Towards the development of eco-friendly disposable polymers: ZnO-initiated thermal and hydrolytic degradation in poly(L-lactide)/ZnO nanocomposites

Abstract

In this work poly(L-lactide)/ZnO nanocomposites with homogeneously distributed nanoparticles have been fabricated by a solvent-precipitation method. The obtained nanocomposites have been submitted to thermal and hydrolytic degradation processes in order to elucidate the catalytic effect of ZnO nanoparticles. The resulting degradation products from the thermally-initiated catalysis have been identified by Fourier transform infrared spectroscopy (FTIR). It has been found that the presence of ZnO gives rise to a 65-fold increase in the formation of CO₂ in comparison to acetaldehyde. FTIR results of hydrolytically degraded nanocomposites show an increased amount of carboxylic acid groups as ZnO concentration increases, while the C=O stretching band splitting denoted larger crystalline regions as degradation proceeds. The obtained results are explained from the viewpoint of lattice oxygen vacancies in ZnO. Upon thermodegradation nanoparticles initiate unzipping depolymerization/intermolecular transesterification reactions in PLLA, while during hydrolytic degradation H₂O is dissociated on oxygen vacancy sites, giving hydroxyl groups that initiate the hydrolysis of ester bonds, and thus reducing PLLA to soluble monomers. The obtained findings are expected to allow the development of eco-friendly disposable polymeric waste by opening new possibilities in the use of naturally-available materials as efficient catalysts for feedstock recycling of biopolymers by common chemical processes.

---

Introduction

Nowadays there is an increasing interest from both academia and private industry towards the development of a sustainable society. In this trend of obtaining sustainable materials, especially interesting is the development of biopolymer-based parts to be used in the packaging industry, where the use of non-biodegradable materials for short-term use applications is very widespread. In this sense, biodegradable polymers such as poly(L-lactide) (PLLA) have been gaining both scientific and commercial significance due to their inherent physicochemical characteristics and their renewable character.^1,2 Besides fulfilling the requirements of the traditionally available thermoplastics such as polyethylene terephthalate (PET), polystyrene (PS) and polypropylene (PP), their economically feasible industrial production makes polylactides interesting candidates to replace current petroleum-based plastics.^3,4

At the present time, the feedstock recycling of biopolymers by common chemical processes remains insufficiently explored and is still open to debate. Thereby, it results highly desirable to develop industrial processes in which the plastic recycling would be carried out at low cost and in a controllable way. In this framework, Fan et al. have shown that alkali earth metal oxides such as calcium oxide (CaO) and magnesium oxide (MgO) selectively produce L,L-lactide and lowered thermal degradation temperature, which would serve for the feed stock recycling of PLLA.⁵ In a similar work, Motoyama et al. found that the catalytic activity of MgO markedly increases in presence of smaller particles because of the increase in the catalytic surface area.⁶ In view to those results, in this work we use more economically feasible catalyst particles as a cost-effective approach to reduce PLLA and in that way develop eco-friendly disposable materials. Among all the available materials, zinc oxide nanoparticles (ZnO) are of special interest since they present higher catalytic efficiency in regard to other materials in both aqueous and air media.^7–9 Since catalytic reactions usually take place at the particle surfaces, the use of large surface area nanorods is desirable for enhancing the activity of zinc oxide.¹⁰ At the same time, by using biocompatible zinc oxide nanoparticles UV-absorbing films with high visible transparency would be achieved,¹¹ preventing food photodegradation when PLLA is used as packaging material.^12–15

Spectroscopic and non-destructive techniques such as Fourier transform infrared spectroscopy (FTIR) are of key relevance to understand degradative processes of polymers since they allow identifying compounds while enable the determination of crystalline phases when intermolecular interactions take place.¹⁶ In this sense, the degradation mechanism induced by the presence of ZnO nanoparticles could be identified by the FTIR analysis of the evolved gases resulting from the chain cleavage of the PLLA/ZnO nanocomposites. Moreover, the analysis of the carbonyl (C=O) stretching band, which remains almost uncoupled from the vibrational modes of the backbone chain,¹⁷ would provide information about the conformational changes occurring during the hydrolytic degradation of PLLA-based materials.

ZnO nanoparticles have been previously used to improve the photocatalytic activity of thermoplastic polymers such as polypropylene (PP).¹⁸ However, to the best of our knowledge, no previous work have been devoted to the systematic study of the reactions occurring in both thermally and hydrolytically catalyzed reactions in biopolymers in presence of ZnO. Thereby, in this work we attempt to elucidate how the presence of ZnO nanoparticles affects the resulting catalytic processes of PLLA, which would help the packaging industry to develop new disposable materials. FTIR has been carried out to identify thermodegradation compounds together with the conformational changes occurring during the hydrolytic degradation of PLLA-based materials. According to the 12 Principles of Green Chemistry,¹⁹ the development of renewable PLLA-based nanocomposites with well-dispersed ZnO nanoparticles would open new insights towards the development of eco-friendly disposable polymeric waste.

Experimental

Materials and sample preparation

PLLA with a number-average molecular weight (M_n) of 100.000 g mol⁻¹ and a polydispersity index (M_w/M_n) of 1.85 (see molecular weight distribution in the ESI, Fig. S1†) has been supplied by Purac Biochem. Phosphate buffered saline (PBS) tablets were supplied by Sigma Aldrich, chloroform (reagent ≥ 99.8%) was purchased from LabScan and methanol (reagent ≥ 99.5%) was purchased by Panreac. Zinc oxide (ZnO) nanoparticles, with a purity of ≥99.5% according to obtained X-ray diffraction pattern (Fig. 1a), have been kindly purchased by L'Urederra technological centre (Spain).

