Τhe effect of silica nanoparticles on the thermomechanical properties and degradation behavior of polylactic acid

Abstract

In this work a series of polylactic acid/SiO₂ nanocomposites have been prepared by a melt mixing procedure. The dispersion quality was examined by scanning electron microscopy. To study the degradation behavior of the polylactic acid/nanocomposites prepared, the samples were immersed in a buffer solution at a temperature of 37℃ with a pH of 7.4 for a time period of up to 23 weeks. These conditions simulate those in the human body, appropriate in medical applications. In order to assess their suitability in biomedical applications, we investigated the biocompatibility of these materials in terms of cell viability, growth, and morphology. A good initial cell adhesion has been detected, supporting their potential use in bone tissue engineering applications. The hydrolytic degradation of polylactic acid, under the prescribed conditions, was studied by the molecular weight reduction in terms of size exclusion chromatography, whereas the progress of thermal stability of polylactic acid and polylactic acid/nanocomposites during aging was tested by thermogravimetric analysis. The evolution of the materials’ thermomechanical properties during aging was studied by differential scanning calorimetry, dynamic mechanical analysis, and tensile testing. The crystallization behavior in polylactic acid and the way it is affected by the presence of nanofillers during degradation procedure has been studied and values of 44% crystallinity increment have been found. At the specific aging conditions studied, silica nanoparticles accelerate the degradability of polylactic acid, having a higher impact on Young’s modulus, under the specified aging conditions, for 7 weeks and hereafter this acceleration is retarded, due to the crystallinity increment, as a result of the molecular weight reduction.

Keywords

Biodegradation cytotoxicity testing nanocomposites mechanical properties crystallinity

Introduction

Polylactic acid (PLA) is nowadays one of the most important polymers, due to its combined properties, such as biocompatibility, biodegradability, and adequate mechanical performance. PLA consists of an alternative to petrochemical-based polymers, because it can be produced from natural sources such as corn and wheat and is therefore related to a wide range of medical, textile, and packaging applications. The (bio)degradability of PLA though has both negative and positive aspects. The positive aspect is the (bio)degradation to form nonhazardous products when PLA polymers or articles are discarded or composted after completing their useful life.¹ The negative aspect is the degradation of PLA polymers during processing. Thus, the same properties that make PLA polymers desirable as replacements for nondegradable petrochemical polymers also create undesirable effects which must be overcome. For several applications the degradation rate of PLA is still low as compared to the waste accumulation rate. On the other hand, the fact that PLA degrades slowly—over a period of several weeks up to about 1 year—is advantageous for some other applications such as biomedical, since it leads to a relatively good shelf life.

PLA degrades during thermal processing or under hydrolytic conditions, giving a reduction of molecular weight that affects the final properties of the material, such as the mechanical strength.^2,3 The degradation behavior depends strongly on the molecular weight and the crystallinity of the PLA.⁴ Previous studies have shown that degradation of PLA is enhanced by an increase in temperature and relative humidity (RH).⁵ The temperature, in presence of oxygen, triggers thermal oxidation, while the moisture promotes hydrolytic degradation. PLA degradation driven by hydrolysis needs higher temperature in order to take place (T > 50℃).

Chemical hydrolysis of the hydrolytically unstable backbone is the primary mechanism for degradation of the PLA polymer. Degradation occurs first by water penetrating the bulk of the polymer and hydrolyzing the ester bonds, preferentially those in the amorphous phase, and converting long chains into shorter water-soluble fragments.⁶

On the other hand, the thermomechanical and barrier properties of PLA can be enhanced by the incorporation either of biocompatible fibers⁷ or of a fairly low amount of nanosized fillers (1–5 wt%).^8–10 Many studies have been reported on the preparation and characterization of a wide range of PLA nanocomposites.^10–12 A large variety of inorganic nanoparticles, including silica,¹³ layered silicates,¹⁴ hydroxyapatite,¹⁵ nanotubes,¹⁶ have been investigated for this purpose. Among the various types of nanoparticles, layered silicates, which are naturally abundant, economic, and benign to the environment, have been most extensively used for the preparation of PLA nanocomposites.¹⁷

Besides the thermomechanical properties, the hydrolytic degradation behavior of PLA is also greatly affected by the presence of nanofillers. The manner in which the thermomechanical properties of PLA and PLA/nanocomposites alter over time is still an issue of great importance. Depending on the applications, acceleration of degradation (medical devices) is required, while in some other cases (packaging) suppression of degradation is needed. The effect of various nanofiller types on the PLA degradation behavior has been extensively studied^18–22 and it has been reported that the hydrolytic degradation of PLA takes place first at the interface between PLA matrix and nanofillers rather than in the bulk polymer.¹⁸ A complex effect on the degradation behavior of PLA has been revealed in Kontou et al.,¹⁹ where the effect of two different nanofiller types (silica and montmorillonite) and their mixtures has been studied on the degradability of PLA.

