New blends of ethylene-butyl acrylate copolymers with thermoplastic starch. Characterization and bacterial biodegradation

Carbohydrate Polymers 149 (2016) 68–76

Contents lists available at ScienceDirect

Carbohydrate Polymers journal homepage: www.elsevier.com/locate/carbpol

New blends of ethylene-butyl acrylate copolymers with thermoplastic starch. Characterization and bacterial biodegradation A. Morro b , F. Catalina a,∗ , T. Corrales a , J.L. Pablos a , I. Marin b , C. Abrusci b,∗ a b

Department of Applied Macromolecular Chemistry, Instituto de Ciencia y Tecnología de Polímeros, C.S.I.C. Juan de la Cierva 3, 28006 Madrid, Spain Departamento de Biología Molecular, Facultad de Ciencias, Universidad Autónoma de Madrid-UAM, Cantoblanco, 28049 Madrid, Spain

a r t i c l e

i n f o

Article history: Received 15 January 2016 Received in revised form 13 April 2016 Accepted 16 April 2016 Available online 19 April 2016 Keywords: Thermoplastic starch Ethylene-butyl acrylate copolymer Starch blend chemiluminescence Bacterial biodegradation

a b s t r a c t Ethylene-butyl acrylate copolymer (EBA) with 13% of butyl acrylate content was used to produce blends with 10, 30 and 60% of thermoplastic starch (TPS) plasticized with glycerol. Ethylene-acrylic acid copolymer (EAA) was used as compatibilizer at 20% content with respect to EBA. The blends were characterized by X-ray diffraction, ATR-Fourier Transform Infrared Spectroscopy (ATR-FTIR), Scanning Electron Microscopy (SEM), water-Contact Angle measurements (CA), Differential Scanning Calorimetry (DSC) and Stress–strain mechanical tests. Initiated autoxidation of the polymer blends was studied by chemiluminescence (CL) confirming that the presence of the polyolefin-TPS interphase did not substantially affect the oxidative thermostability of the materials. Three bacterial species have been isolated from the blend films buried in soil and identified as Bacillus subtilis, Bacillus borstelensis and Bacillus licheniformis. Biodegradation of the blends (28 days at 45 ◦ C) was evaluated by carbon dioxide measurement using the indirect impedance technique. © 2016 Elsevier Ltd. All rights reserved.

1. Introduction In recent times, biodegradable materials have gained academic and industrial importance particularly for the protection of the environment from ever-increasing plastics waste. Partially biodegradable materials prepared by blending natural biodegradable and non-biodegradable synthetic polymers can reduce the volume of plastics waste by their non-complete environmental biodegradation. These blends exhibit economic advantages and superior properties than the completely biodegradable ones due to the overall properties imparted by the commercial polymer used as a blending component. Thermoplastic starch (TPS) is a promising biodegradable natural polymer of low cost with traditional polymers and particularly in polyolefins (Da Róz, Ferrera, Yamajai, & Carvalho, 2012; Rodriguez-Gonzalez, Ramsay, & Favis, 2003; Taguet, Huneault, & Favis, 2009; Cerclé, Sarazin, & Favis, 2013; Taguet, Bureau, Huneault, & Favis, 2014) such as polyethylene (LDPE), ethylene-vinyl acetate (EVA) or ethylene-vinyl alcohol (EVOH) copolymers. The main drawback of these materials is the poor compatibility between the hydrophilic starch and hydrophobic polymers. One way to increase compatibility in starch blends

∗ Corresponding authors. E-mail addresses: [email protected] (F. Catalina), [email protected] (C. Abrusci). http://dx.doi.org/10.1016/j.carbpol.2016.04.075 0144-8617/© 2016 Elsevier Ltd. All rights reserved.

is to use a compatibilizer containing groups capable of hydrogen bonding with the starch hydroxyl groups. Poly (ethylene-co-acrylic acid) (EAA) is such an example and compatibilization with starch is due to a helical V-type complex formation (Shogren, Thompson, Green, Gordon, & Cote, 1991) and this copolymer has been used as a compatibilizer, typically of 20% of acrylic acid content. Recently, ethylene-butyl acrylate copolymers (EBA) have been increasingly used in outdoor applications as a base polymer film due to some advantages compared to polyethylene. Owing to the presence of butyl ester polar groups in their structure, EBA performance improves significantly in terms of cohesion and adhesion to different substrates, resulting from its greater polarity and lower crystallinity. These properties can provide higher compatibility with the majority of organic additives and polymers. This fact also increases the dispersibility and the maximum concentration of additives in the polymeric matrix, improving film quality and enhancing service life. Numerous studies have been done to investigate the microbial biodegradation of starch- plastic materials (Swanson, Shogren, Fanta, & Imam, 1993). These studies involve blends with polyolefins like LDPE (Danjaji, Nawang, Ishiaku, & Mohd, 2002; Tena-Salcido, Rodríguez-González, Méndez-Hernández, & Contreras-Esquivel, 2008), EVOH (Simmons & Thomas, 1995) and EVA (Da Róz et al., 2012), but there is no related work made with EBA-starch blends.

