filtered off and the residues washed three times with water and then 5 drops of hydrochloric acid. The ground residues are washed free from chlorides, transferred to a crucible and weighed (V2). Separate one gram samples of the same materials are ignited and losses in weight recorded (Z). The percentage loss in weight due to sulphuric acid treatment is then calculated using the following formula:1922, August. The Journal of the American Ceramic Society, The Control of Glaze-Fit by means of Tensile Test Specimens. A study completed by F.H. Riddle and J.S. Laird of the Research Laboratories of the Champion Porcelain and Jeffery-Dewitt Insulator Companies. Previous to this study glazes formulas were selected by a systematic means of trial and error. However, the problem with this system is that it doesnt often reveal the tendency for glazes to slow craze (crazing that occurs after a specimen has been out of a kiln for quiet some time). A new method was invented to test for glaze fit by using tensile stress. It is based on the effect, which a glaze exerts on the strength of a porcelain tensile stress specimen. The test specimens were of diameters between 0.4-0.5 inches, similar to those used on testing the tensile strength of metals. Each specimen was made from the same clay, had a different glaze applied and then fired in the same kiln. For future uses of this method the researchers suggested making numerous samples of each. The glazes were applied all over the body of the clay except on the ends and then a tensile stress was applied, using the same machine that was used in cement testing (ASTM 21, p. 105056, 1921). It was found that there was a decrease in strength with an increase in crazing, caused by the strain produced on the surface of the clay due to the poor fit of the glaze.1922, August. Journal of the American Ceramic Society, Notes on Shivering of Terra Cotta. John L. Carruthers of the Denver Terra Cotta Company conducted a study on the shivering of terracotta glazes using one clay known to cause bad shivering of glaze and the other non-shivering. He notes that there seems to be some discrepancy among ceramicscientists over the causes of shivering. Carruthers uses test specimens shaped similar to those used in the 1912 experimentation. In his terra-cotta factory he had found that most glaze shivering occurred mainly on 45 bevel edges and on drove or tool finished areas. His procedure was as follows:FL A N 20 D R 09 OFigure 7: Test Specimen7 Figure 8: Test Specimens87 81. Prepare two sets of terracotta test specimens, one using the non-shivering clay and the other with the shivering clay. All specimens are made by pressing the clay into a plaster mold. 2. After their removal from the mold the pieces were finished and dried. 4. The specimens were then fired to cone 3 in a commercial terra cotta 3. The test pieces were then sprayed with three coats of a factory white matte glaze that had shivered badly in factory use.He also had microscopic examinations done on the raw clay bodies to determine their composition. Knowing the compositions of the raw clay and the results of his firing tests he was able to determine that shivering is caused by the following; the presence of highlyImage from Journal of the American Ceramic Society, Notes on Shivering of Terra Cotta. 1922. Image from the Journal of the American Ceramic Society, Notes on Shivering of Terra Cotta. 1922.47- FLANDROFL A N 20 D R 09 O19231. General Information48- FLANDROsiliceous clay in the body, the inclusion of finely ground grog of a siliceous nature or the presence of soluble salts in the clay, grog or water used for tempering. These conditions can be respectively overcome by the following; the use of feldspar, felsites or rock having similar composition to make the body denser, the use of coarse sand or grog to make the body more open, or the use of barium carbonate can eliminate shivering caused by soluble salts. Comments on this publication were very favorable, the use of felsite was of particular interest to the society, as it was common practice to change the glaze formula to fit the clay body and not visa versa. Mr. C.W. Hill (a ceramic engineer) makes a very interesting proposal. He states that it is taken for granted that shivering is caused by poor fit, differences in thermal expansion, of the glaze verses the body. He suggest that finding a correlation between the physical properties of a material and the changes in composition of a material, with the Bureau of Standards equipped to handle such a test it would seem a comparatively simple matter to secure such data. 1922, November. The Journal of the American Ceramic Society, The Use of Pulsichrome From A Manufactures Standpoint, Pulsichrometer vs. the Old Method of Applying Glazes. These two articles signify the beginning of the use of the Pulsichrometer. It was an automatic glaze material weighing machine, which was used in conjunction with a conveyor belt. It produced a specific soft finished look on the terracotta required less handling experience to perfect than an atomizer and increased the speed of production by almost 50%.1923, September. Standard Specification for the Manufacturing, Furnishing and Setting of Terra Cotta 1st Edition. Published by the National Terra Cotta Society. This was a modification to their 1914 specifications, in which many important standards were missing. It contains ninety-one standards that are to be met by all those who were members of the National Terra Cotta Society, twenty-six companies nation wide. At the time of this publication the first material quality tests were being determined by the society working in conjunction with the National Bureau of Standards. The document contains a total of eight sections:2. Material (There is a note in this portion that they will be providing specific quality tests3. Design and Structurein terms of crushing strengths, densities and elasticity in a follow-up document to be inserted into the Bureau of National Standards)FL A N 20 D R 09 O6. General Conditions for Manufactures 7. Materials and Workmanship 8. Suggestions for Corollary Clauses4. Transportation Storage and Protection5. Erection (specification 49 allows for the cleaning down of the terracotta using 1.5 pints of muriatic acid to a gallon of water, post construction completion).19241924, February. The Journal of the American Ceramic Society, Papers and Discussions: The Pulsichrometer. Francis A.H. Schepers of The Advance Terra Cotta Company in Illinois discusses his factories experience and results when using the pulsichrometer. Pieces produced using the Pulsichrometer were examined under the microscope because the glaze appeared to be very thin. However, it was found that the surface of the terra-cotta body was thoroughly covered.1924, May. The Journal of the American Ceramic Society, An Apparatus for Measuring the Abrasive Hardness of Glazes. Submitted to the society by W.J. Scott from the National Bureau of Standards, it describes a machine for testing abrasive hardness of glazes. The machine is standardized so the results are also standardized. The hardness of glazes was to be determined so that further research into the causes/ control of glaze hardness could be figured out. It was found that the use of a stream of falling sand gave very consistent results on material of the same kind, and, when several brands were tested, the loss in weight seemed to be a satisfactory index of the relative resistance to abrasion. Standard Ottawa sand is used for the abrasive, it has to be able to pass through a 20-Mesh but remain on top of a 30-Mesh. The sand is funneled through two pieces of sheet metal set at a 60 angle to each other, the width of the opening through which the sand falls can be adjusted by screws. The sand drops 4.5 feet to the surface of the test piece, it is allowed to fall onto the piece at a rate of 64 pounds in ten minutes, ten minutes is the standard test time. Losses of glaze are measured in milligrams.49- FLANDROFL A N 20 D R 09 ORecipe 1 Powder: 75% calcined zinc oxide by weight, 25% potters flint by weight Liquid: 75% by volume of zinc chloride solution (specific gravity=1.82), 25% by volume of barium chloride solution (sp.gr.=1.27) Recipe 2 Powder: Same as recipe 150- FLANDRO1924, May. The Journal of the American Ceramic Society, The Investigation of Terra Cotta Work at the Bureau of Standards. Eckardt V. Eskeen of the New Jersey Terra Cotta Company was asked by the society to give a brief description of the work currently being undertaken by the Bureau of Standards. He lists reasons why terra-cotta production must be standardized, including the introduction of high-rise construction, the use of concrete and the adaptation of materials to different types of environments. Eskeen also writes that extended tests are being undertaken in the areas of strength, breakage point, porosity (and its coefficient in regards to glazes) and freeze thaw. The best part of his writing is his ability to predict that there will, perhaps, never be any final result or conclusion as to the question of whether a hard body is better than a soft, or a material burned to a very high degree of heat is more durable than a somewhat lower burned material. 1924, July. The Journal of the American Ceramic Society, Address on Terra Cotta. William H. Powell president of the Atlantic Terra Cotta Company, draws attention to the 1923 publication of standards by the National Terra Cotta Society. He wants ceramic engineers in the field of terra cotta to pay particular attention to the official definition of terms and to adopt them to secure uniformity in the published literature. He also mentions that there is a need for the development of a satisfactory patching cement which could be used with equal facility at the factory or the job when necessary and which would retains its color and strength. This is the first mention the use of cementitious terracotta repairs. 1924, September. (See April 1, 1927) December 1925. Circular No. 3, Tentative Recommendations for Producing Patching Cements (see April 1, 1927). Printed by the National Terra Cotta Society and the National Bureau of Standards. This report includes four different recipes for the creation of patching cements specifically for terra cotta repairs. The recipes are as follows:Liquid: Same as recipe 1, but replace the barium chloride solution with 25% by volume of magnesium chloride solution (sp.gr.=1.82) Recipe 3 Powder: 70% calcined zinc oxide by weight, 30% potters flint by weight Liquid: 100% by volume of zinc chloride solution (sp.gr.=1.82) Recipe 4FL A N 20 D R 09 OThrough a 100 mesh sieve Liquid: Magnesium chloride solution of sp.gr. 1.3151- FLANDROPowder: 30% by weight of caustic magnesite, 70% by weight of flintIt goes on to make further suggestions to produce high quality patching mortars: use pure raw materials, do not change the raw materials, conduct frequent test of the raw materials, do not add too much filler and store the cement in an air-tight vessel.1924, September. (See April 1, 1927) December 1925. Circular No. 4, Tentative Recommendations for Improving Certain Qualities of Terra Cotta by Better Manufacture. Printed by the National Terra Cotta Society and the National Bureau of Standards. This piece contains a flow chart of possibilities for improving certain defects or flaws with terra cotta. Subjects include: To decrease the absorption of bodies, to prevent cracking on corners, to prevent fire cracking, to prevent cracking of end pediment pieces between two smaller units, to prevent cracking and shearing of projecting units, to prevent cracking of rusticated pieces, to prevent scum cracks on symmetrical shapes, to prevent crawling in crevices, to prevent crawling on surfaces, to prevent crow footing, to prevent crazing on un-weathered terra cotta, to prevent laminations and cavities in terra cotta, to prevent peeling on glazed terra cotta, to strengthen plaster molds, to reduce popping due to pyrites and to reduce spalling on free standing terra cotta. 1924, September. (See April 1, 1927). December 1925. Circular No. 4, Tentative Recommendations for Improving the Serviceability of Terra Cotta by Better Construction. Printed by the National Terra Cotta Society and the National Bureau of Standards. This includes a flow chart of possibilities for improvements of certain conditions present in construction. Subjects include: To reduce peeling of glazes, to maintain strength of terra cottapatched with cements, to reduce deflects on corners, to reduce cracking on pediment pieces wedged between two smaller units.19251925, June. Department of Commerce Technical News Bulletin of the Bureau of Standards, Architectural Terra-Cotta Investigations. It was written here that the National Terra Cotta Society had been working for several years with the support of the National Bureau of Standards to determine the qualities of architectural terra cotta that would give the best performance in buildings. The article states that the physical properties have been determined, including compressive strength, resistance to freezing and the coefficient of expansion. The expansions of glaze, body and slips were determined using the interferometer method. Crazing was determined to be cause by the glaze having a greater coefficient of expansion than the body. In addition to the laboratory studies, terracotta that had been in service on buildings for over 30 years was examined on more than 535 buildings in several cities east of Kansas City, Missouri. Keeping water out of the terracotta proved to be crucial to its performance, and cements used in the joints connecting two pieces of terracotta that contained zincoxychloride proved to be the most promising. To further the possibility of standardization, fourteen terracotta manufactures were visited and the good and bad practices were pointed out.FL A N 20 D R 09 O52- FLANDRO1925, November. The Journal of the American Ceramic Society, Comparison of Methods Used in Estimating the Maturing of Terra Cotta. An experiment undertaken by Louis Anderson at the American Terra Cotta Company to determine which brand of pyrometric cones was most accurate or if the old method of red slip trial was better. Pyrometric cone packs of the Orton brand and the Mayer brand were used along with two different recipes of red slip (one with manganese). The red slip trials consisted of bars of terracotta clay with half of the bar dipped into a red slip. The red slip changes color as the heat surrounding it increases. It was found that at lower temperatures the cone packs were often incorrect and unreliable, with some cones falling and others not. However, at higher temperatures the readings were correct. The red slip made with manganese was determined to be the best predictor of heat at lower temperatures. The experiment was conducted so that the most accurate results could be used to regulate the evenness of heat in the kiln.19261926, January. The Journal of the American Ceramic Society, The Use of Over Glazes for Polychrome Terra Cotta. A. Lee Bennett of The Gladding McBean Company reveals that their company has recently started using overglazes to produce their polychrome terracotta. The overglaze consisted of very little clay materials and therefore was mainly a mixture of ingredients that would fuse and vitrify on the surface. It was sprayed on over polychrome glazes at a psi of 7-10, it could be clear or colored. If the spray got on surfaces where it was unwanted it could be easily brushed off leaving the under glaze in place. The problems with overglazes were that they were usually translucent, which was not always wanted. 1926, February. The Journal of the American Ceramic Society, Monograph and Bibliography on Terra Cotta. Compiled by Hewitt Wilson, a professor in the Ceramic Engineering department at the University of Washington and the current treasurer for the society. This document is priceless when doing terracotta research. I believe it to be more useful and informative than any standard specifications published up to this point by the National Terra Cotta Society or by others. It covers all parts of terracotta manufacturing from the development of plants and sources of clays to the jointing mortars and concrete needed to set it. In reference to the fit of glazes on a terracotta body Wilson lists the five top defects the definition and their causes: 1. Thin glaze- too thin a coating, caused by spraying troubles, too much water in the dispersion, or low viscosity of the glaze.FL A N 20 D R 09 O53- FLANDRO2. Glaze cracking- Crawling of the glaze from small cracks in the body or underslip, indicative of surface tension activity after fusion. The initial cracks can occur in the body, under slip or glaze.3. Glaze peeling- Although synonymous in some writings with shivering, we limit peeling to the separation of the glaze, or underslip and glaze, from the body when the surface of the body is not broken. It can be caused by a variety of occurrences such as soluble salts present on the surface of the body, insufficient vitrification or too high drying or firing shrinkage.4. Crazing- cracking of the glaze after solidification on cooling, caused by excessive tension. The cracks may be (1) only in the upper surface of the glaze,FL A N 20 D R 09 O54- FLANDRO5. Shivering- A separation of the glaze from the body when portions of the body are removed with the broken piece of glaze. This condition is caused by greater compressive strength of the glaze in unison with weaker tensile strength of the body.(2) throughout the thickness of the glaze or (3) may continue into the body and weaken it. Often crazing cracks occur at right angles or in chicken wire patterns1926, April. Department of Commerce Technical News Bulletin of the Bureau of Standards. New Method for Measuring Thermal Dilation of Glazed Ware. The bulletin states that one of the main problems with glazed ceramics used in the building trades is the cracking of glazes due to different coefficients of thermal expansion between the glaze and the surface below it (body or slip), creating stress in the glaze. In addition to this it was discovered that the stresses introduced into the ware by different rate of contraction between body and glaze during the last cooling in the furnace are enough to rupture the glaze at once or to reduce its resistance to subsequent stresses. As far as the apparatus or methods used to measure these differences in expansion none is listed in this article, it just states: The bureau has developed a method for measuring this property of a glaze as it exists under the conditions of manufacture. Later in my research this machine was determined to be the interferometer.1926, June. The Journal of the American Ceramic Society, Interferometer Measurements of Thermal Dilation of Glazed Ware. Published with permission from the Director of The National Bureau of Standards, written by G.E. Merritt and C.G. Peters (both members of the National Terra Cotta Society). The operation of the interfermetric method is fully described in the Bureau of Standards, Scientific Papers Numbers 393 and 485. The thermal expansion of specimens can be measured with great accuracy in temperature intervals ranging from 20-1000C. In this study six different methods of preparing samples were done, however, all terra-cotta samples were taken from already finished pieces. 1926, June 10th - The 12th Report Relating to the Technical Work of National Terra Cotta Society. This report was printed for members of the society only. The following are key points in the report. 1. There was an update given by Mr. H.G. Schurecht about the outdoor service tests being conducted at the National Bureau of Standards. He states that the more vitreous clay bodies are standing up the best. In addition to this bodies2. Working with these results the Bureau had just started another test where several terra-cotta bodies were impregnated with waterproofing materials; using sodium silicate, china wood oil, castile soap and alum, sodium silicate and magnesium fluo-silicate, and another with sodium silicate and calcium chloride.that had been saturated with paraffin wax were holding up well.FL A N 20 D R 09 O55- FLANDRO3. The next item mentioned was the effect of different forms of silica on the coefficient of expansion of clays. So far it had been determined that using a 200-mesh sieve was not an appropriate way to measure silica in a clay body and that the silica content is measured in terms of the calcined weight. It was found that the addition of crystalline silica to a clay caused a rapid expansion in clays because of the alpha-beta conversion of quartz (rapid expansion could lead to crazing in the glaze). The same test was done on clays containing colloidal silica and a slower expansion was achieved. The clays of all the manufactures in the society were tested for silica content. Future tests would determine the nature of the silica present.4. As well as these tests, the use of patching cements on terracotta test was tried, in hopes to find a good one. The ordinary zinc oxychlorides when calcines were used with concentrated chloride set too fast. He added borax in hopes to lower the set time, but it was not sufficient. However, the zinc-oxychloride cements could be colored well with coloring minerals. 5. Schurecht also reported on a visual survey he completed of glazed terracotta. He found that terracotta in wet parts of the country (those with high annual average humidity and precipitation) had much more crazing than those in areas with lower annual humidity. The field observations determined that the wetness of the climate had more to do with crazing than freeze thaw. 6. Along with his visual survey he took samples of glazed terracotta that all had crazing from various manufactures in the United States and tested them to determine the coefficient of thermal expansions of the body, underglaze and glaze using an interferometer. Much to his curiosity, he found that the expansion of the clay bodies in most of the samples was greater than that of the glaze, meaning the glaze would be in compression, but that the glazes had crazed which is a reaction to tension. So more studies needed to be done. 7. In conclusion all of the charted and numerical data is included in the back ofthis document. 1926, November 5th .- The 13th Report Relating to the Technical Work of National Terra Cotta Society. This report was printed for members of the society only. The following are key points in the report:FL A N 20 D R 09 O56- FLANDRO1. The two fellows to the Bureau of Standards are Mr. Schurecht and Mr. Klinefelter. As presented in the report in June of the same year Mr. Schurecht had undertaken a visual survey of in service terracotta. He had extended his survey to 1260 buildings across the country and found that of the 813 glazed terracotta buildings 300 of them had chicken-wire crazing (36.9%), and 144 had cracking (17.7%). 2. Preliminary data on the influence of waterproofing parapet backings and joints above terracotta work was returned to the society. Asphalt paint (the question remains as to what paint they are referring to as it was noted it was red in color) was used as the waterproofing agent. The data retrieved from 41 buildings with this treatment proved that the waterproofing paint made spalling and crazing more frequent than on those buildings without waterproofing paint. The conclusion was that it was holding moisture in behind the terracotta.3. Previously, there had been some questions as to the severity of the freeze thaw tests experimental procedure that was being employed at the Bureau of Standards. The procedure included soaking the entire specimen in water for 12 hours before being frozen. The Bureau decided to redo the test using cylinders of terracotta placed upright in less than inch of water for 12 hours before the freeze thaw cycles were started. The samples all failed eventually as they had before (spalled or cracked) but with the submersion failed sooner than those done by capillary uptake. 4. As per the discussion in the 12th report, Mr. Schurect had the preliminary results from his test of impregnating terracotta with water proofing agents. He submerged the glazed terra cotta pieces in the mixture being tested for 24 hours then started subjecting the samples to outdoor weather conditions for seven months (starting in March) in Washington DC., and comparing the damages to untreated terracotta. All of the treated blocks had less spalling than the untreated blocks, those treated with sodium silicate alone or sodium silicate and calcium chloride showed the most promising results.FL A N 20 D R 09 O57- FLANDRO5. A study was undertaken to see if there was a correlation between grog size, the amount of grog and water absorption. It was found that bodies containing 40-80 mesh grog have the lowest absorption factor for the bodies containing between 25-45% grog; but, when you get to bodies with 55-77% grog, it seems like those containing 80-100% mesh grog have the lowest absorption factor.6. It was also asked by the Bureau to test using slag in place of grog, as it had been conjectured that slag would make terracotta bodies be more resistance to freezing and thawing, thereby reducing spalling. They tried three different types of slag; granular, glassy and crystalline; all available from St. Louis, Missouri. Up to this point their study on this topic had only progressed to the point of mixing up clays. 