Graphene Oxide versus Carbon Nanofibers in Poly(3-hydroxybutyrate-co-3-hydroxyvalerate) Films: Degradation in Simulated Intestinal Environments

Poly(3-hydroxybutyrate-co-3-hydroxyvalerate) (PHBV) is a microbial biodegradable polymer with a broad range of promising industrial applications. The effect of incorporation of low amounts (1% w/w) of carbon nanomaterials (CBNs) such as 1D carbon nanofibers (CNFs) or 2D graphene oxide (GO) nanosheets into the PHBV polymer matrix affects its degradation properties, as it is reported here for the first time. The study was performed in simulated gut conditions using two different media: an acidic aqueous medium (pH 6) and Gifu anaerobic medium. The results of this study showed that the incorporation of low amounts of filamentous 1D hydrophobic CNFs significantly increased the degradability of the hydrophobic PHBV after 3 months in simulated intestinal conditions as confirmed by weight loss (~20.5% w/w in acidic medium) and electron microscopy. We can attribute these results to the fact that the long hydrophobic carbon nanochannels created in the PHBV matrix with the incorporation of the CNFs allowed the degradation medium to penetrate at ultrafast diffusion speed increasing the area exposed to degradation. However, the hydrogen bonds formed between the 2D hydrophilic GO nanosheets and the hydrophobic PHBV polymer chains produced a homogeneous composite structure that exhibits lower degradation (weight loss of ~4.5% w/w after three months in acidic aqueous medium). Moreover, the water molecules present in both degradation media can be linked to the hydroxyl (-OH) and carboxyl (-COOH) groups present on the basal planes and at the edges of the GO nanosheets, reducing their degradation potential.

The Raman spectra of the GO and the CNFs used in this study were also previously measured [60] (Figure 1h,i). The GO can be classified as multilayer GO according to the GRAPHENE Flagship Project of the European Union for the unequivocal classification of these materials [61]. The Raman spectroscopy of the multilayer GO showed a D band intensity/G band intensity ratio (ID/IG ratio) of 0.92, while the CNFs showed a ID/IG ratio of 1.51, which can be attributed to the higher degree of disorder of CNFs [62]. The highresolution transmission electron microscopy showed the CNFs as one-dimensional hollow filaments with a wide range of diameters (22.7 ± 11.9 nm) and lengths (737.8 ± 522.4 nm). However, the multilayer GO exhibited a morphology of 2D nanosheets with average lateral dimension of 153.8 ± 57.2 nm [60]. The energy-disperse X-ray spectroscopy results showed C/O ratios of 31.3 and 15.4 for the CNFs and GO, respectively. The zeta potential of these GO nanosheets showed a surface charge of −28.94 mV (measured in in water/methanol 98/2) [63]. However, the surface charge of the CNFs used in this study showed a much lower value of −15.4 mV (measured in water at PH = 7) [64]. The dynamic Figure 1. High-resolution transmission electron microscopy images of 2D graphene oxide (GO) nanosheets (a) and 1D carbon nanofibers (CNFs) (c), transmission electron microscopy images of PHBV/GO (b) and PHBV/CNFs (d), and photographs of the PHBV (e), PHBV/GO (f) and PHBV/CNFs (g) films. Reprinted with permission from ref. [43]. Copyright 2020 Elsevier. Raman spectra of CNFs (h) and GO nanosheets (i). Reprinted with permission under a Creative Commons CC BY 4.0 License from ref [60]. Copyright 2020 MDPI.
The Raman spectra of the GO and the CNFs used in this study were also previously measured [60] (Figure 1h,i). The GO can be classified as multilayer GO according to the GRAPHENE Flagship Project of the European Union for the unequivocal classification of these materials [61]. The Raman spectroscopy of the multilayer GO showed a D band intensity/G band intensity ratio (ID/IG ratio) of 0.92, while the CNFs showed a ID/IG ratio of 1.51, which can be attributed to the higher degree of disorder of CNFs [62]. The highresolution transmission electron microscopy showed the CNFs as one-dimensional hollow filaments with a wide range of diameters (22.7 ± 11.9 nm) and lengths (737.8 ± 522.4 nm). However, the multilayer GO exhibited a morphology of 2D nanosheets with average lateral dimension of 153.8 ± 57.2 nm [60]. The energy-disperse X-ray spectroscopy results showed C/O ratios of 31.3 and 15.4 for the CNFs and GO, respectively. The zeta potential of these GO nanosheets showed a surface charge of −28.94 mV (measured in in water/methanol 98/2) [63]. However, the surface charge of the CNFs used in this study showed a much lower value of −15.4 mV (measured in water at PH = 7) [64]. The dynamic light scattering (DLS) technique showed a particle hydrodynamic size of the GO nanosheets that ranged from 300 to 500 nm, depending on the nanofluid (polyethylene glycol, 1-octadecanothiol, or Triton X-100 surfactant) used for these measurements [65]. However, the DLS hydro-dynamic sizes of the 1D nanomaterial showed to be much larger, ranging from 811.2 to 1142 nm, depending also on the nanofluid used (DMEM or water, respectively) [66].

