Preparation of activated carbon from needle coke via two-stage steam activation process

Preparation of activated carbon from needle coke via two-stage steam activation process

Accepted Manuscript Preparation of activated carbon from needle coke via two-stage steam activation process Ui-Su Im, Jiyoung Kim, Seon Ho Lee, Song m...

2MB Sizes 1 Downloads 66 Views

Accepted Manuscript Preparation of activated carbon from needle coke via two-stage steam activation process Ui-Su Im, Jiyoung Kim, Seon Ho Lee, Song mi Lee, Byung-Rok Lee, DongHyun Peck, Doo-Hwan Jung PII: DOI: Reference:

S0167-577X(18)31560-X https://doi.org/10.1016/j.matlet.2018.09.171 MLBLUE 25043

To appear in:

Materials Letters

Received Date: Revised Date: Accepted Date:

20 June 2018 20 September 2018 30 September 2018

Please cite this article as: U-S. Im, J. Kim, S.H. Lee, S.m. Lee, B-R. Lee, D-H. Peck, D-H. Jung, Preparation of activated carbon from needle coke via two-stage steam activation process, Materials Letters (2018), doi: https:// doi.org/10.1016/j.matlet.2018.09.171

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Preparation of activated carbon from needle coke via two-stage steam activation process Ui-Su Ima,b, Jiyoung Kimc, Seon Ho Leea,d, Song mi Leea,b, Byung-Rok Leea, Dong-Hyun Pecka,b, Doo-Hwan Junga,b,* a

New & Renewable Energy Research Division, Korea Institute of Energy Research,

Yuseong-gu, Daejeon 305350, Republic of Korea b

Advanced Energy and System Engineering, Korea University of Science and Technology,

Yuseong-gu, Daejeon 305350, Republic of Korea c

School of Chemical Engineering, Sungkyunkwan University, Jangan-gu, Suwon-si,

Gyeonggi-do 440746, Republic of Korea d

Department of Chemical and Biomolecular Engineering, Yonsei University, Seoul 120749,

Republic of Korea. * Corresponding Author: E-mail: [email protected]

Abstract Mesoporous activated carbon was prepared from needle coke via two-stage steam activation process under high temperature. Two-stage steam activation method involves three main steps: 1st steam activation to generate mesopores and macropores above 2 nm; coating and impregnation of substances with a low molecular weight into the pores; 2nd steam activation to expand the pore size and micropore volume. The needle coke was characterized by elemental analysis, polarized light microscopy, and thermogravimetric analysis, and the porosity structure was analyzed by the Brunauer-Emmett-Teller (BET) method, the t-plot method, the Barrett-Joyner-Halenda (BJH) method, and the scanning electron microscopy (SEM). The maximum specific surface area and mesopore volume ratio of activated carbon manufactured by two-stage steam activation method was 1134 m2/g and 78 %, respectively. The mesopore volume and average pore diameter increased as more coating and impregnation agents were added.

Keywords needle coke, steam activation, activated carbon, naphthalene, pitch

1. Introduction Recently, the demand for activated carbon has increased due to serious environmental problems, such as air pollution and water pollution and demand for electric double-layer capacitors (EDLCs)[1-3]. Activated carbon is manufactured from wood, sawdust, black liquor lignin, coconut shell, coal, petroleum heavy oil, coal and petroleum pitch, coke, etc[4-6]. The activation methods are generally divided into two methods[7, 8]. One is a physical activation method[9, 10] using steam, H2O2, CO2, and air, and the other is a chemical activation method using KOH, NaOH, ZnCl2, etc[11-13]. The chemical activation method has drawbacks of secondary contamination, complexity of the manufacturing process, and corrosion of the device, but activated carbon with exceptionally high specific surface area can be produced using this method[7]. On the other hand, the physical activation method has a disadvantage of requiring a high temperature of 700-1000 °C to expand the pore structure and increase the specific surface area[8]. Thus, it is important factor to select economical raw materials and active methods depending on the purpose of use for activated carbon. Needle coke, as a carbon material exhibiting excellent electrical conductivity and highdensity characteristics, is used as a raw material for high-performance graphite electrodes, and needle coke can be used as a high-performance carbon material for electric double-layer capacitors (EDLCs) due to its excellent electrical conductivity[14, 15]. However, high-density anisotropic carbon materials, such as needle-shaped coke, are generally difficult to produce as activated carbon having a high specific surface area through physical activation with steam, H2O2, CO2, and air[14, 15]. Recently, an activation method to produce activated carbon having

a high specific surface area using needle coke was reported through chemical activation with KOH and NaOH[13, 15]. This study was designed to economically produce excellent activated carbon for use in electrode materials. Accordingly, activated carbon was developed from needle coke having excellent electrical conductivity through the steam activation method which is economical and easy to control the manufacturing process. In this study, we studied the activated carbon with a high specific surface area from needle coke by inducing a change of the surface structure via two-stage steam activation method, as shown in Fig. 1.