A. Morro et al. / Carbohydrate Polymers 149 (2016) 68–76

The aim of this study was to prepare a series of EBA/TPS blends using the copolymer EAA as compatibilizer and varying the starch content in order to investigate their behaviour in microbial biodegradation using previously isolated bacterial strains from biofilm formation under exposure to soils. The materials were studied by means of X-ray diffraction, DSC, FTIR, DMTA and Stress–strain mechanical tests. Also, the blends were studied by the chemiluminescence technique and results pertaining to the thermal oxidative stability of EBA/TPS blends are presented. Three bacterial species have been isolated from films buried in soil located in Murcia, Spain. These were found attached to the polymer and identified as Bacillus subtilis, Bacillus borstelensis and Bacillus licheniformis. Biodegradation of the materials was studied by determining the carbon dioxide during their mineralization by the bacteria using an indirect impedance technique. This technique has been previously employed and was proven useful in studying polymer biodegradation (Abrusci, Marquina, Del Amo, & Catalina, 2007; Abrusci et al., 2011).

2. Experimental 2.1. Materials and polymer characterization The Ethylene/butyl acrylate (EBA) with a content of 13% (w/w) in butyl acrylate (BA) was supplied by Repsol (ALCUDIA® EBA PA1303, density 0.925 g ml−1 , Melt Flow Index MFI = 0.3 g/10 min). The Ethylene/acrylic acid (EAA) of 20% of acrylic acid content (PRIMACOR® 5980I, density 0.958 g ml-1, Melt Flow Index MFI = 14 g/10 min) was supplied by Dow Ibérica (Tarragona, Spain). Native potato starch, 20% of amylose and 80% amylopectin (Meritena® 400), was supplied by Syral Ibérica, S.A.U. (Zaragoza, Spain) and Glycerol (CAS N◦ 56815) and Magnesium nitrate hexahydrate (CAS N◦ 13446-18-9) were purchased from Aldrich. Diffraction X-Ray patterns (XRD) were obtained using an X Bruker D8 Advance diffractometer, with CuK␣ radiation. The scanning speed and the step size were 0.5◦ min−1 and 0.02◦ min−1 , respectively. All the experiments were carried out with 2␪ varying from 10◦ to 40◦ . Scanning Electron Microscopy (SEM). Polymer surfaces were examined employing a SEM Philips XL30 model. The samples were coated with approx. 3 nm of gold/palladium using a Polaron SC 7640 sputter coater. Contact Angle (CA) was measured using a CAM200 KSV equipment using Millipore grade distilled water as wetting liquid. Five independent advanced contact angles were measured and the average values were taken. The surface tension (ST) was calculated as described in the literature (Fowkes, 1964). Dynamic Mechanical Thermal Analysis (DMTA) was performed on a METTLER-TOLEDO DMA/SDTA861e instrument (tensile mode, frequency of 1 Hz) in the temperature range from −130 ◦ C to 65 ◦ C and heating rate of 2 ◦ C min−1 . Attenuated Total Reflectance/FT-Infrared Spectroscopy (ATR-FTIR). IR spectra were obtained using a Perkin Elmer BX-FTIR spectrometer coupled with an ATR accessory, MIRacleTM-ATR from PIKE Technologies. Tensile properties. Tensile strength (␴) and retention of elongation at break (␧) were determined using a dynamo-metric MTS Q-Elite apparatus (room temperature, stretching speed of 50 mm/min). Before testing, all compression moulding films (200 ␮m thickness) were stored under controlled atmosphere by a saturated Mg(NO3 )2· 6H2 O solution (R.H. 53% at 25 ◦ C). Standard dumbbell probes (5 specimens of 35 mm) were tested and the average values were taken. Differential Scanning Calorimetry (DSC) was performed on a METTLER DSC-823e instrument (30–180◦ C) previously calibrated with

69

an indium standard (Tm = 429 K, Hm = 25.75 Jg−1 ). Film samples (10 mg) were heating or cooling at 5 ◦ C/min rates under nitrogen. Polyethylene crystallinity (␹c %) were determined using the area of the melting endothermic peak and related to a reference of 293 Jg−1 for crystalline polyethylene (Flory & Vrij, 1963). 2.2. Preparation of blends TPS was obtained by mixing starch powder, water and glycerol in the ratio of 100:20:30 (w/v/v), respectively, for 30 min to obtain a paste which was transformed to TPS by heating at 120 ◦ C in thermostatic bath with continuous stirring for 30 min. EBA and TPS were mixed in different ratios maintaining a constant EAA at 20% (see Table 1 for compositions). The blends were prepared by melt processing in a Haake MiniLab mixer extruder with two counter-rotating screws at 135 ◦ C of temperature. The rotating speed of the rotor and the mixing time were respectively 80 rpm and 10 min. Polymer films (200 ␮m) were made by compression moulding (1 g of blended powder) in a Collin-200 press 170 ◦ C and pressure cycle: 2 min at 0 bar and 1 min at 200 bar. All blends contained a 20% of EAA respect to EBA as compatibilizer and samples will be referred as EBA% TPS, being% the percentage of TPS content (w/w) in the blend. 2.3. Isolation of bacterial strains Film samples were scattered in November in an agricultural soil (Torre Pacheco, Murcia, Spain). Films collected after 7 days for microbial identification were cultured in trypticase soya agar (TSA) and grown at different ranges of temperature and their ability to hydrolyze starch and glycerol was tested as described in Abrusci et al. (2004). 2.4. PCR amplification of 16S rRNA DNA from the enrichment culture was extracted using an Ultra Clean microbial DNA isolation kit (MO BIO Labs., Inc., Solana Beach, CA), following the manufacturer’s directions, and purified using a DNA purification JetQuick kit (Genomed). The sequences obtained (about 1453 nt) were compared to those available in the GenBank database (National Center for Biotechnology Information) using the Basic Local Aligment Search Tool (BLAST) (http://www.ncbi.nlm. nih.gov/blast/) algorithm to identify known sequences with high similarity. The selected sequences were aligned with CLUSTAL X (Thompson, Gibson, Plewniak, Jeanmougin, & Higgins, 1997). The bacteria isolated from the EBA60 samples and identified were B. subtilis, B. borstelensis, B. licheniformis. A mixture of the three identified bacterial strains, noted herein as MIX, was used for the biodegradation studies. 2.5. Bioassay procedure and indirect impedance technique Aerobic biodegradation of film samples by bacteria were conducted at 45 ◦ C in bioreactors of 7-ml of capacity filled with 1 g of sterile silica and 1.5 ml of bacterial suspension in minimal growth medium of 2.5 × 107 cells/ml concentration, prepared as described in Abrusci et al. (2011). After that, discs of film samples (4 mg) were added to the medium. These containers were introduced in 20-ml disposable cylindrical cells charged with 1.5 ml of 2 g/l KOH aqueous solution and provided with electrodes to measure impedance on a Bac-Trac 4300 (SY-LAB Geräte GmbH, Neupurkerdorf, Austria). The device monitors the relative change in the initial impedance value of KOH solution which is converted to concentration of carbon dioxide by a calibration curve of impedance variation versus