Nanocomposite Film Preparation
Dichloromethane was used to dissolve the PHBV (with and without 1% w/w of GO or CNFs). The PHBV-based films were prepared by solvent casting in Petri dishes as previously described [43]. Sonication of CBNs for 30 min was performed in dichloromethane before mixing these dispersions with the PHBV solution. The produced films were finally vacuumdried at 50 • C for 48 h to constant weight. The neat PHBV sample film and nanocomposite films with GO nanosheets or CNFs are hereafter referred to as PHBV, PHBV/GO and PHBV/CNFs respectively.

Degradation Analysis
The quantitative results of this study were based on the weight control of the three types of material films (PHBV, PHBV/GO and PHBV/CNFs) for 3 months in an acid aqueous degradation medium (pH 6). The acid aqueous medium was composed of deionized water with a pH adjusted to 6 with a 10% HCl solution in water. Sodium azide (0.4 mg/mL) was added to the degradation media as fungicide and the pH was monitored during the study to remain constant. Initially, the materials were vacuum dried at 60 • C to determine the dry mass of each material. Samples were introduced into 50 mL falcon tubes containing the degradation medium (20 mL per flask) and placed in the shaking bath with a shaking frequency of 20 cycles per minute at 37 • C. The experimental degradation times were 1, 2 and 3 months. Three replicates of each material per time were analyzed to provide reproducible results. After each selected degradation time, the materials were extracted and dried to constant weight to determine the weight loss and morphological changes.
Gifu anaerobic medium (GAM) broth was additionally used as degradation medium. GAM composition is as follows: 1.0% (w/v) peptic digest of animal tissue, 0.3% papaic digest of soybean meal, 1.0% protease peptone, 1.35% digested serum, 0.5% yeast extract, 0.22% beef extract, 0.12% liver extract, 0.3% glucose, 0.25% potassium dihydrogen phosphate, 0.3% sodium chloride, 0.5% soluble starch, 0.03% L-cysteine hydrochloride, and 0.03% sodium thioglycollate, final pH 7.3 ± 0.1 [25]. Samples were introduced into 50 mL falcon tubes containing GAM medium (20 mL per flask) and sodium azide (0.4 mg/mL) and placed in the shaking bath with a frequency of 20 cycles/min at 37 • C. The degradation times were 1, 2 and 3 months with three replicates of each material. After each selected degradation time, the samples were extracted and dried to analyze changes in the surface and cross-section morphology.

Electron Microscopy Observation
The qualitative results of the degradation study were evaluated analyzing the morphology (surface and cross-section) of neat PHBV and PHBV with 1% w/w of GO and CNFs by high-resolution field emission scanning electron microscope (HRFESEM) using a Gemi-niSEM 500 microscope (Carl Zeiss, Jena, Germany) at the beginning and at the end of the degradation time (0 and 3 months). The samples were previously coated with a platinum layer by an EM MED020 sputter coater (Leica, Wetzlar, Germany). The cross-section was observed after cryofracture.