The present work is complementary to our previous works^1,19 on the degradation effects on crystalline and thermomechanical properties of PLA/nanocomposites. In the present work, the degradation ability of PLA reinforced with nanosilica particles at three different contents, 2, 3, and 5% wt, under specific environmental conditions, namely immersion in a buffer solution at temperature of 37℃ with a pH of 7.4 has been studied experimentally. These conditions simulate those in the human body, appropriate in medical applications. The majority of prior art studies on the stability of PLA have been carried out at similar physiological conditions,^4,6,23 but there are still some issues that need to be addressed. A matter of importance has been proved to be the crystallization behavior in PLA and the way it is affected by the presence of nanofillers during degradation procedure. As it is mentioned in Chen et al.¹⁸ the effect of silica on the hydrolytic degradation behavior of PLA is less researched. Apart from this, the role of nanofillers on the evolution of thermomechanical properties of PLA/nanocomposites at various stages of degradation has been less examined, and this work is a contribution toward this direction. The stability of PLA and PLA/nanocomposites was monitored by a number of techniques, including size exclusion chromatography (SEC), scanning electron microscopy (SEM), thermogravimetric analysis (TGA), differential scanning calorimetry (DSC), dynamic mechanical analysis (DMA), and tensile measurements. In order to assess their suitability in biomedical applications, we investigated the biocompatibility of the materials tested in terms of cell viability, growth, and morphology. We therefore cultured MC3T3-E1 preosteoblastic cells on films made either from PLA or PLA with 2% wt silica nanocomposite, visualized the morphology of the adhered cells, and quantified the cell proliferation on them.

Materials

PLA was employed for the preparation of the nanocomposites. The selected grade of PLA (supplied by Nature works) has a D content of 4.25%, a residual monomer content of 0.3%, and a density of 1.24 g/cm³. The material in pellets form was dried at 45℃ for a minimum of 8 h prior to use in a desiccating dryer.

For the preparation of the PLA nanocomposites, Silica Aerosil R972 (supplied by Degussa Chemicals) has been employed, which is hydrophobic fumed silica after treated with dimethyldichlorosilane and based on hydrophilic fumed silica, with a specific surface area of 130 m²/g. The average primary size is 16 nm. A series of PLA nanocomposites with a filler content of 2, 3, and 5 wt.% of silica (Si) have been prepared and designated as PLA/Si/2, PLA/Si/3, and PLA/Si/5.

Experimental

For the preparation of the materials, melt mixing of nanofillers with matrix was performed with a Brabender mixer at a temperature of 180℃, while the rotation speed of the screws was 40 r/min. The temperature for melt mixing was kept as low as possible in order to minimize the effect of degradation. Hereafter, the material was compression molded at 130℃, using a thermo-press and a special mold of 2 mm thickness. The material was then cooled slowly down at ambient temperature. For each material type, two samples were prepared in the same manner, which provided reproducible results.

To study the effect of nanofillers on the degradability of PLA, PLA/nanocomposites were exposed to specific aging conditions, namely immersed in a buffer solution at a temperature of 37℃ with a pH of 7.4, for a time period of up to 23 weeks. The effect of the above-mentioned aging conditions has been studied at time intervals of this period.

For the in vitro biocompatibility assessment, we employed thin films on round glass substrates of 15 mm in diameter, by drop casting the materials’ PLA or PLA with 2% per weight nanosilica. The glass substrate surface was covered with the material and dried in an oven at 60℃. Samples were disinfected in 70% ethanol and air-dried in a laminar flow bench prior to cell seeding on them.

The evolution of molecular weight of the PLA at various aging time intervals was recorded by SEC. All aged and nonaged samples were dissolved in tetrahydrofuran (THF) without any insoluble residues being visible to the naked eye, suggesting that no crosslinking took place. The molecular weights and molecular weight distributions of the PLA samples were determined by SEC using a Waters system, composed of a Waters 1515 isocratic pump, a set of three µ-Styragel mixed bed columns, with a porosity range of 10²–10⁶Å, a Waters 2414 refractive index detector (at 40℃) and operated/controlled through Breeze software. THF, containing 3% v/v triethylamine, was the mobile phase used at a flow rate of 1.0 ml/min at 30℃. The setup was calibrated with linear polystyrene standards having weight average molecular weights in the range 1250–900,000 g/mol. The PLA samples were dissolved directly to the carrier solvent overnight at concentrations ca. 0.1% w/v and characterized the next day.

TGA was performed by using a TGA Q500 V20.2 Build 27 instrument by TA Instruments (New Castle, DE) in an inert atmosphere of nitrogen. In a typical experiment, 10 mg of the material was placed in the sample pan, and the temperature was equilibrated at 25℃. The temperature was then increased to 600℃ at a rate of 20℃/min.

SEM images were obtained by direct observation of the topography of the sample’s surface with a Nova™ NanoSEM 230 (FEI, Hillsboro, Oregon, USA) scanning microscope operated at an acceleration voltage of 5 or 15 kV. All the studied samples were coated with gold to avoid charging under the electron beam. Τhe studied surface of the samples was the fractured cross-section obtained from the tensile test.

Cell culture

Minimum essential Eagle’s medium (α-MEM), penicillin/streptomycin, fetal bovine serum (FBS), and trypsin/ethylenediaminetraacetic acid (EDTA) were purchased from Sigma-Aldrich. PrestoBlue™ reagent and carboxyfluorescein diacetate succinimidyl ester (CFSE) cell viability assay were purchased from Invitrogen Life Technologies. Cell culture plates were obtained from Corning.

Early passages of MC3T3-E1 preosteoblastic cells (DSZM, Germany) were cultured in alpha-MEM supplemented with 10% FBS and 1% penicillin/streptomycin and maintained in an incubator (Thermo Scientific) at 37℃ in a humidified atmosphere with 5% CO₂. Cultures were medium changed every 2–3 days and subcultured using trypsin/EDTA. Prior to seeding on the different substrates, confluent cells were harvested using trypsin/EDTA and counted in a Neubauer chamber.

