Atmospheric and soil degradation of aliphatic–aromatic polyester films

lable at ScienceDirect

Polymer Degradation and Stability 95 (2010) 99e107

Contents lists avai

Polymer Degradation and Stability

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

Atmospheric and soil degradation of aliphaticearomatic polyester films

Thitisilp Kijchavengkul a, Rafael Auras a,*, Maria Rubino a, Edgar Alvarado b,José Roberto Camacho Montero b, Jorge Mario Rosales b

a School of Packaging, Michigan State University, East Lansing, MI 48824, United Statesb EARTH University, Guácimo, Limón, Costa Rica

a r t i c l e i n f o

Article history:Received 4 September 2009Received in revised form20 November 2009Accepted 26 November 2009Available online 2 December 2009

Keywords:Aliphaticearomatic polyesterPBATPhotodegradationBiodegradationMulch films

* Corresponding author. Tel.: þ1 517 432 3254.E-mail address: [email protected] (R. Auras).

0141-3910/$ e see front matter � 2009 Elsevier Ltd.doi:10.1016/j.polymdegradstab.2009.11.048

a b s t r a c t

The degradation of an aliphaticearomatic biodegradable polyester film was studied under conditions ofsolar exposure and soil burial in a tropical area. Film samples were evaluated for changes over 40 weeksby visual examination, scanning electronic microscopy (SEM), Fourier transform infrared (FTIR)spectroscopy, mechanical properties, molecular weight, gel content, and thermal properties. Photo-degradation played a major role in the atmospheric degradation of the film, causing it to lose integrityand mechanical properties after week 8 due to main chain scission and crosslinking. SEM micrographsand FTIR spectra indicated that photodegradation started at the exposed side of the film and propagatedthrough the polymer matrix after week 8. FTIR spectra also indicated that subsequent photooxidationprocesses took place. The reduction of molecular weight of the soil burial samples was much slower thanthat of the non-crosslinked portion of solar exposed film samples. The reduction of number averagemolecular weight of the non-crosslinked solar exposed samples followed a first order reaction, whereasthe soil burial samples show a surface erosion biodegradation behavior. The relationship among totalsolar radiation, gel content and number average molecular weight indicated that an accumulated totalsolar radiation of 800 MJ/m2, reached in approximately 7 weeks at the exposure site, is required for PBATmulch film integrity loss.

� 2009 Elsevier Ltd. All rights reserved.

1. Introduction

Due to economic and environmental issues such as wastemanagement and carbon emission, the use of biodegradable poly-ester from renewable or fossil sources, or a combination of both, hasgained research and industry attention in recent years. One of themost difficult wastes to manage (recover or dispose of) in the agri-cultural industry is mulch film because it is contaminated withvarious substances, including soil, debris, water, and chemicals suchas pesticides, insecticides, and herbicides. Biodegradable mulchfilms have been introduced to mitigate this issue since they canbiodegrade in the field after plowing, thus eliminating film recoveryand disposal [1,2].

While biodegradable mulch films are in use over the soil beds infields, they must endure several types of atmospheric degradation,particularly photodegradation from ultraviolet (UV) exposure [1].Photodegradation can affect these films in two ways. First, it cancause random main chain scission, either via Norrish I or Norrish IImechanisms. This main chain scission poses a threat to mulch film

All rights reserved.

application since it can cause a reduction of mechanical properties,such as tensile strength, and a decrease in film integrity, which isnecessary to provide soil protection and prevent weed growth. Thechromophoric carbonyl groups present in biodegradable polyestersin high amounts make these films susceptible to photodegradation.Since the mulch films are mainly used outdoors, the oxidation(photooxidation) process always follows after the free radicalsare formed in an auto-oxidation process [1,3]. Second, photo-degradation can cause crosslinking within the film, due to therecombination of generated free radicals from Norrish I [4e6]. Thecrosslinking reaction of the polymer structure causes thefilm to loseits ductility; the film becomes more brittle and fails in brittle modeinstead of ductile mode, which is undesirable due to being unpre-dictable. Mulch films should possess ductility in order to allow thefilm to stretch whenwinds blow and lift it up from the soil surface.

The biodegradation process begins once the biodegradablepolyester is in increased contact with microorganism-rich envi-ronments, such as after plowing (soil burial) or in a compost pile.Microorganisms such as bacteria, fungi, and algae degrade the filmby using it as their food source [7e10].

