Preparation of Carbon Powder from HPC Derived from Sub-Bituminous Coal (SB-HPC) And Discussion of Its Properties as An EDLC Electrode Material

Okuma S, Moriyasu Y, Inoue T, Shishido T, Hamaguchi M and Toyoda M

Published on: 2024-03-06

Abstract

Carbon precursor powders were prepared by the precipitation method using HPC derived from subbituminous coal as a raw material, carbonized at different temperatures, and the carbonization temperature dependence of the surface properties and EDLC properties of the obtained carbon powders was investigated. The specific surface area obtained carbon material increased with increasing carbonization temperature, reaching a maximum of 1268 m² g^-1 at a carbonization temperature of 1200^oC without activation. Its micropores (D_ave.) formed in the obtained carbon powder was 0.67 to 0.71 nm, suggesting that it has ultra-micropores, which are extremely fine among the micropores. On the other hand, at the carbonization temperature of 1300 ^oC, Dave increased to 0.78 nm and the specific surface area decreased. It was inferred that this was due to the expansion of the pore diameter due to the connection of micropores. Carbon powders treated at different carbonization temperatures were applied as electrode materials in a water-based EDLC with 40% H₂SO₄ as the electrolyte. At a current density of 50 mA g^-1, the capacities per weight were 284 F g^-1, 265 F g^-1, 281 F g^-1, 254 F g^-1 and 212 F g^-1 at carbonization temperatures from 900 to 1300 ^oC. A comparison of the oxidation/reduction peak areas obtained by CV measurement showed that the oxidation/reduction peak areas obtained by CV measurement showed the maximum at a carbonization temperature of 900^oC, suggesting that the surface oxygen-containing functional groups. We investigated the pore characteristics and EDLC characteristics as the effect of CO₂ activation on carbon powder derived from SB-HPC. Micropores were formed in the SB-HPC-derived carbon powder after activation, and the specific surface area increased to 1675 m² g^-1 after 3.0 h of activation. When the activated carbon powder was applied to the EDLC electrode material, an increase in the capacitance per unit weight was observed as the specific surface area increased. However, since the capacitance per volume decreased, it was inferred that the unactivated carbon powder on SB-HPC, which showed excellent capacitance per weight and per volume, is suitable for the EDLC electrode material.

Keywords

HPC; EDLC; Activation

Introduction

Carbon materials are used as EDLC electrode materials because of their excellent electrical conductivity, chemical and thermal stability, easy availability and processing, and low cost. In particular, activated carbon (AC) is most commonly used as an EDLC electrode material. In Japan and abroad, the search for the optimal pore structure for EDLC was conducted to increase the effective electrode surface area of activated carbon [1-4]. The preparation of carbon materials such as activated carbon, carbon fiber, and carbon powder using HPC as a carbon precursor has been reported. X.Y.Zhao et al. prepared activated carbon by carbonizing HPC and chemically activated it with KOH and NaOH, and reported its EDLC properties [5]. HPC was heat-treated and carbonized at 600 ^oC, and the carbonized HPC and NaOH were mixed so that the weight ratio was 1 : 4. We have obtained activated carbon with a specific surface area of 3110 m² g^-1. They also reported that its EDLC properties showed 43.9 F g^-1 in 0.5 M TEABF₄/PC electrolyte solution. The preparation of activated carbon using HPC as a raw material has the advantage that it does not require ash removal compared to using coal as a raw material, and its usefulness as an EDLC electrode material has been clarified. In addition, J. Yang et al. reported the preparation of carbon fibers by melt spinning using HPC as a raw material [6]. The carbon fiber obtained by heat-treating the pre-prepared precursor fiber at 800 ^oC for 5 min has an average tensile strength of 1150 MPa, and the carbon fiber made from HPC has a general performance from the hot-melt spinning method. It has been clarified that carbon fiber can be obtained at low cost.

