Abstract

A series of zeolite X/activated carbon (AC) composites were prepared from the same starting materials at various activation time. The corresponding modified samples were obtained by being treated with diluted NH4Cl solution. The relationship between porosity development, surface properties, and CH4/N2 adsorption performance was investigated. The increase of micropore volume is beneficial to the improvement of CH4 and N2 adsorption capacity, but more sensitive for CH4. In addition, the polar functional groups of zeolite X/AC composites may enhance CH4 adsorption capacity. More importantly, both developing micropore structure and surface modification contributed to enhance the adsorption selectivity . As the optimum sample of these studies, HZAC(24) showed CH4 adsorption capacity of 17.3 cm3/g and the highest adsorption selectivity of 3.4. The CH4 and N2 adsorption isotherms of all samples can be well fitted by the Langmuir–Freundlich model. HZAC(24) showed an excellent cyclability of adsorption/desorption of CH4 with a neglectable capacity loss after subsequent cycles. Moreover, HZAC(24) displayed relatively rapid adsorption kinetics. These properties of zeolite X/AC composites are essential for the adsorptive separation of CH4 from N2 in the pressure swing adsorption (PSA) process.

1. Introduction

The coal bed methane (CBM) is an unconventional gas with a main composition of CH4, N2, and CO2, which reserve is about two times higher than that known to natural gas [1]. Nowadays, due to the CH4 content in the drainage gas of coal mine is only 20–45%, CBM is usually extracted into the atmosphere, which not only is a waste of energy source but also pollutes the environment as CH4 is one of the major contributors to the global warming with 20 times higher global warming potential than that of CO2 [2, 3]. So, it is of great significance to develop and utilize the CBM. The content of CH4 higher than 80% is required for the application of chemical raw stock and 90% for merging into the civil gas system [4]. For pipeline quality natural gas, the impurities of N2 and CO2 content should not exceed 4% and 2%, respectively [5, 6]. So, CH4 separation from N2 and CO2 is one of the important industrial separation processes [7, 8]. However, CH4 and N2 possess extremely similar physicochemical properties and kinetic diameter. Therefore, it is a really large challenge to enrich CH4 from the mixture of CH4 and N2 such as CBM.

Generally, membrane [9], cryogenic [10, 11], and adsorption separation [12] are applied to separate the CH4 and N2 mixture. Membrane separation has the drawbacks of low selectivity and the strongly dependence of the membrane which is liable to damage and block; thus, it is not economical in scale separation [13, 14]. Cryogenic separation requires high-energy consumption and is helpless for low flow rates. In comparison with other methods, adsorption separation with easy operation, lower energy requirement, lower operational cost, running continuous at ambient temperature, and so on attracts increasingly attention. During the past three decades, there has been a rapid growth in the development of adsorption-based technologies for separation and purification of different gas mixtures, so that it can be applied to medium-scale CH4/N2 separation. However, it is still a big challenge as for the larger-scale CH4/N2 adsorption separation due to the lack of satisfactory adsorbent with high adsorption capacity and selectivity.

Many materials have been developed for gas selective separation. Zeolites, as one of the candidates for enriching CH4 from gas mixture, have shown great prospect and can potentially be used in the pressure swing adsorption (PSA) process. In order to obtain an appropriate zeolite for gas adsorption separation, the effect factors, including the pore structure of zeolite and the strength of the electric fields caused by the presence of exchangeable cations in the frameworks, have been investigated [5, 14]. It has been found that zeolite X [1521] is one of the most suitable zeolite adsorbents for adsorption and separation due to its large pore diameter of 0.74 nm which can accommodate large molecules and low Si/Al ratio with the presence of extra framework of cations that produces electric field which interacts strongly with the high polarizability CH4 molecule or quadrupolar CO2 molecules. Activated carbon [2224], another promising candidate for gas separation, possesses lots of advantages such as tunable pore size, easy regeneration, and low cost. In general, activated carbons have higher equilibrium selectivity for CH4 over N2 but smaller adsorption capacities for CH4 than zeolites. As a consequence, in order to integrate the advantages of both zeolite and activated carbon in industrial application, increasingly attentions have been concentrated on the synthesis of the novel porous material of zeolite/AC composites in recent years [25, 26]. Meanwhile, the environmental application such as wastewater treatment as well as gas separation using of zeolite/AC composites has been explored preliminarily [27, 28]. In our previous work, the zeolite X/AC composites from elutrilithe are prepared by adding pitch powder and precipitate silicon dioxide as an additional carbonaceous and silica source, respectively [25, 29, 30].

