Effect of Lauric Acid on the Thermal and Mechanical Properties of Polyhydroxybutyrate (PHB)/Starch Composite Biofilms

Department of Engineering Science, College of Engineering and Agro-Industrial Technology, University of the Philippines Los Baños, College, Los Baños, Laguna 4031, Philippines Department of Mining, Metallurgical and Materials Engineering, College of Engineering, University of the Philippines Diliman, Diliman, Quezon City 1101, Philippines Materials Science and Engineering Program, College of Science, University of the Philippines Diliman, Diliman, Quezon City 1101, Philippines Institute of Environmental Engineering, National Chiao Tung University, 1001 University Rd, Hsinchu 30010, Taiwan Institute of Analytical and Environmental Sciences, National Tsing Hua University, 30013, Taiwan University of the Philippines Rural High School, College of Arts and Science, University of the Philippines Los Baños, Paciano Rizal, 4033 Bay, Laguna, Philippines


Introduction
The management of plastic wastes is a prevailing concern worldwide. Commercially used synthetic plastics degrade very slowly when disposed and have a relatively unknown decomposition rate [1]. Recent research efforts have focused on biobased polymers and composites as a sustainable alter-native to synthetic/petroleum-based polymers, especially single-use plastics [2]. The biodegradability and biocompatibility of these polymers are essential for advanced applications that are critical to the biopolymer's purity. These include medical applications such as sustained drug release carriers, scaffolding for tissue engineering, and durable and biocompatible medical aids [3,4]. Industrial applications of biobased polymers are also sought after in textile manufacturing, construction fillers, composites, foams, etc. [5,6]. Likewise, these polymers found significance in agriculture in the form of horticultural crop components, soilretention sheeting and containers, and agricultural films, among others [7].
Biobased polymers can be produced from a wide variety of natural materials, such as plant (cellulose, starch, lignin, etc.) and animal matter (chitin, chitosan, collagen, etc.). Selected microorganisms can synthesize biopolymers through the metabolism of these source components. Typical examples of these polymers are polylactic acid (PLA), polyhydroxyalkanoates (PHA), and their copolymers [8,9]. Polyhydroxybutyrate (PHB) gained the highest interest among PHA types [10][11][12][13][14][15][16][17]. PHB is produced through carbon assimilation by certain microorganisms that are under physiological stress [18]. It is a biodegradable polyester, characterized as highly crystalline aliphatic thermoplastic [19] that exhibits physical properties comparable to that of petroleum-based synthetics, such as polyethylene terephthalate (PET) and polypropylene (PP) [20]. However, PHB is still in the developmental stage due to its relatively high production cost [21], poor mechanical properties due to brittleness [22], and poor formability during processing [23]. Addressing these problems can be carried chemically by blending with compounds that can improve PHB properties, thus forming a composite. A common route is by incorporating cheap biomass fillers that are locally available in order to significantly reduce the overall production cost. However, previous studies showed that the addition of biomass fillers could negatively affect the mechanical properties of the resulting composite [24][25][26]. Interestingly, the addition of plasticizers along with biomass fillers can offset property change by increasing the flexibility and processability of bioplastic composites through lowering the glass transition temperature (T g ) [27].
In this study, PHB is incorporated with starch (S) as a biomass filler and lauric acid (LA) as a plasticizer to form a PHB/S/LA composite. The resulting blend is investigated for its thermal, physicochemical, and mechanical properties. Zhang and Thomas [28] reported that starch acts both as a filler and as a nucleating agent in PHB/starch blends, significantly reducing the size of PHB spherulites. However, Innocentini-Mei et al. [29] still observed poor mechanical properties after incorporating PHB with starch due to the immiscibility between hydrophilic starch and hydrophobic PHB. Hence, LA is used as a potential plasticizer to address this problem. LA, or dodecanoic acid, is a saturated fatty acid ester with hydroxyl moieties from its carboxylic acid groups that can form bonds with other polymers. In this bioplastic composite, LA promotes hydrogen bond formation to reduce the intramolecular forces between PHB and starch molecules, effectively decreasing the glass temperature of the material [27]. Strong interactions caused by rigid bonds within the polymer groups can be disrupted and can increase the miscibility of the bioplastic composite. As an effect, the polymer molecules will have increased mobility and improved interfacial adhesion [30]. The results would be helpful in the development involving PHA-based bioplastics, especially since the use of LA as a plasticizer in PHB/S has not been explored by existing literature.

