Spectroscopic evidence of successful in situ insertion of sol–gel silica in a poly(3-hydroxybutyrate) matrix

Abstract

Silica-based (SiO₂) poly(3-hydroxybutyrate) (PHB) nanocomposites were obtained via an in situ sol–gel route in three distinct particle concentrations (1, 5, and 7.5% by weight of PHB). The polymer hybrids formed were analyzed via wide-angle X-ray diffraction (WAXD), small-angle X-ray scattering (SAXS), Fourier transform infrared (FTIR) spectroscopy, and time-domain nuclear magnetic resonance relaxometry (TD-NMR). The SiO₂ inorganic structure displayed surface fractal features at low concentration (1 wt%) and denser agglomerates at higher concentrations (5 and 7.5 wt%). FTIR and SAXS results confirmed the formation of the inorganic matrix amid the polymer chains with different levels of distribution and organization. WAXD and TD-NMR results suggested the SiO₂ influence on the PHB crystallinity degree, which was reflected on the polymer’s molecular dynamics with a nonlinear dependence of particles concentration in the PHB matrix.

Keywords

Sol–gel synthesis PHB SAXS time-domain NMR

Introduction

Poly(3-hydroxybutyrate) (PHB) is an aliphatic biodegradable polyester that can be obtained through several sources such as bacterial,^1,2 vegetables,^3,4 and oils.^5,6 There have been extensive studies on PHB over the years due to the fact that this polymer has physical properties similar to polypropylene.⁷ This polymer also presents extensive attractive features, such as biocompatibility, extremely high biodegradability rates, and can be converted to carbon dioxide and energy by common microorganisms, such as bacteria, fungi, and algae.⁸

With properties such as high crystallinity degree (50–70%), excellent gas barrier (H₂O vapor permeability ≈ 560 g.µm m⁻² day⁻¹), and notable mechanical properties (elastic modulus of 3 GPa and tensile strength at break of 25 MPa), PHB is well sought after in the industry. However, PHB presents a technological challenge: it is a brittle polymer, with low elongation at break, low thermal stability when above its melting point, and low degradation temperature.⁹ To enhance PHB degradability rates, Silva et al.¹⁰ created blends of PHB/polyethyleneglycol (PEG) and nanocomposites through solution casting method containing different amounts of vermiculite clay (VMT). The group concluded that the addition of both PEG and VMT created a more ideal environment for increasing water permeability and, consequently, a quicker hydrolytic degradation. The use of PHB with 5 wt% Nanomer I44.P, an organically modified clay with dimethyl dialkyl (C14–18) ammonium,¹¹ presented the same quick hydrolytic reaction.

PEG-grafted silica (SiO₂) nanoparticles were inserted on a PHB matrix, which yielded a better nanoparticle distribution due to good interaction between PEG/PHB matrixes and slowed the crystallization rate and the crystallinity degree of the nanocomposites, thereby broadening the degradation rate of this polymeric matrix.¹² Lan and Sun¹³ reported several cases of different SiO₂ nanoparticles which highlighted the influence of the type and source of a given filler (viz neat commercial SiO₂ (S5130 and R812) vs. sol–gel obtained SiO₂ nanoparticles) upon the properties of the obtained material. Whereas commercial SiO₂ had the tendency of forming aggregates, sol–gel synthesized PEG-grafted silica nanoparticles showed themselves better distributed through the polymeric matrix.

SiO₂ present many advantages, namely high surface area, reliable thermal stability, and increased biocompatibility. SiO₂ is cheap and easy to synthesize, which makes it even more attractive.¹⁴ Wu et al.¹⁵ obtained flame-retardant/ethylene–octene copolymer/polypropylene composites, enhancing its flame-retardant properties by increasing the SiO₂ content. Mechanical properties of this nanocomposite were also increased when increasing the SiO₂ content. Similarly, when used in a cross-linked polylactide (PLA) hydrogel, SiO₂ greatly improved the thermal resistance of this material, but decreased its mechanical properties when compared to the neat hydrogel.¹⁶

