A Biodegradable Composite of Poly (3-hydroxybutyrate-co-3-hydroxyvalerate) (PHBV) with Short Cellulose Fiber for Packaging

,


Introduction
Because of the global growth in plastic use and poor plastic waste management, most plastic wastes end up on the earth's surface, followed by marine and fresh water bodies.Over long-time scales, plastics degrade into particles with sizes smaller than 5 mm, known as microplastics 1 that cause a much greater problem in microbiology or plants than can be observed 2,3 .
Thus, the replacement with biodegradable polymers is a good alternative to mitigate this problem, especially for articles whose period of use and disposal is short, mainly those that are discarded after a single use, such as packaging, which accounts for the majority of polymeric waste.Some biopolymers are already used on an industrial scale, such as PLA, but despite their biodegradability, they are only compostable in an industrial environment and thus can still generate polymeric waste.In this context, poly (3-hydroxybutyrate-co-3-hydroxyvalerate) (PHBV) has characteristics similar to polypropylene (PP), is nontoxic, compostable and has high crystallinity, which makes it competitive to replace PP and has been applied as food packaging material for example [4][5][6] .Compostable polymers can be completely degraded, generating water molecules and carbon oxide as byproducts, making them ideal to avoid polymer waste generation 7 .
In terms of pilot scale production and thermal processability, studies of PHBV indicate its potential for packaging and foodservice applications due to its good oxygen/water vapor barrier properties, but some drawbacks still greatly limit its processing and application, such as inherent brittleness, physical aging at room temperature and narrow processing windows.In addition, its degradation can occur even during storage, further reducing the molar mass and consequently the properties, making commercial use of this polymer difficult.The short processability window and fluid instability are other negative features of the application of this pure polymer, making it difficult to expand its use in packaging 8,9 , but its excellent gas barrier property makes it attractive in packaging application 10 .
The formation of composite material has been pointed out as a possibility for using this polymer, in addition to reducing the amount of polymeric matrix.Natural fibers can be used as reinforcement of biocomposites in the PHBV matrix, producing completely biodegradable composites with a rapidly biodegradable matrix, which increases their applications and attractiveness [11][12][13] .Cellulose is an attractive source of filler, since in addition to being widely used, in the polymer matrix in packaging [14][15][16][17] .Recycled paper has potential as a source of low-cost cellulose fibers 18,19 .
In the process of reusing cellulose fibers, reversible and irreversible chemical and physical changes occur.These changes are inherent to the fibers themselves, and can include loss of swelling capacity and reduction of moisture, these being especially important in packaging applications.
Paper is a resource with high abundance of cellulose fiber, in particular, China produces hundreds of millions of tons of waste paper every year, which can be used with great economic benefit 20,21 .
The aim of this work was to produce fully biodegradable rigid packaging using a compostable polymeric matrix with short fiber cellulose residues.

Materials
PHBV powder with 4 wt.% of HV was supplied by PHB Industrial S.A. (São Paulo, Brazil).The number (Mn) and weight (Mw) average molar mass, and polydispersion index (I=Mw/Mn) were measured by gel permeation chromatography (GPC) with a Shimadzu LC Solution chromatograph using a chloroform column.For this, 6.0 mg of powdered PHBV was evaluated with an injected volume of approximately 2 µL.The evaluation of molar mass by GPC indicated Mn and Mw values of 80.10 3 and 170.10 3 g.mol - , respectively.Cellulose paper (sulphite A4 sheet of paper) after use was reused by passing through a simple fiber separation step.The shredded paper was placed in a beaker with water and after soaking for 30 minutes it was submitted to agitation with an IKA RW 20 digital propeller agitator for 30 minutes at 30 rpm until the fibers were separated in solution.Then the fibers were dried in an oven for 40 minutes at 80 °C.

Preparation of composites
A pre-mixing step of the PHBV and Cellulose (CE), PHBV/CE composite, was performed employing two systems.First, a Marconi MA500 mill with sintered ceramic balls was used, and a second attempt was made by replacing the ceramic balls with metal rollers.The other premixing method was applied using an Ika A11 analytical mill for 30 s, this was adopted as the method of premixing the composite.Compositions with 0, 5, 10, 15 and 20% by mass of cellulose were obtained to produce formulations with PHBV/CE proportions of 100/00, 95/05, 90/10, 85/15 and 80/20.Each formulation was extruded in a twin-screw extruder (TeckTril DCT model) equipped with 10 temperature zones, ranging from 190 to 110 °C from the feed to die, and rotating at 43 rpm.The test specimens for characterization were obtained by pressing the bulk samples from the extrusion at a temperature of 200 ° C for 300 s, with pressure of 5 tons, followed by cooling in a cold press for 200 s.