Preparation of Bioplastic Composite Films.
Composite films made from PHB/starch (S)/lauric acid (LA) were prepared using a heat-assisted solvent casting method derived from Barud et al.'s study [24]. Chemicals have been predried in a forced air circulation oven (Biobase, China) at a temperature of 60°C at least overnight to remove as much moisture as possible. In an airtight 50 mL Teflon tube, 600 mg of the polymer blend was mixed with 15 mL chloroform to form a 4% (w/v) suspension. PHB/S/LA films were prepared using the amounts shown in Table 1.
The PHB-starch ratio has been maintained at 80 : 20 (w/w). This ratio was found to be the optimum ratio to avoid film breakage (data not shown). The suspension was subjected to a high-speed vortex mixer (Biobase, China) at a maximum speed for 3 minutes. The suspension was then subjected in a water bath maintained at a temperature of 60 ± 5°C for 30 minutes. The water bath was dispersed evenly using moderate speed magnetic stirring. Depressurization of the Teflon tubes was done every 10 minutes to prevent excessive pressure buildup. The solution was cast in a 100 mm × 15 mm borosilicate petri dish inside a well-ventilated fume hood. The composite films were obtained after a 4-6 h evaporation of chloroform.

Characterization of the Bioplastic Composites
2.3.1. Fourier Transform Infrared Spectroscopy (FTIR). Fourier Transform Infrared Spectroscopy (FTIR) was done to determine the different functional groups present in the bioplastic composites. The analysis was carried out using a Shimadzu IR Prestige-21 spectrometer (Tokyo, Japan) with attenuated total reflectance spectroscopy. The final spectrum was processed using ATR correction. The FTIR spectra were acquired in the range of 600-4000 cm -1 and a resolution at 4 cm -1 . 2.3.5. Differential Scanning Calorimetry (DSC). DSC experiments were carried out in Perkin Elmer DSC 800 (Massachusetts, USA). The heating and cooling rate for the runs was 10°C/min in nitrogen (N 2 ) atmosphere (50 mL min -1 ). Around 5-6 mg of the sample was put into sealed aluminum pans. Calibration was done using an indium (In) sample. The experiment consisted of a heating stage from 0°C to 350°C. The glass transition temperature (T g ) was measured as the onset of the baseline change in the heating run. The melting temperature (T m ) was obtained from the heating stage. The degree of crystallinity (x c ) was determined using the following equation [14]: where ΔH m is the enthalpy of fusion, ΔH°m is assumed to be the enthalpy of fusion of purely crystalline PHB (146 J/g) [24], and w PHB is the weight fraction of PHB in the blend. The crosshead speed used is 5 mm/min with the extensometer disabled. 10 sets of stress-strain curves were generated per bioplastic composite to account for repeatability of results. The maximum stress (σ max ), elongation at break (ε), and Young's modulus (E) were all recorded.  Figure 1. While pure PHB is transparent (Figure 1(a)), the biocomposites appear to be slightly translucent and milky white. LA0 was observed to have visible white spots mainly attributed to the starch particles on the PHB surface ( Figure 1(b)). The incorporation of

