lable at ScienceDirect

Polymer Degradation and Stability 102 (2014) 195e203

Contents lists avai

Polymer Degradation and Stability

journal homepage: www.elsevier .com/locate /polydegstab

Biodegradability of crude glycerol-based polyurethane foams duringcomposting, anaerobic digestion and soil incubation

Eddie F. Gómez, Xiaolan Luo, Cong Li, Frederick C. Michel Jr. **, Yebo Li*

Department of Food, Agricultural and Biological Engineering, The Ohio State University, Ohio Agricultural Research and Development Center, 1680 MadisonAve., Wooster, OH 44691-4096, USA

a r t i c l e i n f o

Article history:Received 6 December 2013Received in revised form2 January 2014Accepted 11 January 2014Available online 22 January 2014

Keywords:Polyurethane foamsBiodegradabilityCompostingSoilAnaerobic digestionBiopolyols

* Corresponding author. Tel.: þ1 330 263 3855; fax** Corresponding author. Tel.: þ1 330 263 3859; fax

E-mail addresses: Michel.36@osu.edu (F.C. Michel)

0141-3910/$ e see front matter � 2014 Elsevier Ltd.http://dx.doi.org/10.1016/j.polymdegradstab.2014.01.0

a b s t r a c t

In this study, the biodegradability of polyurethane (PU) foams made from crude glycerol- and petroleum-based polyols was compared during composting, anaerobic digestion (AD), and soil incubation. Chemicalchanges in the PU foams before and after composting were further analyzed using ThermogravimetricAnalysis (TGA), Scanning ElectronMicroscopy (SEM), EvolvedGasAnalysiseMass Spectrometry (EGAeMS),and Fourier Transform Infrared Spectroscopy (FT-IR). The results showed that during composting, AD, andsoil incubation conditions, none of the PU foams were mineralized more than 10%. PU foams made from100% crude glycerol-based polyols were mineralized during 320 days of soil incubation at rates faster thanthose observed for the petroleum-based analogs. However, no significant differences in soil mineralizationrates were observed between PU foams made from blend polyols, which contained 50% crude glycerol-based polyols and 100% petroleum-based polyols. SEM analysis showed that some surface deteriorationoccurred in the PU samples made from bio-based and blend polyols during composting. Minor differenceswere observed in the TGA curves of the PU foams made from petroleum-based polyols before and aftercomposting and pronounced differences occurred in PU foams made from both crude glycerol-based andblend polyols in the thermal regions of urethane and ester segments. In terms of EGAeMS analyses, themajor degradation of PU foams made from crude glycerol-based and blend polyols was attributed to thedecomposition of FAMEs and fatty acid chains in the polyol side of the polymer. FT-IR analysis showed thatlittle degradation of urethane and ester segments of the polymer occurred during composting ofpetroleum-based PU foam. FT-IR analysis of PU foams made from 100% crude glycerol-based polyolsrevealed that the ester segments (eCOOe) of the material were the preferred sites of microbial attack. ThePU foamsmade from blend polyols showed some structural changes in urethane linkages (eNHCOOe) butdegradation was more noticeable in the ester segments (eCOOe) of the polymer, similar to that observedfor 100% bio-based polyols.

� 2014 Elsevier Ltd. All rights reserved.

1. Introduction

It is estimated that the polyurethane industry in the U.S. con-tributes approximately $59.9 billion to the national economy. Threemajor sectors where polyurethanes are used are building andconstruction, transportation, and furniture and bedding [1]. Poly-urethane (PU) foams, such as rigid and flexible foams, are one typeof widely used PU products that include insulation panels, seatingcushions, adhesives, coatings, sealants, etc.

PUs are generally produced by the reaction of two chemicalfeedstocks; polyols and isocyanates. Nearly all of these feedstocks

: þ1 330 263 3670.: þ1 330 263 3670., li.851@osu.edu (Y. Li).

All rights reserved.08

aremade from petroleum [2]. Global concerns about the availabilityand rising price of petrochemical-derived products have led to in-vestigations into substitutes that are produced from renewablesources and can meet cost and performance requirements of themarket [3]. This represents both challenges and opportunities forrenewable-based chemicals to supply this billion dollar industry.

Due to the limited choice of isocyanates, a majority of theresearch on renewable substitutes used for the production of PUshas focused on the polyol component. Polyols are alcohols con-taining two or more hydroxyl functional groups. Currently, mostbio-based polyols (biopolyols) are produced from lignocellulosicbiomass or vegetable oils [4]. Lignocellulosic feedstocks such ascornstalks, wheat straw, and dried distillers grains have beenstudied for bio-based polyols [5,6]. Vegetable oils including castoroil, soybean oil, and palm oil have also been studied as sources ofbio-based polyols [7,8]. Recently, increasing interest has been

Table 1Properties of the petroleum- and bio-based polyols used to produce the PU foamssamples.

Polyol Major components Acid numberb

(mg KOH/gsample)Hydroxylnumberb

(mg KOH/gsample)

Petroleum-based

2.5 � 1 315 � 4

Bio-based

a

5.0 � 1 524 � 7

a Oleic acid is used as a representative fatty acid in crude glycerol.b The acid and hydroxyl numbers were determined in accordance with ASTM

D4662-08 and ASTM D4274-05D, respectively. Values are means � standard devi-ation of three determinations.

E.F. Gómez et al. / Polymer Degradation and Stability 102 (2014) 195e203196

focused on the value-added uses of crude glycerol, a byproduct ofbiodiesel production and a potential renewable substitute forpetroleum-based feedstocks. Several studies have reported theproduction of crude glycerol-based biopolyols with propertiessuitable for PU applications [9e11].