Characterization 2.3.1. Scanning electron microscopy (SEM)
A FEI Quanta 400 scanning electron microscope equipped with energy dispersive spectroscopy (EDS) with a Bruker-AXS Quantax detector, acceleration voltage of 20 kV and a working distance of 11 mm was employed to analyze samples coated with gold.

Fourier-transform infrared spectroscopy (FTIR)
The composition of the samples of PHBV/CE was evaluated using Fourier-transform infrared spectroscopy (FTIR) with a Nicolet 6000 model from Thermo Fisher Scientific equipped with an attenuated total reflectance (ATR) accessory.
The analysis was performed in the region of 4000-650 cm -1 , with 64 scans.

Thermogravimetric analysis (TGA)
The thermal stability and mass loss of the PHBV/CE composites were evaluated by thermogravimetric analysis using a PerkinElmer Pyris 1 analyzer with a heating range of 25 °C to 500 °C and a heating rate of 10 °C/min under N 2 atmosphere with a flow rate of 60 mL/min.

Differential scanning calorimetry (DSC)
The thermal behavior of the PHBV/CE composites with different mass ratios was evaluated by differential scanning calorimetry (DSC) with a Hitachi DSC 7020 thermal analysis system using heating and cooling rates of 10 °C/min and nitrogen atmosphere flow (50 mL/min).Two heating and one cooling cycle were carried out.The first heating cycle was conducted from 0 to 200 °C, followed by cooling to -20 °C and subsequent heating from -20 °C to 200 °C.The first heating run was used to erase the thermal history of the samples, while their thermal parameters were obtained from the second heating run.The crystallinity degree of the samples was calculated by the following Equation 1: Where is the enthalpy of crystalline PHBV 100% (146.6 J.g -1 ) 22 , f H ∆ is the enthalpy of the sample, and W is the cellulose fiber (CE) fraction present in the sample.

Contact angle
Two methods were used to measure the contact angle: the force-based method (or Wilhelmy method) and the sessile drop method using a Ramé-Hart NRL A 100-00 goniometer, measuring speed at 3 mm/min.For the force-based method, the advancing contact angle was measured by obtaining the perimeter of rectangular samples, which were then placed in a tensiometer (Kruss K100K) along with the surface tension of the liquid (water) to determine the wettability.The sessile drop method involved depositing a 2 μl drop of distilled water onto the surface of each sample at room temperature, and measuring the contact angles in triplicate.

Impact resistance
The impact tests were carried out in accordance with ASTM D256 23 with an impact pendulum (CEAST 9050), with a 2.7 J impact hammer.

Vicat softening temperature
The test was performed according to ASTM D-1525 24 , with a CEAST HV3 6911.000tester.At least three Vicat softening temperature test repetitions were conducted for each formulation and the averages were calculated.The test parameters were imposed stress of 50 N and initial temperature of 25 o C.

ANOVA
To verify the effect of adding of cellulose waste (CE) on the PHBV, an analysis of variance (ANOVA) Post hoc Test-Fisher's Least Significant difference (LSD) was performed adopting a 0.05 significance level (p < 0.05) 25 .