The main aim in this work is to introduce amine modifications in a series of zeolite X/AC composites with different activation time in order to increase the interactions with CH4 without improving those with gases. This strategy will improve the potential of these materials to separate CH4/N2 mixtures, making these materials candidates for natural gas upgrading. The effect of pore texture and surface properties of the adsorbents on the adsorption performance of CH4 and N2 was investigated in detail.

2. Experimental

2.1. Preparation of Zeolite X/AC Composites

Zeolite X/AC composites were prepared by the following two steps. First, the locally available Elutrilithe chunk, with major chemical composition of 41.0 wt.% SiO2, 35.5 wt.% Al2O3, and 7.0 wt.% C, was crashed and sieved in order to collect the grains with an average size below 200 meshes. The elutrilithe powder was then simply physically mixed with precipitated silica (ca.93% SiO2) and pitch powder in order to get a constant molecular ratio SiO2/Al2O3 = 4.5 and a final mass composition of 50 wt.% carbon. The obtained blend powder was kneaded with defined amount of deionized water and extruded into a cylinder shape with dimensions of φ2.0 mm × 6.0 mm by extruder. After drying overnight, the extrudates were calcined in a tubular furnace under a stream of N2 with a flow rate of 150 ml/min at 723 K for 2 h and following 1123 K for 2 h. After that, the temperature of 1123 K was maintained for time ta, and the atmosphere was switched from N2 to CO2 (200 ml/min) for activation. ta is a variable activation parameter, ranging from 4 h to 32 h. A heating rate of 5 K/min was applied up to the appointed temperature in the above procedure. Subsequently, the activated samples were treated with a NaOH hydrothermal system in a flash (Al2O3 : SiO2 : Na2O : H2O = 1 : 4.5 : 4.5 : 135), and then, the products were filtered, washed with deionized water at 323–343 K until the pH of the products reached around 7, and dried at 373 K for at least 6 h. ZAC(ta) (ta = 4, 16, 24, and 32, which represent the activation time of 4 h, 16 h, 24 h, and 32 h) is used to mark the composites, and the corresponding burn-off of ZAC(ta) is 42, 29, 21, and 17%, respectively.

The surface modification of the composites was carried out by the conventional soaking method. A typical soaking method involved adding 250 mL of 0.3 mol/L NH4Cl solution to about 10 g composite. The solution was heated to 340 K for 0.5 h. After that, the solution was decanted, and fresh solution was added. This procedure was repeated three times. After modification, the samples were filtered and washed with copious amount of deionized water until the effluent was free from chloride ions as tested with AgNO3 solution. The resulting products were dried at 373 K for more than 6 h and calcined in a tube furnace at 623 K for 2 h under vacuum to dislodge the NH3 in the products. The corresponding products after modification were denoted as HZAC(ta) (ta = 4, 16, 24, and 32, which represent the activation time of 4 h, 16 h, 24 h, and 32 h).

2.2. Characterization

X-ray powder diffraction (XRD) patterns of the materials were obtained using a Shimadzu LabX XRD-6000 system in the 2θ range of 5–35° using CuKα1 (λ = 1.54056 Å) radiation operated with 40 kV and 30 mA. The surface area and pore volume of the products were calculated from N2 adsorption isotherms measured on a surface area and pore size analyzer, QUADRASORB SI, (Quantachrome Inc., USA) after degassing the samples at 623 K for at least 4 h under high vacuum. The surface areas (SBET) were calculated by using the Brunauer–Emmett–Teller (BET) equation at the relative pressure in the range of 0.01–0.2. The micropore volume (Vmic) and micropore surface (Smic) area were determined by the t-plot method. The amount of N2 absorbed at P/P0 = 0.98 was employed to calculate the total pore volume (Vtotal). The total basicity of the samples was determined by temperature-programmed desorption of carbon dioxide (CO2-TPD) using a type of TP-5076 TPR/TPD automatic temperature-programmed chemical adsorption apparatus. The surface properties of the samples were analyzed by employing a conventional Boehm titration procedure. Aqueous solutions of NaOH, HCl, Na2CO3, and NaHCO3 (0.1 mol/L) with the volume of 25 mL were decanted into conical flasks with 1 g sample each, respectively, and after shaking for 1 h, the conical flasks were kept standing for 48 h, and then filtered. The filtrates were diluted and employed back titration with standard solution to calculate the amount of surface function groups on the samples.