Fig. 1 WAXD pattern of ZnO nanoparticles (a) and representative transmission electron microscopy (TEM) image showing the hexagonal structure of ZnO nanoparticles at a 230.000× magnification (b).

Samples have been prepared by solvent-precipitation followed by compression moulding. Firstly, ZnO nanoparticles were homogeneously dispersed in chloroform (CHCl₃) via mild-sonication (20% output for 1 minute) in a Vibra-Cell™ CV 334 ultrasonic processor. Finely dispersed nanoparticles were added to dissolved PLLA (in chloroform) to obtain ZnO concentrations ranging from 0.05 wt% to 5 wt%. To obtain a good dispersion of nanoparticles within the polymer, PLLA–ZnO dispersions were submitted to an additional sonication step for 5 minutes before precipitating them in an excess of methanol. This method ensures a homogeneous distribution of nanoparticles (NPs) within the PLLA matrix. Once resulting materials were dried during 48 h at 60 °C in a vacuum-oven, films with a thickness of 80 ± 10 μm were fabricated in a hydraulic hot press by compression moulding at 200 °C for 4 minutes under a pressure of 150 MPa.

Wide angle X-ray diffraction (WAXD)

Room temperature wide angle X-ray diffraction (WAXD) has been conducted in order to elucidate the crystalline structure of ZnO nanoparticles and PLLA/ZnO nanocomposites during their hydrolytic degradation. X-ray powder diffraction patterns were collected in a PHILIPS X'PERT PRO automatic diffractometer in theta–theta configuration, secondary monochromator with Cu-Kα radiation (λ = 1.5418 Å) and a PIXcel solid state detector. The sample was mounted on a zero background silicon wafer fixed in a generic sample holder. Data were collected from 20 to 75° 2θ (step size = 0.026) at RT. The presence of oxygen vacancies in the ZnO lattice has been determined by WAXD measurements on a Bruker D8 Advance diffractometer operating at 30 kV and 20 mA, equipped with a Cu tube (λ = 1.5418 Å), a Vantec-1 PSD detector, and an Anton Parr HTK2000 high-temperature furnace. The powder patterns were recorded in 2θ steps of 0.0167° in the 10 ≤ 2θ ≤ 60 range, counting for 2 s per step (total time 2 h). Data sets were recorded at 30, 200, 225 and 250 °C with a 0.2 °C s⁻¹ heating rate.

Morphological characterization

ZnO nanoparticles have been analyzed by transmission electron microscopy (TEM). A droplet of diluted ZnO suspension (0.1% (w/w)) in distilled water was deposited on a carbon-coated grid. TEM was carried out using a Philips CM120 Biofilter apparatus with STEM module at an acceleration voltage of 120 kV. Cryofractured surfaces of nanocomposites have been analyzed in a Hitachi S-4800 field emission scanning electron microscope (FE-SEM) at an acceleration voltage of 5 kV. Surfaces were chromium-coated in a Quorum Q150T ES turbo-pumped sputter coater (5 nm thick coating).

Thermogravimetric analysis (TGA)

Thermo-degradative behaviour of nanocomposites was studied by means of Thermal Gravimetric Analysis (TGA METTLER TOLEDO 822e) in alumina pans by heating the samples from room temperature to 500 °C at 10 °C min⁻¹ with nitrogen flux of 50 ml min⁻¹. Thermodegradation products (exhaust gases upon heating the samples) were analyzed using a Thermo Nicolet Nexus 670 FTIR spectrophotometer connected through an interface to DTG-60 Shimadzu thermobalance. Neat PLLA and its PLLA/ZnO 1 wt% nanocomposite were heated from room temperature to 500 °C at 10 °C min⁻¹ under nitrogen atmosphere. FTIR scans were taken every 16 s.

Contact angle measurements

Water was used as the probe liquid for the determination of hydrophobicity at the nanocomposite surface. Measurements were carried out by sessile drop method (2 μL per drop at a rate of 2 μL s⁻¹) using a Neurtek Instruments OCA 15 EC at room temperature. The average values were calculated using six measurements of each composition.

Hydrolytic degradation

Neat PLLA and its ZnO nanocomposites (square-shaped; 1 cm²) were hydrolytically degraded in an oven at 80 °C. 80 μm thick samples with a surface area to volume ratio of 0.2 cm⁻¹ were immersed in a PBS solution (PBS tablets dissolved in Milli-Q water; pH = 7.4) and they were removed after 1, 3, 6, 10, 12, 18 and 25 days. The pH evolution of the medium has been monitored using a 691 pH meter (Metrohm).

Fourier transform infrared spectroscopy (FTIR)

Infrared spectra in transmission KBr mode were recorded on a Thermo Nicolet Nexus 670 FTIR spectrophotometer. Samples were prepared by spreading two drops of a 2 wt% solution over KBr pellets. The solution was dried in vacuum at 60 °C for 24 h before characterization. Each IR spectrum consisted of 64 scans taken in the range 400–4000 cm⁻¹ with a resolution of 2 cm⁻¹. The average values over three replicates are reported for each material.

Results and discussion

ZnO nanoparticle characterization

Both X-ray diffraction (WAXD) and Transmission Electron Microscopy (TEM) have been carried out to determine the structure of ZnO nanoparticles, which affects the physicochemical properties of PLLA/ZnO nanocomposites. As depicted in Fig. 1a, WAXD data exhibits featured reflections of hexagonal crystal system of ZnO, known as wurtzite (P63mc space group, with crystal lattice dimensions shown in the insert).²⁰ TEM micrographs confirm the nanoscale nature of ZnO particles, which present a rod-shaped structure of 15–30 nm wide and 25–85 nm long.