While significant achievements in the production of PUs frombio-based polyols have been made [12], there are still numerousuncertainties about the interaction of these polymers with theenvironment. Understanding the biodegradability of PUs is partic-ularly important since substantial amounts end up in waste man-agement facilities and in natural environments [13]. From anengineering standpoint, biodegradability is desirable for single-useand short lifespan applications, such as packagingmaterials, but areundesirable for long lifespan applications, such as automotive andconstruction uses [14]. From an environmental standpoint, thepotential biodegradability of products and their interaction withthe environment need to be understood in addition to the mate-rial’s life cycle [15].

Biodegradation is the transformation of materials as a result ofthe action of naturally-occurring microorganisms such as bacteriaand fungi [16]. The biodegradation of PUs is limited bymany factorsincluding chemical properties of the polymer such as chemicalstructure, cross-linking density, crystallinity, and the fact that mostplastics are xenobiotic [17]. Microbially-induced deterioration ofthe urethane bonds in PUs has been reported to be influenced bythe polyol type, i.e. polyether or polyester polyols [18]. PUs madefrom polyether polyols have been found to be relatively resistant tomicrobial degradation, whereas the polyester analogs are reportedto be more vulnerable to biodegradation processes [19]. Differencesin biodegradation rates have been attributed to the degradationmechanism in which endo- and exo-type enzymatic depolymer-ization pathways are undergone for polyester and polyether PUs,respectively [18]. In polyesters, microbial degradation has beenpostulated to occur mainly due to the hydrolysis of the ester bondsby membrane-bound and extra-cellular polyurethanases [20].

Research has been conducted to evaluate the biodegradability ofPUs made from vegetable oil-based and liquefied biomass-basedpolyols. Wang et al. [21] studied the biodegradability of PU foamsmade from castor oil-derived polyols by measuring weight lossduring soil incubation and reported reductions ranging between 10and 40% in a 4-month period. In addition, microbial deterioration ofPUs made from castor oil has been reported to occur in the esterbonds of the polymer [8]. Shogren et al. [22] reported that PUsmade from different vegetable oil-based polyols with hydrolyzablebonds such as esters were biodegradable to an appreciable extent.In studies conducted by Zhang et al. [23], 16% mass loss in a 12-month period during soil burial experiments was reported andthe hydrolysis of the ester and urethane bonds due to microbialactivity in PUs made from liquefied wood-based polyols wasobserved. Weight losses ranging between 6 and 14% after 6 monthsof soil incubation were reported for PUs produced from liquefiedwaste paper [24] and wheat straw [25] and were attributed to thedegradation of the biomass portion of the polymers.

Studies on the biodegradation of PUs produced from crudeglycerol-based polyols have not been reported. More researchneeds to be conducted to understand the extent towhich PUsmadefrom crude glycerol-based polyols will biodegrade in differentwaste management scenarios, such as composting, anaerobicdigestion (AD) or natural settings. The objective of this study was tocompare the relative biodegradability of PU foams produced frompetroleum- and crude glycerol-based polyols under composting,AD, and soil conditions. The hypothesis of this research is that PUfoams made from polyols with higher bio-based content arechemically different and will biodegrade to a greater extent thantheir petroleum-based analogs.

2. Materials and methods

2.1. Materials

Petroleum-based (Stepanpol�) and bio-based (crude glycerol-derived) polyester polyols were provided by Bio100 Technologies,LLC. (Mansfield, OH, U.S.). Bio-based polyol is produced via a ther-mochemical conversion of crude glycerol. It is commonly composedof glycerol, monoglycerides, diglycerides, and unreacted fatty acidmethyl esters [11]. Chemical properties of the polyols are shown inTable 1. Polycat 5, Polycat 8, and DABCODC5357 used in PU foamingwere obtained from Air Products & Chemicals, Inc. (Allentown, PA,U.S.). Polymeric methylene-4,40-diphenyl diisocyanate (pMDI) wasobtained from Bayer Material Science (Pittsburgh, PA, U.S.). Threetypes of PU foamswere prepared from petroleum-based, bio-based,and their blend polyols (50/50, w/w) via their polyaddition reactionwith pMDI (Scheme 1) following a commonly used formula [26]and ASTM D7487-08.

PU foams were cut in 10 � 10 mm squares with an averagethickness of 2.1 � 0.4 mm for composting, AD, and soil incubationtests. Physical and chemical properties of the foams are shown inTable 2. Cellulose paper (Fisher Scientific, PA, U.S.) was used as thepositive control for all experiments.

2.2. Biodegradation of polyurethane foams

Laboratory-scale experiments were conducted to measure therelative biodegradability of PU foams during soil incubation, com-posting, and AD.

Experiments to determine the biodegradability of PU foams insoil were conducted using an assay based on a standard protocol(ASTM D5988-03). A more complete description of the methodol-ogy to study themineralization of polymeric materials in soil can befound elsewhere [27]. The soil media used for the experiments wasa mixture of 43% certified organic top soil, 43% no-till farm soil, and14% sand. The chemical properties of the soil mixture are shown inTable 3. The soil media was amended with ammonium phosphate

NCONCO NCO

CH2 CH2

n

R OHHO

R O NH

CH2 CH2

n

NH

O

ONH

OOO

R

R

Polyol pMDI Polyurethanen=0_4 n=0_4

Scheme 1. Production of PU via a polyaddition reaction of polyols with pMDI.

Table 3Initial properties of the aerobic and anaerobic organic substrates.a

Organicsubstrate

Total solids(%wbg)

Volatilesolids(%dbg)

Totalcarbon(%db)

Totalnitrogen(%db)

pH

Compostb

inoculum43.7 � 0.8 71.5 � 12 38.9 � 3.5 2.98 � 0.4 8.19 � 0.02

Soil mixturec 90.5 � 0.1 2.8 � 0.03 0.9 � 0.4 0.12 � 0.03 6.63 � 0.40

E.F. Gómez et al. / Polymer Degradation and Stability 102 (2014) 195e203 197

(Fisher Scientific, PA, U.S.) to maintain a C:N ratio of 20:1 includingthe carbon content of the PU foams.

