Fabrication of Activated Multiporous Carbon Nanoﬁbers Using Vacuum Plasma for High-Capacity Energy Storage

: Porous carbon nanoﬁbers are widely used as supercapacitor electrode materials due to their excellent physical adsorption/desorption operation and smooth transport of ions. The acid/base activation method is commonly used to generate micropores on the surface of carbon nanoﬁbers, but controlling the activation level and minimizing the release of harmful chemicals pose challenges. This study proposed a method for producing activated multiporous carbon nanoﬁbers that is easier to operate and more environmentally friendly. It utilizes the vacuum plasma process to enhance surface area and introduce functional groups onto the electrospun polymer nanoﬁbers. Subsequent heat treatment results in the formation of activated multiporous carbon nanoﬁbers. The type and density of the functional group introduced into the carbon structure were adjusted to the type of plasma gas (O 2 , NH 3 and C 4 F 8 ) being exposed. Among them, oxygen plasma-treated carbon nanoﬁbers (O-MPCNFs) not only have a much larger active surface (517.84 m 2 g − 1 ) than other gases (290.62 m 2 g − 1 for NH 3 and 159.29 m 2 g − 1 for C 4 F 8 ), but also generate a lot of micropores, promoting rapid adsorption/desorption-inducted charges; therefore, they have excellent energy storage capacity. The O-MPCNF-based symmetrical two-electrode supercapacitor has a high speciﬁc capacitance (173.28 F g − 1 ), rate capability and cycle stability (94.57% after 5000 cycles).


Introduction
Supercapacitors and batteries are two popular types of energy storage being stimulated for development to meet the demand for portable electronics and electric vehicles.Researchers aim for higher power and energy density, that depend significantly on the materials used in these devices [1][2][3][4].As a versatile method, electrospinning has been actively used to produce one-dimensional (1D) nanomaterials such as polymers, inorganic materials and sp 2 carbon with diameters ranging from nanometers to several micrometers [5][6][7][8].In particular, 1D nanomaterials formed using electrospinning have a larger specific surface area and aspect ratio and more pore interconnections than those manufactured using other processes, making them suitable for energy storage and as conversion materials [9,10].In addition, the electrospinning method is environmentally friendly and offers a simple principle of operation as well as production flexibility and cost reduction [11].Among the electrospinning 1D materials, carbon nanofibers are not only easily manufactured in large quantities, but are also widely used as supercapacitor electrode materials due to their physical adsorption/desorption operation mechanism [12][13][14].Many polymers are used as precursors for synthesizing carbon nanofibers using electrospinning, such as polyimide, cellulose, polyacrylonitrile (PAN), etc.Among them, PAN is the most commonly used precursor because of its high carbon yield and excellent compatibility with electrospinning systems [15][16][17].
Porous carbon nanofibers not only efficiently transport electrons in the longitudinal direction, but also enable the smooth transport of ions due to their high surface-to-volume ratio with the electrolyte [18][19][20].Therefore, research on the production of porous carbon nanofibers is actively conducted in order to maximize the energy storage capacity of carbon nanofibers [21,22].The most commonly used method is to induce a chemical reaction using a strong acid or a strong base aqueous solution with carbon nanofibers derived from PAN in order to prepare surface-modified activated carbon nanofibers [23,24].Recently, an attempt has been made to prepare and form porous carbon nanofibers using electrospinning from a mixed solution of a carbon precursor (e.g., PAN) and a pore-forming polymer (poly(methyl methacrylate) (PMMA) and poly(styrene) (PS)) [25][26][27][28].Jang's group prepared composite porous carbon nanofibers to which MnO 2 nano-hairs were attached by means of chemical oxidation and applied as a supercapacitor electrode material [29].However, most of the processes that attempt to activate carbon fibers not only fail to control the amount and type of functional groups introduced using solution-based methods but also have the disadvantage of discharging harmful substances into the environment.
The plasma method is a widely used process of modifying the surface of a substrate using high-energy plasma gas [30][31][32].Recently, research has been attempted that prepares activated carbon by exposing a high-energy plasma gas to carbon materials [33,34].Specifically, the goal is to develop a carbon material with micropores and structural defects due to introducing functional groups by inducing a reaction between chemically unreactive sp 2 carbon and high-energy plasma gas [35,36].Adhamash's group improved the storage capacity of a supercapacitor electrode by activating biochar using a plasma process [37].However, so far, research on the activation of carbon materials using the plasma method has not reached the stage of precise control of the activation and reaction degree of nanostructures.
In this work, we fabricated activated multiporous carbon nanofibers using modified electrospinning and vacuum plasma processes and applied them as supercapacitor electrode materials with excellent energy storage performance.First, the electrospun nanofibers were exposed to three different types of plasma gas (O 2 , NH 3 , C 4 F 8 ) to generate micropores and finely controlled functional groups on their structure.Afterward, heat treatments were conducted to transform activated nanofibers into activated multiporous carbon nanofibers.Among them, carbon nanofibers with oxygen-related functional groups (O-MPCNFs) not only maximized the active surface area (517.84 m 2 g −1 ) but also increased the volume of micropores, allowing for the rapid absorption/desorption of ions, resulting in excellent energy storage performance.When O-MPCNFs are used as a symmetrical two-electrode supercapacitor electrode material, they have a high specific capacitance of 173 F g −1 and show excellent speed capability and cycle stability (94.57% after 5000 cycles).

Fabrication of Activated Multiporous Carbon Nanofibers
The PAN solution was mixed with the DMF solvent in a weight ratio of 10% and stirred for 12 h at a temperature of 60 • C. The PS solution was prepared in the same manner as the PAN solution.The PAN and PS solutions were then mixed in the same weight ratio and stirred for 3 h.The mixed polymer solution was electrospun using a 23 G needle at an applied voltage of 10 kV and a flow rate of 10 µL min −1 .The distance from the tip to the collector was fixed at 15 cm [38].The electrospun nanofibers were subjected to vacuum plasma treatment of different gases (O 2 , NH 3 and C 4 F 8 ).The vacuum plasma process was conducted using a COVANCE-3MPR-RF 13.56 MHz rf plasma chamber with an internal pressure of 5 × 10 −3 torr at 16 sccm at a power of 100 W for 5 min.Then, the nanofibers subjected to plasma treatment were carbonized at a heating rate of 5 • C min −1 in an Ar atmosphere at 800 • C for 1 h [39].

Characterization
Field emission scanning electron microscopy (FE-SEM) images were obtained using a Hitachi S-4300SE.X-ray photoelectron spectrometer (XPS) spectra were obtained using a thermo-scientific K-alpha+.The surface area and pore size distribution were obtained using a Micromeritics ASAP 2020.Fourier-transform infrared (FTIR) spectra were recorded using a Spotlight 200i Frontier (PerkinElmer, Waltham, MA, USA).The Raman spectrum was reported using a BX51 (Olympus, Tokyo, Japan) spectrometer.

Three-Electrode Electrochemical Performance Measurement
In order to measure the electrode capacitance, an active material powder, carbon black and poly(vinylidene fluoride) (PVDF) polymer binder were dissolved in N-methylpyrrolidone to prepare a homogeneous paste.The mass ratio of the homogeneous paste was 85:5:10 for active material powder, carbon black and PVDF, respectively.The mixture was coated onto a stainless-steel mesh (1 mg of active materials per unit area (cm −2 )), and dried in a 60 • C oven for 12 h.The electrochemical properties of the carbon nanostructure samples (working electrode) were measured using a three-electrode system, a Hg/HgO reference electrode, a platinum counter electrode and KOH 1 M as an electrolyte.The capacitance of the electrodes of the active materials was calculated using galvanostatic measurements according to Equation (1) [40]: where I, ∆V, ∆t and m are the discharge current, voltage range, discharge time and mass of active material, respectively.

Symmetric Supercapacitor Device with Electrochemical Performance
Symmetric supercapacitor devices were generated to confirm the capacity of the activated carbon nanofibers.The supercapacitor device consists of two working electrodes (activated carbon nanofibers), filter papers soaked in 1 M KOH as separators and two layers of acrylic on the bottom and top for securing the devices.The capacitance of the symmetric cell is calculated in galvanostatic measurements according to Equation (2) [41]: where I is the discharge current, ∆V is the voltage difference of the discharge, ∆t is the discharge time and m is the mass of active material in each electrode.