Thermal degradation of PLLA and PLLA/ZnO nanocomposites

PLLA is one of the most sensitive thermoplastic to high temperatures in comparison with other polyesters such as poly(ε-caprolactone) (PCL) or poly(L-lactide-co-ε-caprolactone) (PLCL).²¹ Fig. 2 shows thermogravimetric (a) and weight loss rates (b) of neat PLLA and its ZnO nanocomposites with a concentration up to 5 wt%, being their characteristic thermodegradation temperatures summarized in Table 1. Results show a continuous reduction on the thermal stability because of the catalytic behaviour of ZnO.

Fig. 2 Thermogravimetric traces (a) and weight lost rates (or derivative TGA curve) (b) of PLLA/ZnO nanocomposites as a function of ZnO concentration.

Table 1 Characteristic thermodegradation temperatures (T_5% and T_peak) and maximum degradation rates (α_max) of PLLA/ZnO nanocomposites

ZnO wt%	0	0.05	0.1	0.2	0.5	1	2	5
T5% (°C)	340.2	328.3	302.2	279.8	271.1	264	258.2	243.4
T10% (°C)	347.2	337.8	310.9	288.2	280	273.4	266.2	252.9
Tpeak (°C)	366.8	363.5	348.2	331.5	312.9	306.7	295.4	282.5
FWHM (°C)	21.9	24	26.3	30.5	21.6	21.9	20.4	18.5

---

As shown in Table 1, thermal degradation of unreinforced polymer begins (T_{5 wt%}) at 340.2 °C and reaches its maximum rate (T_peak) at 366.8 °C. Conversely, PLLA/ZnO 5 wt% nanocomposite begins to degrade almost 100 °C below, at 243.4 °C, to attain its maximum rate at 282.5 °C (note that the char residue increases with NP loading due to the high thermal stability of ZnO). As could be seen in Fig. 2b, while the derivative TGA curve (DTG) of neat PLLA shows a bell-shaped curve (representing a single degradation step), nanocomposites containing 0.1 wt% and 0.2 wt% ZnO present a shoulder in their weight lost rates located at temperatures of 307 °C and 294 °C respectively. Accordingly, the maximum weight lost rate decreases with increasing ZnO content from 3.48% per °C for neat PLLA to a minimum value of 2.63% per °C for its 0.2 wt% nanocomposite, which reflects slower degradation rate for these compositions. This fact is accompanied by wider curve as denoted by a full width at half maximum (FWHM) of 21.9 °C for neat polymer in regard to 30.5 °C obtained in the case of 0.2 wt% counterpart. At larger concentrations, both curve-shape and FWHM reach similar values to those obtained for PLLA. It should be noted that the first degradation step in nanocomposites containing low ZnO concentration;²² i.e. 0.1 and 0.2 wt%, may be related to the thermodegradation of PLLA close to ZnO surfaces, the second stage, achieved at higher temperatures, could be associated with the degradation of PLLA that is far from nanoparticles. This behaviour has been previously found in several processes of polymer-based nanocomposites involving chain mobility, where the presence of nanoparticles results in the development of kinetically different phases (for example, in processes concerning crystallization,^23,24 structural relaxation and mechanical reinforcement).^25,26

On the contrary, at large nanoparticle concentration, the increased exposition of L-lactide units to ZnO surfaces results in the presence of a single degradation step, in which thermodegradation proceeds even in a narrower temperature-range than in the case of neat PLLA as confirmed by a FWHM of 18.5 °C for 5 wt% nanocomposite counterpart with respect to 21.9 °C in the case of neat polymer. This remarkable continuous decrease in thermal stability of PLLA with ZnO concentration, especially up to 0.2 wt%, may arise from the fact that metallic nanoparticles such as Fe, Al, Zn and Sn catalyze depolymerization reactions of the matrix, boosting the occurring thermodegradation process.^27–29 Indeed, ZnO nanoparticles coordinate PLLA backbone ester groups, accelerating the unzipping depolymerization reactions as would be further discussed in the following section.³⁰

In order to further understand the catalytic effect of ZnO nanoparticles during the heating of PLLA, the resulting degradation products from the PLLA thermodegradation have been analyzed by FTIR. The spectra corresponding to exhaust gases obtained during the thermal degradation of neat PLLA and PLLA/ZnO 1 wt% nanocomposite are shown in Fig. 3. It should be pointed out that these spectra correspond to 10, 50, 80 and 95 wt% weight loss of the samples, thus, and according to TGA thermograms shown above, the temperatures at which have been collected differ substantially from each another. The main degradation products obtained during the thermal degradation of neat PLLA ((C₃H₄O₂)_n) could be identified as acetaldehyde (C₂H₄O; 2733 cm⁻¹, 1759 cm⁻¹, 1370 cm⁻¹ and 1106 cm⁻¹ bands), carbon monoxide (CO; two sharp peaks at 2176 cm⁻¹ and 2120 cm⁻¹), and carbon dioxide (CO₂; three absorption bands positioned at 2362 cm⁻¹, 2334 cm⁻¹ and 669 cm⁻¹). Furthermore, small traces of L-lactic acid (C₃H₆O₃; narrow peak at 1739 cm⁻¹ and a broad band in 1000–1500 cm⁻¹ range with maximum at 1457 cm⁻¹, 1212 cm⁻¹, 1128 cm⁻¹ and 1045 cm⁻¹) and low-molecular-weight by-products such as formic acid (CH₂O₂; main three bands at 1700–1820 cm⁻¹, 1050–1130 cm⁻¹ and 550–750 cm⁻¹) could be observed. The spectra corresponding to the gaseous products evolved from neat PLLA and PLLA ZnO 1 wt% at weight loss of 40% and their assignments are depictured in the Fig. 4 (enlarged spectra within the 1600–2000 cm⁻¹ region is provided in the ESI, Fig. S2†). These observations agree with the investigations carried out by McNeil and Leiper,³¹ who found that PLLA chain cleavage produces cyclic oligomers, lactide, acetaldehyde, carbon monoxide and carbon dioxide.

