Charging Properties of Electrospun Poly(l‐lactic acid) Submicrofiber Mat and Its Electrical Applications

Wearable pressure sensors have attracted significant attention owing to their potential applications in health monitoring and connectivity to internet‐based apps. Polymers such as poly(vinylidene fluoride) have been used in sensors. However, being petroleum‐derived materials, they do not decompose and remain in the soil when disposed. Poly(l‐lactic acid) (PLLA) is a promising material because of its biodegradable nature and its derivation from plant‐based materials. In addition, the electrospun PLLA fiber mat contains real charges and exhibits electromechanical properties. However, the detailed charging properties of the PLLA fiber mats remain unclear. Herein, the charge distribution of these fiber mat is presented, and a charging model of the fiber mat and a numerical model of the output charges from the fiber mats with electrodes are proposed. Additionally, the retention properties of the stored charges are determined using surface potential measurements at different temperatures. In addition, a self‐power‐generating touch sensor and mask‐type sensor are developed using biodegradable materials produced from biomass. These studies contribute to the improvement in the charge properties of PLLA fiber mats and the resulting wearable biodegradable sensors.

shellfish ingest and accumulate.Hence, there are concerns about their adverse effects on ecosystems and human health.
Poly(lactic acid) (PLA) is one of the most promising materials for solving the problems such as natural resource depletion and nonbiodegradability because it is biodegradable and produced from plant-derived materials such as corn, cassava, sugarcane, and sugar beet pulp. [21]The lactic acid monomer has two optically active configurations, L-lactic acid and D-lactic acid.These enantiomers lead PLA to poly(L-lactic acid) (PLLA), mainly comprising L-lactic acid units; poly(D-lactic acid) (PDLA), mainly comprising D-lactic acid units; or poly(D,L-lactic acid) (PDLLA), comprising a mixture of L-lactic acid and D-lactic acid units.Highly crystallized and oriented PLLA or PDLA films produced by drawing are known to exhibit piezoelectricity in shear direction.This shear piezoelectricity is attributed to the change in the position of the C═O dipoles arranged circularly around the helical structure of the polymer main chain when stress is applied to the PLLA film in the direction of fiber. [22]25][26][27][28] Sencadas et al. reported that a single-electrospun PLLA microfiber exhibits direct electromechanical property. [23]Zhu et al. reported that an electrospun PLLA fiber mat with an orientation demonstrated direct electromechanical properties along the direction of the fiber orientation when subjected to bending deformation. [24]Zhao et al. reported that electrospun PLLA fibers exhibit direct electromechanical properties in the direction of the thickness of fiber. [25]Lee et al. reported that electrospun PLLA fiber mats with random fiber orientations demonstrated direct electromechanical properties in the direction of the thickness of the film. [26]Iumsrivun et al. reported that several electrospun submicro/microfiber mats made of biodegradable polymers with random fiber orientation demonstrated direct electromechanical properties in the direction of the thickness of film; in particular PLLA submicron fiber mats demonstrated a significantly high apparent piezoelectric constant d, as high as 1533 pC N À1 . [27]n addition, they demonstrated that the direct electromechanical properties originated not from dipolar charges, which originate from the polarization of permanent dipoles in PLLA, but from real charges, including surface charges and space charges in the fibers, stored in the fiber mat.However, the detailed charging properties of the PLLA fiber mat, including the charge distribution in the fiber mat and the retention properties of the stored charges, remain unclear.Moreover, numerical modeling of the output charges from charged electrospun PLLA fiber mats with electrodes has not been reported.
Therefore, in this study, the charging properties of electrospun PLLA submicrofiber mats were investigated in detail, and a numerical model of the output charges from charged electrospun PLLA fiber mats with electrodes is proposed.Additionally, applications of PLLA fiber mats, in touch sensors, and mask-type acoustic sensors comprising mainly biodegradable materials produced from biomass, are demonstrated.This study promotes a deeper understanding of the charging of electrospun PLLA fiber mats and paves the way for the development of wearable pressure sensors that are self-powergenerating, plant-derived, and biodegradable.

