Polybutylene Succinate Auxetic Membrane Produced by Solution Electrospinning

Herein


Introduction
Metamaterials represent a class of artificially engineered materials showing exceptional properties and performances surpassing those of conventional materials.Metamaterials have remarkable properties based on their microstructural geometry rather than their constituent material.Furthermore, metamaterials have exceptional design flexibility allowing for the development of specific properties and tunable responses to external stimuli. [1]ne category of metamaterials is associated with auxeticity.These materials show unusual deformation behavior by contracting (expanding) in at least one lateral dimension when subjected to longitudinal compression (tension).This behavior is opposite to that of classical materials and is attributed to negative Poisson's ratios, a physical parameter defined as the negative ratio of transverse to longitudinal (loading direction) strains.Due to this atypical behavior, auxetic materials have superior mechanical properties, including shear resistance, indentation resistance, fracture resistance, and energy absorption, as well as tunable properties such as permeability and conductivity. [2]s a result, they have a broad range of applications in diverse fields such as sports and protective devices, [3] aerospace, [4] civil engineering, [5] intelligent filtering, and [6] biomedical and tissue engineering. [7]n 1987, Lakes produced the first synthetic polymer-based auxetic materials using polyurethane (PUR) with a negative Poisson's ratio (NPR) of À0.7 in tension. [8]ollowing this pioneering work, researchers explored a multitude of architectures and designs for auxetic materials, [2] leading to a variety of techniques for their manufacturing. [3]For instance, a fabrication method of auxetic foams was developed through optimized heating, compression, and stretching stages before cooling. [9,10]Chemo-mechanical methods were also introduced as alternatives to thermo-mechanical methods, which involved a solvent [11] or a softening agent [12] rather than heating.For example, Critchley et al. [13] introduced the production of 3D-printed auxetic foams using a commercial thermoplastic polyurethane (TPU) resin with re-entrant cells to achieve an auxetic effect.The scientific community also showed significant interest in developing auxetic fibers, textiles, and scaffolds as another category of auxetic materials.Alderson et al. [14] successfully produced the first auxetic polypropylene (PP) fibers by modifying the conventional melt spinning process.This method was applied for the production of auxetic polyester and Nylon fibers.Other techniques, such as knitting and weaving, were proposed to fabricate auxetic fabrics. [15]The combination of 3D printing (3DP) and multiphoton lithography (MPL) was done to produce a 3D auxetic scaffold for tissue engineering. [16]In another work, Jin et al. [17] developed a 3DP scaffold with multi-scale patterns using melt-electro writing.The scaffold was fabricated by integrating two different dimensions of fibers.Thick fibers with an auxetic shape were used to control the Poisson's ratio of the scaffold, while thin electrospun fibers were incorporated to fill the unit cell of the auxetic structures, promoting cell growth.The authors used precise process control parameters to produce the scaffolds.Pneumatic force was applied to produce the thick fibers, while electrostatic force was used to generate thin fibers.
Domaschke et al. [18] recently reported on the potential of solution electrospinning to produce auxetic membranes.Their study only used solution electrospinning to fabricate nano-fibrous auxetic membranes having significant negative engineering Poisson's ratios (-400) under tension.The unique behavior of their membranes was attributed to the realignment of some fibers with respect to the stretching direction (uniaxial extension), generating lateral contraction within the network.Consequently, transverse fibers under compression buckled and deflected out of the plane leading to an overall increase in the network volume through the collective response of a high number of these fibers.
Solution electrospinning is a technique using electrical force and a whipping instability thinning mechanism to create fibers of small diameter (nanometer range).The formation of solid fibers depends on the generation of a charged jet, which consists of a viscous polymer solution expelled from a small die as a result of a balance between electrostatic forces and surface tension.Through careful control of the processing parameters (concentration, voltage, flow rate, needle-to-collector distance, and collector speed), electrospinning can be optimized to produce fibers with important properties including unique characteristics such as large specific surface area, high surface-area-to-volume ratio, and a highly porous structure with remarkable interconnectivity.Furthermore, electrospinning is very good to produce very small (nano)fibers from a variety of materials including polymers, ceramics, and metals.Nevertheless, organic polymers are the most commonly used materials since they can be dissolved in appropriate solvents. [19]According to the literature, a wide range of organic polymers, encompassing both synthetic and natural, as well as biocompatible and biodegradable, were effectively used for solution electrospinning. [20]Furthermore, polymer melts can also electrospun. [21]n previous works, the design of auxetic materials was achieved by manipulating their microstructure.However, this approach restricted the choice of suitable materials to those having specific microstructural properties, [2] or those requiring complex and expensive equipment. [16,17]Conversely, electrospinning represents a different mechanism (buckling of transversely oriented fibers) to generate an auxetic behavior allowing for a broader selection of polymers to be used. [18]Furthermore, electrospinning is a simple and cost-effective technique to produce auxetic nanofibers with tunable physico-mechanical properties for diverse applications, especially in the field of biomedical and tissue engineering.This is because electrospun membranes can closely mimic the architecture of a native extracellular matrix (ECM). [22]n this study, polybutylene succinate (PBS) was selected as a biobased and sustainable polymer with desirable properties such as excellent biodegradability, thermoplastic processability, and low toxicity. [23]The significance of this research lies in highlighting the importance of controlling auxeticity in electrospun membranes by considering the direction and orientation of the fibers.The main objective was to investigate the mechanical properties and Poisson's ratio of these membranes, with a specific emphasis on understanding the effect of fiber orientation and direction.To the best of the authors' knowledge, no previous reports have been published showing the relations between the fabrication of auxetic biobased and biodegradable membranes using solution electrospinning, while quantifying the role of fiber orientation and direction on the auxetic behavior.