2. Experimental 2.1 Materials The raw material of activated carbon was a coal-based needle coke produced by POSCO ChemTech Co., Ltd. in Korea. Prior to use, needle coke was ground and sieved under 75 μm using a 200 mesh filter. A coal-based pitch for coating and the impregnation agent were obtained from JFE Chemical Co., Ltd. in Japan. Additionally, naphthalene (C10H8; Samchun Pure Chemical Co., Ltd.) was used for coating and impregnation in these experiments. 2.2 Preparation of ANC, N-ANC and P-ANC, AN-ANC and AP-ANC Fig. S1a shows the process flow chart of two-stage steam activation method. The activation device illustrated in Fig. S2b was utilized for both of 1st steam activation and 2nd steam activation. ANC is a sample in which needle coke is activated by 1st steam activation. N-ANC and P-ANC are samples that are coated and impregnated with naphthalene or pitch to ANC, respectively. In addition, AN-ANC is a sample in which N-ANC is activated by 2nd steam activation. Please refer to Supplementary Materials for details of the experimental conditions and sample names. 2.3 Analyses

The as-prepared samples were characterized by elemental analysis, thermogravimetric analysis, the pore and surface structure, and X-ray diffraction (XRD). Please refer to the Supplementary Materials.

3. Results and discussion 3.1 Characterization of the raw material and schematic diagram A schematic diagram of this study describing the surface structure modification of activated carbon is shown in Fig. 1. Needle coke with excellent electrical conductivity is difficult to be used as an activated carbon precursor via general steam activation due to its highly graphitized carbon nature. However, it is possible to develop an activated carbon having an increased specific surface area through the two-stage steam activation after coating and impregnation of an additive with ANC having mesopore and macropore. In terms of the pore structure, the mesopore and macropore properties in 1st steam activation are the key points for the coating and impregnation processes. Ultimate analysis and proximate analysis for needle coke and pitch are listed in Table S1. Thermogravimetric analysis of the needle coke is presented in Fig. S2a. Optical microscopy of the needle coke revealed an anisotropic domain structure, as shown in Fig. S2b. The softening point of the pitch used in this study is approximately 120 °C. 3.2 Effects of 1st steam activation on the porosity development As the water feed rate of 1st steam activation and the reaction temperature increased, the specific surface area of the ANCs increased, as indicated in Fig. 2a. In particular, the rate of the specific surface area increase started to decrease after the reaction temperature was above 850 °C. On the other hand, the yield of the ANC consistently decreased as the temperature increased (Fig. 2b). Fig. 2c, d show the specific surface area and pore size distribution as a function of the reaction time. ANC/850/4/0.12 exhibited the highest specific surface area of 643.3 m2/g and the most developed mesopore volume. From these results, it was found that the