Properties of Electrospun PLLA Fiber Mats
Electrospun PLLA fiber mats were produced using the setup shown in Figure 1.The setup included seven multineedles and an aluminum plate that moved circularly using an automatic XY stage.The PLLA fiber mats were directly deposited on an indium tin oxide (ITO) glass substrate placed on an aluminum plate.The details of the production of the fiber mats are explained in Section 4.
Field-emission scanning electron microscopy (FESEM) image of electrospun PLLA single fiber and a scanning electron microscopy (SEM) image of the electrospun PLLA fiber mat are shown in Figure 1b.Randomly oriented submicrofibers with a uniform diameter, that is, without a nonuniform beaded morphology, were obtained.The geometrical properties of the individual fibers and mats are summarized in Table 1.The fiber mat is significantly light in weight and highly sparse with a density of 0.127 AE 0.029 g cm À3 .If as-received PLLA pellets with the density of 1.21 g cm À3 (measured value) are assumed as the PLLA filling ratio of 100%, the PLLA filling ratio of the fiber mat calculated is 10.6%.
The crystallinity of the PLLA fiber mat, X c , was evaluated from its X-ray diffraction (XRD) pattern shown in Figure 1c.[31][32] X c was calculated using the following equation: [33] where A c is the area of the crystalline peaks of the XRD pattern and A a is the area of the amorphous region of the pattern.X c was calculated to be 0.72%, which shows that the electrospun PLLA submicrofibers were predominantly amorphous.This high degree of amorphousness of the electrospun PLLA fibers was also reported in previous studies, [34,35] which results from the rapid evaporation of the solvents during electrospinning and the resulting rapid formation of the fiber structure.
The thermal properties of PLLA fiber mat were investigated with differential scanning calorimetry (DSC) analysis (Figure 1d), where the glass transition temperature (T g ), cold crystallization temperature (T c ), and melting temperature (T m ) were determined to be 67, 76, and 173 °C, respectively.DSC measurement could be done just once because the PLLA fiber mat melted after the measurement.The average relative permittivity of the fiber mats was measured to be 1.22 AE 0.08 using an LCR meter.[38][39] because of its highly sparse structure.

Electrical Properties
The electrospun PLLA submicrofiber mats stored charges without any postcharging treatment including polling; this is evidenced by a surface potential of approximately þ1000 V on the fiber mat and an average total charge of þ3.58 AE 1.20 nC measured with a Coulomb meter as shown in Table 1.
First, we investigated the types of charges stored in the fiber mats, including surface, space, and dipolar charges.Isopropyl alcohol (IPA) was sprayed onto the surface of the fiber mat and allowed to dry.Subsequently, the surface potential of the fiber mat decreased significantly to approximately 0 V.If dipolar charges that originate from unidirectionally aligned permanent dipoles of C=O in the PLLA chains are predominantly stored in the fiber mat, the surface potential should not decrease to 0 V even after spraying IPA because dipolar charges remain after IPA spraying.This result confirms that the electrospun PLLA fiber mats predominantly store real charges.Furthermore, the significantly positive charge amount stored in the fiber mat, þ3.58 AE 1.20 nC, as shown in Table 1, supports that real charges are predominantly stored in the PLLA fiber mat because the charge amount should be nearly 0 nC if dipole charges comprising equivalent positive and negative charges are mainly stored in the fiber mat.
Although the surface potential disappeared after being sprayed with IPA, the helical crystalline structure, which causes shear piezoelectricity in conventional piezoelectric PLLA films, may have been present in the individual PLLA fibers.Generally, the helical crystalline structure does not cause dipolar charges and resulting surface potential.This is because the permanent dipoles of C=O in the PLLA chains along the perpendicular direction of the carbon chain cancels out the macroscopic dipole moments in the helical crystalline structure. [25]In the present electrospun PLLA fiber mat, the helical crystalline structure is hardly present because the crystallinity of the PLLA fiber mat was quite low at 0.72%.In addition, the direct electromechanical properties of the PLLA fiber mat disappeared after the being sprayed with IPA, as discussed in Section 2.4.Moreover, this result supports little evidence of the helical  crystalline structure in the present PLLA fiber mat because the fiber mat should show the electromechanical properties even after spraying IPA if the helical crystalline structures, those remain even after spraying IPA, predominately exist in the fiber mat.Second, to evaluate the vertical distribution of the real charges stored in the PLLA fiber mat, the surface potentials of the top and bottom sides were measured twice after each fiber mat was peeled off carefully from the ITO/glass substrate (Table 2), and placed on an electrically grounded metal plate covered with an insulating tape to prevent the leakage of stored charges of the fiber mats into the metal plate.The surface potential of the insulating tape was confirmed to be 0 V prior to each measurement.
As shown in Table 2, the surface potential of the top side of each of the three fiber mats (fiber mats A-C) was positive, whereas that of bottom side was negative.In addition, the absolute value of the surface potential of the top side was higher than that of the bottom side.This result confirms the model of stored charge distribution in the electrospun PLLA submicrofiber mats as summarized in Figure 2a too: 1) The fiber mats store both real positive and negative charges.2) The real positive charges are distributed mainly in the upper part of the fiber mat, and the real negative charges in the lower part of the fiber mat.3) The amount of stored real positive charges is higher than that of real negative charges (this is also evidenced from the amount of the stored positive charges in the fiber mats measured with a Coulomb meter, þ3.58 AE 1.20 nC (Table 1)).The stored charges may leak when each peeled fiber mat is placed on the insulating tape during the measurement of the surface potentials.However, the magnitude relationship of each top and bottom surface potentials shown in Table 2 does not change in first and second measurements, and hence, leakage is negligible to pursue a discussion on the present qualitative stored charge distribution.A previous study assumed that the amount of stored positive charge is equal to that of the stored negative charge, as shown in Figure 2b. [14]owever, this study demonstrated experimentally that the amount of stored positive charge was higher than that of the stored negative charge.
Iumsrivun et al. recently proposed a model for stored real charge distribution for electrospun PS microfiber mats similar to those shown in Figure 2a by distinguishing three phases (phases I, II, and III) based on electrospinning time. [40]Phase I: Although positively charged electrospun fibers are initially deposited on a grounded electrode, the positive charges leak to the electrode in such a way that an uncharged insulating fiber mat layer is formed at a certain thickness.Phase II: Positively charged electrospun fibers are deposited on the insulating fiber mat layer forming a positively charged fiber mat layer.Simultaneously, negative charges were induced at the bottom electrode owing to electrostatic induction by positive charges in the positively charged fiber mat layer.Phase III: As positively charged electrospun fibers keep depositing, an electrical breakdown occurs between the positive charges in the positively charged fiber mat layer and negatively induced charges in the bottom electrode.As a result, the induced negative charges are injected into the lower part of the fiber mat.In the electrospun PLLA submicrofiber mats developed in this study, the stored charge distribution, as shown in Figure 2a, can occur through phases I, II, and III.

