Abstract
A series of open-tip carbon nanotubes (CNTs) were obtained by HNO3 modification with various concentrations of as-prepared carbon nanotubes via the CVD method, and this work aimed at investigating the structural features of open-tip CNTs for the methane capability. Modified CNTs had higher specific surface area and larger total pore volume, and importantly, greater micropore volume was obtained through HNO3 modification of the as-prepared CNTs. The remarkably high methane adsorption capacities were measured on the modified CNTs under pressure ranges of 0∼4.0 MPa at 298 K. The resulted H-CNT, which exhibited highest specific surface area and micropore volume, showed high methane uptake of 26.15 mg/g from the D-A model. This value was nearly as twice as the methane uptake of original CNTs (13.62 mg/g), along with an initial adsorption heat of 19.4 kJ/mol at lower coverage and 9.5 kJ/mol at higher methane coverage for H-CNT, indicating the physical nature for methane adsorption over open-tip CNTs.
1. Introduction
Methane (CH4) produced by coal seam and shale is widely regarded as one of the main greenhouse gases contributing to global warming, which is one of the most significant challenges today [1–3]. The ongoing use of fossil fuels makes it compelling to store CH4 via the adsorption technology. In recent decades, there has been an increasing interest to develop CH4 storage systems for environmental purposes and energy applications [4, 5]. It is therefore highly desirable to develop adsorption materials with moderate adsorption capacity and good stability [6–10].
Structurally carbon materials, depending on their properties, can be used in CH4 storage, due to their high specific surface area and pore volume that is able to host large amounts of CH4 [2, 3, 11, 12]. More importantly, the textural property and surface chemistry of carbon materials for effectively increasing the amounts of stored gas can be appropriately tailored. Among the diverse adsorbents, CNTs are considered to be one of the most promising candidates for CH4 storage due to their low cost, high surface area, excellent thermal stability, and high amenability to pore structure tailoring [13–15]. However, CH4 adsorption on the as-prepared CNTs is generally a weak physisorption process, which means that the uptake capacity can be low. A higher adsorption capacity in CH4 storage of the open-tip CNTs than the as-prepared ones was obtained due to the larger specific surface area. Hence, it is necessary to open the tips of the as-grown CNTs before application. Generally, there are several approaches to achieve the purpose, such as strong acid oxidation and plasma treatment [16, 17]. There has been currently intensive research efforts focused on enhancing the adsorbate-adsorbent interaction and CH4 storage via the increase of the specific surface area of adsorbents. One simple approach is to fabricate the adsorbents having increased surface area by HNO3 modification [18].
In the present study, we investigated the open-tip effect of the as-prepared CNTs on their structure and adsorption performance. The pristine CNTs was synthesized by CVD in a quartz tube reactor, and a series of open-tip carbon nanotubes (CNTs) were obtained by HNO3 modification with various concentrations of as-prepared CNTs. The structural property was determined by N2 adsorption/desorption at 77 K, scanning electron microscopy (SEM), X-ray diffraction (XRD), Raman spectroscopy, Fourier-transform infrared spectroscopy (FT-IR), and thermogravimetric analysis (TGA), respectively. The data of methane adsorption equilibrium were measured by a volumetric method, and experimental results of methane adsorption by CNTs were described by the model isotherms such as Langmuir and Dubinin–Astakhov (D-A) models. Furthermore, heat of adsorption was evaluated based on the Clausius–Clapeyron equation.
2. Experimental
2.1. Synthesis of the Adsorbents
The pristine CNTs was synthesized by CVD in a quartz tube reactor (inner diameter = 3.5 cm) laid in a horizontal furnace with a thermocouple in its central zone [19–21]. The catalyst loaded in a ceramic boat was reduced at 550°C in H2 gas for 1 h and then raised to the reaction temperature of 650°C. Subsequently, the system was switched to methane of high purity (50 mL/min) for 1 h. After the reaction, the as-grown CNTs were cooled to room temperature in N2 atmosphere. The product was collected and labeled as O-CNT.
2.1.1. Synthesis of L-CNTs
3.0 g of O-CNTs was modified in 200 mL of 4.0 M HNO3 refluxing at 110°C for 4 h. After cooling to room temperature, it was filtered and washed with deionized water until the pH of the filtrate was around 7. Finally, the product was dried at 60°C for 12 h and named as L-CNT.
2.1.2. Synthesis of M-CNTs
3.0 g of O-CNTs was modified in 200 mL of 10.0 M HNO3 refluxing at 110°C for 10 h. Others were similar to the above. Finally, the product was dried at 60°C for 12 h and represented by M-CNT.
2.1.3. Synthesis of H-CNTs
3.0 g of O-CNTs was modified in 200 mL of 14.8 M (68 wt.%) HNO3 refluxing at 110°C for 15 h. Others were similar to the above, and the sample was dried at 60°C for 12 h and denoted as H-CNT.