Sol–gel chemistry represents a widely popular synthesis strategyfor a large range of popular organic/inorganic materials because of its convenience and versatility. This method consists of a generation of colloidal suspensions (the sol phase, with particles in the range between 1 nm and 100 nm) which gains viscosity, though the formation of the covalent bonds, leading to a gel material.¹⁷ The first step generally consists of mixing the desired alkoxide or organometallic precursors, followed by hydrolysis and polycondesantion of the inorganic network.¹⁸ Several sol–gel products are used as biomaterials, since this route yields compounds with notable biocompatibility and lower toxicity.^19,20 Furthermore, there are already sol–gel materials applied in soil and pollution control,²¹ sensors,²² and other applications. Huang and Chen²³ obtained poly(vinylidene fluoride)(PVDF)/sol–gel SiO₂ composites. These composites displayed well-distributed SiO₂ particles through the polymeric matrix, which decreased PVDF crystallinity degree due to the in situ insertion of SiO₂ particles. Bioactive SiO₂ materials were created with copolymers of methyl methacrylate, 3-(trimethoxysilyl)propyl methacrylate, and sol–gel synthesized SiO₂. The hybrids, when compared to their non-loaded counterparts, became tougher, more flexible, and exhibited lower toxicity in MC3T3 cell cultures.²⁴ Sol–gel obtained poly(ether imide)(PEI)/SiO₂ hybrids were used to coat magnesium bone implants, since those materials tend to degradate extremely fast. The coating increases their hydrophilicity by 15%, which might have lead to a better cellular interaction and decreased the corrosion rate of the metallic implants.²⁵ Catauro et al.²⁶ also used sol–gel synthetized hybrids for implant coatings. Titanium implants were dipped into polycaprolactone/SiO₂ hybrids solutions. The materials displayed premature biomineralization when immersed in simulated body fluid and increased cellular viability. Carbon nanotubes (CNTs)/SiO₂ sol–gel-derived nanoparticles were incorporated into poly(vinyl alcohol)(PVA) and polyurethane (PU), and both the PVA/CN/SiO₂ and PU/CNT/SiO₂ nanocomposites displayed higher thermal stability when compared to their neat counterpart.²⁷

Taking into account the versatility of sol–gel SiO₂-based materials, in this work, we sought to obtain SiO₂ structures in situ through this method in a PHB matrix and characterize the properties of the hybrids through Fourier transform infrared spectroscopy (FTIR), wide-angle X-ray diffraction (WAXD), small-angle X-ray scattering (SAXS), and time-domain nuclear magnetic resonance (TD-NMR) methods.

Experimental

Preparation of the samples

PHB (10 wt%) solutions were prepared in Schott flasks using PHB (Biocycle Inc., São Paulo, Brazil, weight-average molecular weight 375.831 g mol⁻¹) and chloroform (Sigma-Aldrich, São Paulo, Brazil, 99% purity) under magnetic stirring for 24 h and were heated until complete solubilization of the polymer. After solubilization, sufficient amounts of tetraethyl orthosilicate (TEOS; 98%, Sigma-Aldrich) were added to the polymer solutions to obtain hybrids containing 1, 5, and 7.5 wt% of SiO₂. After stirring, the solutions were poured into petri dishes and left in a fumehood at room temperature for a week. The hybrid PHB films were taken to a vacuum oven to remove the residual chloroform and unreacted TEOS. The samples were then kept in a desiccator and thereafter hydrolyzed with a 10 wv% sodium hydroxide aqueous solution at 50°C for a certain amount of time, then washed with deionized water until neutral pH, and stored in a desiccator for characterization.

Infrared spectroscopy (FTIR)

The infrared spectra of the samples were recorded on a Varian 3100 spectrometer using an attenuated total reflectance accessory (Varian Inc., São Paulo, Brazil, ZnSe crystal), between 4000 cm⁻¹ and 600 cm⁻¹, with 120 scans and a resolution of 4 cm⁻¹.

