Skip to main content
NCBI home page
ACS Omega logoLink to ACS Omega
. 2020 Oct 22;5(43):28158–28167. doi: 10.1021/acsomega.0c03827

Catalytic Mechanism of Liquid-Metal Indium for Direct Dehydrogenative Conversion of Methane to Higher Hydrocarbons

Yuta Nishikawa †, Yuhki Ohtsuka ‡, Hitoshi Ogihara §, Rattanawalee Rattanawan ‡, Min Gao ‡, Akira Nakayama ∥, Jun-ya Hasegawa ‡, Ichiro Yamanaka †,*
PMCID: PMC7643202  PMID: 33163798

Abstract

graphic file with name ao0c03827_0010.jpg

There is a great interest in direct conversion of methane to valuable chemicals. Recently, we reported that silica-supported liquid-metal indium catalysts (In/SiO2) were effective for direct dehydrogenative conversion of methane to higher hydrocarbons. However, the catalytic mechanism of liquid-metal indium has not been clear. Here, we show the catalytic mechanism of the In/SiO2 catalyst in terms of both experiments and calculations in detail. Kinetic studies clearly show that liquid-metal indium activates a C–H bond of methane and converts methane to ethane. The apparent activation energy of the In/SiO2 catalyst is 170 kJ mol–1, which is much lower than that of SiO2, 365 kJ mol–1. Temperature-programmed reactions in CH4, C2H6, and C2H4 and reactivity of C2H6 for the In/SiO2 catalyst indicate that indium selectively activates methane among hydrocarbons. In addition, density functional theory calculations and first-principles molecular dynamics calculations were performed to evaluate activation free energy for methane activation, its reverse reaction, CH3–CH3 coupling via Langmuir–Hinshelwood (LH) and Eley–Rideal mechanisms, and other side reactions. A qualitative level of interpretation is as follows. CH3–In and H–In species form after the activation of methane. The CH3–In species wander on liquid-metal indium surfaces and couple each other with ethane via the LH mechanism. The solubility of H species into the bulk phase of In is important to enhance the coupling of CH3–In species to C2H6 by decreasing the formation of CH4 though the coupling of CH3–In species and H–In species. Results of isotope experiments by combinations of CD4, CH4, D2, and H2 corresponded to the LH mechanism.

1. Introduction

Methane is the main component of natural gas which affluently exists on the earth. The development of new technologies to extract natural gas from the shale has led to global attention of an efficient use of natural gas. Therefore, CH4 is a promising candidate which plays an important role as the future energy and carbon sources.1−3

Conversion of CH4 into chemical feedstocks and liquid fuels is one of the suitable ways to utilize CH4 effectively. Almost all of CH4 is currently used to generate thermal energy and electric power because we do not have industrial chemical technology to convert CH4 into useful chemicals except for methanol synthesis and Fischer–Tropsch process. These processes consume a large amount of energy because of multistep processes via steam reforming of CH4 to synthesis gas (2H2 + CO). It is desirable to develop a new catalyst and reaction for direct conversion of CH4 into useful chemicals and liquid fuels.

High stability of CH4 is a big barrier for the direct conversion of CH4. It is difficult to activate CH4 and form a C–C bond because of a strong C–H bond (438 kJ mol–1) and a symmetrically stable structure of the CH4 molecule. In the early 1980s, attractive and ambitious works for oxidative conversion of CH4 to chemicals were reported by using the well-designed catalysts.4−7 Active oxygen species derived from O2 could cleave a strong C–H bond of CH4. An oxidative coupling of CH4 can directly produce C2H6 and C2H4. In addition, partial oxidation of CH4 can produce methanol and formaldehyde by using various catalysts.8−10 However, it is inevitable to suppress deep oxidation producing CO2 by successive oxidation of products in both reactions. On the other hand, dehydrogenative conversion of CH4 (DCM) can also produce useful chemicals. The catalytic DCM reaction has been strenuously researched since Wang et al. reported Mo/HZSM-5 catalysts.11 The Mo/zeolite catalysts can convert CH4 into benzene (C6H6) with a good yield, although deactivation of the catalyst by carbon deposition is serious. In addition, some effective catalysts such as carbonaceous materials,12−14 Pt–SO4/ZrO2,15 TaH/SiO2,16 GaN,17 and Pt-modified zeolite catalysts18,19 were found for the DCM reaction. Recently, the catalysis of single Fe sites in a silica matrix (Fe@SiO2) was reported for the DCM catalyst.20,21 Fe@SiO2 achieved a high hydrocarbon yield (48%) without carbon deposition at 1363 K. Kjølseth et al. succeeded to suppress the carbon deposition on the Mo/HMCM-22 catalyst by using a coionic membrane reactor.22

Transition metals such as Fe, Ni, and Pd are well-known catalysts for the decomposition of CH4 to C (carbon fiber) and H2.23−26 The transition metals catalyze the cleavage of four C–H bonds of CH4. We considered that the DCM reaction would occur through an activation of one or two C–H bonds over a metal surface and a successive coupling of CH3 or CH2 species. According to the former concept, the catalysis of transition metals was too strong for the DCM reaction. Therefore, we focused on the catalysis of post-transition metals (Ga, In, Sn, Tl, Pb, and Bi). From a previous screening study on the DCM reaction, we found that a SiO2-supported indium catalyst (In/SiO2) was effective for the DCM reaction.27 We have already revealed that liquid-metal indium catalyzed the DCM reaction.27,28 From the density functional theory (DFT) molecular dynamics (MD) simulation study,29 the gas/liquid interface of indium was significantly disordered at operating temperature, and structures that looked like small In clusters appeared/disappeared dynamically. The potential energy profile calculated with DFT also showed the catalytic role of the small In clusters, such as In2, which has reasonably small activation energy for the C–H dissociation of methane.29 It has been already reported that the indium cation species in ZSM-5 (In-ZSM-5) could activate the C–H bonds of CH4 and react with ethylene and benzene;30−35 however, the coupling of CH4 to C2H6 could not proceed by the In-ZSM-5 catalyst. Therefore, it was observed that the catalysis of In/SiO2 is very different from that of In-ZSM-5. Different liquid-metal catalyses of dehydrogenation of butyl alcohol by liquid Zn and Ga,36,37 decomposition of CH4 by molten Pb38 and molten Ni–Bi,39 and dehydrogenation of butane by molten Ga–Pd40 have been reported. Various liquid-metal catalyses and properties have recently received much attention and interest.41 The purposes of this work are to investigate the unique catalysis of liquid-metal indium on SiO2 for the DCM reaction and to clarify the reaction mechanism by first-principles DFT MD calculations42−50 (Calculations S1–S5 and Figure S1–S5 in the Supporting Information), various kinetic studies (Experiments 4.3.1–4.3.8 and definitions of reactant conversions and product selectivities in eqs 13–18), and isotope experiments.

2. Results and Discussion

2.1. Property of the DCM Reaction

Figures 1 and S6 show the durability of the In/SiO2 catalyst for conversion of CH4, the selectivity to sum of hydrocarbons including aromatics, and the selectivity to carbon deposition in the DCM reaction. A selectivity to sum of aromatics was additionally indicated. The CH4 conversion decreased from 5.4 to 3.5% in the initial stage of the reaction, and time courses of the selectivities to hydrocarbons and aromatics were similar to the conversion. On the other hand, the selectivity to the carbon deposition increased with process time until 6 h. The performance of the DCM reaction was stable after 6 h. The CH4 conversion was 3.5% at 24 h and the selectivity to hydrocarbons was 63%, which showed almost the same performance at 6 h. Although the catalytic activity was weakly deactivated at the initial stage, the In/SiO2 catalyst showed a good durability for the DCM reaction and against the carbon deposition. As mentioned above, the selectivity to the carbon deposition including unidentified products was almost constant after 6 h.

Figure 1.

Figure 1

Open in a new tab

Time course of CH4 conversion and selectivity to sum of hydrocarbons including aromatics, sum of aromatics, and cokes in the DCM reaction over 10 wt % In/SiO2 at 1173 K for 24 h; red circle: CH4 conversion, blue square: selectivity to sum of hydrocarbons, black diamond: selectivity to carbon deposition, and green triangle: selectivity to sum of aromatics (benzene, toluene, o-xylene, styrene, indene, naphthalene, biphenyl, anthracene, and phenanthrene).