Toyoda et al. focused on the fact that HPC is soluble in organic solvents, and prepared carbon fibers and carbon powder from a solution of HPC dissolved in pyridine [7,8]. When HPC was dissolved in pyridine, a viscous solution exhibiting conductivity was obtained. After carbon precursor fibers were prepared by electrospinning, they were carbonized by heat treatment at 900 ^oC, and then the carbon fibers formed micropores without activation treatment [8]. It was reported that electrospinning is a new method for preparing microporous carbon fibers that does not require activation treatment, whereas hot melt spinning requires activation treatment to obtain pores. They reported that after dissolving HPC in pyridine, HPC was precipitated in water, which is a poor solvent, and carbon powder was prepared by heat treatment and carbonization at 900 °C [7]. The obtained carbon powder has micropores with an average pore diameter of 0.75 nm and ~ demonstrating, 310 F g^-1, similar performance to activated carbon powder. It was also reported that a capacitance of 210 F cm^-3 (260 F g^-1) was obtained in the same electrolyte by CO₂ activation treatment for 60 or 90 min. Based on this, we reported that carbon powder with very fine ultra-micropores can be obtained by the precipitation method using a poor solvent [8]. Although it has been reported that carbon powder made from bituminous coal-derived HPC as a starting material exhibits a capacitance comparable to that of commercially available activated carbon or exceeds it by CO₂ activation treatment [7], but bituminous coal is used for ironmaking (manufacturing of steel) and fuel applications. There are concerns about price hikes and resource depletion due to the wide range of demand for such materials. Compared to high-grade bituminous coal, sub-bituminous coal has the characteristics of low aromatic carbon fraction, small aromatic ring size, low number of layers and low crystallinity. Since there are many [6], it is expected that a high specific surface area can be obtained. The carbon powder using HPC derived from sub-bituminous coal shows porosity without activation treatment, as in the case of using conventional HPC derived from bituminous coal. Preparation of porous carbon is expected.

In this study, carbon powder was prepared from HPC derived from sub-bituminous coal, and its EDLC properties were investigated.

Experimental

Starting Material

Sub-bituminous coal-derived HPC (SB-HPC) [Kobe Steel, Ltd.] was used as the starting material. This HPC was ground with an agate mortar and an agate pestle, and then sieved to a size of less than 250 μm.

Preparation of HPC Solution

HPC solution was prepared by adding the ground and sieved HPC to pyridine [Fuji Film Wako Pure Chemical Industries, Ltd., purity 99.5 %] so that the weight ratio of HPC: pyridine = 1 : 1. The HPC solution was obtained by stirring with a stirrer at room temperature for 1 h.

Preparation of Carbon Precursor Powder

Carbon powder was prepared from the HPC solution by a precipitation method using water as a poor solvent. The prepared HPC solution was poured into water to precipitate HPC. The precipitate was filtered by suction and vacuum-dried at 110 ^oC for 12 h to recover the carbon precursor powder.

Preparation of Carbon Powder

Carbon powder was prepared by heat-treating the prepared carbon precursor powder. The carbon precursor powder obtained by the precipitation method was infusibilized and carbonized using an electric furnace (KRB-24HH, Isuzu Manufacturing Co., Ltd., MS548, Motoyama Co., Ltd.). In the infusibility treatment, air was supplied into the furnace core tube at a flow rate of 1.0 dm³ min^-1 using an air pump, and the temperature was raised to 300 ^oC at a heating rate of 60 ^oC h^-1. It was allowed to cool. In the carbonization treatment, at a carbonization temperature of 900 to 1100 ^oC, nitrogen gas was flowed at room temperature at a flow rate of 1.0 × 10^-1 dm³ min^-1 for 4 h to replace the inside of the core tube with nitrogen. The temperature was raised from 900 to 1100 ^oC at a heating rate of 1, held at that temperature for 0.5 h, and then naturally cooled to room temperature. At the carbonization temperature of 1200 and 1300 ^oC, Ar gas was flowed at room temperature at a flow rate of 3.0 × 10^-1 dm³ min^-1 for 4 h to replace the inside of the core tube with argon, and then the heating rate was 60 ^oC h^-1. The temperature was raised to 1200 ^oC and 1300 ^oC at , held at that temperature for 0.5 h, and then naturally cooled to room temperature. The yield of the obtained carbon powder was calculated from the following formula (1).