The biodegradation process can be affected by photo-degradation. In the case of biodegradable mulch film designed tobiodegrade by hydrolysis and/or microbial breakdown of the

T. Kijchavengkul et al. / Polymer Degradation and Stability 95 (2010) 99e107100

polymer chain as a food source, solar exposure during the seasoncan affect biodegradation in twoways. First, themain chain scissionfrom photodegradation reduces the number average molecularweight, which provides greater accessibility to the polymer chainby moisture and microorganisms [6,10e12]. These smaller plasticmolecules can be more easily hydrolyzed or utilized by microbes.Second, in the case of aliphaticearomatic polyesters, photo-degradation can result in both main chain scission and crosslinking[13,14]. In biodegradable mulch films, the crosslinked structureslimit the segmental mobility of the plastic molecules as well as theaccessibility of the water and microbes to the polymer chain.Consequently, under such conditions the biodegradation process isinhibited after the film is plowed into the soil [6].

The objective of this study was to monitor the degradation ofan aliphaticearomatic biodegradable polyester in two environ-ments: 1) upon exposure to solar radiationwhen the film is used asa mulch film, and 2) under soil burial conditions to evaluate thedegradation of the film under the soil.

2. Materials and methods

2.1. Experimental details

Polybutylene adipate-co-terephthlate (PBAT) with added carbonblack (Fig. 1) was used as biodegradable mulch film for organicpineapple production on raised beds. PBAT, an aliphaticearomaticpolyester, was selected for this experiment since it is biodegradableand yet its mechanical properties (i.e., tensile strength, elongation,and flexibility) are similar to polyethylene mulch films [2]. Otherbiodegradable polyesters, such as poly(lactic acid), or poly(hydroxybutyrate) and its copolymers, do not offer the requiredmechanical properties for mulch film application at a reasonableprice. In addition, the benzene ring in the aromatic section of PBATabsorbs UV energy and acts as an energy sink to dissipate the UVenergy via electron delocalization [14]. The PBAT film used in thisexperiment was 31.75 � 10�6 m thick (1.25 mil) and was providedby Northern Technologies International (Circle Pines, Minn., USA).Conventional linear low density polyethylene (LDPE) film (AGRI-NOVA, Mexico) with a thickness of 31.75 � 10�6 m was used asa control. The experiment was carried out at the EARTH Universityin Guácimo, Limón, Costa Rica (10� 120 N, 83� 360 W) fromApril 2008to January 2009 for 40 weeks.

The field layout was a complete randomized design (CRD) with20 subplots. Each subplot was 6 � 6 m2 except the film samplingplots, which were 12 � 6 m2. Outer beds or crops of the five beds ineach subplot were guard rows, which were in place to eliminateedge effects and provide consistent data for the sampling area.Weather data were obtained from the EARTH University weatherstation while soil temperature and moisture were monitored with

Fig. 1. Structure of poly(butylene adipate-co-terephthalate) (PBAT) and 1,4 butanediol(B), terephthalic acid (T), and adipic acid (A) components.

a HOBO� Weather Station Data Logger model H21-001 (OnsetComputer, Bourne, Mass., USA) at 4 and 10 cm depths.

At predetermined times during crop production (0, 1, 2, 4, 8, 12,16, 20, 24, 28, 32, and 40 weeks), film pieces (0.2� 0.3 m2) were cutfrom themulch film on the sampling plots, cleaned, vacuum packed(to delay further oxidation), and air shipped to our laboratory at theSchool of Packaging, Michigan State University (MSU) for filmcharacterization.

To measure the soil degradation of the mulch film, four pieces of0.2 � 0.2 m2 PBAT film were placed in a 0.5 � 0.5 m2 frame, aspreviously described by Kale et al. [15]. A total of 12 frames wereburied 0.3 m deep under the soil in the raised beds. At pre-determined times (0, 1, 2, 4, 7, 12, 16, 20, 24, 28, 32, and 39 weeks),one framewas retrievedand thefilmswere cleaned, vacuumpacked,and air shipped to our laboratory at MSU for film characterization.

All film samples were characterized in terms of visual evalua-tion, mechanical properties, FTIR spectra, molecular weight, andthermal properties.

2.2. Visual evaluation

Photographs of cleaned, solar exposed and soil burial filmsamples were taken at each of the predetermined sampling timesusing a 7.1 megapixels digital camera (Canon PowerShot A710IS,Canon USA, Lake Success, N.Y., USA). The following settings wereused: fluorescent lighting and white balance, ISO speed of 400,aperture of f4.5, and shutter speed of 1/15 s. The photographs werearranged in chronological order to visually evaluate the filmdegradation over time.

2.3. Scanning electron microscopy

A scanning electron microscope (SEM) (model JSM-6400V, JeolUSA, Peabody, Mass., USA) was used to examine the surface dete-rioration of film samples after each period of exposure to solarradiation or soil burial. An accelerated voltage of 10 kV and vacuumpressure of approximately 1e10 � 10�6 Pa were used as the oper-ating conditions. The air-dried samples along with the standardsample holders were sputter coated with gold using a current of20 mA for 3 min.

2.4. Mechanical properties

A universal testing machine from Instron, Inc. (Norwood, Mass.,USA) was used to test tensile strength, percent (%) elongation, andtensile modulus on five samples for each production plot mulchfilm in the machine direction (MD) and cross direction (CD) inaccordance with the standard ASTM D882 method [16]. Dependingon the samples, an initial grip length set at 0.05 or 0.125 m anda grip separation rate of 0.0125 or 0.5 m/min were used. Unlikemolecular weight reduction or CO2 evolution, a change inmechanical properties can only be evidence of degradation andcannot be used as a conclusive quantifying tool for polymerbiodegradation. Therefore, mechanical property tests were notperformed on the soil burial film samples.

2.5. Fourier transform infrared (FTIR) spectroscopy

For each film, five samples were scanned from 4000 to 650 cm�1

using a Shimadzu IR-Prestige 21 (Columbia, Md., USA) equippedwith an Attenuated Total Reflectance (ATR) attachment (PIKETechnologies, Madison, Wis., USA). Changes in the spectra inten-sities were correlated with the formation and destruction of func-tional groups in the films.

a

b

c

d

Fig. 2. (a) % Relative humidity (RH), (b) ambient temperature, (c) soil temperature, and(d) average total solar radiation at the experimental site fromApril 2008 to January 2009.

T. Kijchavengkul et al. / Polymer Degradation and Stability 95 (2010) 99e107 101

2.6. Molecular weight

Molecular weight was determined by dissolving PBAT samples(20 mg) in 10 ml of tetrahydrofuran (THF) and then injecting 100 mlof each sample solution into a gel permeation chromatograph(GPC). The GPC was equipped with a Waters 1515 isocratic pump,a Waters 717 autosampler, a series of 3 columns (HR4, HR3, andHR2), and a Waters 2414 refractive index detector interface withWaters Breeze software (Waters Inc., Milford, Mass., USA). A flowrate of 1 ml/min, a runtime of 45 min, and a temperature of 35 �Cwere used.

2.7. Gel content

The gel content of the film samples was measured according tothe standard ASTM D2765 method A using THF as the solvent [17].Gel content was calculated using equation (1)e(3).

%Extract ¼ Ws �Wdf ,Ws

� 100 ¼ Ws �Wdð 1� FÞ,Ws

� 100 (1)

f ¼ 1� F ¼ Total Sample Weight� Filler WeightTotal Sample Weight

(2)

%Gel content ¼ 100� %Extract (3)

where Ws is the weight of the specimen being tested, Wd is theweight of dried gel, f is the polymer fraction (the ratio of the weightof the polymer in the formulation to the total weight of theformulation), and F is the fraction of filler.

2.8. Thermal properties

Glass transition (Tg) and melting (Tm) temperatures of thebiodegradable films were measured with a differential scanningcalorimeter (DSC) from Thermal Analysis Inc., (Model Q 100; NewCastle, Del., USA). The testing temperature was from �60 to 160 �Cwith a ramping rate of 10 �C/min, according to ASTM D3418 [18].Film samples (5e10 mg) were analyzed in triplicate at each testingtime.

A DSC Avrami equation analysis using melting peak onset timedifferences between non-crosslinked and crosslinked samples wascarried out to verify the gel content of the solar exposed samples.Since crosslinking of films causes non-uniformity of the crystallattices, making them melt at different temperatures, the onsettemperature of the melting peak was used as input data for theAvrami equation to verify the gel content as previously reported byIoan et al. [19] and Kijchavengkul et al. [20].

3. Results and discussion

Weather data obtained for the experimental site from April2008 to January 2009 are shown in Fig. 2. Despite variations inambient temperature between 19 and 35 �C during this period, thePBAT mulch film maintained the average soil temperature atapproximately 30 �C during the first 4 months, and prevented theminimum soil temperature from dropping below 24 �C. After thatinitial period the film started to lose integrity, so soil temperaturewas no longer measured. The total solar radiation of the experi-mental site from April 2008 to January 2009 was 4053 MJ/m2.

Based on visual examination, no major tears or cracks werefound on the solar exposed samples until week 8, when thesamples began to lose integrity and developed some cracks. Byweek 16 and later, the film samples were reduced to small friablepieces (Fig. 3a). Thus, the useful lifetime of PBAT mulch film is

approximately 8 weeks with solar exposure. The soil burial samplesremained unchanged until week 24, when visual examinationrevealed some development of holes and surface erosion (Fig. 3b).As degradation proceeded fromweek 32e39 at a faster rate, the soilburial samples started to lose integrity: holes and major cracks

Fig. 3. (a) Degradation of solar exposed PBAT film as a function of time (original sample size, 0.2 � 0.3 m2); and (b) degradation of soil burial PBAT film as a function of time (originalsample size, 0.2 � 0.2 m2).

T. Kijchavengkul et al. / Polymer Degradation and Stability 95 (2010) 99e107102

were found throughout the film, and by week 39 one side of thesample had disappeared (Fig. 3b).

3.1. Effects of solar exposure

During the 40 week study, the tensile strength and percentelongation of the control LDPE film remained stable at approxi-mately 3.2 ksi (22.1 MPa) and 430% and 640% elongation in the MDand CD, respectively (data not shown), and as we also previouslyreported [2]. On the other hand, the tensile strength and percentelongation of PBAT film declined significantly upon solar exposure,from 2.6 ksi (17.9 MPa) and 492% at week 0e0.72 ksi (4.96 MPa)and 29% at week 4 for the film in the MD, and from 2.6 ksi(17.9 MPa) and 626% at week 0e0.73 ksi (5.03 MPa) and 24% at

week 4 for the film in the CD (Fig. 4). After the initial reduction ofmechanical properties during weeks 2e4, both tensile strength andelongation values did not change significantly until week 12, whensamples large enough for tensile testing could still be obtained.After week 12, film samples developed longitudinal cracks in themachine direction, which made it difficult to measure mechanicalproperties in the cross direction.

The significant decreases of tensile strength and percent elon-gation of the solar exposed PBAT film coincided with the amount ofgel content generated during weeks 2e4. Fig. 5 shows that the gelcontent increased rapidly from 5.7% at week 2 to 22.3% at week 4.As a result of crosslinking, the films became more brittle, as wasindicated by the rapid decrease in percent elongation to less than30% at week 4. Crosslinking in PBAT films results from the

a

b

Fig. 4. (a) Tensile strength and (b) percent elongation of solar exposed PBAT film asa function of time and accumulated total solar radiation for machine direction (MD)and cross direction (CD). The accumulated total solar radiation was fitted as a functionof 102.08 times the exposure time with an R2-value of 0.9949.

Time (week)

0 5 10 15 20 25 30 35 40 45

Gel

con

tent

(%)

0

5

10

15

20

25

30

35

Fig. 5. Gel content of solar exposed PBAT film as a function of time; filled circles (C)are means � standard deviation, solid line (d) is the predicted gel content, and dashedlines (eee) are the 95% predicted confidence intervals from the DSC Avrami equation�lnð 1� XgÞ ¼ 0:5717Dt0:7438 with R2-value of 0.9984.

T. Kijchavengkul et al. / Polymer Degradation and Stability 95 (2010) 99e107 103

recombination of the generated free radicals from the Norrish Ireaction and hydrogen abstraction, as previously described byvarious authors [4,5,21e27]. In our study the gel contents of thefilms plateaued out after 8 weeks. Amajor concernwith crosslinkedgels is that they limit the flexibility of the polymer chains [6], whichprevents the penetration of water and the use of the polymer asa microbial food source, thereby delaying the biodegradationprocess of mulch films plowed into the soil. After 4 weeks of solarexposure, about 25% of the film by weight was crosslinked anddeveloped some thermosetting behaviors, such as increased tensilemodulus and brittle mode of failure under stress.

The FTIR absorbance spectra of the solar exposed side of PBATfilm samples indicated evidence of photodegradation and photo-oxidation reactions. In Fig. 6b a reduction of the carbonyl peak(C]O) and a broadening of the peak were observed. The leftshoulder (1790e1750 cm�1) indicates the formation of free C]O,and the right shoulder represents the formation of a lower molec-ular weight ester [28]. These observations indicate a Norrish I chain

scission reaction as suggested in Scheme 1. In Fig. 6c, increasedabsorbance at 1000 cm�1 from the out-of-plane bending of C]Cindicates the formation of terminal double bond compounds, sug-gesting a Norrish II chain scission. Furthermore, in Fig. 6a theformation of free OH at 3620 cm�1, hydrogen-bonded OH at3530 cm�1, free OOH (or peroxide) at 3440 cm�1, and the broadpeak at 3370 cm�1 of the polymer OH first found at week 2 allindicate a photooxidative reaction [28], which is from the auto-catalytic reaction or the carboxylic terminal groups from the Nor-rish II reaction (Scheme 1). These chain scissions would account forthe observed reductions in mechanical properties, especially theloss of tensile strength. As time proceeded, the film samples becametoo dirty andwrinkled to provide good contact with the ATR crystal,and the resulting spectra were unsuitable for comparison.