7. The final test the society was updated on was one of trying to determine the reason for permanent expansion after a piece of terracotta had weathered for some time. Their controls were pieces of terracotta that had been store in the open air for three years. The difficulty was finding a test that produced the same results as weathering on a building had. The first test consisted of boiling a specimen for four hours and allowing it remain in water for forty-eight hours, this was unsuccessful, as the amount of expansion was far less than an in service piece of terracotta experienced. The second experiment was run in an autoclave at 85 pounds, for one hour, which still did not produce the same amount of expansion as the air stored specimens. The next test done was using the autoclave at 150 pounds, which was also unsuccessful. It was decided that the continual wetting and drying below 100C, and not to one-time submersions must cause the permanent expansion. Portions of the specimens that were treated in the autoclave at 150 pounds were then subjected to an ink test. The sections of the specimens that had been in direct contact with the water in the autoclave absorbed more ink than the other portions.1926, November. The Journal of the American Ceramic Society, Some Fundamentals of Terra Cotta. Harry Spurrier, the ceramic engineer for the Northwestern Terracotta Company (a member of the National Terra Cotta Society) had been asked to make certain tests on terra cotta with a specific objective in mind, (it is believed that he was part of the group working in conjunction with the National Bureau of Standards). He undertook a microscopic survey of a number of terra cotta bodies.The first thing he noted was voids. The voids were not those of the material not being packed tightly into a mold, but rather those caused by the mixing of the clay body. He found that the number of voids increased when there were coarse grog particles present, and that the movement of these coarse particles through a body left a trail of voids. Secondly, it was noted that the grog would either separate from the body of the clay or it could flux with the body. These occurrences were both caused from the differences in volume changes at high temperatures between the body and the grog. The areas where the grog had fluxed with the body and had subsequently been pulled apart upon cooling caused cooling cracks to appear. Third he made thin sections, the samples that he took were never hammered or broken out of the body, but rather sawed out using a metal disk and a mixture of 120-mesh Carborundum and oil. He noted that different bodies made different sounds whilst cutting and soft and porous bodies took longer to cut than the denser bodies. The sections were cemented onto a slide and then ground down to one or two thousandths greater than the slide. The section was then removed from the slide with the aid of hot turpentine and a section lifter, followed by being cleaned of the Carborundum and oil. It was washed in ether and dried. He noted that if there were any internal strain in the thin slice of terracotta it would curl, he hoped to prove in later tests a connection between porosity and the flexibility of a specimen. Fourth, he studied osmosis as a cause of terracotta breakdown. As with any porous material if water is introduced into terracotta and then localized heat causes localize evaporation (possibly the sun hitting the glazed surface) and according to the laws of osmosis a flow develops towards this area and away from others, creating osmotic pressure that in turn could cause a rupture. Through his laboratory test (see photo below) he determined that this was not the case with terracotta.FL A N 20 D R 09 O58- FLANDROFL A N 20 D R 09 O19279Figure 9:Apparatus used to determine Osmosis as a mode of terracotta deterioration.9Fifth, he considered that there might be an internal structure change due to freezing and thawing long before spalling on the exterior occurred. This was confirmed by using a similar apparatus as designed for the osmosis test but filling the glass tube with distilled water to measure the rate of absorption by the terracotta body (millimeters/second). The body was then submerged in water and frozen to -10C, thawed, dried and then retested. Upon the retest the results were such that the absorption of water increased more than threefold. This increase continued with subsequent freezing and thawing. To confirm this he had terracotta cubes made (10X10X10cm) and performed porosity tests (method not described) after each freezing and thawing cycle and found that indeed the porosity had increased %1.25 after each freeze.1927. Terra Cotta Standard Construction by the National Terra Cotta Society, revised edition is published. 1927, February. The Journal of the American Ceramic Society, A Preliminary Study of the Resistance to Abrasion of Ceramic Glazes, Its Control and Methods Used of Determination. Published by permission of the Director of the Bureau of Standards. A.C.Image from The Journal of the American Ceramic Society, Some Fundamentals of Terra Cotta. 1926.59- FLANDROHarrison, working as a ceramic engineer for the Bureau of Standards tested two different methods of abrasion. The first apparatus used was the Glarimeter, which was an instrument created initially for the use of measuring the gloss of papers, but the results when used on ceramic surfaces was found to be satisfactory. Gloss readings were done on specimens both before and after abrasion tests were completed. The apparatus described in the May 1924 article entitled, An Apparatus for Measuring the Abrasive Hardness of Glazes, was also used here but slightly modified so that the fall of the sand was reduced to 40 inches and the funnel diameter was reduced from .4 inch in length to .24. Slowing the rate of the sand so gloss measurements could be taken throughout the duration of the test. The alternative test method used for determining abrasion resistance was one of visual comparison to a set of standards. The standards used were prepared by abrading five samples of glazed ware so that the first was only slightly pitted and the following more pitted and so forth until the last sample, which was all but destroyed. It was thought that the ease of use of this method could be useful for manufacturing control. Both abrasive tests were completed on a total of 176 samples, and the following conclusions were determined: The alternative test method was sufficiently accurate for determining glaze resistance, the resistance of well-matured glazes is inversely proportional to the thickness, the resistance decreases as the tendency for a glaze to matte increases, abrasive resistance is markedly higher when the firing temperature is increased, glaze composition and firing temperature effect the abrasive resistance of a glaze more than the type of clay body used. 1927, March 30th Technical News Bulletin of the Bureau of Standards (April, 1927). Meeting of Advisory Committee on Ceramics published by the Department of Commerce Bureau of Standards. This short entry gives a list of attendees and also a list of topics covered. 1927, April 1st . Summary and Index of Reports One to Eight, Relating to the Technical Work of National Terra Cotta Society. This report was printed for members of the society only, but was compiled by The National Terra Cotta Society Fellowship at the National Bureau of Standards in Washington DC. This index is important in terra-cotta research because the full reports of the society from its formation up to 1926 are no longer in existence, having been lost to father time. The printed works here also includes an Appendix with tables and figures. Information included from this document has been placedFL A N 20 D R 09 O60- FLANDROin its proper chronological order. Contents: Circular No.1: May 10, 1918- February 24, 1922 Summary of the Printed Reports 1-8 on Technical Work of the society.FL A N 20 D R 09 OCircular No.2: May 10, 1918- February 24, 1922 Index of the Technical Printed Reports 1-8 Circular No. 3: Monthly Progress Reports Circular No. 4: ly Progress Reports ra Cotta by Better Manufacture. Circular No. 5: Monthly Progress Reports. Terra Cotta by Better Construction Circular No. 6: - September 1924- December 1925, inclusive.61- FLANDROSeptember 1924-December 1925, inclusive. Taken from theTentative Recommendations for Producing Patching CementsSeptember 1945- December 1925, inclusive. Taken from the Month-Tentative Recommendations for Improving Certain Qualities of Ter-September 1924- December 1925, inclusive. Taken from theTentative Recommendations for Improving the Serviceability of- Index of the Monthly Progress Reports 47-62 1927, June 2nd. The 14th Report Relating to the Technical Work of National Terra Cotta Society. This report was printed for members of the society only and included topics presented and discussed at the Semi-Annual Meeting of the National Terra Cotta Society held in Schenectady, NY from June 2nd to June 4th, 1927. The report was compiled by one of the fellows to the National Bureau of Standards, Mr. H.G. Schurecht. The following are key points in the report. 1. Factors causing spallingFL A N 20 D R 09 O2. Factors influencing the resistance of terra cotta to spalling 3. Properties of special bodies62- FLANDROa. The inspection of terra cotta on buildings in different climates revealed that spalling does not occur in non-freezing climates. The hemispherefreezing test was found to be quite accurate in predicting if spalling would occur on a specimen. This test is used to test the resistance of terra cotta to freezing. Hemispheres are immersed in water (without boiling) for 12 hours. They are then placed in a shallow pan with inch of water and placed in the freezer and are frozen at -20F during the next 12 hours. They are then removed and thawed by placing their convex sides over 2 inch circular openings o a steam bath for 20 minutes. This procedure is repeated until the specimen fails and if failure has not occurred after 160 cycles it is removed from the test. b. Freezing two terra cotta bodies saturated with water 25 times decreased the strength by 13.5% c. Terra cotta samples were subjected to accelerated weather by inserting them into a steam pressure autoclave at 150 lbs. for 2 hours.a. A terra cotta body containing 40 to 80 mesh grog was found to be the most resistant to freezing. While those with smaller grog sizes were less resistant.a. It was found that in order to obtain the maximum strength in a terra cotta body, it was necessary to use clays and grogs with similar coefficients of expansions.4. The effect of slag on terra cotta bodiesb. Substituting a vitreous grog for a more porous grog increased the strength of the clay in every trialFL A N 20 D R 09 O1. Peeling of glazed terra cotta63- FLANDRO5. The solubility of terra cotta and other ceramic bodies in sulphuric acida. It was determined that blast furnace slags increase the resistance of terra cotta bodies to freezing (and hence their resistance to spalling). However, large particles of slag cause the glaze to pit. The slag also had to be free of iron particles (passed over a magnetic field).a. This test was undertaken to determine the amount of vitrification a terra cotta body had achieved during firing. Bodies having low solubility in sulphuric acid are in general the most resistant to that type of crazing that is caused by weathering.1927, September 27th. The Technical News Bulletin of the National Bureau of Standards, Fundamental Study of Glaze Fit. This notice states that the Columbus branch of the bureau of standards had been conducting an investigation on the effects of various oxides in the tensile strength, modulus of elasticity, and coefficient of thermal expansion of glazes. At the point of this publication, 85 glazes had been tested. The study also was pursuing an answer to the reason for the crazing of glazes several months after they had been removed from the kiln. It had been determined that expansion of the body occurred due to the re-hydrating of the body.6. Further recommendations for the improvement of terra cotta practices.1927, November 15th. The 15th Report Relating to the Technical Work of the National Terra Cotta Society. This report was printed for members of the society only, and included works presented at the semi-annual meeting of the National Terra Cotta Society held in Chicago Illinois from November 15-17th, 1927. The fellow to the National Bureauof Standards, Mr. H.G. Schurecht, compiled it. The following are key points in the report:a. An examination of buildings revealed that peeling does not take place on terra cotta in non-freezing environments, similar to spalling. b. Peeling rarely occurred on bodies without an under slip, it is recom-mended to do away with the under slip, although this could cause more crawling of the glaze, which would have to be rectified. c. Repairs of peeled terra cotta- because Mr. Schurecht was frequently asked about ways in which to repair peeling glazes he undertook a small study on the repair of such. He applied different coatings to 4x4x1 cubes of unglazed terra cotta and exposed them to the weather for four months. He tested Ripolin gloss white by Glidden & Co.,Vitralite by Pratt & Lambert, Valspar by Valentine & Co., Duco brush-white by E.I. DuPont DeNemours & Co., Duco Green by E.I. DuPint DeNemours & Co., white zinc lead paint, zinc oxychloride cement (applied 1/8 inch thick, after the surface had been roughened) and Nitro cellulose solutions by Van Schaack Brothers Chemical works. The best results were obtained using the Vitralite and the white zinc lead paint.FL A N 20 D R 09 O2. Spalling of free standing terra cottaFigure 10: Sample types used.1010a. In an outdoor service test the vitreous stoneware bodies were found to be resisting weathering the best, however all of the bodies had developed crazing on their glazes. This test was conducted on several specimens that were set outside on June 15, 1923 and retrieved on September 15, 1927. The specimens were inspected ever 30 days. As well as testing different terra cottaImages from The 15th Report Relating to the Technical Work of the National Terra Cotta Society, 1927, page 18.64- FLANDROFL A N 20 D R 09 OFigure 11: Set employed to test for size changes in terra cotta when in contact with water.65- FLANDRO3. The influence of water treatments on changes in size of terra cotta bodiesbodies and glazes, he had some pieces saturated with paraffin wax. The wax saturated trials were the most resistant to weathering.a. When terra cotta was wetted for periods less than an hour, it remained the same size. b. Newly fired specimens when placed in contact with water expanded every time it contacted water for a period between 3 and 5 days after it had been removed from the kiln. c. A new method for testing a glazes resistance to crazing was used: Specimens were placed in an autoclave in about 1 inch of water, the autoclave is then heated so it will produce a steam pressure of 100-125 pounds per square inch for one hour. The trials are then removed and tested for crazing using the ink dye method. The autoclave can be made from ordinary parts found at the hardware store, enabling members to test their own glazes.FL A N 20 D R 09 OFigure 12: Autoclave used to test specimens for crazing.114. Effect of different types of grog on the strength of terra cottad. Open burning pure clay will increase in size after weathering, where as tight burning pure clay will remain the same size. It was determined that this was caused by crystalline silica in the bodies, the more silica the bigger the increase in size was.a. In contradiction to the previous report (14th) it was found that using a grog that is more vitreous than the bond clay produced better strength, differing from the current practice (which was using porous grog with vitreous clay which produced weak bodies).Figure 13: Sketches of different types of body separation from grog and types of fractures found in strength tests employing terra cotta.12 Images from The 15th Report Relating to the Technical Work of the National Terra Cotta Society, 1927, page 48. 12 Images from The 15th Report Relating to the Technical Work of the National Terra Cotta Society, 1927,1166- FLANDRO5. Crazing of terra cottaa. Three types of crazing were determined: i. Crazing upon cooling ii. Crazing due to weatheringFL A N 20 D R 09 Oiii. Crazing due to pressure Cb-Cg+Es-Eb+Mb-Ms=D temperature of the glaze down to 20C Mb= Modulus of elasticity of body Ms= Modulus of elasticity of surface D= Crazing factor Cb-Cg+Es-Eb=Dpage 65 and 71.67- FLANDROb. The following empirical formula was developed to determine the tendency for these three types of crazing to occur:This formula is to be used for specimens subjected to weathering in an autoclaveCb= Percent contraction of body from the solidification temperature of the glaze down to 20C Cg= Percent contraction of glaze from the solidification Es= Percent expansion of surface due to weatheringEb= Percent expansion of body due to weatheringThis formula is to be used for specimens not weathered in an autoclave1927, December. National Bureau of Standards Technical News Bulletin No.128, Fundamental Study of Glaze Fit, conducted at the government laboratories in the Columbus branch the study was completed in hopes to determine the effect of various oxides on the tensile strength, modulus of elasticity and coefficientof expansion of glazes. The Fizeau-Pulfrich (interferometer) method was used to measure changes in size.19281928, February. Journal of the American Ceramic Society, Methods for Testing Crazing of Glazes Caused by Increases in Size of Ceramic Bodies, presented at the annual meeting of the American Ceramic Society and published in the journal in May 1928, presented by H.G. Schurect of the National Bureau of Standards and the National Terra Cotta Society. In this document he states his new method of testing for crazing by use of the autoclave, also he determined that bodies that have the least amount of tendency to craze due to weathering also have a low porosity and are not very soluble in sulphuric acid (using the O. Kallauner and R Barta Methodssee March 1922). 1928, March. National Bureau of Standards Technical News Bulletin No. 131, Cleaners for Terra Cotta, Artificial Vitrifying Agents for Ceramic Bodies, and Crazing of Glazes Caused by Permanent Increases in the Size of Ceramic Bodies. In connection with the bureaus research into architectural terra cotta properties they also tried out five different cleaners. Sodium hydrosulphate, which cleaned well and did not corrode the glaze, Fluosilicic acid, which cleaned well but attacked the glaze a little, trisodium phosphate, which was found to be good on unglazed terra cotta but not on glazed, soap powder, which had the tendency to roughen the surface of glazes and hydrochloric acid which etched the glaze badly and therefore its use is strongly condemned. The bureau experimented with using different types and combinations of fluxes t to create a more vitreous ceramic body at lower temperatures. It was found that compositions containing 10% frits were the most successful. The next test planned was to determine the effects these have on the physical strengths of the body. 1928, May 10th. The 16th Report Relating to the Technical Work of the National Terra Cotta Society. This report was printed for members of the society only, and included works presented at the semi-annual meeting of the National Terra Cotta Society held in Chicago Illinois from May 9-10th, 1928. The fellow to the National Bureau of Standards, Mr. H.G. Schurecht, compiled it. The following are key points in the report: 1. The comparison of some properties of limestone with those of terra cotta (comparisons made to the recent publication of standards for building limestones used in the United States, Standards Technologies Paper No. 349, 1927).FL A N 20 D R 09 O68- FLANDROa. Compressive strength- that of limestone is slightly higher than terra cotta b. Transverse strength- the average transverse strength of terra cotta is much higher than that of limestone c. Tensile strength- The average tensile strength of terra cotta is 1.298 pounds per square inch, much higher than that of limestone at 446-500 pounds per square inch d. Modulus of Elasticity- That of limestone is higher than terra cotta (it being undesirable in terra cotta as it causes crazing) e. Adsorption- the average adsorption of terra cotta is much greater than that of limestone f. Coefficient of expansion- the average coefficients of expansions for terra cotta and limestone are about the same. g. Permeability- Terra cotta without a glaze is much more permeable than limestone, with a glaze it is much less (on that surface) h. Decrease in strength due to wetting- The average decrease in strength due to soaking in water for 48 hours of terra cotta is 39.6%, where as with limestone soaking in water for two weeks it is only 10.2% i. Solubility in carbonated water- limestone is soluble in rain water containing soluble carbon dioxide. Terra cotta is not.FL A N 20 D R 09 O2. Serviceability of terra cotta on buildings 3. Methods for preventing spalling69- FLANDROa. As per the information provided in the 15th report, terra cotta glazes have a stronger tendency to peel in wet climates compared to dry climates. b. As per the information provided in the 11th report, terra cotta glazes have a tendency to craze in both freezing and non-freezing environmentsa. In order to create a body that will develop less than .03% expansion in the autoclave test it is necessary to select a body having a free silica content less than 25% and an absorption less than 6-8%b. Present day terra cottas were found to have free silica contents ranging from 18-42% with absorptions ranging from 5-16.8% c. The addition of slag to a terra cotta body will increase its resistance to freezing.FL A N 20 D R 09 O5. The hydration of terra cotta bodiesFigure 14: Ring test specimens.13134. Methods for preventing strains in terra cottaa. The grog and clay should have similar coefficients of expansion, or the clay slightly higher than the grog b. The absorption of the clay should be the same, or the clay slightly higher c. The strengths of the clay and grog should be the same, or the grog slightly higher d. Strain tests are currently being undertaken on glazed specimens to determine the amount of strain an ill fitting glaze can cause on a body.a. The ignition losses of terra cotta bodies above 110C after subjecting the same sample to the autoclave test, may be used as a rough estimate of its tendency to expand due to weathering b. The autoclave test was standardized: Terra cotta should withstand a stemImage from The 16th Report Relating to the Technical Work of the National Terra Cotta Society. 1928 page 35.70- FLANDROpressure of at least 150 pounds per square inch for one hour without crazing. c. In general those bodies, which are most resistant to expansion by hydration, also have low absorptions and low solubilitys in sulfuric acid.FL A N 20 D R 09 O7. New Experimental works 1. The serviceability of terra cotta on buildings 2. Some causes and methods of preventing spalling71- FLANDRO6. Influence of slag and feldspar in workabilitya. When blast furnace slag is used as a grog it may cause pitting and discoloration of some glazes.a. The use of Portland cement clinker as grog in a terra cotta body. It has been observed that the use of Portland cement clinker when ground until it passes through a 10, 60 and 100 mesh, acts as a flux in the body, its inclusion did prevent glazes from crazing in the autoclave test, but also caused the body to absorb much of the glaze material during firing. Overall the inclusion of Portland cement clinker lowered the absorption of the terra cotta body.1928, December. The 17th Report Relating to the Technical Work of the National Terra Cotta Society. This report was printed for members of the society only. The National Terra Cotta Fellowship to the National Bureau of Standards, Mr. H.G. Schurecht compiled it. It is the final report printed that remains available. The following are key points in the report:a. A study of crazing on buildings in different climates indicates that crazing occurs more in wet warm climates than in cold dry climatesa. The spalling freeze test is the best laboratory test for testing the resistance of terra cotta to spalling. Description of the spalling freeze test: This test is different from other freezing tests being used because the samples are special hemispherical shaped specimens are sued and a comparatively small area of the specimen is thawed on a steam bath. The specimens used are 5.5 inches in diameter with the glaze only on the convexFL A N 20 D R 09 O6. Additional tests72- FLANDROsurface. The specimens are immersed in water (without boiling) for 12 hours, they are then placed in shallow pans containing inch of water and put into the freezer for 12 hours. They are removed and placed with their convex sides over a 2.5-inch circular opening for about 20 minutes. This is repeated for 75-150 freezings, the total number of spalls is used to estimate their resistance to spalling.3. Some causes and methods of reducing peeling and crazing in terra cotta glazesb. Bodies that are most resistant to spalling in the freeze test were those with low free silica and low absorption.4. Some causes and methods of preventing the expansion of terra cotta bodies due to hydration a. It was found that terra cotta bodies continually expand when outdoors, firing the terra cotta to a higher temperature reduces its tendency to expanda. Glazes that use zinc oxide instead of tin oxide as an opacifier in, peel less when used without an under slip.5. Some causes and methods of preventing crazing of terra cotta glazes and other ceramic finishesa. Porous standard finishes and certain matte finishes were found to be under high compression, while vitreous standard finishes and lustrous finishes were only under slight compression.a. Expansion of glazes due to the action of water as a means of fitting glazes to bodies. It was observed that glossy glazes had the tendency to craze more than matte glazes. This would mean that certain finishes and glazes expand from hydration similar to the bodies and therefore do not craze as much. b. Strength tests of walls- It was found that building codes relating to the thickness of walls required for terra cotta work varied widely and wasvague, some cities required an 8 inch wall and others 12. A brick test wall was built to determine if 4 inch terra cotta ashlar facing blocks add to the strength of the wall or decrease it. Terra cotta was set on concrete backed walls and on brick backed walls. All of the walls were tested using a 10,000,000 pound testing machine at the Bureau of Standards.FL A N 20 D R 09 OFigure 15: Image of wall load test.1414When the load was being applied the following displacements were measured: vertical deformation of the terra cotta, vertical deformation of the concrete, horizontal movement of the terra cotta facing away from the backing wall, deflections or bending of the backing wall, strain gage measurements of displacements of terra cotta and backing wall. It was determined through the above test that the brick backing was compressed to a much higher degree on the brick constructed wall than the terra cotta on this wall. In the concrete backed wall the terra cotta facing was compressed more than twice the amount of the concrete. Defects that were visible on the brick wall included the shearing off of the faces of the terra cotta blocks. The conclusions from this testing procedure were such that building codes could be re-written to allow for the use of 14 inch (actually 8 inch solid brick with headers extending into the terra cotta voids) terra cotta faced walls filled with construction as they were strong enough.c. Cleaners for terra cotta- Alkaline cleaners like soap powder (Gold Dust) or tri-sodium phosphate are considered best for cleaning terra cotta as they do not damage the glaze. However, it was found that this sometimes was not strong enough to clean the wall, in which case slightlyThe 17th Report Relating to the Technical Work of the National Terra Cotta Society, 1928 page 69.73- FLANDROacidic cleaners were used such as sodium hydrosulphite, acidified sodium fluosilicate or ammonium bifluoride. Sandblasting or the use of hydrofluoric acid are highly injurious to the surface and are therefore condemned. 