70

A. Morro et al. / Carbohydrate Polymers 149 (2016) 68–76

Table 1 Composition of EBA/TPS blends (% w/w), EBA crystallinity (␹c ), melting temperature (Tm ) and crystallization temperature (Tc ), advancing water contact angle (CA), surface tension (ST), tensile strength (␴) and elongation at break (␧) of the films. Film sample

EBAa (%)

EAA (%)

TPS (%)

Tm ( ◦ C)c

Tc (◦ C)c

␹c (%)b

TPS EBA EBA10% TPS EBA30% TPS EBA60% TPS

0 80 70 50 20

0 20 20 20 20

100 0 10 30 60

– 105 103 104 101

– 89.0 84.5 88.6 85.5

– 19.8 14.4 12.0 7.3

a b c d

Heating

␹c (%)b

Cooling

– 21.2 17.7 12.7 7.1

CAc (◦ )

STc (mN/m)

␴ (MPa)d

␧ (%)d

37.0 101.8 90.6 89.4 85.2

86.3 71.8 62.4 61.5 60.9

– 15.1 12.3 8.5 4.9

– 543 480 180 55

Content of butyl acrylate 13%. PE phase- determined by DSC. Determined by water-tensiometry. Compression moulded films.

Fig. 1. X-ray diffraction patterns of EBA, TPS and their blends.

concentration of CO2 . Under our experimental conditions, in the absence of EBA films the media impedance of the KOH solution remained constant, confirming the absence of organic contaminants. The experimental device and procedure have been described by Abrusci et al. (2004, 2007). 2.6. Chemiluminescence Chemiluminescence emission (CL) of circular film samples (12 mm), were obtained as described earlier, by using a CL400ChemiLUME from ATLAS Co. (Catalina, Peinado, Allen, & Corrales, 2002). Measurements were made under nitrogen or oxygen (50 ml/min.) in isothermal conditions (170 ◦ C) and temperature test-ramp (2 ◦ C/min.). The data collected were processed using the specific software supplied with the instrument. 3. Results and discussion 3.1. Preparation and morphological characterization of EBA-TPS blends Firstly, TPS was produced by mixing the native starch with glycerol as a plasticizer (30%) at a temperature of 120 ◦ C, above the starch gelatinization temperature (70–90 ◦ C). This gelatinization was confirmed by XRD as described in the next section. The extrusion mixing time of 10 min at 135 ◦ C was adequate for TPS and EBA/EAA polymeric materials. Longer periods led to some TPS degradation, as suggested by the development of a light brown

colour. Table 1 shows the composition of the blends prepared in this work. The EBA blends were named EBA%, for the polyolefin mixture of EBA and 20% of EAA followed by the percentage of TPS content. Visual observation of EBA blends with TPS showed that all the compositions were quite homogeneous and translucent. The difference between the refract indexes of semi-crystalline polymer phase of EBA, similar to LDPE ( = 1.51) (Brandrup & Immergut, 1989) and the TPS ( = 1.489) (Rodriguez-Gonzalez et al., 2003) is small and represents a small value resulting in low opacity even at a high content of TPS. The surface analysis obtained by SEM showed, in general, a smooth surface, as shown in Fig. 6A and B corresponding to 90/10 and 40/60 EBA/TPS films and only differing in textures. The starch domain size increases and small TPS particles can be distinguish on the surface. Very small micro voids could be present beyond the range of typical microscopic techniques and it is highly probable that a very thin glycerol layer coats the surface of TPS in these materials as other authors (Baldev, Udaya Sankar, & Siddaramaiah, 2004; Taguet et al., 2009) reported in morphological interface studies of LDPE/TPS blends. Higher TPS content results in the formation of micro voids and can be the other potential cause of opacity in immiscible polymer blends. The presence of a 20% of EAA in the blends produced translucent materials instead of opaque confirming a quite good interfacial contact and a significant reduction in interfacial voiding (Leclair & Favis, 1996). The SEM images shown in Fig. 6A and B revealed a smooth surface in the presence of plasticized TPS caused

A. Morro et al. / Carbohydrate Polymers 149 (2016) 68–76

A

71

B

Fig. 2. Curves of storage modulus (A) and loss factor tg ␦ (B) versus temperature of TPS, EBA-20% of EAA and their blends.