International Journal of Polymer Science
LA decreased the appearance of white spots and increased the transparency of the composites. LA2 (Figure 1(d)) is found to be more transparent compared to LA1 (Figure 1(c)) which implies that a small amount of LA present in the bioplastic can affect its relative opacity. The texture of the films has also improved, with LA1 and LA2 being smoother upon physical contact compared to LA0. and 2(k)). It was observed that starch particles ranging from 5.0 to 15.0 μm in relative size were scattered across the bioplastic unevenly, forming a matrix along the film's surface. These results were consistent with the report of Thiré et al. [32] where starch granules found are polygonal in shape, with a broad numerical distribution of particles from 2.0 to 27.0 μm in equivalent spherical diameter. At low magnifications (200.0 μm), the inhomogeneity seems to have decreased for the LA1 and LA2, suggesting an increased blending in the presence of LA.   International Journal of Polymer Science To properly check for the compatibility of the composites, a cross-sectional SEM was performed for all bioplastic samples. Several cracks and voids can be observed for all samples (Figures 2(c), 2(f), 2(i), and 2(l)). It was observed that pure PHB reference has no phase separation, as shown in Figure 2(c). Starch particles are observed to be dispersed in the continuous PHB matrix in a typical sea-island structure for LA0, as shown in Figure 2(f). The starch granules are not evenly dispersed and exist in the form of agglomerates that are grouped together in clearly demarcated domains [32]. It is worth noting that even though the starch agglomerated to the PHB matrix, it has still adhered strongly. This suggests that the larger starch particles have become a nucleating agent for the recrystallization/precipitation of PHB bioplastic to form the composite [28]. This is also observed in the cross-sectional SEMs of LA1 and LA2 shown in Figures 2(i) and 2(l). It appears that the LA-incorporated bioplastic shows homogenous interfacial boundaries consistent with plasticized polymers, which suggest good polymerfatty acid miscibility.
For a more quantitative comparison, the average surface roughness (R a ) was calculated from the SEM images through ImageJ software [32]. The R a values are presented in Table 2. It was observed that LA0 had increased R a upon the addition of starch, which is attributed to the visible starch granules on the surface of LA0 (Figure 2(b)), significantly contributing to the surface roughness of the bioplastic. This roughness decreased with the addition of LA, with the R a of LA1 and LA2 decreasing by 16.6% and 19.4% compared to LA0. This implies that surface-level blending occurred for the LA1 and LA2, as evidenced by the micrographs shown in Figures 2(i) and 2(l).

Effect of Lauric Acid Addition on the Physicochemical
Properties of the Bioplastic Composite 3.2.1. X-Ray Diffraction. Biopolymers, much like polymers, can also exist in crystalline or amorphous states, and they can have both amorphous and crystalline regions [1,32,33]. X-ray diffraction (XRD) is a surface characterization technique to check the disorder or misalignment of crystal formations or to see if the material is amorphous. The diffraction peaks for the precursors PHB and starch are shown in Figure 3 (222), respectively, for PHB [24,33]. Typical A-type crystallinity pattern can be found for starch, with peaks situated at 2θ = 14:9°and 22.7°and a doublet with reflections at 16.9°and 17.8° [32,34]. Strong peaks centered at 2θ = 20:1°, 21.3°, and 23.7°shown in Figure 3(b) were identified corresponding to the aggregation of lauric acid crystals [34].
The PHB/S/LA composites all depict the same peaks of orthorhombic PHB, as shown in Figure 3(c). It was also observed that the peak at 13.4°increases in intensity as the amount of LA increases, which implies that LA might have a direct effect on the crystallinity of the polymer. Despite the strong intensity of LA, these peaks have not reflected for the diffractograms of LA1 and LA2. This can be attributed to the low concentration of LA on the composite. Moreover, the peaks related to A-type crystallinity of starch particles were not observed in LA0, LA1, and LA2. Traces of B-type or V h -type crystalline patterns have also not been clearly established. These suggest that the presence of starch does not directly affect the PHB crystalline lattice. It may be possible that cocrystallization between the molecules of the blend components has occurred due to starch acting as a nucleating agent. However, due to the similarity of the main peaks of PHB and starch [1,34], establishing a relationship on the incorporation of starch to PHB will be difficult.

Energy-Dispersive X-Ray Spectroscopy (EDS)
. EDS was done to check for the relative purity of the bioplastics. The results are summarized in Table 3. It was shown that there are no nitrogen compounds in the composites, suggesting purity. The incorporation of starch decreased the carbon content slightly, as starch has more carboxyl groups than PHB. Residual chloroform in the form of elemental chlorine has been detected in the samples, denoting that chloroform was not completely removed after blending. This could be a possible concern when used in applications concerning biocompatibility. Other blending methods (melt mixing, melt extrusion, etc.) may be implored to avoid dealing with this issue in the future.