The soil mixture (300 g dry) was placed in the bottom of a 2-L(working volume) wide mouth jar (Ball� Corporation, item #383178, IN, US). Distilled water was added to bring the moisturecontent of the mixture to 60% of its moisture holding capacity. PUfoams (1 g of sample carbon) were then mixed thoroughly with thesoil. A solution containing 20 mL of 0.5 N potassium hydroxide(KOH) (Fisher Scientific, PA, U.S.) was placed in a cup suspendedfrom the lid of each vessel to trap evolved carbon dioxide (CO2).Vessels were incubated at 27 � 1 �C in triplicate for a period of 320days. The amount of CO2 produced was determined by titrating theKOH solution with 0.25 N hydrochloric acid (Fisher Scientific, PA,U.S.) to a phenolphthalein end-point. Vessels were allowed to sitopen for 5 min during titration to prevent oxygen concentration inthe vessel from falling below 18% during the experiment.

The compostability of PU foams under conditions that mimic acommercial scale industrial composting facility was conductedbased on ASTM D5338-98. A more complete description of themethodology used to measure the compostability of polymericmaterials can be found elsewhere [27]. PU foams (15 g sample)were mixed with 350 g (dry weight) of mature compost made fromdairy manure and hardwood sawdust (Table 3). The compostinoculum was collected from a full-scale windrow composting fa-cility at the Ohio Agricultural Research and Development Center(OARDC) inWooster, OH, U.S. Characteristics of this type of compostmixture are described elsewhere [28]. Ammonium phosphate(Fisher Scientific, PA, U.S.) was added to the mixture to give a C:Nratio of 20:1, which included the carbon content of the PU foams.The initial moisture content of the mixture was adjusted to 60%.

The mixtures containing the compost inoculum and PU foamswere incubated in triplicate reactors for a period of 50 days. Re-actors consisted of 4-L (working volume) vessels incubated at 55 �C(BioCold Environmental Inc., MO, U.S.). The composting vesselswere aerated from the bottom of the reactor at 100 þ 1 mL/min[29]. Before entering the vessels, the air was saturated with waterby bubbling through bottles containing deionized water. The off-gas was then analyzed for CO2 percent using an infrared gasanalyzer (Vaisala model GMT 220, range 0e20%). CO2 data wasautomatically recorded using a Campbell Scientific (UT, U.S) model23XL data logger for each vessel every hour.

The anaerobic biodegradation tests of PU foams undercontrolled high-solids mesophilic (37 � 1 �C) AD conditions wereconducted based on a standard protocol (ASTM D5511-02). These

Table 2Initial characteristics of the positive control and PU foam samples.a

Sample Total solids(%wbb)

Volatile solids(%dbb)

Total carbon(%db)

Total nitrogen(%db)

Positive control(cellulose paper)

96.1 � 3.0 98.5 � 1.3 41.8 � 0.1 0.03 � 0.010

Petroleum-based 98.0 � 0.6 99.1 � 1.0 65.8 � 0.2 8.02 � 0.005Bio-based 95.0 � 1.0 97.9 � 1.0 72.3 � 0.4 8.07 � 0.020Blend 97.9 � 1.1 98.8 � 1.6 67.6 � 0.3 7.15 � 0.030

a Values are means � standard deviation of three determinations.b wb and db are wet and dry weight basis, respectively.

tests resemble biologically active landfills and anaerobic digestersand measure the conversion of samples to CO2 and methane (CH4).

AD tests were conducted in triplicate for a period of 105 days in2-L (working volume) laboratory-scale batch reactors. Squares ofPU foamwith a total weight of 10 g were mixed with 450 g (wet) ofmethanogenically active sludge obtained in January 2013 from afull-scale (3000 m3) AD system treating municipal sewage sludge[30]. This was mixed with 60 g of the organic fraction of municipalsolid waste (OFMSW) and 30 g of dried corn stover (sieved to lessthan ½ in. particle size) to achieve the desired total solids contentfor the test (20%). The physical and chemical properties of theorganic substrates are shown in Table 3. Ammonium phosphate(Fisher Scientific, PA, U.S.) was added to the mixture to adjust theC:N ratio to a value of 20:1.

2.3. Analytical methods

The mineralization of PU foams under soil incubation, com-posting, and AD conditions was calculated by measuring theaverage carbon (CO2 and/or CH4) mineralized from each treatment,subtracting the average carbon evolved from the blanks, anddividing this by the total amount of sample carbon added to eachtreatment. The blanks contained mixtures of only inoculum andorganic substrates (no PU foam added).

A variety of analytical measurements were used to monitor theAD, composting, and soil incubation processes and characterize theorganic substrates and PU foams that were tested in the experi-ments. The total solid and volatile solid contents were determinedby drying the samples to constant weight at 80 �C and 500 �C,respectively. The pH was determined using a pH electrode (TMECC04.11-A 1:5 slurry method, mass basis). Carbon (TMECC 04.01-Acombustion with CO2 detection) and nitrogen content (TMECC04.02-D oxidation, Dumas method) were determined using a Var-ioMax N Analyzer (Elementar Americas, NJ, U.S.) by the OARDCSTAR Lab.

Anaerobicinoculumd

9.0 � 0.04 62.0 � 0.3 38.8 � 1.6 6.30 � 0.2 8.06 � 0.07

MedinaCountye

OFMSW

47.2 � 7.2 60.3 � 1.2 89.6 � 1.3 0.92 � 0.2 7.50 � 0.4

Corn stoverf 93.9 � 0.2 91.7 � 0.3 46.9 � 2.1 0.9 � 0.003 e

a Values are means � standard deviation of three determinations.b Dairy manure and hardwood sawdust mature compost.c A mixture of 43% certified organic top soil, 43% of no-till farm soil and 14% sand.d Methanogenically active municipal sewage sludge.e OFMSW ¼ the organic fraction of municipal solid waste.f Dry corn stover ground to ½ inch.g wb and db are wet and dry weight basis, respectively.