If we can remove the carbon deposition on the In/SiO2 catalyst, the catalytic activity will be recovered. To remove the carbon deposition, O2 was introduced over the catalyst instead of CH4 at the same temperature and then the oxidized catalyst was re-reduced with H2 before the DCM reaction. As shown in Figure S7, the initial catalytic activity seems to be recovered after the O2 treatments. Small increases in the CH4 conversion and the selectivity to aromatics were observed in the initial stages after the O2 treatments. When H2 was induced instead of the O2 treatment, the increase in initial catalytic activity was not observed. However, both DCM performances at 8 h at the end of the reaction were very similar. These facts indicated that the steady catalytic activity was not influenced by the carbon deposition.

The N2 adsorption isotherms of the fresh In/SiO2 catalyst and the used In/SiO2 catalyst were measured, as shown in Figure S8 and Table S1. Their specific surface areas were very low, 0.10 and 0.89 m2 g–1, respectively. The low surface area is a unique characteristic of the In/SiO2 catalyst. A small increase in the surface area of the used catalyst should be due to the carbon deposition. Additionally, the amount of N2 adsorption steeply increased at higher relative pressure for the used In/SiO2 catalyst. This indicated the macropore carbon deposition on the In/SiO2 catalyst. The decreases in the initial performance of the CH4 conversion and the selectivity to aromatics may be due to the decrease in the surface area of indium by the carbon deposition, but the carbon deposition did not affect the catalytic activity of indium in the steady state. The amount of carbon deposition was confirmed by the thermogravimetric (TG) analysis as shown in Table S2. The amount of carbon deposition from the TG analysis was much smaller than that from calculation based on the hydrogen balance (Experiment 4.3.8). A black deposition was observed on the reactor wall after the DCM reaction. This indicates that the carbon deposition occurred not only on the catalyst surface but also on the reactor wall.27

Figure 2 shows the effects of partial pressures of CH4 [p(CH4)] on the DCM reaction by the In/SiO2 catalyst (Experiment 4.3.2). The conversion rates of CH4 increased with the first order of p(CH4). The selectivities to hydrocarbons and carbon deposition were almost constant without a dependence on p(CH4). The dependence of the CH4 conversion rate on p(CH4) suggested that an impact of CH4 against indium should participate an activation of a C–H bond of CH4. Apparent activation energies of the DCM reaction by using the In/SiO2 catalyst and the SiO2 support were obtained in the previous work.27 A good linearity in ln[r(CH4)] versus 1/T was observed in both the cases. From their slopes, the apparent activation energies were calculated as 170 kJ mol–1 for the In/SiO2 catalyst and 365 kJ mol–1 for the SiO2 support. It is clear that indium catalyzes the activation of a C–H bond of CH4.

Figure 2.

Figure 2

Open in a new tab

Effect of partial pressure of CH4 on the conversion rate of CH4 and selectivity to hydrocarbons and coke; red circle: CH4 conversion, square: selectivity to sum of hydrocarbons, and diamond: selectivity to carbon deposition.

Figure S9 shows a temperature-programmed reaction (TPR) in CH4 over the SiO2 support and the In/SiO2 catalyst (Experiment 4.3.6).27 In the case of the In/SiO2 catalyst, increases in the mass spectra of m/z = 2 (H2), 28 (C2H4), and 30 (C2H6) were observed with an increase in temperatures lower than 1000 K. Figure S9c shows the onset of the mass spectra of m/z = 2 (H2), 28 (C2H4), and 30 (C2H6) over the In/SiO2 catalyst. These spectra rose up around 800 K and rose exponentially over 900 K. The ratio of the onset intensity of m/z = 28 and 30 was approximately 4:1, which is in agreement with the ratio of a pure C2H6 gas. The onset temperatures of m/z = 41 (C3H6) over 1000 K and 78 (C6H6) over 1050 K shown in Figure S9b were higher than 800 K. This indicated that C3H6 and C6H6 were successively produced from primary products (m/z = 28, 30). In the case of the SiO2 support (Figure S9a), the spectra slowly rose over 900 K. It was clear that indium activated CH4 and catalyzed the dehydrogenation of CH4 to C2H6 over 800 K.