Statistical Analysis
Data expressed as mean ± standard deviation (SD). Statistical analysis was performed with the GraphPad Prism 6 (GraphPad Software Inc., San Diego, CA, USA) by oneway ANOVA and subsequent Tukey's correction. Significant results were assumed at p-values < 0.05 (95% confidence).

Degradation in Simulated Intestinal Conditions
The results of weight loss after the degradation assay in the acid aqueous medium are shown in Figure 2.

Statistical Analysis
Data expressed as mean ± standard deviation (SD). Statistical analysis was performed with the GraphPad Prism 6 (GraphPad Software Inc., San Diego, CA, USA) by one-way ANOVA and subsequent Tukey's correction. Significant results were assumed at p-values < 0.05 (95% confidence).

Degradation in Simulated Intestinal Conditions
The results of weight loss after the degradation assay in the acid aqueous medium are shown in Figure 2.  Weight loss of PHBV, PHBV/GO and PHBV/CNFs after degradation in acid aqueous medium (pH = 6) for 1, 2 and 3 months. Data are shown as mean ± standard deviation. Significant differences between samples were calculated by one-way ANOVA with Tukey's correction. Only the significant differences are indicated in the graph with their level of significance: * p > 0.05; ** p > 0.01; *** p > 0.001.
The weight loss study shows clearly that the incorporation of 1% w/w 1D carbon nanomaterial, CNFs, into the PHBV polymer matrix significantly enhances the degradation properties of the PHBV biopolymer. These results are in good agreement with the HRFESEM cross-section and surface images shown in Figures 3 and 4.
The cross-section images show higher visual degradation in PHBV/CNFs, both in acid aqueous medium (pH = 6) and GAM (Gifu) broth after 3 months (Figure 3). However, the addition of 2D nanomaterial GO (samples PHBV/GO) does not seem to affect substantially the morphology of the samples compared to the images of neat PHBV film. Signs of degradation (changes in the porous structure) can be seen in both neat PHBV and PHBV/GO, although not as evident as those observed in PHBV/CNF samples.
The addition of 1% w/w of CNFs and GO nanosheets produce slight changes in the surface, although the irregular morphology remains.The surface images of PHBV/CNFs show clearly again the remarkable degradation after 3 months in acid aqueous medium (pH = 6) in good agreement with the weight loss reported in Figure 2. The sample with the 2D nanomaterial (GO nanosheets) also shows signs of surface degradation, although not as noticeable as PHBV/CNF. These signs seem to be related to a first stage of degradation, which has not yet translated into a statistically significant weight loss with respect to pristine PHBV in the analyzed period ( Figure 2).
Regarding the surface degradation of PHBV/CNFs films in GAM (Gifu) broth after 3 months, no signs of degradation are observed as evident as in the degradation in the acid aqueous medium. GAM medium contains organic components, such as yeast extract, beef extract and liver extract, among others (see Section 2.3.1) that are deposited on the materials' surface, preventing the clear visualization of the surface structure. However, as stated previously, the cross-section images showed evident morphological differences of PHBV/CNFs after intestinal degradation compared to PHBV/GO and pure PHBV (Figure 3). Due to the presence of aggregates of organic matter on the surface, observed in all the samples, it was not possible to perform the weight loss analysis, so only the morphological analysis of the samples was carried out after their immersion in GAM medium. It can be noted that the organic layer in pristine PHBV is less dense than in the PHBV/GO and PHBV/CNFs nanocomposites. Several studies have shown that carbon nanomaterials, such as GO, reduced GO (rGO) and CNFs, increase protein adsorption, depending on their concentration or the degree of reduction (in the case of rGO) [67,68]. Proteins spontaneously and quickly adsorb onto carbon nanoparticles, forming the socalled protein corona and covering the surface [69]. The organic components of GAM broth contain a high content of proteins, which are adsorbed to a greater extent on the PHBV/GO and PHBV/CNFs nanocomposites, increasing the density of the surface layer.  The cross-section images show higher visual degradation in PHBV/CNFs, both in acid aqueous medium (pH = 6) and GAM (Gifu) broth after 3 months (Figure 3). However, the addition of 2D nanomaterial GO (samples PHBV/GO) does not seem to affect substantially the morphology of the samples compared to the images of neat PHBV film. Signs of degradation (changes in the porous structure) can be seen in both neat PHBV and PHBV/GO, although not as evident as those observed in PHBV/CNF samples.