A

B

C

D

Fig. 3. Chemiluminescence emission from film samples: Temperature-ramping curves obtained under nitrogen (A) and oxygen (B) and CL curve versus time under oxygen at 170 ◦ C (C) and expanded plot (D) showing the initial CL-emission.

by the addition of glycerol, which improves interface compatibility through some migration to the interface EBA/TPS. This also facilitates chain mobility while decreasing the Tg values of the starch. 3.2. Film characterization by XRD, DSC, ATR-FTIR and DMTA Fig. 1 shows the X-ray diffractograms of the EBA/TPS blends compared with that of TPS and EBA containing 20% of EAA as reference materials. A-type characteristic peaks of the native crystalline starch phase (diffraction peaks at 15.0◦ , 18.1◦ and 22.9◦ ) disappeared completely in the TPS processed with the mixture

water/glycerol. Also, the X-ray diffraction peaks associated to Vtype structures were present at 17.5 and 20.1◦ in the TPS material as shown in Fig. 1. In addition upon aging of the water-plasticized TPS, B-type structures can be formed leading to peaks at 2␪ = 22.3◦ and 26.1◦ , a phenomenon associated with retrogradation (Shi et al., 2007). Reference films of EBA with 20% of EAA blends exhibited the characteristic orthorhombic diffraction peaks of polyethylene segments present in EBA and EAA copolymers at 2␪ = 21.4◦ , 23.9◦ and 36.2◦ .

72

A. Morro et al. / Carbohydrate Polymers 149 (2016) 68–76

Fig. 4. ATR-FTIR spectra of EBA 60% TPS blend film: initial and after 7 days of bacterial treatment at 45 ◦ C together with TPS film as reference.

Fig. 5. Percentage of carbon mineralization at 45 ◦ C, determined from carbon dioxide measurements in the biodegradation of EBA/TPS films by: (A) the mixture of Bacillus, MIX including model compound acrylic acid and butyl acrylate, and (B) B. subtilis.

The development of ‘B’-type lattice structures could be attributed to the fast recrystallization of amylose or the outer chains of amylopectin, or both, in the presence of glycerol. X-ray diffractograms of all blends showed similar curves differing in relative intensity of the characteristics peaks of polyethylene and TPS. The intensity of the peak at 21.6◦ increased with increasing concentration of LDPE in the blends. Even in the blend EBA60% TPS it is possible to distinguish the polyethylene crystallinity peaks overlapping with the TPS peaks. This fact confirms the clear phase immiscibility of both polymers. The DSC thermogram of the EBA-20% EAA blend shows a broad endotherm peak in the temperature range 65–110 ◦ C, which corresponds to the melting transition of polyethylene segments of EBA (Tm = 105 ◦ C) and EAA (Tm = 79 ◦ C) copolymers, Table 1. An additional transition at temperatures between 35 and 50 ◦ C can be observed in the EBA/TPS blends due to the Tg of the employed EAA compatibilizer (42 ◦ C). Under our experimental conditions, no endotherms for TPS were observed by DSC suggesting that all the starch had been gelatinized and its amorphous phase was maintained during mixing and processing and that recrystallization was inhibited due to an interaction with the EAA. The crystallinity of the polyethylene segments was calculated in heating and cooling scans respectively,

the data are summarized in Table 1. There is a little difference in melting and crystallization temperatures for all the film samples, suggesting that polyethylene exists as a separate phase, interacting to only a minor extent with the TPS. The decrease in crystallinity content with the increase of TPS indicates that part of the hydrocarbon portion of EAA is interacting strongly with starch. This fact is in agreement with the results reported by other authors (Shogren et al., 1992). EBA/TPS blends were also characterized by ATR-FTIR spectroscopy and the obtained spectrum of the blend of 60% content of TPS was plotted in Fig. 4. Infrared spectra show mainly the characteristic bands of TPS and polyolefin phase and the spectrum plotted in Fig. 4 (EBA60% TPS) did not show any specific band due to bond interaction of TPS and the polyolefin component of the blend. The main peaks observed and assigned to TPS were (Capron, Robert, Colonna, Brogly, & Planchot, 2007; Wang & Huang, 2007; Sankri et al., 2010): O H (3000–3600 cm−1 ) stretch, C H (2800–3000 cm−1 ) stretch, C C, C O and C O C (950–1200 cm−1 ) bending modes and skeletal mode vibration of the glycosidic linkage (900–950 cm−1 ). Also, the characteristic bands of EBA and EAA are: 1735 cm−1 carbonyl butyl ester band and 1705 cm−1 carbonyl acid group bands of EAA.