Fourier Transform Infrared Spectroscopy (FTIR).
FTIR spectroscopy was performed to check for the interactions between the composite components. The FTIR spectra of the bioplastic blends are shown in Figure 4. Figure 4(a) shows the FTIR spectra of PHB, starch, and LA0. It was observed that the absence of broad peaks around 3200 to 3600 cm -1 indicates that hydrogen bonds have formed between the hydroxyl groups of PHB and starch to form LA0 [35]. Reis et al. [1] reported in their FTIR results showed that the PHB/starch composite still possesses broad peaks around 3200-3600 cm -1 , in which the intensity is highly dependent on the starch concentration. This absence suggests that PHB interacted with starch as a nucleating agent during the precipitation process. Moreover, the peak around 1645 cm -1 present in pure PHB and starch corresponds to the intermolecular bonding of the carboxyl group [35]. Also, the absence of this peak in all PHB/S/LA films (Figure 4(b)) indicates that esterification took place between the components.
The addition of lauric acid in blending the PHB/S/LA polymer enhanced the esterification of the carboxyl groups, as shown in Figure 4(b) where the flattening of the broad peaks can be observed. For the composites, the presence of a sharp peak around 1720 cm -1 represents the ester carbonyl group formed after esterification [36]. Additionally, several bands between 980 and 1300 cm -1 that are present in all composite films increase in intensity as the added LA increases, suggesting C-C stretching (980 -1000 cm -1 ) and C-O-C (1228 cm -1 ) ester stretching. This phenomenon indicates that the ordered structure of crystalline PHB was destroyed caused by destructured starch incorporation on the bioplastic. Salleh et al. [37] found that lauric acid is strongly bound to amylose, forming an amylose-lipid inclusion complex during starch gelatinization. This complex suggests that starch can have increased esterification with PHB molecules, denoting the increase of C-O bonds.

Effect of Lauric Acid Addition on the Thermal
Properties of the Bioplastic Composite 3.3.1. Thermogravimetric Analysis (TGA). TGA was performed to study the thermal stability of the composites. The improvement of the thermal processability of plastics desires that the range between its melting point (T m ) and the onset degradation temperature is sufficiently large. The TGA and DTG curves of the synthesized bioplastic composites are shown in Figure 5.
TGA key parameters such as the initial decomposition temperature (IDT), the maximum rate of decomposition temperature (MRDT), and the maximum degradation temperature (MDT) [38] are summarized in Table 4. The onset degradation temperature (or IDT) is the point wherein the material starts to lose a significant portion of its mass through thermal degradation. The MRDT shows the highest rate of degradation or the highest peak of the DTG curve. The

MDT occurs after most of the thermodegradative material is lost. This can be found in the TGA and DTG curves shown in Figures 5(a) and 5(e).
Initial degradation of the films in the range of 50-125°C can be attributed to the evaporation of interstitial water attached to the surface of the bioplastics, specifically on the hydrophilic starch molecules [18]. The small degradation peak around 150°C (Figures 5(a) and 5(b)) corresponds to the volatilization of LA, as only films with LA (LA1 and LA2) exhibited a mass loss accounting for approximately 3% mass loss. Volatilization of pure lauric acid occurs around 150°C-250°C [39]. All of the composites have undergone thermal degradation in the range of 260-270°C and only having less than 10% of their original weight by >305°C. This reflects the observations of the IDT and MDT as shown in Table 4. Studies have found that PHB is degraded through means of nonradical random chain scission (cis-elimination), which involves an aromatic C 6 -derived intermediate/transition state [32]. It was also found that there is no clear trend between the TGA parameters. The mass of residue obtained (Figures 5(a) and 5(d)) after the thermal degradation varied directly with the amount of LA added. The decomposition of the nonvolatile residues of LA is known to occur around 600°C [39]. While the composites retain a small amount of mass, pure PHB has undergone full thermal degradation. The negative value for pure PHB at temperatures > 400°C could be caused by residual oxygen trapped in the sample having thermooxidative reactions affecting the inert pan weight.
The temperature at maximum degradation also shifted to higher temperatures with the addition of LA as shown in the DTG curve in Figure 5(e). This reflects the recorded MRDT shown in Table 4. Due to the complete thermal degradation of PHB, its DTG peak is higher than the synthesized composites. The addition of lauric acid indeed improved the thermal stability of PHB/starch films. Also, it was observed that there is a small shoulder in the latter part of the degradation peak at a range of 300-330°C ( Figure 5(e)). This can be attributed to the different decomposition rates of the amylose and amylopectin components of starch. Amylose degrades at a lower temperature compared to amylopectin, which has a more branched structure [33]. It was also observed that thermal 10    International Journal of Polymer Science condensation and dehydration mechanisms begin in the 300°C range, generating ether and ethylene segments. Aromatic and cross-linked structures can also be found at higher temperatures. These reactions lead to the carbon-rich residue, depending on the starch source [32]. Overall, the improvement in thermal stability with the addition of plasticizer to PHB composites is consistent with previous works [13,40].