Fabrication of Multiporous Activated Carbon Nanofibers
Figure 1 shows the method of fabricating activated multiporous carbon nanofibers (a-MPCNFs) and the structural changes in the nanofibers at each step.First, to produce a template structure for forming a multiporous carbon nanofiber, different polymer mixed solutions (polyacrylonitrile (PAN) and polystyrene (PS)) were electrospun using a single nozzle.In the above step, PAN forms a continuous phase and PS forms a discontinuous phase in the mixed solution due to phase separation resulting from the different physical properties (i.e., surface tension) of the two polymers [42].Then, under the influence of the pressure from the syringe pump applied to the mixed solution and the electric field force between the nozzle and the ground, a multi-core nanofiber is formed that surrounds the PS cylinder through the PAN.The multi-core nanofibers are exposed to high-energy gas molecules using a vacuum plasma device, and functional groups related to exposed gas molecules are introduced onto the surface.Three types of plasma gases (O 2 , NH 3 and C 4 F 8 ) are used for the activation of the multi-core nanofibers to introduce different types of functional groups.Thereafter, the multi-core polymer nanofibers form multiporous carbon nanofibers, to which a functional group is introduced in the process of stabilization and carbonization.During the heat treatment process, the PAN in the polymer nanofibers forms functional groups by interacting with plasma gases introduced during the process of transformation into sp 2 carbon whereas the PS undergoes pyrolysis, creating channels and micropores in the fiber structure [43].
Appl.Sci.2023, 13, x FOR PEER REVIEW 4 of 14 molecules are introduced onto the surface.Three types of plasma gases (O2, NH3 and C4F8) are used for the activation of the multi-core nanofibers to introduce different types of functional groups.Thereafter, the multi-core polymer nanofibers form multiporous carbon nanofibers, to which a functional group is introduced in the process of stabilization and carbonization.During the heat treatment process, the PAN in the polymer nanofibers forms functional groups by interacting with plasma gases introduced during the process of transformation into sp 2 carbon whereas the PS undergoes pyrolysis, creating channels and micropores in the fiber structure [43].The composition of the electrospinning polymer solution and the type of plasma gas were adjusted to confirm the conditions affecting the formation of the carbon nanostructure.First, carbon nanofibers produced using only PAN and without plasma processing form a nonporous smooth surface structure with a uniform diameter of ca.400 nm (Figure S1).On the other hand, the multi-core polymer nanofiber-based carbon nanostructures formed by electrospinning of the PAN and PS mixed polymer solution show a different The composition of the electrospinning polymer solution and the type of plasma gas were adjusted to confirm the conditions affecting the formation of the carbon nanostructure.First, carbon nanofibers produced using only PAN and without plasma processing form a nonporous smooth surface structure with a uniform diameter of ca.400 nm (Figure S1).On the other hand, the multi-core polymer nanofiber-based carbon nanostructures formed by electrospinning of the PAN and PS mixed polymer solution show a different structure.The electrospun polymer nanofibers, which are not subjected to plasma treatment, have a diameter of ca.420 nm and a smooth surface (Figure 2a,b).Under the influence of gas plasma, the polymer surface is etched by high-energy molecules, reducing the fiber diameter to ca. 400 nm and making the surface rough (Figure 2c,d).The morphology of carbon nanofibers formed after the carbonization process also shows different aspects Appl.Sci.2023, 13, 10840 5 of 14 depending on the exposure to plasma gas.When not subjected to plasma treatment, carbon fibers with a relatively smooth surface and an average diameter of ca.360 nm are formed in the channel structure (Figure 3a,b).However, in the case of carbon fibers treated with oxygen plasma, the channel is formed evenly in the structure, but the diameter is reduced to ca. 320 nm and the surface roughness is significantly increased (Figure 3c,d).This structure change occurs in the same way when other plasma gases (NH 3 and C 4 F 8 ) are used (Figure S2).In other words, the etching phenomenon caused by the high-energy plasma gas also affects the structure of the final carbon structure, resulting in an uneven surface formation.
structure.The electrospun polymer nanofibers, which are not subjected to plasma treatment, have a diameter of ca.420 nm and a smooth surface (Figure 2a,b).Under the influence of gas plasma, the polymer surface is etched by high-energy molecules, reducing the fiber diameter to ca. 400 nm and making the surface rough (Figure 2c,d).The morphology of carbon nanofibers formed after the carbonization process also shows different aspects depending on the exposure to plasma gas.When not subjected to plasma treatment, carbon fibers with a relatively smooth surface and an average diameter of ca.360 nm are formed in the channel structure (Figure 3a,b).However, in the case of carbon fibers treated with oxygen plasma, the channel is formed evenly in the structure, but the diameter is reduced to ca. 320 nm and the surface roughness is significantly increased (Figure 3c,d).This structure change occurs in the same way when other plasma gases (NH3 and C4F8) are used (Figure S2).In other words, the etching phenomenon caused by the high-energy plasma gas also affects the structure of the final carbon structure, resulting in an uneven surface formation.

