Antimicrobial nanocomposites and electrospun coatings based on poly(3‐hydroxybutyrate‐co‐3‐hydroxyvalerate) and copper oxide nanoparticles for active packaging and coating applications

Active biodegradable poly(3-hydroxybutyrate-co-3-hydroxyvalerate) (PHBV) melt mixed nanocomposites and bilayer structures containing copper oxide (CuO) nanoparticles were developed and characterized. The bilayer structures consisted of a bottom layer of compression molded PHBV3 (3% mol valerate) coated with an active electrospun fibers mat made with CuO nanoparticles and PHBV18 (18% valerate) derived from microbial mixed cultures and cheese whey. The results showed that the water vapor permeability increased with the CuO addition while the oxygen barrier properties were slightly enhanced by the addition of 0.05 wt % CuO nanoparticles to nanocomposite films but a negligible effect was registered for the bilayer structures. However, the mechanical properties were modified by the addition of CuO nanoparticles. Interestingly, by incorporating highly dispersed and distributed CuO nanoparticles in a coating by electrospinning, a lower metal oxide loading was required to exhibit significant bactericidal and virucidal performance against the food-borne pathogens Salmonella enterica, Listeria monocytogenes, and murine norovirus. The biodisintegration tests of the samples under composting conditions showed that even the 0.05% CuO-coated structures biodegraded within 35 days. © 2017 Wiley Periodicals, Inc. J. Appl. Polym. Sci. 2017, 134, 45673.


INTRODUCTION
Among the different kinds of nanoparticles applied in many fields of human life, which range from energy production to industrial production processes and biomedical applications, those having antimicrobial activity are highlighted as they can be incorporated into a variety of polymer matrices used in daily life, producing novel biocide nanomaterials [1][2][3] . Metal and metal oxide nanoparticles based on gold, zinc, silver and copper have been reported to have extensive antimicrobial activity in a wide range of microorganisms, including viruses, fungi and bacteria 4 .
In particular, it has been used for centuries as biocide compound to disinfect liquids, solids and human tissues and more recently used as antifouling agent, water purifier, algaecide, fungicide, and bactericide for several applications 5 . In the biomedical and cosmetics industries, copper has been used in the forms of wound dressing 6 and facial mask 7 ; In the food area, metallic copper sheets have been evaluated as antimicrobial surfaces to inhibit the growth of enteropathogens 8 .
Copper and copper oxide bulk materials (micro-sized) have been also physically and chemically characterized and investigated as antimicrobial agents 9,10 , with the advantage of being easily mixed with polar liquids and polymers, relatively stable in terms of chemical and physical properties, and cheaper and less toxic for living organism than silver 11,12 . However, the antimicrobial use of copper could be successfully extended by using it in the nano-sized form and by embedding into polymeric matrices 13,14 . In fact, the incorporation of antimicrobials in a polymer matrix allows the gradual release of the biocide substance from package, controlling the microbial contamination along the shelf life of the food product at the same time that while reducing the amount of preservatives added within the bulk of food 15 .
Due to the suitability of the electrospinning technique to entrap active compounds within polymeric fibers and the intrinsic characteristics of these fibres such as very high specific surface and porosity, the electrospun materials are provide excellent candidates for a wide variety of applications 16 . Moreover, biopolymer-based fibers constitute ideal carriers for antimicrobials in active packaging applications because of their capacity to release antimicrobial substances in a controlled manner and the possibility of developing blends and multilayer systems 17 .
Biopolymers such as polyhydroxyalkanoates (PHAs) have been receiving much attention in recent years as biocompatible and biodegradable thermoplastics with potential packaging applications 18 . The most extensively studied polymer from the PHAs group is the poly(3-hydroxybutyrate), PHB. PHB is partially crystalline with a high melting temperature and high degree of crystallinity and rigidity 19,20 . To overcome these aspects, the copolymer obtained with the insertion of 3hydroxyvalerate (HV) units, known as poly(3-hydroxybutyrate-co-3hydroxyvalerate), PHBV, is usually employed to improve the handling properties of PHB films.
Although some studies about the incorporation of copper and copper oxide nanoparticles into polymeric and biopolymeric matrices such as polypropylene 21 , cotton 22 and others carbohydrate-based matrices such as chitosan, carrageenan, chitosan and carboxymethyl cellulose 23

Development of active PHBV films
Active films based on CuO nanoparticles and PHBV were developed using two different methods of preparation and two different loading. The samples code and composition are summarized in Table 1. Later, a post-annealing step was applied to form a continuous film by fibers coalescence. Fibers mats (14 wt.%) of c.a. 40 µm of thickness were placed onto PHBV3 films prepared by compression moulding as described above. This assembly was put in between hot plates hydraulic press (Carver 4122, USA) at

Morphology
The morphology of the electrospun fibers and the active films were studied by To obtain an accurate estimation of the average fibers and nanoparticles diameter, 200 to 300 measurements were done by means of the Adobe Photoshop CS4 software from the SEM and TEM micrographs in their original magnification.