Y_Carbon = W₂/ W₁ x 100 (1)

where Y_Carbon is the yield of carbon powder (%), W₁ is the weight of carbon precursor before heat treatment (g), and W₂ is the weight of carbon powder after heat treatment (g).

Preparation of Activated Carbon Powder

Activated carbon powder was prepared by CO₂ activation treatment after heat treatment of the prepared carbon precursor powder. The carbon precursor powder obtained by the precipitation method was infusibilized, carbonized, and activated with CO₂ using an electric furnace (KRB-24HH, Isuzu Manufacturing Co., Ltd.). In the infusibility treatment, air was supplied into the furnace core tube at a flow rate of 1.0 dm³ min^-1 using an air pump, and the temperature was raised to 300 ^oC at a heating rate of 60 ^oC h^-1. After that, it was allowed to cool. In the carbonization treatment, the inside of the core tube was replaced with nitrogen by flowing nitrogen gas at a flow rate of 1.0 × 10^-1 dm³ min^-1 for 4 h at room temperature, and then held at that temperature for 0.5h. After that, the temperature was raised to 950^oC at a heating rate of 200 ^oC h^-1, and the supplied gas was switched to a mixed gas of nitrogen and CO₂. The gas flow rate was adjusted to 0.05 dm³ min^-1 for nitrogen and 0.05 dm³ min^-1 for CO₂ gas, and the flow rate of the entire mixed gas was 1.0 dm³ min^-1. CO₂ activation treatment was performed by holding at 950 ^oC for 0.5 to 1.5 h in this mixed gas atmosphere, and then the supply gas was switched to nitrogen again and allowed to cool naturally to room temperature. The yield of the obtained carbon powder was calculated from the following formula (2).

Y_Carbon = W₂/ W₁ x 100 (2)

where Y_Carbon is the yield of activated carbon powder or carbon powder (%), W₁ is the weight of carbon precursor before heat treatment (g), and W₂ is the weight of activated carbon powder or carbon powder after heat treatment (g).

Characterization

Surface Characterization

The pore characteristics of the obtained carbon powder were evaluated using a 77K nitrogen gas adsorption/desorption analyzer (Autosorb-3B, Quantachrome instruments Japan LLC). All samples used for measurement were degassed in advance at 200 ^oC for 18 h or more using a vacuum pump. The specific surface area was calculated from the adsorption isotherm of the obtained carbon powder using the α_s analysis method. Total specific surface area obtained by the α_s analysis method and the average micro pore diameter (D_ave.) were calculated.

Evaluation of Crystallinity

Crystallinity of the obtained carbon powder was evaluated using an X-Ray Diffraction (XRD) device (Rigaku Rint-2000).

EDLC Characterization

Electrode Fabrication

EDLC electrode material was prepared by mixing the obtained carbon powder: acetylene black: PTFE at a weight ratio of 8 : 1 : 1 and processed into a circular sheet with a diameter of 1 cm. The volume of the electrode was calculated by measuring the thickness of the electrode at 5 or more points and using the average value.

Charge/Discharge Measurement

Charge/discharge measurements were performed using a battery charge/discharge device (HJ1001 SD8, Hokuto Denko Co., Ltd.). A three-electrode cell was used as the cell, carbon material of HPC derived from sub-bituminous coal (SB-HPC) was used as the working electrode. Platinum foil was used as the current collector for the working and counter electrodes, and a saturated KCl silver-silver chloride reference electrode (BAS Inc. RE-1C) was used as the reference electrode. 40 % sulfuric acid was used as the electrolyte.

Figure 1 (a): Cell configuration used for EDLC measurement, (b) Scheme of electrochemical cell for the EDLC measurement in aqueous H₂SO_4.