Membrane Morphology and Fiber Orientation
The effect of the collector's rotation speed on the morphology of the PBS electrospun fibers was conducted using four velocities (4.03, 7.11, 9.96, and 11.01 m s À1 ) to understand the effect of morphology on the mechanical properties, especially the Poisson's ratio of electrospun membranes.The solution concentration and applied voltage were held constant at 12% w/v and 28 Kv, respectively, and the distance between the needle tip and collector was fixed at 12 cm.SEM images of the electrospun membranes at the selected velocities, their FFT output images and converted angular distribution plots are presented in Figure 1.It can be seen that increasing the velocity led to higher fiber alignment.When the rotating speed was low (4.03 m s À1 ), the fibers have a more random orientation (Figure 1a), but the fibers are more oriented (parallel to the collector rotating direction) at the highest speed (11.01 m s À1 ) (Figure 1j).In order to get a deeper understanding of the structural properties of electrospun membranes, FFT analysis was used to evaluate anisotropy and quantitatively determine the degree of fiber alignment (orientation).In each plot, the peak represents the main orientation.So, the height and shape of this peak can be used to quantify the degree of alignment and level of orientation of the fibers.A higher peak intensity indicates a higher degree of alignment along a single axis of orientation, this indicates that the fibers are more uniformly oriented in a particular direction.A lower intensity of the peak suggests a lower degree of alignment, implying that the fibers are more randomly oriented.Furthermore, the width of the peak reflects the level of orientation of the fibers.A narrow peak width indicates that the fibers are more uniformly oriented along a single axis with a high degree of alignment.Conversely, a broad peak width suggests that the fibers are more randomly oriented in different directions, with a lower degree of alignment. [24]The normalized intensity values (FFT alignment value) of a sample prepared at low rotating speed (4.03 m s À1 ) is 0.026 (Figure 1c) and this value increases as the rotation speed increases, reaching a value of 0.069 at a rotation speed of 7.11 m s À1 which is almost constant at higher rotation speeds: 0.094 at 9.96 m s À1 and 0.110 at 11.01 m s À1 .It can be inferred from these results that the FFT alignment values of PBS electrospun membranes showed a positive correlation with the rotation speed (Figure 2a).Several groups, using different synthetic polymers, also observed that fiber orientated at 90°increased as the collector speed increased up to a critical threshold.For example, Sian et al. produced electrospun polyacrylonitrile (PAN) nanofibers mats by varying the linear velocity of the collector from 3.5 to 12.5 m s À1 . [25]hey observed that the fibers orientation at 90°decreased when the linear velocity of the collector exceeded 8.6 m s À1 .This optimum can be explained as described next.
Generally, the optimal conditions for generating aligned fibers involve maintaining a linear speed of rotation for the drum matching the deposition speed of the polymer jet, resulting in a consistent deposition and collection of fibers on the collector's surface. [19]The fiber orientation is mainly influenced by the rotation of the collector.If the rotation is slower than the deposition speed of the polymer jet, the fibers will be randomly oriented on the collector.However, increasing the rotational speed leads to a corresponding improvement in fiber orientation due to the centrifugal force generated near the perimeter of the collector.This force acts to elongate the fibers before they are collected on the surface.But if the rotational speed exceeds a certain threshold, the high-speed motion of the collector creates air turbulence that can negatively affect the fiber deposition, leading to a random collection of fibers due to instabilities.
Additionally, it was observed that the average fiber diameters decreased as the rotation speed of the collector increased (Figure 2b): from 0.70 AE 0.15 μm (4.03 m s À1 ) to 0.43 AE 0.13 μm (11.01 m s À1 ).According to a statistical analysis (see Supporting Information), the effect of collector speed on the fiber diameter is found to be significant.The explanation is that the electric field applies an elongational stress (stretching) on the fibers as they approach the collector.Furthermore, as the fibers adhered to the surface of the spinning drum and experienced high rotational force, they experience even more stretching (elongational strain).
It should be noted that increasing the collector speed to 11.01 m s À1 resulted in the presence of beads in the fibers (marked in red in Figure 1j) due to increased air turbulence (see discussion above).As a stable electric field is needed to produce uniform fibers, turbulences interfere with this condition.Consequently, air turbulence led to the formation of irregular fibers or beads and limits fiber alignment.