2.2. Reactivity of C2H6

As mentioned above, C2H6 is the primary product in the DCM reaction by the In/SiO2 catalyst. In order to know the reactivity of C2H6, dehydrogenation of C2H6 and hydrogenation of C2H6 were conducted under different conditions. Figure 3a,b shows the TPR spectra in 5% C2H6/Ar on the SiO2 support and the In/SiO2 catalyst (Experiment 4.3.6). The decrease in the mass spectra of m/z = 28 and 30 started over 900 K on both the cases, which indicated that the dehydrogenation of C2H6 started around 900 K. To be noted, the onset temperatures of the spectrum of m/z = 2 (H2), 16 (CH4), and 78 (C6H6) on the In/SiO2 catalyst were higher than that on the SiO2 support. This indicates that the In/SiO2 catalyst weakly suppresses the dehydrogenation of C2H6. Additionally, the onset temperature of the conversion of C2H6 in TPR was higher than that of the CH4 conversion in TPR, though C2H6 was generally more reactive than CH4. These results indicate that C2H6 is noncatalytically converted at 900 K and indium does not activate C2H6. Figure 3c,d shows the TPR spectra in 5% C2H4/Ar on the SiO2 support and the In/SiO2 catalyst. The onset temperature of conversion of C2H4 was 950 K. There was no difference between both the cases. Although indium catalyzes the conversion of CH4, it cannot catalyze the conversion of C2H4, similar to C2H6. The TPR results proposed that indium is inert for the hydrocarbon formation from C2H6.

Figure 3.

Figure 3

Open in a new tab

Mass spectra of TPR in C2H6 over (a) SiO2 and (b) 10 wt % In/SiO2 and in C2H4 over (c) SiO2 and (d) 10 wt % In/SiO2 at 4 K min–1.

Additional experiments were conducted to clarify the reactivity of C2H6 for the In/SiO2 catalyst under a steady state (Experiment 4.3.5). Figure 4 shows conversions of C2H6 using (a) a gas mixture of C2H6 and Ar and (b) a gas mixture of C2H6 and H2 with the In/SiO2 catalyst, a Ni/SiO2 catalyst, and the blank test. In the case of (a), dehydrogenation of C2H6 proceeded. 43% of the C2H6 conversion by the In/SiO2 catalyst was observed. This value was lower than 58% of the blank test (without catalysts). These observations were consistent with the results of C2H6 TPR, which is a higher onset temperature of m/z = 2 (H2) on the In/SiO2 catalyst than that on the SiO2 support. These facts proposed that indium showed negative catalysis for the dehydrogenation of C2H6. In the case of (b), the dehydrogenation of C2H6 to C2H4 mainly proceeded and a part of hydrogenolysis of C2H6 to CH4 proceeded using the In/SiO2 catalyst and no catalyst. Conversions of C2H6 in the two systems were approximately 60% and selectivities to C2H4 and CH4 were very similar. On the other hand, when the Ni/SiO2 catalyst was used, the hydrogenolysis of C2H6 to CH4 mainly proceeded. 91% of the C2H6 conversion and 67% of the CH4 selectivity were observed. Ni/SiO2 catalyzed the cleavage of a C–C bond of C2H6, followed by the hydrogenation of methyl species. In the case of the In/SiO2 catalyst, indium does not catalyze a cleavage of C–C and C–H bonds of C2H6, though indium can activate a C–H bond of CH4. This unique catalysis of liquid-metal indium to activate only CH4 results in a higher selectivity to hydrocarbons in the DCM reaction.

Figure 4.

Figure 4

Open in a new tab

C2H6 conversions using (a) a gas mixture of C2H6 and Ar with the In/SiO2 catalyst and no catalyst at 973 K and (b) a gas mixture of C2H6 and H2 with the In/SiO2 catalyst, Ni/SiO2 catalyst, and no catalyst at 1073 K; gray: coke, red: C2H4, and green: CH4.