Fig. 3 FTIR spectra of thermodegradation products obtained of neat PLLA and PLLA/ZnO 1 wt% at different weight loss percent: 10, 50, 80 and 95 wt%.

Fig. 4 FTIR spectra of the evolved gases of neat PLLA and PLLA/ZnO 1 wt% at 40 wt%, recorded at 294 °C and 357 °C, respectively.

Additionally, during the last part of thermodegradation process water is produced as a consequence of L-lactic acid oligomer fragmentation.³² The results obtained from the analysis of the evolved gases upon thermal decomposition suggest that the degradation of the PLLA/ZnO nanocomposites combines unzipping depolymerization and random transesterification processes. In FTIR spectra of PLLA/ZnO 1 wt% nanocomposite enlarged for 1000–2000 cm⁻¹ region (Fig. S3†) an overlapped presence of aldehyde and L-lactic could be observed. The observation of acetaldehyde and formic acid degradation exhausts corresponds to a transesterification process. Besides, beyond 300 °C a characteristic lactide peak is observed at 940 cm⁻¹ in Fig. S3,† indicating an unzipping depolymerization reaction. In the Scheme 1 a possible reaction mechanisms for PLLA/ZnO nanocomposites is proposed. These results agree with the previous studies of PLLA/metal oxide composites with MgO, CaO and SnO,^5,6,32,33 in which a catalytic effect of these oxides have been analysed. The degradation of PLLA catalyzed by ZnO nanoparticles present similarities to MgO catalyzed process since lactides are present at higher temperature than 250 °C observed for CaO.^5,6 On the other hand, at Sn-catalyzed degradation a random transesterification have been observed.

Scheme 1 Expected degradation mechanism in PLLA/ZnO nanocomposites.

Although in both PLLA and PLLA/ZnO 1 wt% degradation processes the same products are formed, there are considerable divergences when considering the relative intensity of absorption bands. Indeed, while acetaldehyde, carbon dioxide and carbon monoxide are formed from the beginning of PLLA thermodegradation (Fig. 3) and their intensity-ratios remains fairly unchanged over the entire decomposition reaction, thermodegradation products of PLLA/ZnO 1 wt% nanocomposite strongly depends on the conversion rate. Table 2 shows the ratio between evolved acetaldehyde and CO₂ calculated from the intensity ratio between absorption bands located at 1759 cm⁻¹ and 2362 cm⁻¹. In this sense, it can be clearly observed that acetaldehyde represents the main degradation product of neat polymer, while in presence of 1 wt% of ZnO nanoparticles much larger quantities of CO and CO₂ are evolved during the thermodegradation process. This is in line with the work developed by Saleh et al.,³⁴ where it has been showed that ZnO generates ˙OH radicals that could react with gaseous acetaldehyde (CH₃CHO) to generate carbon dioxide and water. From the beginning of the PLLA degradation, acetaldehyde/CO₂ ratio is found to be about 5, which gradually increases up to 10.54 when 70 wt% has been degraded (at 367 °C). Then, an increase of temperature up to 385 °C reduces the acetaldehyde/CO₂ ratio to 5.67. At 95 wt% loss mainly formic acid is generated, which is significantly less volatile compound than acetaldehyde. On the contrary, as denoted by the increase of the absorption bands located at 2362 cm⁻¹ and 2334 cm⁻¹, CO₂ yield dramatically rises in presence of ZnO nanoparticles, being the amount of evolved CO₂ much larger than that of acetaldehyde (acetaldehyde/CO₂ ratio is about 0.08) during the initial stages of thermodegradation, which represents a 65-fold change in comparison with neat polymer.

Table 2 Evolved acetaldehyde/CO₂ ratio of neat PLLA and PLLA/ZnO 1 wt% nanocomposite as depending on the mass loss

ZnO concentration	Mass loss (wt%)
	20	30	40	50	60	70	80	90	95
0 wt%	5.23	5.78	6.82	9.31	9.79	10.54	10.27	8.91	5.67
1 wt%	0.08	0.09	0.13	0.21	0.31	0.48	1.06	1.04	0.73

---

As shown in Fig. S4,† concerning to the volatility of compounds it is important to note that, for the same temperature, the vapour pressure of CO₂ is 42 753 mmHg, being the vapour pressure of acetaldehyde 57 times smaller (42 753 mmHg vs. 740 mmHg).³⁵ Thus, according to Raoult's law it is reasonable to suppose that the primary reason of the reduced thermal stability of PLLA/ZnO nanocomposites regarding neat PLLA arises from the generation of more volatile thermodegradation products in the presence of ZnO nanoparticles.³⁶ To the best of our knowledge, this is the first report in which TGA-FTIR measurements show that the presence of nanofillers modifies the thermodegradation products of PLLA. The developed CO and CO₂ gases could be used for active packaging applications since they could effectively extend the self-life of packed foods. Indeed, FDA has approved the use of carbon monoxide as primary packaging system,³⁷ while carbon dioxide could be used as oxygen scavenger to remove, or at least decrease the amount of oxygen gas into the package.³⁸ Moreover, it has been proven that CO₂ limits the microbial growth in foods such a fish, cheese and meat.³⁹ Another interesting application is the development of methanol as an alternative to petro-based fuels since it could be produced by the hydrogenation of both carbon monoxide and carbon dioxide (for example, by using bimetallic Cu–ZnO and Pd–Cu catalysts).⁴⁰

Surface hydrophobicity

It is well known that the affinity between the sample and the solvent affects the resulting hydrolytic degradation behaviour of materials. In this sense water contact angle (WCA) measurements were carried out to further elucidate the occurring degradation mechanism. As illustrated in Fig. 5, an increased surface hydrophobicity with the addition of ZnO from 81.2 ± 3.9° for neat polymer to 90.8 ± 2.4° for 5 wt% nanocomposite has been obtained. This behaviour has been previously reported for different polymer/ZnO nanocomposites.⁴¹

Fig. 5 Representative images of a water drop at the surface of neat PLLA (left) and 5 wt% nanocomposite film (right).