The surface of neat PHBV, shown in Figure 4, shows an irregular morphology, characteristic of semicrystalline polymers [30,63].  The addition of 1% w/w of CNFs and GO nanosheets produce slight changes in the surface, although the irregular morphology remains.The surface images of PHBV/CNFs show clearly again the remarkable degradation after 3 months in acid aqueous medium (pH = 6) in good agreement with the weight loss reported in Figure 2. The sample with the 2D nanomaterial (GO nanosheets) also shows signs of surface degradation, although not as noticeable as PHBV/CNF. These signs seem to be related to a first stage of degradation, which has not yet translated into a statistically significant weight loss with respect to pristine PHBV in the analyzed period ( Figure 2).
Regarding the surface degradation of PHBV/CNFs films in GAM (Gifu) broth after 3 months, no signs of degradation are observed as evident as in the degradation in the acid aqueous medium. GAM medium contains organic components, such as yeast extract, beef extract and liver extract, among others (see Section 2.3.1) that are deposited on the materials' surface, preventing the clear visualization of the surface structure. However, as stated previously, the cross-section images showed evident morphological differences of PHBV/CNFs after intestinal degradation compared to PHBV/GO and pure PHBV ( Figure  3). Due to the presence of aggregates of organic matter on the surface, observed in all the samples, it was not possible to perform the weight loss analysis, so only the morphological analysis of the samples was carried out after their immersion in GAM medium. It can be noted that the organic layer in pristine PHBV is less dense than in the PHBV/GO and PHBV/CNFs nanocomposites. Several studies have shown that carbon nanomaterials, such as GO, reduced GO (rGO) and CNFs, increase protein adsorption, depending on their concentration or the degree of reduction (in the case of rGO) [67,68]. Proteins spontaneously and quickly adsorb onto carbon nanoparticles, forming the so-called protein corona and covering the surface [69]. The organic components of GAM broth contain a Therefore, the results of this study show that the incorporation of a low amount (1% w/w) of filamentous 1D hydrophobic CNFs, which are longer nanoparticles and possess lower negative charge than the GO nanosheets [63][64][65][66], significantly increased the degradability of the hydrophobic PHBV after 3 months in simulated intestinal media (both acid aqueous and GAM media). The long filamentous hydrophobic CNFs incorporated into the PHBV polymer matrix creates carbon-based nanochannels through which water (in this case aqueous medium) can penetrate at ultrafast speed according to previous studies [70][71][72], increasing the total area of PHBV chains exposed to degradation. In addition, the hydroxyl groups (-OH) and carboxyl (-COOH) present on the basal planes and at the edges of the GO nanosheets can retain water molecules present in the degradation medium, which reduces their degradation potential. These reactive groups form hydrogen bonds with the PHBV polymer chains producing a homogeneous stable composite structure that prevents the degradation. The weight loss in this nanocomposite presents a similar degradation behavior compared to neat PHBV films, which remains unaltered after the degradation period analyzed in this study.

Conclusions
PHBV is a microbial renewable biopolymer with great promise in a wide range of biomedical applications, such as controlled drug delivery and tissue engineering, due to its biodegradation and biocompatibility properties. We have demonstrated previously that the physical and biological properties of PHBV can be enhanced with the incorporation of low amounts (1% w/w) of 2D or 1D CBNs such as GO and CNFs, respectively. In this study, we show that the degradation time of PHBV in simulated intestinal environments can be reduced with the incorporation of this low amount of CNFs, while the addition of GO nanosheets does not affect the degradability of PHBV in such a significant way. This different degradation behavior can be attributed to the different chemical interactions between hydrophobic PHBV with the hydrophobic CNFs or with the hydrophilic GO. The presence of CNFs in the PHBV polymer produces changes in the structure of the PHBV that accelerate the degradation process, as we have confirmed by the weight loss and electron microscopy analysis.