DSC Result
To investigate the effect of the collector speed and fiber orientation on the crystalline structure of the electrospun PBS membranes, as well as the effect of the crystalline structure on mechanical properties, DSC measurements were conducted.Table 1 reports the thermal properties of the electrospun membranes, including glass transition temperature (T g ), crystallization temperature (T c ), melting temperature (T m ), melting enthalpy (ΔH m ), crystallization enthalpy (ΔH c ), and degree of crystallinity (X c ).The results indicate that the collector speed had a direct effect on increasing the crystalline structure of electrospun membranes (up to 9.8%).
During electrospinning, some polymer chains crystallized, forming small crystalline structures called lamellae, which are responsible for increased crystallinity.The remaining chains form the amorphous phase.In response to shear and elongation forces, these lamellae are aligned and organized into fibrils.Tie chain molecules passing through adjacent crystalline regions act as a bridge between the lamellae, helping to organize and stabilize fibrils formation, ultimately leading to the formation of small-sized bundles.Crystal growth can be facilitated by these small bundles by providing more nucleation sites.The organized orientation of the lamellae into fibrils can also increase the overall crystallinity of the fibers.Therefore, forming bundles of small-sized fibrils can contribute to improve the crystallinity of electrospun fibers.In a related study, Kongkhlang et al. [26] reported on the molecular orientation and the crystal morphology of electrospun polyoxymethylene (POM) fibers.It was observed that electrospinning can produce POM fibers with highly oriented extended chain crystal (ECC) and relatively high crystallinity, even without additional tension via the rotating disk.However, by applying tensile forces during the disk rotation and increasing the rotating speed, the degree of molecular orientation and crystallinity of the POM fibers were substantially increased (from about 55 to 62%).