Hydrolytic degradation

It is well established that polyesters such as poly(L-lactide) undergo their hydrolytic degradation through random-scission of ester linkages.⁴² Since PLLA could be considered as a heterogeneous material where amorphous and crystalline domains present different molecular mobility,^43,44 in order to avoid the effect introduced by the crystalline phases on the resulting hydrolytic degradation of PLLA/ZnO nanocomposites (crystalline phases present an increased resistance to hydrolytic degradation in regard to amorphous ones),⁴² samples with a crystallinity degree around 11–13% (see corresponding Differential Scanning Calorimetry (DSC) traces in ESI, Fig. S5†) have been submitted to an accelerated hydrolytic degradation at 80 °C in phosphate buffered saline (PBS). Fig. 6a shows the physical aspect of PLLA/ZnO nanocomposites once they were removed from PBS solution (after 24 h at 80 °C) and after being completely dried. It could be observed that nanocomposites containing larger ZnO concentrations undergo physical disintegration to a larger extent than neat PLLA. This fact could be associated with a ZnO-initiated hydrolysis mechanism, in which metallic nanoparticles accelerate the ester-bond hydrolysis of PLLA. To further confirm the catalytic effect of zinc oxide, as shown in Fig. 6b, the pH evolution of the medium for a composition range of 0-2 wt% over 10 days of degradation at 80 °C has been measured (all the nanocomposite films have been immersed with a surface area to volume ratio of 0.2 cm⁻¹ at a starting pH of 7.4). Since the pH decrease over the time is due to the release of acidic by-products that have been expelled to the outside medium during the degradation, the determination of the pH of the medium present in Falcon tubes would allow us to determine the extent of hydrolytic degradation. It could be seen that for all the compositions, after 10 days the pH of the medium decreases from 7.4 up to 3.5–3.8, although this decrease progressively occurs at shorter times when increasing nanoparticle content.

Fig. 6 Photograph showing physical disintegration of PLLA/ZnO films with concentrations up to 5 wt% after 24 h in PBS medium at 80 °C (a) and pH evolution of the PBS degradation medium over 10 days at 80 °C. Nanocomposites with a concentration up to 2 wt% have been submitted to hydrolytic degradation (b).

For instance, after one day, the pH decreases from 7.4 to 5.6 with the addition of 1 wt% of nanoparticles, while the sample containing neat PLLA present a pH value of 6.5, suggesting that the presence of ZnO accelerates the ester hydrolysis of the PLLA matrix. Moreover, the released acidic products are accumulated in the PBS solution and further autocatalyze the degradation by the generated carboxylic end-groups,⁴⁵ speeding up the overall degradation kinetic. For instance, after three days the pH difference of the solution containing neat PLLA and its 1 wt% nanocomposite is even larger than that achieved after one day (1.7 vs. 0.7). After the sixth day, all the nanocomposites have been almost completely diffused into the medium, reaching a nearly constant pH value of 3.5–3.9 depending on the ZnO concentration. As reported by Höglund et al.,⁴⁶ oligomers become water-soluble when the molar mass falls below 1000 g mol⁻¹, which corresponds to 13 repeating LA units.⁴⁷ Resulting lactic acid could be again easily converted into PLLA, closing the loop towards a greener production of biopolymers.

To better understand how this catalytic reaction takes place in PLLA/ZnO nanocomposites, as depicted in Fig. 7a, samples have been analyzed by Fourier transform infrared spectroscopy (FTIR). FTIR spectra have been normalized to the characteristic PLLA carbonyl absorption band located at 1759 cm⁻¹. Although they present similar structures, some differences could be observed in their corresponding FTIR spectra. During the hydrolytic degradation of PLLA/ZnO nanocomposites, as a result of the scission of the main chain ester bonds, shorter chains would be found, yielding more carboxylic acid and alcohol groups per mass sample regarding non-degraded sample. Thus, the relative ratio between the ester bonds and end functional groups, i.e., carboxylic acid/alcohol groups, could indicate the extent of the degradation process. Upon degradation the intensity of –CH stretching (achieved at 2997 cm⁻¹ and 2945 cm⁻¹) and –CO stretching (1268 cm⁻¹) absorption bands decrease as ZnO concentration increases, indicating a reduction on the amount of ester bonds.^48,49 A shoulder centred at 1715 cm⁻¹ appears in the case of neat PLLA and its 0.05 wt% nanocomposite, which signifies an increased amount of ester groups in comparison with highly filler nanocomposites.⁵⁰ In addition, the band centred at 922 cm⁻¹, which corresponds to a combination of CH₃ rocking mode of PLLA crystals and C–C backbone notably decreases its intensity as ZnO concentration increases,⁵¹ i.e., as degradation further evolves.

Fig. 7 FTIR spectra of PLLA/ZnO nanocomposites hydrolytically degraded for 24 h in PBS medium at 80 °C (a); carbonyl (C=O) stretching region of hydrolytically degraded PLLA/ZnO nanocomposites (b) and its corresponding second derivative (c). The insert in part a shows obtained FWHM values.

