Available online at www.sciencedirect.com

www.elsevier.com/locate/chemosphere

Chemosphere 71 (2008) 942–953

Assessment of aliphatic–aromatic copolyester biodegradablemulch films. Part I: Field study

Thitisilp Kijchavengkul a, Rafael Auras a,*, Maria Rubino a,Mathieu Ngouajio b, R. Thomas Fernandez b

a School of Packaging, Packaging Building, Michigan State University, East Lansing, MI 48824-1223, United Statesb Department of Horticulture, Plant and Soil Science Building, Michigan State University, East Lansing, MI 48824-1223, United States

Received 5 June 2007; received in revised form 23 October 2007; accepted 23 October 2007Available online 8 February 2008

Abstract

The objective of this work was to study the use of new biodegradable films in agriculture under open field conditions. Three biode-gradable mulch films made from modified biodegradable polyester of different thicknesses and colors (black and white) and a conven-tional low density polyethylene (LDPE) mulch film were used to cover the beds of tomato plants. Changes in physical appearance of thefilms were recorded as well as changes in their mechanical, optical, and physical properties. Once tomato harvest was completed, theconventional LDPE mulch film was removed and all the tomato plants were cut using a mower. The biodegradable mulch films wereplowed into the soil. The change in the appearance of the film was recorded and samples of each film after plowing were characterizedaccording to the properties mentioned above.

After the biodegradable films photodegraded, cross-link formation occurred within the films which promoted brittleness. Titaniumdioxide, an additive used to produce white color in the films, catalyzed the photodegradation, while carbon black used for black colorstabilized the photodegradation. The white films started to degrade after two weeks while it took about eight weeks for the black films tosignificantly degrade. The black biodegradable film seems to be a more promising alternative as a mulch film because of the comparableyields and weed suppression ability to conventional mulch film.� 2008 Elsevier Ltd. All rights reserved.

Keywords: Plasticulture; Aliphatic–aromatic copolyester; Mulch films; Biopolymers; Compostability; Degradation

1. Introduction

The term ‘Plasticulture’ is defined by the American Soci-ety of Plasticulture as the use of plastics in agriculture forboth plant and animal production including; plastic mulch,drip irrigation, row covers, low tunnels, high tunnels,greenhouses, silage bags, hay bales and in food packagingand nursery pots and containers for growing transplants(Lamont and Orzolek, 2004). Mulching by putting a thinplastic film directly over the soil surface with drip irrigationhas become a standard technique for vegetable growers inthe US and around the world. In 2003, more than 1.2 mil-

0045-6535/$ - see front matter � 2008 Elsevier Ltd. All rights reserved.

doi:10.1016/j.chemosphere.2007.10.074

* Corresponding author. Tel.: +1 517 432 3254; fax: +1 517 353 8999.E-mail address: aurasraf@msu.edu (R. Auras).

lion ha in the US were covered with plastic mulch (Miles,2003). The estimate of world mulch film application is700000 tons per year (Espi et al., 2006), and the US marketin 2001 for mulch films is 150000 tons (Shogren, 2001).

The advantages and purposes of using mulch films are to(1) control or increase the soil temperature, (2) maintainsoil humidity, (3) maintain raised-bed soil structure, (4)reduce germination time, (5) provide early or out of seasoncrop production for better market value produce, (6)reduce weeds and plant diseases, (7) provide efficiency inthe usage of water and fertilizers, and most importantly,and (8) increase yields and improve produce quality(Lamont, 1999; Jensen, 2004; Espi et al., 2006). Despiteall these advantages, there are several major concernsabout using mulch films which are the costs of removal

T. Kijchavengkul et al. / Chemosphere 71 (2008) 942–953 943

and disposal of the used plastics and environmental issues.Most of the used mulch films are not typically recycled,since they are often soiled and wet, which make them diffi-cult to recycle; therefore, mostly all of them end up in thelandfill (Lamont, 1999). To solve these problems, the use ofbiodegradable mulch films seems to be a promising solu-tion because the films can degrade right in the field; there-fore, the costs of removal and disposal, approximately $147million annually in 2003, will disappear, and the amount ofwaste ending up in landfills can be avoided.

There are two main degradation mechanisms which maybe considered for biodegradable mulch films: (1) photodeg-radation and (2) biodegradation. Photodegradation is thedegradation process due to the exposure of films to certainamount of solar radiation. As a consequence of this expo-sure, the films become more brittle, with cracks, tears,holes, occurring before finally disintegrating into smallflakes. Biodegradation, however, is a degradation processresulting from the action of naturally occurring microor-ganisms such as bacteria, fungi, and algae (ASTM, 2004).Appropriate degradation mechanisms must be triggeredfor biodegradable mulch films to work.