Mechanical Properties
Figure 3 presents the tensile properties of the electrospun PBS membranes evaluated in both the parallel (P) and transverse (T ) directions with respect to the collector speed.The results show that samples produced at low speed (4.03 m s À1 ) have mechanical isotropy, i.e., similar tensile strength and Young's modulus in both directions (P and T ) due to a random distribution of fiber orientation in each direction (Figure 1c).Conversely, the samples produced at higher speeds (especially at 9.96 and 11.01 m s À1 ) showed mechanical anisotropy as the tensile strengths and   Young's modulus in the P-direction were substantially higher than for the T-direction (Figure 3c,d).For example, the sample prepared at 11.01 m s À1 showed the highest tensile strength of 3.54 AE 0.16 MPa (P-direction), while this sample showed the lowest value for the T-direction: 1.32 AE 0.15 MPa (only about one-third).Xu et al. also reported that mechanical properties can be tailored via anisotropically and heterogeneously aligned patterns. [27]t was also observed that for samples tested in the P-direction (Figure 3a), increasing the collctor speed from 4.03 to 11.01 m s À1 led to a significant increase in the tensile strength (77%) and Young's modulus (88%) (Figure 3c,d), but the elongation at break decreased from 150 AE 16% to 78 AE 13% (Figure 3e).This behavior is attributed to increased fiber orientation with increasing speed (Figure 1).Oriented fibers are more effective in reinforcing material (improved tensile modulus and strength) when the solliciations are applied in the fiber direction.Moreover, when the fibers are aligned, the total load is more equally distributed, so the fibers resist more easily and deform less, resulting in a stiffer material.Furthermore, the diameter of the electrospun fibers can also have an important effect on the mechanical properties of the resulting material.As the average fiber diameter decreased with increasing speed (Figure 2b), the mechanical properties of the material also increased due to a higher surface area to volume ratio, leading to improved interfacial interactions between the fibers, resulting in higher strength and better load transfer.Additionally, the porosity of the electrospun fibers that can be controlled by adjusting the collector speed can also contribute to their mechanical properties.Electrospun fibers have a lower porosity (more compact and dense) at higher collector speeds due to increase layer density.Lower porosity can result in higher stiffness and strength. [28]Xiang and Frey showed that the mechanical properties of Nylon 6 electrospun mats could be improved by increasing their packing density. [29]Crystallinity, which is also affected by the speed (Table 1), can influence the mechanical properties of electrospun fibers.It is known that higher crystallinity usually results in improved tensile strength and stiffness, but lower ductility. [30,31]This is because the crystalline regions provide more dimensional stability, while amorphous regions imparts better elastic properties. [32]owever, for the membranes tested in the T-direction (Figure 3b), increasing the speed had the opposite effect on mechanical properties.The tensile strength and Young's modulus decreased by 34% and 98%, respectively (Figure 3c,d), while the elongation at break slightly increased from 163% to 179%.The reduced number of fibers oriented in the loading direction can explain the observed reduction in mechanical properties for samples tested in the T-direction.In fact, increasing the collector speed leads to a lower number of fibers oriented in the loading direction as they are mainly oriented in the perpendicular direction with respect to the load.Consequently, the force applied cannot be transferred to the fibers, leading to lower mechanical properties.
It is noteworthy that the sample prepared at the highest speed 11.01 m s À1 showed a lower Young's modulus (10.58 AE 1.52 MPa) than the sample prepared at 9.96 m s À1 (11.58 AE 0.64 MPa).This can be attributed to the formation of beads resulting from the generation of air turbulence during processing.Under stress, the beads act as defects as they experience higher deformation and disturb the surrounding fibers.This observation is in agreement with the results of Abdullah et al. on PLA beads-on-a-string fibers. [33]his study also investigated the mechanical hysteresis of electrospun membranes.To determine the resilience of the electrospun membrane, single tensile load-unload strain cycle was conducted and the effect of fiber orientation on mechanical hysteresis was evaluated.Representative mechanical hysteresis tensile loops for up to 3% strains are shown in Figure 4, while the results, including the damping factor (Ψ), dissipated energy (E d ), and strain energy (E p ), are listed in Table 2. Samples taken in different directions and having different fiber orientations showed different hysteresis loop areas.For samples tested in the P-direction, Figure 4a shows that smaller hysteresis loop areas are generated compared to the same samples measured in the T-direction (Figure 4b).This indicates that less energy is dissipated for the former compared to the latter.However, the samples prepared at low speed (4.03 m s À1 ) did not follow this trend as they mostly have randomly oriented fibers instead of aligned ones.The origin of dissipated energy can mainly be attributed to internal friction caused by polymeric chain slippage during deformation (stretching), [34] as well as fiber break-up at junction points. [35]The frictional force, opposing the fibers motion, converts mechanical work into heat (dissipated energy), leading to significant residual strain when the stress is removed.Consequently, overlapped loading and unloading curves indicate low dissipation energy and high resilience. [34]Based on the results presented in Table 2 and Figure 4, it can be concluded that the orientation of electrospun fibers in the loading direction is an important factor affecting the dissipated energy.For samples tested in the P-direction, increasing the speed of fiber deposition leads to higher fiber orientation resulting in lower dissipated energy and damping factor.In particular, the samples prepared at 11.01 m s À1 showed the lowest dissipated energy (6.5 mJ cm À3 ) and damping factor (1.1%).This result can be attributed to the improved efficiency of load transfer along the fiber direction when the fibers are parallel to the loading direction and the reduced number of network junction points limiting stress dissipation in multiple directions, resulting in lower deformation and energy dissipation.Additionally, oriented fibers are more likely to slide past each in all directions.This reduces the interfiber friction and decreases the energy dissipated during deformation.Similarly, Guan et al. [36] investigated the effect of nanofibers' orientation (angle) on the recovery and mechanical properties of shape memory polyurethane (SMPU) fibers.Their findings indicated that SMPU fibers oriented at 0°(parallel to the tensile direction) showed a much better ability to recover from the applied strain after unloading.On the other hand, samples tested in the T-direction, where most of the fibers are aligned perpendicular to the loading direction, showed higher dissipated energy and damping factor with increasing speed up to 9.96 m s À1 as this condition generated the highest dissipated energy (40.8 mJ cm À3 ) and damping factor (25.2%).This result can be explained by the fact that fibers not oriented in the loading direction were bent during deformation, resulting in energy dissipation through interfiber friction and shear forces generated by sliding between the fibers.Moreover, these fibers generate interfiber gaps during deformation, leading to fiber rupture and further energy dissipation. [34]owever, this trend of increasing dissipated energy and damping factor with increasing speed did not hold for the sample prepared at the highest speed (11.01 m s À1 ).In this case, both parameters were decreased because of the lower number of parallel fibers beyond a certain threshold, which can affect the ability of bending.
The strain energy of samples tested in the P-direction was higher than that of the T-direction (Table 2).Strain energy is defined as the potential energy stored within a material when deformed, which is proportional to the amount of work required to deform the material. [37]Therefore, the P-direction, due to its lower flexibility (more rigid structure), required more work to be deformed.For instance, the highest value of strain energy (591.1 kPa) was observed for the sample prepared at 11.01 m s À1 and tested in the P-direction.