Fig. 1. Cumulative carbon loss (CO2-C) as a percentage of initial carbon (average ofthree determinations � cumulative standard error) for PU foams made from petro-leum, bio-based and blend polyols during (a) 320 days of soil incubation, (b) 50 days ofcomposting, and (c) 105 days of anaerobic digestion. For some data points standarderror bars are smaller than markers.

E.F. Gómez et al. / Polymer Degradation and Stability 102 (2014) 195e203198

Volumetric biogas production and composition during AD ex-periments were measured using a water displacement method andgas chromatography, respectively, as described by Gómez et al. [30].

Structural changes of PU foams before and after the compostingexperiments were analyzed by Fourier Transform Infrared Spec-troscopy (FT-IR), Scanning Electron Microscopy (SEM), Thermog-ravimetric Analysis (TGA), and Evolved Gas AnalysiseMassSpectrometry (EGAeMS). At the end of each experiment, PU foamcuboids were carefully rinsed with DDI water and vacuum dried at75 psi at 40 �C for 48 h (Isotemp� Vaccum Ovenmodel 282A, FisherScientific, PA, U.S.).

For the SEM analysis, samples were first coatedwith platinum toa thickness of 0.2 kÅ using a Hummer� 6.2 sputtering system(Anatech USA, CA, U.S.). SEM images were taken using a Hitachi S-3500N microscope (Hitachi High Technologies America, Inc., CA,U.S.) with a 15 kV electron beam at the Molecular and CellularImaging Center at the OARDC.

FT-IR spectra were recorded using a Spectrum Two� (PerkinElmer Inc., MA, U.S.) FTIR spectrophotomer equipped with a uni-versal attenuated total reflectance accessory (UATR). This FT-IR/UATR has a diamond crystal that allows for directly recording thespectra of the PU foams without sample preparation. FT-IR spectrawere averaged over 16 scans from 4000 to 450 cm�1 wavenumberwith a resolution of 4 cm�1. A background scan of the clean dia-mondwas recorded before scanning the samples. Spectra datawerenormalized and the baseline was corrected using Perkin ElmerSpectrum software (application version 10.03.07.0112, MA, U.S.).

Thermogravimetric analyses were carried out at a heating rateof 20 �C min�1 from 50 to 600 �C under nitrogen atmosphere usinga TA Q 50 thermogravimetric analyzer (TA Instruments, DE, U.S.).Thermogravimetric curves were obtained using approximately2 mg of sample. The Universal Analysis 2000 software was used forcurve analyses, version 4.5A (TA Instruments, DE, U.S.).

The compounds evolved during the thermal decomposition ofthe PU foams were characterized by evolved gas analysisemassspectrometry (EGAeMS) using a multi-shot pyrolyzer (EGA/PY-3030 D, Frontier Lab, Fukushima, Japan) coupled with a GCeMS(GCMS-QP2010 SE, Shimadzu, MD, U.S.) via a UADTM-2.5N column(2.5 m, 0.15 mm i.d., Frontier Lab, Fukushima, Japan). Helium wasused as the carrier gas at a flow rate of 1 mL min�1 at a split ratio of30. The column oven temperaturewas kept at 300 �C. The pyrolyzertemperature was held at 70 �C for 0.5 min, then heated to 600 �C ata rate of 20 �C min�1, and held at 600 �C for 1 min. The massspectrometer was operated in the electron ionization (EI) modewith m/z scan range of 0e500, and its ion source temperature waskept at 200 �C. The pyrolyzates were identified by comparing theirmass spectra with those reported in the NIST library.

2.4. Statistical analysis

Three independent replicates were used for each treatment.Analysis of variance (ANOVA) was calculated for the average finalcumulative percent of initial carbon loss for each of the studies.Comparisons for all pairs of final cumulative biodegradation meanswere performed using Student’s t method. All conclusions werebased on a significant difference level of a ¼ 0.05. The statisticalanalyses were performed using JMP statistical program version 9(SAS Institute Inc., SAS Campus Drive, NC, U.S.).

3. Results and discussion

3.1. Biodegradation studies

The mineralization of PU foams incubated in soil was measuredfor a period of 320 days (Fig. 1a). The highest initial extent of

mineralization was obtained with the PU foams made from bio-based polyols with 8.5 � 2.0% during the first 87 days (Fig. 1a).Mineralization of the bio-based PU foams continued through day146 reaching 10.9 � 2.2% (Fig. 1a). This was followed by a period ofslow mineralization that continued until the termination of theexperiment (Fig. 1a). Slower initial mineralization rates wereobserved for the PU foams made from petroleum-based and blendpolyols with values ranging between 1.0 and 2.5% during the first146 days of the experiment (Fig. 1a). The final (320 days) extents ofmineralization for the PU foams made from petroleum-based, bio-based, and blend polyols were 1.7�1.4%,11.2� 4.3%, and 2.9�1.0%,respectively (Fig. 1a). The positive control (cellulose paper)exhibited 65.4 � 6.0% mineralization during the same period(Fig. 1a).

E.F. Gómez et al. / Polymer Degradation and Stability 102 (2014) 195e203 199

ANOVA statistical analysis revealed that significant differences(p < 0.05) between group means of the total carbon loss for foamsmade from different polyol types existed during soil incubation.Further analysis revealed that the extent of mineralizationmeasured for the PU foams made from bio-based polyols wassignificantly different (p< 0.05) from those of the petroleum-basedand blend polyols. The difference between the petroleum-basedand the blend PU foams was not significant (p > 0.05).