Differential Scanning Calorimetry (DSC).
To check for the thermal profile of the bioplastic composites (T g , T m , and crystallinity), differential scanning calorimetry (DSC) was done as shown in Figure 6. The miscibility of the bioplastic can be determined through DSC analysis. Blends are considered miscible if only one glass transition temperature (T g ) can be detected. As shown in Figure 6(a), there are only four endotherms present with clear distinctions supported by the literature [1,24,32]. The first endotherm was identified to be the sole T g for all composites that were analyzed. Defined endotherms confirm that the PHB/S/LA blends exhibited single, composition-dependent T g values, indicating that a single homogenous amorphous phase was    [25,39].
The reported values of T g are summarized in Table 5, along with other thermal properties. It was shown that the T g was found in the first endotherm (0-10°C). This parameter slightly increased upon the incorporation of starch. LA addition to the biopolymer decreased the T g , with a downward trend [27]. Theoretically, this suggests that lauric acid can still be added to attain a smaller T g . There are three other endotherms for PHB/starch/LA films, as shown in Figures 6(a), 6(c), and 6(d). Figure 6(c) shows endotherm 2 for PHB/S/LA blends. Blending of starch to PHB shifted the amylose gelatinization peak to approximately 80°C for all composites. For comparison, in the thermogram of pure starch (Figure 6(a)), the recognizable peak at 86.59°C denotes amylose gelatinization [18]. The second irregular peak found at 307°C for starch is attributed to the phase transition of the amylose-lipid complex [41] Endotherm 3 in PHB/S/LA films shown in Figure 6(d) corresponds to the melting of PHB in the blends [42]. As observed, the melting temperature of the films shifted to lower temperatures with an increasing amount of lauric acid. This clearly helps with the thermal processability of the biopolymer composites, as they are much further from the onset degradation temperature, as discussed in the TGA results (Table 4). Only one melting peak was observed for pure PHB, while two peaks were observed for PHB/S/LA composites. The presence of double endothermic melting peaks in the PHB composites was ascribed to the melt recrystallization mechanism. This behavior is common among polymeric materials and has been explained based on two theories: (1) double lamellar thickness population model [43] and (2) melting and recrystallization model [44]. Hong et al. [45] refer to the smaller peak as a melting temperature of small and imperfect crystallites that formed during the evaporation of the solvent (metastable crystals). The high-temperature endotherm was classified to  International Journal of Polymer Science be the melting of ordered polymer crystals [32]. As a result, even though endotherm 3 consists of a pair of peaks, they are considered one due to this behavior. Lastly, endotherm 4 for pure PHB (296.22°C) and PHB/S/LA films (271-277°C) is due to degradation of the polymer, which occurs at breaking of ester bonds and thus decreasing the molar mass [46]. Lastly, the degree of crystallinity of the samples was determined from Equation (1). ΔH m is the apparent enthalpy of fusion observed for PHB and PHB/S/LA blends in endotherm 3 from the DSC thermograms. The computed degree of crystallinity of pure PHB and PHB blends are summarized in Table 5. A decrease in the degree of crystallinity of samples was then observed with the addition of starch. The presence of a plasticizer in the polymer blend interfered with the intermolecular forces between chains such that the folding/arrangement among polymeric chains is altered. The addition of a plasticizer increased the crystallinity of the polymer, making it similar to the crystallinity of pure PHB. However, no certain trend was established between the degree of crystallinity and the amount of lauric acid added. Also, the crystallinity results directly correlate to the XRD results (Figure 3(c)), wherein LA1 and LA2 have an increased peak compared to LA0, which suggests increased crystallinity.