Characterization of Multiporous Activated Carbon Nanofibers
X-ray photoelectron spectroscopy (XPS) was used to confirm the change in the composition of surface elements and functional groups of the carbon structure depending on the plasma gas exposure.Figure 4a shows a carbon structure that is not exposed to plasma treatment (MPCNFs), which is decomposed into four peaks as viewed from the C1 peak of the material.Specifically, the four peaks represent the following functional groups: 284.4 eV for a C-C or C=C aromatic ring; 285.6 eV for a C and N sp 2 single bond; 286.7 eV for a C and O single bond (e.g., epoxy and hydroxyl groups); 288.3 eV for a C and O double bond (e.g., carbonyl and carboxyl groups) [44][45][46].When exposed to plasma, not only does the ratio of the four peaks change, but peaks associated with the elements of the exposed gas are also generated.For the carbon structure generated via exposure to oxygen plasma (O-MPCNFs), the intensities of the 286.7 eV and 288.3 eV peaks increase, which indicates the bond between carbon and oxygen atoms (Figure 4b).Among the two peaks, 288.3 eV is more clearly increased, indicating that the carbonyl group is superior to the other functional groups (i.e., hydroxyl and quinone) generated via oxygen plasma exposure.In the case of carbon fibers treated with ammonia gas plasma (N-MPCNFs), the 285.6 eV peak, which indicates the bonding of sp 2 carbon and nitrogen, was clearly increased (Figure 4c).This is because the high-energy ammonia plasma provides positively excited nitrogen components (nitrogen-containing molecules or atoms) that introduce nitrogen into the polymer chain structure.After that, going through the carbonization process, nitrogen is directly doped into the sp 2 carbon chain.Carbon nanofibers exposed to C 4 F 8 plasma (F-MPCNFs) show new peaks at 290.2 eV and 292.8 eV in addition to the existing four peaks (Figure 4d).The 290.2 eV and 292.8 eV peaks show the C-F-C bond and the F-C-F bond, respectively, indicating that the F element is doped into the sp 2 carbon structure.To gain insight into the changes that occurred in nanofibers after exposure to plasma gas, Fourier-transform infrared (FTIR) spectra were investigated (Figure S3a).In the case of O 2 plasma (O-MPCNFs), an increase in the intensity of the peaks at 3433 cm −1 and 1220 cm −1 was observed, which corresponded to the -OH and -C-O, respectively.When NH 3 was exposed, the intensity of the peaks at 2356 cm −1 and 1606 cm −1 characterizing -C-N and =C-N, respectively, increased.The small C-F bonds at 2921 cm −1 and 816 cm −1 were also detected in the case of C 4 F 8 plasma (F-MPCNFs).Furthermore, a difference in the D band intensity of a-MPCNF samples compared with MPCNFs was observed, which meant that functional groups derived after the plasma process contributed to the formation of defects in the carbon structure, leading to the increases in the intensity of the D band (Figure S3b).

Characterization of Multiporous Activated Carbon Nanofibers
X-ray photoelectron spectroscopy (XPS) was used to confirm the change in the composition of surface elements and functional groups of the carbon structure depending on the plasma gas exposure.Figure 4a shows a carbon structure that is not exposed to plasma treatment (MPCNFs), which is decomposed into four peaks as viewed from the C1 peak of the material.Specifically, the four peaks represent the following functional groups: 284.4 eV for a C-C or C=C aromatic ring; 285.6 eV for a C and N sp 2 single bond; 286.7 eV for a C and O single bond (e.g., epoxy and hydroxyl groups); 288.3 eV for a C and O double bond (e.g., carbonyl and carboxyl groups) [44][45][46].When exposed to plasma, not only does the ratio of the four peaks change, but peaks associated with the elements of the exposed gas are also generated.For the carbon structure generated via exposure to oxygen plasma (O-MPCNFs), the intensities of the 286.7 eV and 288.3 eV peaks increase, which indicates the bond between carbon and oxygen atoms (Figure 4b).Among the two peaks, 288.3 eV is more clearly increased, indicating that the carbonyl group is superior to the other functional groups (i.e., hydroxyl and quinone) generated via oxygen plasma exposure.In the case of carbon fibers treated with ammonia gas plasma (N-MPCNFs), the 285.6 eV peak, Functional groups introduced into the polymer chain after plasma treatment increase the surface roughness and create various structural defects after the carbonization step and change the surface area of the carbon structure.The method of adsorption and desorption using nitrogen gas was used to confirm the difference in surface area and micropore distribution of carbon nanofibers depending on plasma exposure and gas type (Figure 5).The active surface area and pore distribution of the material were illustrated using the Brunauer-Emmett-Teller (BET) and Barrett-Joyner-Halenda (BJH) methods, respectively.First, when comparing the BET curve of each material, the adsorption in the low-pressure range (0~0.2P/P 0 ) of the micropores is clearly visible when plasma treatment is performed, and is greatest when oxygen gas is used.The shape of the curve in the range of relatively higher pressures (>0.8 P/P 0 ) also shows the greatest change value when oxygen plasma treatment is performed [47].The calculated surface area values for each material are as follows: 24.98 m 2 g −1 for MPCNFs; 517.84 m 2 g −1 for O-MPCNFs; 290.62 m 2 g −1 for N-MPCNFs; 159.29 m 2 g −1 for F-MPCNFs.When the pore distributions of the carbon nanofibers are compared, in the case without plasma, the distribution of pores other than the 30 nm diameter channel generated by PS pyrolysis is not shown.However, during the oxygen plasma treatment, a large number of micropores with a diameter of ca. 2 nm was produced, whereas in the cases of NH 3 and C 4 F 8 plasma, mesopores with a diameter of ca. 10 nm are formed rather than micropores.Therefore, the O 2 plasma process confirmed that oxygen atoms could be most easily intercalated into carbon to activate the surface and minimize the collapse of the structure.
Appl.Sci.2023, 13, x FOR PEER REVIEW 7 (O-MPCNFs), an increase in the intensity of the peaks at 3433 cm −1 and 1220 cm −1 observed, which corresponded to the -OH and -C-O, respectively.When NH3 wa posed, the intensity of the peaks at 2356 cm −1 and 1606 cm −1 characterizing -C-N and N, respectively, increased.The small C-F bonds at 2921 cm −1 and 816 cm −1 were also tected in the case of C4F8 plasma (F-MPCNFs).Furthermore, a difference in the D b intensity of a-MPCNF samples compared with MPCNFs was observed, which meant functional groups derived after the plasma process contributed to the formation of de in the carbon structure, leading to the increases in the intensity of the D band (Figure Functional groups introduced into the polymer chain after plasma treatment incr the surface roughness and create various structural defects after the carbonization and change the surface area of the carbon structure.The method of adsorption and sorption using nitrogen gas was used to confirm the difference in surface area and cropore distribution of carbon nanofibers depending on plasma exposure and gas (Figure 5).The active surface area and pore distribution of the material were illustr using the Brunauer-Emmett-Teller (BET) and Barrett-Joyner-Halenda (BJH) method spectively.First, when comparing the BET curve of each material, the adsorption in low-pressure range (0~0.2P/P0) of the micropores is clearly visible when plasma treatm is performed, and is greatest when oxygen gas is used.The shape of the curve in the r of relatively higher pressures (>0.8 P/P0) also shows the greatest change value when gen plasma treatment is performed [47].The calculated surface area values for each terial are as follows: 24.98 m 2 g −1 for MPCNFs; 517.84 m 2 g −1 for O-MPCNFs; 290.62 m for N-MPCNFs; 159.29 m 2 g −1 for F-MPCNFs.When the pore distributions of the ca nanofibers are compared, in the case without plasma, the distribution of pores other

Electrochemical Performance of the Multiporous Activated Carbon Nanofiber-Based Electrodes
A three-electrode system was used to compare the energy storage performance of porous carbon nanofibers according to the plasma process.The three-electrode system consisted of the multiporous carbon fibers as the working electrode, Hg/HgO as the reference electrode, and Pt as the counter electrode in the 1 M KOH electrolyte.Figure 6a shows the cyclic voltammetric (CV) curve according to the constant voltage rate (5 mV s −1 ) of each carbon fiber electrode (−1 V to 0 V vs. Hg/HgO).Each CV curve is similar in shape, but its area increases in the order of C 4 F 8 , NH 3 and O 2 plasma.This is because the introduction of O atoms into the carbon structure via O 2 gas plasma is smoother than in the other cases (N atoms via NH 3 and F atoms via C 4 F 8 ).A galvanostatic charge/discharge measurement was also conducted to confirm the charge storage performance of each carbon electrode (Figure 6b).The voltage range for measurement is from −1 to 0 V for a current density of 1.0 A g −1 .All charge/discharge curves have linear symmetrical shapes with materials that exhibit excellent capacitive behavior.The specific capacitance of each carbon electrode calculated using the discharge time from the curve is as follows: the 30 nm diameter channel generated by PS pyrolysis is not shown.However, during the oxygen plasma treatment, a large number of micropores with a diameter of ca. 2 nm was produced, whereas in the cases of NH3 and C4F8 plasma, mesopores with a diameter of ca. 10 nm are formed rather than micropores.Therefore, the O2 plasma process confirmed that oxygen atoms could be most easily intercalated into carbon to activate the surface and minimize the collapse of the structure.

Electrochemical Performance of the Multiporous Activated Carbon Nanofiber-Based Electrodes
A three-electrode system was used to compare the energy storage performance of porous carbon nanofibers according to the plasma process.The three-electrode system consisted of the multiporous carbon fibers as the working electrode, Hg/HgO as the reference electrode, and Pt as the counter electrode in the 1 M KOH electrolyte.Figure 6a shows the cyclic voltammetric (CV) curve according to the constant voltage rate (5 mV s −1 ) of each carbon fiber electrode (−1 V to 0 V vs. Hg/HgO).Each CV curve is similar in shape, but its area increases in the order of C4F8, NH3 and O2 plasma.This is because the introduction of O atoms into the carbon structure via O2 gas plasma is smoother than in the other cases (N atoms via NH3 and F atoms via C4F8).A galvanostatic charge/discharge measurement was also conducted to confirm the charge storage performance of each carbon electrode (Figure 6b).The voltage range for measurement is from −1 to 0 V for a current density of 1.0 A g −1 .All charge/discharge curves have linear symmetrical shapes with materials that exhibit excellent capacitive behavior.The specific capacitance of each carbon electrode calculated using the discharge time from the curve is as follows: 202.4 F g −1 for MPCNFs, 358.2 F g −1 for O-MPCNFs, 286.1 F g −1 for N-MPCNFs and 268.2 F g −1 for F- To confirm the rate capability of each carbon electrode, CV curves were suggested according to the voltage scan rate (from 5 mV s −1 to 100 mV s −1 ). Figure 6c shows the CV curve of the O-MPCNF electrode, and as the voltage scan rate increases, the intensity of the responding current steadily increases, and thus the area increases.Other porous carbon nanofiber electrodes show the same phenomenon, but the intensity of the corresponding current is lower than that of oxygen plasma (Figure S4).Even on the galvanostatic charge/discharge curve of each carbon electrode, the reduction in the discharge time according to the increase in current density does not decrease sharply and remains constant (Figures 6d and S5).As the current density of each carbon electrode increases (from 1 A g −1 to 5 A g −1 ), the calculated specific capacitance based on the discharge time changes as follows: 202.41 F g −1 to 154.87 F g −1 for MPCNFs; 358.21 F g −1 to 299.72 F g −1 for O-MPCNFs; 286.13 F g −1 to 234.76 F g −1 for N-MPCNFs; 268.2 F g −1 to 219.24 F g −1 for F-MPCNFs (Figure 6e).Comparing the reduction ratio of each carbon electrode, the oxygen-plasma-treated electrode and non-plasma-treated electrode show the minimum and maximum values, respectively (83.4% for O-MPCNFs, 81.8% for N-MPCNFs, 80.5% for F-MPCNFs and 76.5% for MPCNFs).
MPCNFs.For a better comparison, the energy storage performances of the O-MPCNF based electrode and other carbon fiber-based electrodes are listed in Table 1.

Carbon Nanofibers Specific Capacitance (Three-Electrode System) Reference
PAN-derived nitrogen-doped carbon fibers To investigate the electrochemical and structural performance of the electrode materials, electrochemical impedance spectroscopy (EIS) measurements were performed in the frequency range of 100 kHz to 10 mHz. Figure 6f shows the Nyquist plots of the carbon nanofiber-based electrodes that suggest the semicircle is amplified in the highfrequency impedance region, a linear shape at low frequencies and the x-intercept.In the plot, the x-intercept starting with the semicircle and the diameter of the semicircle represent the equivalent series resistance (ESR) and the charge transfer resistance at the electrode-electrolyte interface (R ct ), respectively.The ERS value, i.e., the charge transfer resistance in the electrode structure, does not differ significantly depending on the material (0.82 Ω for MPCNFs, 0.33 Ω for O-MPCNFs, 0.45 Ω for N-MPCNFs and 0.58 Ω for F-MPCNFs).On the other hand, the R ct value, which can be calculated using the diameter of the semicircle, tends to decrease with increasing surface area of the carbon structure (0.95 Ω for O-MPCNFs, 1.18 Ω for N-MPCNFs, 1.54 Ω for F-MPCNFs and 2.14 Ω for MPCNFs).This is because the contact area between the electrode and the electrolyte is increased due to the formation of micropores in the carbon by the introduction of heterogeneous elements, thereby facilitating charge transport.