In vitro cytotoxicity testing/assessment

Optical and fluorescence microscopy

A suspension of 2 × 10⁴ cells in α-MEM cell culture medium was seeded on the PLA-based material films and placed in the cell culture incubator at 37℃. The cells on the samples were examined daily and visualized by means of a Zeiss Axiovert 200 microscope. Images were taken by a ProgRes® CFscan Jenoptik camera (Jena, Germany) using the ProgRes® CapturePro 2.0 software and an objective lens for the 20-fold magnification. The visualization of the living cells by means of the CFSE fluorescent dye was performed in the same microscope using a fluorescence lamp and a filter for emission at 520 nm. Images were taken with the above-mentioned software with an exposure time set at 152 ms.

Cell viability and proliferation on the materials

Cell viability and proliferation of MC3T3-E1 preosteoblasts were assessed using the Presto Blue™ assay, which is a resazurin-based non-toxic metabolic indicator for viable cells.²⁴ PLA-coated samples were disinfected with ethanol and transferred to an empty plate before assaying. A total of 2 × 10⁴ cells in complete alpha-MEM culture medium were seeded on the samples and placed in the cell culture incubator at 37℃. On days 1, 3, and 7 postseeding cell viability and proliferation assay was performed with the PrestoBlue™ reagent according to the manufacturer’s instructions. The absorbance was measured in a spectrophotometer (Molecular Devices SpectraMax M2) and cell number quantification was performed by means of a calibration curve. Error bars represent the average of triplicates ± standard deviation in two independent experiments for each time point.

Examination of cell metabolic activity on the materials was performed by staining with CFSE, which is a nonfluorescent, cell-permeable dye that diffuses into cells and the acetate groups are cleaved by intracellular esterases to yield highly fluorescent carboxyfluorescein succinimidyl ester. After cleavage of the acetate groups, it has a peak excitation of 494 nm and peak emission of 521 nm. Cells (2 × 10⁴) were cultured on PLA-containing samples for 7 days, then washed with phosphate buffered saline (PBS), stained for 15 min with 20 µM CFSE dye in PBS and viewed by epifluorescence microscopy.

DSCs were carried out using a Setaram DSC 141 instrument, calibrated with an Indium standard. Each PLA sample (aged and nonaged) was heated at a constant heating rate of 10℃/min from 20℃ up to 170℃, and the thermogram was recorded. The degree of crystallinity was calculated by considering a melting enthalpy of 93.1 J/g for 100% crystalline PLA.²⁵ Three sheet specimens of 2 mm thickness were tested for every aging procedure. DSC samples were taken from both the center and the tip of every PLA sheet, and it was found that the DSC results were repeatable, with an average scatter lower than 5%. In our study, especially for DSC tests, the aging time period was 36 weeks.

The DSC samples were washed with distilled water two times and dried in the ambient for 2 h, before testing.

DMA experiments were performed using the TA Instruments DMA Q800 instrument. The mode of deformation applied was the single cantilever beam, and the mean dimensions of sample plaques were 12.6 mm × 2 mm × 17.5 mm. Ιn order to obtain a direct measure of the storage modulus evolution versus aging time, the test temperature was kept constant equal to 20℃, so no further thermal effects could take place. The experiments were performed at a constant frequency of 1 Hz.

Tensile measurements were performed with an Instron 1121 type tester, at room temperature. The dumbbell type specimens were of a gauge length of 30 mm, and the applied crosshead speed was 0.2 mm/min. This value corresponds to an effective strain rate of 1.1 × 10^–4s^–1. The deformation could be measured very accurately with an experimental procedure, which is based on a noncontact method with a laser extensometer, described in detail in a previous work.²⁶ Tensile stress–strain curves were then obtained up to the breaking point. Five specimens were tested for each aging procedure, and the scatter between experimental data was lower than 8%.

Results and discussion

SEC results

The degradation of PLA matrix has been tested by examining the molecular weight distribution after exposure to the prescribed aging conditions, for different time periods. The weight average molecular weight

{\bar{M}}_{w}

versus aging time is given in Table 1. A substantial molecular weight decrement is obtained after 163 days of aging.

Table 1.

PLA molecular weight variation with aging time.

Aging time (weeks)	Molecular weight (g/mol)
0	194,873
9.0	169,439.5
23	89,609
48	69,600

As can be seen in Table 1, after aging at 37℃ for 65 days, PLA underwent little change in its molecular weight. Significantly lower molecular weights were measured for samples after 163 days (23 weeks) of aging. This is exemplified in Figure 1, which shows the elution curves of PLA samples for different time intervals. As can be seen in Figure 1, the elution peak shifted to lower molecular weight with increasing aging time. Furthermore, the elution peaks of the aged samples display a higher contribution of low molecular weight fractions, which is attributed mainly to chain scission due to hydrolysis.

The degradation rate was quantified in terms of the average hydrolytic degradation rate constant (k_t). The (k_t) values were evaluated assuming an exponential decrease of M_n using the following equation⁴

{lnM}_{n} (t_{2}) {= lnM}_{n} (t_{1}) {- k}_{t} t

(1)

where M_n(t₂) and M_n(t₁) are the M_n values at the hydrolytic degradation times of t₂ and t₁, respectively. In Figure 2, the molecular weight changes are plotted logarithmically with varying the aging time. In the same plot, analogous results for PLA, degraded under different conditions¹ are depicted for comparison. The experimental procedure applied for aging PLA at 80% RH and various temperatures is presented in detail in Niaounakis et al.¹ Following Figure 2, an average value

{\bar{k}}_{t}

for the hydrolytic degradation rate has been calculated. The estimated

{\bar{k}}_{t}

values for the different time intervals were 3.73 × 10^–4, 2.96 × 10^–3, and 20 × 10^–3days^–1 at 20, 40, and 50℃, and 80% RH, respectively. The sample immersed in the buffer solution at 37℃ had an average degradation rate equal to 4.9 × 10^–3days^–1 for a similar time period of aging with the other samples. When the time period is extended to 340 days, the value of

{\bar{k}}_{t}

is reduced to 3.185 × 10^–3days^–1. The results showed that the degradation rate was higher at the beginning (first 60 days) for every temperature and decreased significantly after 30 days. This finding was in agreement with previous results.² Regarding the aging temperature effect, the degradation process was very slow at temperatures equal or below 40℃, but the speed was greatly enhanced at 50℃. The hydrolytic degradation rate at the buffer solution 37℃ appears to be almost the same with the conditions of 40℃ and 80% RH.