1928, December. The Technical News Bulletin of the National Bureau of Standards No. 140, Tests on Terra-Cotta faced Walls, another description of the tests undertaken and published in The 17th Report Relating to the Technical Work of the National Terra Cotta Society. It was found that terra cotta facing does add to the strength of the wall, this is significant because it means that the structural frame work of a building would not have to carry as high a load and could be constructed more economically.FL A N 20 D R 09 O192915 at the same time.1929, February. The National Bureau of Standards Technical News Bulletin No. 142, Fundamental Study of Glaze Fit. As reported in bulletins No. 125 and 128 a study of the effect of oxides on glaze fit has been underway. It has now be determined that the coefficient of expansion increases slightly with temperature. In the future the following tests hope to be completed: Resistance to thermal shock, the effect on glaze fit and its position relative to the alpha-beta inversion point and the variation with temperatures of the strain in a glaze attached to a body and its effect on crazing and shivering. The final tests will be undertaken with a thermal microscope that has recently been constructed at the Columbus station. 1929, May. The Journal of the American Ceramic Society, Effect of Water in Expanding Ceramic Bodies of Different Compositions, supplied to the journal by H.G. Schurecht and G.R. Pole at the National Bureau of Standards and working in conjunction with the National Terra Cotta Society. As determined earlier delayed crazing found on ceramic ware could be caused by the gradual expansion of the body during outdoor storage or use. This type of expansion, termed moisture expansion does not occur in glazes. Moisture expansion is less likely to occur on vitreous bodies in comparison with porous bodies. Schurecht and Pole ran several tests using the autoclave method of accelerated expansion on various clay body compositions and determined that an increase in free silica would cause an increase in moisture expansion; a decrease in moisture expansion of a siliceous fireclay body could be achieved with the addition of finely-ground magnesite or* Note: Investigations into the Durability of Architectural Terra Cotta and Faience by W.A. McIntyre was published in London. This publication includes similar studies to the tests being performed by HG Schurecht at the National Bureau of Standards. However, they are much less thorough also less technologically advanced than those being performed in the US74- FLANDROblast furnace slag. Increasing the firing temperature of ceramic bodies also reduced moisture expansion. 1929, May. The Journal of the American Ceramic Society, A Sandblast Abrasion Test for Glazes. Entered to the journal by Mr. Edward Schramm of the Onongada Pottery Laboratory, the article suggests using sandblasting as a way to measure abrasive resistance of a glaze. In his design the sand could be reused and the sample would be rotated so that a larger area would be abraded. Differences in weight from the start of the test compared with those at the end would determine a glazes resistance. 1929, September. The Journal of the American Ceramic Society, Notes on Terra Cotta Glaze Consistency, submitted by H. Spurrier of the Northwestern Terra Cotta Company. Spurrier makes the case that the use of a hydrometer for determining the viscosity of a glaze solution is reprehensible, incorrect and should not be done (this was a common practice in factories at the time) as the hydrometer is designed to weigh specific gravities of pure solutions, of which glaze is not. He suggest instead the use of a viscosimeter after a batch of glaze has been made at the desired specific gravity, then future batches should be made to meet that same viscosity.FL A N 20 D R 09 O193075- FLANDRO1929, October. The Journal of the American Ceramic Society, A Standard Unit for Masonry, Fred T. Heath suggests that standard unit sizes for terra cotta would be a great benefit to the construction industry, at the present moment terra cotta units were designed specifically for each job site.1930, May. The Journal of the American Ceramic Society, The Influence of Chemical Composition on the Physical Properties of Glazes, submitted to the journal by F.P. Hall and approved by the Director of the Bureau of Standards. Hill submitted some preliminary work on the study of factors affecting glaze fit on ceramic bodies. In order the examine glazes first without the influence of a body he made glazes in batches of 2 lbs, melted them in a clay crucible in a pot furnace and were stirred frequently with a nickel or fused silica rod. The glaze was then drawn into rods with aid of nickel or fused silica rods depending on the temperature of the molten glaze. Selections of rods from the batch were chosen based on their uniformity of cross sectional area and freedom of bubbles. Those selected were then annealed in an electric furnace. The annealing and cooling required, on average, thirty hours.The properties of glazes studied were tensile strength, Youngs modulus of elasticity, and coefficient of thermal expansion. To determine the chemical composition of the glaze through chemical analysis the methods used for soda-lime glass were used (these can be found in the Journal of the American Ceramic Society, November 1927, The Analysis of Soda-Lime Glass by G.E.F. Lundell and H.B. Knowles). The tensile strength was determined by sealing one end of a rod into a brass cup with DeKhotinsky cement, this was then secured to the table by a small chain. The other cup was connected to a chain run over a frictionless pulley. Load was added at the rate of 3500 grams/minute in the form of lead shot until the rod ruptured.FL A N 20 D R 09 OFigure 16: Apparatus/method used to determine the tensile strength of glazes.1616Youngs modulus of elasticity was determined on the glazes by the use of the apparatus shown here:Image from The Journal of the American Ceramic Society, The Influence of Chemical Composition on the Physical Properties of Glazes, 1930, page 190.76- FLANDROFL A N 20 D R 09 O17Figure 17: Apparatus used to determine Youngs modulus of elasticity of glazes.17and the following formula: E= (Wl3)/(L12r4) where W= weight in kilograms, L= deflection in mm, l=length of span in mm, r=radius in mm and E= Modulus of elasticity in kg./ mm2. The Fitzeau-Pulfrich method (interferometer) was used to measure small dilations relating to thermal expansion. 1930, June. The Journal of the American Ceramic Society, Some Simulative Service Tests for Glazed Building Materials, submitted by Otis Everhart of the Ohio State University Engineering Experiment Station (National Bureau of Standards). Everhart looked at several different test methods being used to simulate weather on building glazed building materials. He determined that the autoclave test simulated only one mode of nature, in particular, the differential expansion of the body and glaze, so the common failure in the autoclave is crazing. Next he looked at the freezing tests, both actual freezing and the use of crystallization of sodium sulphate. Freezing tests produce failure through spalling and also through the gradual expansion of the pores. Everhart makes recommendations for using these tests in factory settings. He suggests that no one test be used alone to determine failure. He suggests using the autoclave tests and also one or both of the freezing test methods. 1930, June. The Journal of the American Ceramic Society, Method of Measuring Strains Between Glazes and Ceramic Bodies, submitted to the journal by H.G. Schurecht and G.R. Pole fellows for the National Terra Cotta Society at the National Bureau of Standards. Schurecht and Pole composed a new test in which to compare the strains existingImage from The Journal of the American Ceramic Society, The Influence of Chemical Composition on the Physical Properties of Glazes, 1930, page 193.77- FLANDRObetween the body of terra cotta and the glaze. Its use would enable ceramic chemists to ensure that a proper fit was made between the glaze and body, eliminating crazing, peeling and shivering. The new test was particularly valuable because it used the same terra cotta bodies and glazes used for regular ware. The specimens used for the test are two-inch diameter hollow cylinder rings with the glaze or other ceramic finish to be tested applied to the outside only.FL A N 20 D R 09 O18Figure 18: Drawings of rings and apparatus used to determine contraction and expansion of the rings.18After firing, two grooves or holes are cut in one end, and are large enough to hold two capillary tubes that are 1/6 inch long and 1/32 inch in diameter. The tubes are cemented into the holes with litharge glycerin. The distance between the two holes is then measured using a micrometer microscope. The area between the two holes is then cut open using a diamond saw. The re-measuring of the space between the tubes will reveal the expansion or contraction of the ring. The same process is done on specimens without glaze made of the same body. The same procedures are performed. The subtraction of the expansion/ contraction of the unglazed terracotta body from the glazed will reveal the strain placed or released from the ring by the glaze alone. If the rings are in tension the ring will expand when cut, when in compression it will contract. The maximum stress that may have been released when cutting the ring can also be determined by Merriams curved beam formula.Images from The Journal of the American Ceramic Society, Method of Measuring Strains Between Glazes and Ceramic Bodies, 1930.78- FLANDROFL A N 20 D R 09 O19311. Lustrous glazes had an average expansion of .004% 2. Matte glazes had an average expansion of .011% 3. Vitreous slip finishes had an average expansion of .005% 4. Porous slip finishes had an average expansion of .033%79- FLANDROThe conclusions from this trial were that glazes could be under considerable compression without shivering but that they can craze when under comparatively small amounts of tension. The ring tests could also be used to compare moisture expansions of ceramic bodies and in general it was determined that mat glazes (due to their low thermal expansion) are in stronger compression than glossy glazes and are more resistant to crazing. 1930, December 31st.- The Report Relating to the Technical Work of the NationalTerra Cotta Society. Standardization of Tests for Terra Cotta Bodies and Glazes by R.L. Clare see December 1934.1931, April. The Journal of the American Ceramic Society, Moisture Expansion of Glazes and Other Ceramic Finishes, submitted by H.G. Schurecht and G.R. Pole fellows from the National Terra Cotta Society to the National Bureau of Standards. Schurecht and Pole used the autoclave test and the pressure test to examine several glaze specimens for moisture expansion. They obtained the following results:1931, June. The Journal of the American Ceramic Society, The Effect of Furnace Atmosphere on the Quality of Certain Types of Glazes, submitted by Arthur P. Watts, American Gas Association Research Fellow, Department of Ceramics Rutgers University. The point of this research was to determine the effects that different gasses have on glazes. All of the gases tested were those which those which could be found in actual kiln conditions. The atmosphere was the only variable. The following conclusions were found:1. An atmosphere of air produces good glaze surfaces (provided the glaze is of a good mixture) 2. Atmospheres containing carbon monoxide produce a gray or yellow discoloration in white glazes3. Unburned hydrocarbons may break down in a furnace atmosphere and darken a white glaze and leave a fine deposition of finely-grained carbon 4. Sulfur dioxide produces scumming and dull patches on the surface 5. Water vapor has a very harmful effect on glazes.FL A N 20 D R 09 O193380- FLANDRO19321932, July. The Journal of the American Ceramic Society, Miscellaneous Terra Cotta Laboratory Notes, submitted by H.R. Goodrich of Gladding, McBean & Company, he points out the cause and solution of green scumming. Scumming is caused by the deposition of water-soluble materials on the surface of terra cotta by evaporation of any moisture that has entered the body. The characteristic color of scum is first a pale greenish yellow, which deepens over a period of days to a dark green when it become very hard to remove. Scum forming salts are often found in the clay deposits themselves, such as diatomaceous earth and Lompoc and Ione. Scumming can be avoided by dipping in a potassium carbonate solution. At Gladding, McBean & Co. the dipping was simplified by equipping all the trucks used in unloading the kilns with a moveable shelf, where a motor hoist could be used to lift the load and dipped into a tank of the solution. The removal of green scum on older bricks and terra cotta (fabricated before the dipping method was discovered) used to be removed using a nitric acid, hydrogen peroxide solution and a potassium carbonate solution was applied afterwards. The potassium carbonate solution failed to prevent the reappearance of the scum, so water-proofing with a product called Minwax was applied with success. The method that has more recently been used is to brush the surface with a 5% nitric acid, 5% hydrogen peroxide solution until the scum disappears and while still damp the brick/ terra cotta and mortar joints were saturated with clear Minwax.1933, January. The Journal of the American Ceramic Society, A Method for the Determination of the Coefficient of Expansion, submitted by C.J. Kinzie and C.H. Commons, Jr. of the Titanium Alloy Manufacturing Company. A full description of a furnace and parts needed to determine the coefficient of expansion of glass in high temperature ranges. What makes this set up different from others is that there is a direct connection made from the sample to the micrometer, eliminating the need for error correction calcu-lations caused by the expansion and contraction of connecting parts (previous set ups had the rod-expansion test had the micrometer attached to the Ames dial on the furnace instead of having the Ames dial connected to the sample itself). 1933, December. The Journal of the American Ceramic Society, Terra Cotta Slip Coatings, submitted by Samuel J.McDowell of the American Terra Cotta Company. Used a triangular chart to compose different recipes for terra cotta slips (clay, feldspar, Cornwall stone) he then tested the 10 best results in compression (autoclave test) and for porosity. Porosity was determined by soaking the samples in an alcoholic dye solution for twentyfour hours, breaking them, and the porosity noted by the dye penetration. 1933, December. The Journal of the American Ceramic Society, A Low-Temper-FL A N 20 D R 09 O1. Cementing Materialssurfaces of terra cotta.81- FLANDROature Comparator for Coefficient of Expansion of Ceramic Bodies, submitted by H.E. Davis of the Federal Seaboard Terra Cotta Company. He developed a new apparatus to measure thermal expansion of fired clays and body mixtures from room temperature to 300C in hopes of predetermining their adaptability to a satisfactory glaze fit. While there was other more accurate testing procedures to reach the same result (see, January 1933) this one could be easily assembled for use at a terra cotta factory. 1933, December. The Journal of the American Ceramic Society, Water-Tight Terra Cotta Masonry, explored by F.O. Anderegg of Pittsburg, Philadelphia. This submission is unique in that it actually explored and determined the best way to make terra cotta masonry water-tight, knowing that the failure of terra cotta on buildings is usually caused by water. Suggestions for future constructions include:a. the mortar should be plastic enough to spread over the water absorbentb. It should have a bond strength of no less than 30 pounds per square inch in 28 days. c. A section of terra cotta wall should show, when loaded in flexure to two-thirds is breaking point a deflection of .001 inch per inch without breaking d. The mortar should shrink has the moisture leaves it, the loss of water prior due to premature evaporation can create small capillaries, connecting to the terra cotta body and enabling water flow.e. Setting rate- to help control shrinkage, waterproofers such as stearates or Magnesia, found in dolomitic limes are also helpful, so the lowest amount of water required to make the mortar workable is used. f. The cementing material recommended is composed of equal parts lime putty (soaked for 10 days) and portland cement to which has been added lb. of stearic acid per bag of cement.FL A N 20 D R 09 O4. Workmanshipa. Joint thickness- ideal thickness is in the range of 3/16-1/4 inch 5. Masonry Back up for Terra Cotta Walls82- FLANDRO2. Sand- the selection should have all different sizes present3. Mortar Mixing- Where the job size allows, a mechanical mixer should be used.b. Shoved joints- The mortar should be spread on the top of the lower block in an amount greater than that needed, and then pressed into the joint removing an excess mortar. c. Tool finishing joints- the use of any tool that will tightly pack the joint should be used d. Pointing exposed horizontal joints- for these joints it is recommended that a mortar containing equal parts by volume of lime putty and portland cement with four to five part sand and allowed to set for two hours prior to use is recommended, and the finishing of the joint should be completed with a rounded tool. e. Glazed surfaces- Two different methods many be used. One option is to use the hydrated mortar described above. The other is to apply a coating of portland cement paint to the glazed surfaces prior to pointinga. The standard practice is to back up with common brick; this has disadvantages as it is difficult to obtain complete water tightness, and if water is allowed to penetrate from the exterior it will enter the wall itself. Because of these problems it is suggested that a hollow tile back- up unit be designed and manufactured to use with terra cotta.b. Parapet walls- it is strongly recommended to build all parapet walls with hollow units, a flashing should be extended through the wall just above the roof-line to divert moisture away from the wall. 1933, December. The Journal of the American Ceramic Society, Terra Cotta Cracking, submitted by Fred B. Ortman of Gladding McBean and Company. This investigation was concerned solely with the causes and prevention of a certain type of cracking found on terra cotta buildings. Cracking was determined to be cause by the moisture expansion of the mortar used in the brick backing. The mortar used was determined to be too rich in ration of cementing ingredients to sand and too high a percentage of high magnesium hydrated lime. The recipe for this mortar was approximately 1 part by volume of portland cement, 1 part of high magnesium hydrated lime and 3 parts by volume of sand. Gladding, McBean and Company determined that if a high magnesium lime was to be used in conjunction with brick backup work on terra cotta a special hearing was required to determine why it was necessary. In general the following recipe for mortars was required for use with terra cotta brick backing: 1 part by volume of portland cement, 1 part by volume of high calcium lime (14 days old putty in preference to hydrated) and no less than 5 parts by volume of clean sharp sand. For pointing mortars: 1 part by volume of portland cement, 2 parts by volume of clean sharp sand and part by volume of fire clay and no lime of any character. For grout filling the following was required: 1 part by volume of portland cement, 6 parts by volume top gravel or well graded sand and no lime of any character.FL A N 20 D R 09 O19341934. The National Terra Cotta Society dissolves.83- FLANDRO1934. The United States National Bureau of Standards takes a severe budget cut, resulting in the research station at Columbus, Ohio being closed and a severe curtailment in the development of new standards, refined measuring methods and the determination of the physical properties of materials. 1934. The Bulletin of the American Ceramic Society, New Principles of Tunnel Kiln Design Applied in a Kiln for Architectural Terra Cotta, Mr. Hull describes the process for firing architectural terra cotta in the new tunnel kiln. The advantage of this new kiln was its ability to completely fire terracotta in a period of 4 or 5 days instead of 10 or 12. 1934, December. The Journal of the American Ceramics Society, Blistering andFL A N 20 D R 09 O193884- FLANDROCrawling of Glazes caused by Underslips, submitted by H.G. Schurecht former fellow to the National Bureau of Standards for the National Terra Cotta Society, now employed at The Eastern Terra Cotta Company. He tested various slip formulas to determine why with some slips blistering and crawling during firing occurred more frequently. He determined that the blistering of glazes during spraying may be reduced by lowering the amount of non-plastic ingredients in the slip and increasing the clay content. In addition, reducing the shrinkage and increasing the dry strength of the under-slips may prevent crawling of gazes caused by under-slips.1934, December. The Journal of the American Ceramics Society, Standardization of Tests for Terra Cotta Bodies and Glazes, presented by R.L. Clare of the Federal Seaboard Terra Cotta Corporation and also printed in The National Terra Cotta Society Technical Report December 31, 1930. Clare points out the lack of uniformity of testing in the manufacturing of terra cotta, so if a pieces is not to have absorption more than 10% how is this tested? Even with the autoclave test which was generally approved as a way to test for crazing, Schurecht at the National Bureau of Standards suggests one hour at 150 pounds, but M.I.T. suggests five cycles of one hour durations at 200 pounds is necessary. Clare further suggests that a committee should be set up to standardize the following testing procedures for terra cotta: absorption, freezing, thermal shock and modulus of rupture.1938, July. The Journal of the American Ceramic Society, Life History of a Glaze: Measurement of Stress in a Cooling Glaze, submitted by A.M. Blakely at the Laboratory of Ceramics at Massachusetts Institute of Technology. In this experiment Blakely comes up with a way to look at glazes in a quantitative way instead of just qualitative. Using a slight alteration on Stegers method (also called the tuning fork method) (see image below) for measuring stresses in glazes at high temperatures he came to the following conclusions: 1. Glazes in high compression (8-10kg/mm2) show no signs of failure after being removed from the kiln or autoclave, however those even in the slightest tension show the tendency to craze. 