Poisson's Ratio
Out-of-plane Poisson's ratios of the samples were measured for both directions (P and T ).The relation between the Poisson's ratio and engineering strain is presented in Figure 5. From these plots, Table 2 reports on the minimum Poisson's ratio as a function of the collector speed.These results clearly show that all the samples showed an auxetic behavior, which was characterized by a NPR.However, the values in the T-direction are more negative compared to the P-direction.In particular, the sample prepared at 9.96 m s À1 and tested in the T-direction showed the lowest value of À5.73 as this sample produced the largest thickness change in excess of 232% (Figure 5d).The minimum value of Poisson's ratio observed for samples tested in the P-direction was À2.90 for the sample processed at 7.11 m s À1 with a thickness increase of 58% (Figure 5c).The mechanism of auxetic behavior in electrospun membranes can be explained by the fact that during uniaxial tensile loading, parallel fibers resist deformation and become thinner in the loading direction.This thinning leads to a decrease in the lateral dimension resulting in the generation of transverse forces in the perpendicular fibers.Hearle and Sultan [38,39] suggested that a relatively stable balance is maintained between the transverse forces developed in directions perpendicular to the fiber axis during tensile loading.In order to maintain this balance of transverse forces, significant levels of width contraction and thickness increase are necessary.So, the perpendicular fibers must bend (buckle) to increase the networks thickness due to the generation of high transverse forces by parallel fibers (Figure 6).Therefore, buckling of transverse fibers is a key factor contributing to thickness increase and represents the origin of a NPR as observed in these electrospun membranes.
For samples tested in the T-direction, there is a higher number of fibers oriented perpendicular to the applied load compared to the parallel direction.This higher number results in a more NPR, as the network expands through the collective buckling response of these fibers (Figure 6).Another factor that may contribute to the more NPR for the T-direction is the lower elastic modulus and tensile strength of these samples, as well as their higher elongation at break as shown in Figure 3. So, these samples have less resistance to deformation and are more flexible.
Moreover, in terms of samples tested in T-direction, increasing the speed up to 9.96 m s À1 resulted in the generation of samples with a more NPR.However, this trend was not observed in the sample prepared at a higher speed (11.01 m s À1 ).In this case, a decrease was observed in both the thickness change (197%) and NPR (-4.89).Improved auxeticity with increasing speed may be attributed to a higher proportion of fibers oriented perpendicular to the loading direction, as well as the presence of thinner fibers promoting fiber buckling.In the transverse direction, samples produced at 11.01 m s À1 do not follow this trend due to the presence of beads limiting fiber buckling.Additionally, as speed increases, the number of fibers in the perpendicular direction increases, while the number of fibers in the parallel direction decreases.When the number of fibers in the parallel direction falls below a certain threshold, there may not be sufficient force to cause the transverse fibers to buckle.It is important to note that these trends were observed for energy dissipation.The sample prepared at 9.6 m s À1 showed the highest NPR (-5.73) and thickness change (232%) in the T-direction, as well as the highest dissipated energy (40.8 mJ cm À3 ) and damping factor (25.2%).However, a decrease in energy dissipation and damping was observed at 11.01 m s À1 , also leading to lower Poisson's ratio.This suggests that a critical number of parallel fibers is necessary for effective buckling and thickness change, which in turn affects Poisson's ratio.
In contrast, samples tested in the P-direction showed that increasing the speed from 7.11 to 11.01 m s À1 produced samples with higher Poisson's ratio values (from À2.90 to À1.80) and lower auxeticity (Figure 5a).Therefore, an inverse relationship exists between auxeticity and speed when tested in the P-direction.The underlying mechanisms responsible for this observation can be attributed to an increase in the number of fibers oriented in the loading direction with increasing speed, leading to higher mechanical properties (Figure 3) and decreased buckling.For samples tested in the P-direction, the same trend was observed for the dissipating energy and damping factor as both properties decreased with increasing speed (less auxetic) (Table 2).This can be explained by lower thickness change during tension resulting in lower deformation and therefore less energy dissipation through damping.