Under composting condition the initial mineralization rates forthe three different PU foams were similar for the first 18 days withvalues ranging between 4 and 8% (Fig. 1b). This was followed by aperiod of continuous increase in the extent of mineralization for thePU foams made from bio-based and blend polyols until day 46 ofthe experiment (Fig. 1b). PU foams made from petroleum-basedpolyols on the other hand, showed almost no mineralization from

Fig. 2. SEM before and after incubation during 50 days of composting for PU foams made fbefore, f: after) polyols.

day 18 until the end of the experiment (Fig. 1b). The final (50 days)mineralization values for PU foams made from petroleum-based,bio-based, and blend polyols were 3.6 � 5.5%, 12.6 � 6.4%, and16.2 � 8.0%, respectively (Fig. 1b). The positive control (cellulosepaper) exhibited 77.2� 8.7%mineralization during the same 50 daycomposting period (Fig. 1b). Statistical analyses revealed that dif-ferences between groupmeans of the PU foams’ cumulative carbonloss observed during composting were not significant (p > 0.05).

SEM images of PU foams before and after composting are shownin Fig. 2. After composting, rough surfaces appeared for the PUfoams made from bio-based and blend polyols, while almost nosignificant surface change was observed for the petroleum-basedpolyols derived PU foams.

The biodegradation of PU foams under anaerobic conditions wasconducted over a period of 105 days in a batch incubation (Fig. 1c).

rom petroleum-based (a: before, b: after), bio-based (c: before, d: after), and blend (e:

Fig. 3. Thermogravimetric curves of PU foams made from (a) petroleum-based, (b)bio-based, and (c) blend polyols before and after 50 days of composting.

E.F. Gómez et al. / Polymer Degradation and Stability 102 (2014) 195e203200

Mineralization (total CO2 plus CH4 production) during the first 15days of study was slow for all treatments (Fig. 1c) followed by in-creases in the mineralization rate for the foams made from bio-based and blend polyols between days 22 and 29 of the studywith final mineralization values reaching 5.4 � 1.2 and 6.6 � 1.4%,respectively (Fig. 1c). The mineralization of PU foams made frompetroleum-based polyols proceeded at a slower rate through day67. The final extent of mineralization of the petroleum-based PUfoams was 3.5 � 3.3% (Fig. 1c). Final cumulative carbon loss valuesfor the PU foams made from bio-based and blend polyols, and thepositive control were 8.9 � 1.8%, 8.5 � 2.3%, and 79.4 � 5.6%,respectively (Fig. 1c).

Differences in mineralization observed between group meansduring AD studies were not significant (p > 0.05).

The extent of mineralization of the PU foams made frompetroleum-based polyols was similar under composting and ADconditions but a much lower extent was observed during soil in-cubation for the same material. For the PU foams made from bio-based and blend polyols, the highest extent of mineralization wasobserved during composting. This was followed by soil incubationfor the foamsmade from bio-based polyols and by the AD conditionfor the foams made from the blend analog. The lowest extent ofmineralization for the PU foams made from bio-based and blendpolyols was observed during AD and soil incubation, respectively.The relative mineralization rates were consistent in all three en-vironments. The 100% bio-based polyol derived PU foams miner-alized to a greater extent than 50% bio-based polyol derived PUfoams which in turn mineralized more rapidly than petroleum-based ones.

3.2. Thermogravimetric analysis

Before and after 50 days of composting, the thermal decompo-sition profiles of the PU foams made from petroleum-based polyolswere nearly identical (Fig. 3a). They were characterized by single-stage weight loss at a temperature range between 240 and400 �C (Fig. 3a). This thermally-induced single-step weight lossbehavior is characteristic of petroleum-based PU materials [31,32].Weight losses occurring in temperatures ranging between 150 and400 �C in petroleum-based PUs have been attributed to decom-position of urethane segments [33]. However, the onset of thethermal decomposition has been reported to occur anywhere be-tween 150 and 260 �C and can be influenced by many factorsincluding the types of substituents on the isocyanate and polyolsides [34], as well as the atmosphere used during thermaldecomposition analyses [31].

The PU foam samples made from bio-based polyols obtainedbefore and after composting showed different degradation profiles.They were both characterized by weight losses at two differenttemperature ranges (Fig. 3b). The first stage of thermal decompo-sition occurred at temperature ranges of 190e320 �C and 240e340 �C for the samples analyzed before and after composting,respectively (Fig. 3b). In the first stage, decomposition of urethanegroups in PUs made from vegetable oil-derived polyols has beenreported [8,35,36]. The greatest thermal decomposition for the bio-based PU foams before and after composting was observed duringthe second temperature range with approximately 50% weight loss(Fig. 3b). This stage occurred over temperature ranges of 340e530 �C and 360e540 �C for the samples analyzed before and aftercomposting, respectively (Fig. 3b). The weight losses in thesetemperature ranges might be mainly attributed to the decompo-sition of ester linkages and fatty acid chains [8,37] which havehigher stability against thermal decomposition than the urethanesegments [38]. This two-stage thermal decomposition behavior ofcrude glycerol-based PU foams was similar to that of PUs made

from vegetable oil-derived polyols [31]. PU foams made from blendpolyols showed thermogravimetric analysis (TGA) curves similar tothose from bio-based polyols (Fig. 3b and c). Two stages of weightlosses were observed for samples both before and after composting.The first stage in the blend PU foams occurred over temperatureranges of 180e320 �C for the samples before composting and 200e340 �C for the samples after composting (Fig. 3c). The second stageoccurred over temperature ranges of 350e540 �C and 370e550 �C(Fig. 3c) for the samples obtained before and after composting,respectively (Fig. 3c).

The thermal decomposition profiles of PU foams made frompetroleum-based polyols were nearly identical before and aftercomposting, indicating a low degree of degradability. PU foamsmade from bio-based and blend polyols showed higher thermalstability after undergoing a composting process, which indicatedthe occurrence of structural changes during biodegradation pro-cesses. Other studies conducted with PU foams made from vege-table oil- and liquefied biomass-derived polyols have also reported

E.F. Gómez et al. / Polymer Degradation and Stability 102 (2014) 195e203 201

shifts to higher temperature thermal decomposition after samplesunderwent a biodegradation process [8,23]. The increase in thethermal stability of PU foams from bio-based and blend polyolsafter compostingmight be attributed to a decrease in the content ofurethane segments during the composting process [39].