Electrochemical Performance of the Multiporous Activated Carbon Nanofiber in a Symmetric Supercapacitor
To further confirm the electrochemical performance of the multiporous carbon nanofibers, they were applied as electrodes in a symmetrical supercapacitor device.The structure of the symmetrical supercapacitor consists of a separator with a 1 M KOH aqueous solution between two carbon fiber-based working electrodes placed onto a basic frame designed using acrylic (Figure S6). Figure 7 shows the electrochemical performance of each carbon electrode in a symmetrical supercapacitor device.The CV curve of each device was measured from 0 V to 1 V.The current of the symmetrical supercapacitor device varied with increasing voltage scan rate, while all four types of carbon electrode maintained a similar rectangular shape (Figure 7a and Figure S7).However, the plasma-treated carbon electrode shows a wider curved area than the non-plasma-treated carbon electrode, and the oxygen-plasma-treated carbon electrode has the largest curved area.Comparing the galvanostatic curve of each supercapacitor device depending on the change in current density, the change in the graph shape according to the increase in current density is not shown, and the shortening of the discharge time is also maintained in a constant ratio (Figure 7b and Figure S8).The specific capacitance of each device at 1 A g −1 is as follows: 96 F g −1 (MPCNFs); 173.28 F g −1 (O-MPCNFs); 138.12 F g −1 (N-MPCNFs); 128.2 F g −1 (F-MPCNFs) (Figure S9).In addition, the specific capacitance change in each device with increasing current density (from 1 A g −1 to 5 A g −1 ) is 73.4% (93.44 F g −1 ) for MPCNFs, 84.2% (145.91 F g −1 ) for O-MPCNFs, 81.7% (112.74F g −1 ) for N-MPCNFs and 80.1% (102.68F g −1 ) for F-MPCNFs.
The EIS method was applied to confirm the charge transfer of each storage device (Figure 7c).The ESR and R ct values of each storage device show similar patterns to those of the three electrodes.However, the slope value of the straight line at low frequencies varies depending on the material.The devices based on non-plasma-treated carbon electrodes (MPCNFs) exhibit a straight-line slope of about 45 • at low frequency.However, when the carbon electrodes exposed to plasma are treated in sequence with C 4 F 8 , NH 3 and O 2 plasma, the slope of the straight line approaches 90 • .This is because the supercapacitor device behaves like an ideal capacitor as the movement of charges becomes smooth due to the micropores generated by the plasma treatment.
The verification of the repeatable energy storage performance of a supercapacitor device is an essential factor for the actual application of the device.Figure 7d shows the specific capacity changes in the energy systems according to 5000 repeated charging/discharging experiments with constant current (1 A g −1 ).All four types of carbon-based device maintained excellent energy storage performance, with specific values as follows: 91.76% for MPCNFs, 94.57% for O-MPCNFs, 91.88% for N-MPCNFs, 91.83% for F-MPCNFs.This is because carbon-based devices operate on the principle of physical adsorption/desorption of charges, and even plasma-treated carbon-based devices maintain a uniform operating principle.
(Figures 7b and S8).The specific capacitance of each device at 1 A g −1 is as follows: 96 F g −1 (MPCNFs); 173.28 F g −1 (O-MPCNFs); 138.12 F g −1 (N-MPCNFs); 128.2 F g −1 (F-MPCNFs) (Figure S9).In addition, the specific capacitance change in each device with increasing current density (from 1 A g −1 to 5 A g −1 ) is 73.4% (93.44 F g −1 ) for MPCNFs, 84.2% (145.91 F g −1 ) for O-MPCNFs, 81.7% (112.74F g −1 ) for N-MPCNFs and 80.1% (102.68F g −1 ) for F-MPCNFs.The EIS method was applied to confirm the charge transfer of each storage device (Figure 7c).The ESR and Rct values of each storage device show similar patterns to those of the three electrodes.However, the slope value of the straight line at low frequencies varies depending on the material.The devices based on non-plasma-treated carbon electrodes (MPCNFs) exhibit a straight-line slope of about 45° at low frequency.However, when the carbon electrodes exposed to plasma are treated in sequence with C4F8, NH3 and O2 plasma, the slope of the straight line approaches 90°.This is because the supercapacitor device behaves like an ideal capacitor as the movement of charges becomes smooth due to the micropores generated by the plasma treatment.
The verification of the repeatable energy storage performance of a supercapacitor device is an essential factor for the actual application of the device.Figure 7d shows the specific capacity changes in the energy systems according to 5000 repeated charging/discharging experiments with constant current (1 A g −1 ).All four types of carbon-based device maintained excellent energy storage performance, with specific values as follows: 91.76% for MPCNFs, 94.57% for O-MPCNFs, 91.88% for N-MPCNFs, 91.83% for F-