Charge Retention Property
The retention properties of real charges stored in the PLLA fiber mats were investigated by measuring their surface potentials over time (Figure 3).The fiber mats were stored in two different environments: a low humidity environment (nitrogen gas atmosphere, 0.7-7.5%RH)and high humidity environment (air atmosphere, humidity: ≈90%RH).The temperature under Table 2. Surface potentials of the top and bottom side of three PLLA fiber mats (A-C) after peeling them off from the ITO/Glass substrates 3 days after electrospinning.The measurements were repeated twice.
Figure 2. a) The current proposed model of stored charge distribution in the electrospun PLLA submicrofiber mat.b) The previously proposed model of stored charge distribution in the electrospun PS microfiber mat. [14]oth environments was maintained at ≈30 °C.As shown in Figure 3, the surface potentials decreased rapidly after 3 days under both the storage conditions.Subsequently, the surface potential of the fiber mats stored under a low-humidity environment decreased gradually from 3 to 50 days; thereafter, the surface potential remained constant even after 240 days.This demonstrates the excellent charge retention properties of the electrospun PLLA fiber mats in a low-humidity environment.However, the surface potential of the fiber mats stored under the high-humidity environment decreased gradually from the 3rd day to approximately the 50th day; the surface potential then continued to decease to approximately 2% of its initial value in 240 days.This significant decrease in the surface potential under high-humidity conditions could be because of leakage of the stored real charges through water molecules. [41]o understand the charge retention properties of the PLLA mats further, the surface potential of the fiber mats was measured when heated to different temperatures.Figure 4 shows the measured surface potentials normalized to those before heating at different temperatures.In this case, electrospun PLLA fiber mats were stored under a high-humidity environment for 2, 80, and 240 days.In 2 days, the surface potential decreased moderately with increase in temperature from 24 °C (room temperature) to 50 °C; the surface potential then decreased significantly when heated from 50 to 70 °C.This temperature range includes T g (67 °C) of the PLLA fiber mats (Table 1).The surface potential decreased gradually at over 70 °C and reached approximately zero at 110 °C.However, for the fiber mats stored for 80 and 240 days, the temperature at which the surface potential began to decrease shifted to higher temperatures than that of the fiber mats stored for 2 days.Finally, the surface potentials of these fiber mats decreased to approximately zero when heated to 110 °C.A decrease in the surface potential occurred because of the leakage of the stored real charges in the PLLA fiber mat resulting from the energization of the stored charges by heating.Therefore, the decrease in surface potential at higher temperatures results from the leakage of stored charges trapped at the energy levels corresponding to that temperature.As shown in Figure 4, the surface potential decreased with increase in storage time at higher heating temperatures.This result indicates that the stored real charges trapped at shallower energy levels selectively leak with increase in storage time, and the stored real charges trapped at deeper energy levels remain.This leakage of the stored charges trapped at shallower energy levels occurs faster than those at deeper energy levels, due to which the surface potential decreases rapidly in 3 days as shown in Figure 3. Subsequently, most of the charge trapped at the energy level lower than that at the temperature of 60 °C leaked after 80 days.This leakage corresponds to a decrease in the surface potential from 3 to 50 days, as shown in Figure 3.However, even after 80 days, a significant decrease in the surface potential was observed at temperatures between 60 and 70 °C, around T g .This decrease in the surface potential due to leakage of stored charge occurred not only because it was energized by heating, but also because of shallowing of the charge-trapped energy level due to Brownian motion in the amorphous phase, which is mainly composed of the PLLA fiber mat.After 240 days, the decrease in the surface potential around T g is lower.This means that the charge trapped in the energy level around T g leaks over time.