3.3. Evolved gas analysis

Mass spectrometry (MS) was used to identify evolved com-pounds at different stages of the thermal decomposition process ofPU foam samples before and after 50 days of composting. Thetemperature program used for this analysis was similar to that usedfor the TGA.

The thermograms of the petroleum-based PU foams showedevolution of compounds (Fig. 4a) over temperature ranges similarto decompositions observed during TGA (Fig. 3a). Differences

Fig. 4. EGA thermograms of PU foams made from (a) petroleum-based, (b) bio-based,and (c) blend polyols before and after incubation during 50 days of composting.

between the thermograms of the petroleum-based PU foams beforeand after composting were marginal. The thermograms of the PUfoams made from petroleum-based polyols were characterized bytwo peaks. The first small peaks of evolved compounds wereobserved in temperature ranges of 100e200 �C for samples testedbefore and after composting. MS analysis revealed that evolvedcompounds had characteristics of aromatic compounds, such asphthalic acid derivatives. The intensity of evolved compounds inthis thermal region was lower for the sample tested after com-posting. This reduction could be attributed to the degradation ofaromatic compounds with lowmolecular weights. The second peakof evolved compounds in the petroleum-based PU foams wasobserved at a temperature range from 220 to 400 �C (Fig. 4a). MSanalysis showed that the evolved compounds contained methylenediphenyl diisocyanate and diphenylmethane diamine, which aregenerated from the rupture of urethane bonds [31].

The thermograms of PU foams made from bio-based polyolsrevealed that major structural changes occurred during the com-posting process (Fig. 4b). For example, the peak at the temperaturerange between 110 and 200 �C for samples before composting,almost disappeared after composting. MS analysis revealed that theevolved compounds in this thermal region were fatty acid methylesters (FAMEs). These likely were residual compounds that werenot completely consumed in the production of bio-based polyols[11]. The result indicates that residual FAMEs in bio-based polyolswere degraded during the composting process. This could alsoexplain the initial minor weight loss observed in the TGA curve ofthe uncomposted PU foams made from bio-based polyols (Fig. 3b).A secondmajor shift occurred over the temperature range from 200to 320 �C and the onset temperature of the corresponding peakincreased by 30 �C for the samples analyzed after composting(Fig. 4b). MS analysis revealed that evolved compounds at thistemperature rangewere characteristic of fragments similar to thosereported during the breakage of urethane bonds. The third differ-ence in the thermograms of the bio-based PU foams before andafter composting occurred over the temperature range from 330 to540 �C. A significant reduction in peak intensity was observed afterthe composting process. An increase in the onset temperature ofthe corresponding peak was also observed for the composted PUfoams. Compounds evolved in this thermal region showedcompositional characteristics of fatty acid chains in the polyol sideof PU foams and diphenylmethane diamine which was generateddue to the intensified dissociation of urethane bonds formed frompMDI and hydroxyl groups of polyols.

The thermograms of the PU foams made from blend polyolsbefore composting were mainly characterized by three peaks(Fig. 4c). The first peak, which was attributed to the decompositionof FAME-like products, was observed over the temperature rangesimilar to that found for bio-based PU foams (Fig. 4b) and alsodisappeared after the composting process. The second peak in theblend PU foams was attributed to the decomposition of urethanegroups. The third peak at temperatures ranging from approxi-mately 345e540 �C also changed as a result of composting in bothintensity and onset temperature (Fig. 4c). Overall, these resultssuggest that structural changes in the PU foams made from eitherbio-based or blend polyols occurred during the composting processas a result of the degradation of FAMEs and ester compounds withfatty acid chains.

3.4. FT-IR analysis

FT-IR spectroscopy was further used to analyze structuralchanges in PU foams made from petroleum-based, bio-based, andblend polyols before and after 50 days of composting (Fig. 5). Thespecific location of important bands in the FT-IR spectra of PU

Fig. 5. FT-IR spectra of PU foams produced from (a) petroleum-based, (b) bio-based,and (c) blend polyols before and after incubation during 50 days of composting.

E.F. Gómez et al. / Polymer Degradation and Stability 102 (2014) 195e203202

foams was assigned according to previous reports [40,41]. Char-acteristic bands for PU foams at 1721e1722, 1529e1536, 1510e1511,1281e1306, and 1204e1220 cm�1 were assigned to hydrogen-bonded carbonyl (C]O) stretching, hydrogen-bonded and freeNeH bending, and urethane and ester CeO stretching vibrations,respectively (Fig. 5). The band at 2277e2280 cm�1 was assigned tothe asymmetric stretching of isocyanate groups (Fig. 5).

The band at 2277e2280 cm�1 in the FT-IR spectra of the PUfoams made from petroleum-based, bio-based, and blend polyolsbefore composting indicated that there were still unreacted isocy-anate groups present in these samples (Fig. 5). The intensity of thisband decreased substantially after composting suggesting thatisocyanate groups underwent further reactions during composting.Reactions may have included the formation of additional urethanelinkages, and in environments with high moisture content, such asin composts, these unreacted isocyanate groups may have reacted

with water to form urea linkages [42]. Formation of urea groups inthe hard segments of the PU structure could have been one of thefactors influencing the thermal degradation profiles describedabove (Figs. 3 and 4). These groups have been reported to be morethermally stable than urethane bonds [43].

Slight reductions in the intensity of the bands at 1721, 1529, and1281 cm�1 were observed in the spectra of the PU foamsmade frompetroleum-based polyols after composting (Fig. 5a). These resultssuggest that deterioration occurred in the urethane bonds of the PUstructure during composting. Biodegradation of these bonds due tomicrobial hydrolytic enzyme activity has been previously reportedto occur in PUs made from petroleum-based polyester polyols[44,45]. Breakage of these bonds could also be one of the factorsinfluencing the thermal degradation profiles (Figs. 3a and 4a) asthese bonds are less thermally stable than other major parts of thepolymer structure [39].