optimum conditions were a reaction temperature of 850 °C, a reaction time of 4 h, and a water feed rate of 0.12 ml/min•gANC. 3.3 Effects of naphthalene coating and 2nd steam activation The specific surface area and total pore volume of the N-ANC were reduced by approximately 300 m2/g and 0.15 cm3/g, more than those ANC/850/4/0.12, as summarized in Table 1. Hence, naphthalene is believed to have properly coated and impregnated the carbon. The specific surface areas of the AN-ANCs were reduced by approximately 20-40 m2/g compared with ANC/850/4/0.12; however, the mesopore surface area and total pore volume increased. The highest average pore diameter was 2.54 nm in AN-ANC/1:4, and the average pore diameter increased as the coating and impregnation ratio of naphthalene increased. Thus, the micropores of ANC/850/4/0.12 expanded and transformed into mesopores during 2nd steam activation. N-ANC/1:4 and AN-ANC/1:4 showed the lowest pore volume and largest pore volume, respectively, as illustrated in Fig. 3a. In addition, pore sizes of more than 100 nm were developed in AN-ANC1:1 and AN-ANC1:4 through 2nd steam activation, as depicted in Fig. S3, which shows the pore structures of needle coke, ANC, and AN-ANC. 3.4 Effects of pitch coating and 2nd steam activation Table 1 shows the porosity characteristics of P-ANC prepared using different impregnation ratios and AP-ANC through 2nd steam activation. The specific surface area of P-ANC coated and impregnated with pitch considerably decreased in comparison with N-ANC. As the coating and impregnation ratio of the pitch increased, the total pore volume and average pore diameter were increased. The highest specific surface area and average pore diameter were 1134.3 m2/g and 2.77 nm observed in AP-ANC/1:3. In particular, the mesopore volume ratios of APANC/1:2 and AP-ANC/1:3 were more than 70 % due to the rapid increase in the mesopore volume. The distribution curves of the AP-ANCs are depicted in Fig. 3b. The mesopores of the AP-ANCs developed as the ratio of the coating and impregnation agents increased. Thus, the

AP-ANCs and AN-ANCs prepared using pitch and naphthalene as the coating and impregnation agents, respectively, using two-stage steam activation exhibited a consistent trend of extending micropores into mesopores[4]. The XRD diagram of needle coke, ANC and AP-ANC are shown in Fig S4. ANC and AP-ANC were found to have the same crystallinity as needle coke, which has excellent electrical conductivity.

4. Conclusion In summary, the results discussed in this paper show that an outstanding mesoporous activated carbon could be prepared via two-stage steam activation method from needle coke, which is difficult to convert into activated carbon by steam activation. The as-prepared activated carbon had a high specific surface area of 1134.3 m2/g and high mesopore volume ratio of 78 %. Thus, the obtained results prove that needle coke with excellent electrical conductivity can be utilized as a precursor for activated carbon, which is a significant asset of preparing activated carbon via two-stage steam activation process.

Acknowledgements This work is sponsored by Korea Institute of Energy Research (Grant No. B7-5521).

References [1] E. Menya, P.W. Olupot, H. Storz, M. Lubwama, Y. Kiros, Chem. Eng. Res. Des. 129 (2018) 271-296. [2] A. Clifford, L. Lang, R. Chen, K.J. Anstey, A. Seaton, Environ. Res. 147 (2016) 383-398. [3] P. González-García, Renewable Sustainable Energy Rev. 82 (2018) 1393-1414. [4] B. Tian, P. Li, D. Li, Y. Qiao, D. Xu, Y. Tian, J. Porous Mater. (2017). [5] N.A. Rashidi, S. Yusup, Chem. Eng. J. 314 (2017) 277-290. [6] C. Zhong, S. Gong, L.e. Jin, P. Li, Q. Cao, Mater. Lett. 156 (2015) 1-6.

[7] D.-W. Kim, H.-S. Kil, K. Nakabayashi, S.-H. Yoon, J. Miyawaki, Carbon 114 (2017) 98105. [8] J.A. Maciá-Agulló, B.C. Moore, D. Cazorla-Amorós, A. Linares-Solano, Carbon 42 (2004) 1367-1370. [9] B.-k. Choi, J.-K. Ko, S.-J. Park, M.-K. Seo, Carbon letters 22 (2017) 96-100. [10] I. Lupul, J. Yperman, R. Carleer, G. Gryglewicz, J. Porous Mater. 22 (2015) 283-289. [11] A.H. Jawad, R.A. Rashid, M.A.M. Ishak, L.D. Wilson, Desalin. Water Treat. 57 (2016) 25194-25206. [12] A.H. Jawad, S. Sabar, M.A.M. Ishak, L.D. Wilson, S.S. Ahmad Norrahma, M.K. Talari, A.M. Farhan, Chem. Eng. Commun. 204 (2017) 1143-1156. [13] Y.H. Park, U.-S. Im, B.-R. Lee, D.-H. Peck, S.-K. Kim, Y.W. Rhee, D.-H. Jung, Carbon Letters 20 (2016) 47-52. [14] S. Mitani, S.-I. Lee, K. Saito, S.-H. Yoon, Y. Korai, I. Mochida, Carbon 43 (2005) 29602967. [15] W. Qiao, S.-H. Yoon, I. Mochida, Energy Fuels 20 (2006) 1680-1684. [16] Z. Hu, M.P. Srinivasan, Y. Ni, Adv. Mater. 12 (2000) 62-65.