2. Delayed crazing is caused by a delayed contraction of the glaze, the change of glaze stress caused by this contraction is always in the direction of tension. This type of contraction can take weeks or months to happen after the sample has been removed from the kiln.3. Autoclaving always moves the glaze stress into tension, by using moisture to expand the body. Therefore, bodies with higher porosity have a tendency to craze more in this test. 4. Low-fired bodies show a greater tendency to have the glaze be in tension (crazing).FL A N 20 D R 09 OFigure 19: Glaze Stress Furnace195. The difference between the thermal expansion curves of the glaze and body is only a qualitative test.19401940, June. The Journal of the American Ceramics Society, Evaluation of GlazeFit Test Methods, completed by W.C.Bell at the National Bureau of Standards Engineering Station in Columbus Ohio. Bell tested the three methods currently available to check for proper glaze fit on a body; the ring test (see June 1930 & 1928), the flat plate test (the flat plate test is only used for items being fired twice) and the tuning fork method (ca. 1928). The results of all the tests were in agreement with each other. He found that the ring test was good for qualitative measurements, the flat plate test was rapid and required little equipment but could only be used for two-fire glazes and that the tuning fork test was really19Image from The Journal of the American Ceramic Society, Life History of a Glaze: Measurement of Stress in a Cooling Glaze, 1938.85- FLANDROonly good in laboratories specifically studying the development of stress in glazes upon cooling due to the expensive nature of the equipment.19411941, April. The Journal of the American Ceramics Society, Study of Glaze and Body Interface, by C.W. Parmelee and Paul E. Buckles of the Department of Ceramic Engineering at the University of Illinois. Using a microscopic with white light and X325 magnification thin sections of cross sections of lead glazes on bisque tiles were examined. The authors concluded that the interfacial layer is not necessarily intermediate in expansion conduct.FL A N 20 D R 09 O1943Figure 20: Device used to cause parallel crazing.20201943, March. The Journal of the American Ceramic Society, Fitting Glazes to Ceramic Bodies, submitted by H.G. Schurecht at the Ceramic Experiment Station, New York State College of Ceramics (Alfred). Presenting new information pertaining to fitting glazes to ceramic bodies. Schurecht found that looking at thin sections of just the glaze (cut perpendicular to the body) and examined under polarized light would reveal that an unstressed glaze would be isotropic and a stressed glaze anisotropic. The birefringence caused by the stressed glaze could be measured with a Berek compensator. He also notes that parallel crazing on in service buildings is caused by pressure and can be tested using the following device:Image from The Journal of the American Ceramic Society, Fitting Glazes to Ceramic Bodies, 1943.86- FLANDRO1943, May. The Journal of the American Ceramic Society, Simplified Thermal Expansion Rings for Production Control in Terra Cotta Manufacture, by T.E. Nicholson and Fred L. Goin of N. Clark and Son. Nicholson and Goin created a simplified and more controlled method for making thermal expansion rings by altering a meat grinder. The clay could be extruded and then cut with a wire.FL A N 20 D R 09 O1946 195021Figure 21: Converted meat grinder to prepare expansion rings for testing.21The usual tests using expansion rings were performed using 4 different clay bodies and glazes and then compared with expansion ring tests not made through extrusion. The die extruded pieces proved to be more accurate, and this machine could be made with very little cost to the manufacture.1946, July. The Journal of the American Ceramic Society, Investigations of Crazing- Delayed Thermal Contracting and Crazing of Ceramic Glazes. Submitted by L. Mattyasovszky-Zsolnay he made experimental rods consisting of sixteen earthenware glazes, which were then tested for stresses using the Steger apparatus, stored for five years in an air-tight container and then re-tested. The stresses remained consistent, indicating that delayed thermal contraction is negligible.1950. The American Society for Testing and Materials, ASTM C-242, Standard Terminology of Ceramic Whitewares and Related Products, this document standardized the definitions of over two hundred words relating to ceramics. It was revised and reapImage from The Journal of the American Ceramic Society, Simplified Thermal Expansion Rings for Production Control in Terra Cotta Manufacture, 1943.87- FLANDROproved in 2001.19611961, October. AIA File No. 9, Public Works Specifications Ceramic Veneer, presented as a guide, by the architectural terra cotta institute, for preparing contract specifications covering ceramic veneer in government financed project. This document gives suggestions and standards (ASTM) for testing ceramic veneer. However, it was used (and possibly still used) by Gladding, McBean and Company for architectural terra cotta, as there are no standards relating specifically to terra cotta, key points are listed below: 1. Physical requirements (does not include sculpture), 5 specimens of 1 square inch have to be tested. a. Compressive strength- 5000 psi (average of 5), individual 3500 psi. b. Maximum absorption by 5 hour boiling (percent)- 16 (average of 5), individual 18 c. Maximum saturation coefficient (is the ratio of absorption by 24-hour submersion in cold water to that after 5 hour submersion in boiling water)- .80 (average of 5), individual .82FL A N 20 D R 09 O2. Properties of glazes and vitrified ceramic finishes 3. Standard methods of sampling and testing ceramic veneer88- FLANDROa. Imperviousness- after the imperviousness test (see testing methods below), no stain that can be seen from a distance of 5 feet shall remain on or beneath the surface, except slight discolorations in the depressions of mottled finishes b. Chemical resistance- the color of the glaze shall not change under the chemical resistance test c. Resistance to crazing- the glaze shall not craze, spall or crack when subjected to one cycle in the autoclave crazing test d. Adhesion- adhesion tests shall show no clean separation of finish and body.a. Compressive Strength Test- ASTM C67 Standard Methods of Sampling and Testing Brick b. Absorption Test- ASTM C 67 c. Imperviousness Test- This test shall be done on 3 specimens, 10x10inch face size and the full thickness of the exposed face of the unit. Permanent blue-black fountain pen ink shall be applied liberally to the glazed surface and allowed to remain for 5 minutes. The surface will then be washed with a wet cloth and running water, and examined from a distance of 5 feet for staining. d. Chemical Resistance Test- This test shall be made on 3 specimens, 6x6 inch face size and the full thickness of the exposed face of the unit. Subjects subjected to the absorption tests may be used. A portion of the glazed face shall be placed in a 10% solution of hydrochloric acid for 3 hours. Another portion of the glazed surface of the same specimen shall be submerged in a 10% solution of potassium hydroxide for 3 hours. The solutions shall be kept at a temperature between 60-80F. The specimens will then be rinsed, dried and examined for changes in color. e. Crazing test- to be done on 3 specimens, 4x4 inch face sizes and the full thickness of the exposed face. The autoclave used for testing shall be equipped with a safety valve, blow off valve, thermometer, pressure gage (accurate), and a heater or other means of sufficient capacity to insure that the steam pressure remains consistent. The specimen should be placed loosely above the water in the autoclave at room temperature. The blow off valve will be kept open as the unit is heated allowing most of the air to escape, after which it can be closed. The steam pressure should increase at a uniform rate until 75 psi is reached in a period not less than 60 minutes or more than 90. Consistent steam pressure at 75 psi is to then be maintained for an additional hour. The heater then can be shut off and the steam released (in not less than 30 minutes). The autoclave head can be loosened but not released, allowing the specimens to cool to room temperature (not more than 120 F) in a period of no less than 3 hours. The specimens can then be removed and a standard blue-black fountain pen ink or water-soluble dye solution rubbed upon the glazed surface to aid in the detection and examination of failures.FL A N 20 D R 09 O89- FLANDROFL A N 20 D R 09 Ob. Mortar materials shall conform to the following: i. Portland cement: ASTM C 150, Type 1 iii. Hydrate lime: ASTM C 206, Type S iv. Sand: Clean, sharp sand, ASTM C1444. Mortarf. Adhesion test- vinyl acetate dissolved in methyl iso-butyl keytone is applied to the surface to be tested and also to the end of a 1 by 1 by 4 inch vitrified test bar. The adhesive is heated by means of an infrared lamp until bubbling ceases. The two surfaces are pressed together until cool and then the 1 by 1 inch bar is knocked or pried off.a. Setting mortar: 1 part portland cement, high-calcium lime putty or Type S hydrate lime and 4-4.5 parts sandii. Lime Putty: approved high-calcium slaked lime putty screened and aged at least 20 days, containing not more than 4% magnesium oxide1962-20091969. American Society for Testing and Materials, Standard F 109-04, Standard Terminology Relating to Surface Imperfections on Ceramics, twenty-nine definitions relating to ceramics surfaces were standardized. This standard was revised and reapproved in 2004. 1979. National Park Service, Preservation Brief 7: The Preservation of Historic Glazed Architectural Terra Cotta, by de Teel Patterson Tiller is created for general guidelines to be followed when undertaking a historic terra cotta restoration. These guidelines have never been updated since their publication.1996. American Society for Testing and Materials International, Special Technical Publications, STP- 1258, Testing an Analysis of Terra Cotta Glaze Repairs, by R. Viera of Building Conservation Associates. Viera tested coatings and combinations to determine a suitable repair for losses on otherwise sound terra cotta units. He tested urethane-based materials, acrylic based, epoxy based and mineral based materials. Viera completed tests90- FLANDROon these materials to determine their stability to ultraviolet radiation (using ASTM G 5384), resistance to freeze-thaw cycles (ASTM C 67-83) and adhesive performance (ASTM D3359). After the conclusions of these tests he determined that acrylic based coatings suffered the least amount of damage. 2001. Originally published as ASTM C 242-50, Standard Terminology of Ceramic Whitewares and Related Products, is revised, renumbered and approved as ASTM C24201. 2004. ASTM F 109-04 revised and reapproved.FL A N 20 D R 09 O91- FLANDRO2004. American Society for Testing and Materials International, Special Technical Publications, STP- 1444, Terra Cotta Facades, presented by K.R. Hoigard, G.R. Mulholland and R.C. Haukohl at the Symposium on Building Faade Maintenance, Repair and Inspection. They discuss the historical uses of terra cotta in the United States, three misconceptions about the material properties and what should be noted in a thorough examination. The first being that terra cotta glazing is impervious to liquids, glazing only provides protection against water on one face; the others do not have this protection. Secondly, that terra cotta facades are weatherproof, historically manufactures would market this point; however, mortar joints, roofing, building projects and other faade discontinuities all provided points of entry for weathering and future deterioration. The third misconception was that moisture expansion did not occur with terra cotta, this did occur, but on historic buildings expansion joints were rarely provided. In an examination of in service terra cotta should include (but not be limited to) close up visual inspection, sounding, penetrative examination and a review of any available drawings. 2009, May. Xsusha Flandro, a brilliant graduate student at Columbia University complies a history of testing procedures created throughout history to gather more information about architectural terra cotta as well as performing water vapor transmission testing on pieces of glazed terra cotta with crazing and/or crawling.FL A N 20 D R 09 OSECTION II: MATERIAL TESTING92- FLANDROSection III: Material TestingAs is noticeable through the previously supplied time line of testing procedures, there was a substantial amount of research and experimentation done on terracotta. Interestingly, very little of this sort of examination has been continued in the preservation field. Twentyone years ago, Susan Tindall published an article in the Association for Preservation (APT) Bulletin, titled Terra Cotta Replacement, listing twelve tests that should be done to assure that properties of terra cotta are correctly identified,1 in reference to original terracotta material. Her list included: Water Absorption (24 hour) Boil Absorption (1 hour) Firing Temperature Moisture Expansion (reheat test) Water Uptake with and Without Glaze Compressive Strength Shear Strength Petrographic Examination X-ray Diffraction and Fluorescence Glaze Adhesion Coefficient of Thermal Expansion Youngs Modulus of ElasticityFL A N 20 D R 09 O1. Glazed Surfaces- Typical Conditions1Rarely are these tests completed, a conservator should know as much as they can about the original material before performing any conservation treatments or suggesting replacement and when it comes to glazed terra cotta the practice of ignoring original conditions or immediately assuming that the block is faulty because of existing conditions in the glaze (crazing or crawling) must be halted. To help alleviate the current practice of assumming terra-cotta material conditions without testing them first, five different examinations where choosen to be completed in this experimental design. These experiments are realatively quick, easy and inexpensive to complete. Two of the tests, water vapor transmission and inverted cup water vapor transmission have not ever been used before on glazed terracotta (it has always been assummed that glaze is impermeable to water). The following tests provided a wealth of information about terracotta. Five investigations:Tindall, Susan M. Terra Cotta Replacement, Association for Preservation Technology Bulletin. Volume 20, No. 3, 1988, p. 12.93- FLANDRO2. Cross Sections (of glaze and underlying body) 3. Petrographic Examination 4. Water Vapor Transmission 5. Inverted Cup Water Vapor Transmission While the ideal testing program would allow for innumerable samples, one of the inherent limitations with testing terracotta is finding enough samples to work with. Six terra-cotta blocks taken from buildings in New York City for the following material tests were kindly provided by Douglas Schickler at Essex Works Limited in Brooklyn, New York.FL A N 20 D R 09 O94- FLANDROTesting MethodologiesAll of the specimens were washed in lukewarm water (approximately 86F) and scrubbed free of loose debris with a soft synthetic bristle scrub brush. The specimens were dried indoors at room temperature (75-85F) for six weeks prior to taking smaller samples for testing. Extracting Samples: Because the glaze and the condition of the glaze was being tested it was crucial to make sure that the extraction of smaller samples from the blocks did not disturb the glaze. It was first attempted to chisel samples out, but this was determined to be insufficient because there was concern that the amount of direct impact pressure caused by the chisel and hammer could possibly cause crazing and cracking in the specimens. The second method used a 1 inch diameter Neiko Professional Diamond Dust Hole Saw Drill Bit 1 on a Black and Decker Variable Speed Drill (120v, 0-1340 RPM, Model # 4.5A DR 200) with 7/8 (approximately 22.23 mm) inch diameter cores being taken dry. The problem with dry drilling is the amount of dust produced and the inability to see how deep the drill bit was going due to the dust. This method was then employed using an intermittent spray of de-ionized water directed onto the drill bit (Figure 22) and was found to be sufficient. Coring was done to an approximate depth of 7-10 mm. The removal of the core from the rest of the block required a hammer and chisel, which was used to excavate an area adjacent to the core to the same depth as the core. The core was knocked loose from the body, using the chisel, with very little force (Figure 23). Following extraction, the cores were rinsed of debris with de-ionized water and dried in a Precision Economy oven (manufactured by Jouan, Inc.) at 30C for 10 hours.FL A N 20 D R 09 OFigure 22- Intermittent spray of de-ionized water directly onto the core drill bit.95- FLANDROFigure 23- Excavation of terracotta adjacent to core, to allow core to be freed without a lot of force.Specimen Labeling: Each specimen was labeled using a two digit number for each experiment. For example the first sample taken from block #1 was labeled 1.1, the second sample 1.2 and so forth. Because different samples were used for different experiments this numbering system was repeated for each experiment. Visual Analysis: After the samples were taken, the glazed surface was examined and photographed at different magnifications. A VHX KEYENCE Digital Microscope Model: VHX-500, was used for this analysis. This particular microscope allows for samples to be photographed at any magnification between 25x and 175x The purpose of this was to document typical glaze conditions. Petrography: Samples from each block were sent to American Petrographics in Roslyn Heights, NY to be ground, (two were sent from block #2 because of the extensive amount of crazing visible) such that the interface between the glaze and the body was visible. The remaining portions of the samples that had not been ground down were returned from American Petrographics and used for the cross section analysis. The visual analysis and photographing of the thin sections was done under transmitted light, using a Zeiss Binocular Microscope (Model Axioplan 2 ), with a mounted camera using Pursuit using Advanced SPOT Software, by Diagnostic Instruments Inc. (Model#: 164 4MP Slider). Photographs of the thin sections were taken both under plane polarized light (PPL) and cross polarized light (XPL), at 50x magnification and 100x magnification. The specimens were oriented so that when photographed the glaze would be at the top of the image (see Figure 24). One of the main purposes for inspecting thin sections of terracotta is to determine how much of the ceramic material has actually reached a glassy state, which would be impermeable to water. The more vitrified a substance is the darker it will appear in plane polarized light, with glass being completely black. The transitional phase between un-vitrified and vitrified has a muddy (blurry) brown appearance. The other important purpose for thin sections is to identify particle orientation, which is critical in situations of delamination.FL A N 20 D R 09 O96- FLANDROFL A N 20 D R 09 OFigure 24- Specimen oriented with glaze pointing upwards.97- FLANDROFigure 25- Exterior surface of the terracotta block on the left-hand side.Cross-Section Analysis: After the thin sections were made, the remaining portion of the sample was mounted in resin (Castin Craft Clear Liquid Plastic Casting Resin). After the samples were mounted they were allowed to harden for 48 hours. After which a portion of the resin and the sample were cut off to expose a cross section of the sample. The sample was progressively polished using Micro-MeshTM abrasive sheets (Numbers 400, 1500, 1800, 2400, 3200, 3600, 4000 and 8000). The sample was placed on a glass slide (to ease maneuvering while under the microscope objective). It was stabilized on the slide using an oil clay and made level with a Zeiss specimen press level. Looking at the samples under the microscope it was apparent that a small drop of oil was needed on top of the sample, which was then covered with a glass slide coversheet to improve the visibility. All samples were photographed at 25x, 50x and 100x magnification. With the exterior facing surface on the left hand side, (see Figure 25). The cross sections were observed and photographed under reflective light, using a Zeiss Binocular Microscope (Axioplan 2 ), with a mounted camera using Pursuit using Advanced SPOT Software, by Diagnostic Instruments Inc. (Model#: 164 4MP Slider).The purpose of a cross section examination is to determine if the crazing/crawling (if applicable) continues into the terracotta body. It is also important for determining the thickness of the glaze layer and the presence or absence of a slip layer. Water Vapor Transmission: Note: Photographs of the surface of each sample have been provided in the appendix as well as all of the daily weight data. Because there are currently no pre-made water vapor transmission cups for 7/8 diameter samples, small 50cc containers were purchased from the local drug store. The containers came with threaded lids. The threads were coated with 5 minute epoxy (Devcon/Versachem High Strength Clear 5-Minute Epoxy) and allowed to set. A one inch diameter hole was cut in the bottom of the container.FL A N 20 D R 09 O98- FLANDROEight cores were taken from three blocks (blocks 1, 2 and 3). Blocks 4, 5 and 6 were not used for this laboratory (which is explained further in the analysis of each block) for a total of twenty-four cores. The cores were then glued into the holes with the glazed surface facing outwards, using the five minute epoxy. As the epoxy set, they were set glaze face downwards on filter paper so that the glazed surface remained level with the container as the glue hardened. A small hole was drilled on the side of the containers to enable 20ml of de-ionized water to be injected into each container using a large syringe. Care was take during the injection to not get any liquid on the sample (Figures 26 & 27). The fill hole was sealed with the epoxy. The samples were each weighed using a Sartorius LP 4200 S (1/100 gram measurements with a maximum weight of 4200g) scale, this weight was recorded as the starting weight. The samples were stored in a Secador Desiccator humidity control container, with an approximate humidity between 25-30%. Every day for the following twenty five consecutive days the samples were weighed. The date, time, humidity and temperature inside of the container were also recorded.FL A N 20 D R 09 O99- FLANDROFigure 26- Small container with terracotta Figure 27- Injection of de-ionized water core in place. into container. Inverted Cup Water Vapor Transmission: Similar to water vapor transmission, the inverse water vapor transmission tests was used to determine if water in direct contact with the terracotta body had any effect on its water vapor transmission rate. Immediately following the completion of the regular water vapor transmission test (where the liquid is not in contact with the specimen) the samples were emptied of any remaining water by drilling a small hole in the bottom of the container. Each item was then flipped over, re-filled with 20ml of de-ionized water and the hole resealed with 5-minute epoxy. The specimens were allowed to sit glaze face down on filter paper for four hours to ensure that no leaks had formed in the epoxy seal around the terracotta core during the previous experiment (Figure 28). During this period small 1 foot squares of fiberglass 18/16 screen mesh were sewed around three separate, 6 diameter pyrex petri dishes. Each sample was weighed and this weight was noted as the start weight. The samples were then placed onto the mesh covered dish (Figure 29). The mesh was used to elevate the samples, so they were not sitting in direct contact with a solid surface (which could potentially block air flow and slow water vapor transmission. The specimens were returned to the dessicator. Every day for twenty five consecutive days each specimen was removed from the dessicator for a short period of time to be weighed.FL A N 20 D R 09 OFigure 28- Specimens on filter paper to ensure water tightness of the epoxy seal.100- FLANDROFigure 29- Prepared samples elevated by fiberglass mesh, ready to be inserted into the dessicator. In addition to the weight, the time, date, temperature and humidity in the container were also recorded. This raw data can be found in the appendix. Normalized data: In order to determine if the testing method was working and to gain an average weight loss per sample per day, the data was normalized and graphed. The x-axis was the cumulative change in mass (g) {start weight minus current weight} divided by the number of the day, and the y-axis the day number. For example on day nine item 1.1 had a weight of 37.09g, its starting weight was 37.33g, the normalized data point would be (37.33-37.09)/9 = .03 g/day . When the data is flat line it means that a constant amount of water vapor is being transmitted on a day to day basis, and the experiment is successful and can be halted. The data was normalized for both the water vapor transmission and also the inverted cup water vapor transmission.The normalized data point on day twenty-five in both experiments was used along with the surface area of the core (which is 19.71mm2 for every sample) to gain the water vapor transmission per day (in grams) per surface area for each sample. This number was then used to extrapolate the amount of water vapor transmission (in grams) per day per meter squared, to make the data useful for the larger surface areas that are dealt with in architectural preservation.Once the water vapor transmission in grams/day/m2 had been determined for each test, Dixons Q-Test was applied separately to each group of samples taken from one individual block to identify and eliminate outliers with 95% confidence.FL A N 20 D R 09 O101- FLANDROWater Vapor Transmissions Comparative Analysis: Used as a further comparative analysis technique the twenty-four samples were separated into five different categories based on the glaze surface conditions visible with the naked eye. The categories were: glazes with no visible fit defects, crazes only, crawl only, crawls and crazes and scratched glaze. The idea behind the use of this type of analysis was to determine if different surface conditions (especially crazing and crawling) would yield higher water vapor transmission rates (it was hypothesized that it would). After the group was completed the Q-Test was used at 95% certainty to eliminate any outliers from each category. In relation to the materials testing the following information can be found in the appendix: Daily weights for each sample for both the water vapor transmission and inverted cup water vapor transmission, the normalized data and graphs of for both tests, a chart of all the samples along with their grams of water vapor transmission rate and the mathematics used in conjunction with Dixons Q-Test.Block #1FL A N 20 D R 09 OFigure 30: Terra-cotta Block #1102- FLANDROFigure 31: Building from which Block #1 originated. Building: The Alameda 255 West 84th Street (or 2321 Broadway), New York, NY 10024 Year Built: 1914 Architect: Gaetan Ajello Munsell Color (Glaze): 10YR, 8/1* Munsell Color (Body): 10YR, 8/3 Glaze: Matte * All Munsell colors were taken using the Munsell Soil Color Charts 1975 EditionBlock #1- Visual AnalysisFL A N 20 D R 09 OFigure 32- Block #1, 25x magnification. Figure 33- Block #1, 50x magnification. Figure 34- Block #1, 100x magnification. Figure 35- Block #1, 175x magnification. Block #1 surface observations: The surface of Block #1 is rough due to a highly pitted surface. Interspersed in the glaze are peaks of a glassy, transparent material. One such peak has been noted in Figure 35. Iron-red colored speckles are visible all over the surface (both inside and outside of the pitting). No crazing or crawling is visible on this specimen.103- FLANDROBlock #1- Thin Section AnalysisFL A N 20 D R 09 OFigure 36- Block #1, PPL, 50x magnification. Figure 37- Block #1, PPL, 100x magnification.104- FLANDROFigure 38- Block #1, XPL, Figure 39- Block #1, XPL, 50x magnification. 