Optical properties
Transparency of the neat PHBV3 film and coated systems was determined through the surface reflectance spectra in a spectrocolorimeter CM-3600d (Minolta Co., Tokyo, Japan) with a 10 mm illuminated sample area. The internal transmittance (Ti) of the samples was determined by applying the Kubelka-Munk theory 25 for multiple scattering to the reflection spectra where an increase in the spectral distribution of transmittance is associated with more homogeneous and transparent samples. Measurements were taken in triplicate for each sample by using both a white and black background.
Moreover, CIE-L* a* b* coordinates (CIE, 1986) were obtained by the infinite reflection spectra of the samples, using D65 illuminant/10 observer. Samples were evaluated per duplicate and three measurements were taken at random locations on each of the studied films.

Wide Angle X-Ray Diffraction Analysis (WAXD)
X-ray diffractograms of the CuO powder samples and films were recorded at room temperature using a Bruker AXSD4 Endeavour diffractometer with a Cu-Kα source (wavelength = 1.54178 Å). Peak fitting was carried out using IgorPro software package (Wavemetrics, Lake Oswego, Oregon). Gaussian function was used to fit the experimental diffraction profiles obtained. The crystallinity degree (X c ) of the films was taken as the ratio of the sum of areas under the crystalline diffraction peaks to the total area under the curve between 2θ = 5°and 40°.

Differential Scanning Calorimetry (DSC)
Thermal properties of the neat PHBVs films and its active nanocomposites were evaluated by DSC using a Perkin-Elmer DSC 8000 thermal analysis system under nitrogen atmosphere. The sample treatment consisted of a first heating step from 0°C to 200°, a subsequent cooling down to -50°C and a second heating step up to 200°C. The heating and cooling rates for the runs were 10°C/min and the typical sample weight was 3 mg. The first melting endotherm, and the controlled crystallization at 10°C/min from the melt was were analysed. To ensure reliability of the data obtained, heat flow and temperature were calibrated using indium as a standard. The tests were done, at least, in triplicate.

Mechanical properties
Tensile tests were performed according to ASTM Standard D638 using a Universal Testing Machine (Shimadzu AGS-X 500N). Before testing, the samples were allowed to reach the equilibrium under ambient conditions (25°C and 50% R.H. for 24 hours) and cut in dumbbell shaped specimen. Elastic modulus, tensile strength, and elongation at break were determined from the stress-strain curves, estimated from force-distance data obtained for the different films. At least, three specimens of each film were tensile tested as to obtain statistically meaningful results.

Water Vapour Permeability (WVP)
Water vapour permeability (WVP) was determined according to the ASTM E96 (ASTM 2011) gravimetric method, using Payne permeability cups (Elcometer SPRL, Hermelle/s Argenteau, Belgium). Cells containing distillate water were placed inside a desiccator at 0% RH and 24°C and the water weight loss through a film area of 0.001 m 2 was monitored. WVP was calculated from the steady-state permeation slopes obtained from the regression analysis of weight loss data over time. All measurements were performed in triplicate.

Oxygen transmission rate (OTR) measurements
The oxygen permeability coefficient was derived from oxygen transmission rate (OTR) measurements recorded using an Oxygen Permeation Analyzer M8001 (Systech Illinois, UK). Experiments were carried out at 23°C and 80% RH. The samples were previously purged with nitrogen in the humidity equilibrated samples, before exposure to an oxygen flow of 10 mL min -1 . The exposure area during the test was 5 cm 2 for each sample. In order to obtain the oxygen permeability, film thickness and gas partial pressure were considered in each case. The measurements were done in triplicate.

Antibacterial activity of active films
The antibacterial activity of the materials was evaluated against two food-borne pathogens, Salmonella enterica CECT 4300 and Listeria monocytogenes CECT 7467. Bacterial strains were obtained from the Spanish Type Culture Collection (CECT: Valencia, Spain) and stored in phosphate buffered saline (PBS, Sigma Aldrich) with 10 wt.% tryptic soy broth (TSB, Conda Laboratories) and 10 wt. % glycerol at -80˚C until needed. The stock culture was maintained by regular subculture to tryptone soy agar (TSA) slants at 4°C and transferred monthly.
Previous to each antimicrobial assay, a loopful of bacteria was transferred to 10 mL of TSB and incubated at 37˚C overnight and an aliquot was again transferred to TSB and grown at 37˚C and 120 rpm to the mid-exponential phase of growth and this cultured was used as inoculum.
The antibacterial activity of the active films was evaluated according to the

Statistical analysis
The statistical analysis was carried out by means of StatGraphics Plus version 5.1 (Statistical Graphics Corp.) through the analysis of variance (ANOVA).
Homogeneous sample groups were obtained by using Tukey's Honestly Significant Difference (HSD) (95% significant level).