Effect of Lauric Acid Addition on the Mechanical
Properties of the Bioplastic Composite. Flexibility is an important property for plastics, especially for single-use packaging, where the plastic is subjected to continuous wear and tear that forces the material to be stretched during usage. It is very important for the mechanical properties of the synthesized bioplastics to properly dictate their range of applications.
UTM results for the mechanical properties and their summary are shown in Figure 7.
The stress-strain curves obtained from tensile testing of the bioplastic composites are shown in Figure 7(a). All curves depict a behavior characteristic of a ductile material and not of a polymeric material. Still, it can be observed that there is a reduction in the slope of the initial linear region (corresponding to Young's modulus) with an increasing amount of LA (Figure 7(d)), in comparison with LA0. This observation verifies the plasticizing effect of LA on PHB/starch films, due to the aforementioned formation of hydrogen bonds between the two matrix molecules (PHB and starch) and the plasticizer (LA) [27]. The effect of LA addition on the tensile strength and extension at break of the films is shown in Figures 7(b) and 7(c). As expected, the tensile strength slightly decreased with the incorporation of starch, amounting to at most 20% loss of tensile strength with no clear trend. The presence of heterogeneity through crystallite formation can vitrify amorphous polymer chains. This leads to the indication that a level of interfacial adhesion lacked between PHB and starch, albeit its nucleation-induced blending states otherwise. This implies that the PHB-starch mixture behaves as a thermoplastic, but small amounts of loose starch granules in the composite matrix provide avenues for tensile fracture.
It has also been observed that the tensile strength has decreased with increasing plasticizer amount, which has been reported elsewhere [14,30,37,39]. This is due to the plasticizer's effect on the promotion of intermolecular forces, consequently diminishing the strong intramolecular forces within the PHB and starch polymer chains. Moreover, although Young's modulus has no observable trend   International Journal of Polymer Science (Figure 7(d)), the elongation at break has increased considerably with the addition of LA (Figure 7(c)). From previous studies, it is expected that increasing plasticizer concentration allows for a greater elongation of the film before it fractures [14,27,47]. The elongation accounted for an approximately 200-250% increase from LA0 (6:201 ± 1:64%), although there is only a slight difference when comparing LA1 (15:985 ± 4:236%) and LA2 (17:955 ± 4:76%). Reis et al. [1] have incorporated maize starch to PHB. Their mechanical properties show that a 20% starch incorporation in PHB led to a 67% tensile strength loss and a low elongation at break (1.0%). However, this resulted in a relatively high reported Young's modulus (E = 0:75 GPa). Other studies, as summarized by Yeo et al. [17], have reported tensile strength values ranging from 3 to 31.50 MPa, elongation at break of 0.8-15.5%, and a reported Young's modulus of approximately 0.12-4 GPa. From Figures 7(b)-7(d), the reported tensile strength and Young's modulus are within the average to the high end of current literature results. Interestingly, the elongation at break values was higher than reported from existing studies. Overall, the mechanical properties of the PHB/S/LA composite are competitive with the data reported from other existing studies.

Conclusions
PHB/S/LA bioplastic composite films were prepared, and the morphology, physicochemical, thermal, and mechanical properties of the bioplastic films were investigated. The characteristic peaks in the FTIR spectra showed that the addition of LA as a plasticizer promoted the formation of hydrogen bonds between the PHB, starch, and LA components. This can also be observed in the SEM images where a decrease in the inhomogeneity of the samples can be seen, as well as a decrease in the surface roughness as the amount of LA added was increased. The DSC profiles of the films also indicate an improvement in the miscibility of the polymer blends. A decrease in the glass transition temperature (T g ) suggests an improvement in the flexibility of the samples. Moreover, the TGA and DTG profiles also exhibited increased thermal stability. The temperature at the maximum degradation of the PHB/S/LA films was likewise higher with the addition of LA. The plasticizing effect of LA was further verified when Young's modulus was decreased and the elongation at break (maximum 250% increase) was significantly increased. The addition of LA has therefore exhibited promising results in the thermal and mechanical properties of PHB/starch films. Further studies are still necessary to further improve the utility of the aforementioned blends to make the product suitable for practical packaging applications.

Data Availability
The data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest
The authors declare no competing interest.