Properties of Output Charge
The output charge of the PLLA fiber mat was measured using a sequential approaching/loading technique [42] using a piezoelectric constant measurement apparatus (PF-02B, Lead Techno Co., Ltd.) equipped with a laser confocal displacement meter (LT-9010M/LT-9500, Keyence Corp.).For this measurement, a metal disk probe with an area, S, of approximately 0.5 cm 2 placed approximately 100 μm away from the surface of the fiber mat on the ITO/glass substrate is perpendicularly and sequentially approached/loaded onto the fiber mat.During approach and loading, the amount of output charge from the metal disk probe, position of the probe, and pressure applied from the probe to the fiber mat were continuously measured.The metal disk probe and ITO acted as upper and bottom electrodes, respectively; the position of the metal disk probe at 0 μm, corresponding to the position of the surface of the fiber mat, is defined as the position where the pressure applied by the probe onto the fiber mat is  approximately 60 Pa.As shown in Figure 5a, when the probe sequentially approached and was loaded onto the fiber mat, charges were outputted onto the probe without an external power source.A similar charge output property was reported for electrospun PS microfiber mats. [42,43]The output charges originate from the compensation charges due to electrostatic induction from the stored charges in the fiber mat.In particular, once the probe comes in contact with the fiber mat, the charge output with the applied pressure is regarded as a direct electromechanical property.Figure 5b shows the amount of output charge from the fiber mat under different pressures.By applying pressure, the electrospun PLLA fiber mat outputs a charge; thus, the fiber mat demonstrates direct electromechanical properties.Using the data in Figure 5b, the apparent piezoelectric constant (d app ) was calculated using the following equation: [42] where and Here, Q(P) is the output charge amount when the pressure P is applied to the fiber mat.P pre is the pressure applied to the fiber mat when the disk probe position is 0, that is, P pre ≈ 60 Pa. Figure 5c shows the value of d app calculated with different pressures applied; it is ≈2.5 nC N À1 when a pressure of 1 kPa is applied.Such a high value can be attributed to very low prepressure of approximately 60 Pa [42] and the soft nature of the electrospun PLLA fiber mats. [17]o confirm that the direct electromechanical properties originate not from the shear piezoelectricity of crystalline PLLA in the fiber mat but from real charges stored in the fiber mat, d app was evaluated before and after spraying with IPA.Spraying with IPA selectively discharges the real charges; however, the helical crystalline PLLA is retained in the fiber mat and it exhibits direct electromechanical properties even after spraying with IPA.As shown in Figure 5c, the d app values decreased drastically to approximately zero after spraying with IPA.This result evidences that the direct electromechanical properties do not originate from the shear piezoelectricity of crystalline PLLA in the fiber mat but from real charges stored in the fiber mat.Additionally, the significantly higher value of d app shown in Figure 5c compared with the piezoelectric constant of highly crystalline PLLA films (d 14 : ≈16 pC N À1 [44] ) and same polarity of the output charges from electrospun PDLA, PDLLA, PLLA fiber mats (see Supporting Information) support the origin of the direct electromechanical properties.Previously, a sheet of oriented electrospun PLLA fibers produced using a disk collector that rotates with a high rotation speed of 1800 rpm was reported to demonstrate crystallinity and voltage/current output with strain deformation. [24]PLLA in this sheet crystalized owing to mechanical drawing by the rotating collector such that shear piezoelectricity of crystalline PLLA may appear.Nevertheless, the randomly oriented electrospun PLLA fiber mat, without any mechanical drawing, in this study demonstrates its direct electromechanical property that originates from real charges that are stored in the fiber mat, although electrospinning conditions may affect the type of charge stored in electrospun PLLA fibers.Note that the electrospun PLLA fiber mat after spray with IPA was tested with the sequential approach/loading technique, and repeated 10 times as shown in Figure 5c; however, no significant charge output was observed, and the charge output due to triboelectrification during the testing is negligible.
Next, we propose a numerical model for the output charge from the upper electrode (i.e., the metal disk probe in the present measurement).In a previous numerical model, [43] in which an electrospun PS microfiber mat was used, the electrospun fiber mat was modeled as a rectangular electret that stored positive and negative real charges on its upper and bottom surfaces, respectively, with the same absolute surface charge density, as shown in Figure 2b.In the present numerical model, the electrospun PLLA fiber mat is modeled as a rectangular electret as shown in Figure 2a, where the absolute surface charge density on the upper surface, i.e., positive real charges (þσ EFFU0 ), is higher than that of the bottom surface, i.e., negative real charges (Àσ EFFL0 ).We assume the fiber mat to be an infinitely wide rectangle of thickness T F0 .
The position of the upper electrode is represented by h (Figure 6), and placed at the position of h = d (d ≥ 0) and Here, ε F0 is the relative permittivity of the fiber mat.Because of electrostatic shielding by both the upper and lower electrodes, there was no electric field above the upper electrode or below the lower electrode.Thus, the summation of σ U (d), σ L (d), σ EFFU0 , and σ EFFL0 is zero basing on Gauss's equation, and can be depicted as follows: From Equation ( 5) and ( 6), σ U (d) and σ L (d) are calculated as follows: When σ EFFU0 and σ EFFL0 are the same value, σ U (d) and σ L (d) are calculated to be the same value as follows: The calculated σ U (d) and σ L (d) in Equation ( 8) is the same as that in the previous numerical model, where σ EFFU0 and σ EFFL0 are of same value. [40]hen the position of the upper electrode changes from d 0 , its initial position, to d (Figure 6a,b), the amount of output charge from the upper electrode, ΔQ U (d), is calculated using Equation (7).
where S is the area of the upper electrode.
When the upper electrode presses the fiber mat, as shown in Figure 6c, the decrease in the its thickness induces an increase in its charge density.Therefore, the effective surface charge density at the upper surface of the fiber mat (σ EFFU ) after pressing the fiber mat can simply be assumed as follows: According to Gauss's equation and Kirchhoff 's second law, σ EFFU and σ U (d) satisfy the following equation: Therefore, when the upper electrode presses the fiber mat, ΔQ U (d) can be calculated from the following equations: From Equation ( 9) and ( 12) we can conclude that the amount of output charge from the upper electrode depends on σ EFFU0 , and not σ EFFL0 or both σ EFFU0 and σ EFFL0 .From these equations we can conclude the following: 1) A positive current (its direction is shown in Figure 6b) is observed when the upper electrode approaches and presses the fiber mat.2) The amount of output charge depends on the area of the upper electrode and the effective surface charge density at the top of the fiber mat.