In FTIR spectra of the unexposed side of the PBAT film sampleexposed to solar radiation (i.e., opposite side), peaks of free OH at3620 cm�1, hydrogen-bonded OH at 3530 cm�1, free OOH at3450 cm�1, and the polymer OH peak at 3380 cm�1 were alsoobserved starting at week 2 (Fig. 6d). These results suggest thatphotooxidation occurred throughout the sample matrix due to thepropagation of photodegradation, which is a surface phenomenon[14,29]. Unlike for the exposed side of the film (Fig. 6b), carbonylpeaks at 1710 and 1730 cm�1 did not become broader (Fig. 6e). Anincreased peak at 1000 cm�1 for C]C was also observed (Fig. 6f),but this change of absorbance for the unexposed side of the filmwas not as intense as that of the exposed side. Other notablechanges in the unexposed side were the reduced absorbance peaksat 2920 and 2850 cm�1 from asymmetric and symmetric CeHstretching of the methylene (CH2) group in the butanediol section(Fig. 6d).

Chain scission from the photodegradation and photooxidationreactions caused a reduction in molecular weight of the PBATsamples (Fig. 7). A rapid reduction of Mn as well as a moderatelyreduced Mw were observed at week 4. As a result, the poly-dispersity index (PI) spiked up to 6.9 at week 4, which corre-sponded to the time at which the reduction in mechanicalproperties was detected. The PI values started to decline slowlyfrom 4.1 at week 12e2.1 at week 40; at the same time theMn valuesslowly decreased almost to a plateau stage at 5.8 kDa, while Mwdeclined at greater rate and also showed a tendency to reacha plateau stage.

Abso

rban

ce

0.00

0.05

0.10

0.15

0.20Week 0 Week 2 Week 4 Week 8

0.0

0.2

0.4

0.6

0.8

0.0

0.2

0.4

0.6

0.8

1.0

Abso

rban

ce

0.00

0.05

0.10

0.15

0.20Week 0 Week 2 Week 4 Week 8

0.0

0.2

0.4

0.6

0.8

0.0

0.2

0.4

0.6

0.8

1.0

Wavenumber (cm-1) Wavenumber (cm-1) Wavenumber (cm-1)

280030003200340036003800

Abso

rban

ce

0.00

0.05

0.10

0.15

0.20Week 0 Week 1 Week 2 Week 4 Week 7

16001700180019000.0

0.2

0.4

0.6

0.8

80090010001100120013000.0

0.2

0.4

0.6

0.8

1.0

a b

e f

c

d

hg i

Fig. 6. FTIR absorbance spectra of PBAT film: Solar exposed side of film from 0, 2, 4, and 8 weeks in the wavenumber range of (a) 3800e2750 cm�1, (b) 1900e1550 cm�1, and(c) 1350e750 cm�1; solar unexposed side of film from 0, 2, 4, and 8 weeks in the wavenumber range of (d) 3800e2750 cm�1, (e) 1900e1550 cm�1, and (f) 1350e750 cm�1; and soilburial film from 0, 1, 2, 4, and 7 weeks in the wavenumber range of (g) 3800e2750 cm�1, (h) 1900e1550 cm�1, and (i) 1350e750 cm�1.

T. Kijchavengkul et al. / Polymer Degradation and Stability 95 (2010) 99e107104

As expected, the SEM micrographs revealed that the exposedside of the PBAT film (Fig. 8a) underwent more surface degradationfrom photodegradation than did the unexposed side (Fig. 8b). Boththe FTIR spectra and SEM micrographs demonstrated that photo-degradation of PBAT started as a surface degradation and thenpropagated throughout the polymer matrix.

According to Rivaton and Gardette [4] and Halim et al. [30], thechromophore groups in aromatic polyesters such as PBAT absorbUV from 300 to 350 nm, with maximum sensitivity at 325 nmcorresponding to the ester group. This spectral component is likelyresponsible for the photodegradation of PBAT. The crosslinking ofPBAT is observed regardless of the presence or absence of oxygen,but it is enhanced by the absence of oxygen [4,24].