RESULTS AND DISCUSSION
In the present work, commercial CuO nanoparticles were incorporated into PHBV by direct melt-mixing at two different concentrations (0.05 and 0.1% CuO) and, the lower loading (0.05%) was also used for the formation of an annealed coating of electrospun PHBV18/CuO fibers mats put over compression molded PHBV3 films for comparison purposes. The physicochemical properties of both systems were determined and are described below.

Morphology and optical properties of CuO nanoparticles and active films
Firstly, the morphology and the particle size of the commercial CuO nanoparticles were characterized. From Figure 1, the morphology of the particles appears to be rather flaky (Figure 1a) and the particle size distribution shows a bimodal distribution with a mean weight size D 4,3 of 191nm (Figure 1b).  Figure 2. PHBVs films showed a compacted structure (cf. Fig 2a) which was not significantly altered by the addition of 0.05% CuO by melt-mixing (cf. Fig 2b). However, the incorporation of CuO at 0.1% promoted some changes in the morphology of the PHBVs matrix since a rough surface was observed in the cross-section of films (cf. Fig 2c) and a marked agglomeration pattern was evidenced in the TEM images of the ultrathin sections (cf. Fig 2d). Moreover, it is noteworthy to remark that all samples showed  The fibers mats used as coating in the 0.05% ES films were also observed by SEM, Figure 3 shows representative images of electrospun PHBV18/CuO fibers before the annealing step (cf. Fig 3a) exhibiting an average diameter of 1.01 ± 0.2 µm. After the thermal annealing, the electrospun fibers formed a continuous layer (cf. Fig3b) which was strongly adhered to the surface of PHBV3 (cf. Fig 3c).   Therefore, the surface morphology and thus, the strategy followed to develop the films (nanocomposites or coating) play an important role on their optical properties.
The lower Ti values observed in nanocomposite films, implies a greater light dispersion, greater opacity and thus, more heterogeneous matrices probably due to the presence of PHBV18 which seems to be thermally degraded to some extent during the melt-mixing process, making the film darker with a brown hue as it can also be deduced from the contact transparency images (cf. Fig. 4). Color differences were also quantitatively assessed by means of lightness (L*), hue (hab*) and Chroma (Cab)*, obtained from the reflectance spectra of an infinity thickness film and the results are gathered in  27 , this color differences can be ascribed to PHBV18 impurities from the fermentation process (close to 30 wt.%, Castro-Mayorga et al. 28 , which induce Maillard reactions between amide groups and residual reducing sugars during the melt-mixing process. Interestingly, while the addition of ZnO mitigated the effect of the unpurified PHBV18, improving the appearance of the melt-mixed nanocomposite films 29 , this seems not to occur for CuO nanoparticles as it can be observed in Figure 4.

Thermal properties
DSC assays were carried out to evaluate how the CuO nanoparticles affected the thermal properties of the PHBVs matrix. Table 3 gathers thermal parameters (melting/crystallization temperature and melting enthalpy) of the developed samples which were determined from the first heating run, offering information related to the thermal characteristics of the so obtained materials. Figure 6 shows representative first heating thermograms for each sample. The first clear observation is that the method of CuO incorporation, either by melt-mixing or as an electrospun coating, affected the thermal behavior of the PHBVs films. Thus, when CuO nanoparticles were directly added in the melt mixing process, the DSC curves of the first heating scan showed two distinguishable melting peaks, which occurred between 165°C and 172°C. Multiple melting peaks occurring during the DSC first heating scan of PHBV and PHBV containing metal nanoparticles have been previously interpreted as a result of a melting-recrystallization process during the thermal run [29][30][31] . Nevertheless, compared to the neat PHBVs, the active films prepared by melt compounding, i.e. 0.1% and 0.05% films, did not present any significant difference in their melting enthalpy (ΔH m ) or crystallization temperature (T c ) suggesting that the nanoparticles did not interfere with crystallization.
In contrast, the addition of CuO within the coated structures led to a significant decrease in the melting enthalpy with regard to the neat PHBVs and their nanocomposites. This agrees with the results reported earlier by Castro-Mayorga et. al. 32 which ascribed the thermal behavior variations to the more homogeneous melting of the fibers due to the high surface to volume ratio of the electrospun materials as compared to a thicker continuous film. These results do not necessarily translate into crystallinity due to the differential behavior during melting of the materials and the crystallinity development observed during the DSC run.  In order to have a more reliable source of crystallinity information, the effect of the CuO addition and the processing type on the PHBVs crystallinity was also analyzed by WAXD (Fig. 7). The X-Ray patterns of all samples showed the typical peaks at 2Ɵ angles of 13.6°, 17.1° and 22.6°, corresponding to the (0 2 0) , ( 1 1 0) and (1 0 1) lattice planes of the orthorhombic unit cell of PHBV 33,34 . A more intense peak at 2Ɵ = 26° was found and associated to the (0 2 2) reflection of boron nitride, which has been recently reported by Sanchez-Safont et al. 35 as a nucleating agent used in the commercial PHBV3 grade used as matrix. No significant differences in the diffraction peak positions were observed between the neat and the active film. This suggests that the crystal morphology of the PHBVs did not noticeably change with the CuO addition neither with the processing type.
However, the lower intensity of the diffraction peaks of 0.05% coated films pointed out a lower crystallinity degree for the electrospun coating layer (also supported by the DSC results shown in the Table 3).