Figure 1(a) shows the configuration of the three-electrode cell used for the electrochemical measurements. In order to sufficiently impregnate the carbon electrode with the electrolyte, the working electrode of the three-electrode cell is placed on a glass filter paper, a platinum foil is placed on top of it, and both sides are covered with Teflon sheets. It was fixed by sandwiching it between plates. For the counter electrode, a platinum foil was placed on a glass filter paper, and both sides were sandwiched between Teflon plates. After assembling the cell, put the cell in a beaker and put the beaker into a 0.5 dm³ separable flask. The separable flask was covered with a separable cover, fixed with a clamp, and then evacuated for 60 min using a vacuum pump. Then, using a 0.1 dm³ cylindrical separating funnel, 0.05 dm³ of 40 % sulfuric acid was poured into the beaker inside the separable flask, and after nitrogen was sent into the separable flask, 0.01 dm³ of 40 % sulfuric acid was added. Finally, the beaker inside was taken out and the reference electrode was inserted into the center of the cell {Figure 1(b)}. Measurements were performed by connecting + current and + voltage to the working electrode, - current to the counter electrode, and - voltage to the reference electrode. Nitrogen was bubbled during the measurement. Charge-discharge measurements were performed in a potential range of “0 – 1” V and current densities of 50 mA g^-1, 100 mA g^-1, 500 mA g^-1, and 1000 mA g^-1, three cycles each. The capacitance per unit weight was calculated using Eq. (3) in the range of 0.2 – 0.8 V of the obtained discharge curve. In addition, the capacitance per unit volume was calculated using equation (4).

C_m = (I Δt )/( m ΔV) (3)

C_v = ( I Δt )/( d ΔV) (4)

where, C_m [F g^-1] : capacitance per unit weight, C_v [F cm^-3] : capacitance per unit volume,

I [A] : current value, Δt [s] : time, m [g] : electrode weight, ΔV [V] : potential difference, d [cm³] : electrode volume.

Cyclic Voltammetry (CV) Measurement

CV measurement was performed by AUTOMATIC POLARIZATION SYSTEM HSV-3000 (Hokuto Denko Co., Ltd.) using the 3-electrode cell. Its CV measurement was performed for 5 cycles under the measurement conditions of a potential range of 0 to 1.0 V and a sweep rate of 1 mV s^-1.

Sample Code

The obtained carbon powder was expressed as SB - (carbonization temperature) based on the type of HPC used as the raw material and the carbonization temperature.

Results And Discussions

Carbonization Yield Of SB-HPC-Derived Carbon Powder

SB-HPC-derived carbon precursor powder was prepared by dissolving SB-HPC in pyridine, a good solvent, and adding it to water, a poor solvent. It was obtained by heating at 110 ^oC for 12 h. The recovery rate of the obtained carbon precursor powder was 96 %, suggesting complete precipitation and recovery, similar to the previously reported case of using HPC derived from bituminous coal [9,10].

Table 1 shows the carbonization yield of each carbon powder at carbonization temperatures of 900 to 1200 ^oC. At carbonization temperatures of 900 and 1000 ^oC, its yield was high, approximately 50 % and although the yield decreased with increasing carbonization temperature, it was still approximately 30 % even at the highest carbonization temperature of 1300 ^oC. The yield was maintained, showing high productivity.

Table 1: Carbonization yield of SB-HPC derived Carbon powder.

Sample	Carbonization yield / ^o/_o
SB-900	50.2
SB-1000	50.8
SB-1100	40.3
SB-1200	39.9
SB-1300	33.3

Crystallinity Evaluation Of SB-HPC-Derived Carbon Powder

Figure 2 shows XRD patterns of carbon powder carbonized at 900 to 1300 ^oC. A broad peak corresponding to the carbon 002 diffraction line was observed around 2θ = 26° for all carbon powders. In addition, no significant change was observed in the diffraction pattern with increasing heat treatment temperature, and all of the peaks were broad. It was confirmed that the carbon powder of HPC derived from subbituminous coal has low crystallinity even when the carbonization temperature reaches 1300 ^oC.

Figure 2: XRD patterns of carbon powder heat treated at 900 ~ 1300 ^oC.