Results and Discussion
Figure 1, shows the SEM images of the cellulose fibers from recycled paper (Figure 1a) and the cross-section surface of the PHBV/CE composites obtained by varying the concentration of cellulose fibers, Figure 1b-f.The increase in the polymer/filler mass ratio (PHBV/CE) generated larger defects on the composite surface, such as a rougher surface as a function of rising CE concentration, revealing the incorporation of cellulose fibers in the polymeric matrix.Morphologies characteristic for this filler as reported in the literature 26,27 .
The FTIR spectra obtained for PHBV, cellulose fibers and composites in the wavenumber regions between 4000 cm -1 and 750 cm -1 and between 1500 cm -1 and 600 cm -1 are shown in Figures 2 and 3 respectively.Peaks near 2975 and 2745 cm -1 (C-H groups) were observed.A strong and sharp band at 1724 cm -1 , characteristic of the axial deformation of the ester group (C=O), of PHBV was observed.The band at 1380 cm -1 corresponds to the asymmetric deformation of the C-H bond of the methyl group.The characteristic band of symmetrical and asymmetrical stretching of the C-O-C group is present at 1132 cm -1 .
The bands representing the axial deformation of the C-O and C-C bonds are present at 1275, 1052 cm -1 and 979 cm -1 respectively (Figure 3), whereas this broadening and increased intensity in the bands allocated at 1050 cm -1 may suggest some interaction and presence of cellulose in the PHBV/CE composite 28,29 .Studies of PHBV loaded with performed by Yu et al. 30 , Yu and Qin 31 and Ponjavic et al. 32 , presented similar spectra to that observed by us.
Figure 4a and b and Table 1 shows the thermal transitions and crystallization behavior of neat PHBV and PHBV/CE composites during the first and the second heating runs of DSC assay.The first heating destroyed the entire crystalline structure, so the thermal history of the samples was eliminated.In the second heating, the increase in cellulose content (5 to 20 wt%) promoted a slight reduction in the melting point of PHBV (T m2 ) compared with the neat PHBV (169.4 °C) showed in Table 1.Also, the incorporation of cellulose may favor the formation of a small population of more imperfect crystals which are consistent with previous reports in the literature 33 .This imperferct crystals, melt at a lower temperature (~150 °C) as observed in Figure 4b.The addition of cellulose (5 to 15 wt%) improved the crystallization degree of PHBV in aproximatly 15 wt%, for these formulations.However, the increase in cellulose content for 20 wt% showed a similar value for the neat PHBV (~ 42 wt%).
Figure 4c and d and Table 2 shows the TGA results of PHBV, cellulose fibers and composites.The cellulose had lower thermal stability than neat PHBV and its composites, considering the T onset values.The cellulose fibers presented two thermal events related to mass loss.At a temperature of 41 °C there was mass loss, which corresponds to the release of volatiles, mainly absorbed water.
The second mass loss process started at a temperature of 293 °C, with maximum degradation temperature of 349 °C, ending at 394 °C.According to Cordeiro et al. 34 , this temperature range may be related to the thermal degradation of fiber components liganocellulosic substances, mainly cellulose, which occurs at 340 ºC.For PHBV, the mass loss was gradual and uniform with increasing temperature.The thermal degradation of PHBV occurred in two stages of mass loss.The first mass loss process started at a temperature at 276 °C and ended close to 332 °C, with maximum degradation temperature of 294 °C.The second mass loss process started at 354 °C, with maximum degradation temperature of 362 °C, ending at 388 °C.This temperature is almost 26 and 104 °C, respectively, higher than that reported in literature (T onset = 250 °C) 35 .
With respect to the first degradation event of the PHBV/CE composites, the increase in cellulose content (5 to 15 wt%) did not change the thermal stability of the polymer.The TGA curves of the composites showed practically the same thermal degradation behavior, with mass variation onset temperature around 270 ºC and maximum degradation temperature around 293 °C.The PHBV/CE composites with the highest cellulose content (20 wt%) were less thermally stable.The increase in the cellulose percentage caused the approximation of molecular chains, leading to the formation of agglomerates 36 .Similar results were observed by Bhardwaj et al. 37 , who reported that the TGA curves of PHBV composites undergo thermal degradation above 250°C and that T onset of thermal degradation of the PHBV composites is like that of pure PHBV except for the composite with higher cellulose content.Considering the second degradation event, all materials showed similar behavior, with reduced thermal stability compared to pure PHBV.Different amounts of residues were formed in the PHBV composites.The amount of residue was higher in composites with a higher amount of cellulose fibers.
Figure 5 shows the XRD results.The XRD of PHBV showed two main diffraction peaks, which are typical of a structure of PHB in 2θ≈13.4°and 16.8° relative to crystallographic planes (020) and (110).The other peaks in 2θ near to 20.0°; 21.4°; 22.5°; 25.4° and 27.1°, are related to crystallographic planes (021), (100), (101), (111), ( 130) and (040).Studies of PHBV performed by Macedo et al. 38 , and Wang et al. 39 , presented diffraction peaks like that observed by us.Cellulose diffractograms showed typical behavior of semicrystalline structures.Two well-defined peaks referring to the amorphous and crystalline phases are visualized.The lowest intensity peak refers to the amorphous phase and is close to 2Ө=17°, whereas the most intense peak is related to the crystalline portion and is close to 2Ө=21°4 0,41 .The increase in the cellulose content (5 to 20% in weight) promoted a small decrease in the intensity of crystalline phases referring to the plane of PHBV in the composites as the load concentration increased.The composite with the lowest cellulose content (95/5) showed a variation in the peaks close to 2Ө=22°, suggesting a variation in the crystalline structure of the PHBV polymeric matrix.
Roughness and porosity are features that make the analysis of contact angle of cellulose samples difficult 42,43 .Therefore, performing contact angle measurements by tensiometry on porous materials was indicated as a good alternative 44 .Figure 6 displays the force-tensiometry curve profile as a function of sample displacement, highlighting that the curve of pure PHBV and the sample containing 5% cellulose exhibited similar behavior, whereas the remaining compositions displayed similarity differently.Similarly, such behavior was observed by Szymczyk and Podkościelna 45 in a cellulose composite.The dynamic contact angle results measured by using force tensiometry and drop method can be seen in Table 3.The contact angle of 78° of the pure polymer sample indicated its hydrophilic character, but the insertion of cellulose promoted characteristics of hydrophobicity, as indicated by the increase in the angle to 100.52 o .As indicated by the micrographs, the insertion of the cellulose promoted an increase in surface roughness, drastically altering the surface characteristics, which was determinant in the wettability of the composites 46 .Similarly, according to the drop method, the increase in cellulose content indicated a trend towards increased hydrophobicity of the material, with values ranging from 77.7 ± 2.0 for pure PHBV to 90.0 ± 3.8 for the composition with 5% cellulose.
Figure 7a shows the impact resistance results of the PHBV and composites.The PHBV sample had higher impact resistance.For PHBV/CE composites, the increase in cellulose content (5 to 20 wt%) reduced the impact resistance.For PHBV composites containing 20% cellulose, the impact strength reduced by 40% compared with the neat PHBV.This decrease occurred because the addition of cellulose causes a decrease in the mobility of the polymeric chains, reducing the capacity of the composite to absorb energy.This decrease in impact resistance may be a consequence of cellulose agglomerations in the composites, acting as stress concentration points, from which cracks propagate.Similar results were observed by Ahankari et al. 47 who reported that the addition of lignocellulosic fibers to the PHBV matrix promoted reduction impact strength of composites.According to the authors, this was due to the formation of fiber agglomerates in the matrix, increasing the stresses and making the material more brittle, and consequently reducing the energy needed to break the samples.On the other hand, the impact resistance shown significance variation for samples with major amount of filler (CE).Samples 80/20 and 85/15 presented LDS > 0.0168, referring statistically significancy compared with neat PHBV (100/00) sample.However, for the samples with minor amount of CE, the variation of impact resistance was minor significative, this significance was calculated by Analyze of Variance (ANOVA) post Hoc Test -Fisher's Least Significant Difference (LSD) (calculated LSD = 0.995).
The thermal resistance of the neat PHBV and PHBV/CE composites was evaluated by measuring the Vicat softening temperature (VST).Figure 7b shows the relationship between Vicat and the incorporated cellulose content in the composite.The PHBV showed Vicat temperature of 95 °C, while the composite with 5% CE showed Vicat of 98°C and the composites with 10, 15 and 20% presented Vicat softening temperature close to 103 °C, increases of 3% and 8%, respectively, in relation to the Vicat temperature of the neat polymer.The increase in Vicat results from the restriction of the mobility of the polymeric chains in the presence of the cellulose, increasing the dimensional stability of the PHBV, so higher temperatures are necessary to reach a given deformation level, increasing the dimensional stability of the polymeric matrix 48 .The increase in Vicat softening point caused by the presence of cellulose is an important technological contribution to the production of composites, since one of the limitations of PHBV for technological applications is its short processability.Therefore, the composites can be used in environments with higher temperatures than possible with PHBV alone while maintaining the dimensional stability.
ANOVA -(LSD) (calculated LSD = 0.869) for Vicat softening temperature indicates that increasing levels of cellulose fibers as the filler on PHBV matrix were statistically significant between 100/00 and major amount of CE 90/10, 85/15, 80/20 samples (LSD > 0.0168).On the other hand, the increase of cellulose for minor relation (95/05) shows not influential on the impact resistance (LSD < 0.883) demonstrating that with 5 wt.% of CC don't has influence in the variation on softening temperature.