2.2. Characterization
The textural characterization of the samples was obtained by N2 adsorption/desorption isotherms, determined at 77 K with a NOVA1000e surface area and pore size analyzer (Quantachrome Company) [19, 22]. The specific surface areas were determined according to the BET method. Total pore volume was calculated from the nitrogen volume held at = 0.98∼0.99, and the micropore volume was estimated using the HK equation.
Surface morphology was investigated by scanning electron microscopy (SEM) (Hitachi S-4800, Japan).
X-ray diffraction of the samples was recorded in DX-1000 powder diffractometer equipped with Cu Kα X-ray source and an internal standard of silicon power.
Raman studies of the CNTs samples were conducted using a Renishaw inVia Raman spectrometer (UK) with an excitation source of 532 nm.
FT-IR analysis of the samples was performed on a Nicolet DXC20 FTIR spectrometer.
TGA of the samples was performed on a thermogravimetric analyzer (TGA Q500). The samples were heated at a rate of 10°C/min under N2 atmosphere.
2.3. Methane Adsorption Evaluation
Methane adsorption isotherms were conducted using a volumetric method similar to that previously described [10, 13, 22]. Prior to each experiment, the samples were degassed for 3 h at 323 K, and after the samples were cooled down to the required temperature, helium was introduced into the system in order to calibrate the void volume in the adsorption setup via helium expansion. The purities of helium and methane were 99.999% and 99.99%, respectively. Methane adsorption test was operated under pressure ranges of 0∼4.0 MPa at 298 K, and the adsorption process was repeated three times to guarantee the validity of the experimental data.
3. Results and Discussion
3.1. Textural and Structural Characteristics
The porosity parameters of the samples were studied by adsorption analysis using N2 as the adsorbate molecule [1]. Figure 1 shows the N2 adsorption/desorption isotherms at 77 K for the as-prepared CNTs and the corresponding open-tip samples. From the shapes and type of the obtained isotherms (type II of the IUPAC classification), it can be concluded that all the samples presented a micro-mesoporous character, for the isotherms displayed small hysteresis loops reflecting a minimum presence of mesopores. Furthermore, the HNO3 modification process carried out provoked a continuous increase in the adsorption capacity of N2. As exhibited in Figure 1, the volume amount adsorbed (cm3/g) increased with the increase of HNO3 concentration.
Table 1 summarizes the textural characterization results for the pristine and modified adsorbents, as calculated from N2 adsorption/desorption isotherms. The specific surface area for the pristine CNTs (O-CNT) was the lowest (98.7 m2/g) with a micropore volume of 0.0392 cm3/g which was less than half of the total pore volume (0.120 cm3/g). The modified CNTs suffered a drastic increase in all textural properties, indicating that HNO3 may have opened the tips of the as-prepared sample. For the sample H-CNT modified with 14.8 M HNO3 refluxing at 110°C for 15 h, a BET specific surface area of 174.4 m2/g and a HK micropore volume of 0.0682 cm3/g were reached.
To investigate the effect of HNO3 modification with various concentrations and treating times on the modified CNTs, SEM observation was employed to obtain the morphology of the resulted samples. As shown in Figure 2, some tips of the modified CNTs were open, and the external diameter of CNTs was mainly distributed in the range of 10∼34 nm, illustrating the carbon nanotubes were multiwalled. It could be seen that, in spite of various HNO3 concentrations, the samples of L-CNT, M-CNT, and H-CNT presented no apparent difference in morphology. Comparatively, the H-CNT sample displayed a much larger amount of open-tip due to the accelerated activation process at higher HNO3 concentration.
In order to study the influence of HNO3 concentrations on crystalline phase, the X-ray diffraction (XRD) analysis was conducted to characterize the structural property of the samples [23–26], and the results are shown in Figure 3. The (002) peak emerged at 26°, corresponding to the interplanar spacing of 0.34 nm. Also, the (100) peak at 43° was observed. Furthermore, the crystalline structure of the samples was not significantly altered by HNO3 concentrations, suggesting the concentration of HNO3 had no obvious effect on the structural property of the modified CNTs.
Raman spectra of the samples, shown in Figure 4, presented a prominent D-band (defects) at 1350 cm−1 and a G-band at 1575 cm−1. The increase of the G-band relative intensity indicated an improvement of the crystallite orientation; however, the increasing D-band relative intensity suggested more defects and an increasing amount of disordered structure. The pertinent Raman parameter ID/IG, mentioned to account for the degree of structural order, as displayed in Figure 4, confirmed the structure development of the samples (ID/IG values of O-CNT, L-CNT, M-CNT, and H-CNT were 0.983, 1.113, 1.222, and 1.313, respectively). The HNO3 oxidation of CNTs opened the tube tips, and according to the results of Raman spectra, HNO3 modification can produce some structural defects in addition to eliminating tube ends.