The band at 1721, together with the one at 1220 cm�1 in thepetroleum-based PU foams, corresponded to ester groups in thepolyol segments of the PU structure. A slight decrease in the CeOstretching intensity of the aromatic ester indicates that these estersegments were not preferred sites for microbial attack duringcomposting.

Characteristic bands of urethane bonds in the bio-based PUfoams were located at 1722, 1536, and 1512 cm�1 (Fig. 5b). Nosubstantial decreases were observed except for the 1720 cm�1 band(Fig. 5b). Decreases in the carbonyl (C]O) bond could have beenrelated to the ester segments of the foams. Further analysisrevealed that bands assigned to the ester CeO bond (1204 cm�1)also showed significant decreases in intensity after composting(Fig. 5b). These results suggest that the ester bonds of glycerideand/or FAME structures were preferred sites of attack for biodeg-radation of bio-based PU foams during composting. Other studieson the biodegradation of PUs made from bio-based polyols,including castor oil [8,21] and soybean oil [22], have reported thatthese glyceride-based ester linkages are susceptible to microbialattack.

The FT-IR spectra of the PU foams made from blend polyolsbefore and after composting are shown in Fig. 5c. Small decreases inthe intensities of the bands assigned to the urethane segments at1536 (NeH), 1510 (NeH), and 1301 cm�1 (urethane CeO) vibrationswere observed. Obvious decreases were observed in those bandsassociated with the ester segments at 1722 (C]O) and 1214 (CeO) cm�1 stretching vibrations (Fig. 5c). These results indicate thatester segments in the PU foams made from blend polyols were thepreferred site for microbial attack, whereas little degradationoccurred in the urethane linkages.

4. Conclusion

This study showed that during composting, anaerobic digestion,and soil incubation a limited amount of mineralization of PU foamsoccurred. PU foams made from 100% bio-based polyols exhibited agreater extent of biodegradation than petroleum-based PU foamduring extended (320 days) soil incubation. No significant differ-ences in mineralization were observed between PU foams fromeither bio-based or blend polyols and their petroleum-basedcounterparts during composting or anaerobic digestion.

SEM analysis showed that some surface deterioration occurredin the PU foam samples made from bio-based and blend polyolsduring composting. Although some differences were observed inthe TGA curves of the petroleum-based foams analyzed before andafter composting, the most prominent structural changes occurredin the PU foamsmade from bio-based and blend polyols. In terms ofEGAeMS analyses, the major degradation of PU foams made frombio-based and blend polyols during composting was attributed to

E.F. Gómez et al. / Polymer Degradation and Stability 102 (2014) 195e203 203

the decomposition of FAMEs and partial decomposition of glyceridestructures in the PU foams. FT-IR analytical results further showedthat ester segments (C]O and CeO) were preferred sites of attackfor the degradation of PU foams made from 100% bio-based andblend polyols and very little degradation of urethane or ester seg-ments in petroleum-based PU foams occurred during composting.

Acknowledgments

The authors thank Department of Food, Agricultural and Bio-logical Engineering (FABE), The Ohio State University and USDA-NIFA for financial support. The authors would like to thank Mrs.Mary Wicks (FABE, OSU) for reading through the manuscript andproviding useful suggestions.

References

[1] ACC. The economic impact of the polyurethanes industry in 2010. Economics& Statistics Department, American Chemistry Council; 2012.

[2] Desroches M, Escouvois M, Auvergne R, Caillol S, Boutevin B. From vegetableoils to polyurethanes: synthetic routes to polyols and main industrial prod-ucts. Polym Rev 2012;52(1):38e79.

[3] Babb DA. Polyurethanes from renewable resources. Adv Polym Sci 2012;245:315e60.

[4] Li Y. Development of polyurethane foam and its potential within the biofuelsmarket. Biofuels 2011;2(4):357e9.

[5] Liang L, Mao Z, Li Y, Wan C, Wang T, Zhang L, et al. Liquefaction of crop res-idues for polyol production. Bioresources 2006;1(2):248e56.

[6] Xu J, Jiang J, Hse C, Shupe TF. Renewable chemical feedstocks from integratedliquefaction processing of lignocellulosic materials using microwave energy.Green Chem 2012;14(10):2821e30.

[7] Campanella A, Bonnaillie LM, Wool RP. Polyurethane foams from soyoil-basedpolyols. J Appl Polym Sci 2009;112(4):2567e78.

[8] Cangemi JM, Neto SC, Chierice GO, dos Santos AM. Study of the biodegradationof a polymer derived from castor oil by scanning electron microscopy, ther-mogravimetry and infrared spectroscopy. Polímeros: Ciencia Tecnol2006;16(2):129e35.

[9] Hu S, Wan C, Li Y. Production and characterization of biopolyols and poly-urethane foams from crude glycerol based liquefaction of soybean straw.Bioresour Technol 2012;103(1):227e33.

[10] Li Y, Zhou Y. Methods for producing polyols using crude glycerin. In: UnitedStates patent and trademark office. United States: The Ohio State UniversityResearch Foundation; 2011. p. 9.

[11] Luo X, Hu S, Zhang X, Li Y. Thermochemical conversion of crude glycerol tobiopolyols for the production of polyurethane foams. Bioresour Technol2013;139:323e9.

[12] Petrovic ZS. Polyurethanes from vegetable oils. Polym Rev 2008;48(1):109e55.

[13] Hammer J, Kraak MHS, Parsons JR. Plastics in the marine environment: thedark side of a modern gift. Rev Environ Contam Toxicol 2012;220:1e44.

[14] Urgun-Demirtas M, Singh D, Pagilla K. Laboratory investigation of biode-gradability of a polyurethane foam under anaerobic conditions. Polym DegradStabil 2007;92(8):1599e610.

[15] Narayan R. Carbon footprint of bioplastics using biocarbon content analysisand life-cycle assessment. MRS Bull 2011;36(9):716e21.