A. Morro et al. / Carbohydrate Polymers 149 (2016) 68–76

The tensile strength (␴) and elongation at break (␧) of the films previously conditioned were measured and the obtained data are shown in Table 1. Maximum stress decreased with increasing TPS contents from 15.1 MPa for film without TPS (EBA-20% EAA) to approximately 4.9 MPa for blend film containing 60% of TPS. Similar effect was observed for the corresponding values of the elongation at break. The change of elongation at break with the content of TPS was significant at 30% of TPS and became especially significant for blend with 60% starch content. The obtained values of the elongation at break for the compression moulded films of EBA/TPS blends are much lower than those reported for other films used as packaging materials (Wang et al., 2015; Wei et al., 2016). The glass transition was studied by DMTA observing the storage modulus and the loss factor (tg ␦) as a function of temperature as shown in Fig. 2. In the case of plasticizer starch/glycerol systems (TPS), DMTA highlights the presence of two relaxation phenomena characterized by two peaks on the loss factor curve, around −47 and −7 ◦ C revealing the heterogeneity of the TPS system (30% of glycerol) and a phase separation. The first peak is associated with the relaxation of glycerol-rich domains, and the second one with starch-rich domains; similar results were reported by other authors (Loudin, Bizot, & Colonna, 1997; Anglés, & Dufresne, 2000). As expected, a sharp decrease occurs in storage modulus, Fig. 2A, accompanied by the peak in tg ␦ due to the glycerol reach domain relaxation. As is well known the loss factor (tg ␦) of low density polyethylene, as shown in Fig. 2B, reveals three peaks (Khanna, Turi, Taylor Vickroy, & Abbott, 1985; López-Vilanova, Martinez, Corrales, & Catalina, 2014) called the ␣, ␤ and ␥ relaxations, which appear for the mixture EBA-20% of EAA at T␣ = 25 ◦ C, T␤ = −20 ◦ C and T␥ = −105 ◦ C. In Fig. 2, the DMTA results for the tg ␦ and storage modulus of the blends are intermediate between those of the two component polymers. The temperature of the ␣ relaxation (T␣) of the polyethylene phase in the blends decreases with the content of TPS due the lower crystallinity (Table 1). The value of T␣ = 25 ◦ C for the EBA-20%EAA decreased to 18 ◦ C for the blend EBA60% TPS. In the blends, the relaxation at −47 ◦ C is the same as that observed for the glycerolrich domain in pure TPS. The tan ␦ peak of the starch-rich region at −7 ◦ C observed for the TPS cannot be observed for the blends due to the contribution of polyethylene ␤ relaxation, although traces of this peak are discernible for the 60% TPS blend.

3.3. Chemiluminescence analysis It is generally accepted that CL-emission can be related to the hydroperoxide (POOH) content in the polymer. Hydroperoxides are important intermediate products in the oxidation process and their decomposition rate greatly affects the overall oxidation rate. (Catalina et al., 2002; Jacobson, Eriksson, Reitberger, & Stenberg, 2004). In spite of the numerous studies on chemiluminescence accompanying thermooxidation of many synthetic polymers (Corrales, Peinado, Abrusci, Allen, & Catalina, 2010), the light emission from polysaccharides was investigated to a lesser extent (Strlic, Kolar, Philar, Rychly, & Matisova-Rychla, 2001; Rychly et al., 2004). Chemiluminescence of TPS in an inert atmosphere is relatively intense especially at elevated temperatures (Fig. 3A), although low when compared with the signal obtained in similar conditions in an oxygen atmosphere (Fig. 5B ), as reported for cellulose. (Strlic et al., 2001). This phenomenon was shown to originate in a thermolysis (transglycosidation) reaction. A weak peak appears with a maximum situated at approximately 120 ◦ C. This emission has been observed in cellulose and related to the presence of peroxide content in pulp.

73

As shown in Fig. 3A, the auto-oxidation temperatures of the blends under nitrogen were lower than that observed for the EBA-20%EAA. This result indicates that for the blends the autoxidation under nitrogen atmosphere started in the EBA/TPS interfacial region. Also, the CL emission under nitrogen confirmed that no significant amounts of CL species were induced by oxidation during film processing The CL-temperature-ramping emissions of the EBA/TPS blends under oxygen (ICL-max) decreased with increasing TPS content, Fig. 3B. The onset temperature (TOnset ), temperature of the emission maxima (TCL-peak ) and Intensity of the peak (ICL-max ) are summarized in Table 2. The CL emission of TPS sample under oxygen at 170 ◦ C is weak since the steady-state content of hydroperoxides is much lower than that found in polyolefins Fig. 3C and D. Similar behaviour has been reported for cellulose (Strlic et al., 2001). In blends, CL-emission maxima (ICL-max ) decrease when the content of TPS increased. In contrast, the initiation of oxidation named as Oxidation Induction Time (OITCL-170 ) remained at a similar value (Table 2) for the EBA-20%EAA samples and the blends with TPS except for the case of the EBA60% TPS. In this blend (Fig. 3C), the values of OITCL-170 and ICL-max are significantly lower than those obtained for samples with a higher content of polyolefin phase. The results revel that the presence of polyolefin-TPS interphase did not substantially affect the oxidative thermostability of the materials. 3.4. Bacterial isolation and identification After exposure in an agricultural field, films contaminated were cultured (37 ◦ C, 24 h with TSA) and the different colonies of bacteria developed were isolated. Also, optimal temperatures for growth were tested using different temperature conditions in a range of temperatures between 5 and 55 ◦ C and the optimal growth was observed in the range between 30 and 45 ◦ C. Three different bacterial strains were obtained and all of them were able to hydrolyze starch and glycerol. The bacterial strains were identified by means of their 16S rDNA sequence obtained after PCR amplification and sequencing. Comparison of the 16S rDNA sequences of the isolated strains with the sequences available in the GenBank database showed that the isolated bacteria were B. borstelensis (GU125631), B. subtilis (AF318900) and B. licheniformis (GU137297), with a similarity equal or over 99%. The three Bacilli species isolated from the EBA/TPS blend films exposed to the agricultural field are widespread in soil and air and form endospores which are ubiquitous in nature and commonly found in soil and vegetation (Claus & Berkeley, 1986). 3.5. Biofilm formation and biodegradation In order to study the biofilm formation, the blend films were collected from the bioassay reactors after one week and analysed by FTIR-ATR. The addition of starch in the blends facilitates early adherence and biofilm formation due, in a first step, to the increasing hydrophilic character of the surface. The water contact angles decreased as the TPS content of the blends increased (Table 1). In the case of the hydrophobic polyolefin film EBA-20% EAA (CA = 101.8◦ ) there is no bacterial adhesion. In Fig. 4, the ATR-FTIR spectra for 7-days EBA60-exposed to B. subtilis and the mixture of Bacillus (MIX) in the bioassay reactor is shown together with the references TPS and initial EBA60% TPS. Incubation of the EBA60% TPS blend with B. subtilis and MIX for 7 days showed a reduction in the amount of TPS as evidenced by the decrease of typical TPS bands in the blend. The broad band in the region 1260–925 cm−1 peaking at 1070 cm−1 (overlapping with the stretching vibration bands of TPS) shows the presence of polysaccharides, which are the major constituents of the biofilm