In the past, the problems of using biodegradable or pho-todegradable mulch films were either premature break-down (the films disintegrate before the harvesting), orpostmature breakdown (the films degrade slower than theexpected rate) (Lamont, 1999; Jensen, 2004). The normalage of the mulch film should be designed according the veg-

Fig. 1. Biodegradable mulch films cycle, starting from raising the bed and appfall, and associated degradation processes.

etables to growth, 2–6 months depending on the crops,starting from the laying of the film to the harvesting ofthe produce. However, there are many factors affectingthe photodegradation and biodegradation rate of films thatshould be considered during the design and production inorder to control the right degradation rates of the biode-gradable mulch films. Fig. 1 summarizes the processesand properties required and needed in designing biode-gradable mulch films. Mulch films cycle starts when theherbicide is applied to the field and the beds are raised toform the plots in spring. Then the mulch films are laid tocover the soil beds together with drip irrigation tubes. Atthe same time or one week after, the crops are planted intothe mulch cover beds. During the growing season of two tosix months, fertilizer, pesticide, and stake and twine areapplied to the beds and plants. After the harvest in fall,there are three options to dispose the used biodegradablemulch films: (1) remove and dispose them in landfill; (2)remove and compost the films in a compost pile; and (3)plow the tomato plants and the biodegradable mulch filmsinto the field. If the third option is used, the films must begone before the next season.

According to Fig. 1, in the photodegradation period,factors affecting the photodegradation rate are categorizedas: (1) formulation and quality of the film, (2) season, (3)geographical region, (4) amount of solar radiation andday length, (5) cloud cover, and (6) sun angle (Giacomellet al., 2000).

lying herbicide in spring to harvesting and the disposal of the films in late

944 T. Kijchavengkul et al. / Chemosphere 71 (2008) 942–953

In the biodegradation period, factors affecting biodegra-dation rate are categorized as: (1) the exposure conditions,including moisture, pH, temperature, and aerobic or anaer-obic conditions of the soil and (2) the polymer characteristic,such as polymer structure and chain flexibility, crystallinity,molecular weight, copolymer composition, and thickness,size, and shape of the exposed film (Kale et al., 2007).

The goal of this project was to study the performance ofnew biodegradable mulch films in agricultural production.

Fig. 2. Characteristics of air and soil of the experimental station from May 200exposure, (d) soil temperature, and (e) soil moisture content.

Scheme 1. Structure of poly(butylene adipate-co-terephthalate) (PBAT)or Ecoflex and places where main chain scission could occur.

In order to achieve this goal this paper focuses on the fol-lowing specific objectives: (1) to determine the mainrequired properties that lead to successful mulching, (2)to determine and develop the right biodegradable mulchfilms for agricultural applications, (3) to study any struc-tural changes occurring in the structure of the biodegrad-able films due to degradations, and (4) to determine theduration the film keeps its integrity after laying in the field.

Part I of this paper series focuses on the film character-ization of samples obtained from the field experiment.

2. Materials and methods

Three biodegradable mulch polyester films, modified Eco-flex� or poly(butylene adipate-co-terephthalate) (PBAT),

6 to April 2007: (a) air temperature, (b) relative humidity, (c) solar radiant

T. Kijchavengkul et al. / Chemosphere 71 (2008) 942–953 945

see Scheme 1, were provided from Northern TechnologiesInternational (Circle Pines, MN). PBAT are truly biodegrad-able in the composting environments without adverse eco-toxicological effect as reported by Muller et al. (2001) andWitt et al. (2001). The thickness of the biodegradable filmswas 25 lm for white film (25w), 35 lm for white film (35w)and 35 lm for black film (B). Additionally, a conventionallow density polyethylene (LDPE) mulch film with 25 lm(M) was also included in the test. These films were used tocover the beds of tomato plots in the state of Michigan(42�440N, 84�290W) during the months of May through Sep-tember, 2006. Air temperature, relative humidity, soil mois-ture content, soil temperature, and solar radiation during theexperiment are shown in Fig. 2a–e. Spectral irradiance pro-files of UV light from June, 2006 to April, 2007 in Michiganare shown in Fig. 3, and the total radiant exposure the filmsexperienced was 2178 MJ m�2 during the photodegradationperiod as shown in Fig. 1 starting from laying the mulch filmsin June to harvesting in September 2006. A plot layout wasdesigned based on a Latin square with four treatments (threebiodegradable films and one control LDPE film) was usedwith an additional guard row on each side of the plot. Nodata were collected from guard rows since they were designedto protect the experimental plots from pests and edge effects.Every week the changes in the physical appearance of thefilms and tomato plants were visually observed and recordedusing a digital SLR (single-lens-reflex) camera, Canon RebelDigital with six megapixels (Canon, New York City, NY).Every two weeks, film samples were taken from no-cropplots, and their mechanical, optical, physical, and thermalproperties were characterized, and the changes in properties