Application of PLLA in Self-Power-Generative Sensors
A self-power-generative touch sensor was developed using biodegradable materials produced from biomass (Figure 7a,b) with PLLA films as top and bottom substrates, an electrospun PLLA submicrofiber mat as a ferroelectret, PDLLA films as spacers, and a thermoadhesive layer between the top and bottom substrates.The developed sensor outputs a positive voltage when pressed with a finger even with no external power supply and a negative voltage when the finger is released (Figure 7c).The sensor without a fiber mat was also pressed and released with no significant output voltage as shown in Figure 7c.Thus, it was confirmed that the output voltage originated from the fiber mat.When the sensor is pressed with a finger, d is lower (d < d 0 ) and ΔQ U (d) is positive (Equation ( 9)).That is, the sensor outputs a positive voltage.Note that the polarity of the output voltage from each touch sensor comprising the electrospun PDLA fibers or electrospun PDLLA fibers is same as that from the touch sensor comprising the electrospun PLLA fiber mat (see Supporting Information).This result provides evidence that the electrospun PLLA fiber mats predominantly store real charges.
Additionally, a mask-type acoustic sensor mostly made of biodegradable materials was developed (Figure 8a,b).This mask was reconstructed from a commercially available disposable mask consisting of nonwoven PLA fabric.It consists of an outer electrode layer partially coated with conductive poly(3,4-ethlene dioxythiophene)/poly(styrene sulfonate) (PEDOD/PSS) without a fiber mat, a middle sensor layer partially coated with PEDOD/PSS, on which the electrospun PLLA fiber mat was deposited, and an inner insulation layer of pure nonwoven PLA fabric preventing a short between the electrode and the skin of person wearing the mask.
Figure 8c shows the output voltage from the mask-type acoustic sensor when the sensor was worn by a 40 year old male  vocalizing "Kyoto Institute of Technology."The sensor shows an output voltage in response to person's voice with no external power supply connected to it.This result demonstrates that the proposed sensor behaves like a self-powered generative acoustic sensor.The fiber mat and/or electrode vibrates in response to the voice; hence, the charge output is due to the change in the distance between the fiber mat and the electrode.
The mask-type sensor is connected to a tablet via a phone cable/plug.A load of only 1 kΩ was introduced to recognize the cable/plug connection for the tablet.Voice recognition software was launched on the tablet to characterize the person's voice.When the mask-type sensor was worn by a 40 year old male vocalizing at "Kyoto Institute of Technology," the software displayed the same words (Figure 8d).This demonstrates that the mask-type sensor behaves like a microphone.When the person put off the mask-type sensor and spoke, the software did not respond, which proves that the software recognized the signal from the mask and not from the inner microphone on the tablet.Although several mask-type sensors using petroleum-derived materials have been reported, [45][46][47][48] this is the first time that biomass materials have been used as the sensor material.