74

A. Morro et al. / Carbohydrate Polymers 149 (2016) 68–76

Fig. 6. Scanning electron micrographs of blends films with different contents of TPS: initial blends (A) EBA10% TPS and (B) EBA60% TPS and collected from the bioassay reactors after exposure to the consortium of Bacillus, MIX: (C) EBA10% TPS and (D) EBA60% TPS after 21 days of treatment; (E) and (F) EBA60% TPS after 14 days of treatment and without rinsing with water.

Table 2 CL data obtained under oxygen atmosphere for EBA/TPS blends. Film Sample

CL-T/ramping ◦

TPS EBA-20%EAA EBA10 EBA30 EBA60 a b

under O2 ◦

CL-T/isotherm

under O2 at 170 ◦ C

TOnset ( C)

TCL-peak ( C)

ICL-max (mV)

ICL-max (mV)

OITCL-170 (min)

120a 183 187 189 194

– 225 226 214 205

– 31898 21786 8191 4033

573b 8550 7943 4949 1955

5 22 23 22 17

Slight slope, no peak observed. Plateau in the CL-emission.

(Kavita, Mishra, & Jha, 2013; Abrusci et al., 2011). This strong band that is similar in intensity was observed in biofilms produced by MIX and B. subtilis. The relative intensity of this polysaccharide band is much stronger than that observed in earlier works in other Bacillus biofilms (Abrusci et al., 2011). This fact can be due to the increase in compatibility between the TPS of the blend and bacterial exopolysaccharides. Also, in Fig. 4 bands at 1643 cm−1 and 1550 cm−1 typical of amide I and II bands, respectively, confirms the presence of protein material on the polymer surface. Biodegradation of the EBA/TPS blends was studied by determining the carbon dioxide produced in the metabolic action of the isolated bacteria, using the indirect impedance technique. The time

periods of carbon mineralization at 45 ◦ C of the EBA/TPS films are reported in Fig. 5 for the mixture of Bacillus MIX and B. subtilis. From the results plotted in Fig. 5, biodegradation of the films started from the beginning of the bioassays and the production of carbon dioxide was increased with the content of TPS in the materials. The metabolic action of the consortium MIX was efficient in the biodegradation of low molecular-weight products included in the study as model compounds (Fig. 5A), acrylic acid (76% mineralization) and butyl acrylate 53% mineralization. In contrast, no detectable production of carbon dioxide was found for both series of bioassays on EBA-20% EAA films during the 28 days of bioassay. This result is in agreement with the absence of biofilm formation on EBA-20% EAA.

A. Morro et al. / Carbohydrate Polymers 149 (2016) 68–76

In general, the consortium MIX was more efficient in degrading the EBA/TPS blends than the isolate strain of B. subtilis which was obtained from the same consortium, indicating a synergistic effect of the mixture of Bacillus. The good blend compatibility seen here would also to explain the resistance of the EBA/TPS blend films with low content of TPS (10 and 30%) to mineralization of the TPS phase. Steric interferences of EAA macromolecules neighbouring the TPS domains inhibit the enzymatic action. Exoenzimes would not be able to penetrate easily to the whole thickness of the film due to its hydrophobicity (Table 1) that diminishes their swelling in water. Hence, the mineralization percentages reached after 28 days of bioassays (Fig. 5) are 6.6% for EBA10% TPS and approximately 15% for EBA30% TPS. At these lower TPS concentrations, only TPS close to the surface would be accessible to a direct facile attack by microorganisms. Similar results were found by other authors (Baldev et al., 2004; Tena-Salcido et al., 2008). Only in the blend with high content of TPS, EBA60% TPS, the mineralization percentage reaches 60% with MIX in 18 days and 23 days with B. subtilis, confirming the total biodegradation of the TPS phase. To illustrate the biodeterioration process of the materials, micrographs of EBA10% EBA60% TPS blend films exposed to the consortium of bacteria MIX after three weeks are presented in Fig. 6C and D respectively, and for EBA60% TPS blend films after two week without rinsing with water (Fig. 6E and F). The SEM micrograph of the EBA10% TPS sample after three weeks of the test shows small perforations on the surface of the film (Fig. 6C). In contrast, EBA60% TPS exhibits important modifications on the film surface and hole formations (Fig. 6D). Also, Fig. 6E and F show the bacterial colonization around the holes in the later material after two week of treatment. The high TPS content in the blend (EBA60% TPS) allows access of all the starch to microorganisms and TPS domains would be completely biodegraded like the raw TPS. This fact could indicate a continuous nature of the gelatinized starch phase. As expected, the polyolefin phase remained unchanged after 28 days of bioassay. It has been found by other authors (Peansky, Long, & Wool, 1991) in the study of hydrolytic degradation of polyethylene-starch blends that, almost 40% by weight of starch (30% by volume) is required in the formulation to assure interconnectivity of the starch domains. Our results indicate that the concentration of TPS in these EBA/TPS blends influences its availability to bacterial attack and demonstrate that selection of microorganisms is a fundamental factor to biodegrade these blends as a sole carbon source.