Fig. 3. Spectral irradiance profiles of UV light from August 2006 to April2007 (data obtained from The US Department of Agriculture (USDA)UV-B Monitoring and Research Program weather station located inPellston, MI).

were plotted against time. Furthermore, the biodegradationtest of all the mulch film samples in laboratory conditionswas conducted based on ASTM D 5338 standard (ASTM,2003b) using a direct measurement respirometric (DMR)system.

2.1. Film characterization

2.1.1. Optical properties

2.1.1.1. Color. Color changes of each film were measuredusing a HunterLab colorimeter (Reston, VA) in L a b scale.‘L’ referred to brightness and darkness of the sample, posi-tive ‘a’ referred to red color, while negative ‘a’ referred togreen, and positive ‘b’ referred to yellow color, while nega-tive ‘b’ referred to blue. Five samples were tested for repe-tition. The color change (DE value) was calculated usingEq. (1), and using the L, a, and b values from day zeroas reference.

DE ¼ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiðDLÞ2 þ ðDaÞ2 þ ðDbÞ2

qð1Þ

2.1.1.2. Light transmission. Percent light transmission wasmeasured by using a Lambda 25 UV/Visible spectropho-tometer from PerkinElmer (Wellesley, MA). For each film,five samples were scanned from 400 to 700 nm wavelength,which is the Photosynthetically Active Radiation range(Ngouajio and Ernest, 2004).

2.1.2. Physical properties

2.1.2.1. Molecular weight. For each film, three samples weredissolved in tetrahydrofuran (THF) at a concentration of20 mg of film per 10 cm3 of THF. One hundred microlitersof each sample solution was injected into a gel permeationchromatography system equipped with a Waters 1515 iso-cratic pump, a Waters 717 autosampler, a series of threecolumns (HR2, HR3, and HR4), and a Waters 2414 refrac-tive index detector interface with a Waters Breeze softwarefrom Waters Inc. (Milford, MA), using a flow rate of1 cm3 min�1, a runtime of 45 min, and a temperature of35 �C.

2.1.2.2. Gel content. The gel content of the film samples wasmeasured according to the standard ASTM D 2765 methodA (ASTM, 2006) using THF as the solvent. The gel contentwas calculated using the following equations:

% Extract ¼ W s � W d

fW s

� 100 ¼ W s � W d

ð1� F ÞW s

� 100 ð2Þ

f ¼ 1� F ð3Þ% Gel content ¼ 100�% Extract ð4Þ

where Ws is the initial weight of the specimen being tested,Wd is the weight of dried gel, f is polymer factor (the ratioof the weight of the polymer in the formulation to the totalweight of the formulation), and F is the Fraction of fillerwhich is determined by using method ASTM D 1603.

946 T. Kijchavengkul et al. / Chemosphere 71 (2008) 942–953

2.1.2.3. Fourier transform infrared spectrophotometry

(FTIR). For each film, five samples were scanned using aShimadzu IR-Prestige 21 (Columbia, MD) with an Attenu-ated Total Reflectance attachment (PIKE Technologies,Madison, WI) from 4000 to 650 cm�1 to measure anychanges in the spectra intensities which correlated to theformation and destruction of functional groups in thefilms.

2.1.3. Thermal properties

Glass transition (Tg) and melting (Tm) temperatures ofthe biodegradable films were measured using a DifferentialScanning Calorimeter (DSC) model Q 100 from ThermalAnalysis Inc (New Castle, DE). The testing temperaturewas from �60 �C to 160 �C with a ramping rate of10 �C min�1, according to ASTM D 3418 (ASTM,2003a). Three samples were used for each film.