Conclusions
An analytical mill was used to promote the dispersion of cellulose fibers making PHBV powder and cellulose fibers more dispersed and homogeneous.In the SEM images, it was possible to observe the increase of irregularity of the material surface due to the increase of roughness that was a dominant effect of the increase of the contact angle.The FTIR analysis suggested that the cellulose fibers acted as nucleating agents based on the lower intensity of the peaks in the regions of 980 and 1230 cm -1 .The TGA results indicated two thermal events concerning the polymer and cellulose where the composite samples presented close T onset values, except for the 80/20 composition, which started degrading at lower temperature.The DSC results indicated increased crystallinity of the polymer, in agreement with the FTIR results.The XRD peaks of the pure cellulose were less intense in relation to the polymer matrix, so the cellulose peaks were not prominent in the composite samples.In particular in sample 95/05 a change in the curve profile was noticeable.The increase in the amount of cellulose caused a reduction in the mobility of the molecules.In addition to the increase of crystallinity, there was a reduction in impact strength as well as an increase in the formation of agglomerates, which acted as stress concentrators.Finally, the Vicat softening temperature increased with increase of cellulose, which is very positive for PHBV, which has a short processability window.

Figure 4 .
Figure 4. Thermal behavior from cellulose fiber and formulations: DSC curves for samples with different relation PHBV/CE.(a) First heart cycle and (b) Second heat cycle; TG (c) and DTG (d) curves.

Figure 6 .
Figure 6.Contact angle measurements using the Wilhelmy method.

Table 1 .
DSC results -Thermal transitions of the second heating cycle for formulations.

Table 3 .
Contact angle results of formulations by Drop and Wilhelmy method: