Methane Adsorption Study Using Activated Carbon Fiber and Coal Based Activated C

Guo Deyong; Li Fei; Liu Wenge

(1. School of Resource and Safety Engineering, China Uniνersity of Mining and Technology(Beijing), Beijing 100083; 2. China Coal Information Institute, Beijing 100029)

Methane Adsorption Study Using Activated Carbon Fiber and Coal Based Activated Carbon

Guo Deyong1; Li Fei1; Liu Wenge2

(1. School of Resource and Safety Engineering, China Uniνersity of Mining and Technology(Beijing), Beijing 100083; 2. China Coal Information Institute, Beijing 100029)

Influence of ammonium salt treatment and alkali treatment of the coal based activated carbon (AC) and activated carbon fiber (ACF) adsorbents on methane adsorption capacity was studied via high-pressure adsorption experiment. Surface functional groups and pore structure of two types of adsorbents were characterized by the application of infrared absorption spectroscopy (IR) and low temperature liquid nitrogen adsorption method. The results show that both ammonium salt treatment and alkali treatment have obvious effect on changing BET, pore volume as well as pore size distribution of adsorbents; and methane adsorption capacity of the activated carbon fiber is the maximum after the ammonium salt treatment.

methane; adsorbents; ammonium salt and alkali treatment; microstructure; high-pressure adsorption

1 Introduction

Methane is the main component of coal mine gas. Direct emission of large amount of low concentration CH4to the atmosphere not only seriously influences the environment, but also wastes huge energy. Therefore, it is very important to effectively utilize low concentration methane[1]. In recent years, new progress has been made in the study of CH4refinement and concentration technique both at home and abroad[2-3]. CH4/N2separating development technologies mainly include cryogenic separation at low temperature, pressure swing adsorption and membrane separation technique[4], among which the pressure swing adsorption (PSA) has successfully realized hydrogen recovery and oxygen and nitrogen separation[5]thanks to its advantages of strong applicability, flexible operation, low cost, etc. The core of PSA is the selection of adsorbent, while the adsorptive property of the adsorbent is mainly determined by its microstructure. This paper discusses the microstructure features of the adsorbent including surface functional group and micropore structure, and studies the influence of ammonium salt treatment and alkali treatment of adsorbents on methane adsorbing capacity via high-pressure adsorption experiment.

2 Experimental

2.1 Materials

The selected experimental materials and chemical reagents are shown in Table 1.

Table 1 Materials and chemical reagents used in experiments

2.2 Preparation of experimental sample

Six experimental samples were prepared through ammonium salt treatment and alkali treatment of the original experimental samples, with the specific conditions depicted in Table 2.

2.3 Characterization of adsorbent property

Nicolet iS10 Fourier transform infrared spectrometer wasadopted to determine the surface functional group of the adsorbents. The scan range of the instrument covered from 400 cm-1to 4 000 cm-1, with a resolution ratio of 4.00 cm-1and a scanning number of 32.

Table 2 The sample number and treatment methods

Specific surface area and pore structure parameters of the adsorbents were obtained by measuring N2(77K) adsorption isotherms under a relative pressure of 10-6—1 MPa on the ASAP2020 specific area and porosity analysis meter manufactured by the Micromeritics Instrument Corp. of U.S.A. The specific area of adsorbents was calculated by means of the standard BET equation, and specifically the pore volume and pore size distribution of adsorbents were determined by the DFT method[6-9].

2.4 High-pressure adsorption experiment

High-pressure adsorption device developed by the Key Laboratory of High-Pressure Adsorption of the School of Chemical Engineering and Technology (SCET) at Tianjin University was utilized to measure the methane adsorption isotherms of adsorbent samples. The adsorbent was firstly placed in the electro-thermal blowing dry box, and heated for 4 hours at 120 ℃ to remove large amount of water in the samples, and then dried for 24 hours at 120 ℃ in a vacuum oven to obtain final samples. Finally, the adsorbent samples were placed in the adsorption slot of the highpressure adsorption device after being weighed.

3 Results and Discussion

3.1 Functional group characterization of adsorbent

The oxygen-containing functional groups[10-11]on activated carbon surface mainly include carboxyl group, phenolic hydroxyl group, lactone group or carbonyl group, which are all recognized as acidic oxygen-containing functional groups and alkaline oxygen-containing functional groups. The acidic oxygen-containing functional group on the surface of the activated carbon provides the activated carbon with polarity that is beneficial to adsorption of compounds with strong polarity. As methane is a non-polar molecule, the acidity of oxygen-containing functional group on the adsorbent surface should be lowered according to the similitude theory. Therefore, in this paper, the two adsorbents, viz. the coal based activated carbon (AC) and the activated carbon fiber (ACF), were treated by ammonium salt and alkali, which would be conductive to enhancing their ability to adsorb methane.

Figure 1 FTIR spectra of AC samples

As shown in Figure 1, the samples have obvious spectral peaks within the range of 3 300—3 500 cm-1, which explains the stretching vibration peak of phenolic hydroxyl group, and the AC2 peak intensity is sharper and more steep than that of AC1 and AC3, which shows that the corresponding phenolic hydroxyl group vibration is more violent and has a higher content. The three kinds of activated carbon have a stretching vibration peak of lactone group C=O at 1 632 cm-1, at which the high AC2 intensity means a larger content of the lactone group. It can be known from the comprehensive analysis that the most obvious change is the increased intensity of surface oxygencontaining functional group of AC2.

Figure 2 FTIR spectra of ACF samples

It can be seen from Figure 2 that the shapes and the peak shapes of the three activated carbon fibers in the infrared spectral analysis are similar to each other, which explains that no obvious change occurs to the group types contained in the samples after the ammonium salt and alkali treatment. It can be observed that the absorption peak of ACF3 is longer and sharper at 3 460 cm-1, which indicates that the phenolic hydroxyl group vibration of ACF3 is more violent and more frequent, while the absorption peak intensities of ACF1 and ACF2 are almost the same at this point, which explains that the contents of the phenolic hydroxyl groups are nearly identical with each other. Therefore, the oxygen-containing functional group of the activated carbon fiber (ACF3) after the alkali treatment increases obviously.

3.2 Pore structural characterization of adsorbent

The adsorption and desorption isotherms of three kinds of coal-based activated carbon samples are shown in Figure 3. According to adsorption isotherm types classified by IUPAC, this kind of adsorption isotherm is Type IV. Whenp/p0is low, the adsorption capacity rapidly increases with the increase of pressure, because when the relative pressure is small, the absorption ability of the adsorbent is mainly originated from micropores, and the superposition of its pore wall force fields can strengthen the adsorption potential, leading to strong adsorption capacity. With the increase ofp/p0,the adsorption capacities of the samples almost increase linearly, which indicates that a certain amount of mesopores and large pores are contained in the sample structures, which produces capillary condensation phenomenon and shows tailing on the isotherm.

Figure 3 Adsorption/desorption isotherms of AC samples

The adsorption isotherms of all the three kinds of adsorbents show similar trend, but there are certain differences in their adsorption capacities. Compared with AC1, the capacity for adsorption of nitrogen by the samples (AC2 and AC3) are reduced after ammonium salt and alkali treatment, among which AC3 is the most obvious one, which explains that the ammonium salt and alkali treatment has changed the pore structure of the adsorbent to some extent. As shown in Figure 4, adsorption isotherms of ACF1 and ACF3 is a characteristic Type I adsorption isotherm, which shows that this kind of adsorbent belongs to microporous adsorbent[12]. When the relative pressure is small, the adsorption capacities of ACF1 and ACF3 show a rapid increase tendency. When the relative pressure exceeds 0.2, the adsorption isotherm gradually tends to be smooth, and finally becomes a horizontal line, which indicates that the adsorption capacity is saturated. This kind of phenomenon fully explains that the micropores of the two kinds of adsorbents are concentrated in distribution without mesopores and large pores on the whole.

Figure 4 Adsorption/desorption isotherms of ACF samples

It is clear that ACF2 adsorption isotherm is a typical Type IV adsorption isotherm. With the increase of the relative pressure, the adsorption capacity increases rapidly which means that a certain amount of mesopores and large pores appear in the adsorbent structure after the alkali treatment. These pores enable the capillary condensation phenomenon to take place during the adsorption process, and it can be understood that the capillary condensation effect of mesopores produces the filling phenomenon after the completion of filling of micropores.

The specific area and the total pore volume of the samples can be directly obtained by the ASAP2020 analysis software. The results are shown in Table 3.

Table 3 Pore structure of AC and ACF samples

Compared with the original sample (AC1), the microporous distribution peaks of the activated carbon samples (AC2 and AC3) changed after the ammonium salt and alkali treatment. The microporous pore volumes at pore diameters of 0.8 nm and 1.26 nm increased obviously, and micropores with a diameter of about 0.4 nm appeared in AC3 due to activation, but its total pore volume has decreased, indicating to the change of AC3 adsorption volume at different pore diameters (Figure 5).

Figure 5 Microporous distribution of AC samples

Compared with ACF1, the distribution of microporous pore diameter of ACF3 does not change on the whole; the micropore diameters are concentrated within the range of 1.5—2.0 nm, and the microporous pore volume of ACF3 with a pore diameter of above 1.6 nm increases slightly. But the microporous structure of ACF2 changes greatly, and the pore volume fluctuates when the diameter of micropores is less than 1.5 nm, with the strong peaks of pore volume (Figure 6) identified at micropore diameter of 0.53 nm and 0.80 nm, respectively, which means that the immersion treatment of ammonium salt has produced corrosion of the activated carbon fiber to certain degree. Large amount of uniform distribution of “active points” are formed on the whole surface of the fiber and its internal part, resulting in changes of the pore diameter distribution of the adsorbent.

Figure 6 Microporous distribution of ACF samples

3.3 High-pressure adsorption experiment

Figure 7 CH4adsorption isotherms of AC samples

The methane adsorption isotherms of the three kinds of coal-based activated carbon are almost the same; all of them belong to simple convex Type I adsorption isotherms. At the lower pressure section, the methane adsorption capacity increases rapidly; and with an increasing pressure, the adsorption curve gradually becomes smooth (Figure 7). Compared with AC1, the activated carbonafter the ammonium salt and alkali treatment reduces the adsorption capacity of methane, especially for the case of activated carbon after alkali treatment. The adsorption capacities of AC1 and AC2 are similar at the pressure of 5 MPa; when the pressure rises to 6 MPa, the adsorption capacity of AC2 exceeds that of AC1. This phenomenon indicates that with a continuous increase of the pressure, the adsorption capacity of AC2 will exceed that of AC1. The study shows[13-14]that when the pore diameter of the activated carbon is less than 0.7 nm, the adsorption potential energy of methane is substantially higher than that of nitrogen. The results of high-pressure adsorption experiment indicate that the methane adsorption capacity of all the three kinds of activated carbon is much higher than the capacity for adsorption of nitrogen. Based on the analysis of activated carbon pores distribution (Figure 5), it can be seen that the specific areas of the activated carbon (AC2 and AC3) after ammonium salt and alkali treatment increase to different degrees, and the micropore ratio contained in AC2 also increases, but the methane adsorption capacity of AC2 decreases. This phenomenon explains that the increase of specific area and micropore ratio of the adsorbent has little impact on the increase of methane adsorption capacity.

The tendencies of the methane adsorption isotherms of the activated carbon fiber and the coal-based activated carbon are the same, and both of them gradually increase and finally tend to be smooth with an increasing pressure. The difference is that methane adsorption capacity of the three kinds of activated carbon fibers (ACF) is about two times greater than that of activated carbon (AC). It means that compared with the activated carbon, the activated carbon fiber is more favorable for the adsorption of methane. As shown in Figure 8, the methane adsorption capacity of three activated carbon fiber samples increases in the following order: ACF3＜ACF1＜ACF2. The adsorption capacity of the activated carbon fiber after ammonium salt treatment (ACF2) has increased to certain degree greater than that of the original sample (ACF1), while the adsorption capacity of the activated carbon fiber after alkali treatment (ACF3) has decreased slightly.

It can be seen from Figure 6 that the ammonium salt and alkali treatment can result in a significant increase of specific area of the activated carbon fiber; meanwhile, the total pore volume and the micropore ratio also change greatly, wherein the total pore volume of ACF2 increases greatly; and the micropores under 1.5 nm also appear. The comprehensive analysis shows that ACF2 has the highest methane adsorption capacity, because the proper pore size distribution of ACF2 contributes greatly to improvement of the methane adsorption capacity, which shows that the methane adsorption capacity is mainly dependent on the distribution of the suitable pore sizes of the adsorbent, while the specific area and the micropore content have little impact on it.