Conclusion
This study provided evidence that solution electrospinning can be a promising approach for the production of polymer-based materials with exceptional mechanical properties (metamaterials, auxeticity, etc.).In the past, auxetic materials often had complex structures requiring specialized manufacturing processes and making them challenging to cost-effectively produce these materials for a broad range of applications.However, this study showed that solution electrospinning can produce biobased (PBS) auxetic membranes by a careful control of the processing conditions to get an appropriate degree of fiber orientation and alignment.In particular, collector speed was investigated, and the optimized membranes showed a NPR down to À5.73.
As expected, fiber orientation and alignment induced some anisotropy in the membrane.To quantify this effect, mechanical properties were measured in both transverse and parallel directions to the collectors' speed.The samples tested in the transverse direction had more NPR (-5.73) compared to samples tested in the parallel direction (-2.90).But the latter generated higher tensile properties (modulus and strength) due to a higher number of more aligned fibers in the loading direction.
The results obtained open the doors for further development by using other polymers (biobased or not).The concept can also be applied to melt electrospinning.But more investigations should be performed to determine the effect of fiber dimensions and spacing (porosity) on the final properties.Finally, studying the mechanical properties at different angles relative to the membrane fabrication direction (collector rotation) should be of interest to optimize the overall properties for a wide range of applications such as impact/shock mitigation (pads), military/ sporting protective devices (fabrics), intelligent filtering (membranes), as well as biomedical, chemical, civil, and tissue engineering.