100x magnification. Block #1 Thin section observations: The glaze on this specimen looks to be of similar composition to the clay body, but with smaller sized grog. The glaze has not reached a vitrified state, meaning that it was not fired to maturity. They clay body had a lower melting point than the glaze, distinguishable by areas in the clay body that have reached maturity. The areas of most developed maturity in the clay body coincide with a white round piece of aggregate (in PPL). This occurrence is denoted in Figure 37. Voids in the clay body can bee seen in close proximity to areas that are the most vitrified.Block #1- Cross Section AnalysisFL A N 20 D R 09 OFigure 40- Block #1, 25x magnification. Figure 41- Block #1, 50x magnification. Figure 42- Block #1, 100x magnification.105- FLANDROBlock #1 Cross Section Observations: Block #1 has a translucent glaze and no slip layer, these findings are in accordance with the observations noted in thin section. The interface between the glaze and the body is rough, suggesting that the clay body was not smoothed before the glaze was applied. The glaze layer is 200 m thick.Block #1Water Vapor Transmission Rates & Inverted Cup Water Vapor Transmission RatesFL A N 20 D R 09 OFigure 43- Sample 1.1 WVT rate: 51.48 grams/day/m2 Inverted WVT rate:51.48 grams/day/m2 Figure 45- Sample 1.3 WVT rate: 77.22 grams/day/m2 Inverted WVT rate: 77.22 grams/day/m2106- FLANDROFigure 44- Sample 1.2 WVT rate: 25.74 grams/day/m2 Inverted WVT rate: 51.48 grams/day/m2Figure 46- Sample 1.4 WVT rate: 51.48 grams/day/m2 Inverted WVT rate: 25.74 grams/day/m2Block #1 (continued)Water Vapor Transmission Rates & Inverted Cup Water Vapor Transmission RatesFL A N 20 D R 09 OFigure 47- Sample 1.5 WVT rate: 0 grams/day/m2 Inverted WVT rate: 25.74 grams/day/m2 Figure 49- Sample 1.7 WVT rate: 77.22 grams/day/m2 Inverted WVT rate: 25.74 grams/day/m2107- FLANDROFigure 48- Sample 1.6 WVT rate: 51.48 grams/day/m2 Inverted WVT rate: 25.74 grams/day/m2Figure 50- Sample 1.8 WVT rate: 25.74 grams/day/m2 Inverted WVT rate: 102.96 grams/day/m2Block #1- Water Vapor Transmission Rates & Inverted Cup Water Vapor Transmission Rates Analysis The application of the Q-Test determined there were no outliers from this set of data that needed to be eliminated. The average water vapor transmission rate for this block was, 45.05 g/day/m2. The average inverted cup water vapor transmission rate was, 48.26 g/day/m2. The rate increased only slightly when the water was in direct contact with the specimen. Block #1- ConclusionsFL A N 20 D R 09 O108- FLANDROThese tests revealed interesting information about Block #1. Both the body and the slip were not fully vitrified, which would imply that the water vapor transmission rates would be higher than specimens with a completly vitrified coating (the more glass like a substance is the lower the water vapor transmission rates are). However, this was found to be an inccorect assumption in this case. The water vapor transmission rates were low in comparisons to the other samples. Looking at the surface texture of the glaze it could also be assumed that the increase in surface area caused by the roughness would also increase the water vapor transmission rates, again this theory was proven incorrect with the lower water vapor transmission rates. The glaze thickness on this block was approximately 200 m thick and there was no slip layer present. The interface between the clay body and the slip was rough, perhaps providing better adhesion between the two.Block #2FL A N 20 D R 09 OFigure 51: Terra-cotta Block #2 Figure 52: Building from which Block #2 originated. Building: The Capitol 12 East 87th Street, New York, NY 10128 Year Built: 1910 Architects: George and Edward Blum Munsell Color (Glaze): 2.5Y, 8/2 Munsell Color (Body): 10YR, 8/3 Glaze: Glossy109- FLANDROBlock #2- Visual AnalysisFL A N 20 D R 09 OFigure 53- Crazing on Block #2, 25x magnification. Figure 54- Crazing on Block #2, 50x magnification. Figure 55- Crazing on Block #2, 100x magnification. Craze Crawl Intersection of a craze and a crawl Figure 57- Crawling on Block #2, 100x magnification. Figure 58- Crawling on Block #2, 175x magnification.110- FLANDROFigure 56- Crazing and crawling on Block #2, 25x magnification.Block #2 surface observations: The surface of this glaze is smooth, there are significant amounts of crazing and crawling. The appearance of these glaze fit defects were easily visible due to the soiling inside of the interstices. In addition it was noted that there are areas on the surface where both crazing and crawling intersect, as is demonstrated in Figure 57.This is an interesting observation as crawling is caused by glaze adhesion problems and crazing by a glaze being under tensile stresses.FL A N 20 D R 09 OFigure 59- Block #2, crawling glaze, PPL, 50x magnification. Figure 61- Block #2, crawling glaze, XPL, 50x magnification111- FLANDROBlock #2- Thin Section Analysis (taken from an area with crawls)Figure 60- Block #2, crawling glaze, PPL, 100x magnificationFigure 62- Block #2, crawling glaze, XPL, 100x magnification.FL A N 20 D R 09 OBlock #2- Thin Section Analysis (taken from an area with crazes)Void Figure 63- Block #2, crazing glaze, PPL, 50x magnification Figure 65- Block #2, crazing glaze, XPL, 50x magnification.112- FLANDROBlock #2 thin section observations- crawl: The glaze on this specimen is close to complete vitrification. A slip coat layer is visible, separating the glaze from the clay body. In areas where crawling has occurred the glaze has taken a mass of slip with it as it contracted during the firing and subsequent cooling, forming small mountains of slip topped with glaze. The glaze in these areas of crawling does not completely cover the sides of these mountains, it remained only on the top most portions. Visible in Figure 60, are the distinct layers with little intermingling (there is no absorption of the slip layer into the body of the clay, or into the glaze). The thin section also proves that the crawling of the glaze does expose the terra-cotta body in some areas and the slip layer in others.Figure 64 Block #2, crazing glaze, PPL, 100x magnification.Figure 66- Block #2, crazing glaze, XPL, 100x magnificationBlock #2 thin section observations- Craze The glaze in areas of crazing is thin in comparison with the glaze ares that show crawls The crazing does continue into the clay body and coexists with an uneven layer of slip and a large void in the clay body (Figure 63). In addition the exterior surface of the glaze on this sample has higher roughgosity in comparison to the exterior surface of the glaze that crawled from the same block.FL A N 20 D R 09 OCrawl Figure 67- Block #2, crawling glaze 25x magnification. Figure 69- Block #2, crawling glaze, 100x magnification.113- FLANDROBlock #2- Cross Section Analysis (area with crawl)Figure 68- Block #2, crawling glaze, 50x magnification.Block #2- Cross Section Analysis (Area with craze)FL A N 20 D R 09 OCraze Figure 70- Block #2, crazed glaze 25x magnification. Figure 71- Block #2, crazed glaze, 50x magnification. Figure 72- Block #2, crazed glaze, 100x magnification.114- FLANDROBlock #2 cross section observations- This block has three distinct layers. The outermost layer is opaque, while the slip layer is translucent. The glaze layer is approximately 180 m thick, while the slip layer is closer to 400 m thick. In the area adjacent to the valley caused by the crawling of the slip and glaze, the slip layer is thick and the glaze is sitting directly on top of the slip, this pheomenom was also noted in the thin sections. There is not any residual slip or glaze in the valley and the clay body is exposed. The crazing is subtle in this cross section and difficult to see, it has been marked in Figure 70. The crazing does not appear to penetrate the terra-cotta body in cross section, however the penetration is visible in the thin section.Block #2Water Vapor Transmission Rates & Inverted Cup Water Vapor Transmission RatesFL A N 20 D R 09 OFigure 73- Sample 2.1 WVT rate: 489.10 grams/day/m2 Inverted WVT rate: 102.96 grams/day/m2 Figure 75- Sample 2.3 WVT rate: 386.12 grams/day/m2 Inverted WVT rate: 77.22 grams/day/m2115- FLANDROFigure 74- Sample 2.2 WVT rate: 180.19 grams/day/m2 Inverted WVT rate: 347.60 grams/day/m2Figure 76- Sample 2.4 WVT rate: 411.86 grams/day/m2 Inverted WVT rate: 1003.90 grams/day/m2Block #2 (continued)Water Vapor Transmission Rates & Inverted Cup Water Vapor Transmission RatesFL A N 20 D R 09 OFigure 77- Sample 2.5 WVT rate: 257.41 grams/day/m2 Inverted WVT rate: 154.45 grams/day/m2 Figure 79- Sample 2.7 WVT rate: 180.19 grams/day/m2 Inverted WVT rate: 308.89 grams/day/m2116- FLANDROFigure 78- Sample 2.6 WVT rate: 102.96 grams/day/m2 Inverted WVT rate: 102.96 grams/day/m2Figure 80- Sample 2.8 WVT rate: 128.71 grams/day/m2 Inverted WVT rate: 77.22 grams/day/m2Block #2- Water Vapor Transmission Rates & Inverted Cup Water Vapor Transmission Rates Analysis The Q-Test was used to eliminate any outliers in this data. The data for sample 2.4 on the inverted cup water vapor transmission was eliminated from the following analysis. The average water vapor transmission rate for this block was, 267.07 g/day/m2. The average inverted cup water vapor transmission rate was, 167.33 g/day/m2.FL A N 20 D R 09 O117- FLANDROIn accordance with the hypothesis, the data proved that crazing and crawling of a glaze will significantly increase the rate of water vapor transmissions in a terra-cotta block. The crazing and crawling provides a direct route from the body through the glaze layer for water vapor to exit the specimen. Block #2- Conclusions The experimentations performed here showed that crawling and crazing does expose the terracotta body to the exterior elements and crazing does continue through the glaze layer into the body. The crawling appeared to be a symptom of an ill fitting glaze layer, while the crazing appeared to be caused by a characteristic present in the clay body. Despite the fact that the glaze and body of this specimen were both significantly vitrified the water vapor transmission rates were high.Block #3FL A N 20 D R 09 OFigure 81- Terra-cotta Block #3118- FLANDROFigure 83- Building from which Block Figure 82- Building from which Block #3 originated. #3 originated. Building: 900 Riverside Drive, New York, NY 10032 Year Built: 1914 Architects: Harold L. Young Munsell Color (Glaze): 10YR, 8/1 Munsell Color (Body): 10YR, 8/2 Glaze: Matte with glossy black flecksBlock #3- Visual AnalysisFL A N 20 D R 09 OFigure 84- Crawling on Block #3, 25x magnification. Figure 85- Block #3, 50x magnification. Brown inclusion Figure 86- Block #3, 100x magnification.119- FLANDROGlossy black surface fleckBlock #3 surface observations: The glazed surface on this block appeared smooth at 25x magnification, however, at 50x magnification closely packed pits resulting in an overall rough surface is visible. There are two types of concentrated dark colors noted on the specimen. The brown colored dots are inclusions in the glaze body and are evenly shaped, sized and distributed. The black dots are more reflective, only on the surface, and also evenly shaped, sized and distributed. These observations suggest that the brown dotted areas are something that was mixed into the glaze body prior to its application, and that the black flecks were applied on top of the main glaze body, most likely with a machine due to their controlled characteristics. The visual analysis of this specimen also showed areas of crawling in the glaze. The crawls were all linear in nature and ranged from 1-3mm in size.Block #3- Thin Section AnalysisFL A N 20 D R 09 OFigure 87- Block #3, PPL, 50x magnification. Figure 88- Block #3, PPL, 100x magnification. Figure 89- Block #3, XPL, 50x magnification. Figure 90- Block #3, XPL, 100x magnification. Thin section observations Block #3: The glaze on this block is less vitrified than the clay body, and the clay body is more vitrified than previously examined specimens, with more and larger partially vitrified areas. In addition, it is unclear if the larger pieces of grog in the glaze have settled onto the clay body, creating a layer of larger sized glaze surrounded aggregate directly on top of the clay body surface or if this is a thin layer of a coarse slip. The visible (not vitrified) grog in this clay is also large in comparison with other samples (100-200 m).120- FLANDROBlock #3- Cross Section AnalysisFL A N 20 D R 09 OFigure 91- Block #3, 25x magnification. Figure 92- Block #3, 50x magnification. Figure 93- Block #3, 100x magnification.121- FLANDROBlock #3 cross section observations: Visible in cross section is just a layer of glaze, no slip layer, as is suggested by the thin-sections. The glaze is translucent in reflected light. The interface between the clay body and the glaze is smooth, suggesting that the body was smoothed prior to the glaze application. The glaze layer is approximately 200 m thick.Block #3Water Vapor Transmission Rates & Inverted Cup Water Vapor Transmission RatesFL A N 20 D R 09 OFigure 94- Sample 3.1 WVT rate: 386.12 grams/day/m2 Inverted WVT rate: 25.74 grams/day/m2 Figure 96- Sample 3.3 WVT rate: 102.96 grams/day/m2 Inverted WVT rate: 154.45 grams/day/m2122- FLANDROFigure 95- Sample 3.2 WVT rate: 77.22 grams/day/m2 Inverted WVT rate: 51.48 grams/day/m2Figure 97- Sample 3.4 WVT rate: 25.74 grams/day/m2 Inverted WVT rate: 25.74 grams/day/m2Block #3 (continued)Water Vapor Transmission Rates & Inverted Cup Water Vapor Transmission RatesFL A N 20 D R 09 OFigure 98- Sample 3.5 WVT rate: 77.22 grams/day/m2 Inverted WVT rate: 102.96 grams/day/m2 Figure 100- Sample 3.7 WVT rate: 540.56 grams/day/m2 Inverted WVT rate: 205.93 grams/day/m2123- FLANDROFigure 99- Sample 3.6 WVT rate: 25.74 grams/day/m2 Inverted WVT rate: 51.48 grams/day/m2Figure 101- Sample 3.8 WVT rate: 25.74 grams/day/m2 Inverted WVT rate: 77.22 grams/day/m2Block #3- Water Vapor Transmission Rates & Inverted Cup Water Vapor Transmission Rates Analysis The Q-Tests was used to eliminate any outliers in this data set. The data for sample 3.7 from the inverted cup water vapor transmission was eliminated from the following analysis. The average water vapor transmission rate for this block was, 157.66 g/day/m2 The average inverted cup water vapor transmission rate was, 69.87 g/day/m2FL A N 20 D R 09 OBlock #3- Conclusions124- FLANDROThe water vapor transmission rates demonstrated by this block are puzzling. The average inverted cup water vapor transmission rate is more similar to that of Block #1 in which no glaze fit defects are visible, however, the regular water vapor transmission rate was closer in comparison to those of Block #2, on which there were many types of glaze defects. This block is exhibiting water vapor transmission characteristics of both glaze fit situations.The examination of this block revealed that the body of the specimen was largely vitrified, and the glaze was not. The block does not have a slip layer, so any water vapor that is transmitted through the vitrified body can exit the specimen with realative ease through the un-vitrified glaze. This perhaps may be the reason for the dual characteristics exposed through the water vapor transmission rate tests.Block #4FL A N 20 D R 09 OFigure 102- Terra-cotta Block #4 Figure 103- Building from which Block #4 originated. Building: The Alameda 255 West 84th Street (or 2321 Broadway), New York, NY 10024 Year Built: 1914 Architect: Gaetan Ajello Munsell Color (Glaze): 10YR, 8/1 Munsell Color (Body): 10YR, 8/3 Glaze: Matte125- FLANDROBlock #4- Visual AnalysisFL A N 20 D R 09 OFigure 104- Block #4, 25x magnification. Figure 105- Block #4, 50x magnification. Figure 106- Block #4, 100x magnification. Figure 107- Block #4, 175x magnification. Block #4 surface observations: The surface of the glaze is pitted and rough. Similar to Block #1, there are raised areas of a translucent glassy material. At 175x magnification a white powdery looking substance could be seen in the inside of the pitting. A yellowing was also noted in some areas (visible at 100x magnification in Figure 106). Minimal crazing is visible (Figure 104).126- FLANDROBlock #4- Thin Section AnalysisFL A N 20 D R 09 OFigure 108- Block #4, PPL, 50x magnification. Figure 109- Block #4, PPL, 100x magnification.127- FLANDROFigure 110- Block #4, XPL, Figure 111- Block #4, XPL, 50x magnification. 100x magnification. Thin section observations Block #4: The glaze on this clay body is close to a completly vitrified state, showing up as almost completley black under PPL. Also visible in XPL is an area of transformation between the clay body and the glaze, there is no vitrification of materials in this area (visible in Figure 111). Interestingly, the glaze on this piece of terracotta is much more vitrified than Block #1 which came from the same building, even the clay bodies look to be of different compositions. The difference in vitrification and composition is not in any way distinguishable by the naked eye.Block #4- Cross section AnalysisFL A N 20 D R 09 OFigure 112- Block #4, 25x magnification. Figure 113- Block #4, 50x magnification. Figure 114- Block #4, 100x magnification.128- FLANDROBlock #4 cross section observations: This specimen has only a glaze layer with no slip layer. The glaze is an opaque gray color, with a rough exterior facing surface. The interface of the glaze and the clay body is smooth. The glaze thickness is 200 m thick.Note: Water vapor transmission and inverted cup water vapor transmission tests were not performed on this block of terracotta. The majority of the glazed surface area of this block had been attacked and the glaze had severely pitted and bubbled. There was only enough undamaged area to extract samples for thin section and cross section analysis.Block #4- Conclusions Although water vapor transmission testing was not done on this specimen, there was great value in completing the other tests. The thickness of glaze and also the complete vitrification of the glaze would not have been otherwise determinable. The glaze on this specimen look very powdery and matte with limited magnification and without the thin section analysis would have probably have been assumed to be not completely vitrified. In addition an important concept to take away from this blocks analysis is that characteristics noted by the naked eye can be deceiving. Block Blocks #1 and #4 look identical in visual analysis, but completely different in microscopic examinations.FL A N 20 D R 09 O129- FLANDROBlocks #5 & #6Block #6FL A N 20 D R 09 OFigure 115- Terra-cotta Blocks #5 and #6. Figure 116- Building from which Blocks #5 and #6 originated. Building: 53 Park Place (or 51 West Broadway), New York, NY 10007 Year Built: 1921 Architects: Shape, Beady & Peterkin, and Cross & Cross Munsell Color (Glaze): 10YR, 7/4 Munsell Color (Body): 10YR, 8/3 Glaze: Matte with black and white glossy flecks.130- FLANDROBlock #5Block #6- Visual AnalysisFL A N 20 D R 09 OFigure 117- Block #6, 25x magnification. Figure 118- Block #6, 50x magnification.131- FLANDROFigure 119- Block #6, Figure 120- Block #6, 100x magnification. 175x magnification. Block #6* surface observations: The glazed surface area is a compilation of three different glazes. There is the base glaze, a white fleck glaze and also a black fleck glaze. The white flecks are higher than the average surface level, however they are more linear in shape than the black flecks. The white flecks are not glossy. The black flecks are convex and are higher than the average surface level and have a glossy sheen. The black flecks appear to be a black pigment of sorts encapsulated in a glassy matrix. Both types of flecks are not evenly spaced, sized or shaped, suggesting the possibility of hand application after the base glaze was applied. The base glaze is not glossy but has a relatively smooth surface when compared with blocks 1, 3 and 4. It was also noted that the walls of glaze created by the crawling were very glossy, (visible in Figure 120 ).*Block #5 was not photographed as a cleaner surface area was available on block #6 and they came from the same building. Block 5 and 6 both had a greasy mold release agent on them prior to their acquisition and were therefore not used for water vapor transmission and inverted cup water vapor transmission. The photographs here of block #6 were taken from the side of the block, where no release agent was present.Block #6- Thin Section AnalysisLayer-1 Layer-2? Layer-3?FL A N 20 D R 09 OFigure 121- Block #6, PPL, 50x magnification. Figure 122- Block #6, PPL, 100x magnification. Figure 123- Block #6, XPL, 50x magnification. Figure 124- Block #6, XPL, 100x magnification. Block #6 thin section observations: The glaze on this block is not vitrified, but the clay body does have large areas that are muddy and transitioning into a more glassy state. There is a slip coat visible but it is unclear if the larger grog in the glaze has migrated to the bottom of the glaze layer or if this is an additional layer of aggregate between the slip coat and the glaze (noted in Figure 122). The different types of grog, distinguished in XPL, in this specimen are finely ground and evenly distributed throughout the clay and glaze bodies.132- FLANDROBlock #6- Cross section AnalysisFL A N 20 D R 09 OFigure 125- Block #6, 25x magnification. Figure 126- Block #6, 50x magnification. Figure 127- Block #6, 100x magnification.133- FLANDROBlock #6 cross section observations: The glaze on this sample is translucent and also has consistent oscillations between thicknesses, some areas are 200 m thick, while the thinner areas are around 150 m thick. The mechanisim for developing these oscillations in the glaze in unknown. The slip layer noted in the thin-section analysis is not visible in cross-section.Block #5- Thin Section AnalysisFL A N 20 D R 09 OFigure 128- Block #5, PPL, 50x magnification. Figure 129- Block #5, PPL, 100x magnification. Figure 130- Block #5, XPL, 50x magnification. Figure 131- Block #5, XPL, 100x magnification. Block #5 Thin section observations: The glaze on this specimen is not vitrified. There is a slip coat visible between the glaze and the clay body, marked by a distinct brown line (Figure 129)*. The clay body does have large areas of partial vitrification (visible in Figures 129 & 131). The grog ingredients in both the body and the glaze are a range of different sizes, which are evenly dispersed. * Upon the examination of the specimen in XPL it was determined that the slip coat specified was in fact not a slip coat but the top portion of the terra-cotta body.134- FLANDROBrown lineBlock #5- Cross section AnalysisFL A N 20 D R 09 OFigure 132- Block #5, 25x magnification. Figure 133- Block #5, 50x magnification. Glaze Transitional body Body Figure 134- Block #5, 100x magnification.135- FLANDROBlock #5 cross section observations: The glaze on this specimen in reflected light is a mixture of opaqueness in some areas and translucency in others. The glaze thickness varies between 100-200 m thick, with the thinner areas being more translucent. While the slip layer is not readily noticeable upon closer examination (moving the specimen around to look at different areas while under the microscope) it appeared that it is not a slip layer but rather a transitional phase (figure 134) of the clay body (possibly a fire skin). What is not distinguishable is the layer of large aggregate in the glaze which was noted in thin section. There is no slip layer on this specimen.Block #5 and #6- ConclusionsSpecimens #5 & #6 both had similar degrees of vitrification, and also similar layer organizations. The glaze layer thickness is 200 m. What should be taken from this examination is that both XPL and PPL are necessary to determine if a layer is a slip coat or a fireskin and the cross-section can be used to confirm these findings.FL A N 20 D R 09 O136- FLANDROFL A N 20 D R 09 OCategory 1 Specimens: No Fit DefectFigure 135- Sample 1.1: WVT rate: 51.48 grams/day/m2 Inverted WVT rate:51.48 grams/day/m2Water Vapor Transmissions Comparative Analysis: The twenty four specimens were divided into five categories based on their glaze fit characteristics visible with the naked eye. The categories were: Category 1: no glaze fit defects visible Category 2: Crazes only Category 3: Crawls only Category 4: Crawls and Crazes Category 5: Surface Scratches on the glaze (body exposed)Figure 136- Sample 1.2 WVT rate: 25.74 grams/day/m2 Inverted WVT rate: 51.48 grams/day/m2Figure 137- Sample 1.3 WVT rate: 77.22 grams/day/m2 Inverted WVT rate: 77.22 grams/day/m2Figure 138- Sample 1.4 WVT rate: 51.48 grams/day/m2 Inverted WVT rate: 25.74 grams/day/m2Figure 139- Sample 1.5 WVT rate: 0 grams/day/m2 Inverted WVT rate: 25.74 grams/day/m2Figure 140- Sample 1.6 WVT rate: 51.48 grams/day/m2 Inverted WVT rate: 25.74 grams/day/m2Figure 141- Sample 1.7 WVT rate: 77.22 grams/day/m2 Inverted WVT rate: 25.74 grams/day/m2Figure 142 - Sample 1.8 WVT rate: 25.74 grams/day/m2 Inverted WVT rate: 102.96 grams/day/m2Figure 143- Sample 3.2 WVT rate: 77.22 grams/day/m2 Inverted WVT rate: 51.48 grams/day/m2137- FLANDROCategory 1 Specimens: No fit Defect (continued)FL A N 20 D R 09 OFigure 144- Sample 3.4 WVT rate: 25.74 grams/day/m2 Inverted WVT rate: 25.74 grams/day/m2Figure 145- Sample 3.5 WVT rate: 77.22 grams/day/m2 Inverted WVT rate: 102.96 grams/day/m2Figure 147- Sample 3.8 WVT rate: 25.74 grams/day/m2 Inverted WVT rate: 77.22 grams/day/m2Figure 146- Sample 3.6 WVT rate: 25.74 grams/day/m2 Inverted WVT rate: 51.48 grams/day/m2Category 2 Specimens: Crazes onlyFigure 148- Sample 2.4 WVT rate: 411.86 grams/day/m2 Inverted WVT rate: 1003.90 grams/day/m2Figure 149- Sample 2.6 WVT rate: 102.96 grams/day/m2 Inverted WVT rate: 102.96 grams/day/m2Figure 150- Sample 2.7 WVT rate: 180.19 grams/day/m2 Inverted WVT rate: 308.89 grams/day/m2Figure 151- Sample 3.1 WVT rate: 386.12 grams/day/m2 Inverted WVT rate: 25.74 grams/day/m2138- FLANDROCategory 3 Specimens: Crawls OnlyFL A N 20 D R 09 OFigure 152- Sample 2.1 WVT rate: 489.10 grams/day/m2 Inverted WVT rate: 102.96 grams/day/m2Figure 153- Sample 2.2 WVT rate: 180.19 grams/day/m2 Inverted WVT rate: 347.60 grams/day/m2Figure 154- Sample 2.3 WVT rate: 386.12 grams/day/m2 Inverted WVT rate: 77.22 grams/day/m2Category 4 Specimens: Crazes and CrawlsFigure 155- Sample 2.5 WVT rate: 257.41 grams/day/m2 Inverted WVT rate: 154.45 grams/day/m2Figure 156- Sample 2.8 WVT rate: 128.71 grams/day/m2 Inverted WVT rate: 77.22 grams/day/m2Figure 157- Sample 3.3 WVT rate: 102.96 grams/day/m2 Inverted WVT rate: 154.45 grams/day/m2Category 5 Specimen: Surface scratches on the glaze (body exposed)Figure 158- Sample 3.7 WVT rate: 540.56 grams/day/m2 Inverted WVT rate: 205.93 grams/day/m2139- FLANDROWater Vapor Transmissions Comparative Analysis: After using the Q-Test to detect outliers and none were located, the following averages were calculated for each category: Category 1: No fit defect Average water vapor transmission rate: 45.54 g/day/m2 Average inverted cup water vapor transmission rate: 53.46 g/day/m2FL A N 20 D R 09 OCategory 2: Crazes only Average water vapor transmission rate: 270.28 g/day/m2 Average inverted cup water vapor transmission rate: 360.38 g/day/m2 Category 3: Crawls only Average water vapor transmission rate: 163.03 g/day/m2 Average inverted cup water vapor transmission rate: 128.71 g/day/m2 Category 4: Crazes and Crawls Average water vapor transmission rate: 351.80 g/day/m2 Average inverted cup water vapor transmission