The PBAT film used in this study retained some integrity for upto 8weeks. During this period, photodegradationwas the dominantfactor in the deterioration of the films. Marten et al. reported thehydrolysis rate at pH 7 of PBAT containing different isomers to be

less than 0.24 mmol/day at 37 and 50 �C [31]. Hydrolysis of PBATdoes not play a major role in its initial deterioration [32]. Therefore,if PBAT films are exposed to or used in dry weather conditions likein the Southwest U.S., we would not expect any significant differ-ences in their performance than when exposed to similar solarradiation in a moist environment. On the other hand, if the filmintegrity can be prolonged beyond 8 weeks, the film degradationunder simultaneous exposure to UV and moisture would be muchfaster in higher humidity conditions. Hydrolysis of the ester bondcan cause the main chain scission of the PBAT, resulting ina reduction of molecular weight. Photooxidation is then promoteddue to the generation of reactive hydroxyl free radicals, whichinduces more free radicals and enhances further crosslinking andoxidative chain scission [30]. In addition, moisture absorptionwithin the amorphous regions of the aromatic polyester, in thiscase PBAT, causes swelling and produces stress across the amor-phousecrystalline interface resulting in microcavitation [33].

Time (week)

0 5 10 15 20 25 30 35 40 45

Mol

ecul

ar w

eigh

t (kD

a)

0

20

40

60

80

100MnMw

PI

0.02.55.07.5

10.0a

b

Fig. 7. Polydispersity index (PI), number average molecular weight (Mn), and weightaverage molecular weight (Mw) of solar exposed PBAT film as a function of time.

Scheme 1. Norrish I and II reactions for PBAT: (1) ¼ terminal carboxylic compound,and (2) ¼ terminal double bond compound.

T. Kijchavengkul et al. / Polymer Degradation and Stability 95 (2010) 99e107 105

3.2. Effects of soil burial

Molecular weight determination of the PBAT films exposed tosoil burial conditions, where mainly hydrolysis and biodegradationcan occur, showed no evidence of gelation and, therefore, nocrosslinking. As the hydrolytic reaction took place, the formation ofOH, free OH at 3620 cm�1, hydrogen-bonded OH at 3530 cm�1, anda broadened peak at 3370 cm�1 for polymer OH were apparent atweek 1 (Fig. 6g), and these changes were due to random chainscission. Reduced absorbance peaks at 2920 and 2850 cm�1 fromasymmetric and symmetric CeH stretching of the methylene (CH2)group in the butanediol section were also observed (Fig. 6g).Moreover, reduced ester peaks were observed at 1730 and1710 cm�1 for aromatic and aliphatic C]O (Fig. 6h), and at 1270and 1250 cm�1 for aromatic and aliphatic CeO (Fig. 6i). Theformation of OH, as well as reduced CeH and ester peaks suggestthe occurrence of random chain scission from hydrolysis.

Fig. 8. SEM micrographs of (a) solar exposed side and (b) unexposed side of PBAT film at wdeterioration.

A steady reduction in Mn was observed in the soil burial PBATsamples especially after week 12 (Fig. 9). Similarly, a linear reduc-tion in Mw was observed until week 20, followed by an increasedreduction rate (Fig. 9), which corresponded to the results fromthe visual evaluation (Fig. 3b). The PI of the soil burial samplesremained stable at a value of approximately 2 for about 20 weeksand then increased to 3.0 by week 39. The reduction of Mn for thenon-crosslinked portion of the solar exposed samples followeda first order reaction. However, for the soil burial samples thereduction of Mn is from the hydrolysis associated with bulk erosionbiodegradation and also shows the pattern of surface erosionbiodegradation (Fig. 10).

Soil burial conditions resulted in slow degradation of the PBATsamples. Reductions in Mn and Mw were only 68.4 and 48.5%,respectively, after 40 weeks, and a reduction in molecular weight ofthe PBAT samples to the state of oligomers did not occur. Accordingto the fitted equation of the exponential reduction Mn shown inFig. 10, the non-crosslinked PBAT sample may completely degradein 52 weeks under soil burial conditions; more time would berequired if crosslinked structures are present in the buried films.