Wide-angle X-ray diffraction

WAXD profiles of samples were obtained on a Rigaku Ultima IV equipment (Japan). The parameters set were copper K_α (λ = 0.154 nm) radiation, at 40 kV and 30 mA, and the 2θ angle was varied between 1° and 50° by 0.01° steps. The crystallinity degree (% X_c, equation 1) of the sample was calculated through the Ruland-Vonk method²⁸ with the aid of the Fityk® 0.9.8 software.

X_{c} = (\frac{\sum Area of the crystallization}{\sum Total area}) \times 100

Small-angle X-ray scattering

SAXS measurements were carried out on a laboratory SAXS instrument Bruker-Nanostar™ (Bruker Inc., São Paulo, Brazil), located at the Institute of Physics at University of São Paulo (USP), Brazil. This equipment is improved by the use of microfocus source Genix 3D coupled with Fox 3D multilayer optics and two sets of scatterless slits for beam definition, all provided by Xenocs™. The film samples were cut in the adequate shape and kept at sealed plastic holders until needed. A Hi-Star detector (Bruker) was used to record the SAXS intensity, I(q), as a function of modulus of scattering vector q, q = (4π/λ) sin (θ/2), with θ being the scattering angle and λ = 0.15418 nm. Each SAXS pattern corresponds to a data collection time of 2 h. The range of the scattering vector, q, was from 0.009 Å⁻¹ to 0.48 Å⁻¹. The scattering intensity was normalized by subtracting the background scattering.

Time-domain nuclear magnetic resonance

A film of sample was taken to a Maran Ultra equipment, operating at 0.54 T (23.4 MHz proton Larmor frequency) equipped with a 18-mm diameter probe and an air flow temperature control unit (±0.1°C). Measurements were performed at 27°C using the Solid Echo pulse sequence $[{(RD - 90 x - t - 90 y - FID)}_{n}]$ . The 90° pulses were automatically calibrated by the equipment at 7.5 μs, with a total dead time of 11.5 μs. The interval between 90° pulses in Solid Echo was 5 μs. Four signal accumulations were performed, with a recycle delay of 2 s and a total of 2048 acquisition points. The data processing was performed with the aid of WinFit 2.4.0.0 (Resonance Instruments Ltd) and Origin 8.5 (Origin Lab Co.) software.

The Solid Echo pulse sequence was employed to eliminate the influence of the probe dead time on the signal acquisition, thus affording better interpretation of the contributions to the NMR signal arising from intensely coupled proton nuclei, such as those likely to be found in the polymeric crystalline phase. The total NMR signal can be fitted by a function combining a Gaussian and an exponential decay, accounting for rigid and mobile phases, respectively. Such function shown given in equation (2):

A (t) = A_{g} exp [- {(\frac{- t}{T_{2}^{g}})}^{2}] + A_{e} exp [- (\frac{- t}{T_{2}^{e}})]

where A_g is the signal amplitude of the rigid fraction and A_e is the signal amplitude of the mobile fraction. Each fraction decays with a characteristic $T_{2}^{*}$ time constant.

The percentage contribution of each fraction of the total signal is given by equation (3):

A_{x} = \frac{A_{g} vs. A_{e}}{I_{FID}} \times 100

Where A_x is the desired fraction (either Gaussian or exponential) and I_FID is the total signal intensity of the sample.

Results and discussion

Opaque, off-white, self-sustained polymeric films were obtained after complete solvent evaporation. The samples showed good stability upon handling, with no sign of brittleness. The alkaline hydrolysis step was employed to accelerate the polycondensation of the remaining unreacted TEOS on the polymer’s surface, thus affording SiO₂ nanoparticles amid PHB chains.

The efficiency of the hydrolysis step was gauged via TD-NMR by measuring the magnetization (M₀) arising from the hydrogen nuclei in the sample before and after the treatment with the alkaline solution (Figure 1).

Figure 1.

Comparative analysis of the samples before and after alkaline treatment. The reduction in signal intensity is related to the consumption of TEOS by the sol–gel reaction.

For better comparison, the TD-NMR signals were normalized by the mass of each sample. The results indicate an overall reduction of the signal intensity after the hydrolysis step, which can be attributed to the consumption of TEOS and its replacement by silica chains that cannot be detected by the TD-NMR equipment. It should be noted that some residual TEOS must have stayed trapped inside the polymer matrix, as denoted by the positive dependence of the signal observed on the amount of TEOS employed in each formulation.