4. Conclusions These results have shown that films made with thermoplastic starch plasticized with glycerol (TPS), EBA and EAA present a compatible TPS-EAA phase and a separate incompatible EBA phase. The melt temperature and crystallization temperature are slightly changed due to the presence of the starch and there was gradual decrease in crystallization due to the increasing concentration of TPS in the EBA/starch system indicating that part of the hydrocarbon portion of EAA is interacting strongly with the starch. Low temperature transitions were observed in TPS (30% glycerol) by DMTA and two relaxation phenomena characterized by two peaks on the loss factor curve, the first one at −47 ◦ C associated with the relaxation of glycerol-rich domains and the second one at −7 ◦ C with starch-rich domains. In the blends, the results are intermediate between those of the two components. Chemiluminescence (CL) of TPS in an inert atmosphere is relatively intense especially at elevated temperatures, although low

75

when compared with the signal obtained in similar conditions in an oxygen atmosphere. The CL analysis of the EBA/TPS blends revealed that the presence of polyolefin-TPS interphase did not substantially affect the oxidative thermostability of the materials. Three bacteria adhering to EBA/TPS blend films buried in soil were isolated and identified as B. borstelensis (GU125631), B.subtilis (AF318900) and B. licheniformis (GU137297). Biodegradation of the EBA/TPS blends was studied by evaluating the mineralization at 45 ◦ C using the indirect impedance technique. The consortium of the three Bacilli, MIX, was more efficient in degrading the EBA/TPS blends than the isolated strain of B. subtilis, indicating a synergistic effect of the enzyme complex produced by the mixture of bacteria The mineralization percentages reached after 28 days of bioassays are 6.6% for EBA10% TPS and approximately 15% for EBA30% TPS. The blend with a high content of TPS, EBA60% TPS reached 60% of mineralization confirming the total biodegradation of the TPS phase The high degree of TPS continuity demonstrated in this work indicates that these blend systems have the potential to render all TPS domains in the blend accessible for bacterial biodegradation. These EBA/TPS materials had the added benefit of containing large quantities of a renewable resource completely accessible for biodegradation and hence represent a more sustainable alternative to pure synthetic polymers. Acknowledgments The authors would like to thank the MINECO (Spain) for financial support (MAT2012-31709). One of the authors, C.A., would also like to thank the Ramón y Cajal program. References Abrusci, C., Marquina, D., Del Amo, A., & Catalina, F. (2007). Biodegradation of cinematographic gelatin emulsion by bacteria and filamentous fungi using indirect impedance technique. International Biodeterioration and Biodegradation, 60, 137–143. Abrusci, C., Martin Gonzalez, A., Del Amo, A., Corrales, T., & Catalina, F. (2004). Biodegradation of type-B gelatine by bacteria isolated: a viscosimetry study. Polymer Degradation and Stability, 86, 283–291. Abrusci, C., Pablos, J. L., Corrales, T., López Marín, J., Marín, I., & Catalina, F. (2011). Biodegradation of photo-degraded mulching films bases on polyethylenes stearates of calcium and iron as pro-oxidant additives. International Biodeterioration and Biodegradation, 65, 451–459. Anglés, M. N., & Dufresne, A. (2000). Plasticized starch/tunicin whiskers nanocomposites. 1. Structural Analysis. Macromolecules, 33, 8344–8353. Baldev, R., Udaya Sankar, K., & Siddaramaiah. (2004). Low density polyethylene/starch blends films for food packaging applications. Advances in Polymer Technology, 23, 32–45. Brandrup, J., & Immergut, E. H. (1989). Polymer handbook (3rd ed.). New York: Wiley. Capron, I., Robert, P., Colonna, P., Brogly, M., & Planchot, V. (2007). Starch in rubbery and glassy states by FTIR spectroscopy. Carbohydrate Polymers, 68, 249–259. Catalina, F., Peinado, C., Allen, N. S., & Corrales, T. (2002). Chemiluminescence of polyethylene: the comparative antioxidant effectiveness of phenolic stabilizers in low-density polyethylene. Journal of Polymer Science Part A: Polymer Chemistry, 40, 3312–3326. Cerclé, C., Sarazin, P., & Favis, B. D. (2013). High performance polyethylene/thermoplastic starch blends through controlled emulsification phenomena. Carbohydrate Polymers, 92, 138–148. Claus, D., & Berkeley, R. C. W. (1986). The genus Bacillus. In P. H. A. Sneath, M. E. Sharpe, & J. G. Holt (Eds.), Bergey’s manual of systematic bacteriology (pp. 1105–1139). Baltimore: Williams and Wilkins. Corrales, T., Peinado, C., Abrusci, C., Allen, N. S., & Catalina, F. (2010). Chemiluminescence processes in polymeric materials. In N. S. Allen (Ed.), Photochemistry and Photophysics of polymer materials (pp. 93–135). Hoboken, New Jersey: Wiley & Sons. Da Róz, A. L., Ferreira, A. M., Yamaji, F. M., & Carvalho, A. J. F. (2012). Compatible blends of thermoplastic starch and hydrolyzed ethylene-vinyl acetate copolymers. Carbohydrate Polymers, 90, 34–40. Danjaji, I. D., Nawang, U. S., Ishiaku, H., & Mohd, Z. A. (2002). Degradation studies and moisture uptake of sago-starch-filled linear low-density polyethylene composites. Polymer Testing, 21, 75–81. Flory, P. J., & Vrij, A. (1963). Melting points of linear-chain homologs: the normal paraffin hydrocarbons. Journal of the American Chemical Society, 85, 3548–3553.