Pore Characteristics of Carbon Powder Derived From SB-HPC And Its Temperature Dependence

N₂ gas adsorption/desorption isotherms of carbon powders carbonized at 900 to1300°C were measured. In all isotherms, an increase in adsorption was observed in the low relative pressure region (around P/P_o = 0), suggesting the formation of micropores. The amount of adsorption in the low relative pressure region for each carbon powder was 130 cc g^-1 for SB-900, 162 cc g^-1 for SB-1000, 185 cc g^-1 for SB-1100, and 206 cc g^-1for SB-1200, indicating that the amount of adsorption increased as the heat treatment temperature increased. On the other hand, SB-1300 showed 183 cc g^-1, indicating a decrease in adsorption and a difference in isotherms. The adsorption behavior of N₂ molecules on the micropores in the low relative pressure region was investigated, and it was found that all the carbon powders started to adsorb at a relative pressure of around 10^-3. It was presumed that this was caused by the stabilization effect (micropore filling) due to the interaction potential of the pore walls of the micropores, which promoted the adsorption of N₂ molecules inside the micropores. Therefore, the SB-HPC carbon powder was considered to have not only supermicropores (0.7 to 2 nm) but also ultramicropores (0.7 nm < ) within the micropores. Therefore, by performing α_s analysis using the adsorption isotherm of each carbon powder, the total specific surface area (S_total), micropore specific surface area (S_micro), external specific surface area (S_ext.), and average pore diameter of micropores (D_ave.) were calculated and the micropores of each carbon powder were investigated. Figure 3 shows their dependence on the carbonization temperature, and Table 2 shows the calculated values.

Total specific surface area of each carbon powder is 737 m² g^-1 for SB-900, 935 m² g^-1 for SB-1000, 1101 m² g^-1 for SB-1100, 1268 m² g^-1 for SB-1200 and 1194 m² g^-1 for SB-1300 at carbonization temperatures from 900 to 1200 ^oC, and its specific surface area increased with increasing heat treatment temperature. The increase in S_total in this temperature range was not significantly affected by the carbonization temperature, and the increase in S_micro was due to the increase in S_ext. On the other hand, when carbonized at 1300 ^oC, S_micro decreased and S_ext. increased compared to 1200 ^oC. In addition, although D_ave. also varies between 900 and 1200 ^oC, there is no significant difference.

Therefore, up to a carbonization temperature of 1200 ^oC, the surface area increased due to the development of micropores as the heat treatment temperature increased. From this, when the carbon powder carbonized at 900 ^oC to 1200 ^oC is applied to the EDLC electrode material, the capacitance does not necessarily depend on the surface area value, but it is expected that the capacitance will increase.

EDLC Characterization

EDLC electrodes were fabricated using SB-HPC-derived carbon powders prepared by heat treatment and carbonization at 900 ^oC to 1300 ^oC, and their EDLC properties were evaluated. Carbon powder carbonized at 900, 1000, 1100, 1200 and 1300 ^oC was applied to the EDLC electrode. The charge-discharge curves at current densities of 50 mA g^-1 to 1000 mA g^-1 when applied as materials are shown in Figure 4, 1100 ^oC for (a) and 1200 ^oC for (b). The obtained charge-discharge curves showed an improvement in linearity as the current density increased. The contribution of the electric double layer capacitance due to adsorption/desorption was suggested. In addition, a decrease in linearity was observed around 0.4 V to 0.6 V in the discharge curve, suggesting the contribution of pseudo-capacitance due to the surface-containing oxygen functional groups of the carbon powder. Figure 5 shows a plot of the capacitance calculated from the range of 0.2 V to 0.8 V of the discharge curve at each current density against the current density.

Figure 3: Dependence of specific surface area of carbon powders.

Table 2: Pore parameter and Carbonization yield of carbon powders.

Sample	S_total / m² g^-1	S_micro / m² g^-1	S_ext. / m² g^-1	S_micro / S_total	D_ave / nm
SB-900	737	537	200	0.73	0.69
SB-1000	935	745	190	0.8	0.67
SB-1100	1101	899	202	0.82	0.69
SB-1200	1268	1093	175	0.86	0.71
SB-1300	1194	938	258	0.79	0.78

Figure 4: CD curves of (a) SB-1100 and (b) SB-1200.