2.1.4. Mechanical properties

A universal tester machine from Instron, Inc. (Nor-wood, MA) was used to test tensile strength, percentageof elongation, and tensile modulus on five samples for each

Fig. 4. Pictures of the white biodegradable films during a

Fig. 5. Pictures of the black biodegradable films during a

film in machine direction (MD) using ASTM D 882 stan-dard (ASTM, 1998). An initial grip length was set at 5 or12.5 cm, and a grip separation rate was 1.25 or 50 cmmin�1 depending on the samples. The grip length of 12.5cm and separation rate of 1.25 cm min�1 were used fordetermining the tensile strength, elongation, moduli ofthe biodegradable films. The grip length of 5 cm and sepa-ration rate of 50 cm min�1 were used for measuring tensilestrength and elongation of the conventional LDPE mulch.

2.1.5. Biodegradability

A DMR consisted of a total of 18 bioreactors, three forblank (compost only), three for positive control (cornstarch, denoted as CS), three for negative control (conven-tional mulch film), three for each biodegradable film (25w,35w, and B) was used to determine the biodegradation ofthe films according to ASTM D 5338 (ASTM, 2003b).Each reactor was filled with approximately 200 g of matureyard compost, 50 g of vermiculite, and 6 g of sample orcontrol materials. All the bioreactors were incubated indark at 58 �C and 50–60% relative humidity, with the airsupply flow rate of 40 cm3 min�1. Amount of cumulative

nd after the growing season (Bar length equals 5 cm).

nd after the growing season (bar length equals 5 cm).

Fig. 6. Color changes of the film sample: (a) L value (full scale 0–100), (b)b value (full scale –100–100), and (c) DE value.

Fig. 7. Changes in % light transmission of all the film samples.

T. Kijchavengkul et al. / Chemosphere 71 (2008) 942–953 947

CO2 evolution of all the samples was measured and plottedagainst time. Detail of the apparatus and testing conditionscan be found in Kijchavengkul et al. (2006).

2.2. After season film characterization

Once tomato harvest was completed, the conventionalLDPE mulch films and the drip irrigation tape wereremoved and all the tomato plants were cut. Then, thethree biodegradable mulch films were plowed into the soil.The changes in the films were visually evaluated and thefilm samples after plowing were characterized accordingto the properties mentioned earlier.

3. Results and discussion

It was visually observed that both of the white filmsdegraded faster than the black film. The white films startedbreaking down and forming cracks within the first twoweeks of testing (Fig. 4) while the black biodegradable filmstarted breaking down slowly at around the 8th week(Fig. 5). The conventional mulch film did not show anyvisual degradation during the test period.

Samples of the white films large enough for optical anal-ysis were able to be obtained only until 8th week. Therewere overall color changes in both white films, sincethey became slightly darker (i.e., lower L value) and more

Fig. 8. Changes in mechanical properties of all the films against time: (a)tensile strength and (b) % elongation.

Table 1Changes in molecular weight of the biodegradable films

Time span Wk number Molecular weight (kDa)

25w 35w B

June 6, 2006 0 86.3 ± 2.2 89.3 ± 1.9 84.4 ± 1.7June 23, 2006 2 N/A N/A 60.0 ± 8.5July 5, 2006 4 N/A N/A N/AJuly 19, 2006 6 N/A N/A N/A

September 20, 2006 15 N/A N/A N/AAfter plowed December 20, 2006 28 55.4 ± 0.1 60.5 ± 0.5 51.0 ± 0.5February 21, 2007 37 48.5 ± 0.4 50.9 ± 0.6 43.2 ± 0.3April 25, 2007 46 46.7 ± 5.5 54.4 ± 2.9 45.2 ± 2.5

Fig. 9. Changes in gel content of all the sample films against time.

948 T. Kijchavengkul et al. / Chemosphere 71 (2008) 942–953

yellow (greater b value) (Fig. 6a–c). These changes might beattributed to accumulation of dirt and/or photodegrada-tion. On the other hand, there was no significant changein ‘a’ value for all the films (data not shown).

The percentage of light transmission of the white biode-gradable films decreased throughout the experiment from93% at 0th week to 81% at 8th week, but no change wasobserved from the black biodegradable film and conven-tional film (Fig. 7). The white films did not prevent weedgrowth, which may contribute to the disintegration of thefilm since they were able to penetrate the films and createcracks, since both white biodegradable films still had veryhigh% light transmission rates (all above 80%). Ngouajioand Ernest (2004) reported in their study done in thetomato plots at Benton Harbor, MI that the mulch filmswith higher% light transmission were correlated with thegreater weed density and weed biomass.