Another evidence of the successful sol–gel reaction is the appearance of an infrared absorption mode typical of Si–O symmetric and asymmetric deformations at 795 and 1090 cm⁻¹, respectively,²⁹ indicating that there was formation of the inorganic phase (Figure 2). The absorptions were only clearly discernible at the highest nominal SiO₂ concentration studied in this work due to the equipment’s resolution. Furthermore, the absorption arising from chloroform at 825 cm⁻¹,^30,31 was not detected, meaning that the solvent was sufficiently removed by the film preparation steps, therefore not contributing to the TD-NMR signal intensity analysis discussed previously. Aside from the mentioned features, no other relevant modification was identified in the sample’s infrared spectra.

Figure 2.

FTIR spectra of the samples showing Si–O absorption modes arising from the inorganic phase.

The incorporation of the inorganic phase amid the PHB matrix should effect changes on the polymer’s crystallinity degree by positive and/or negative nucleation events. Figure 3 shows WAXD patterns obtained for the samples. A qualitative analysis of the diffractograms’ profiles demonstrates a sudden increase in peak intensity at 1 wt% of SiO₂ loading with respect to neat PHB. Higher SiO₂ loads seemed to have provoked the disappearance of the typical PHB diffraction pattern at 2θ = 13.5°, 17°, 22°, and 25.5° corresponding to planes (020), (110), (101), and (121), respectively.^32,33 The WAXD crystallinity degree reduction was previously seen on sol–gel obtained PLA-SiO₂ nanocomposites and attributed to the formation of amorphous SiO_2,²⁴ and the same characteristic was seen for phenolphthalein polyethersulfone/SiO₂ nanofibrous membranes.³⁴

Figure 3.

WAXD patterns obtained for all the samples. Qualitative analysis indicates SiO2-load dependent.

The calculated X_c of all samples is presented in Table 1. It should be noted that SiO₂ is an amorphous oxide at room temperature if it has not undergone annealing.³⁵ Therefore, it could not have contributed to the increase in PHB crystallinity degree observed for the PHB/SiO₂ 1 wt% hybrid. This observation is better discussed with the aid of the SAXS results presented ahead in this article.

Table 1.

X_c calculated via equation (1) for the samples studied in this work.

Sample	X_c (%)
PHB	51.8
PHB/SiO₂ 1 wt%	56.8
PHB/SiO₂ 5 wt%	39.7
PHB/SiO₂ 7.5 wt%	24.5

X_c: degree of crystallinity; PHB: poly(3-hydroxybutyrate); SiO₂: silica.

Probing the PHB molecular dynamics by means of the Solid Echo sequence, we verified that the rigid fraction of the NMR magnetization, typically arising from rigid and crystalline phases in the polymer matrix, was not affected by the incorporation of SiO₂ in all concentrations studied in this work. The mobile fraction, attributed to polymeric amorphous regions, however, showed substantial increase in relaxation time with the addition of SiO₂ in high concentrations. This increase can be related to the extinction of portions of the crystalline phase brought by the presence of dense silica networks in the polymeric matrix.^36,37 The results are presented in Table 2.

Table 2.

Rigid and mobile phase contributions to the decay of the NMR signal of the samples obtained by equation (2).

Sample	$T_{2}^{*}$ /rigid fraction (µs)	$T_{2}^{*}$ /mobile fraction (µs)
PHB	13.0	130
PHB/SiO₂ 1 wt%	13.1	180
PHB/SiO₂ 5 wt%	12.9	254
PHB/SiO₂ 7.5 wt%	13.2	263

NMR: nuclear magnetic resonance; PHB: poly(3-hydroxybutyrate); SiO₂: silica.

A convenient way to visualize the similarities between the WAXD results and those obtained through TD-NMR^32,38 is presented in Figure 4. A nonlinear dependence of both variables on the loading of SiO₂ in the PHB matrix is clearly seen. The results show that the rigid fraction (A_g) can be attributed to the crystalline phase in the polymeric matrix.