2.3. Adsorption Measurements

The presence of water in the composites significantly affects the adsorption isotherm; therefore, all the samples were calcined at 573 K for 4 h under N2-flow in a tubular furnace. Prior to adsorption measurements, the samples were degassed under vacuum by heating up to 623 K for at least 4 h. The CH4 and N2 adsorption isotherms were measured at 273 K using a static volumetric apparatus of NOVE1200e instrument (Quantachrome Inc., USA). The adsorption kinetics curve was measured by releasing a small amount of CO2 with the type NOVE1000e instrument (Quantachrome Inc., USA). During the adsorption, measurement temperature was maintained by circulating ethanediol-water from a constant temperature bath. The adsorption capacity was determined from the adsorption isotherm measured at 273 K, and the pure component selectivity of CH4 and N2 () was calculated by using the following equation:where and are the adsorption amount (at 273 K, 100 kPa) of CH4 and N2, respectively.

3. Results and Discussion

3.1. Zeolite X/AC Composites

The XRD patterns of zeolite X/AC composites before and after modification are shown in Figure 1. It can be seen that all samples are highly crystalline materials giving the reflections in the range of 5–35° and exhibit characteristic peaks of zeolite X without other impurity phases. The crystallinity of zeolite X decreases after surface modification as evidenced from the decrease of the intensity of the major characteristic peaks of zeolite X. This is due to the partially collapse of the crystalline structure, which is consistent with the conclusion from others [31, 32].

Figure 2 displays N2 adsorption-desorption isotherms at 77 K on a series of samples. All isotherms exhibit the characteristic combination of the type-I and type-IV morphology, indicating the coexistence of micropores and mesopores in the composites [33]. Table 1 summarizes their pore structure parameters calculated from the isotherms. It can be seen that, as far as ta ≤ 24 h, the SBET, Smic, Vtotal, and Vmic of ZAC(ta) samples presented a significant rising trend, and it occurred sharply decrease thereafter, while the Sext was increasing with lengthening the activated time. This means that extending the activation time when ta ≤ 24 h mainly induces the creation of new pores, while widening of pores becomes the main mechanism of porosity development with longer activation time [25]. In addition, the treatment of NH4Cl solution has a significant effect on the textual properties. In comparison with ZAC(ta), the pore structure parameters of HZAC(ta) decrease except Sext, due to the partially collapse of the crystalline structure in the composites during surface modification [32].

The CO2-TPD results of ZAC(ta) and HZAC(ta) are shown in Figure 3. For ZAC(ta) samples, a larger desorption peak at 573–673 K and a little shoulder peak take place at around 383 K, corresponding to the surface basic site; however, there is only one desorption peak of HZAC(ta) samples at about 473–573 K. From Figure 3, it can be seen that CO2 desorption peak areas increase and CO2 desorption peak moves to higher temperature for ZAC(ta) samples when increasing activation time, indicating that the amount and strength of surface basic groups rising, respectively. The variation tendency of the surface basic properties of HZAC(ta) is similar with ZAC(ta). However, both the amount of the surface basic groups and basic strength decline for HZAC(ta) compared with ZAC(ta), which resulted from the treatment by NH4Cl solution. In addition, Figure 4 shows the representative Boehm titration results of ZAC(24) and HZAC(24), which shows that the basic groups decrease almost half, and phenolic, carboxylic, and the total acidic groups sharply increase. These results are similar to the reports from references [3436].