Conclusions
In summary, the activated multiporous carbon nanofibers with functional groups were fabricated using a simple plasma process of multi-core nanofibers that resulted from single-nozzle co-electrospinning.The activated multiporous carbon nanofibers formed using this process can not only control the functional groups introduced according to the type of plasma gas (O 2 , NH 3 and C 4 F 8 ), but can also form micropores to maximize the active surface area.In particular, oxygen plasma-introduced multiporous carbon nanofibers (O-MPCNFs) exhibit high energy storage performance due to their large surface area and a large number of micropores.The symmetrical supercapacitor device manufactured using O-MPCNFs has a high specific capacitance (173.28F g −1 at 1 A g −1 ) and promotes the rapid movement of charges due to micropores into which oxygen functional groups are introduced, thereby exhibiting an excellent rate capability.In addition, due to the unique physical adsorption/desorption operation principle of the carbon material, it has a high cycle stability of 94.57% even after 5000 repetitions of charging/discharging. Therefore, the results of this study suggest an effective method for easily producing porous carbon nanofibers with a stable structure.

Figure 1 .
Figure 1.(a) Schematic diagram of the fabrication stage of activated multiporous carbon nanofibers.(b) Structural changes in and chemical structures of nanofibers according to each fabrication step.

Figure 1 .
Figure 1.(a) Schematic diagram of the fabrication stage of activated multiporous carbon nanofibers.(b) Structural changes in and chemical structures of nanofibers according to each fabrication step.

Figure 2 .
Figure 2. Field emission scanning electron microscopy (FE-SEM) images of electrospun polymer nanofibers (a,b) before and (c,d) after the plasma process.

Figure 2 .
Figure 2. Field emission scanning electron microscopy (FE-SEM) images of electrospun polymer nanofibers (a,b) before and (c,d) after the plasma process.

14 Figure 3 .
Figure 3. FE-SEM images of multiporous carbon nanofibers (a,b) without plasma and (c,d) after the O2 plasma process.

Figure 3 .
Figure 3. FE-SEM images of multiporous carbon nanofibers (a,b) without plasma and (c,d) after the O 2 plasma process.

Figure 7 .
Figure 7. (a) Cyclic voltammetry curves at various scan rates (5 to 100 mV s −1 ) and (b) galvanostatic charge/discharge curves for different current densities (1 to 5 A g −1 ) of O-MPCNFs.(c) Electrochemical impedance spectroscopy (EIS) plots for different MPCNFs in the frequency range from 100 kHz to 10 mHz.(d) Long-term cycling performance and coulombic efficiency of different MPCNFs with charge/discharge numbers at 1 A g −1 current density.

Figure 7 .
Figure 7. (a) Cyclic voltammetry curves at various scan rates (5 to 100 mV s −1 ) and (b) galvanostatic charge/discharge curves for different current densities (1 to 5 A g −1 ) of O-MPCNFs.(c) Electrochemical impedance spectroscopy (EIS) plots for different MPCNFs in the frequency range from 100 kHz to 10 mHz.(d) Long-term cycling performance and coulombic efficiency of different MPCNFs with charge/discharge numbers at 1 A g −1 current density.

FTIR
and Raman spectra of the different MPCNFs.Figure S4: Rate capability of other multiporous carbon nanofibers.Figure S5: Capacitance maintenance of different multiporous carbon nanofibers.

Figure S6 :
Structure of symmetric supercapacitor.FigureS7: CV curves of other multiporous carbon nanofibers in the two-electrode system.FigureS8: Galvanostatic charge/discharge curves of other multiporous carbon nanofibers in the two-electrode system.
Figure S9: Specific capacitance change of symmetric supercapacitors.

Table 1 .
Comparison of specific capacitance of the O-MPCNF-based electrode with other carbo electrodes.

Table 1 .
Comparison of specific capacitance of the O-MPCNF-based electrode with other carbon electrodes.