Figure with captions

Figure 1. Comparison of the schematic diagrams between general steam activation and twostage steam activation

Figure 2. Effect of activation temperature and time on 1st steam activation: (a) BET surface area and (b) yield according to activation temperature; (c) BET surface area with activation time; (d) change of pore volume distributions

Figure 3. Influence on the pore volume distribution according to the coating and impregnation ratio of (a) naphthalene, (b) pitch.

Table with captions Table 1. Pore structure parameters of ANC, N-ANC, AN-ANC, P-ANC and AP-ANC calculated from the nitrogen adsorption isotherms. SBET Sme Vtot Vmi Vme Vme/Vtot Sample 2 2 3 3 3 (m /g) (m /g) (cm /g) (cm /g) (cm /g) (%) ANC/850/4/0.12 N-ANC/1:2

643.3 366.2

86.9 51.0

0.340 0.192

0.235 0.138

0.105 0.054

31 28

d [nm] 2.11 2.09

N-ANC/1:4 375.6 59.2 0.193 0.135 0.058 30 AN-ANC/1:2 601.4 139.2 0.349 0.202 0.147 42 AN-ANC/1:4 621.0 197.2 0.395 0.182 0.213 54 P-ANC/2:1 40.3 30.6 P-ANC/1:1 5.2 4.9 AP-ANC/2:1 492.6 95.7 0.266 0.173 0.093 35 AP-ANC/1:1 602.7 155.0 0.343 0.194 0.149 43 AP-ANC/1:2 1013.0 668.1 0.706 0.152 0.554 78 AP-ANC/1:3 1134.3 665.9 0.789 0.220 0.569 72 * by difference, SBET BET surface area, Sme mesopore surface area, Vtot total volume, Vmi micropore volume, Vme mesopore volume, d average pore diameter, Vme/Vtot mesopore volume ratio

2.05 2.32 2.54 2.16 2.28 2.58 2.77

Supplementary data Preparation of samples The pulverized needle coke was activated at reaction temperatures of 800, 850, and 900 °C and reaction times of 2, 4, and 6 h through the activation device illustrated in Fig. S2b. In addition, 1st steam activation experiments were performed by supplying water at 0.06, 0.09, 0.12, and 0.15 ml/min•gANC during the active reaction time. The prepared activated needle cokes (ANCs) were labeled ANC followed by the numbers indicating the reaction temperature, reaction time, and feed rate of water. For example, ANC/850/4/0.12 refers to the ANC activated at 850 °C for 4 h under a water feed rate of 0.12 ml/min•gANC. Coated and impregnated samples (N-ANC and P-ANC) were prepared by mixing ANC/850/4/0.12 and ‘naphthalene or pitch’ under vacuum. Coated and impregnated activated carbons (N-ANC and P-ANC) were mixed with ANC/850/4/0.12 in mass ratios of 2:1, 1:1, 1:2, 1:3, and 1:4. In addition, depending on the properties of the coating and impregnating agents, naphthalene and pitch were heat treated at 200 °C for 1 h and at 150 °C for 1 h respectively under vacuum. The sample names of N-ANC and P-ANC are given by listing the mass ratio of the coating and impregnation agents after N-ANC or P-ANC. N-ANC and P-ANC that were