Figure 4.

Comparison between TD-NMR and WAXD results showing a similar trend for all samples studied. The TD-NMR values were obtained by equation (3).

SAXS is a powerful and well-established technique for probing the detailed microstructure of semicrystalline polymers and provides information describing structural features of the order of 10–1000 Å in size.^39,40 In samples possessing a regular, ordered morphology (such as PHB), a correlation peak is observed in the SAXS intensity at a characteristic value of q (denoted by q_max), which is related to the distance between domains or heterogeneities in the material. In this way, SAXS usually allows lamellar period lengths (L) in polymer spherulites and drawn polymers to be resolved. In a simplified way, the scattering may be treated according to Bragg’s law and hence the following equation (4):

q_{max} = \frac{2 π}{d}

where d is the average interdomain spacing (d-spacing) and q is the magnitude of the scattering vector, defined as (4π sin θ)/λ), where θ is the scattering angle and λ is the X-ray wavelength. In our case, the PHB possesses a lamellar morphology which is isotropically distributed throughout the material. The Lorentz correction can then be applied to the SAXS intensity, aiming at the determination of peak position. The Lorentz correction involves the multiplication of the intensity function by q² and has the effect of shifting the position of a broad correlation peak to higher values of q (Figure 5).

Figure 5.

Lorentz-corrected (iq² vs. q) SAXS patterns of PHB samples with 0, 1, 5, and 7.5 wt% of SiO2.

In this way, assuming an ideal lamellar model used to describe certain semicrystalline synthetic polymers, the d-spacing represents the distance between adjacent crystalline lamellae which are separated by amorphous domains and arranged in space-filling stacks. The long period (L) can be also estimated by Lorentz-corrected SAXS patterns, represented by d-spacing (d). The long period of the PHB samples with 0, 1, 5, and 7.5 wt% of SiO₂ according to Bragg’s Law were 59.7, 57.3, 55.9, and 55.2 Å (Table 3). The decrease in the long period with increasing of SiO₂ concentration may be related to the level of structural disorganization of the polymer, as previously indicated by WAXD and TD-NMR.

Table 3.

SAXS parameters obtained from Teixeira’s model and Bragg’s law.^a

Sample	r₀ (Å)	ξ (Å)	D	L (Å)
PHB	20	145	3.08	59.7
PHB/SiO₂ 1 wt%	11	115	2.90	57.3
PHB/SiO₂ 5 wt%	5	146	2.98	55.9
PHB/SiO₂ 7.5 wt%	4.9	195	2.96	55.2

SAXS: small-angle X-ray scattering; PHB: poly(3-hydroxybutyrate); SiO₂: silica.

^a SAXS parameters obtained from Teixeira include r₀, ξ, D, and Rg. SAXS parameter obtained by Bragg’ law: L.

From Figure 6 it is also possible to observe that the characteristic peak of the lamellar periodicity of PHB at high q-values (around 0.1 Å⁻¹) becomes flattened, indicating the increase of structural disorder might be caused by inorganic phase formation by sol–gel method. Another information can be visualized in the range of q between 0.2 Å⁻¹ and 0.3 Å⁻¹ (highlighted dotted region), where a not very intense and broad peak appears. This peak is possibly attributed to the correlation distance of domains rich in SiO₂, indicating an excess of inorganic phase in PHB samples containing 5 and 7.5 wt% of SiO₂.

Figure 6.

Small-angle X-ray scattering patterns of PHB samples with 0, 1, 5 and 7.5 wt% of SiO2.

Owen and Bergmann⁴¹ reported the presence of structures showing fractal geometry over a certain range of lengths in PHB samples. They concluded that the units making up the fractal structure are much larger fibrillar units to which individual crystal lamellae. Sol–gel systems also commonly present fractal structures.^42,43 For fractal objects, it is well-known that the scattered intensity I(q) follows a power law, I(q) ≈ q^−D, where D is a fractal dimension. For mass fractals, 1 < D < 3, while for surface fractals 3 < D < 4. In this present work, we have found an exponent of approximately 2.9 in the sample with 1 wt% of SiO₂.