The results of the TPR in Figure 3 and the C2H6 conversion in Figure 4 clearly indicated that alkenes and aromatics were thermally produced from C2H6 in the gas phase.51 To clarify contribution of the gas-phase reactions on the conversion of CH4 to C2H6, we studied the effects of the volume of the quartz reactor on the DCM reaction. The results using a narrow reactor (i.d. 8 mm) and the standard one [i.d. 12 mm] are shown in Figure S10. A conversion of CH4 slightly decreased using the i.d. 8 mm reactor, but the difference between the two reactors was within an experimental error. The C2H6 selectivity using the i.d. 8 mm reactor was higher than that using the i.d. 12 mm one. In contrast, selectivities to C2H4 and C3H6 for the former were lower than that for the latter. Although the gas hourly space velocity (GHSV) values were the same using the two reactors, the linear velocity of the former was 2.25 times faster than that of the latter. If the conversion of CH4 was promoted by CH3• in the gas phase, the conversion should drastically increase with the decrease in linear velocity; however, the conversions were almost the same. Hence, the conversion of CH4 to C2H6 proceeds on the In/SiO2 catalyst but not in the gas phase. In addition, conversion of C2H6 to other hydrocarbons proceeds in the gas phase (Figure 4).

2.3. Reaction Path

Figure 5 shows the effects of the contact time (W/F) on the conversion of CH4 by using the In/SiO2 catalyst (4.3.3. Experiment). A reaction temperature of this study was decreased to 1073 from 1173 K to suppress the successive reactions in the gas phase. The major products were C2H6 and C2H4, although C2H2 and C3H6 were detected at a long contact time. When the weight of the In/SiO2 catalyst was fixed at 100 mg and the flow rate of CH4 was varied from 5 to 30 mL min–1, the conversion of CH4 increased with the contact times but slightly decelerated with 300 mg of the In/SiO2 catalyst. To be noted, a slope of the line at 300 mg of the catalyst was distinctly different from that at 100 mg. Different values of conversion at 300 and 100 mg were obtained at the same contact time. When the flow rate of CH4 was fixed at 20 mL min–1 and the weight of the In/SiO2 catalyst was increased from 50 to 300 mg, the conversion of CH4 increased with the contact time (catalyst weight) but reached to an upper limit (ca. 0.4%) while the thermodynamic equilibrium conversion of CH4 to C2H6 and H2 is 3.5% under the condition shown in Figure 5. The effects of W/F values on product selectivities are shown in Figure S11. When the contact time (W/F) was increased, the selectivity to C2H6 decreased from around 50 to 20% and that to C2H4 increased from around 15 to 40%. The selectivity to carbon deposition was constant with the contact time. This indicates that successive conversion of C2H6 to C2H4 proceeded at high W/F. CH4 conversion shown in Figure 5 is much lower than the equilibrium conversion of CH4. The tendency of CH4 conversion to an upper limit as shown in Figure 5 is strange in terms of thermodynamics.

Figure 5.

Figure 5

Open in a new tab

Effects of contact time on CH4 conversion; (red circle) changing the catalyst weight at 20 mL min–1 of CH4, (black triangle) changing the flow rate of CH4 at 100 mg of catalyst weight, and (blue diamond) 300 mg of catalyst weight. T = 1073 K, In/SiO2: 50–300 mg, CH4: 5–30 mL min–1.

Generally, the conversion increases linearly with the contact time in a usual catalytic reaction far from the thermodynamic equilibrium. There are conflicts between the results in Figure 5 and the usual results. We suggest that an intermediate species is involved in the conversion of CH4 to C2H6 and an equilibrium reaction between CH4 and an intermediate (eq 1) is concerned in the DCM reaction. This is the reason why CH4 conversion does not achieve the thermodynamic equilibrium of conversion of CH4 to C2H6 despite the increase in contact time. A kinetic equation of the formation rate of C2H6 based on eq 1 was derived in eq 2, which depended on a first order to p(CH4) and indium concentration. eq 2 is in agreement with the experimental result of the effect of p(CH4) in Figure 2.