[16] ASTM. Standard terminology relating to plastics (standard D883-11). WestConshohocken, PA: ASTM International; 2011.

[17] Howard GT. Biodegradation of polyurethane: a review. Int Biodeterior Bio-degrad 2002;49(4):245e52.

[18] Nakajima-Kambe T, Shigeno-Akutsu Y, Nomura N, Onuma F, Nakahara T.Microbial degradation of polyurethane, polyester polyurethanes and poly-ether polyurethanes. Appl Microbiol Biotechnol 1999;51(2):134e40.

[19] Darby RT, Kaplan AM. Fungal susceptibility of polyurethanes. Appl Microbiol1968;16(6):900e5.

[20] Zheng Y, Yanful EK, Bassi AS. A review of plastic waste biodegradation. CritRev Biotechnol 2005;25(4):243e50.

[21] Wang HJ, Rong MZ, Zhang MQ, Hu J, Chen HW, Czigany T. Biodegradable foamplastics based on castor oil. Biomacromolecules 2008;9(2):615e23.

[22] Shogren RL, Petrovic Z, Liu Z, Erhan SZ. Biodegradation behavior of somevegetable oil-based polymers. J Polym Environ 2004;12(3):173e8.

[23] Zhang HR, Pang H, Zhang L, Chen X, Liao B. Biodegradability of polyurethanefoam from liquefied wood based polyols. J Polym Environ 2012;21(2):329e34.

[24] Lee S-H, Teramoto Y, Shiraishi N. Biodegradable polyurethane foam fromliquefied waste paper and its thermal stability, biodegradability, and geno-toxicity. J Appl Polym Sci 2002;83(7):1482e9.

[25] Chen F, Lu Z. Liquefaction of wheat straw and preparation of rigid poly-urethane foam from the liquefaction products. J Appl Polym Sci 2009;111(1):508e16.

[26] Tu Y-C, Kiatsimkul P, Suppes G, Hsieh F-H. Physical properties of water-blownrigid polyurethane foams from vegetable oil-based polyols. J Appl Polym Sci2007;105(2):453e9.

[27] Gómez EF, Michel Jr FC. Biodegradability of conventional and bio-basedplastics and natural fiber composites during composting, anaerobic diges-tion and long-term soil incubation. Polym Degrad Stabil 2013;98(12):2583e91.

[28] Michel FC, Pecchia JA, Rigot J, Keener HM. Mass and nutrient losses during thecomposting of dairy manure amended with sawdust or straw. Compost SciUtil 2004;12(4):323e34.

[29] Grewal SK, Rajeev S, Sreevatsan S, Michel FC. Persistence of Mycobacteriumavium subsp paratuberculosis and other zoonotic pathogens during simulatedcomposting, manure packing, and liquid storage of dairy manure. Appl En-viron Microbiol 2006;72(1):565e74.

[30] Gómez E, Martin J, Michel FC. Effects of organic loading rate on reactor per-formance and archaeal community structure in mesophilic anaerobic di-gesters treating municipal sewage sludge. Waste Manag Res 2011;29(11):1117e23.

[31] Javni I, Petrovi�c ZS, Guo A, Fuller R. Thermal stability of polyurethanes basedon vegetable oils. J Appl Polym Sci 2000;77(8):1723e34.

[32] Zhang C, Feng S. Effect of glycols on the properties of polyester polyols and ofroom-temperature-curable casting polyurethanes. Polym Int 2004;53(12):1936e40.

[33] Petrovi�c ZS, Zavargo Z, Flyn JH, Macknight WJ. Thermal degradation ofsegmented polyurethanes. J Appl Polym Sci 1994;51(6):1087e95.

[34] Wirpsza Z, Kemp TJ. Polyurethanes: chemistry, technology, and applications.New York: E. Horwood; 1993.

[35] Guo A, Javni I, Petrovi�c ZS. Rigid polyurethane foams based on soybean oil.J Appl Polym Sci 2000;77:467e73.

[36] Monteavaro LL, Riegel IC, Petzhold CL, Samios D. Thermal stability of soy-based polyurethanes. Polímeros: Ciencia Tecnol 2005;15:151e5.

[37] Claro Neto S. Physical and chemical characterization of castor-oil derivedpolyurethanes used in bone implants [Doctoral thesis]. Brazil: University ofSão Paulo/São Carlos Institute of Chemistry; 1997.

[38] Simon J, Barla F, Kelemen-Haller A, Farkas F, Kraxner M. Thermal stability ofpolyurethanes. Chromatographia 1988;25(2):99e106.

[39] Lu Y, Larock RC. Soybean-oil-based waterborne polyurethane dispersions:effects of polyol functionality and hard segment content on properties. Bio-macromolecules 2008;9(11):3332e40.

[40] de Haseth JA, Andrews JE, McClusky JV, Priester RD, Harthcock MA, Davis BL.Characterization of polyurethane foams by mid-infrared fiber/FT-IR spec-trometry. Appl Spectrosc 1993;47(2):173e9.

[41] Lobo H, Bonilla JV. Handbook of plastics analysis. New York: Marcel Dekker;2003.

[42] Leventis N, Sotiriou-Leventis C, Chandrasekaran N, Mulik S, Larimore ZJ, Lu H,et al. Multifunctional polyurea aerogels from isocyanates and water. Astructureeproperty case study. Chem Mater 2010;22(24):6692e710.

[43] Chang L-C, Xue Y, Hsieh F-H. Dynamic-mechanical study of water-blown rigidpolyurethane foams with and without soy flour. J Appl Polym Sci 2001;81(8):2027e35.

[44] Filip Z. Decomposition of polyurethane in a garbage landfill leakage water andby soil microorganisms. Eur J Appl Microbiol 1978;5(3):225e31.

[45] Ozsagiroglu E, Iyisan B, Guvenilir YA. Biodegradation and characterizationstudies of different kinds of polyurethanes with several enzyme solutions.Polym J Environ Stud 2012;21(6):1777e82.