A linear decreasing intensity regime is observed at medium and low q-values, with a slope lower than 3, which invalidates the use of the Beaucage model to determine the structural parameters of the samples’ secondary domains. Freltoft et al.⁴⁴ and Teixeira⁴⁵ have shown that the nanostructure of samples presenting SAXS profiles with both tendency to plateau at low q-range and said linear regime at medium q-range is fractal and that the experimental SAXS curve can be fitted by equation (5):

I (q) = C {1 + {[\frac{1}{{(q r_{0})}^{D}}] [\frac{D Γ (D - 1)}{1 + \frac{1}{q^{2} ξ^{2}}}] sin [(D - 1) {tan}^{- 1} (q ξ)]}}

Equation (5) is valid for a fractal object composed of elementary particles of size r₀, forming fractal aggregates of average size ξ. Γ is the gamma function and C is a constant. This expression becomes I(q) ≈ q^−D when 1/ξ < q <1/r₀. Equation (5) fits well the experimental curves at low and medium q-range. This reveals that the secondary domains present in all samples have a fractal structure. The value of D (around 3, Table 3) evidences the ramified character of the fractal secondary aggregates commonly obtained by the sol–gel method.

The data in Table 3 show that the structural parameters obtained by fitting the SAXS data through Teireixa’s model indicate that a more dense inorganic structure is formed from the concentration of 5 and 7.5 wt% of SiO₂. The sample containing the least amount of SiO₂ had the smallest aggregate size (115 Å), which may have been low enough that small nuclei were formed, with a short reach of interconnectivity, favoring the crystallization of PHB. Somehow it appears that the lower volume of branched inorganic structure positively interferes with the crystallinity degree, facilitating the growth of PHB crystals. On the other hand, for the samples containing 5 and 7.5 wt% of SiO₂, the formation of a dense inorganic network would be responsible for the decrease in the packaging of the crystalline structure of the PHB, reducing sharply the crystallinity degree of the samples due to the greater extension of the structural disorder. The comparative analysis of the WAXD and TD-NMR data allowed us to infer that the variation in crystallinity degree was arising from the different mobility regimes experienced by the PHB chains.

Conclusion

In situ sol–gel synthesis of SiO₂ nanoparticles have been successfully conducted in a PHB matrix. The alkaline hydrolysis of the TEOS precursor was detected via TD-NMR showing a convenient and fast analysis method for this kind of reaction. The formation of the SiO₂ network on the polymer was verified through FTIR and SAXS measurements. The hybrid’s crystallinity degree was very sensitive to SiO₂ loading and demonstrated a nonlinear dependency with this parameter, as detected by TD-NMR and WAXD. Structural parameters of the hybrids were determined via SAXS which revealed fractal structures composed of SiO₂ networks in the hybrids, typical of this kind of sol–gel synthesis. Besides, trends in parameters such as the inorganic network´s level of aggregation were elucidated and shown to be dependent on SiO₂ loading in the PHB matrix.

Footnotes

Acknowledgements

The authors would like to thank the Institute of Physics at University of São Paulo for providing the means to conduct the SAXS measurements.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors would like to thank Brazilian agencies CAPES and CNPq for their financial support.

ORCID iD

Anna Lecticia Martinez Martinez Toledo

References

Fluechter

Follonier

Schiel-Bengelsdorf

, et al. Anaerobic production of poly (3-hydroxybutyrate) and its precursor 3-hydroxybutyrate from synthesis gas by autotrophic clostridia. Biomacromolecules 2019; 20: 3271–3282.

Anjum

Zuber

Zia

, et al. Microbial production of polyhydroxyalkanoates (PHAs) and its copolymers: a review of recent advancements. Int J Biol Macromol 2016; 89: 161–174.

Dobrogojski

Spychalski

Luciński

, et al. Transgenic plants as a source of polyhydroxyalkanoates. Acta Physiol Plant 2018; 40: 162.

Suriyamongkol

Weselake

Narine

, et al. Biotechnological approaches for the production of polyhydroxyalkanoates in microorganisms and plants—a review. Biotechnol Adv 2007; 25: 148–175.