Figure 8 CH4adsorption isotherms of ACF

4 Conclusions

(1) The Fourier infrared spectrum experiments have revealed that the amount of oxygen-containing functional group on the surface of both coal-based activated carbon and the activated carbon fiber after ammonium salt and alkali treatment increases somewhat; but compared with the adsorbent pore structure, the oxygen-containing functional group has less influence on the methane adsorption capacity.

(2) Nitrogen adsorption experiment indicates that the adsorption isotherms of AC1, AC2, AC3 and ACF2 belong to Type IV, which means that there are a certain number of mesopores and large pores in the absorbents. The adsorption isotherms of ACF1 and ACF3 belong to Type I, which means that the adsorbents are dominated by micropores and contain few mesopores. The ammonium salt and alkali treatment of adsorbent samples have obvious effect on increasing the specific area of the adsorbent.

(3) The methane isothermal adsorption curves obtained in high-pressure adsorption experiments have shown that they are simple convex Type I adsorption isotherms.

With an increasing pressure, the adsorption capacity of adsorbent samples gradually increases and finally reaches a smooth value, wherein the methane adsorption capacity of ACF2 is the largest, because the pore size distribution of ACF2 is favorable for adsorption of methane. It can be ascertained that the methane adsorption capacity is mainly dependent on the pore size distribution of the adsorbents and the ratio of the suitable pore sizes.

Acknowledgement:The project is financially supported by the National Natural Science Foundation of China (Grant No. 41072118) and the Foundation for Key Program of Ministry of Education, China (Grant No. 311022).

[1] Yu B F. Technical Manuals of Disaster Prevention and Application of Coal Mine Gas[M]. Beijing: Coal Industry Press, 2005 (in Chinese)

[2] Grande, Carlos A. Carbon molecular sieves for hydrocarbon separations by adsorption [J]. Industrial and Engineering Chemistry Research, 2005, 44(18): 7218-7227

[3] Liu K W, Gu M, Xian X F. Research progress in concentration of methane from CH4/ N2by PSA[J]. Modern Chemical Industry, 2007, 27(12): 15-20 (in Chinese)

[4] Guo Pu, Li Ming. Research progress of separation of CH4/ N2in coal-bed methane [J]. Chemical Industry and Engineering Progress, 2008, 27(7): 963-965 (in Chinese)

[5] Olajossy A, Gawdzik A, Budner Z, et al. Methane separation from coal mine methane gas by vacuum pressure swing adsorption [J]. Chemical Engineering Research and Design, 2003, 81(4): 474-482

[6] Kruk M, Jaroniec M, Bereznitski Y. Adsorption study of porous structure development in carbon blacks[J]. J Colloid Interface Sci, 1996, 182(1): 282-288

[7] Ryu Z, Zheng J, Wang M. Porous structure of PAN-based activated carbon fibers[J]. Carbon, 1998, 36(4): 427-432

[8] Oliver J P, Conklin W B, Szombathely M V. Characterization of Porous Solid III [M]//Ronquerd J, Rodregaez-Keiso F, Sing K S W, et al. Studies in Surface Science and Catalysis, 1994, Volume 87: 129-139

[9] Barrret E P, Joyner L G, Halenda P H. The determination of pore volume and area distribution in porous substances[J]. J Am Chem Soc, 1951, 73(1): 373-380

[10] Jacques, Lahaye. The chemistry of carbon surfaces [J]. Fuel, 1998, 77(6): 543-547

[11] Wan Mold Ashri Wan Daud, Amir Hossein Houshamnd. Textural characteristics, surface chemistry and oxidation of activated carbon [J]. Journal of Natural Gas Chemistry, 2010, 19(3): 267-279

[12] Greg S J. Pressure Swing Adsorption [M]. 2nd edition. London: Academic Press, 1992

[13] Cui X J, Bustin R M, Dipple G. Selective transport of CO2, CH4and N2in coals insights from modeling of test gas adsorption data [J]. Fuel, 2004, 83(3): 293-303

[14] Wang F Y, Zhu Z H, Massarotto P, et al. Mass transfer in coal seams for CO2sequestration [J]. AIChE J, 2007, 53(4): 1028-1049

Recieved date: 2013-07-12; Accepted date: 2013-10-15.

Professor Dr. Guo Deyong, Telephone: +86-10-62331517; E-mail: kjkfg@cumtb.edu.cn.