As a result of the progressive chain-shortening during hydrolytic degradation, the achieved decrease in molecular entanglement together with increased chain mobility boost the development of highly-ordered crystalline phases.⁵² More precisely, wide angle X-ray diffraction patterns of hydrolytically degraded nanocomposites (Fig. S6a†) denote a progressive increase on the crystalline fraction in regard with amorphous phase as ZnO concentration increases as denoted by a decrease on the amorphous halo at higher ZnO concentrations. Additionally, DSC results (Fig. S6b†) show lower melting temperature of nanocomposites in presence of zinc oxide together with an increase on the X_c crystallinity degree from 50.1% for neat polylactide to 69.1% for its 5 wt% nanocomposite. Those results are in agreement with obtained FTIR spectra and indicate an accelerated chain-scission of polylactide macromolecules in presence of zinc oxide nanoparticles.

Furthermore, it has been previously demonstrated that the carbonyl (C=O) stretching region of polylactides is highly sensitive to conformational changes.^17,53 Although from a theoretical point of view crystallization should not modify the location of absorption bands, their intensity changes because of the induced changes in refractive index and density by new crystalline regions.⁵⁴ Indeed, the increased molecular order in those ordered/semi-ordered domains results in narrower IR bands located at the same wave-number than the original band.¹⁶ Accordingly, as shown in Fig. 7b, the FWHM of the carbonyl band continuously decreases from 27.7 cm⁻¹ in the case of neat PLLA to 13.8 cm⁻¹ for its PLLA/ZnO 2 wt% nanocomposite counterpart. It is worthy to note that, as occurs with the pH values (see Fig. 6b), the main changes are achieved at concentrations up to 1 wt%, suggesting that particle aggregation occurs at larger concentrations that 1 wt%. This fact has been confirmed by FE-SEM micrographs displayed in Fig. 8, where it could be observed that concentrations of 0.5–1 wt% ZnO nanoparticles remain well dispersed within the matrix, whilst nanocomposite containing 5 wt% presents large ZnO aggregates.

Fig. 8 Representative FE-SEM micrographs showing ZnO dispersion in PLLA matrix for concentrations of 0.5 wt% (a), 1 wt% (b) and 5 wt% (c).

It is reported that the splitting in the second derivative of the C=O stretching band of PLLA occurs as a result of ordered phases with different chain conformation. As demonstrated in Fig. 7c, the second derivative for hydrolytically degraded PLLA/ZnO nanocomposites presents four main components, which arise from the presence of four distinct conformations occurring in PLLA known as gauche–gauche (gg), trans–gauche (tg), gauche–trans (gt) and trans–trans conformers, which absorb respectively at 1777, 1767, 1759 and 1749 cm⁻¹.⁵⁵ It is suggested that the relative areas of those peaks depend on the relative population of the conformers. Thereby, it could be concluded that the fraction concerning to gt conformer (showing a narrow derivative peak) notably increased with the addition of ZnO nanoparticles, together with a decrease in the amount of amorphous gg conformers (absorbance at 1776 cm⁻¹).⁵⁶ This gt conformer has the lowest energy with a 10³ helical chain conformation,⁵⁷ denoting that PLLA/ZnO nanocomposites crystallize into α crystal form during the hydrolytic degradation. Those data suggest that as ZnO fraction increases, the initially amorphous nanocomposites evolve towards more crystallized materials as a result of the molecular weight reduction. This crystallinity increase ascribed to a molecular weight reduction provides further evidence about the accelerated hydrolytic degradation in presence of ZnO nanoparticles.

We hypothesize that the degradation reactions are initiated by ZnO nanoparticles. In this sense, wide angle X-ray diffraction experiments have been carried out to investigate the structural changes occurring in ZnO nanoparticles during the degradation of nanocomposites. WAXD patterns of PLLA/ZnO 5 wt% nanocomposite upon degradation are shown in Fig. 9. The reflections corresponding to ZnO (100), (002) and (101) planes of non-degraded nanocomposite show a shoulder (see Fig. S7† for deconvoluted pattern), which has been recently ascribed to the oxygen vacancies present in the lattice.^58,59 During the degradation of nanocomposite this shoulder disappears and diffraction peaks are continuously shifted towards lower 2θ angles, denoting a steady increase of the unit cell parameters. This expansion of the unit cell is attributed to a decrease of the residual stress induced by the decrease of oxygen vacancies in the lattice.⁶⁰

Fig. 9 Wide angle X-ray diffraction patterns of PLLA/ZnO 5 wt% nanocomposite upon degradation.

In the light of obtained X-ray diffraction and spectroscopic results arising from both thermal and hydrolytic degradation, and taking into account the increased surface hydrophobicity of nanocomposite films, a model concerning the catalytic mechanism in PLLA/ZnO nanocomposites is proposed in Fig. 10. In overall, it could be concluded that biopolymer degradation is initiated by zinc oxide nanoparticles. Upon thermodegradation ZnO nanoparticles fill their vacancies with O atoms originating from polylactide macromolecules, leading to unzipping depolymerization and intermolecular transesterification reactions as shown in Scheme 1. On the contrary, hydrolytic degradation proceeds when H₂O is dissociated on oxygen vacancy sites.^61,62 Thus, two hydroxyl groups for each oxygen atom are generated, which initiates the degradation of polymer host by attacking adjacent PLLA chains.

Fig. 10 Schematical representation showing the mechanism of ZnO-initiated degradation.