Since all the white films started breaking within twoweeks, the samples of the white film for mechanical testingcould only be retrieved until 2nd week. In the case of theblack biodegradable films, the sampling continued until8th week. After two weeks, the white films became verybrittle, while the black ones were still ductile, as shown inFig. 8a and b. Fig. 8a shows that there were decreases intensile strength for all the biodegradable films, especiallyduring the first two weeks for the white films. Conversely,there were no changes for the conventional mulch. It canbe seen in Fig. 8b that the % elongation also dropped afterthe first two weeks for all the biodegradable films. The bio-degradable films were more brittle after exposure in thefield resulting in an increase of the tensile modulus; thedata coincided with the initial visual evaluation done onsite. There was no change in the tensile modulus for theconventional mulch film.

Table 1 displays the changes in molecular weight of thebiodegradable films. Samples of the films obtained from2nd to 15th week could not be dissolved in THF, exceptthe black biodegradable film sample from 2nd week. Thischange in THF dissolution from day zero can be attributedto crosslinking that occurred in the films due to UV radia-tion. The molecular weight of the B film decreased from84.4 kDa at 0th week to 60.0 kDa at 2nd week. After allthe biodegradable mulch films were plowed into the field

in November, the molecular weights of 25w, 35w, and Bfilms (only portions that dissolved in THF) were 55.4,60.5, and 51.0 kDa in December, respectively, 48.5, 50.9,and 43.2 kDa in February, respectively, and 46.7, 54.4,and 45.2 kDa, respectively in April. This is an indicationthat there were likely main chain scissions occurring inthe soil after the films were plowed, but at slow degradationrates because of the low ambient/soil temperatures ofbelow 5 �C and the low solar radiation of below 8.4 MJ d�1

during the winter (see Fig. 2a, c, and d).Since the samples after two weeks of field exposure were

difficult to dissolve in THF, the gel content of the sampleswas measured together with the molecular weight. The gelcontent of the white biodegradable films was around 50–60%, while that of black film was at 20–30% (Fig. 9). Thepossible explanation for the increase in gel content is thataddition of TiO2 to the white biodegradable films affectedthe photodegradation rate of these films. The TiO2 photo-catalytic ability could catalyze the photodegradation of thewhite films and increase their degradation rate (Irick, 1972;Gesenhues, 2000; Searle and Worsley, 2002). On the otherhand, the carbon black used as black colorant for the blackfilm stabilized the photodegradation with a screeningmethod, where any light absorbing substance, carbon blackin this case, absorbs the light energy and reduces the inten-sity of the light reacting with the polymer (Shlyapintokh,1984; Schnabel, 1992).

Fig. 10. (a) Changes in the melting temperatures of all the biodegradablefilm over time. (b) DSC thermogram of the changes in the meltingtemperatures of the black biodegradable film over time.

Fig. 11. Avrami Plot of �ln(1 � Xg) versus difference in onset time (Dt)between the unexposed and the exposed samples.

Table 2Summary of Avrami parameters

Films k n R2

B 0.18a 0.69 0.99425w 0.54b 0.42 0.99935w 0.51b 0.44 0.998

Numbers followed by the same letter within a column are not statisticalsignificantly different at P 6 0.05 (using t-test with Fisher LSDadjustment).

T. Kijchavengkul et al. / Chemosphere 71 (2008) 942–953 949

The Tg, the amorphous transition temperature, and Tm

of the unexposed film were �33.4 ± 1.2 �C, 54.8 ± 0.5 �C,and 121.6 ± 0.3 �C, respectively. There was no change inTg with time (data not shown), but Tm of all three biode-gradable films decreased (Fig. 10a). As the photodegrada-tion proceeded, the melting peaks in the thermogramsbecame broader and the area under these peaks and Tm

were more difficult to identify (Fig. 10b). The broadeningof the peak does not mean that the percentage of crystallin-ity increased since the crosslinked structure of the polymercan disrupt the melting process (Ioan et al., 2001). Cross-linking could cause non-uniformity of the crystal lattices,which makes them melt at different temperatures, hence,causing the broader melting peaks in the thermograms.

Furthermore, the relationship between the difference inthe onset time of the melting peak (between the unexposedsample and the field exposed sample) and the gel content ofthe film at different time was adequately described by theAvrami equation (Eq. (5)), see Fig. 11. The Avrami param-eters for each biodegradable film are summarized in Table2, and the R2 values of all the models for each film weregreater than 0.994. The rate constant (k) of the black bio-degradable film was much lower than that of the whitefilms (Table 2), which supported the finding that the whitefilms crosslinked faster than the black one.