Zahari

MAKM

Ariffin

Mokhtar

, et al. Case study for a palm biomass biorefinery utilizing renewable non-food sugars from oil palm frond for the production of poly (3-hydroxybutyrate) bioplastic. J Clean Prod 2015; 87: 284–290.

Kookos

Koutinas

Vlysidis

. Life cycle assessment of bioprocessing schemes for poly (3-hydroxybutyrate) production using soybean oil and sucrose as carbon sources. Resour Conserv Recycl 2019; 141: 317–328.

Griffin

GJL

. Chemistry and technology of biodegradable polymers. Blackie Academic and Professional, 1994.

Holmes

. Applications of PHB—a microbially produced biodegradable thermoplastic. Phys Techn 1985; 16: 32–36.

Pachekoski

Agnelli

JAM

Belem

. Thermal, mechanical and morphological properties of poly (hydroxybutyrate) and polypropylene blends after processing. Mater Res 2009; 12: 159–164.

10.

Silva

RDN

da Silva

LRC

de Morais

ACL

, et al. Study of the hydrolytic degradation of poly-3-hydroxybutyrate in the development of blends and polymeric bionanocomposites. J Thermoplast Compos Mater 2019. Epub ahead of print 11 June 2019. DOI: 10.1177/0892705719856044.

11.

Gunning

Geever

Killion

, et al. The effect of the mixing routes of biodegradable polylactic acid and polyhydroxybutyrate nanocomposites and compatibilised nanocomposites. J Thermoplast Compos Mater 2016; 29: 538–557.

12.

Lan

Sun

. Influence of the surface properties of nano-silica on the dispersion and isothermal crystallization kinetics of PHB/silica nanocomposites. Mater Chem Phys 2017; 199: 88–97.

13.

Lan

Sun

. Dispersion, crystallization behavior, and mechanical properties of poly (3-hydroxybutyrate) nanocomposites with various silica nanoparticles: effect of surface modifiers. J Poly Res 2018; 25: 121.

14.

Waddell

Douglas

. Improved black sidewall compound performance using precipitated silica. Rubber World 1993; 208: 21–25.

15.

Zhao

, et al. Flammable and mechanical effects of silica on intumescent flame retardant/ethylene–octene copolymer/polypropylene composites. J Thermoplast Compos Mater 2015; 28: 981–994.

16.

Prapruddivongs

Rukrabiab

Kulwongwit

, et al. Effect of surface-modified silica on the thermal and mechanical behaviors of poly (lactic acid) and chemically crosslinked poly (lactic acid) composites. J Thermoplast Compos Mater 2019. Epub ahead of print 8 April 2019. DOI: 10.1177/0892705719856044.

17.

Wright

Sommerdijk

. Sol-gel Materials: Chemistry and Applications. Boca Raton: CRC press, 2018.

18.

Kaur

Pickrell

Sriranganathan

, et al. Review and the state of the art: sol–gel and melt quenched bioactive glasses for tissue engineering. J Biomed Mater Res B Part B: Appl Biomater 2016; 104: 1248–1275.

19.

Fathi

Mohammadi

. Preparation and characterization of sol–gel bioactive glass coating for improvement of biocompatibility of human body implant. Mater Sci Eng: A 2008; 474: 128–133.

20.

Gupta

Kumar

. Bioactive materials for biomedical applications using sol–gel technology. Biomed Mater 2008; 3: 034005.

21.

Innocenzi

Kessler

Zub

. Sol-gel Methods for Materials Processing: Focusing on Materials for Pollution Control, Water Purification, and Soil Remediation. New York: Springer, 2008.

22.

Morosanova

. Silica and silica–titania sol–gel materials: synthesis and analytical application. Talanta 2012; 102: 114–122.

23.

Huang

Chen

. The study on poly (vinylidene fluoride)/silica hybrid composites made by sol—gel process. J Thermoplast Compos Mater 2009; 22: 651–660.

24.

Zhang

. Preparation of PLLA-SiO₂ composites using APTMS as a crosslinking agent with the sol-gel technique. Poly Poly Compos 2013; 21: 431–438.