rate: 175.93 g/day/m2 Category 5: Scratches in the glaze (body exposed) Average water vapor transmission rate: 540.56 g/day/m2 Average inverted cup water vapor transmission rate: 205.93 g/day/m2140- FLANDROConclusions: What is made most apparent by this comparative analysis is that any fit defect in the glaze will greatly increase the rate of water vapor transmissions. This is true in both types of transmission.Material Testing- ConclusionThe application of the following tests were found to be highly beneficial and educational about specific pieces of terracotta: visual analysis, petrographic examination, cross-section examinations and both water vapor transmission testing and inverted cup water vapor transmission testing. The visual analysis was useful for learning more about the glaze fit condition as well as the mode of glaze application on a specimen. Just by looking closely at the glazes certain characteristics can be identified and described. Some of these characteristics include crawling, crazing, sheen and hand or machine application. Glaze fit defects result in high rates of water vapor transmissions. The petrographic analysis was used mainly to determine the amount of vitrification that had taken place, operating under the assumption that the more vitrified a material is the less water vapor is transmitted. However, these two characteristics were found to be unrelated in this particular study. However, during the petrographic analysis the thin sections were found to be useful for identifying the depth of penetration of crazes, the exposure of the terracotta body by crawls and the identification of a fireskin and a slip layer. In addition it was noted that the less vitrified the glaze was on a sample the more translucent the layer appeared in reflected light. It should be noted that while only six blocks were tested for this thesis only two had glazes that were vitrified, Block #2 (ca. 1910) and Block #4 (ca. 1914). The thin section examinations were initially done to determine the depth of crazing, but was found to be an insufficient testing method to determine this. What the thin section did aid with was determination of the presence of a slip layer and also the thickness of the layers. In this examination every glaze layer was approximately 200 m thick, suggesting a mechanical means of application (but not the pulsichrometer which was not used until 1922. Water vapor transmission testing was done to prove that water vapor could get through a glaze on a terracotta specimen. What was revealed was that water vapor did indeed get through, but also that the rates at which the water vapor was transmitted was much higher in specimens with glaze fit defects. This test also killed the notion that glazed terracotta on historic buildings breaths only through the joints (although this may still be possible with an absolutely perfect glazed block).FL A N 20 D R 09 O141- FLANDROFL A N 20 D R 09 O142- FLANDROThe inverted cup water vapor transmission was done because of the way historic architectural terracotta was installed on buildings. The blocks were back filled with bricks and/or mortar, often insufficiently to keep water out, water often penetrates into the interior of the block and puddles when gravity doesnt provide an exit route. It was determined that the rate of water vapor transmissions from this sort of condition where the water is in direct contact with the body were equal to or close to the rates in the regular water vapor transmission test, where the water is not in contact with the body. The aforementioned information was gained through relatively small and simple to extract samples, and inexpensive and easy to perform examinations, all of which can be done by conservators prior to specifying treatments.143- FLANDROFL A N 20 D R 09 OFL A N 20 D R 09 ORECOMMENDATIONS144- FLANDROFL A N 20 D R 09 O145- FLANDROFor practicing architectural conservators the following tests should be performed on terracotta units prior to the application of conservation treatments or replacement: Visual analysis petrographic examination, cross-section analysis At least one of the two forms of water vapor transmission testing Visual Analysis! At the very least on the site analysis of in situ blocks with a hand lenses, looking specifically for glaze fit defects, prior to applying a coating or patch of any sort. It would better to take samples back to the laboratory for observations under higher magnifications. In addition to looking at the historic material, inspecting any new replacement units for crazes after they have had time sit and expand due to water of hydration (which is reviewed in Section I). Crazes are easiest to identify when they are soiled so taking a few sample blocks from the delivery and rubbing ink on the surface (non reversible, so dont do this with blocks to be installed on the building) will reveal any crazing. It should also be requested by the conservator that the specifics of the replacement blocks firing practices as well as the mixes used to make the clay and glazes be provided with the delivery of blocks to insure that each block is identical in makeup, which will make spotting and solving problems easier. The other four tests can be completed on samples taken from in situ blocks, using the best practices of sample taking (taking multiple samples from different places exposed and currently under different conditions). The assumption should never be made that all terra cotta is the same. The use of the laboratories employed in this study in addition to the eleven other tests suggested by Susan Tindall will ensure that site specific characteristics of the terracotta are properly identified. The assumption should never be made that all terracotta is the same and, as the testing here has demonstrated, conservators should keep in mind that multiple micro-climates (glaze fit, different levels of vitrification etc.) can co-exsist in one block of terracotta.Recommendations:A few suggestions for further investigations in relation to this topic: The relation between water vapor transmission rates and levels of vitrification, both of the glaze and body. If the use of a slip coat prior to the application of a glaze causes better or worse adhesion of the glaze (specifically if there is increased incidents of crazing and or crawling when a slip coat is employed) Historical research outlining the chronological development of kilns used for glazed terracotta in the United States. Determination of the water vapor transmission rates of new specimens of glazed terracotta, showing no signs of any glaze fit problems. Determination of the rate of deterioration of a terracotta body in direct extended contact with water. A findings comparrison of glazed terra cotta and terra-cotta patching mortars.FL A N 20 D R 09 O146- FLANDRO147- FLANDROFL A N 20 D R 09 OFL A N 20 D R 09 OCONCLUSION148- FLANDROFL A N 20 D R 09 O149- FLANDROConclusion This thesis proved that the glaze fit on architectural terra cotta has been a source of concern and interest for manufactures, researchers and conservators alike for over a hundred years. This paper solved two terra-cotta mysteries. First the mystery behind the development of terracotta testing procedures has been flushed out with the chronological time line of studies relating specifically to architectural terracotta. The time line provides an easily accessible way to read about the progression of terra-cotta testing. Secondly, the previously untried basic examination program developed specifically for glazed terracotta conservation was proven to be successful. Easy tests revealed that water vapor transmissions does occur through the glaze, with an increase in the rate showing as a result of terracotta specimens with glaze fit defects. The problem that was posed at the beginning of this research was one of wanting to submit patching compounds to the exact examinations that glazed terracotta was being subjected to. This direction was immediately halted when it was discovered that glazed terracotta was not being tested! The next logical thing to do was to come up with testing procedures to be applied to terra cotta. Rather than grab randomly at the thousands of testing procedures available in todays conservation and construction industries, it made more sense to look for historic testing trials; what the procedures were? If they could be duplicated? And what the findings were? It was concluded that testing on glazed terracotta started around the turn of the twentieth century. Branching quickly from the brick industry as more builders and architects became interested in the material and its inherent characteristics. The peak in the research studies coincided with the peak of its production between 1915 and 1925, after which the studies were less frequent and completely halted by the early 1960s. Even with the upsurge of interest in the historic built world, the beginning of historic preservation in the late 1960s and a span of sixty years after the last complete testing document was published, glazed terra cotta and most of its basic properties continues to remain a mystery. While every property cannot be determined all at once, conservators can start by doing their part and properly (if belatedly) inserting a testing protocol for glazed terra cotta into their best practices. To allow for the proper and non-destructive conservation of terra cotta on historic structures it is necessary to first complete investigative studies. While specific experiments were listed by Susan Tindall twenty one years ago they are rarely (if ever) completed. It is be-lieved that the practice of having no testing program in place for glazed terra cotta occurred because of the lack of testing procedures provided with this list, not knowing what the test results would do to help the conservation and also the expense of the equipment involved. This study has attempted to rectify some of these issues by providing a more basic and tried testing program, including procedures and expected outcomes. The implementation of this simple program could save thousands of historic glazed terra-cotta blocks from needless treatment or destruction.FL A N 20 D R 09 O150- FLANDRO151- FLANDROFL A N 20 D R 09 OFL A N 20 D R 09 OBIBLIOGRAPHY152- FLANDROBibliographyAmerican Architect and Building News. The Manufacture of Terra-Cotta in Chicago, (December 30, 1876), pages 420-425. American Institute of Architects. Public Works Specifications Ceramic Veneer A.I.A File No. 9. October 1961. American Society for Testing and Materials. Standard Terminology of Ceramic Whitewares and Related Products, Standard C242-01. ASTM Standards. ASTM International, PA. (2005), pages 77-86. American Society for Testing and Materials. Standard Terminology Relating to Surface Imperfections on Ceramics, Standard F109-04. ASTM Standards. ASTM International, PA. (2005), pages 507-511.FL A N 20 D R 09 O153- FLANDROAnderson, Louis. Comparison of Methods Used in Estimating the Maturing of Terra Cotta, The Journal of the American Ceramic Society. (Volume 8, Issue 11, November 1925), pages 762-762. The Architectural Record. The Smaller Houses of the English Suburbs, (July 1876), Page118. The Architectural Record. Glazed and Colored Terra Cotta, (April 1906), pages 313323. Barber, Edwin Atlee. Ceramic Embellishment in Architecture, The Clay-Worker. (Volume 27/28, 1897), pages 511-515.Bell, W.C.. Evaluation of Glaze Fit Test Methods, The Journal of the American Ceramic Society. (Volume 23, Issue 6, June 1940), pages 163-166. Bennett, Lee. The Use of Overglazes for Polychrome Terra Cotta, The Journal of the American Ceramic Society. (Volume 9, Issue 4, January 1926), pages 187-188. Blakely, A.M.. Life History of a Glaze: II, Measurement of Stress in a Cooling Glaze, The Journal of the American Ceramic Society. (Volume 21, Issue 7, July 1938), pages 243-251. Binns, Charles F.. The Development of the Matt Glaze, Transactions of the American Ceramic Society. (Volume V, 1903), pages 50-63. Binns, Charles F.. Lessons from the St. Louis Exposistion, 1904, Transactions of the American Ceramic Society. (Volume VII, 1905), pages 47-55. Bleininger A.V.. Ceramic Investigations Conducted by the States and the Federal Government, The Journal of the American Ceramic Society. (Volume 6, Issue 1, January 1923), pages 145-156.Bleininger A.V. and H.H. Clark. Note on the Viscosity of Clay Slips As Determined by the Clark Apparatus, Transactions of the American Ceramic Society. (Volume 7, February 1910), pages 383-391. Brickbuilder. Architectural Terra-Cotta, Volume 2 (February 1893), pages 11-14. Brickbuilder. Architectural Terra-Cotta, Volume 2 (March 1893), pages 21-23. Brickbuilder. Polychrome Terra Cotta in the Masonic Temple in Brooklyn, (Volume 20, April 1911), pages 79-90. Brickbuilder. Program which Governed the Competition for a Hotel in an American City of Moderate Size. (Volume 20-plates 1911), pages 1-5. {The exterior of the building had to be designed entirely in architectural terra cotta}. Buell, R.N.. Terra Cotta Construction, The Clay-Worker. (Volume 27/28, 1897), pages 191-192. Bulletin of the American Ceramic Society. Bureau of Standards Keeps Prestige Despite Fund Cut, Belgian Ceramic Journal Changes Name, Walter Suter at New York Ceramic Studios, (Volume 13, 1934), pages 142-143. Carruthers, John L.. Notes on Shivering of Terra Cotta, The Journal of the American Ceramic Society. (Volume 5, Issue 8, August 1922), pages 518-527.FL A N 20 D R 09 OThe Clay-Worker. Advantages of Terra Cotta, (Volume 43, 1905), page 353. The Clay-Worker. Beauties of Terra Cotta, (Volume 37/38, 1902), March 1902.154- FLANDROCathedral Stone Products. Jahn M100 Terra Cotta and Brick Repair Mortar, www.cathedralstone.com, 2009. Clare, R.L.. Causes for the Failure of Terra Cotta in the Wall, Transactions of the American Ceramic Society. (Volume 19, 1917), pages 594-596. Clare, R.I.. Discussion of the Effect of De-Airing on the Fired Properties of a Terra Cotta Body. The Bulletin of the American Ceramic Society. (Volume 13, 1934), pages 223-224. Clare, R.L.. Standardization of Tests for Terra Cotta Bodies and Glazes, The Bulletin of the American Ceramic Society. (Volume 13, Issue 12, December 1934), page 333. Clark, Ernest. The Use of Pulsichrome From a Manufactures Standpoint, The Journal of the American Ceramics Society. (Volume 5, Issue 11, November 1922), pages 826-827.The Clay-Worker. Bright Outlook for Terra Cotta, (Volume 45/46, 1906), pages 485486. The Clay-Worker. Buffalos Great Commercial Structures, (Volume 27/28, 1897), page13-14. The Clay-Worker. Chicagos Latest Architectural Achievement, (Volume 27/28, 1897), pages 283 and 332. The Clay-Worker. Clay in Architecture, (Volume 27/28, 1897), page 172. The Clay-Worker. Clay, Art & Architecture, (Volume 29/30, 1898), page 382. The Clay-Worker. Clay in Architecture, (Volume 29/30, 1898), page 136. The Clay-Worker. Clay Products of the United States, (Volume 51, 1909), pages 431433. The Clay-Worker. Columbus, The Convention City, (Volume 49/50, 1908), pages 1 and 22-30. The Clay-Worker. The Corning Brick and Terra Cotta Company, (Volume 47/48, 1907), pages 256-259. The Clay-Worker. The Cost of Terra Cotta and Stone, (Volume 29/30, 1898), page 212. The Clay-Worker. Crazing-Its Causes and Remedy Wanted, (Volume 35/36, 1901), page 43. The Clay-Worker. Enameled Terra Cotta, (Volume 27/28, 1897), page 173. The Clay-Worker. The First Job Turned Out by the Western Terra Cotta Company, (Volume 47/48, 1907), page 514. The Clay-Worker. A Fitting Use for Terra Cotta, (Volume 29/30, 1898), pages 93-94. The Clay-Worker. An Idyll(er) in Terra Cotta, (Volume 27/28, 1897), pages 270-275. The Clay-Worker. The Machinery Annex. (Volume 27/28, 1897), page 161. The Clay-Worker. The Manufacture of Architectural Terra Cotta in America. (Volume 43, 1905), pages 695 and 754. The Clay-Worker. The Maryland Terra Cotta Company, (Volume 49/50, 1908), pages 65-66. The Clay- Worker. Modern American Architecture: Advantage of Enameled Brick and Terra Cotta for Exterior Walls, (Volume 39/40, 1903), pages 69 and 126-127. The Clay-Worker. Modern Terra Cotta Construction- How Good Work is Marred Sometimes, (Volume 27/28, 1897), page 264. The Clay-Worker. More Queries About Glazing, Crazing Sometimes Considered a Merit, Unequal Expansion of Body and Glaze, Cause of Crazing and Remedy, (Volume 35/36, 1901), pages 137-138.FL A N 20 D R 09 O155- FLANDROThe Clay-Worker. A New Architecture, (Volume 29/30, 1898), page 381. The Clay-Worker. The New York Subways: Where Immense, Quantities of Brick, Terra Cotta and Faience Have Been Used, (Volume 41, 1904), pages 705-706. The Clay-Worker. The O.W. Ketcham Terra Cotta Company, (Volume 47/48, 1907). The Clay-Worker. Perfection in Architecture: The Artistic Side of Brick and Terra Cotta Construction Exemplified, (Volume 51, 1909), pages 29-31. The Clay-Worker. A Prosperous New Jersey Terra Cotta Plant, (Volume 42, 1904), page 149. The Clay-Worker. A Self-Cleaning Building, (Volume 39/40, 1903), page 172.FL A N 20 D R 09 OThe Clay-Worker. Terra Cotta Building, (Volume 35/36, 1901), page 508. The Clay-Worker. Terra Cotta Fronts, (Volume 29/30, 1898), pages 261-263. The Clay-Worker. Terra Cotta to the Front, (Volume 45/46, 1906), page 403.156- FLANDROThe Clay-Worker. Terra Cotta and Brick in Architecture, (Volume 37/38, 1902), pages 234-239. The Clay-Worker. Terra Cotta at the Worlds Fair. (Volume 42, 1904), pages 140-141.The Clay-Worker. Terra Cotta Building Blocks, (Volume 27/28, 1897), page 345.The Clay-Worker. Terra Cotta in Architecture. (Volume 49/50, 1908), pages 31-35. The Clay-Worker. Terra Cotta in Landscape Architecture, (Volume 43, 1905), page 467. The Clay-Worker. A Terra Cotta Mans Hopes and Fears, (Volume 29/30, 1898), page 426.The Clay-Worker. Terra CottaThe Material Grows in Usefulness in the Builders Hands, (Volume 27/28, 1897), page 34.The Clay-Worker. Theres Music in the Air at Terra Cotta, (Volume 39/40, 1903), page 441. The Clay- Worker, Tile, Brick and Terra Cotta at the University of Illinois, (Volume 27/28, 1897), page 182. The Clay-Worker. Questions for the Terra Cotta Manufacture, (Volume 42, 1904), page 57. The Clay-Worker. Wants to Buy Bella Robbia, (Volume 47/48, 1907), page 52C. The Clay-Worker. Waterproofing Clay Products, (Volume 35/36, 1901), page 594.The Clay-Worker. Why Staff, Why Not Terra Cotta?, (Volume 37/38, 1902), page 56. The Clay-Worker. Winkler Terra Cotta Works, (Volume 41, 1904), 598- 601. Conkling, I.L.. Kilns for Burning Architectural Terra Cotta, Transactions of the American Ceramic Society. (Volume VIII, 1906), pages 125-146. Conproco Corporation. Terracotta Finish, Mimic TR, www. Conproco.com, 2009. Cox, Paul E.. Vitrified Terra Cotta, The Journal of the American Ceramic Society. (Volume 10, Issue 8, August 1931), pages 265-267. Davis, H.E.. A Low Temperature Comparator for the Coefficient of Expansion of Ceramic Bodies, The Journal of the American Ceramic Society. (Volume 16, Issue 12, December 1933), pages 634-639.FL A N 20 D R 09 O157- FLANDRODavis, H.E. and J.S. Lathrop. A Study of Slips for Standard Finish Terra Cotta, The Journal of the American Ceramic Society. (Volume 8, Issue 1, January 1925), pages 23-28. Durand, A.. Architectural Terra Cotta, Brickbuilder. (Volume 20-plates, 1911), Pages 63-66. Edison Coatings Incorporated. Thin-Fill 55, Custom System 45, Aqua-Spex 220, Aquathane UA-210, Guide to the Repair of Architectural Terra Cotta, www. edisoncoatings.com, 2009. Edison, Michael P.. Color and Long Term Color Retention in Composite Patching Systems for Stone and Masonry, Association for Preservation Technology (APT) Bulletin. (Volume 33, Number 2/3, 2002), pages 23-32. Eskesen, Eckardt V. How Terra-Cotta Is Made, Real Estate Record and Builders Guide. (April 3, 1909), page 643-644. Eskesen, Eckardt V..The Investigation of Terra Cotta Work at the Bureau of Standards, The Journal of the American Ceramic Society. (Volume 7, Issue 5, May 1924), pages 158-161. Eskesen, Eckardt V. The Investigation of Terra Cotta Work at the Bureau of Standards. Journal of the American Ceramic Society. (Volume 4. 1924), page 158. Eskesen, Eckardt V.. Water-Tight Terra Cotta Construction, The Bulletin of the American Ceramic Society, (Volume 13, 1934), pages 154-163. Everhart, Otis. Some Simulative Service Tests for Glazed Building Materials, The Journal of the American Ceramic Society. (Volume 13, Issue 6, June 1930), pages 404410.Ferrand, Gabriel. Ceramic Art and Architecture, The Journal of the American Ceramic Society. (Volume 5, Issue 11, November 1922), pages 743-757.Fetterolf, L.D.. A Study of the Consistencies of Raw Terra Cotta Glazes, The Journal of the American Ceramic Society. (Volume 13, Issue 9, 1930), pages 675-682. Fidler, John. The Conservation of Architectural Terracotta and Faience, Transactions. Association for Studies in the Conservation of Historic Buildings, Volume 6 (1981), pages 2-16. Fidler, John. The Manufacture of Architectural Terracotta and Faience in the United Kingdom. Association for Preservation Technology (APT) Bulletin. Volume 15, No.2 (1983), pages 27-32. Fidler, John. Glassfibre-Reinforced Plastic and Cement Facsimiles in Building Restoration. Transactions. Association for Studies in the Conservation of Historic Buildings, Volume 12 (1987), pages 17-24. Foster, E.S.. Organic Agents as Aids to the Adhesion and Suspension of Glazes, The Journal of the American Ceramic Society. (Volume 12, Issue 4, April 1929), pages 264-269. Frechette V.D. and T.A. Michalske. Glaze-Stress Calculations From Ring-Test Data, The Journal of the American Ceramic Society. (Volume 63, Issue 11, November 1980), pages 712-713. Geijsbeek, S.. The Definition and Application of Terra Cotta, The Clay-Worker. (Volume 49/50, 1908), pages 255-259. Gibson, L.H.. A House Built of Terra Cotta, The Clay-Worker, (Volume 35/36, November 1901). Gingold, J.J. and E.J. Hirsh. Influence of Glaze on Strenght of Steatite Tubes and New Method of Determining Glaze Stresses, The Journal of the American Ceramic Society. (Volume 28, Issue 6, June 1945), pages 168-170. Gladding, McBean. Terra Cotta: Our Process. www.GladdingMcBean.Paccoast.com, 2007.FL A N 20 D R 09 O158- FLANDROGoodrich, H.R.. Miscellaneous Terra Cotta Laboratory Notes, The Journal of the American Ceramic Society. (Volume 15, Issue 7, July 1932), pages 382-385. Geer, Walter. The Story of Terra Cotta. Tobias A. Wright Printer and Publishers, New York, NY, 1920. Hall, F.P. The Influence of Chemical Composistion on the Physical Properties of Glazes, The Journal of the American Ceramic Society. (Volume 13, Issue 3, March 1930), pages 182-199. Harman, C.G.. Suggestions for Solution of Difficult Glaze-Fit Problems, The Journal of the American Ceramic Society. (Volume 27, Issue 8, August 1944), pages 231233.Harrison, A.C.. A Preliminary Study of the Resistance to Abrasion of Ceramic Glazes, Its Control and Methods of Determination, The Journal of the American Ceramic Society. (Volume 10, Issue 2, February 1927), pages 77-89. Heath, Fred T.. A Standard Unit of Dimension for Masonry, The Journal of the American Ceramic Society. (Volume 12, Issue 10, October 1929), pages 605-639. Hensel, Otto.. The Use of Glazed Clay-Ware as an Architectural Decoration in Exterior Vs. Interior Work, Transactions of the American Ceramic Society. (Volume III, 1901), pages 153-158.FL A N 20 D R 09 O159- FLANDROHewlett, Monroe J. Polychrome Terra Cotta in Exterior Architecture, Brickbuilder. (Volume 20, April 1911), pages 71-72. Hill, E.C. The Effect of Glaze Composition on the Crazing of Terra Cotta, The Journal of the American Ceramic Society. (Volume 4, 1921), pages 23-31.Hill, E.C. The Effect of Some Fluxes in the Absorption and Transverse Strength of a Terra Cotta Body, The Journal of the American Ceramic Society. (Volume 19, 1922), pages 832-836. Hill, E.C.. Some Data on the Development of Terra Cotta Glazes, The Journal of the American Ceramic Society. (Volume 3, Issue 1, January 1920)., pages 13-15. Hoigard, Kurt R. and G.R. Mulholland and R.C. Haukohl. Terra Cotta Facades, Symposium on Building Faade Maintenance, Repair and Inspection, ASTM STP 1444. J.L. Erdlv and T.A. Schwartz, Editors, ASTM International, West Conshohocken, PA, (2004), pages 75-90.Horie, C.V.. Materials for Conservation. Massachusetts, Butterworth-Heinemann, 2000. Howe, Raymond and Robert F. Ferguson. A Study of Spalling, The Journal of the American Ceramic Society. (Volume 4, 1921), pages 32-46. Howe, Samuel. Polychrome Terra Cotta, The American Architect. (February 28, 1912), pages 99-105.Hull, Walter. New Principles of Tunnel Kiln Design Applied in a Kiln for Architectural Terra Cotta., The Bulletin of the American Ceramic Society. (Volume 13, 1934), pages 196-200. Humphrey, H.P.. Notes on the Burning of Terra Cotta in Open Kilns, Transactions of the American Ceramic Society, (Volume IX, 1907), pages 661-674. Jacob, Judith and Norman R. Weiss. Laboratory Measurement of Water Vapor Transmission Rates of Masonry Mortars and Paints, APT Bulletin (Vol. 21, No. , 1989) pp. 62-70 Jacob, Judith M. Inpainting Areas of Glaze Loss on Architectural Terra Cotta : A Study ofSix Proprietary Masonry Paint Systems. Columbia University, 1989. Joiner, Joseph. Terra Cotta in Architecture, The Clay-Worker, (Volume 29/30, 1898), pages 416-418. Joiner, Joseph. Terra Cotta in Architecture, The Clay-Worker, (Volume 29/30, 1898), pages 455 and 495-496, June 1898. Journal of the American Ceramic Society. Discussion on Notes on Shivering of Terra Cotta. (Volume 5, Issue 11, November 1922), pages 301-303. Kavenagh, Claudia and George Wheeler. Evaluation of Cleaning Methods for the Exterior Brick at the Brooklyn Historical Society Journal of the American Institute for Conservation Volume 42 Number 1, Architecture Issue, (Spring 2003), pages 97-112. Keller, R.B.. Recent Improvements in Methods of Terra Cotta Mold Making, Transactions of the American Ceramic Society. (Volume 19, 1917), pages 585-592.FL A N 20 D R 09 OKoenig, C.J..Spit-Out, The Journal of the American Ceramic Society. (Volume 19, Issue 1, January 1936), pages 287-294. Kurutz, Gary F. Architectural Terra Cotta by Gladding McBean. California, Windgate Press, 1989.160- FLANDROKingery, W.D. and H.K. Bowen and D.R. Uhlmann. Introduction to Ceramics 2nd Edition. New York, John Wiley & Sons, 1976.Kinzie, C.J.. Apparatus for Testing Enamels and Glazes for Resistance to Crazing Under Thermal Shock, The Journal of the American Ceramic Society. (Volume 15, Issue 2, February 1932), pages 112-122. Kinzie C.J. and C.H. Commons Jr.. A Method for the Determination of the Coefficient of Expansion, The Journal of the American Ceramic Society. (Volume 16, Issue 1, January 1933), pages 1-5.Laurence, F.S. Color in Architecture. New York: National Terra Cotta Society, 1922. Laurence, F.S. Terra Cotta in Architectural Design, The Journal of the American Ceramic Society. (Volume 8, Issue 2, February 1925), pages 79-83. Lefever, Leon. Architectural Pottery. Translation by K.H. Bird and Binns W. Moore. European Commission, France: Scott Greenwood and Company, 1900. Levin, Elaine. The History of American Ceramics. New York: Harry Abrams Inc., 1988. Lockington, W.P.. Architectural Decorations in Terra Cotta, The Clay-Worker. (Volume 39/40, 1903), page 34.Lockington, W.P.. Terra Cotta in Philadelphia Architecture, The Clay-Worker. (Volume 27/28, 1897), page 191. Lundell, G.E.F. and H.B. Knowles. The Analysis of Soda-Lime Glass, The Journal of the American Ceramic Society. (Volume 10, Issue 11, November 1927), Pages 829849. Matero, Frank and Andrew Gustine. Bibliography of Architectural Terra Cotta. University of Pennsylvania, 2001. Mattyasovsky-Zsolnay, L.. Investigations on Crazing, I-III: Delayed Thermal Contraction and Crazing of Ceramic Glazes, The Journal of the American Ceramic Society. (Volume 29, Issue 7, July 1946), pages 200-203. McDowell, Samuel. Terra Cotta Slip Coatings, The Journal of the American Ceramic Society. (Volume 16, Issue 12, December 1933), pages 134-140. McGohan, G.W.. Microscopic Technique in the Investigation of Enamel Coating and Some Observations, Journal of the American Ceramics Society. Volume 15 (1932), pages 389-396. McIntyre. Investigations Into the Durability of Architectural Terra-Cotta and Faience. London: Harrison and Sons Limited, 1929. Merritt, G.E. and C.G. Peters. Interferometer Measurements of Thermal Dilation of Glazed Ware, The Journal of the American Ceramic Society. (Volume 9, Issue 6, June 1926), pages 327-342. Middlesex County Culture and Heritage Commission. Uncommon Clay. New Jersey, The Commission, 2004. Moynehan, C.R., G.C. Allen, I.T. Brown, S.R. Church, J. Beavis AND J. Ashurst. Surface Analysis of Architectural Terracotta Journal of Architectural Conservation (Volume 1, Number 1, March 1995), pages 56-69. Mueller, Herman C.. The Independence of Burned Clay as A Decorative Material, Transactions of the American Ceramic Society, (Volume I, 1899), pages 15-21.FL A N 20 D R 09 O161- FLANDROMunsell Soil Color Charts 1975 Edition. Macbeth Division of Kollmorgen Corp., Baltimore, MD, 1975. Munshaw, L.M.. Pulsichrometer vs. Old Method of Applying Glazes, The Journal of the American Ceramic Society. (Volume 5, Issue 11, November 1922), pages 827-831. National Reporter System. The Pacific Reporter Volume 169 Permanent Edition. Neugebauer vs. Gladding, McBean & Co. West Publishing Company, 1918. National Terra Cotta Society. Architectural Terra Cotta: Standard Construction. Architects Edition. New York, NTCS, 1914.National Terra Cotta Society. Color in Architecture. New York, National Terra Cotta Society, 1922. Reprinted by permission of the American Architect and Architectural Review. National Terra Cotta Society. Standard Specification for the Manufacturing, Furnishing and Setting of Terra Cotta First Edition. New York, National Terra Cotta Society, 1923. National Terra Cotta Society. Standard Specification for the Manufacturing, Furnishing and Setting of Terra Cotta Revised Edition. New York, National Terra Cotta Society, 1927. National Terra Cotta Society. 