2.3. 1
2.3. 2

2.4. Calculations

Here, we also performed DFT calculations to investigate the mechanism. Because the indium catalyst is in liquid state at experimental temperature, which is totally disordered and dynamically fluctuating (for a MD snapshot, see Figure S12), it was very difficult to adopt a cluster model to represent the dynamically disordered catalytic surface. Thus, we performed the first-principles MD calculations under the temperature close to experimentally operating conditions. Before considering the mechanism, we observed the dynamics of CH3 and H species after the C–H bond of methane is broken.55 In the trajectory, CH3 and H form CH3–In and H–In species, respectively. No CH3–In–H species was observed because of the instability, which was also confirmed by nonperiodic DFT calculations (see more details in Figure S13 in the Supporting Information). We observed that CH3–In and H–In diffuse in different ways. Figure 6 shows the position of C and H atoms relative to the indium surface in the trajectory. The CH3–In species stayed on the surface, while the H–In species moved inside the indium catalyst and appeared on the other surface. This result indicates that the CH3–In species accumulates on the liquid surface. In contrast, the H–In species spreads into the liquid, and its concentration at the surface becomes lower than that of the CH3–In species. These trends were also observed in MD calculations with higher CH3 and H concentrations,54 as shown in Figures S3 and S4. This distribution helps to suppress the recombination of CH3 and H to regenerate methane.

Figure 6.

Figure 6

Open in a new tab

Position of H (blue) and C (orange) atoms relative to the liquid indium surface in trajectory. The inset indicates the liquid layer in the computational model.

The reaction mechanism of the DCM reaction by the In/SiO2 catalyst was proposed from the results of kinetic experiments and DFT calculations, as described below. Equations 3–10 show proposed elementary reactions. First, thermally excited indium surface atoms, for which In2 is used as a representative model for active species,54 dissociate a C–H bond of CH4 and CH3(In2)H as an intermediate species form.51 The actual active species may be coordinatively unsaturated and activated indium species on the liquid-metal indium surface. This first reaction is the rate-determining step. The intermediate species decomposes and CH3–In species and H–In species form on the liquid-metal indium surface. CH3–In species and H–In species freely migrate on the liquid-metal surface. C2H6 forms when the CH3–In species encounters the CH3–In species. As mentioned above, migration of CH3–In species and H–In species on the liquid indium surface is unique for liquid-metal catalysis and important for dehydrogenative coupling of CH4 and C2H6. Regeneration of CH4 occurs when CH3–In species encounters the H–In species. Alkenes and aromatics are successively produced through thermal conversion of C2H6 to C2H4.27 The formation rate of C2H6 is expressed as eq 8. When the formation rate of CH4 (k3) on the surface is much faster than the formation rate of C2H6 (k41/2k51/2), we observe pseudo-chemical equilibrium (k1/k3) between CH4 and CH3(In2)H, as shown in eq 9. This equation of formation rate is in proportion to p(CH4) and accountable for pseudo-chemical equilibrium, which is in agreement with the experimental results and our discussion. On the other hand, when the formation rate of CH4 (k3) is much lower than the formation rate of C2H6 (k41/2k51/2), the formation rate of C2H6 is expressed as eq 10. These kinetic equations show that the catalytic performance of indium may be improved by modification of the surface reactions. If we suppress the coupling of CH3(ad) and H(ad) or increase the concentration of active species, the formation rate of C2H6 will increase.

2.4. 3
2.4. 4
2.4. 5
2.4. 6
2.4. 7
2.4. 8
2.4. 9
2.4. 10

The above proposed mechanism was verified by DFT/MD calculations at T = 1200 K, as shown in Figure 7. In addition to the major reaction steps, the Langmuir–Hinshelwood (LH) mechanism (A → B/B′ → C), we also examined a LH-type methane formation with two CH3–In species (B/B′ → D), and an Eley–Rideal (ER) mechanism for ethane production (B/B′ → E → F) in Figure 7. Ethane formation with two methyl radicals in the gas phase can be ruled out because the present DCM reaction showed reactor size independence in the methane conversion (see the description above). The first-principles MD trajectory was used for evaluating activation free energy with the Blue moon ensemble method.

Figure 7.

Figure 7

Open in a new tab

A plausible mechanism of the methane-to-ethane conversion by the liquid indium catalyst. The activation free energy is calculated in kJ/mol unit. For the B/B′ → E step, the CH3 desorption energy calculated with the In14 cluster was given (see the main text).