25.

Kang

Jang

Jung

, et al. Poly (ether imide)-silica hybrid coatings for tunable corrosion behavior and improved biocompatibility of magnesium implants. Biomed Mater 2016; 3: 035003.

26.

Catauro

Bollino

Papale

. Surface modifications of titanium implants by coating with bioactive and biocompatible poly (ε-caprolactone)/SiO₂ hybrids synthesized via sol–gel. Arab J Chem 2018; 11: 1126–1133.

27.

Xie

Liu

, et al. Study on thermal properties of poly (vinyl alcohol) and polyurethane composites with multiwalled carbon nanotubes/silica nanohybrids. J Thermoplast Compos Mater 2017; 30: 404–415.

28.

Vonk

Fagherazzi

. The determination of the crystallinity in glass-ceramic materials by the method of Ruland. J Appl Crystallogr 1983; 16: 274–276.

29.

Beganskienė

Sirutkaitis

Kurtinaitienė

FTIR, TEM and NMR investigations of Stöber silica nanoparticles. Mater Sci (Medžiagotyra) 2004; 10: 287–290.

30.

Colthup

Daly

Wiberly

. Introduction to Infrared and Raman Spectroscopy. New York: Academic Press, 1990.

31.

Giordano

Borland

, et al. Mechanical properties of dense polylactic acid structures fabricated by three dimensional printing. J Biomater Sci Polym Ed 1997; 8: 63–75.

32.

Cavalcante

Toledo

ALMM

Rodrigues

EJR

, et al. Correlation between traditional techniques and TD-NMR to determine the morphology of PHB/PCL blends. Poly Test 2017; 58: 159–165.

33.

Zhijiang

Haizheng

, et al. Poly (hydroxybutyrate)/cellulose acetate blend nanofiber scaffolds: preparation, characterization and cytocompatibility. Mater Sci Eng: C 2016; 58: 757–767.

34.

Song

Chen

Zhu

, et al. Preparation and characterization of phenolphthalein polyethersulfone/silica nanofibrous membranes by solution blowing. J Thermoplast Compos Mater 2019: 32: 746–760.

35.

Musić

Filipović-Vinceković

Sekovanić

. Precipitation of amorphous SiO2 particles and their properties. Braz J Chem Eng 2011; 28: 89–94.

36.

Tavares

Menezes

Dutra Filho

, et al. Surface-coated polycaprolactone nanoparticles with pharmaceutical application: structural and molecular mobility evaluation by TD-NMR. Poly Test 2017; 60: 39–48.

37.

Sebastião

Monteiro

MSSB

Brito

, et al. Conventional and fast field cycling relaxometry study of the molecular dynamics in polymer nanocomposites for use as drug delivery systems. J Nanosci Nanotechnol 2016; 16: 7539–7545.

38.

Rodrigues

EJR

Cavalcante

Tavares

MIB

. Time domain NMR evaluation of poly(vinyl alcohol) xerogels. Polímeros 2016; 26: 221–227.

39.

Glatter

Kratky

. Small angle X-ray scattering. London: Academic Press, 1982.

40.

Balta-Calleja

Vonk

. X-ray scattering of synthetic polymers. Amsterdam: Elsevier, 1989.

41.

Owen

Bergmann

. On the fractal character of polymer spherulites: an ultra-small-angle X-ray scattering study of poly[(R)-3-hydroxybutyrate]. Poly Inter 2004; 53: 12–14.

42.

Brinker

Scherer

. Sol-gel science: the physics and chemistry of sol-gel processing. Cambridge: Academic press, 2013.

43.

Siouffi

. Silica gel-based monoliths prepared by the sol–gel method: facts and figures. J Chromatogr A 2003; 1000: 801–818.

44.

Freltoft

Kjems

Sinha

. Power-law correlations and finite-size effects in silica particle aggregates studied by small-angle neutron scattering. Phys Rev B 1986; 33: 269.

45.

Teixeira

. Small-angle scattering by fractal systems. J Appl Crystallog 1988; 21: 781–785.