The increased degradation kinetics with ZnO concentration may be explained in terms of polymer/filler interfacial features since the reported catalysis changes are related to the amount of available ZnO surfaces. As zinc oxide concentration raises the formation of vats interfacial regions are enough to boost the overall degradation reaction of bulk PLLA because the occurring reactions at ZnO surfaces. Larger concentrations than 0.5–1 wt% result in an increase of direct contacts between adjacent nanoparticles, leading to the formation of bundles that reduce ZnO–PLLA interfacial area per single nanoparticle, which notably limits the catalyzing effect of ZnO.

Conclusions

In this work PLLA/ZnO nanocomposites with homogeneously dispersed nanoparticles have been prepared. TGA results show that during thermodegradation process of nanocomposites ZnO nanoparticles dramatically lower the thermal degradation temperature by 100 °C, being acetaldehyde and CO₂ the main degradation products of neat PLLA and PLLA/ZnO nanocomposites respectively. It is shown that the degradation of the PLLA/ZnO nanocomposites combines unzipping depolymerization and transesterification processes. The accumulation of acidic degradation products in PBS proves that hydrolytic degradation proceeds noticeably faster as ZnO concentration increases. In addition, FTIR spectra demonstrate the presence of shorter polymeric chains when in presence of zinc oxide, with more carboxylic acid groups in comparison to ester groups, suggesting a more marked backbone scission at ester bonds. In addition, according to the splitting in the second derivative of the C=O stretching band, larger crystalline regions are developed as a result of faster hydrolytic degradation as ZnO concentration increases.

WAXD experiments demonstrate that the amounts of oxygen vacancies present in the ZnO lattice are reduced upon degradation. We hypothesize that both thermal and hydrolytic degradation are initiated at ZnO surfaces, leading to unzipping depolymerization/intermolecular transesterification reactions upon thermodegradation or dissociated hydroxyls which initiated the hydrolysis of ester bonds during the hydrolytic degradation. Thereby, the lactic acid resulting from the hydrolytic degradation could be again easily converted into PLLA, closing the loop towards a greener production of biopolymers. It is expected that those findings would open new insights in the use of naturally-available materials as efficient catalyst for the recycling of biopolymers towards the development of eco-friendly disposable polymeric waste.

Acknowledgements

E. L. thanks the University of the Basque Country (UPV/EHU) for a postdoctoral fellowship. In addition, we gratefully acknowledge Corbion-Purac for the kind donation of PLLA polymer. Authors thank the Basque Country Government for financial support (Ayudas para apoyar las actividades de los grupos de investigación del sistema universitario vasco, IT718-13).