76

A. Morro et al. / Carbohydrate Polymers 149 (2016) 68–76

Fowkes, F. M. (1964). Attractive forces and interfaces. Industrial and Engineering Chemistry, 56, 40–52. Jacobson, K., Eriksson, P., Reitberger, T., & Stenberg, B. (2004). Chemiluminescence as a toll for polyolefins oxidation studies. Advances in Polymer Science, 169, 151–176. Kavita, K., Mishra, A., & Jha, B. (2013). Extracellular polymeric substances from two biofilm forming Vibrio species: characterization and applications. Carbohydrate Polymers, 94, 882–888. Khanna, Y. P., Turi, E. A., Taylor, T. J., Vickroy, V. V., & Abbott, F. (1985). Dynamic mechanical relaxations in polyethylene. Macromolecules, 18, 1302–1309. López-Vilanova, L., Martinez, I., Corrales, T., & Catalina, F. (2014). Photoreversible crosslinking of poly-(ethylene-butyl-acrilate) copolymers functionalized with coumarin chromophores using microwave technology. Reactive and Functional Polymers, 85, 28–35. Leclair, A., & Favis, B. D. (1996). The role of interfacial contact in immiscible binary polymer blends and its influence on mechanical properties. Polymer, 37, 4723–4728. Loudin, D., Bizot, H., & Colonna, P. (1997). Antiplastificization in starch-glycerol films. Journal of Applied Polymer Science, 63, 1047–1053. Peansky, J. S., Long, J. M., & Wool, R. P. (1991). Percolation effects in degradable polyethylene-starch blends. Journal of Polymer Science Part B: Polymer Physics, 29, 565–579. Rodriguez-Gonzalez, F. J., Ramsay, B. A., & Favis, B. D. (2003). High performance LDPE/thermoplastic starch blends: a suitable alternative to pure polyethylene. Polymer, 44, 1517–1526. Rychly, J., Matisova-Rychla, L., Slovak, K., Strlic, M., Kocar, D., & Kolar, J. (2004). Thermal oxidation of cellulose investigated by chemiluminescence: the effect of water at temperatures above 100 ◦ C. Carbohydrate Polymers, 58, 301–309. Sankri, A., Arhaliass, A., Dez, I., Gaumont, A. C., Grohens, Y., & Lourdin, D. (2010). Thermoplastic starch plasticized by an ionic liquid. Carbohydrate Polymers, 82, 256–263. Shi, R., Liu, Q. Y., Ding, T., Han, Y. M., Zhang, L. Q., Chen, D. F., et al. (2007). Ageing of soft thermoplastic starch with high glycerol content. Journal of Applied Polymer Science, 103, 574–586. Shogren, R. L., Thompson, A. R., Green, R. V., Gordon, S. H., & Cote, G. (1991). Complexes of starch polysaccharides and poly(ethylene co-acrylic acid): Structural characterization in solid state. Journal of Applied Polymer Science, 47, 2279–2286.

Shogren, R. L., Thomson, A. R., Felker, F. C., Harry-O´ıKuru, R. E., Gordon, S. H., Greene, R. V., et al. (1992). Polymer compatibility and biodegradation of starch-poly(ethylene-co-acrylic acid)—polyethylene blends. Journal of Applied Polymer Science, 44, 1971–1978. Simmons, S., & Thomas, E. L. (1995). Structural characteristics of biodegradable thermoplastic starch/poly(ethylene-vinyl alcohol) blends. Journal of Applied Polymer Science, 58, 2259–2285. Strlic, M., Kolar, J., Pihlar, B., Rychly, J., & Matisova-Rychla, L. (2001). Initial degradation processes of cellulose at elevated temperatures revisited—chemiluminescence evidence. Polymer Degradation and Stability, 72, 157–162. Swanson, C. L., Shogren, R. L., Fanta, G. F., & Imam, C. L. (1993). Starch-plastic materials-preparation, physical properties: and biodegradability (A review of recent USDA Research). Journal of Environmental Polymer Degradation, 1, 155–166. Taguet, A., Bureau, M. N., Huneault, M. A., & Favis, B. D. (2014). Toughening mechanisms in interfacially modified HDPE/thermoplastic starch blends. Carbohydrate Polymers, 114, 222–229. Taguet, A., Huneault, M. A., & Favis, B. D. (2009). Interface/morphology relationships in polymer blends with thermoplastic starch. Polymer, 50, 5733–5743. Tena-Salcido, C. S., Rodríguez-González, F. J., Méndez-Hernández, M. L., & Contreras-Esquivel, J. C. (2008). Effect of morphology on the biodegradation of thermoplastic starch in LDPE/TPS blends. Polymer Bulletin, 60, 677–688. Thompson, J., Gibson, T., Plewniak, F., Jeanmougin, F., & Higgins, H. (1997). The CLUSTAL X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Research, 25, 4879–4882. Wang, H., Wei, D., Ziaee, S., Xiao, H., Zheng, A., & Zhao, Y. (2015). Preparation and properties of nonleaching antimicrobial linear low-Density polyethylene films. Industrial & Engineering Chemistry Research, 54, 1824–1831. Wang, H. Y., & Huang, M. F. (2007). Preparation: characterization and performances of biodegradable thermoplastic starch. Polymers for Advanced Technologies, 18, 910–915. Wei, D., Wanga, H., Ziaee, Z., Chibante, F., Zheg, A., & Xiao, H. (2016). Non-leaching antimicrobial biodegradable PBAT films through a facile and novel approach. Materials Science and Engineering: C, 58, 986–991.