Conclusion
In this study, the charging properties of the electrospun PLLA fiber mats were investigated.First, the type of charges stored in the fiber mats was identified to be predominantly real charges.Subsequently, the charge distribution in the fiber mat is elucidated, and the charging and numerical models of the output charges from the fiber mats with electrodes are proposed.The surface potential of the fiber mat stored in a low-humidity environment was highly stable after a significant decrease in the surface potential immediately after electrospinning.In contrast, the surface potential of the fiber mat stored under a high-humidity environment decreased gradually after a significant decrease immediately after electrospinning and reached approximately 0 V in 240 days.The measurement of the surface potentials of the fiber mat at different temperatures showed that this decrease in the surface potential was due to the leakage of the stored real charges trapped at shallower energy levels with increase in storage time, whereas the stored real charges trapped at deeper energy levels remained with increase in storage time.In addition, a self-generative touch sensor and a mask-type acoustic sensor were developed mostly made of biodegradable materials from biomass.The touch sensor outputs positive and negative voltages when pressed and released.The mask-type sensor outputs a voltage in response to the voice of the person wearing the mask-type sensor, and the tablet connected to the mask-type sensor displays the word spoken.This study will pave the way for the development of wearable pressure sensors with self-power-generating and biodegradable features that will contribute to solving resource depletion and nonbiodegradability issues.