SEM micrographs of the surface of the soil burial PBAT samples(Fig. 11) show that microbial activity began around week 20 and

eek 4 at 2500� magnification; bars ¼ 20 mm; arrows indicate regions of major surface

Time (week)

0 5 10 15 20 25 30 35 40 45

Num

ber a

vera

ge m

olec

ular

wei

ght (

kDa)

0

10

20

30

40

50

60Solar exposed Soil burial

Fig. 10. Number average molecular weight (Mn) of non-crosslinked portion of solarexposed and soil burial PBAT samples as a function of time. Solid lines (d) are thefirst order reduction Mn ¼ 8.028þ41.16exp(�0.4144t) with R2-value of 0.9417 for solarexposed sample and exponential reduction Mn ¼ 79:63� 35:38expð0:0155tÞ withR2-value of 0.9722 for soil burial sample.

Time (week)

0 5 10 15 20 25 30 35 40 45

Mol

ecul

ar w

eigh

t (kD

a)

0

20

40

60

80

100MnMw

PI

1

2

3

4a

b

Fig. 9. Polydispersity index (PI), number average (Mn), and weight average molecularweights (Mw) of soil burial samples as a function of time.

T. Kijchavengkul et al. / Polymer Degradation and Stability 95 (2010) 99e107106

continued progressively until the end of the experiment. Evidenceof progressive biodegradation includes the increased amountof biofilms and numbers of microorganisms adhering to thosebiofilms; no biofilms were found on samples prior to week 20(Fig. 11a and b).

3.3. Final remarks

Fig. 12 summarizes the results obtained from the PBAT biodeg-radation mulch film trials of pineapple production for 40 weeks in

Fig. 11. SEM micrographs of soil burial PBAT samples at (a) week 8, (b) week

Costa Rica (this study) and tomato production for 15 weeks inMichigan (previously reported by Kijchavengkul et al. [2]). Bothstudies indicate that once the PBAT film with carbon black isexposed to an accumulated total solar radiation of 800 MJ/m2, thegel content of the filmwill reach a plateau stage at around 25%, anda rapid reduction of Mn will parallel a plateau stage around 10 kDa.Awareness of this trigger point (800 MJ/m2) can be helpful inimproving film photostability and designing films for differentapplications, especially for outdoor uses. In the case of mulch films,if a crop requires less than 800 MJ/m2 of solar radiation to get toa mature stage, then enough crop and soil protection can beobtained with the PBAT films.

20, (c) week 32, and (d) week 39; 500� magnification; bars ¼ 100 mm.

a

b

Fig. 12. Reduction of Mn, development of gel content, and accumulated solar radiationof the exposed PBAT film as a function of time from (a) pineapple production in CostaRica for 40 weeks, and (b) tomato production in Michigan, USA, for 15 weeks fromKijchavengkul et al. [2].

T. Kijchavengkul et al. / Polymer Degradation and Stability 95 (2010) 99e107 107

Acknowledgements

The authors would like to thank the College of Agriculture andNatural Resources at MSU for partially funding this project, EARTHUniversity forproviding thefieldsand facilities to run thisproject, andNorthern Technologies International for providing the mulch films.

References

[1] Kyrikou I, Briassoulis D. Biodegradation of agricultural plastic films: a criticalreview. J Polym Environ 2007;15:125e50.

[2] Kijchavengkul T, Auras R, Rubino M, Ngouajio M, Fernandez RT. Assessment ofaliphatic-aromatic copolyester biodegradable mulch films. Part I: field study.Chemosphere 2008;71:942e53.

[3] Hussain I, Hamid H. Plastics in agriculture. In: Andrady AL, editor. Plastics andthe environment. Hoboken, NJ: John Wiley & Sons; 2003. p. 185e209.

[4] Rivaton A. Gardette JL. Photo-oxidation of aromatic polymers. Angew Mak-romol Chem 1998;261/262:173e88.

[5] Buxbaum LH. The degradation of poly(ethylene terephthalate). Angew ChemInt Edit 1968;7:182e90.

[6] Kijchavengkul T, Auras R, Rubino M, Ngouajio M, Fernandez RT. Assessment ofaliphatic-aromatic copolyester biodegradable mulch films. Part II: laboratorysimulated conditions. Chemosphere 2008;71:1607e16.

[7] ASTM D6400-04 standard specification for compostable plastics. West Con-shohocken, PA: ASTM International; 2004.

[8] Chandra R, Rustgi R. Biodegradable polymers. Prog Polym Sci 1998;23:1273e335.

[9] Kale G, Kijchavengkul T, Auras R, Rubino M, Selke SE, Singh SP. Compostabilityof bioplastic packaging materials: an overview. Macromol Biosci 2007;7:255e77.