Figure 5 (a) shows the capacitance per weight for each current density. Capacitances at a current density of 50 mA g^-1 are 284 F g^-1 for SB-900, 265 F g^-1 for SB-1000, 281 F g^-1 for SB-1100, 254 F g^-1for SB-1200, and 212 F g^-1 for SB-1300, respectively. From this, it was clarified that when the SB-HPC-derived carbon powder is applied, it has a high capacitance at a relatively low carbonization temperature range from 900 ^oC to 1100 ^oC. In addition, the capacitance at a current density of 1000 mA g^-1 is 249 F g^-1 for SB-900, 236 F g^-1 for SB-1000, 245 F g^-1 for SB-1100, 228 F g^-1 for SB-1200, and 192 F g^-1 for SB-1300, respectively. Although the capacitance decreased as the current density increased, it was confirmed that the capacitance of about 87 % to about 90 % was maintained in both cases. Capacitance per volume for each current density shown in Figure 5 (b) is 207 F cm^-3 for SB-900, 153 F cm^-3 for SB-1000, 146 F cm^-3 for SB-1100, 154 F cm^-3 for SB-1200, and 148 F cm^-3 for SB-1300, respectively, and then SB-900 showed the maximum value compared with the heat treatment temperature. From this, it can be inferred that SB-900 is the most suitable EDLC electrode material for each capacitance per volume. In addition, the dependence of the capacitance of the SB-HPC-derived carbon powder treated at various carbonization temperature is shown in Figures 6 (a) and (b), respectively. Although the specific surface area increased up to the carbonization temperature of 1200 ^oC, no significant difference was observed in the capacitance per unit weight, and SB-900 showed the highest capacitance per volume. This suggests that the capacitance of the SB-HPC-derived carbon powder is largely affected by factors other than the specific surface area.

Figure 5: Current density dependence of capacitance (a) per mass and (b) per unit volume.

Figure 6: Relationship between specific surface area and capacitance (a) per unit mass and (b) per unit volume.

Figure 7 shows the results of CV (cyclic voltammetry) measurements performed at a scanning rate of 1 mV s^-1. Oxidation and reduction peaks were observed at around 0.4 V to 0.6 V, confirming pseudocapacitance due to surface oxygenated functional groups. A comparison of the peak areas revealed that the peak areas decreased with increasing carbonization temperature, suggesting a decrease in surface oxygenated functional groups. From the above discussion, it was considered that the effect of the oxygen-containing functional groups on the surface of the SB-HPC-derived carbon powder was greatest at the carbonization temperature of 900 ^oC. In addition, SB-900 heat-treated at a carbonization temperature of 900 ^oC was inferred to be suitable as an EDLC electrode material.

Figure 7: CV curves of SB-900, SB-1000, SB-1100, SB-1200 and SB-1300.

Pore characteristics and carbonization yield of SB-HPC-derived carbon powder for CO₂ activation

Figure 8 (a) shows the N₂ gas adsorption and desorption isotherms of the carbon powder that was activated at 950^oC for 0.5 to 3.0 h and the carbon powder that was carbonized at 900^oC. In unactivated HPC-0.0, an increase in adsorption amount was observed in the low relative pressure region (around P/P_o = 0), suggesting the formation of micropores. The adsorption amount of each carbon powder in the low relative pressure region is 127 cc g^-1 for HPC-0.0, 187 cc g^-1 for HPC-0.5, 206 cc g^-1 for HPC-1.0, 226 cc g^-1 for HPC-1.5, and 256 cc g^-1 for HPC-3.0, respectively, and then those values increased with CO₂ activation. Figure 8 (b) shows the adsorption and desorption isotherms on a logarithmic scale. It was confirmed that micropores were formed because the amount of adsorption increased at a relative pressure of around 10^-3 by prolonging the activation time. On the other hand, the adsorption behavior of HPC-3.0 was different from that of the activation treatment at a relative pressure of 10^-1 to 10⁰ for 0.5 to 1.5 h. Therefore, we investigated the micropores for each carbon powder activation treatment by α_s analysis. Figure 9 shows the total specific surface area (S_total), micropore specific surface area (S_micro), and external specific surface area (S_ext.) calculated by α_s analysis of each carbon powder versus the activation treatment time. Table 3 shows the calculated pore diameter (D_ave.) and the yield. The total specific surface area of each carbon powder was 729 m² g^-1 for HPC-0.0, 1086 m² g^-1 for HPC-0.5, 1205 m² g^-1 for HPC-1.0, 1381 m² g^-1 for HPC-1.5, and 1675 m² g^-1 for HPC-3.0, respectively. Its specific surface area increased with increasing activation time. This increase in specific surface area increased with an increase in S_micro, and although there was some variation in D_ave. up to an activation time of 1.5 h, no significant difference was observed. On the other hand, in HPC-3.0 with an activation time of 3.0 h, although S_total increased, S_ext. increased slightly, and D_ave. expanded to 0.75 nm. I thought it happened. In addition, although the yield of carbon powder decreased as the activation time increased, it was maintained at approximately 31 % at an activation time of 3.0 h. Even when the activation treatment was performed after the charcoal treatment, the productivity was higher than that of the commercially available activated carbon that had been subjected to the physical activation treatment.