SEM results

The morphological changes incurred in the PLA/nanocomposites after exposure to the prescribed aging conditions were analyzed by SEM micrographs and shown representatively in Figure 3(a) to (d) for PLA/Si/3 and PLA/Si/5 nanocomposites. From these photos, the good quality of nanofiller’s dispersion is apparent, given that the nanoparticles are visible and quite homogeneously dispersed. Nanoparticles of a few nanometers size coexist with larger aggregates, with the aggregates being larger with increasing the nanofiller content. For PLA/Si/3 after 23 weeks of aging, an increasing surface roughness is observed, as well as the existence of holes (dark regions) especially around nanoparticles. Analogous trend is obtained for PLA/Si/5 where, in some cases, holes connecting neighboring nanoparticles are also created.

In vitro cytotoxicity testing/assessment

Initial cell adhesion on PLA and PLA + SiO₂ coatings

The optical microscopy images indicate a good initial cell adhesion 2 h postseeding on both materials, the PLA (Figure 4(b)) and PLA/Si/2 (Figure 4(c)), which is comparable to the tissue culture treated polystyrene control polystyrene (PS) (Figure 4(a)). Cells extend elongated protrusions, displaying a spindle-shape morphology, and attach in a similar number and manner on both PLA-containing coatings as on the preferable PS control. These data depict a characteristic morphology of preosteoblastic cells seeded on a biocompatible material surface without showing any adverse effects.

Cell proliferation on PLA and PLA + SiO₂ coatings

The graph in Figure 5 shows the number of living cells on the tissue culture treated polystyrene control surface, PLA and PLA/Si/2 after 1, 3, and 7 days in culture. Beginning with an initial number of 2 × 10⁴ cells seeded on each sample type, we observe a significantly increasing cell proliferation after 3 and 7 days in culture on both PLA and PLA/Si/2 materials, which is similar to the PS control surface. Among the different material surfaces at each experimental time point, we do not observe any statistically significant proliferation increase.

Cell metabolic activity

Figure 6 shows the epifluorescence of living preosteoblastic cells stained with the CFSE dye on either tissue culture treated polystyrene control (a), or PLA (b), or PLA/Si/2 (c) after 7 days in culture. We observe a comparable number of living cells appearing in green, with a similar flattened morphology on both PLA and PLA/Si/2 materials, as on the polystyrene control. The depicted high cell density and the polygonal cell morphology are evident for well-attached living cells on the material coatings. After 7 days in culture, and compared to the initial adhesion stage shown in Figure 4, cells appear to have increasingly proliferated and formed a confluent live cell layer on all three surfaces. Additionally, these data corroborate with the quantitative cell proliferation analysis shown in Figure 5.

TGA results

The TGA results are summarized in Table 2, where the 10 wt% loss temperature with varying aging time is presented for all samples. From this table it is extracted that the thermal stability of PLA and PLA/Si/5 nanocomposite remains almost unaffected with aging time, while for PLA/Si/2 and PLA/Si/3 it is reduced and shifted toward lower temperatures. This effect denotes that the good quality dispersion of silica at those samples, as postulated by SEM, increases the interfacial region, which is more sensitive to the aging conditions. The formation of large agglomerates in PLA/Si/5 does not have the same effect.

Table 2.

TGA 10 wt% loss temperature in ℃ at various aging times.

Aging time (weeks)	PLA	PLA/Si/2	PLA/Si/3	PLA/Si/5
0	335.8	342	332.7	333
14	333	332.5	331.5	331
24	330.5	327.5	325	330.8

DSC results

The DSC results of the PLA nanocomposites and their evolution with aging time are summarized in Table 3. PLA nanocomposites exhibit three transitions, the glass transition temperature Tg, the cold crystallization temperature T_cc (for PLA matrix), and the melting regime T_m.

Table 3.

DSC properties of PLA/nanocomposites with varying aging time.

Material/aging time (weeks)	T_g (℃)	ΔH^a_r (J/g)	ΔH_cc (J/g)	ΔH^a_m (J/g)	Degree of crystallinity X_{C (%)}	Percentage change crystallinity (%)	T_m (℃) T_m1/T_m2
PLA
0	58.42	3.49	−5.31	22	18.0	0	146.1/154.4
3.0	56.8	5.99	−8.16	22.49	15.4	−14.4	146.25/154.52
12.0	57.1	7.86	−6.9	23.77	18.1	0.55	146/155.2
18.0	58.4	8.32	−9.05	24.66	16.8	−6.6	146/155.8
23.0	60.24	7.9	−9.36	22.45	14.0	−22.2	146.8/157.8
36.0	63.7	8.35	−9.33	27.72	19.8	10.0	146/156.0
PLA/Si/2
0	58.81	1.06		22.55	24.24	0	151/155.3
3.0	59.29	3.88		24.18	26.0	7.26	149.7/155.29
12.0	58.1	4.22		25.52	27.44	13.2	149.54/155.9
18.0	58.15	3.7		27.33	29.38	21.2	149.56/156.4
23.0	60.15	5.0		24.71	26.56	9.57	150/157.74
36.0	59.0	6.25		31.55	33.92	40.0	149.47/156.81
PLA/Si/3
0	58.22	1.15		21.89	23.53	0	150/155.3
3.0	61.0	3.13		24.64	26.49	12.6	150.38/154.8
12.0	58.2	4.11		25.62	27.54	17.0	149.7/155.9
18.0	58.82	4.51		26.39	28.37	20.6	150.1/156.5
23.0	59.6	4.75		27.34	29.39	25.0	149.5/156.3
36.0	62.4	6.0		30.1	32.36	37.52	150/156.54
PLA/Si/5
0	58.15	0.92		20.88	22.45	0	150.3/155.8
3.0	61.5	3.76		23.47	25.23	12.38	149.8/154.7
12.0	57.9	4.0		23.72	25.5	13.58	150.38/156.12
18.0	60.31	3.78		22.79	24.5	9.13	150.1/157.1
23.0	59.8	5.37		27.63	29.7	32.3	150.8/156.19
36.0	62.34	6.67		30.2	32.36	44.14	149.5/156.67