Experimental Section
Materials and Fabrication of Fiber Mats: PLLA (M w ≈ 80 000-100 000, Polyscience, Inc.) was dissolved in a 3:7(V/V) mixture of N,N-dimethyl formamide (Nacalai tesque, Inc.) and chloroform (Nacalai tesque, Inc.) at a concentration of 10 wt% for 24 h at 95 °C.The solution was then loaded into a syringe equipped with a multineedle apparatus (DN30-7n-23G-20L, Unicontrols Co., Ltd.) consisting of seven 23 G stainless-steel needles placed at 5 mm intervals.The PLLA solution was fed constantly at a rate of 3.0 mL h À1 using a syringe pump (KDS-100, KD Scientific, Inc.).The multineedle apparatus was connected to a 22.0 kV supply (HVU-30P100, Mecc Co., Ltd.).Glass plates (30 mm Â 30 mm Â 0.7 mm thickness) coated with an ITO layer of 150 nm thickness were used as collectors, on which the electrospun PLLA fiber mats were deposited.An aluminum plate (300 mm Â 200 mm Â 1 mm thickness) was placed vertically 130 mm above the tip of the multineedles; each ITO glass substrate was then attached to the bottom surface of the aluminum plate, with the ITO surface directed toward the multineedles (Figure 1a).Both the aluminum plate and ITO layer were electrically grounded.The aluminum plate and the ITO/glass substrate attached to the plate were rotated circularly at a speed of 1 cm s À1 and a radius of 1 cm during electrospinning to deposit the electrospun fiber mat homogeneously (Figure 1a).The electrospinning was performed for 30 min at a temperature ranging from 28.1 to 29.0 °C and 30.2 to 33.5%RH.
Geometrical Characterization: SEM images of the electrospun PLLA fiber mats were obtained using TM4000PlusII apparatus (Hitachi High-Tech Corp.) in the backscattered electron detection mode with an acceleration voltage of 10 kV.A 4 nm thick gold layer was coated onto the fiber mat prior to obtaining the SEM images.FESEM images of the singleelectrospun PLLA fibers were obtained using JSE-7001F apparatus (JEOL Ltd.).The fibers were deposited directly on a p-type Si substrate via electrospinning and then coated with a 2 nm-thick gold layer.The average fiber diameter was calculated from 100 diameters measured from the FESEM images.
Charge Output Property: The charge output was measured using a piezoelectric constant measurement apparatus (PF-02B, Lead Techno Co., Ltd.) equipped with a laser confocal displacement meter (LT-9010M/ LT-9500, Keyence Corp.).A disk shape metal probe with a diameter of 8.0 mm was placed at distance of approximately 100 μm from the surface of the fiber mat, approaching toward the fiber mat.When the pressure applied to the fiber mat by the disk probe was 60 Pa, the probe position was defined as d = 0. Subsequently, the disk probe perpendicularly indented until approximately 20 μm.The output charge flowing from the disk probe to the bottom electrode was measured using a charge amplifier.
Electrical Characterization: The surface potential of each PLLA fiber mat was measured using a digital low-voltage static meter (KSD-3000, Kasuga Denki).The probe was placed at a distance of 10 mm from the surface of the fiber mat, and the surface potential was measured over an area of 20 mm Â 20 mm.The bottom ITO electrode was electrically ground for each measurement.To measure the surface potential of the fiber mats after heating to different temperatures, they were heated for an additional 10 min in a thermostatic chamber (AVO-250NB, Asone Corp.) after reaching the target temperature.The average total charge in the fiber mats was evaluated using a Faraday cage (KQ-140, KASUGA DENKI, Inc.) and a Coulomb meter (NK-1001A, KASUGA DENKI, Inc.).The Faraday cage and Coulomb meter were electrically grounded to discharge any residual charge; the inner container of the Faraday cage was connected to the probe of the Coulomb meter, and the outer container was electrically grounded.The fiber mats were carefully peeled from the glass substrate and placed in a Faraday cage.Four fiber mats were measured and averaged.To determine the relative permittivity, a gold foil with a certain area was placed on the fiber mat deposited on the glass substrate as the upper electrode.The exact area of the gold foil was determined from the image captured by a digital camera (DC-G99, Panasonic) using image analysis software ImageJ (Wayne Rasband).The capacitance was measured using LCR meter (IM3536, Hioki E.E.Corp.) between the upper and bottom electrode at 1.0 V and 1.0 kHz.The relative permittivity, ε r , was calculated using following equation: Here, ε 0 , C, T F0 , and S are vacuum permittivity, capacitance, thickness, and area of the gold foil placed on the fiber mat, respectively.XRD Measurement: XRD measurements were carried out using an automated multipurpose X-ray diffractometer (SmartLab, Rigaku) with Cu Ka radiation.The voltage and current in the X-ray tube were set to 45 kV and 200 mA, respectively.The XRD pattern with 2θ was recorded in the angular range of 5°-30°with scan speed of 5°min À1 .
DSC Measurement: DSC measurements were performed using a DSC-60A Plus apparatus (Shimadzu Corp.).An electrospun PLLA fiber mat of 2.06 mg was heated from ≈20 to 200 °C at a rate of 10 °C min À1 .Only a single measurement could be performed because the PLLA fiber mat melted during the first measurement.
Biodegradable Touch Sensor: As shown in Figure 7b, two PLLA films, 30 mm Â 30 mm Â 128 μm in size, were used as the substrates of the touch sensor.In the central square area of 20 mm Â 20 mm on one side of each PLLA film, colloidal graphite paste (G7711, EMJapan Co., Ltd.) was applied as an electrode and dried for 1 day using a vacuum dryer.The electrospun PLLA fiber mat was directly deposited on the colloidal graphite electrode on one of the two PLLA films, with the electrode electrically grounded.The two PLLA films with and without the fiber mat were sandwiched, as shown in Figure 7b.In the PLLA films, the PDLLA films were introduced as spacers and thermoadhesive layers.The PDLLA films maintained distance between the upper colloidal graphite electrode and the surface of the PLLA fiber mat while no pressure was applied; moreover, the PDLLA films adhered to the four sides of the two PLLA films after heating the four sides with a heat sealer (FV-802, Hakko Corp.).PDLLA films with dimensions of 30 mm Â 30 mm were used, and a hole of 20 mm Â 20 mm was made in the central square area of each film.The output voltage was measured using an oscilloscope (GDS-3504, Good Will Instruments Co., Ltd.) with a probe resistance of 10 MΩ.
Biodegradable Mask Type Sensor: Nonwoven PLA fabrics were prepared by decomposing a commercially available 3-layer nonwoven disposable mask (Electronic Mask, Samuraiworks, Inc.) and coating it with PEDOT/ PSS in an area within a diameter of 90 mm from the center as the electrode.A filter was not used in this case because it was made of polypropylene, which is not biodegradable.A conductive yarn for the output signal was sewn from the center to the edge of the nonwoven fabric.The PLLA fiber mat was directly deposited onto the conductive area by electrospinning.Immediately after electrospinning, the mask-type sensor was fabricated using the layers A to C: A-nonwoven PLA fabric without the fiber mat; B-nonwoven PLA fabric with PLLA fiber mat; and C-pure nonwoven PLA fabric to act as an isolating layer between skin of the person wearing it and the electrospun fiber mat or the electrode (Figure 8b).The layers A-C, ear pads, and nose pads of the upper and lower parts were then integrated using a stapler and dried for 2 days in a vacuum dryer.The output voltage from the mask-type sensor was measured using an oscilloscope (GDS-3504, Good Will Instruments Co., Ltd) with a probe resistance of 10 MΩ.A tablet (iPad Air, iOS ver.12. 5.5, Apple.Inc.) was connected to the mask-type sensor via a 3.5 mm diameter, four conductor phone plugs.To recognize the connection to the mask and the tablet, a 1 kΩ metal film resistance was introduced to the cable in parallel with the mask-type sensor.The voice recognition software, Siri, was launched when the output voltage from the mask sensor was input to the tablet.
All experiments involving human participants were carried out upon the approval from the Human Subjects Research Ethics Review Board of Kyoto Institute of Technology (Approval number: 2022-89), and informed written consent was obtained from the participants.