3.2. CH4 and N2 Adsorption Isotherms

The pure component CH4 and N2 adsorption isotherms at 273 K on the samples of ZAC(ta) and HZAC(ta) have been generated and are given in Figure 5. All curves are found to be type-I as per the IUPAC classification, and the low-pressure adsorption isotherms behave linearly [14, 33], which is ideal for sorbent regeneration, since most of the working capacity is recovered at moderate pressure ranges. A summary of adsorption and separation parameters extracted from adsorption isotherms is presented in Table 1. It can be seen that the CH4 adsorption capacity is higher than that of N2 on all of the samples. The CH4 and N2 adsorption capacity and CH4/N2 adsorption selectivity increased with the increase of Vmic. After the composites being treated by NH4Cl, both of the CH4 and N2 adsorption capacity on the composites decreased; however, the CH4/N2 adsorption selectivity increased significantly.

Equilibrium adsorption isotherms in adsorbents are required for the available modeling or simulation of the various PSA separation processes of CH4 and N2 in practical application. Hence, in order to extend its utility, the collected experimental adsorption isotherm data should be fitted to an isotherm model [37]. In the present work, the adsorption isotherms of CH4 and N2 on all samples are fitted using the conventional Langmuir–Freundlich (L-F) (equation (2)) adsorption model [38, 39]:where is the pressure, is the corresponding amount of the adsorbate adsorbed, is the saturation capacity of the adsorbent, is the adsorption affinity defined by the van’t Hoff equation for a heterogeneous solid, and is the reciprocal of the Freundlich heterogeneity factor (the larger the value, the more the deviation from the Langmuir isotherm). Both and are empirical constants to be obtained by the data fitting method. Fitted parameters of the CH4 and N2 adsorption isotherms for all samples with L-F models are listed in Table 2. It can be seen that the Langmuir–Freundlich model is well correlated with the experimental adsorption data over the entire pressure range studied. Obviously, the values of of CH4 adsorption are higher than that of N2 for ZAC(ta), suggesting the stronger interaction between CH4 and the samples than that between N2 and the composites. After NH4Cl treatment, the value of decreases for N2 adsorption, due to a decrease in the interaction between N2 and modified composites. However, increases for CH4 adsorption, and the higher value of for HZAC(ta) is due to an increase in the interaction between CH4 and HZAC(ta). The changes in the values of both CH4 and N2 adsorptions explain well the increase of the CH4/N2 selectivity on HZAC(ta).

3.3. Effect of Pore Structure and Surface Properties on the Adsorption Performance

In order to improve the CH4/N2 adsorption selectivity and maintain the adsorption capacity of CH4, the relationship between CH4 and N2 adsorption properties and the pore texture were studied in detail. ZAC(32) possesses a lower Vmic and a larger Sext than ZAC(24), and the adsorption capacity and selectivity of CH4 and N2 showed the lower values on ZAC(32), indicating that the high adsorption capacity and selectivity of CH4 and N2 results from the large micropore volume rather than Sext [4045]. In Figure 6, the adsorption capacity of CH4 and N2 and adsorption selectivity are plotted versus Vmic. For ZAC(ta), as the Vmic increased from 0.275 cm3/g to 0.357 cm3/g, the adsorption capacity of CH4 and N2 increased from 16.1 cm3/g to 22.6 cm3/g and 7.52 cm3/g to 9.39 cm3/g, respectively. When the Vmic increases from 0.196 cm3/g to 0.273 cm3/g for HZAC(ta), the CH4 adsorption capacity increases from 14.4 cm3/g to 17.3 cm3/g. The adsorption selectivity for ZAC(ta) and HZAC(ta) presents a slight rise tendency. As a summary, with the increase of Vmic, CH4 and N2 adsorption capacity increased, and adsorption selectivity enhances slightly at the same time. However, the CH4 and N2 adsorption properties of ZAC(4) are different from HZAC(24) although the values of Vmic for the two samples are similar, which suggests that the pore structure is not the only factor affecting the gases adsorption performance on the samples.