Experimental Section
Materials: Poly (butylene succinate) grade FD92PM had a melt flow rate MFR (190 °C and 2.16 kg) of 4 g [10 min] À1 and a melt temperature of 84 °C as supplied by PTT MCC BIOCHEM CO., Ltd. (Thailand).Chloroform was purchased from Fisher (USA), N,N-dimethylformamide (DMF) was provided by Sigma-Aldrich (USA), and cetyltrimethylammonium bromide (CTAB) was obtained from TCI America (USA).
Solution Preparation: First, the PBS pellets underwent a predrying step in an oven at 80 °C for 8 h.Then, the solutions for electrospinning were prepared by dissolving PBS and CTAB in a solvent mixture of chloroform/ DMF in a proportion of 90:10 v/v.In parallel, PBS pellets were dissolved in chloroform by stirring for 2 h at room temperature (around 23 °C).Then, DMF was added under continuous stirring (700 rpm) for an additional 0.5 h.The polymer concentration was adjusted to 12% w/v, and the CTAB concentration was set at 0.06% w/v based on our previous work. [40]lectrospinning: A custom-made electrospinning setup was built for our experiments.It consists of a high voltage power supply (Simco, USA), a programmable syringe pump (Harvard Apparatus, USA), a 10 mL plastic syringe, a stainless blunt-tipped needle with an inner diameter of 0.26 mm (26 G), and a rotating drum collector with a diameter of 4.7 cm, a length of 20.3 cm and an adjustable rotating speed up to 660 m min À1 .The collector can also be moved in the transverse direction (speed of 0.4 m min À1 ) to generate uniform fabric mats.The solution was loaded in the syringe and injected through the needle with a controlled feed rate of 1 mL h À1 .The voltage was set at 28 kV causing the jet stream to form and the fibers to deposit on the rotating collector, which was placed 12 cm from the needle's tip.The electrospinning process was carried out for 4 h to achieve the desired thickness (0.3-0.6 mm) at room temperature (around 23 °C).After electrospinning, the nanofibers mats were removed from the rotating drum and dried in a vacuum oven (Cole-Parmer Scientific Co., USA) at 42 °C for 48 h to remove any residual solvent.More information on the optimized set-up and sample production can be obtained from our previous publication. [40]orphology Characterization (Scanning Electron Microscopy): The morphology of the electrospun PBS nanofibers was examined and characterized by a scanning electron microscope (SEM) Inspect F50 (FEI, USA).The dry electrospun mats were coated with a thin layer of carbon (sputtercoating) and images (different magnifications) were acquired at 10 kV.The nanofibers diameters were quantitatively determined using the diameter imaging software (NIH, USA).A minimum of 100 individual fiber was used to compute the average diameter and standard deviation (SD).
Fiber Orientation: Fiber orientation as a function of electrospinning conditions was determined through image processing using a fast Fourier transform (FFT) approach.First, SEM images were transformed into 8-bit grayscale TIF files.Then, these grayscale images were cropped to 2048 Â 1760 pixels for analysis and processed with the FFT function of ImageJ (image-processing software).SEM images of the electrospun membranes representing the fibers' spatial configuration underwent transformation through the FFT function into a mathematically defined frequency domain.This was done by representing the image as a combination of sine and cosine waves with different frequencies, amplitudes, and phases.The FFT function calculates the Fourier coefficients of these waves by decomposing the original image into a sum of complex exponential functions.This frequency domain represents the rate of fluctuation in pixels intensity for the spatial domain.The output image obtained from the FFT, composed of grayscale pixels arranged in a pattern, reflects the degree of fiber orientation in the original data image.At this stage, the FFT output image underwent a 90°rotation because the frequencies generated via FFT are orthogonal (perpendicular) to those in the original image.
The FFT fiber orientation plots were constructed by implementing a circular projection on the FFT output image through the application of the ImageJ "make circular selection" tool, followed by a radial sum of pixel intensities for every degree ranging from 0°to 360°.Pixel summing was performed using ImageJ with an oval profile plug-in (provided by B. O'Connnell). [41]Subsequently, the total pixel intensity for each radius was plotted as a function of orientation, corresponding to the angle of acquisition.The degree of alignment in the original image is illustrated by the height and overall shape of the peak present on the plot. [24]ifferential Scanning Calorimetry (DSC): The melting and crystallization properties of the electrospun membranes were analyzed using a Discovery DSC 25 (TA instruments, USA) under a nitrogen atmosphere.Samples weighing between 1 and 5 mg were subjected to a heating process from 25 °C to 140 °C at a rate of 10 °C min À1 , held for 5 min, and subsequently cooled down to À90 °C at 10 °C min À1 .The samples were then reheated again to 140 °C at 10 °C min À1 .The thermal properties of the membranes were determined from the second heating scan as the first cooling scan erased their thermal history.The glass transition temperature (T g ) was identified as the midpoint of the heat capacity change.The melting temperature (T m ), crystallization temperature (T c ), crystallization enthalpy (ΔH c ) and melting enthalpy (ΔH m ) were determined based on the maximum of the endothermal peak, the maximum of the exothermal peak, the area under the exothermic peak, and the area under the endothermic peak, respectively.Additionally, the degree of crystallinity (X c ) was computed as where ΔH m0 represents the theoretical melting enthalpy of 100% crystalline PBS (110.3Jg À1 ), [42] while ϕ is the weight fraction of CTAB (0.06%).
Mechanical Characterization: The mechanical properties were evaluated using a dynamic mechanical analyzer (DMA) RSA-3 (TA Instruments, USA) at a tensile rate of 0.02 mm s À1 and room temperature (23 °C).Rectangular strips (50 mm Â 10 mm) were cut (single-edge razor blades) from the electrospun membranes in both parallel and transverse directions to the collector rotation and placed in clamps with a 30 mm distance (Figure 7).The tensile modulus, strength, and elongation at break were determined from the stress-strain plot.Four replicates were performed for each formulation to obtain an average with standard deviation.
Additionally, hysteresis tests were carried out using the same set-up in a strain range of 0.1% to 3% at a frequency of 1 rad s À1 .The strain was varied in an "up-and-down" manner to complete the loop.The hysteresis loop formed on a stress (σ)-strain (ε) plot during a single cycle is commonly used to calculate the dissipated energy (E d ).This was done by calculating the area enclosed by the loop as [43,44] Similarly, the strain energy (E p ) was determined by calculating the area under the loading curve from the beginning to the point where the maximum deformation was reached to give [44,45] The damping factor (Ψ) refers to the ability of the material to dissipate strain energy during each loading cycle.When a material is subjected to cyclic loading, it undergoes repeated cycles of deformation and relaxation.During each cycle, the material stores elastic energy and then dissipates this energy as heat during the subsequent relaxation.The damping factor represents the ratio between the dissipated energy to the strain energy stored in the material as [44,46] Ψ ¼ Poisson's Ratio: In this study, the out-of-plane Poisson's ratio of thin electrospun membranes was investigated by performing tensile tests at deformation rate of 0.02 mm s À1 .The test involves applying a tensile load in one direction and measuring the resulting dimensions' changes in the direction perpendicular to the applied load.A digital microscope (Inskam, USA) with a magnification range of 50-1000x was used to capture the video of displacement in the axial and transverse directions until the sample ruptured (Figure 7b).Several points were selected to monitor displacement.Subsequently, ImageJ was used to perform image analyses after dimensional calibration.The average thickness at different loading levels was computed by measuring the thickness at eight distinct points and averaging the results.Also, a small section (length between 5 and 8 mm) was chosen for the length determination.The out-of-plane Poisson's ratio (ν zx ) was calculated as the negative ratio of the transverse strain (ε z ) to axial strain (ε x ) as [47] while the transverse strain (ε z ) and axial strain (ε x ) were calculated as [47] ε x ¼ À l À l 0 l 0 (6) where t 0 and l 0 represent the initial (undeformed) thickness and length, respectively, while t and l refer to the deformed thickness and length, respectively.