12th Report Relating to the Technical Work of the National Terra Cotta Society. NY, NY, National Terra Cotta Society (June 10, 1926).FL A N 20 D R 09 O162- FLANDRONational Terra Cotta Society. 13th Report Relating to the Technical Work of the National Terra Cotta Society. NY, NY, National Terra Cotta Society (November 5, 1926). National Terra Cotta Society and H.G. Schurecht. 14th Report Relating to the Technical Work of the National Terra Cotta Society. NY, NY, National Terra Cotta Society (June 2, 1927). National Terra Cotta Society & H.G Schurecht. 15th Report Relating to the Technical Work of the National Terra Cotta Society. NY, NY, National Terra Cotta Society (November 15, 1927).National Terra Cotta Society and H.G. Schurecht. 16th Report Relating to the Technical Work of the National Terra Cotta Society. NY, NY, National Terra Cotta Society (May 10, 1928). National Terra Cotta Society and H.G. Schurecht. 17th Report Relating to the Technical Work of the National Terra Cotta Society. NY, NY, National Terra Cotta Society (December 1928). National Terra Cotta Society. Summary and Index of Reports 1-8 Relating to the Technical Work of National Terra Cotta Society. NY, NY, National Terra Cotta Society (April 1, 1927). The New York Times. Indict 49 in 1000% Terra Cotta Rise, September 29, 1921. Nicholson, T.E.. Discoloration of a Terra Cotta Glaze, The Journal of the American Ceramic Society. (Volume 15, 1932), pages 386-388. Nicholson, T.E. and Fred L. Goin. Simplified Thermal Expansion Rings for Production Control in Terra Cotta Manufacture, The Journal of the American Ceramic Society. (Volume 26, Issue 5, May 1943), pages 163-167.OBrien, Philip. Colorimetry of Glazes, The Journal of the American Ceramic Society. (Volume 35, Issue 10, April 1953), pages 254-259.Ortman, F.B.. The Composistion of Kiln Gases and Their Effect on Terra Cotta Glazes and Colors, The Journal of the American Ceramic Society. (Volume 3, Issue 6, June 1920), pages 476-488. Ortman, F.B.. Terra Cotta Cracking, The Journal of the American Ceramic Society. (Volume 16, Issue 12, December 1933), pages 639-642. Orton, Edward Jr. E.M.. Clay GlazesNo.1, The Clay-Worker. (Volume 27/28, 1897), pages 14-15.FL A N 20 D R 09 O163- FLANDROParmelee, Cullen W.. Note on Terra Cotta Glazes, Transactions of the American Ceramic Society. (Volume IX, 1907), pages 577-599. Parmelee, Cullen W. and Ralph Heidingsfeld. Note on Terra Cotta Glazes, Transactions of the American Ceramic Society. (March 1912), pages 840-848.Parmelee, Cullen W. and Ralph Heidingsfeld. The Production of Black Spots upon Terra Cotta Glazes, Transactions of the American Ceramic Society. (Volume XIV, March 1912), pages 229-237. Parmelee, C.W. and Paul E. Buckles. Study of Glaze and Body Interface, The Journal Of the American Ceramic Society. (Volume 25, April 1941), pages 11-15.Pence, Forrest K.. A Theory for the Cause of Matteness in Glazes, Transactions of the American Ceramic Society. (March 1912), pages 682-690. Petersen, Susan and Jan. The Craft and Art of Clay. New York, Woodstock Press, 2003. Planje, C.W.. Observations on Crazing in Terra Cotta Glazes, The Journal of the American Ceramic Society. (Volume 14, Issue 10, October 1931), pages 747-750. Powell, William H.. Address on Terra Cotta, The Journal of the American Ceramic Society. (Volume 7, Issue 7, July 1924), pages 255-257.Prudon, Theodore H.M.. Architectural Terra Cotta: Analyzing the Deterioration Problem, Technology and Conservation. Fall 1978, pages 30-40. Prudon, Theodore Henricus Maria. Architectural Terra Cotta and Ceramic Veneer In the United States Prior to World War II. New York, Columbia University, 1981. Pulsch, Herman A., Polychrome Glaze Decoration in Architecture, Transactions of the American Ceramic Society. (March 1907), pages 473-479. Purdy, Ross C.. Matte Glazes, Transactions of the American Ceramic Society. (March 1912), pages 671-681. Putnam, Edward. Architectural Terra Cotta, Its Physical and Structural Properties, Brickbuilder. (Volume 20- Plates, 1911), pages 29-33. Putnam, Edward. Polychrome Terra Cotta, Architecture. Volume 45 (1922), pages 22-24. Rabinowitz, Mark. Repairing Terra-Cotta Surfaces, Traditional Building Magazine. January 21, 2009. Real Estate Record and Builders Guide. Late Examples of Polychrome Terra Cotta, (April 11, 1908), page 653. Real Estate Record and Builders Guide. Development of Polychromatic Exterior Glaze Decoration, (August 1, 1908), page 237. Rhodes, Daniel, revised and expanded by Robin Hopper. Clay and Glazes for the Potter 4th Edition. Wisconsin, Krause Publications, 2000.FL A N 20 D R 09 O164- FLANDRORichardson, Willard D.. The Equipment of A Modern Paving Brick Plant, Transactions of the American Ceramic Society, (Volume IV, 1902), pages 189-199.Riddle, F.H.. The Control of Glaze-Fit by Means of Tensile Test Specimens, The Journal of the American Ceramic Society. (Volume 5, Issue 8, August 1922), pages 500503. Russell, Ralston Jr. and Robert R. Rowlands. Glaze Investigations: I, Effect of Various Silicas in Typical Glazes, The Journal of the American Ceramic Society. (Volume 36, Issue 1, January 1953), pages 1-8. Russell, Ralston Jr. and Robert R. Rowlands. Glaze Investigations: II, Effect of Composistion Variation in Glazes Containing Diatomaceous Silica The Journal of the American Ceramic Society. (Volume 36, Issue 1, January 1953), pages 9-11. Schepers, Francis. The Pulsichrometer, The Journal of the American Ceramic Society. (Volume 7, Issue 2, February 1924), pages 53-55. Schram, Edward. A Sandblast Abrasion Test For Glazes, The Journal of the American Ceramic Society. (Volume 12, Issue 5, May 1929), pages 356-359. Schurecht, H.G.. Blistering and Crawling of Glazes Caused by Underslips, The Journal of the American Ceramic Society. (Volume 17, December 1934), pages 182-184. Schurecht H.G.. Clay Sewer Pipe Manufacture, the Influence of Different Salt Glazing Temperatures Upon the Color and Gloss of Glazes Obtained on Clays with Varying Silica, Alumina, and Iron Oxide Contents, The Journal of the American Ceramic Society. (Volume 7, Issue 6, June 1924), pages 411-422. Schurecht, H.G. and G.R. Pole. Effect of Water in Expanding Ceramic Bodies of Different Compositions. The Journal of the American Ceramics Society. (Volume 12, 1929), pages 596-604. Schurecht, H.G.. Fitting Glazes to Ceramic Bodies, The Journal of the American Ceramic Society. (Volume 26, Issue 3, March 1943), pages 93-98.Schurecht, H.G. and G.R. Pole, Method of Measuring Strains Between Glazes and Ceramic Bodies, The Journal of the American Ceramic Society. (Volume 13, Issue 6, June 1930), pages 369-375. Schurecht, H.G.. Methods for Testing Crazing of Glazes Caused by Increase in Size of Ceramic Bodies, The Journal of the American Ceramics Society. (Volume 11, Issue 5, May 1928), pages 271- 277. Schurecht, H.G. and G.R. Pole. Moisture Expansion of Glazes and Other Ceramic Finishes, The Journal of the American Ceramic Society. (Volume 14, Issue 4, April 1931), pages 313-318.FL A N 20 D R 09 OStratton, Dr. Michael. Understanding and Conserving Terracotta http://www.ihbc.org.uk, 1994.165- FLANDROSchurecht, H.G. and D. H. Fuller. Some Effects of Thermal Shock in Causing Crazing of Glazed Ceramic Ware, The Journal of the American Ceramic Society. (Volume 14, Issue 8, August 1931), pages 565-571. Scott, W.J.. An Apparatus for Measuring the Abrasive Hardness of Glazes, The Journal of the American Ceramic Society. (Volume 7, Issue 5, May 1924), pages 324-346.Searls, Carolyn and Cece Louie. The Good, the Bad, and the Ugly: Twenty Years Of Terra-Cotta Repairs Reexamined APT Bulletin (Volume 32, Number 4, 2001), pages 29-36. Shockley, Jay and Susan Tunick. The Copper Union Building and Architectural Terra Cotta. Winterthur Portfolio Volume 39 Number 4 (Winter 2004). Smithsonian Institution. Smithsonian Folklife Festival: Master of the Building Arts.www. siedu/festival/buildingarts.htm, 2001. Solon, Leon. The Park Avenue Building in New York City, The Architectural Record. (Volume 63, Number 4, April 1928), pages 289-297. Spurrier, Harry. Notes on Terra Cotta Glaze Consistency, The Journal of the American Ceramic Society. (Volume 12, Issue 9, September 1929), pages 577-580. Spurrier, Harry. Some Fundamentals of Terra Cotta, The Journal of the American Ceramic Society. (Volume 9, Issue 11, November 1926), pages 773-778. Staley, Homer F.. The Microscopic Examination of Twelve Matte Glazes, Transactions of the American Ceramic Society. (March 1912), pages 691-708. Stephanie, William J. The Construction and Operation of a Terra Cotta Plant. Transactions of the American Ceramic Society, (March 1907), pages 506-537.Stephanie, William J.. Note on the Muffle Terra Cotta Kiln, Transactions of the American Ceramic Society. (Volume IX, 1907), pages 681-693.Strusholm, A.M.. Overglaze Polychrome Cone No. 6, The Journal of the American Ceramic Society. (Volume 14, Issue 10, October 1931), pages 751-754. Stull, R.T.. Notes on the Manufacture of Enamel Brick with Some Investigations on Enamel Brick Slips, Transactions of the American Ceramic Society. (Volume 7, February 1910), page 711-764. Teft, Dwight. Glost Warping of White Wall Tile, The Journal of the American Ceramic Society. (Volume 12, Issue 12, December 1929), pages 767-770. Teutonico, Jeanne Marie, Editor. Architectural Ceramics: Their History, Manufacuture and Conservation. London, James & James Ltd, 1996. Teutonico, Jeanne Marie and John Fidler. Time for Change: An Overview ofFL A N 20 D R 09 O166- FLANDROBuilding-Materials Research for Conservation of Historic Structures. APT Bulletin, Volume 29, no. (1998), pages 45-49.Thiemecke, H.. Notes on Moisture Expansion, The Journal of the American Ceramic Society. (Volume 27, Issue 12, December 1, 1944), pages 355-358. Thomas, Everett, Milton A. Tuttle and Esther Miller. Study of Glaze Penetration and Its Effect on Glaze Fit: I-III, The Journal of the American Ceramic Society. (Volume 28, Issue 2, February 1945), pages 52-62.Tiller, de Teel Patterson. The Preservation of Historic Glazed Architectural Terra-Cotta, Preservation Briefs Number 7. Washington DC, National Park Service. 1979. Tindall, Susan. How to Prepare Project Specific Terra-Cotta Specifications, Association for Preservation Technology (APT) Bulletin. (Volume 21, Number 1, 1989), pages 26-36. Tindall, Susan. Terra Cotta Replacement, Association for Preservation Technology (APT) Bulletin. (Volume 20, Number 3, 1988), pages 12-14. Tunick, Susan. Terra-Cotta Skyline. New Jersey, Princeton Architectural Press, 1997. Twells, Robert Jr.. Preparing and Spraying a Glaze Slip with Special Reference to the Control of the Various Operations, The Journal of the American Ceramic Society. (Volume 7, Issue 6, June 1924), pages 465-474.United States 81st Congress. Amendment Section 2 of the Act of March 3, 1901. (2nd Session, CHS. 485, 486 July 21, 22, 1950), pages 371-373. United States Department of Commerce. Architectural Terra-Cotta Investigations, Technical News Bulletin of the Bureau of Standards No. 98. Washington, D.C., June 1925. United States Department of Commerce. Method for Measuring Thermal Dilation of Glazed Wares, Technical News Bulletin of the Bureau of Standards No. 108.Washington, D.C. April 1926, page 5. United States Department of Commerce. Meeting of Advisory Committee on Ceramics, Technical News Bulletin of the Bureau of Standards No. 120. Washington, D.C. April 1927, pages1- 5. United States Department of Commerce. Fundamental Study of Glaze Fit, Technical News Bulletin of the Bureau of Standards No. 125. Washington, D.C. September 27, 1927, pages 7. United States Department of Commerce. Physical Properties of Semi-Vitreous Bodies, Technical News Bulletin of the Bureau of Standards No. 126. Washington, D.C. October 1927, page 8. United States Department of Commerce. Fundamental Study of Glaze Fit, Technical News Bulletin of the Bureau of Standards No. 128. Washington, D.C. December 1927, page 3. United States Department of Commerce. Artificial Vitrifying Agents for Ceramic Bodies, Technical News Bulletin of the Bureau of Standards No. 131. Washington, D.C. March 1928, page 33. United States Department of Commerce. Vitrification of Ceramic Bodies, Technical News Bulletin of the Bureau of Standards No. 133. Washington, D.C. May 1928, page 71. United States Department of Commerce. Tests of Terra Cotta Faced Walls, Technical News Bulletin of the Bureau of Standards No. 140. Washington, D.C. December 1928, page 171. United States Department of Commerce. Fundamental Study of Glaze Fit, Technical News Bulletin of the Bureau of Standards No. 142. Washington, D.C. February 1929, page 19. United States Department of Commerce. National Bureau of Standards Technical note 649. Administrative Manual Legislative Authority. Boulder, CO, Department of Commerce, (1974), pages 21- 23. United States General Services Administration. Terra Cotta: Characteristics, Uses and Problems, Historic Preservation Technical Procedures. www.gsa.gov, 2009. United States National Bureau of Standards. National Bureau of Standards Special Publication, Annual Report of the Director of the Bureau of Standards to the Secretary of Commerce for the Fiscal Year Ended June 30, 1919 Miscellaneous publication, Washington, DC. (1919), pages 266-271. Varney, William P.. Enameled Brick, The American Architect. (February 28, 1912), pages 106-109.FL A N 20 D R 09 O167- FLANDROViera, Ricardo. Cold Glazing Terra Cotta: The Repair of Damaged Glazed Surfaces. Columbia University, NY, 1992. Viera, Ricardo. Testing and Analysis of Terra Cotta Glaze Repairs, Standards for Preservation and Rehabilitation, ASTM STP 1258. S.J. Kelly, Ed. American Society for Testing and Materials, (1996), pages 319-336. Warren, B.E. and A.D. Loring. X-Ray Diffraction Study of the Structure of Soda-Silica Glasss, The Journal of the American Ceramic Society. (Volume 18, 1935), pages 269-276. Watts, Arthur P. The Effect of Furnace Atmosphere on the Quality of Certain Types of Glazes, The Journal of the American Ceramic Society. (Volume 14, Issue 6, June 1931), pages 460-468. Weaver, Martin E with Frank Matero. Conserving Buildings: Guides to Techniques and Materials Revised Edition. New Jersey, John Wiley and Sons, 1997. Wilson, Hewitt. Monograph and Bibliography on Terra Cotta, The Journal of the American Ceramic Society. (Volume 5, Issue 2, February 1926), pages 94-145.FL A N 20 D R 09 O168- FLANDROWilson, Hewitt. Note on Pinholing and Peeling on Terra Cotta, Transactions of the American Ceramic Society. (Volume 19, 1917), pages 209-215. Wilson, Hewitt. Polychrome Decoration of Terra Cotta With Soluble Metallic Salts, The Journal of the American Ceramic Society. (Volume 1, Issue 5, May 1918), pages 353-366.169- FLANDROFL A N 20 D R 09 OFL A N 20 D R 09 OLIST OF FIGURES170- FLANDROList of FiguresINTRODUCTION Figure 1: Fred C. Weber Floral Shop, St. Louis, Missouri Figure 2: Polychrome terracotta on the Chambers- Broadway building. Figure 3: Glaze applied with a pulsichrometer Figure 4: Glaze applied by splashing.FL A N 20 D R 09 OFigure 5: The Clark apparatus. Figure 6: Trial piece Figure 7: Test Specimen Figure 8: Test Specimens Figure 10: Sample types used Figure 12: Autoclave used to test specimens for crazing Figure 14: Ring test specimens Figure 15: Image of wall load test. Figure 19: Glaze Stress Furnace Figure 20: Device used to cause parallel crazing Figure 21: Converted meat grinder to prepare expansion rings for testing Figure 24- Specimen oriented with glaze pointing upwards.171- FLANDROSECTION I: PREVIOUS RESEARCH AND KEY HISTORICAL MOMENTSFigure 9:Apparatus used to determine Osmosis as a mode of terracotta deterioration Figure 11: Set employed to test for size changes in terra cotta when in contact with water.Figure 13: Sketches of different types of body separation from grog and types of fractures found in strength tests employing terra cotta.Figure 16: Apparatus/method used to determine the tensile strength of glazes Figure 17: Apparatus used to determine Youngs modulus of elasticity of glazes Figure 18: Drawings of rings and apparatus used to determine contraction and expansion of the rings.Figure 22- Intermittent spray of de-ionized water directly onto the core drill bit. Figure 23- Excavation of terracotta adjacent to core, to allow core to be freed without a lot of force.Figure 25- Exterior surface of the terracotta block on the left-hand side. Figure 26- Small container with terracotta core in place. Figure 27- Injection of de-ionized water into container. Figure 28- Specimens on filter paper to ensure water tightness of the epoxy seal. Figure 29- Prepared samples elevated by fiberglass mesh, ready to be inserted into the dessicator. Figure 30: Terra-cotta Block #1 Figure 31: Building from which Block #1 originated. Figure 32- Block #1, 25x magnification. Figure 33- Block #1, 50x magnification. Figure 34- Block #1, 100x magnification. Figure 35- Block #1, 175x magnification. Figure 36- Block #1, PPL, 50x magnification. Figure 37- Block #1, PPL, 100x magnification. Figure 38- Block #1, XPL, 50x magnification. Figure 39- Block #1, XPL, 100x magnification. Figure 40- Block #1,25x magnification. Figure 41- Block #1,50x magnification. Figure 42- Block #1,100x magnification. Figure 43- Sample 1.1 Figure 44- Sample 1.2 Figure 45- Sample 1.3 Figure 46- Sample 1.4 Figure 47- Sample 1.5 Figure 48- Sample 1.6 Figure 49- Sample 1.7 Figure 50- Sample 1.8 Figure 51: Terra-cotta Block #2 Figure 52: Building from which Block #2 originated. Figure 53- Crazing on Block #2, 25x magnification. Figure 54- Crazing on Block #2, 50x magnification. Figure 55- Crazing on Block #2, 100x magnification. Figure 56- Crazing and crawling on Block #2, 25x magnification. Figure 57- Crawling on Block #2, 100x magnification. Figure 58- Crawling on Block #2, 175x magnification. Figure 59- Block #2, crawling glaze, PPL, 50x magnification.FL A N 20 D R 09 O172- FLANDROFigure 60- Block #2, crawling glaze, PPL, 100x magnification Figure 61- Block #2, crawling glaze, XPL, 50x magnification Figure 62- Block #2, crawling glaze, XPL, 100x magnification. Figure 63- Block #2, crazing glaze, PPL, 50x magnification Figure 64 Block #2, crazing glaze, PPL, 100x magnification. Figure 65- Block #2, crazing glaze, XPL, 50x magnification. Figure 66- Block #2, crazing glaze, XPL, 100x magnification Figure 67- Block #2, crawling glaze 25x magnification. Figure 68- Block #2, crawling glaze, 50x magnification. Figure 69- Block #2, crawling glaze,100x magnification. Figure 70- Block #2, crazed glaze25x magnification. Figure 71- Block #2, crazed glaze, 50x magnification. Figure 72- Block #2, crazed glaze,100x magnification. Figure 73- Sample 2.1 Figure 74- Sample 2.2 Figure 75- Sample 2.3 Figure 76- Sample 2.4 Figure 77- Sample 2.5 Figure 78- Sample 2.6 Figure 79- Sample 2.7 Figure 80- Sample 2.8 Figure 81- Terra-cotta Block #3 Figure 82- Building from which Block #3 originated. Figure 83- Building from which Block #3 originated. Figure 84- Crawling on Block #3, 25x magnification. Figure 85- Block #3, 50x magnification. Figure 86- Block #3, 100x magnification. Figure 87- Block #3, PPL, 50x magnification. Figure 88- Block #3, PPL, 100x magnification. Figure 89- Block #3, XPL, 50x magnification. Figure 90- Block #3, XPL, 100x magnification. Figure 91- Block #3, 25x magnification. Figure 92- Block #3, 50x magnification. Figure 93- Block #3, 100x magnification. Figure 94- Sample 3.1 Figure 95- Sample 3.2FL A N 20 D R 09 O173- FLANDROFigure 96- Sample 3.3 Figure 97- Sample 3.4 Figure 98- Sample 3.5 Figure 99- Sample 3.6 Figure 100- Sample 3.7 Figure 101- Sample 3.8 Figure 102- Terra-cotta Block #4 Figure 103- Building from which Block #4 originated. Figure 104- Block #4, 25x magnification. Figure 105- Block #4, 50x magnification. Figure 106- Block #4, 100x magnification. Figure 107- Block #4, 175x magnification. Figure 108- Block #4, PPL, 50x magnification. Figure 109- Block #4, PPL, 100x magnification. Figure 110- Block #4, XPL, 50x magnification. Figure 111- Block #4, XPL, 100x magnification. Figure 112- Block #4, 25x magnification. Figure 113- Block #4, 50x magnification. Figure 114- Block #4, 100x magnification. Figure 115- Terra-cotta Blocks #5 and #6. Figure 116- Building from which Blocks #5 and #6 originated. Figure 117- Block #6, 25x magnification. Figure 118- Block #6, 50x magnification. Figure 119- Block #6, 100x magnification. Figure 120- Block #6, 175x magnification. Figure 121- Block #6, PPL, 50x magnification. Figure 122- Block #6, PPL, 100x magnification. Figure 123- Block #6, XPL, 50x magnification. Figure 124- Block #6, XPL, 100x magnification. Figure 125- Block #6, 25x magnification. Figure 126- Block #6, 50x magnification. Figure 127- Block #6, 100x magnification. Figure 128- Block #5, PPL, 50x magnification. Figure 129- Block #5, PPL, 100x magnification. Figure 130- Block #5, XPL, 50x magnification. Figure 131- Block #5, XPL, 100x magnification.FL A N 20 D R 09 O174- FLANDROFigure 132- Block #5,25x magnification. Figure 133- Block #5, 50x magnification. Figure 134- Block #5,100x magnification.FL A N 20 D R 09 O175- FLANDROCeramic Nomenclature While the American Society for Testing and Materials (ASTM International) has standardized some terminology relating to ceramic ware, it is wildly insufficient in relation to architectural uses. The following glossary has been compiled from a variety of sources, a bibliography of the sources used are listed at the end of this glossary. - inversion- The change that occurs in quartz at 573 C at this temperature quartz changes during heating from a low temperature alpha form to the high temperature beta form, and reverts back during cooling. The coefficient of thermal expansion of quartz is high in comparison with that of most other minerals but is particularly pronounced at this temperature, jumping from 3.76% volume expansion at 570 C to 4.55% at 580 C and of course, undergoing a corresponding decrease in size during cooling. Changes in the size of quartz inclusions during the cooling stage after firing are the main cause of dunting. (Gibson and Woods, p. 83).FL A N 20 D R 09 Ometal. high dry strength, long Vitrification range, and a light color when fired. (ASTM C242-01, p. 77). body, or in a glaze or other coating. (ASTM C242-01, p. 78).176- FLANDROAdsorption- The capacity of a substance to accept and retain on its surface a layer of another substance, usually a gas or a liquid. (ASTM C242-01, p. 77). Anchor- The mechanical device by which a piece of architectural terracotta is held in place. Usually Architectural Terracotta- Terracotta used for any architectural purposes. The term is mainly used inreference to outdoor installations. Terracotta refers to a high grade of weathered or aged clay which, when mixed with sand or with pulverized fired clay, can be molded and fired at high temperatures to a hardness and compactness not obtainable with brick. Terracotta was usually hollow cast in blocks which were open to the back, like boxes, with internal compartment-like stiffeners called webbing. Webbing substantially