� lnð1� X gÞ ¼ kDtn ð5Þ

where Xg is the gel content of the sample, K is the Avramirate constant, Dt is the difference of the onset time of themelting peak between the unexposed and the field exposedsamples, and n is the Avrami exponent.

The use of DSC associated with the Avrami equationcan be a useful tool for predicting the amount of % gel con-tents in the film samples. The predicted gel content valuesusing the Avrami equation were very close to the actualmeasured values obtained from the ASTM D 2765 testmethod with the greatest percentage difference of �10.3%at 2nd week for the black biodegradable film (B), whilethe rest were less than 5% (Table 3).

From the data on the visual evaluation, the changes inmechanical properties, and increase in gel content of thesamples, it is confirmed that there was crosslinking occur-ring within the exposed biodegradable films, both blackand white. To further understand the mechanism of thecrosslinking, FTIR was used to detect any changes in thefunctional group intensities of the film samples.

There were decreases in the intensities of the C–O bonds(1270 cm�1) and C@O bonds (1710 cm�1) from the estergroups for all the biodegradable films as function of time(Fig. 12a–f). This could be attributed to the high suscepti-bility to photodegradation of the carbonyl group (C@O),the chromophoric group. In addition, there were forma-tions of small amounts of hydroxyl groups (–O–H) at3740 cm�1 starting from 2nd week for all the biodegradable

Table 3Predicted gel content values of different biodegradable films using Avrami equation

Film 25w 35w B

Time (week) Actual Xg (%) Predicted Xg (%) % Diff Actual Xg (%) Predicted Xg (%) % Diff Actual Xg (%) Predicted Xg (%) % Diff

0 0.0 0.0 N/A 0.0 0.0 N/A 0.0 0.0 N/A2 48.7 49.3 1.3 57.3 57.4 0.2 8.0 7.2 �10.34 51.5 50.6 �1.7 51.3 51.3 �0.1 26.6 26.3 �1.06 52.0 52.1 0.3 47.8 48.8 2.0 21.7 22.3 3.08 53.6 53.8 0.3 50.8 49.8 �2.0 23.2 24.3 5.010 N/A N/A N/A N/A N/A N/A 28.3 27.7 �2.312 N/A N/A N/A N/A N/A N/A 22.3 22.6 1.015 N/A N/A N/A N/A N/A N/A 25.5 24.8 �2.6

Fig. 12. FTIR spectra indicated C–O bond at 1200–1315 cm�1 and C@O bond at 1650–1780 cm�1 in ester group: (a) C–O of 25w samples, (b) C–O of 35wsamples, (c) C–O of B samples, (d) C@O of 25w samples, (e) C@O of 35w samples, (f) C@O of B samples.

950 T. Kijchavengkul et al. / Chemosphere 71 (2008) 942–953

films, since the absorbance intensities were slightly higherthan the noise level. The intensities of this hydroxyl peak

remained the same until the end of the season. The smallformation of the hydroxyl groups was attributed to the for-

T. Kijchavengkul et al. / Chemosphere 71 (2008) 942–953 951

mation of the alcohol groups caused by the main chain scis-sion at the ester groups (data not shown).

The increases in the absorbance of the ‘out of plane’bending of benzene groups at 796 and 780 cm�1, withexposure time indicated that there were changes of the ben-zene ring substitutes due to formations of 1,3 meta disub-stitute benzene (corresponded to a wave number of780 cm�1 (Robinson et al., 2005)) and 1,2,4 trisubstitutebenzene (corresponded to the wave number of 796 cm�1

(Robinson et al., 2005)) as photodegradation proceeded(Fig. 13). Therefore, the crosslink structure could resultfrom a recombination of the generated free radicals ofthe 1,3 meta disubstitute benzene and the 1,2,4 trisubstitutebenzene from hydrogen abstraction induced by other free

Fig. 13. FTIR spectra indicated out-of-plane bending of the benzene ringsubstitutes at 765–820 cm�1 (left peaks at 796 cm�1 represent 1,2,4trisubstitution benzene, and right peaks at 780 cm�1 represent 1,3 metadisubstitution benzene): (a) 25w sample, (b) 35w sample, and (c) B sample.