^aEnthalpy values are normalized with respect to pure polymer.

The Tgs of the PLA/nanocomposites were not affected by the presence of silica, meaning that no restriction of silica nanoparticles on the segmental and long-range chain mobility of the matrix takes place, which would lead to a Tg increment of the nanocomposites. On the other hand, no indication of poor adhesion between matrix and nanofillers exists, since in that case a Tg reduction should be obtained.²⁷ A similar effect, where the addition of 2% vol. of fumed silica on PLA does not significantly affect the Tg of the material has been found.²⁸

In the following paragraphs, the evolution of thermal properties with aging time of PLA/nanocomposites is analytically presented. Following Table 3, the Tg of PLA matrix is slightly decreasing during the first 12 weeks of aging, and hereafter starts increasing. The Tg of PLA/nanocomposites does not substantially change with aging time, exhibiting a slight increment after 36 weeks of aging. Generally, the Tg decrement is attributed to the molecular weight reduction of the samples. In previous works it has been found that the Tg is either not significantly affected by aging²⁰ or it is decreased with exposure time.²⁹ In our case, the Tg is generally increased after a long time period of aging for PLA and PLA/nanocomposites, with this increment being more essential in the PLA matrix. The Tg increment with aging has also been detected for similar materials in Georgiopoulos et al.³⁰

From Figures 7 to 10 where DSC curves are illustrated for PLA and PLA/nanocomposites correspondingly, it is shown that the Tg region is expressed by an endothermic peak, corresponding to an enthalpy relaxation ΔH_r. The existence of such an endothermic peak is related to the melting of a mesophase, which is a region with some kind of molecular ordering.^31,32

As presented in Table 3, for the unaged materials, a substantial ΔH_r decrement with the addition of nanoparticles, which implies a confinement of a fraction of PLA chains around nanofillers surface, exists, which cannot contribute to a relaxation around Tg.³³ In contrast, ΔH_r of all samples is increased with aging time, and this increment is enlarged at PLA/nanocomposites compared to that of PLA matrix. This effect denotes that after the sample has been hydrolyzed for a certain period of time, the formation of this locally ordered structure is enhanced.

As shown in Figure 7, PLA matrix exhibits the exothermic effect, attributed to the cold crystallization of PLA during the DSC heating process. It has been found that the cold crystallization temperature T_cc is decreasing with aging time, while the corresponding enthalpy ΔH_cc is increasing. When cold crystallization occurs, imperfect crystals are formed, which melt during the DSC heating scan. In addition, the ΔH_cc increment, related to the enlarged cold crystallization area of PLA matrix after 36 weeks of aging (Figure 7), indicates the presence of more polymeric chains involved in the cold crystallization process, thus strengthening the interpretation of morphological changes previously drawn from the ΔHr assessment. Regarding crystallinity, the PLA/nanocomposites, in the unaged stage, exhibit a percentage change crystallinity, namely 34.7, 30.7, and 24.7% for PLA/Si/2, PLA/Si/3, and PLA/Si/5 correspondingly. These values are calculated with respect to the crystallinity content of PLA matrix, and this effect shows that silica nanoparticles act as nucleating agents.

As far as crystalline behavior of PLA and PLA/nanocomposites under aging conditions is concerned, it must be noted that for a semicrystalline polymer, the initial crystalline structure strongly affects the hydrolytic degradation of PLA.^18,34

Generally, hydrolytic degradation occurs first in the amorphous region. The degree of crystallinity X_c(%) of PLA with increasing aging time exhibits a nonmonotonic behavior, but generally has a tendency to increase up to a value of 10% (Table 3). PLA/Si nanocomposites, on the other hand, appear as monotonic increment of the degree of crystallinity with aging time, up to values of the order of 40–44% with respect to the initial value before aging. This effect consists of a strong indication that silica is an effective nucleating agent, promoting the crystallization of PLA matrix, during the hydrolytic degradation.^18,34 It has been reported that the increased heat of fusion (and therefore degree of crystallinity increment) is attributed to plasticization of PLA by water molecules as well as by lactic oligomers that would give sufficient mobility to the polymer chains to organize and further crystallize.²²

This is the result of two competitive effects: The silica content leads to the increase of the hydrophilicity of the sample, which results in the damage of more crystalline structure during the hydrolytic degradation process. On the other hand, due to degradation, we have easier crystal formation, and the nucleation of new crystalline regions, where smaller chains are involved.