Besides pore structure, the surface properties are also considered to be a significant factor on the gas adsorption process of the studied samples. After the treatment by NH4Cl, the basic groups decreased severely and the acidic groups including phenolic and carboxylic groups sharply increased. Although CH4 is a nonpolar molecule, it has a higher polarizability of 26 × 10−25 cm3 [46]. Such polarizability causes a momentary shift in the time when the neutral electrostatic field of CH4 is in close proximity to the larger polar function groups such as phenolic and carboxylic groups within the composites. The polarizability of N2 (17.6 × 10−25 cm3) is much lower than that of CH4. Hence, the increasing of polar functional groups is more conducive to enhance the interaction between CH4 molecule and the adsorbents [32, 39, 40, 47]. The reduction of the CH4 adsorption capacity resulted from the decrease of Vmic in HZAC(ta) was offset partially by the increase in interaction strongly between CH4 molecule and HZAC(ta). These characteristics lead to a moderate decrease in CH4 adsorption capacity and a drastic decrease in N2 adsorption capacity after NH4Cl treatment; therefore, the adsorption selectivity on the HZAC(ta) is extremely higher than that on the ZAC(ta). In conclusion, there is a positive correlation between CH4 adsorption capacity and adsorption selectivity and the Vmic. When the Vmic of samples is similar, the CH4 adsorption capacity and the adsorption selectivity depended on the surface properties of the samples. HZAC(24) with a considerable Vmic and a higher amount of acidic groups has satisfactory CH4 adsorption capacity and the maximum adsorption selectivity among the studied samples.

3.4. Cycle Adsorption and Time-Dependent Adsorption Capacity of CH4

As adsorption-driven gas separation processes depend on the cyclic regeneration of the adsorbent, it is important to study the cyclic adsorption performance of the adsorbents during long-term cyclical operation [48]. The successive adsorption/desorption cycles on HZAC(24) were carried out for nine times, and the results are shown in Figure 7. Within each cycle, the HZAC(24) was regenerated by reducing the pressure of CH4 along the desorption branch and finally reached the conditions of a dynamic vacuum. The maximum CH4 adsorbed amount was almost constant in nine cycles of adsorption and desorption, indicating that the CH4 adsorption performance of HZAC(24) is stable. The results suggest that the zeolite X/AC composite can be easily regenerated by vacuum in the CH4 adsorption system and be reused for many circles without any decay for the adsorption capacity.

For potential applications of zeolite X/AC composite in adsorption-driven separation of CH4 from CH4/N2 mixture, it is also crucial that the adsorption of CH4 is rapid. Typically, the industrial adsorption process is shorter than 1 min [49]. Here, we studied the time-dependent adsorption of CH4 on HZAC(24) by releasing a small amount of CH4 and studying the adsorbed amount as a function of time as shown in Figure 8. HZAC(24) displayed a relatively rapid adsorption kinetics, reaching 96% of the CH4 capacity within 40 s. The rapid uptake indicates that zeolite X/AC composite can meet the requirements for an adsorbent used in the PSA process for the separation of CH4.

4. Conclusions

Zeolite X/AC composites with different pore textures were prepared and then treated by diluted NH4Cl solution. With the increase of activation time, the micropore volume increased, which reaches the maximum at the activation time of 24 hours. Furthermore, the increase of the micropore volume is beneficial to enhance the adsorption capacity of CH4 and the adsorption selectivity . The increase of acidic oxygen-containing functional groups on sample surface significantly contributed to enhancing the adsorption selectivity , but leading to the decrease of the adsorption capacity of CH4. The HZAC(24) sample showed the highest adsorption selectivity of 3.4 among the studied samples and a high CH4 adsorption capacity of 17.3 cm3/g. The isotherms of CH4 and N2 on the composites before and after NH4Cl treatment can be well fitted by the Langmuir–Freundlich model. Moreover, the HZAC(24) showed excellent cyclability of adsorption/desorption of CH4 and relatively rapid adsorption kinetics. The present work provides a promising sight of applying zeolite X/AC composites from an economic and simple synthesis route as CH4 adsorbent with high adsorption selectivity .

Data Availability

The data used to support the findings of this study are included within the article.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Acknowledgments

We gratefully appreciate the financial support from the National Natural Science Foundation of China (No. 51204120), Natural Science Foundation of Shanxi (No. 2014021014-1), and Key Scientific and Technological Project of Coal Fund of Shanxi province (No. FT201402-03).