Notes and references

1. R. Auras, B. Harte and S. Selke, Macromol. Biosci., 2004, 4, 835–864.
2. R. E. Drumright, P. R. Gruber and D. E. Henton, Adv. Mater., 2000, 12, 1841–1846.
3. P. Gruber and M. O'Brien, Polylactides NatureWorks™ PLA, Biopolymers in 10 Volumes, Wiley-VCH, Weinheim, 2002.
4. X. Helan, and Y. Yang, Degradable Polymers and Materials: Principles and Practice, ACS Symposium Series, 2nd edn, 2012.
5. Y. Fan, H. Nishida, T. Mori, Y. Shirai and T. Endo, Polymer, 2004, 45, 1197–1205.
6. T. Motoyama, T. Tsukegi, Y. Shirai, H. Nishida and T. Endo, Polym. Degrad. Stab., 2007, 92, 1350–1358.
7. J. Huang, X. Xu, C. Gu, W. Wang, B. Geng, Y. Sun and J. Liu, Sens. Actuators, B, 2012, 173, 599–606.
8. A. M. Ali, E. A. C. Emanuelsson and D. A. Patterson, Appl. Catal., B, 2010, 97, 168–181.
9. R. Wahab, S. K. Tripathy, H. S. Shin, M. Mohapatra, J. Musarrat, A. A. Al-Khedhairy and N. K. Kaushik, Chem. Eng. J., 2013, 226, 154–160.
10. B. Cheng and E. T. Samulski, Chem. Commun., 2004, 227, 986–987.
11. F. Werner, J. F. Gnichwitz, R. Marczak, E. Palomares, W. Peukert, A. Hirsch and D. M. Guldi, J. Phys. Chem. B, 2010, 114, 14671–14678.
12. E. Lizundia, L. Ruiz-Rubio, J. L. Vilas and L. M. León, J. Appl. Polym. Sci., 2015, 132, 42426.
13. M. Mathlouthi, Food Packaging and Preservation, Blackie Academic, London 1994.
14. Y. Q. Li, S. Y. Fu and Y. W. Mai, Polymer, 2006, 47, 2127–2132.
15. B. M. Novak, Adv. Mater., 1993, 5, 422–433.
16. P. C. Painter, M. Coleman and J. L. Koening, Acta Polym., 1983, 34, 66.
17. E. Meaurio, I. Martinez de Arenaza, E. Lizundia and J. R. Sarasua, Macromolecules, 2009, 42, 5717–5727.
18. J. C. Colmenares, E. Kuna, S. Jakubiak, J. Michalski and K. J. Kurzydłowski, Appl. Catal., B, 2015, 170–171, 273–282.
19. P. T. Anastas and J. C. Warner, Green chemistry: theory and practice, Oxford University Press, New York, 1998.
20. H. Schulz and K. H. Thiemann, Solid State Commun., 1979, 32, 783–785.
21. A. Larrañaga and J. R. Sarasua, Polym. Degrad. Stab., 2013, 98, 751–758.
22. T. Ozawa, J. Therm. Anal. Calorim., 1970, 2, 301–324.
23. J. Loosa, T. Schimanski, J. Hofman, T. Peijs and P. J. Lemstr, Polymer, 2001, 42, 3827–3834.
24. A. Dasari, Z. Z. Yu and Y. W. Mai, Macromolecules, 2007, 40, 123–130.
25. E. Lizundia and J. R. Sarasua, Macromol. Symp., 2012, 321–322, 118–123.
26. J. N. Coleman, M. Cadek, K. P. Ryan, A. Fonseca, J. B. Nagy, W. J. Blau and M. S. Ferreira, Polymer, 2006, 47, 8556–8561.
27. E. Lizundia, J. L. Vilas and L. M. León, Carbohydr. Polym., 2015, 123, 256–265.
28. D. Cam and M. Marucci, Polymer, 1997, 38, 1879–1884.
29. J. Fernandez, A. Etxeberria and J. R. Sarasua, Polym. Degrad. Stab., 2013, 98, 1293–1299.
30. H. Abe, N. Takahashi, K. J. Kim, M. Mochizuki and Y. Doi, Biomacromolecules, 2004, 5, 1606–1614.
31. I. C. McNeil and H. A. Leiper, Polym. Degrad. Stab., 1985, 11, 267–285.
32. Poly(lactic acid): synthesis, structures, properties, processing, and applications (Wiley Series on Polymer Engineering and Technology), ed. R. Auras, L. T. Lim, S. E. M. Selke and H. Tsuji, New Jersey, John Wiley & Sons, Inc., 2010.
33. Y. Fan, H. Nishida, Y. Shirai and T. Endo, Polym. Degrad. Stab., 2003, 80, 503–511.
34. T. A. Saleh, M. A. Gondal, Q. A. Drmosh, Z. H. Yamani and A. Al-Yamani, Chem. Eng. J., 2011, 166, 407–412.
35. N. A. Lange and J. A. Dean, Lange's Handbook of Chemistry, McGraw-Hill, 10th edn, 1967.
36. R. H. Petrucci, et al., General Chemistry Principles & Modern Applications, Pearson Prentice Hall, Upper Saddle River, NJ, 9th edn, 2007.
37. E. J. Eilert, Meat Sci., 2005, 71, 122–127.
38. T. P. Labuza, Food Technol., 1996, 50, 68–71.
39. A. Lopez-Rubio, E. Almenar, P. Hernandez-Munoz, J. M. Lagaron, R. Catala and R. R. Gavara, Food Rev. Int., 2004, 20, 357–387.
40. X. Jiang, N. Koizumi, X. Guo and C. Song, Appl. Catal., B, 2015, 170–171, 173–185.
41. P. Kanmani and J. W. Rhim, Carbohydr. Polym., 2014, 106, 190–199.
42. A. Larrañaga, P. Aldazabal, F. J. Martin and J. R. Sarasua, Polym. Degrad. Stab., 2014, 110, 121–128.
43. E. Lizundia, S. Petisco and J. R. Sarasua, J. Mech. Behav. Biomed. Mater., 2013, 17, 242–251.
44. J. Del Rio, A. Etxeberria, N. Lopez-Rodriguez, E. Lizundia and J. R. Sarasua, Macromolecules, 2010, 43, 4698–4707.
45. C. G. Pitt, F. I. Chasalow, Y. M. Hibionada, D. M. Klimas and A. J. Schindler, J. Appl. Polym. Sci., 1981, 26, 3779–3787.
46. A. Höglund, K. Odelius and A. C. Albertsson, ACS Appl. Mater. Interfaces, 2012, 5, 2788–2793.
47. A. Höglund, M. Hakkarainen, U. Edlund and A. C. Albertsson, Langmuir, 2010, 26, 378–383.
48. J. Coates, Interpretation of Infrared Spectra, A Practical Approach, Encyclopedia of Analytical Chemistry, John Wiley & Sons, 2006.
49. F. Achmad, K. Yamane, S. Quan and T. Kokugan, Chem. Eng. J., 2009, 151, 342–350.
50. J. Palacio, V. H. Orozco and B. L. López, J. Braz. Chem. Soc., 2011, 22, 2304–2311.
51. J. M. Zhang, H. Tsuji, I. Noda and Y. Ozaki, Macromolecules, 2004, 37, 6433–6439.
52. Y. Gong, Q. Zhou, C. Gao and J. Shen, Acta Biomater., 2007, 3, 531–540.
53. G. Kister, G. Cassanas and M. Vert, Polymer, 1998, 39, 267–273.
54. R. G. Snyder, M. Maroncelli, H. L. Strauss and V. M. Hallmark, J. Phys. Chem., 1986, 90, 5623–5630.
55. E. Meaurio, E. Zuza, N. Lopez Rodriguez and J. R. Sarasua, J. Phys. Chem. B, 2006, 110, 5790–5800.
56. K. Aou and S. L. Hsu, Macromolecules, 2006, 39, 3337–3344.
57. J. Zhang, Y. Duan, H. Sato, H. Tsuji, I. Noda, S. Yan and Y. Ozaki, Macromolecules, 2005, 38, 8012–8021.
58. F. Kayaci, S. Vempati, C. Ozgit-Akgun, N. Biyikli and T. Uyar, Appl. Catal., B, 2014, 156–157, 173–183.
59. F. Kayaci, S. Vempati, I. Donmez, N. Biyikli and T. Uyar, Nanoscale, 2014, 17, 10224–10234.
60. S. A. Ansari, M. M. Khan, S. Kalathil, A. N. Khan, J. Lee and M. H. Cho, Nanoscale, 2013, 5, 9238–9246.
61. H. Noei, H. Qiu, Y. Wang, E. Löffler, C. Wöll and M. Muhler, Phys. Chem. Chem. Phys., 2008, 10, 7092–7097.
62. M. Qu, H. Tu, M. Amarante, Y. Q. Song and S. S. Zhu, J. Appl. Polym. Sci., 2014, 131, 40287.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra24604k

---