radicals created from photodegradation via Norrish I pro-cess (Buxbaum, 1968; Schnabel, 1992; Rivaton and Gar-dette, 1998) (Figs. 12 and 13). A tentative crosslinkingmechanism previously presented by (Rivaton and Gar-dette, 1998) for poly(butylene terephthalate), which is ahomopolymer produced from one of the comonomers ofPBAT that can be appropriately applied to this finding ispresented in Scheme 2. In this mechanism, one of thehydrogen atoms in the benzene ring in the polymer mainchain is abstracted by a free radical generated from theNorrish I photodegradation reaction and a benzene freeradical is produced. Two of those new radicals then recom-bine to form an H-link structure. This H-link structure isstill susceptible to the photodegradation due to the pres-ence of carbonyl in the ester group. Further exposure toUV light causes those structures to breakdown at the estergroup and make them more susceptible to oxidation.Therefore, the H-link structure is reduced into a Y-linkstructure after further exposed to UV or sunlight. Theseobservations showed that all the biodegradable films pres-ent a high degree of photodegradation which led to cross-linking due to sunlight or UV light exposure. Furtherexperiments to confirm the crosslinking mechanism ofthese biodegradable films due to UV photodegradationare being carried out.

Fig. 14 represents the amount of CO2 evolution of thesamples measured by the DMR system after an incubationperiod of 120 d. From the curve, all the biodegradable filmsbiodegraded at comparable rates to that of the corn starch(positive control), and the percent mineralizations are allgreater than 60% with that of corn starch being greaterthan 70%.

From this experiment, it is possible to conclude that themain degradation mechanism observed for the biodegrad-able mulch films was photodegradation. Therefore, as aconsequence of exposure of the film to sun light, crosslink-ing within the biodegradable films took place as supportedby the following facts: (1) the brittleness of the film firstdetected from the visual evaluation and the changes inthe mechanical properties, (2) the difficulty in dissolvingthe films samples in THF after two weeks of exposure,(3) the FTIR spectra that corresponded to the changes ofthe benzene substitutions and the ester groups, and (4)the broader melting peaks from the DSC thermograms.

As determined by the visual inspection, and Fig. 8a andb, the white biodegradable films, with TiO2, degradedmuch faster than the black one. The biodegradability testusing the DMR under the dark condition (Fig. 14) showedthat all the initial films biodegraded at the same rate; eventhough, they have undergone photodegradation differently.Additives, such as colorant, can affect the photodegrada-tion rate due to catalysis or stabilization. Therefore, thesuccess of the biodegradable mulch films will be dependenton the right balance between photodegradation andbiodegradation.

From the horticulture aspect, the black biodegradablefilm had the ability to suppress weeds comparable to the

Scheme 2. Tentative crosslinking mechanism (Rivaton and Gardette, 1998).

Fig. 14. Amount of carbon dioxide evolution of all the film samplesagainst incubation time.

952 T. Kijchavengkul et al. / Chemosphere 71 (2008) 942–953

conventional mulch. There were no differences among allthe films including the conventional one in the meantomato yield. These findings will be presented in a furtherpublication.

From this experiment, the crosslinking of the biodegrad-able films has lead to one major issue regarding whether itwill affect the biodegradability of the biodegradable films.

Therefore, the biodegradation test of crosslinked samplesis being carried out and will be discussed in a followingpaper (Kijchavengkul et al., accepted for publication).

Acknowledgments

The authors would like to thank to the Michigan Agri-cultural Experiment Station for funding this researchthrough Project GREEEN Grants Number GR05-020Dand GR06-089; Kathryn Severin for facilitating the DSCequipment; Dr. Guangyao Wang and all summers studentsfor their assistance with field activities; Claudio Javier deLa Fluente for helping with Fig. 1; Dr. Ramani Narayanfor comments and suggestions about the DMR systemand the biodegradable films; and Northern Technologiesfor producing the modified Ecoflex mulch films.

References

ASTM, 1998. ASTM D 882-98 Standard Test Method for TensileProperties of Thin Plastic Sheeting. ASTM International, WestConshohocken, PA.

ASTM, 2003a. ASTM D 3418-03 Standard Test Method for TransitionTemperatures and Enthalpies of Fusion and Crystallization ofPolymers by Differential Scanning Calorimetry. ASTM International,West Conshohocken, PA.

T. Kijchavengkul et al. / Chemosphere 71 (2008) 942–953 953

ASTM, 2003b. ASTM D 5338-98 Standard Test Method for DeterminingAerobic Biodegradation of Plastic Materials under Controlled Com-posting Conditions. ASTM International, West Conshohocken, PA.