[10] Gopferich A. Mechanisms of polymer degradation and elimination. In:Domb AJ, Kost J, Wiseman D, editors. Handbook of biodegradable polymers.Boca Raton, FL: CRC; 1998. p. 451e71.

[11] Albertsson AC. Biodegradation of polymers. In: Hamid SH, Amin MB,Maadhah AG, editors. Handbook of polymer degradation. New York, NY:Marcel Dekker; 1992. p. 345e63.

[12] Stevens ES. What makes green plastics green? Biocycle 2003;24:24e7.[13] Osawa Z. Photoinduced degradation of polymers. In: Hamid SH, Amin MB,

Maadhah AG, editors. Handbook of polymer degradation. New York, NY:Marcel Dekker; 1992. p. 169e217.

[14] Schnabel W. Polymer degradation: principles and practical applications. NewYork, NY: Hanser; 1992.

[15] Kale G, Auras R, Singh SP, Narayan R. Biodegradability of polylactide bottles inreal and simulated composting conditions. Polym Test 2007;26:1049e61.

[16] ASTM D882-98 Standard test method for tensile properties of thin plasticsheeting. West Conshohocken, PA: ASTM International; 1998.

[17] ASTM D2765-01 Standard test methods for determination of gel content andswell ratio of crosslinked ethylene plastics. West Conshohocken, PA: ASTMInternational; 2006.

[18] ASTM D3418-03 Standard test method for transition temperatures andenthalpies of fusion and crystallization of polymers by differential scanningcalorimetry. West Conshohocken, PA: ASTM International; 2003.

[19] Ioan S, Grigorescu G, Stanciu A. Dynamic-mechanical and differential scanningcalorimetry measurements on crosslinked poly(ester-siloxane)-urethanes.Polymer 2001;42:3633e9.

[20] Kijchavengkul T, Auras R, Rubino M. Measuring gel content of aromaticpolyesters using ftir spectrophotometry and dsc. Polym Test 2008;27:55e60.

[21] Grossetete T, Rivaton A, Gardette JL, Hoyle CE, Ziemer M, Fagerburg DR, et al.Photochemical degradation of poly(ethylene terephthalate)-modified copol-ymer. Polymer 2000;41:3541e54.

[22] Fechine GJM, Rabello MS, Souto Maior RM, Catalani LH. Surface character-ization of photodegraded poly(ethylene terephthalate). The effect of ultravi-olet absorbers. Polymer 2004;45:2303e8.

[23] Marcotte FB, Campbell D, Cleaveland JA. Photolysis of poly(ethylene tere-phthalate). J Polym Sci Pol Chem 1967;5:481e501.

[24] Tabankia M, Gardette JL. Photo-chemical degradation of polybutylenetere-phthalate: part 1-photo-oxidation and photolysis at long wavelengths. PolymDegrad Stabil 1986;14:351e65.

[25] Tabankia M, Gardette JL. Photo-oxidation of block copoly(ether-ester) ther-moplastic elastomers: part 2-origins of the photo-yellowing. Polym DegradStabil 1987;19:113e23.

[26] Tabankia M, Philippart JL, Gardette JL. Photo-oxidation of block copoly(ether-ester) thermoplastic elastomers. Polym Degrad Stabil 1985;12:349e62.

[27] Valk VG, Kehren ML, Daamen I. Photooxidation of poly(ethyleneglycol tere-phthalate). Angew Makromol Chem 1970;13:97e107.

[28] Pretsch E, Buhlmann P, Affolter C. Structure determination of organiccompounds: tables of spectral data. New York, NY: Springler-Verlag BerlinHeidelberg; 2000.

[29] Shlyapintokh V. Photochemical conversion and stabilization of polymers. NewYork, NY: Hanser; 1984.

[30] Hamid SH, Maadmah AG, Amin MB. Weathering degradation of polyethylene.In: Hamid SH, Amin MB, Maadhah AG, editors. Handbook of polymer degra-dation. New York, NY: Marcel Dekker; 1992. p. 219e60.

[31] Marten E, Muller RJ, Deckwer WD. Studies on the enzymatic hydrolysis ofpolyesters. II. aliphatic-aromatic copolyesters. Polym Degrad Stabil 2005;88:371e81.

[32] Andrady AL, Hamid SH, Torikai A. Effects of climate change and UV-B onmaterials. Photochem Photobiol Sci 2003;2:68e72.

[33] Kamal MR, Huang B. Natural and artifical weathering of polymers. In:Hamid SH, Amin MB, Maadhah AG, editors. Handbook of polymer degradation.New York, NY: Marcel Dekker; 1992. p. 127e68.