treated by 2nd steam activation were labeled AN-ANC or AP-ANC and named by listing the mass fraction of the coating and impregnation agents. The conditions of 2nd steam activation were fixed at a reaction temperature of 900 °C, a heating rate of 10 °C/min, a reaction time of 2 h, and a water feed rate of 0.12 ml/min•gANC in a nitrogen atmosphere Thus, all of the sample names imply step-by-step process of two-stage steam activation. For example, AP-ANC/1:3 is described as follows: first, ANC was activated at 850 °C for 4 h under a water feed rate of 0.12 ml/min•gANC; second, ANC/850/4/0.12 and pitch were mixed at a mass rate of 1:3 and heat treated at 150 °C for 1h under vacuum; finally, as-prepared P-ANC/1:3 was again activated at 900 °C for 2 h under a water feed rate of 0.12 ml/min•gANC. Characterization The chemical composition of C, H, N, and S were analyzed using elemental analysis (TruSpec Elemental Analyzer and SC-432DR Sulfur Analyzer, LECO Corp., USA). The oxygen contents were calculated by the contents of C, H, N, and S. Polarization microscopy analysis was performed by polarized light microscopy (BX51M, Olympus Corp., Japan). Thermal characteristics were measured by thermogravimetric analysis (STA 409 PC, Netzsch Corp., Germany) at a heating rate of 5 °C/min to 900 °C under a nitrogen flow of 20 ml/min. The pore structure was determined from a nitrogen adsorption-desorption isotherm at -196 °C using a porosity analyzer (Belsorp-mini II, MicrotracBEL Corp., Japan). Prior to the measurements, all of the samples were sieved through a 200 mesh filter under 75 µm and were degassed at 150 °C under vacuum for 3h. The specific surface area (SBET) and the total pore volume (Vtot) were measured using the Brunauer-Emmett-Teller (BET) method. The mesopore surface area (Sme) and the micropore volume (Vmi) were calculated using the t-plot method[16]. The mesopore volume was determined by the difference between Vmi and Vtot. Curves of the pore size distribution were obtained using the Barrett-Joyner-Halenda (BJH)

method according to the BELSORP analysis program software. The surface structure was observed using scanning electron microscopy (SEM, JSM-6700F, JEOL Ltd., Japan). The crystallinities of needle coke, ANC and AP-ANC were analyzed using X-ray diffractometry (XRD, RTP 300 RC, Rigaku Corp., Japan).

Figure S1. Flow diagram and apparatus for two-stage steam activation method: (a) flow diagram; (b) apparatus, (1) temperature control, (2) nitrogen gas system, (3) pre-heating system, (4) water pump, (5) tube furnace for activation reactor, (6) stores for distillates, (7) thermocouple, (8) activation reactor in detail.

Figure S2. Thermogravimetric curve and polarization microscopic analysis for needle coke

Figure S3. SEM images of (a) needle coke, (b) ANC/850/4/0.12, (c) AN-ANC/1:1, and (d) AN-ANC/1:4

Figure S4. X-ray diffraction patterns of needle coke, ANC/850/4/0.12 and AP-ANC/1:3 Table S1. Characteristics of needle coke and pitch Sample Proximate analysis (wt%) Volatile Fixed Moisture Ash matters carbon

Ultimate analysis (wt%) C

H

N

S

O

Needle coke

1.38

0.17

7.39

91.06

96.2

2.9

0.5

0.2

0.2

Pitch

1.42

0.29

44.35

53.94

95.2

4.1

0.5

0.1

0.1

- Mesoporous activated carbon was prepared from needle coke via two-stage steam activation process. - The activated needle cokes treated by 1st steam activation exhibited the highest specific surface area of 643.3 m2/g. - 2nd steam activation using the naphthalene makes the average pore diameter to increase. - The highest specific surface area and average pore diameter showed 1134.3 m2/g and 2.77 nm via two-stage steam activation.

Table with captions Table 1. Pore structure parameters of ANC, N-ANC, AN-ANC, P-ANC and AP-ANC calculated from the nitrogen adsorption isotherms.

Sample

SBET (m2/g)

Sme (m2/g)

Vtot (cm3/g)

Vmi (cm3/g)

Vme (cm3/g)

Vme/Vtot (%)

ANC/850/4/0.12 643.3 86.9 0.340 0.235 0.105 31 N-ANC/1:2 366.2 51.0 0.192 0.138 0.054 28 N-ANC/1:4 375.6 59.2 0.193 0.135 0.058 30 AN-ANC/1:2 601.4 139.2 0.349 0.202 0.147 42 AN-ANC/1:4 621.0 197.2 0.395 0.182 0.213 54 P-ANC/2:1 40.3 30.6 P-ANC/1:1 5.2 4.9 AP-ANC/2:1 492.6 95.7 0.266 0.173 0.093 35 AP-ANC/1:1 602.7 155.0 0.343 0.194 0.149 43 AP-ANC/1:2 1013.0 668.1 0.706 0.152 0.554 78 AP-ANC/1:3 1134.3 665.9 0.789 0.220 0.569 72 * by difference, SBET BET surface area, Sme mesopore surface area, Vtot total volume, Vmi micropore volume, Vme mesopore volume, d average pore diameter, Vme/Vtot mesopore volume ratio

d [nm] 2.11 2.09 2.05 2.32 2.54 2.16 2.28 2.58 2.77