Figure 1 .
Figure 1.Representative SEM images of the PBS electrospun membranes prepared at different collector speeds 4.03, 7.11, 9.96, and 11.01 m s À1 (a,d,g,j, respectively) with their respective FFT output image (b,e,h,k) and normalized intensity plots against the angle of acquisition (c,f,i,l).

Figure 2 .
Figure 2. Morphological analysis of the electrospun membranes: a) Maximum alignment as a function of collector speed and b) average fiber diameter as a function of collector speed.

Figure 3 .
Figure 3.Typical tensile stress-strain curves for electrospun membranes prepared using different collector speeds and tested in both a) parallel (P) and b) transverse (T ) directions with respect to the collector's rotation.The other plots present: c) the tensile strength, d) Young's modulus, and e) elongation at break of the electrospun membranes at different speeds and for both directions (P and T ).

Figure 4 .
Figure 4. Hysteresis loop for electrospun samples produced using different collector speeds and tested in: a) parallel (P) and b) transverse (T ) directions.

Figure 5 .
Figure 5. Poisson's ratio as a function of tensile engineering strain for samples produced at different collector speeds and tested in: a) parallel (P) and b) transverse (T ) directions.Thickness change as a function of time for samples produced at different speeds and tested in: c) parallel (P) and d) transverse (T ) directions.

Figure 6 .
Figure 6.Schematic representation of the auxetic mechanism involving three points represented by ○ , □, and Δ, selected to exhibit changes under tensile force.The points a, b, and c represent the three selected points at rest (without tension), while points a', b', and c' represent the same points under tension.When transverse fibers are oriented at 90°(a), they exhibit a higher amount of bending (a') under tension compared to transverse fibers oriented at an angle less than 90°(b').When the number of transverse fibers increases (c), the bending force also increases under tension (c').

Table 2 .
Mechanical properties of the electrospun PBS membranes.