ASTM, 2004. ASTM D 6400-04 Standard Specification for CompostablePlastics. ASTM International, West Conshohocken, PA.

ASTM, 2006. ASTM D 2765-01 Standard Test Methods for Determina-tion of Gel Content and Swell Ratio of Crosslinked Ethylene Plastics.ASTM International, West Conshohocken, PA.

Buxbaum, L.H., 1968. The degradation of poly(ethylene terephthalate).Angew. Chem. Int. Ed. 7, 182–190.

Espi, E., Salmeron, A., Fontecha, A., Garcia, Y., Real, A.I., 2006. Plasticfilms for agricultural applications. J. Plast. Film Sheet. 22, 85–102.

Gesenhues, U., 2000. Influence of titanium dioxide pigments on thephotodegradation of poly(vinyl chloride). Polym. Degrad. Stabil. 68,185–196.

Giacomell, G.A., Garrison, S.A., Jensen M., Mears D.R., Paterson, J.W.,Roberts, W.J., Wells, O.S., 2000. Advances of Plasticulture Technol-ogies 1977–2000. The 15th International Congress for Plastics inAgriculture, Hershey, PA.

Ioan, S., Grigorescu, G., Stanciu, A., 2001. Dynamic-mechanical anddifferential scanning calorimetry measurements on crosslinked poly(e-ster-siloxane)-urethanes. Polymer 42, 3633–3639.

Irick, G., 1972. Determination of the photocatalytic activities of titaniumdioxides and other white pigments. J. Appl. Polym. Sci. 16, 2387–2395.

Jensen, M.H., 2004. Plasticulture in the Global Community – View of thePast and Future. American Society for Plasticulture, Bellefonte, PA.

Kale, G., Kijchavengkul, T., Auras, R., Rubino, M., Selke, S.E., Singh,S.P., 2007. Compostability of bioplastic packaging materials: anoverview. Macromol. Biosci. 7, 255–277.

Kijchavengkul, T., Auras, R., Rubino, M., Ngouajio, M., Fernandez, R.T.,2006. Development of an automatic laboratory-scale respirometric systemto measure polymer biodegradability. Polym. test. 25, 1006–1016.

Kijchavengkul, T., Auras, R., Rubino, M., Ngouajio, M., Fernandez,R.T., accepted for publication. Assessment of aliphatic–aromaticcopolyester biodegradable mulch films. Part II: laboratory simulatedconditions. Chemosphere.

Lamont, W., 1999. Vegetable Production Using Plasticulture. Food &Fertilizer Technology Center, Taipei, Taiwan.

Lamont, W., Orzolek, M., 2004. Plasticulture Glossary of Terms. TheAmerican Society for Plasticulture, Bellefonte, PA.

Miles, C., 2003. Alternative to Plastic Mulch. Washington State Univer-sity, Mount Vernon, WA.

Muller, R.-J., Kleeberg, I., Deckwer, W.-D., 2001. Biodegradation ofpolyesters containing aromatic constituents. J. Biotechnol. 86, 87–95.

Ngouajio, M., Ernest, J., 2004. Light transmission though coloredpolyehtylene mulches affects weed populations. Hortscience 39,1302–1304.

Rivaton, A., Gardette, J.-L., 1998. Photo-oxidation of aromatic polymers.Angew. Makromol. Chem. (261/262), 173–188.

Robinson, J.W., Skelly Frame, E.M., Frame II, G.M., 2005. Undergrad-uate Instrumental Analysis. Marcel Dekker, New York, NY.

Schnabel, W., 1992. Polymer Degradation: Principles and PracticalApplications. Hanser, New York, NY.

Searle, J., Worsley, D., 2002. Titanium dioxide photocatalysed oxidationof plasticisers in thin poly(vinyl chloride) films. Plast. Rubber Compos.31, 329–335.

Shlyapintokh, V., 1984. Photochemical Conversion and Stabilization ofPolymers. Hanser, New York, NY.

Shogren, R.L., 2001. Biodegradable Mulch. Agricultural ResearchService, Department of Agriculture, United States.

Witt, U., Einig, T., Yamamoto, M., Kleeberg, I., Deckwer, W.-D., Muller,R.-J., 2001. Biodegradation of aliphatic–aromatic copolyesters: eval-uation of the final biodegradability and ecotoxicological impact ofdegradation intermediates. Chemosphere 44, 289–299.