Figure 1 .
Figure 1.a) Schematic of the electrospinning setup to produce PLLA fiber mats.b) SEM of the electrospun PLLA fiber mat.Inset shows an enlarged FESEM image of a single-electrospun PLLA fiber.c) XRD pattern of the electrospun PLLA fiber mat.d) DSC thermogram of the electrospun PLLA fiber mat.

Figure 3 .
Figure3.Average surface potential of the PLLA fiber mats over time.Each plot and standard deviation errors were calculated from the surface potentials measured from four fiber mats.

Figure 4 .
Figure 4.The surface potential of the PLLA fiber mat stored for 2, 80, and 240 days under high-humidity conditions and at different temperatures.

Figure 5 .
Figure 5. Relationship between amount of output charge from the metal disk probe and a) the position of the disk probe and b) the pressure applied to the fiber mat when the disk probe approached and loaded onto the fiber mat.c) Calculated d app of the fiber mat before and after spraying with IPA.The d app of the fiber mat after spraying with IPA and pressing/releasing for 10 times is also plotted.

Figure 6 .
Figure 6.Schematic explaining the charge output from the upper electrode: a) the upper electrode is placed at a position h = d 0 ; b) the upper electrode moved to h = d; and c) the upper electrode presses the fiber mat at h = d (d < 0).

Figure 7 .
Figure 7. a) The photograph of the top view of the developed touch sensor.Two electrical leads were connected to upper and bottom colloidal graphite electrodes.b) The schematic of the cross section of the touch sensor.c) The output voltage from the touch sensor with time when the sensor was pressed and released repeatedly.Here, the output voltage from the sensor without the PLLA fiber mat is also shown.

Figure 8 .
Figure 8. a) Photograph and b) cross section of the schematic of the developed mask-type acoustic sensor.c) The output voltage from the sensor when the person wearing a mask vocalized "Kyoto Institute of Technology."d) Photograph of the tablet displaying the phrase vocalized by the person wearing the mask.

Table 1 .
Properties of the individual PLLA fibers and their fiber mats.Error values represent the standard deviation.