FT-IR profiles for the modified CNTs are displayed in Figure 5. The samples exhibited similar FT-IR spectra, suggesting that they had similar structures and functional groups. A strong absorption band appeared at around 3400 cm−1 which was assigned to -OH stretching. The adsorption peaks observed at 2900 cm−1∼3000 cm−1 and 1670 cm−1∼1760 cm−1 were attributed to -COOH group vibrations. The peak appearing at 1000 cm−1∼1200 cm−1 can be attributed to esters, such as those in ethers, phenols, and hydroxyl groups. However, all those peaks were more pronounced in H-CNT, suggesting the presence of more oxygen functional groups.
Thermogravimetric analysis (TGA) measured from room temperature to 800°C at N2 atmosphere with a heating rate of 10°C/min has been used to study the structural changes for the samples modified by HNO3 modification with various concentrations [27]. The initial sample weight was measured at room temperature. As confirmed in Figure 6, the weight loss for H-CNT (13.62%) due to oxygen functional groups was generally higher than those for other CNTs (9.45% for L-CNT and 12.36% for M-CNT), which was consistent with the FT-IR results.
3.2. Methane Adsorption Performance
The CH4 adsorption data measured at 298 K and up to 4.0 MPa on the CNT samples are displayed in Figure 7. It was noteworthy that the H-CNT sample exhibited a higher CH4 uptake due to its higher surface area and abundant micropores, followed by M-CNT and L-CNT, and the sample O-CNT had the smallest adsorption capacity among all the prepared adsorbents. The CH4 adsorption isotherms were type I of IUPAC and described by Langmuir and Dubinin–Astakhov (D-A) models, respectively. The corresponding adsorption parameters of the models are given in Tables 2 and 3, and it can be concluded that the D-A model fitted the data better than the Langmuir model, for the mean relative deviation δ for D-A model was smaller than that for Langmuir model.
(a)
(b)
The parameter t in the D-A model describes the surface heterogeneity. It has been reported that the surface of the adsorbent is less heterogeneous when the t value is approaching 3 more (t < 3). Table 3 shows the value of t decreased with the modification concentration increasing. The higher the HNO3 concentration was, the more heterogeneous the sample was, which ultimately led to a greater CH4 uptake.
It is well known that the methane adsorption is a physical process, mainly dependent on the specific surface area and micropore volume (shown in Figure 8). The relationship between n0 (from the D-A model) and BET specific surface area (BET SSA) is presented in Figure 8. The result showed BET SSA was roughly proportional to n0 and the correlation coefficient R2 was 0.962. Also, the relationship between n0 and micropore volume is exhibited in Figure 8, and the correlation coefficient R2 was 0.989, which was higher than that for BET SSA. This was because the micropore volume not specific surface area was a more important factor for gas adsorption on carbon materials.
Figure 9 shows the heat of adsorption Hads as a function of surface loading . As shown in Figure 9, the Hads varied largely with the surface loading for all the samples, and it lay in the range of 9.5∼19.4 kJ/mol, indicating the physical nature of the adsorption process. Due to the surface heterogeneity of the adsorbents and the stronger interaction between the CNTs and methane, the Hads decreased with surface loading increasing. However, when approached above 0.75, the Hads of the O-CNT, L-CNT, and M-CNT increased to a little higher value, indicating that the interaction between the adsorbent and methane was weaker and methane/methane interactions within the adsorption layer became obvious.
4. Conclusions
To investigate the structural features of open-tip CNTs for the methane adsorption capacity, the as-prepared CNTs synthesized via the CVD method were treated by HNO3 modification with various concentrations. The modified CNTs had higher specific surface area, larger total pore volume, and greater micropore volume than those of the as-prepared CNTs. Such unique structural features allowed remarkably high methane uptakes, which was an effective way to improve the adsorption capacity when compared with the original CNTs. The resultant H-CNT exhibited high methane uptake of 26.15 mg/g from the D-A model, which was nearly as twice as the uptake of original CNTs (13.62 mg/g). The heat of adsorption was initially 19.4 kJ/mol at lower coverage and 9.5 kJ/mol at higher methane coverage for H-CNT, indicating the physical nature of methane adsorption over open-tip CNTs. The current study may have an impact on the advance of CNTs for practical application of CH4 storage.
Data Availability
The data used to support the findings of this study are available from the corresponding author upon request.
Conflicts of Interest
The authors declare that they have no conflicts of interest.
Acknowledgments
This work was supported by the National Natural Science Foundation of China (Grant no. 21606058), the Natural Science Foundation of Guangxi (2017GXNSFBA198193 and 2017GXNSFBA198124), the Special Funding for Distinguished Expert from Guangxi Zhuang Autonomous Region, and the Startup Foundation for Doctors of Guilin University of Technology (no. 002401003510).