Mechanical properties
Elastic modulus (E), elongation at break (EAB) and tensile strength (TS) were measured for the different materials and the results are shown in Table 4

Barrier properties
Barrier properties often drive decisions on the selection of materials for food applications and the specific barrier requirements depend on the food product characteristics and the intended end-use application. For instance, materials used for carbonated beverage containers should have a high oxygen barrier in order to prevent oxidation of the beverage contents, whereas in packages for fresh fruits and vegetables, high barriers to oxygen are undesirable 37 . The WVP and OP of the nanocomposites and coated systems are summarized in Table 5. As observed, the WVP of the obtained films increased with the addition of CuO, more especially when the fibers mats were put as an antimicrobial active coating onto PHBV3 matrix. This behavior might be related with the more hydrophilic character of the electrospun fibers mats prepared with PHBV18/CuO as compared to the neat PHBV3 film used as substrate in the coated system. In fact, Castro-Mayorga et. al 27 determined that the water uptake values of the neat PHBV18 (32.1 ± 1.2%) was significantly higher than that for the neat PHBV3 (9.8 ± 0.6%). However, these results differ from thoses found in the literature in which the addition of CuO nanoparticles increased water barrier properties 23 2017  in the CuO loading did not improve the oxygen barrier properties of the PHBVs mixture. The oxygen permeability of the coated system (0.05%ES) was higher than its counterpart prepared by melt-compounding (0.05%), which might be associated with its lower crystallinity degree.

Antimicrobial properties
In the present estudy, the antibacterial activity of the nanocomposites films and the coating structures containing CuO nanoparticles was explored against two foodborne pathogens: the Gram-negative S. enterica and the Gram-positive L.monocytogenes, and the results are gathered in Table 6

Biodegradability
Nanocomposite materials may bring some issues when recycled or incinerated due to the uncertain fate of the entraped nanoparticles. Nanobiocomposites on the other hand, offer an advantage in this regard since they undergo biodesintegration in soil during composting conditions. Thus, PHBV is known to readily biodegrade during composting, being this one of the main benefits of these materials in applications . However, in this work an antimicrobial agent has been incorporated in the PHBV which may potentially interfere in its biodegradation; thus, the disintegrability of the most effective antimicrobial materials and their counterpart (without CuO) were assessed by measuring the weight loss over composting time according to the ISO 20200 standard. Figure 8 shows the evolution of the disintegration (%) over time for the PHBVs films, 0.05% ES films and bilayer PHBVs systems consisting of a PHBV3 matrix coated with electrospun PHBV18 fibers mat (PHBVs-ES) prepared for comparatives purposes.

CONCLUSIONS
Antimicrobial PHBV nanocomposites and coated structures containing CuO nanoparticles were developed and characterized. It was confirmed that the antimicrobial activity increase when the CuO was loaded into electrospun fibers mats and deposited as a coating onto PHBV3 films. For instance, the antiviral activity against of the nanocomposites vs. the coated structures, both containing 0.05% of CuO improved by more than 225%; the bacterial reduction against S. enterica by at least 23%; and no viable cells of L. monocytogenes were detected in either of them. It was furthermore demonstrated that the CuO addition by means of electrospinning enhanced the nanoparticles dispersion and did not modify significantly the oxygen permeability, mechanical or optical properties. Instead, the water vapour permeability increased due to the more hydrophilic character of the PHBV18 coating and the poor dispersion of the higher nanoparticles loading.
Interestingly, biodisintegration tests showed that the coated structures were fully biodegraded in a period of 35 days at composting condition.