Analyzing further the evolution of crystalline structure during aging, the samples obtained at various aging time periods were examined by DSC. From Figures 7 to 10, illustrating PLA and PLA/nanocomposites, as far as the melting region is concerned, two endothermic peaks appear in all material types. The melting point generally is directly related to the quality of the crystalline region, and consequently the thickness of the lamellae; the bigger the lamellae thickness, the higher the melting peak temperature. Therefore, double melting peaks are generally associated with the presence of two main crystal populations, with different size and quality.^35–37 Following Table 3, the low temperature peak T_m1 for PLA/nanocomposites is higher than that of PLA matrix, whereas the second melting peak T_m2 is almost the same for all materials examined. The first peak T_m1 has been related to the fusion of rather imperfect crystals, formed by primary crystallization, whereas the second melting peak T_m2 can be related to the melting of more perfect crystals, which were formed during a reorganization procedure upon heating.

With increasing aging time, it can be observed from Figures 7 to 10 and Table 3 that T_m1 is almost constant with varying aging time (with an exception for PLA/Si/2, where T_m1 slightly decreases), while T_m2 exhibits a slight increment with aging time, for all material types. It has been reported³⁸ that when T_m decreases with aging, crystalline regions are partially degraded, specially the interlamellar amorphous phases into the crystalline regions, even to a less extent than in the amorphous regions.

In our case, following Figures 7 to 10, the T_m2 peak becomes more intense with aging time for PLA/nanocomposites. It probably happens, due to a mechanism of melting and recrystallization of less perfect crystallites into thicker ones, which hereafter melt at a higher temperature. Aging procedure seems to facilitate this mechanism. This fact combined with the enhancement of the total heat of fusion with aging, denotes that silica can promote the crystallization of PLA matrix during the hydrolytic degradation process for the specific aging conditions applied.

Tensile testing results

The experimental tensile properties of PLA/Si nanocomposites are extensively analyzed in Kontou et al.¹⁹ and Georgiopoulos et al.³⁰ The corresponding evolution of the tensile properties of pristine PLA and its nanocomposites with aging at the specific conditions examined, at various time periods are presented in Table 4 and in Figures 11 and 12 representatively for PLA and PLA/Si/3. From Table 4, PLA/Si nanocomposites exhibit a 39% Young’s modulus increment compared to the Young’s modulus of PLA matrix. From Figure 11 of pristine PLA, it can be observed that in all stages of aging, a yield stress and a subsequent strain softening is exhibited. Following Table 4, it can be noticed that after 3 weeks of aging, PLA matrix undergoes a 14% reduction in the Young’s modulus, while PLA/nanocomposites exhibit a 26% reduction. Hereafter, all studied samples exhibited almost the same trend in tensile properties during aging, i.e. their Young’s modulus appears to be not further greatly affected, with increasing aging time. Finally, after 23 weeks of aging, the Young’s modulus reduction for PLA is about 11%, while it is about 23% for PLA/nanocomposites. Therefore, it can be extracted that silica enhances degradability of PLA, while this enhancement slows down with increasing aging time, probably due to the crystallinity increment for PLA/nanocomposites. Similar trend is obtained for the yield stress variation. Initially, PLA/nanocomposites have a slightly yield stress increment (of the order of 7%) compared to that of PLA matrix. With varying aging time, the yield stress of all samples is gradually decreased, with the PLA/nanocomposites demonstrating higher decrement (up to 37% for PLA/Si/3). On the other hand, after 18 weeks of aging, yield stress of the nanocomposites turns to be lower than that of the matrix. It can be extracted that in this case, the entire load is carried by the polymeric matrix and the load-bearing cross-section is lowering with increasing nanofiller content.²⁸ This effect is an indication that the reinforcing effect, observed initially, is no longer active, revealing that degradation is more intense in the matrix–nanofiller interface. This effect is in agreement with SEM results, where the degradation appears to be more intense around nanoparticles. In our previous study,¹⁹ with the same materials system, and different aging conditions, namely 40℃ and 80% RH, yield stress of PLA/Si nanocomposites was generally higher than that of the PLA matrix. Comparing the two different aging conditions, it is revealed that at 40℃ and 80% RH, the mechanical properties substantially decrease in the first 3 weeks and hereafter they are recovering, while under immersion in a buffer solution at 37℃, the same properties after the exhibited decrement in 3 weeks are slightly recovered and remain almost stable for a long time period of aging. It must also be noted that PLA exhibits a lower degradability in all cases, compared to the PLA/nanocomposites. In the present study, the degradation under immersion in a buffer solution at 37℃ is enhanced, compared to 40℃ and 80% RH, which is also related to the higher hydrolytic degradation rate, calculated in the previous paragraph.

Table 4.

Tensile properties of PLA/nanocomposites with varying aging time.

Material/aging time (weeks)	Young’s modulus (GPa)	Yield stress (MPa)	Strain at break
PLA
0	3.5	54.9	0.055
3.0	3.0	52.65	0.074
7.0	3.03	54.4	0.033
14.0	3.3	54.3	0.04
23.0	3.11	49.6	0.025
PLA/Si/2
0	4.87	56.2	0.055
3.0	3.6	53.32	0.062
7.0	3.6	54.3	0.048
14.0	3.64	51.0	0.02
23.0	3.75		0.012
PLA/Si/3
0	4.83	59.0	0.044
3.0	3.67	53.2	0.061
7.0	3.78	54.35	0.075
14.0	4.0	49.5	0.016
23.0	3.7	37.2	0.012
PLA/Si/5
0	4.85	58.7	0.07
3.0	3.58	55.9	0.032
7.0	3.62	53.62	0.06
14.0	3.65	47.2	0.02
23.0	3.87	39.3	0.012

It can be concluded that silica nanoparticles accelerate the degradability of PLA, having a higher impact on Young’s modulus, under the specified aging conditions, for 7 weeks and hereafter this acceleration is retarded, due to the crystallinity increment, as a result of the molecular weight reduction.

DMA

In Figure 13 the storage modulus at a frequency of 1Hz and 37℃ is depicted with varying aging time. From this plot it is revealed that the storage modulus is always higher for PLA/nanocomposites compared to that of PLA matrix, while the reinforcing effect is higher for PLA/Si/2, probably due to the finer dispersion of nanoparticles. The evolution of storage modulus with aging time is monotonically decreasing, appearing to be different than that of the Young’s modulus. This is due to the different type of deformation applied. The slope of the storage modulus decrement is lower for PLA matrix, while PLA/Si/5 exhibits the higher slope of degradation compared to PLA/Si/2 and PLA/Si/3.

Conclusions

A series of biobased polymer matrix nanocomposites based on PLA with varying nanosilica content have been prepared and studied experimentally. A 39% Young’s modulus enhancement has been found for PLA/nanocomposites with respect to the pure PLA matrix. In addition, 33% crystallinity increment in average has been detected for PLA/nanocomposites, indicating that nanosilica fillers act as nucleating agents. Hereafter, this enhancement does not change with varying nanofiller loading.

The cell culture investigation demonstrates that both materials PLA and the PLA/nanocomposites facilitate a good initial cell adhesion, increase preosteoblastic cell proliferation after 3 and 7 days in culture, and therefore support their potential use in bone tissue engineering applications.

A study of the change of the thermomechanical properties of PLA and PLA/nanocomposites with aging time after exposure at a buffer solution at a temperature of 37℃ with a pH of 7.4, simulating the human body, has been performed. Comparative study with various aging conditions, previously executed has also been made. Significantly lower molecular weights for PLA matrix were measured for samples after 23 weeks of aging. The aged samples display a higher contribution of low molecular weight fractions, which is attributed mainly to chain scission due to hydrolysis.

The crystallization behavior in PLA and the way it is affected by the presence of nanofillers during degradation procedure has been studied and values of 44% crystallinity increment have been found. Double melting endotherms at 146 and 156℃, associated with different morphologies have been detected for all material types studied. The high temperature melting peak was more intense with increasing aging time, and this effect has been attributed to a recrystallization procedure followed by a subsequent melting.

At the particular aging conditions studied, silica nanoparticles accelerate the degradability of PLA, having a higher impact on Young’s modulus, under the specified aging conditions, for 7 weeks and hereafter this acceleration is retarded, due to the crystallinity increment, as a result of the molecular weight reduction.

Footnotes

Declaration of conflicting interests

None declared.

Funding

This research has been co-financed by the European Union (European Social Fund, ESF) and Greek national funds through the Operational Program “Education and Lifelong Learning”, Research Funding Program Aristeia (E.K.), and Aristeia II ‘Osteobiomimesis’ (MC).

References

Niaounakis

Kontou

Xanthis

. Effects of aging on the thermomechanical properties of poly(lactic acid). J Appl Polym Sci 1966; 119: 472–481.

Signori

Coltelli

M-B

Bronco

. Thermal degradation of poly(lactic acid) (PLA) and poly(butylene adipate-co-terephthalate) (PBAT) and their blends upon melt processing. Polym Degrad Stab 2009; 94: 74–82.

Pillin

Montrelay

Bourmaud

. Effect of thermo-mechanical cycles on the physic-chemical properties of poly(lactic acid). Polym Degrad Stab 2008; 93: 321–328.

Saha

Tsuji

. Effects of molecular weight and small amounts of D-lactide units on hydrolytic degradation of poly(L-lactic acid)s. Polym Degrad Stab 2006; 91: 1665–1673.

KLG

Pometto

Hinz

. Effects of temperature and relative humidity on polylactic acid plastic degradation. Polym Environ 1999; 7: 83–92.

Middleton

Tipton

. Synthetic biodegradable polymers as orthopedic devices. Biomaterials 2000; 21: 2335–2346.

Felfel

Ahmed

Parsons

. Investigation of crystallinity, molecular weight change and mechanical properties of PLA/PBG bioresorbable composites as bone fracture fixation plates. J Biomater Appl 2012; 26: 765–789.

Fukushima

Abbate

Tabuani

. Biodegradation of poly(lactic acid) and its nanocomposites. Polym Degrad Stab 2009; 94: 1646–1655.

Ray

Bousmina

. Biodegradable polymers and their layered silicate nanocomposites: In greening the 21^st century materials world. Prog Mater Sci 2005; 50: 962–1079.

10.

MacNeill

Leiper

. Degradation studies of some polyesters and polycarbonates-2. Polylactide: Degradation under isothermal conditions, thermal degradation mechanism and photolysis of the polymer. Polym Degrad Stab 1985; 11: 309–326.

11.

Feijoo

Cabedo

Gimenez

. Development of amorphous PLA-montmorillonite nanocomposites. J Mater Sci 2005; 40: 1785–1788.

12.

Jungraithmayr

Laube

Hild

. Bioactive nanocomposite for chest-wall replacement: Cellular response in amurine model. J Biomater Appl 2014; 29: 36–45.

13.

Wen X, Lin Y, Han C, et al. Thermomechanical and optical properties of biodegradable poly(L-lactide)/silica nanocomposites by melt compounding. J Appl Polym Sci 2009; 114: 3379–3388.

14.

Ray SS, Maiti P, Okamoto M, et al. New polylactide/layered silicate nanocomposites. 1. Preparation, characterization, and properties. Macromolecules 2002; 35: 3104–3110.

15.

Delabarde

Plummer

CJG

Bourban

. Accelerated ageing and degradation in poly-L-lactide/hydroxyapatite nanocomposites. Polym Degrad Stab 2011; 96: 595–607.

16.

Zhao

Qiu

Yang

. Effect of functionalization of multiwalled nanotubes on the crystallization and hydrolytic degradation of biodegradable poly(L-lactide). J Phys Chem B 2008; 112: 16461–16468.

17.

Ray

Yamada

Okamoto

. New polylactide-layered silicate nanocomposites. 2. Concurrent improvements of material properties, biodegradability and melt rheology. Polymer 2003; 44: 857–866.

18.

Chen

Wang

Chen

. Hydrolytic degradation behavior of poly(L-lactide)/SiO2 composite. Polym Degrad Stab 2013; 98: 2672–2679.

19.

Kontou

Georgiopoulos

Niaounakis

. The role of nanofillers on the degradation behavior of polylactic acid. Polym Compos 2012; 33: 282–294.

20.

Chen

. Effect of organic montmorillonite on cold crystallization and hydrolytic degradation of poly(L-lactide). Polym Degrad Stab 2012; 97: 2273–2283.

21.

Paul

Delcourt

Alexandre

. Polylactide/montmorillonite nanocomposites: Study of the hydrolytic degradation. Polym Degrad Stab 2005; 87: 535–542.

22.

Fukushima

Tabuani

Dottori

. Effect of temperature and nanoparticle type on hydrolytic degradation of poly(lactic acid) nanocomposites. Polym Degrad Stab 2011; 96: 2120–2129.

23.

Zhang

Espiritu

Bilyk

. Morphological behavior of poly(lactic acid) during hydrolytic degradation. Polym Degrad Stab 2008; 93: 1964–1970.

24.

Terzaki

Kissamitaki

Skarmoutsou

. Pre-osteoblastic cell response on three-dimensional, organic-inorganic hybrid material scaffolds for bone tissue engineering. J Biomed Mater Res Part A 2013; 101A: 2283–2294.

25.

Fischer

Sterzel

Wegner

. Investigation of the structure of solution grown crystals of lactide copolymers by means of chemical reactions. Colloid Polym Sci 1973; 251: 980–990.

26.

Kontou

Farasoglou

. Determination of the true stress-strain behaviour of polypropylene. J Mater Sci 1998; 33: 147–153.

27.

Dacko

Kowalczuk

Janeczek

. Physical properties of the biodegradable polymer compositions containing natural polyesters and their synthetic analogues. Macromol Symp 2006; 239: 209–216.

28.

Dorigato

Sebastiani

Pegoretti

. Effect of silica nanoparticles on the mechanical performance of poly(lactic acid). J Polym Environ 2012; 20: 713–725.

29.

Vasanthan

. Effect of microstructure on hydrolytic degradation studies of poly(L-lactic acid) by FTIR spectroscopy and differential scanning calorimetry. Polym Degrad Stab 2009; 94: 1364–1372.

30.

Georgiopoulos P, Kontou E and Niaounakis M. Thermomechanical properties and rheological behavior of biodegradable composites. Polym Compos 2014; 35: 1140–1149.

31.

Chen

Feng

Zhang

. Hydrolytic degradation behavior of poly(L-lactide)/carbon nanotubes nanocomposites. Polym Degrad Stab 2013; 98: 198–208.

32.

Zhang

Duan

. Physical aging enhanced mesomorphic structure in melt-quenched poly(L-lactic acid). J Phys Chem B 2011; 115: 13835–13841.

33.

Pluta

Murariu

Alexandre

. Polylactide compositions. The influence of ageing on the structure, thermal and viscoelastic properties of PLA/calcium sulfate composites. Polym Degrad Stab 2008; 93: 925–931.

34.

Tsuji

Ikarashi

Fukuda

. Poly(L-lactide): XII. Formation, growth, and morphology of crystalline residues as extended-chain crystallites through hydrolysis of poly(L-lactide) films in phosphate-buffered solution. Polym Degrad Stab 2004; 84: 515–523.

35.

Huang

Hung

Wen

. Polylactide/nano- and microscale silica composite films. II. Melting behavior and cold crystallization. J Appl Polym Sci 2009; 112: 3149–3156.

36.

Shieh

Liu

. Effects of carbon nanotubes on crystallization and melting behavior of poly(L-lactide) via DSC and TMDSC studies. J Polym Sci Part B Polym Phys 2007; 45: 1870–1881.

37.

Chen

Zhang

. Crystallization kinetics and melting behaviors of poly(L-lactide)-graphene oxides composites. Thermochim Acta 2013; 566: 57–70.

38.

Ray

Yamada

Okamoto

. Polylactide-layered silicate nanocomposite: A novel biodegradable material. Nano Lett 2002; 2: 1093–1096.

Τhe effect of silica nanoparticles on the thermomechanical properties and degradation behavior of polylactic acid

Abstract

Keywords

Introduction

Materials

Experimental

Cell culture

In vitro cytotoxicity testing/assessment

Optical and fluorescence microscopy

Cell viability and proliferation on the materials

Results and discussion

SEC results

SEM results

In vitro cytotoxicity testing/assessment

Initial cell adhesion on PLA and PLA + SiO2 coatings

Cell proliferation on PLA and PLA + SiO2 coatings

Cell metabolic activity

TGA results

DSC results

Tensile testing results

DMA

Conclusions

Footnotes

Declaration of conflicting interests

Funding

References

Initial cell adhesion on PLA and PLA + SiO₂ coatings

Cell proliferation on PLA and PLA + SiO₂ coatings