strengthened the load-bearing capacity of the hollow terra-cotta block without greatly increasing its weight. (Tiller, Preservation Brief 7).Ball Clay- A secondary clay, commonly characterized by the presence of organic matter, high plasticity,Black coring- The dark zone sometimes found in the middle of sherds. It is caused by localized reduction during firing and really only occurs in kiln-fired vessels. Care must be taken to distinguish between real black coring, which is caused by the reduction of iron oxides within the pot walls, and the black zone that can be found in many open-fired pots and which is the result of incomplete oxidation of the carbonaceous matter present in the clay; the later is an indicator of short firing (as there has been insufficient time to burn out this material) and therefore, frequently, of open firing. (Gibson & Woods, p. 109). Blistering- The development during firing of enclosed or broken macroscopic vesicles or bubbles in aBloating- Substantial swelling produced by a heat treatment that causes the formation of a vesicular structure. (ASTM C242-01, p. 78). Body- The structural portion of a ceramic article, or the material or mixture from which it is made. (ASTM C242-01, p. 78). Brownstone Terra Cotta- Used primarily in the mid to late 1800s, and was the earliest type of terracotta used on American buildings. It was hollow cast, unglazed or slip-glazed, and was typically dark red to brown in color. (ASTM STP- 1444, p. 78).Calcareous inclusions- Inclusions composed of calcium carbonate limestone or shell. Such materialsfrequently occur naturally in sedimentary clays but were also often added by potters. Problems are incurred in the firing of these inclusions at temperatures between 650C and 900C because of lime blowing. (Gibson and Woods, p. 116).FL A N 20 D R 09 O177- FLANDROCeramic Process- The production of articles or coatings from essential inorganic, nonmetallic materials, the articles or coating being made permanent for utilitarian and decorative purposes by the action by heat at temperatures sufficient to cause sintering, solid-state reactions, bonding, or conversion partially or wholly to a glassy state. (ASTM C242-01, p. 78).Ceramics- A general term applied to the art or technique of producing articles by a ceramic process, or to the articles so produced. (ASTM C242-01, p. 78). Ceramic change- The point at which clay becomes ceramic. It is brought about by the removal of thehydroxyl groups from the chemically combined water in the clay molecules. The point at which this occurs varies according to the type of clay mineral involved but is generally considered to take place at 550C600C; afterwards, the clay is fired and will not regain plasticity when in contact with water. (Gibson and Woods, p. 120).Ceramic Veneer- Consist of glazed ceramic tile attached to a backup substrate of clay tile or a grid of metal ties. Ceramic veneer units were not hollow cast, but were sold cast and ribbed in the back. The ribbing was cast in order to back-parge the units onto the substrate, it was a system that was developed during the 1930s. (Derived from Tiller, Preservation Brief 7).Machine-Extruded Ceramic Veneer- A relatively thin, solid slab of burned clay with a ceramic vitreous or glazed surface. The back must be either scored or ribbed. (AIA File No. 9., p. 2). Adhesion Type Ceramic Veneer- Ceramic veneer which is attached to the backing by the bond between mortar and masonry without the use of metal ties. Its thickness, including ribs, is not more than inch. (AIA File No. 9, p. 2). Anchored Type Ceramic Veneer- Ceramic veneer which is attached to the backing by nonferrous metal anchors. The minimum net thickness, exclusive of ribs, is 1 inch. (AIA File No. 9, p. 2).Mold-Pressed Ceramic Veneer- Ornamental panels and special shapes made from molds into which plastic clay is pressed. The minimum thickness of the exposed face is 1 inch. (AIA File No. 9, p. 2). Sculpture- Ornamental forms sculpted in plastic clay and burned; also may be mold-pressed. (AIA File No. 9, p. 2) Glazed Ceramic Veneer- Units whose exposed surface faces are covered by an inseparable ceramic finish, vitrified to a glassy state, of matte, stain or gloss finish. (AIA File No. 9, p. 2). Vitrified Ceramic Veneer- Units whose exposed surfaces are covered by an inseparable ceramic finish, rendered impervious by firing. (AIA File No. 9, p. 2).Clay- A natural mineral agglomerate, consisting essentially of hydrous alumina silicate; plastic when sufficiently wetted, rigid when dried en masse, and vitrified when sufficiently fired to a high temperature. (AIA File No. 9, p. 2). Presented by the general chemical formula Al2O32SiO22H2O. Crawling- A parting and contraction of the glaze on the surface of ceramic ware during drying or firing, resulting in unglazed areas bordered by coalesced glaze. (ASTM C242-01, p. 79).Crazing- The cracking that occurs in fired glazes or other ceramic coatings as a result of tensile stresses.May also occur in the surface portion of unglazed whiteware bodies. (ASTM C242-01, p. 79).Crow footing- Thin hairline cracking which occurs particularly in unglazed ware. It is different fromcrazing in that the body is more exposed through the cracks. (National Terra Cotta Society, 13th Report, p. 5).FL A N 20 D R 09 Oevaporates. Woods). face concentration of fine colloidal particles in the clay. (Fidler, p. 30). ture, to develop the desired properties. Bisque Fire- The process of kiln-firing ceramic ware before glazing. Firing Cycle- The time required for one complete firing operation (cold to cold). Glost Fire- The process of kiln-firing bisque ware to which glaze has been applied. Single Fire- The process of maturing an unfired ceramic body and its glaze in one firing operation. (ASTM C242-01, p. 80).Cristobalite- One of the three basic structures of silica (see quartz & tridymite) it is formed between 1470C- 1710C (Kingery). Drying shrinkage- The shrinking that occurs during the drying of a clay body as the water of plasticityDunting- Cracking that occurs in fired ceramic bodies as a result of thermally induced stresses. (ASTM C242-01, p. 80). It is thought to be caused by the change in volume of free silica as it cools through the - inversion, and therefore vessels that contain abundant quartz are most at risk. (Gibson & Woods, p.148). Engobe- A term of wide meaning often interchangeable with slip but including other materials. An engobe is used to cover a clay, produce a bugger layer and give a different surface, texture and colour. It is applied by brush, dip, spray etc. Clay slips and glaze slops are essentially fluid in forms. An engobe can be in a jelly or stiff form. It is often half-way between a clay and a glaze in composition and contains materials which are normally considered glaze materials. It therefore fires to a more vitreous state than the body which it covers. However, since it does not fuse to a glassy state, it cannot be called a glaze. And since it can be composed entirely of non-clay materials, it cannot be called a slip. The term engobe is often the only one that is suitable. (Hamer, p. 126).Filler- Inclusions deliberately added to the clay by the potter (see also temper) (Derived from Gibson & Fire Skin- A thin and extremely hard integral vitreous coating formed upon firing and induced by a sur-Fire Spalls- The round flakes of clay that are blown out of the walls of clay vessels when the temperature is raised too quickly during the early stages of firing. Corresponding scars are left on the surfaces of vessels. (Gibson & Woods, p. 156). Firing- The controlled heat treatment of ceramic ware in a kiln of furnace, during the process of manufac-Flux- A substance that promotes fusion in a given ceramic mixture. (ASTM C242-01, p. 81). Glaze- A ceramic coating matured to the glassy state on a formed ceramic article, or the material or mixture from which the coating is made.178- FLANDROBright glaze- A colorless or colored ceramic glaze having high gloss. Clear glaze- A colorless or colored transparent ceramic glaze. Crystalline glaze- A glaze containing macroscopic crystals. Fritted glaze- A glaze in which a part or all of the fluxing materials are prefused. Matt glaze- A colorless or colored ceramic glaze having low gloss. Opaque glaze- A nontransparent colored or colorless glaze.FL A N 20 D R 09 OSemi-mat glaze- a colorless or colored glaze having moderate gloss. Slip glaze- a glaze consisting primarily or readily fusible clay or silt. face tension forces have been active after fusion. (Wilson, p.117).Raw glaze- A glaze compound primarily from raw constituents, that is, containing no prefused materials. Satin glaze- a glaze which exhibits a non-zero specular reflection reduced by at least 50 percent.Semi-glaze- A term that was used prior to 1911 when it was eliminated because of confusion, it was used to specify matt glaze.1Vellum glaze- a semi-mat glaze having a satin-like appearance. (ASTM C242-01, p 81).Glaze Cracking- Crawling of the glaze from small cracks in the body or underslip, indicative that surGlaze Fit- The stress relationship between the glaze and body of a fired ceramic product. Glaze Peeling- Although synonymous in some writings with shivering, peeling is limited to the separation of the glaze, or underslip and glaze, from the body when the surface of the body is not broken. (Wilson, p. 118).Grog- Crushed, previously fired ceramic used as an opening material. Macroscopically, a difference incolor between grog and the surrounding matrix often serves to facilitate its identification. In thin section, grog is usually angular and is often surrounded by voids created when the wet and plastic clay shrinks away from the inert grog during the drying of the vessel. Grog may be of clay similar to or different from the matrix, and this can also be seen in thin section. Frequently, grog within grog, indicating recycling and continuity of ceramic tradition, can also be observed. (Gibson & Woods, p. 178).Illite- A three-layer clay mineral, often used for paints or slips. (Sinopoli, p. 228).Imperviousness- That degree of vitrification evidenced visually by complete resistance to dye penetration. The term impervious generally signifies zero absorption, except for floor and wall tile, which are considered impervious up to .5% water absorption. (ASTM C242-01, 82). Inclusions- The term used to describe all non-clay and/or plastic materials present in a clay body or firedfabric. They may be naturally occurring or added by the potter. (Gibson & Woods, p. 192).Leather-hard- Clay dried sufficiently to be stiff, but still damp enough to be joined to other pieces with slip (Rhodes & Hopper, p. 329) and altered.1Putnam, Edward H. The Brickbuilder, Plates, Volume 20, 1911.179- FLANDROLime blowing- Post-firing defect normally taking the form of small spalls pushed out of the walls of vessels containing calcareous inclusions. When fired between 650C and 890C, and in particular at temperatures in excess of 750C. (Gibson & Woods, p. 203). Machine-Extruded Ceramic Veneer- See Ceramic Veneer Maturing Range- The time-temperature range within which a ceramic body, glaze or other composition may be fired to yield specific properties. (ASTM C242-01, p. 82).FL A N 20 D R 09 O01, p. 83). 01, 84). (Kingery). sence of oxygen. (Gibson & Woods, p. 234).180- FLANDROMottled color/ mottling- The presence in the surface of a glaze or body of irregularly shaped, randomly distributed areas that vary in color, gloss or sheen causing the surface to be non-uniform in appearance. (ASTM C242-01, P. 82).Non-vitreous- (Non-vitrified) That degree of vitrification evidence by relatively high water absorption. The term non-vitreous generally signifies more than 10.0% water absorption, except for floor and wall tile, which are considered non-vitreous when water absorption exceeds 7.0%. (ASTM C242-01, p. 83).Over fire- Fired to or above the point at which defects such as warping, bloating and blistering occur. (Gibson & Wood, p. 214). Oxidation (atmosphere/environment)- An oxygen rich firing atmosphere. Under oxidizing conditions, iron oxides present in the clay will be brought to their highest state of oxidation and, depending on the amount present, will give a yellow or, more commonly a red or reddish-brown color to the fired clay. (Gibson & Wood, p. 216).Particle Orientation- The clay particles orientation is such that their long axes are parallel to the direction of the forces employed to create the object. This orientation can be detected in tangential thin sections and also through x-radiography. (Gibson & Wood, p. 217).Permeability- The measure through a material of fluid flow, gas or liquid. (ASTM C242-01, p. 83). Pinholes- Imperfections in the surface of a ceramic body or glaze resembling pin pricks. (ASTM C242Plastic- A descriptive term applied to a material that exhibits the property of plasticity or stickiness, whereplasticity is the ability of a material to undergo substantial deformation without fracturing. (ASTM C24201, p. 83).Polychrome- The application of two or more colors to a single piece. The term does not denote certain speckled and mottled finishes produced by the intermingling of two or more colors for a ground color to be used without other applied coloring. (Color in Architecture National Terra Cotta Society). Porosity- The volume fraction of voids contained in a solid, often expressed as a percent. (ASTM C242Quartz-One of the three forms of silica (see cristobalite and tridymite), it occurs below 573C to 867C Reduction (atmosphere/environment)- Firing atmosphere characterized by the shortage or abScumming- Also termed scum or green scum. Caused by water soluble materials depositing on the surface of clays by the evaporation of any water that has soaked into the ware. Its characteristic color is first palegreen yellow which deepens over a period of days to a dark green, when it become very difficult to remove. (Goodrich, p. 382).Semivitreous- That degree of vitrification evidence by a moderate or intermediate water absorption. The term semivitreous generally signifies 0.5 to 10.0% water absorption, except for floor and wall tile which are considered semivitreous when water absorption is between 3.0-7.0%. (ASTM C242-01, p. 84). Sinter- A ceramic material or mixture fired to less than complete fusion, resulting in a coherent mass, or the process involved. (ASTM C242-01. p. 85). Sintering- The stage in firing when the edges of the clay particles begin to soften and melt and stick together. It represents the initial stage of the vitrification process and results in a harder, denser and more rigid body. (Gibson & Woods, p. 248).FL A N 20 D R 09 Ois incorrect. Woods, p. 213).Shiver- Also termed peeling, the splintering that occurs in fired glazes or other ceramic coatings as a resultof critical compressive stresses. (ASTM C242-01, p. 85). A separation of glaze from the body when portions of the body are removed with the broken pieces of glaze. (Wilson, p. 118). It looks like a paint chip.Slip- A viscous mixture of clay suspended in water, often the term is used synonymously with engobe whichSlip glaze- A surface treatment involving the application of a suspension of clay in water to a vessel in the leather-hard stage. It is not absorbed to any great extent by the clay body and in thin section are easily distinguishable. (Gibson & Woods, p. 243).Spall- The breaking off of a piece of ceramic body (often taking the glaze with it) caused by thermal or mechanical stresses. Temper- One of the terms used to describe opening materials added to clay by the potter. (Gibson &Thermal Shock- A condition of stress brought about by a large temperature difference across a body or glaze. (ASTM C242-01, p. 85). Thermal Shock Failure- The mechanical failure of a glaze or body, as a result of the stress caused by a large temperature difference across the ware. (ASTM C242-01, p. 85). Thermal Shock Resistance Testing- The act of exposing ware to a rapid temperature change to determine the temperature difference a glaze or body can withstand with mechanical failures. (ASTM C24201, p. 85).Thin Glaze- Too thin a coating of glaze, often it is accompanied by a dry looking surface and occurs mainly on edges. (Wilson, p. 117).Tridymite- One of the three basic forms of silica (see quartz and cristobalite), it occurs between 867C to 1470C (Kingery). Underslip- A term used to signify a slip or engobe underneath a glaze. Vitreous- That degree of vitrification evidenced by low water absorption. The term vitreous generally signifies less than 0.5% absorption, except for floor and wall tile and low-voltage electrical porcelain which are considered vitreous up to 3.0% absorption. (ASTM C242-01, p. 86).181- FLANDROVitrification- The progressive reduction and elimination of porosity of a ceramic composition, with the formation of a glass phase, as a result of heat treatment. (ASTM C242-01, p. 86). Warpage- Curvature of a flat specimen measured as a deviation of the specimen surface from a trueplane along the edges or the diagonals and at the mid-length of an edge or diagonal, expressed as a percent of the length of the edge or diagonal, and called convex or concave with respect to the face of the specimen. (ASTM C242-01, p. 86).FL A N 20 D R 09 OWoods, p. 276). turns to steam (Gibson & Woods, p. 276). 15).182- FLANDROWater of Plasticity- The water that is mixed with the clay to enable it to become plastic. (Gibson &Water of hydration or combined water- The water in a material that cannot be removed by drying at 110C, as it is chemically bound, expressed as a percent of the weight of the material. (ASTM C242-01, p. 86 ). Water Smoking- The stage in firing around 100C when remnant water of plasticity in the clay boils and Webbing- Internal compartment-like stiffeners called. Webbing substantially strengthened the load-bearing capacity of the hollow terra-cotta block without greatly increasing its weight. (Tiller, Preservation Brief 7).Yield Point-The shearing force that will just cause flow of a plastic mass. (Derived from Shephard, p.American Institute of Architects. Public Works Specifications Ceramic Veneer A.I.A File No. 9. October 1961. Glossary Bibliography American Society for Testing and Materials. Standard Terminology of Ceramic Whitewares and Related Products, Standard C242-01. ASTM Standards. ASTM International, PA. (2005), pages 77-86. American Society for Testing and Materials. Standard Terminology Relating to Surface Imperfections on Ceramics, Standard F109-04. ASTM Standards. ASTM International, PA. (2005), pages 507-511.FL A N 20 D R 09 OLeicester University Press, London. 1997.183- FLANDROGibson, Alex and Ann Woods. Prehistoric Pottery for the Archaeologist (2nd Edition).Fidler, John. The Manufacture of Architectural Terracotta and Faience in the United Kingdom. Association for Preservation Technology (APT) Bulletin. Volume 15, No.2 (1983), page 30.Hoigard, Kurt R. and G.R. Mulholland and R.C. Haukohl. Terra Cotta Facades, Symposium on Building Faade Maintenance, Repair and Inspection, ASTM STP 1444. J.L. Erdlv and T.A. Schwartz, Editors, ASTM International, West Conshohocken, PA, (2004), pages 75-90. Kingery, W.D. and H.K. Bowen and D.R. Uhlmann. Introduction to Ceramics 2nd Edition. New York, John Wiley & Sons, 1976. Mayer, R. A Dictionary of Art Terms and Techniques. New York, Harper and Row, 1969.Goodrich, H.R.. Miscellaneous Terra Cotta Laboratory Notes, The Journal of the American Ceramic Society. (Volume 15, Issue 7, July 1932), pages 382-385. Hamer, Janet. The Potters Dictionary of Materials and Techniques. University of Pennsylvania Press, Philadelphia. 2004.National Terra Cotta Society. Color in Architecture. New York, National Terra Cotta Society, 1922. Reprinted by permission of the American Architect and Architectural Review. National Terra Cotta Society. 13th Report Relating to the Technical Work of the National Terra Cotta Society. NY, NY, National Terra Cotta Society (November 5, 1926).Rhodes, Daniel and Robin Harper. Clay and Glazes for the Potter. KP Craft Publishers, 2000.Shepard, Anna O. Ceramics for the Archaeologist. Carnegie Institution of Washington, Publication 609, Washington D.C.. 1956. Sinopoli, Carla. Approaches to Archaeological Ceramics. Plenum Press, NY. 1991. Tiller, de Teel Patterson. The Preservation of Historic Glazed Architectural TerraCotta, Preservation Briefs Number 7. Washington DC, National Park Service. 1979. Wilson, Hewitt. Monograph and Bibliography on Terra Cotta, The Journal of the American Ceramic Society. (Volume 5, Issue 2, February 1926), pages 94-145.FL A N 20 D R 09 O184- FLANDROFL A N 20 D R 09 OAppendix185- FLANDROFL A N 20 D R 09 O186- FLANDROWATER VAPOR TRANSMISSION TEST- RAW DATAFLN 20 D R 09 ODaily Water Vapor Transmission Data187- FLANDROAWater Vapor Transmission Rates (g/day)- normalized dataFLN 20 D R 09 O A188- FLANDROFLN 20 D R 09 OINVERTED CUP WATER VAPOR TRANSMISSION- RAW DATA189- FLANDROAFLN 20 D R 09 ODaily Inverted Cup Water Vapor Transmission Data190- FLANDROAInverted Cup Water Vapor Transmission Rates (g/day)- normalized dataFLN 20 D R 09 O191- FLANDROAFLN 20 D R 09 ORATE OF WATER VAPOR TRANSMISSIONS192- FLANDROARates of water vapor transmissions (both regular and inverted cup)Grams/Day/ Surface areaFLN 20 D R 09 O A193- FLANDROFL A N 20 D R 09 O194- FLANDROCOMPARATIVE ANALYSIS WATER VAPOR TRANSMISSION RATE GRAPHSAverage daily weight loss (g/day):Comparison between regular water vapor transmission and inverted cup* water vapor transmission (organized by glaze fit)[ daily weight (g)]/N)0.14 0.12 0.1 0.08FL A N 20 D R 09 O0.06 0.04 0.02 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 N= Number of days elapsedNo glaze fit defect1.1 1.1i[ daily weight (g)]/N)0.03 0.020.01No glaze fit defect1.2 1.2i0.000.011 2 3 4 5 6 7 8 9 10111213141516171819202122232425 N= Number of days elapsed[ daily weight (g)]/N)0.20 0.16 0.12 0.04 0.00No glaze fit defect1.3 1.3i0.081 2 3 4 5 6 7 8 9 10111213141516171819202122232425 N= Number of days elapsed0.04 0.03 0.02[ daily weight (g)]/N)No glaze fit defect1.4 1.4i0.01 0.00 0.01 1 2 3 4 5 6 7 8 9 10111213141516171819202122232425 N= Number of days elapsed*i denotes data from inverted cup water vapor transmission195- FLANDROAverage daily weight loss (g/day):[ daily weight (g)]/N)Comparison between regular water vapor transmission and inverted cup* water vapor transmission (organized by glaze fit)0.03 0.02 0.01 0.00 0.01 0.02 0.03 0.04No glaze fit defect1.5 1.5iFL A N 20 D R 09 O1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 N= Number of days elapsed 0.05[ daily weight (g)]/N)0.04 0.03 0.02No glaze fit defect1.6 1.6i0.010.000.011 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 N= Number of days elapsed[ daily weight (g)]/N)0.04 0.03 0.02No glaze fit defect1.7 1.7i0.010.001 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 N= Number of days elapsed[ daily weight (g)]/N)0.06 0.04 0.03 0.020.05No glaze fit defect1.8 1.8i0.01 0.00 0.01 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 N= Number of days elapsed196- FLANDROAverage daily weight loss (g/day):Comparison between regular water vapor transmission and inverted cup* water vapor transmission (organized by glaze fit)[ daily weight (g)]/N)0.09 0.07 0.05 0.03 0.01 0.01No glaze fit defect3.2 3.2iFL A N 20 D R 09 O1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 N= Number of days elapsed 0.03 0.02 0.00 0.01No glaze fit defect3.4 3.4i[ daily weight (g)]/N)0.010.021 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 N= Number of days elapsed0.080.06 0.04 0.02 0.00No glaze fit defect3.5 3.5i[ daily weight (g)]/N)1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 N= Number of days elapsed0.030.01No glaze fit defect3.6 3.6i0.011 2 3 4 5 6 7 8 9 10111213141516171819202122232425 N= Number of days elapsed[ daily weight (g)]/N)0.06 0.04 0.02 0.00 0.02 1 2 3 4 5 6 7 8 9 10111213141516171819202122232425 N= Number of days elapsed197- FLANDRONo glaze fit defect3.8 3.8iAverage daily weight loss (g/day):Comparison between regular water vapor transmission and inverted cup* water vapor transmission (organized by glaze fit)[ daily weight (g)]/N)0.60 0.40 0.20 0.00FL A N 20 D R 09 O[ daily weight (g)]/N)0.19 0.14 0.09 0.04 0.01Craze Only2.4 2.4i1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 N= Number of days elapsedCraze Only2.6 2.6i1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 N= Number of days elapsed[ daily weight (g)]/N)0.40 0.30 0.20 0.10 0.00Craze Only2.7 2.7i0.101 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 N= Number of days elapsed[ daily weight (g)]/N)0.19 0.14 0.09 0.04Craze Only3.1 3.1i0.010 1 2 3 4 5 6 7 8 9 10111213141516171819202122232425 N= Number of days elapsed198- FLANDROAverage daily weight loss (g/day):Comparison between regular water vapor transmission and inverted cup* water vapor transmission (organized by glaze fit)[ daily weight (g)]/N)FL A N 20 D R 09 O0.20 0.10 0.00 N= Number of days elapsed 0.300.30Craze & Crawl2.1 2.1i1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25[ daily weight (g)]/N)0.250.200.15Craze & Crawl2.2 2.2i0.100.050.001 2 3 4 5 6 7 8 9 10111213141516171819202122232425 N= Number of days elapsed[ daily weight (g)]/N)0.16 0.14 0.12 0.10 0.08 0.06 0.04 0.02 0.00Craze & Crawl2.3 2.3i1 2 3 4 5 6 7 8 9 10111213141516171819202122232425 N= Number of days elapsed199- FLANDROAverage daily weight loss (g/day):Comparison between regular water vapor transmission and inverted cup* water vapor transmission (organized by glaze fit)[ daily weight (g)]/N)0.15 0.13 0.11 0.09 0.07 0.05FL A N 20 D R 09 O1 2 3 4 5 6 7 8 9 10111213141516171819202122232425 N= Number of days elapsedCrawl only2.5 2.5i[ daily weight (g)]/N)0.07 0.050.06 0.04 0.03 0.02 0.00Crawl only2.8 2.8i0.011 2 3 4 5 6 7 8 9 10111213141516171819202122232425 N= Number of days elapsed0.09[ daily weight (g)]/N)0.07Crawl only3.3 3.3i0.050.03 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 N= Number of days elapsed200- FLANDROAverage daily weight loss (g/day):Comparison between regular water vapor transmission and inverted cup* water vapor transmission (organized by glaze fit)[ daily weight (g)]/N)0.35 0.30 0.20 0.10 0.25 0.15 0.05FL A N 20 D R 09 O1 2 3 4 5 6 7 8 9 10111213141516171819202122232425 N= Number of days elapsed201- FLANDROGlaze scratch3.7 3.7iFL A N 20 D R 09 OQ-TEST: MATHEMATICS202- FLANDRO203- FLANDROFL A N 20 D R 09 O204- FLANDROFL A N 20 D R 09 O205- FLANDROFL A N 20 D R 09 O206- FLANDROFL A N 20 D R 09 O207- FLANDROFL A N 20 D R 09 O208- FLANDROFL A N 20 D R 09 O

Recommended

View more >