Figure 8: (a) N₂ adsorption and desorption of carbon powder, (b) Logarithmic of N₂ adsorption and desorption isotherms.

Figure 9: Dependence of specific surface area of carbon powders.

Table 3: Pore parameter and Carbonization yield of carbon powders.

Sample	S_total / m² g^-1	S_micro / m² g^-1	S_ext. / m² g^-1	S_micro / S_total	D_ave / nm	Carbonization Yield / %
SB-0.0	729	531	198	0.73	0.68	55.6
SB-0.5	1086	1035	52	0.95	0.67	43.6
SB-1.0	1205	1152	53	0.96	0.67	43.7
SB-1.5	1381	1326	55	0.96	0.69	37.2
SB-3.0	1675	1588	87	0.95	0.75	30.8

EDLC characterization of SB-HPC-derived carbon powder for CO₂ activation

Using activated carbon powder prepared by CO₂ activation treatment for 0.5 h to 3.0 h and carbon powder carbonized at 900 ^oC without activation, EDLC electrodes were fabricated and their EDLC properties were evaluated. Figure 10 shows the charge-discharge curves at a current density of 50 mA g^-1 for each carbon powder. In the activation treatment time of 0.5 h to 1.5 h, variations in charge/discharge time were observed. However, when the activation treatment was performed for 3.0 h, the charge/discharge time was prolonged. When the capacitance was calculated in the range of 0.2 V to 0.8 V for each discharge curve, it was shown that 284 F g^-1 for unactivated, 280 F g^-1 for 0.5 h, 304 F g^-1 for 1.0 h, 277 F g^-1 for 1.5 F g^-1, and 277 F g^-1 for 3.0 h activation, respectively. Figure 11 (a) shows the capacitance per weight for each current density of 50 mA g^-1 to 1000 mA g^-1, 249 F g^-1 for unactivated, 251 F g^-1 for 0.5 h, 263 F g^-1 for 1.0 h, 253 F g^-1 for 1.5 h, and 308 F g^-1 for 3.0 h. Although the electric capacitance decreased, it maintained a capacitance of approximately 87 to 92 %. The capacitance per volume at each current density shown in Figure 11 (b) is 207 F cm^-3 for unactivated, 170 F cm^-3 for 0.5 h activated, 138 F cm^-3 for 1.0 h activated, and 141 F cm^-3 for 1.5 h activated, and 125 F cm^-3 at 3.0 h activation. Although the capacitance per weight of HPC-3.0 increased, the capacitance per volume decreased. It was inferred that the unactivated carbon powder is suitable for the EDLC electrode material because it showed excellent capacitance per volume.

Figure 10: CD curves of carbon powder and activated carbon powder at current density 50 mA g^-1.

Figure 11: Current density dependence of capacitance (a) per mass and (b) per unit volume.

Relationship between pore properties and capacitance for CO₂ activation

Figure 12 (a) shows the specific surface area and the capacitance per weight of each carbon powder with respect to the CO₂ activation time. As the activation time increased, the specific surface area increased, but there was no significant difference in capacitance per weight between unactivated and 0.5-1.5 h activation. On the other hand, Figure 12 (b) shows the specific surface area and the capacitance per volume of each carbon powder with respect to the CO₂ activation time, and it was confirmed that the capacitance decreased as the activation time increased. Therefore, although CO activation of SB-HPC-derived carbon powder increased the specific surface area, when it was applied to the EDLC electrode material, the capacitance per weight increased, but the capacitance per volume increased. It did not lead to an increase in capacitance. Therefore, we investigated the relationship between the pore characteristics of carbon powder and the capacitance with respect to CO₂ activation. As a study on mesopores, Figure 13 (a) shows the pore size distribution expressed by the BJH method using the N₂ adsorption isotherm of each carbon powder. No increase in pore volume was observed in the mesopore region (2 ~ 50 nm) due to the activation treatment. As an examination of micropores, Figure 13 (b) shows the pore size distribution expressed by the HK method. Activation treatment increased the number of micropores, reaching a maximum around 0.7 nm. After 0.5 h and 1.0 h of activation, there was no significant increase in the number of pores compared to the unactivated sample, but after 1.5 h and 3.0 h of activation, an increase in pores was observed within the same range, reaching a maximum at 3.0 h. rice field. From this, it was inferred that the increase in the capacitance per weight at 3.0 h was due to the increase in micropores, while the formation of pores by the activation treatment decreased the capacitance per volume.

Figure 12: Activation times relationship of capacitance and specific surface area (a) per unit mass, (b) per unit volume.

Figure 13: Pore size distribution of carbon powder (a) by HK method and (b) BJH method.

Conclusion

D_ave.of the micropores formed in the obtained carbon powder was 0.67 to 0.71 nm, suggesting that it has ultra-micropores, which are extremely fine among the micropores. The specific surface area increased with increasing carbonization temperature, reaching a maximum of 1268 m² g^-1 at a carbonization temperature of 1200^oC. On the other hand, at the carbonization temperature of 1300^oC, D_ave increased to 0.78 nm and the specific surface area decreased. It was inferred that this was due to the expansion of the pore diameter due to the connection of micropores. Although changes in surface properties were observed, there was no significant difference in crystallinity, suggesting low crystallinity. Carbon powders treated at different carbonization temperatures were applied as electrode materials in a water-based EDLC with 40% H₂SO₄ as the electrolyte.

At a current density of 50 mA g^-1, the capacities per weight were 284 F g^-1, 265 F g^-1, 281 F g^-1, 254 F g^-1, and 212 F g^-1 at carbonization temperatures of 900 to 1300 ^oC. Although the specific surface area is maximum at 1200 ^oC, it was suggested that it has a high capacitance at 900 to 1100 ^oC. Capacities per unit volume are shown 207 F cm^-3, 153 F cm^-3 146 F cm^-3, 154 F cm^-3, and 148 F cm^-3. It showed the highest capacitance at 900 ^oC. A comparison of the oxidation/reduction peak areas obtained by CV measurement showed that the oxidation/reduction peak areas obtained by CV measurement showed the maximum at a carbonization temperature of 900 ^oC, suggesting that the surface oxygen-containing functional groups. From these results, carbonized powder was considered suitable for EDLC electrode material.

We investigated the pore characteristics and EDLC characteristics as the effect of CO₂ activation on carbon powder derived from SB-HPC. Micropores were formed in the SB-HPC-derived carbon powder after activation, and the specific surface area increased to 1675 m² g^-1 after 3.0 h of activation. However, although the yield decreased with increasing activation time, it was maintained at approximately 31 % at the maximum activation time of 3.0 h. Even when treated, the productivity was higher than that of commercially available activated carbon that had undergone physical activation. When the activated carbon powder was applied to the EDLC electrode material, an increase in the capacitance per unit weight was observed as the specific surface area increased. However, since the capacitance per volume decreased, it was inferred that the unactivated carbon powder on SB-HPC, which showed excellent capacitance per weight and per volume, is suitable for the EDLC electrode material.

Acknowledgement

This research was supported by JSPSGrant -in-Aid fpr Scientific Research (C) (JP 22K04689) from the Japan society for the Promotion of Science, and Kobe Steel, Ltd.

Journal of Physics and Chemistry Research