For the C–H bond cleavage of methane (A → B/B′), the activation free energy was 222.6 kJ mol–1, as shown in Figure 7. This is the largest activation free energy among all the steps in the present mechanism. As mentioned above, the experimentally observed apparent activation energy, which would correspond to that of the C–H dissociation, was 170 kJ mol–1. The overestimation in the calculation is probably due to the difficulty in the sampling. In the trajectory for evaluating the activation free energy, the C–H dissociation in the gas phase, which is a high energy component, contaminates to that in the surface. In addition, the use of SZV basis sets tends to increase the endothermicity (for numerical comparison of the computational setup, see the Supporting Information), which would also increase the calculated barrier. The activation free energy of C–H reformation (B/B′ → A) was calculated to be 91.2 kJ mol–1.

For the production of ethane from the two CH3–In species on the surface via the LH mechanism (B/B′ → C), the calculated activation free energy is 91.6 kJ mol–1. This is comparable to the reverse reaction to form methane (B/B′ → A). For the hydrogen abstraction from another CH3–In species (B/B′ → D), the activation free energy is 133.9 kJ mol–1, which is 42.3 kJ mol–1 larger than the ethane formation. In the ER mechanism, the CH3–In bond dissociation (B/B′ → E), which is endothermic (196.2 kJ mol–1), should precede before CH3 attacks the CH3–In species. In addition, the radical attack was prioritized to H, not to C, because of the proximity to the H atom. The activation free energy of H2 formation on the surface was calculated to be only 50.2 kJ mol–1.

These results suggest that ethane formation by the indium liquid catalyst proceeds by the LH mechanism. The rate-determining step is the C–H dissociation. The reverse reaction to regenerate methane would be suppressed because the concentration of H–In species on the liquid surface is reduced by the diffusion of H–In species inside the indium catalyst and also by the H2 formation.

2.5. Isotopic Studies

To investigate the hypothesis about elementary reactions on liquid-metal indium (eqs 5–7), isotope studies using CD4 were conducted (4.3.7. Experiment). CHD3 and C2D6 would be produced by the In/SiO2 catalyst. The CD4 pulse peaks were almost constant during the experiments (Figure S14) and used as a standard against various H–D exchanged products. Each component was distinct because parent peaks of CHD3 (m/z = 19), C2D6 (m/z = 36), and CD4 (m/z = 20) are different from each fragment peak.52,53Figure 8 shows the intensity ratios of HD (m/z = 3), CHD3 (m/z = 19), and C2D6 (m/z = 36) to CD4 (m/z = 20) for CD4 pulse with 2% H2/N2 carrier. There was no difference between SiO2 and the In/SiO2 catalyst at room temperature before and after heating. The intensity ratios were not zero at 298 K, probably because of the contamination in CD4. On the other hand, all intensity ratios of the In/SiO2 catalyst were higher than that of the SiO2 support at 1073 K. That is, C2D6 formation from CD4 and the H–D exchange reaction between CD4 and H2 were accelerated by the In/SiO2 catalyst under the reaction condition (eqs 11 and 12).

2.5. 11
2.5. 12

Figure 8.

Figure 8

Open in a new tab

Intensity ratio of (a) m/z = 3 (HD), (b) m/z = 19 (CHD3), and (c) m/z = 36 (C2D6) to m/z = 20 (CD4) in the H–D exchange tests. T = RT and 1073 K, catalyst: SiO2 and 10 wt % In/SiO2 (300 mg), carrier gas: 2% H2/N2 (20 mL min–1), and CD4 pulse: 2.5 mL.

Although we also observed the H–D exchange between CD4 and H2 in the gas phase as shown in Figure S15, the results of Figure 8 show the significant differences between the SiO2 support and the In/SiO2 catalyst. This shows that one C–H bond of CH4 was activated on the liquid-metal indium surface because of no influence of the gas phase on the conversion of CH4. These results correspond to eqs 5–7 and suggest a CH3 intermediate species on the liquid-metal indium surface.

In the case of a comparative experiment using the In/SiO2 catalyst with 0.2% H2/N2 carrier gas at 1073 K (Figure S16), the intensity ratio of HD and CHD3 to CD4 decreased from 0.02 to 0.004 and from 0.028 to 0.013, respectively. In addition, the intensity ratio of C2D6 to CD4 increased from 2 × 10–5 to 6 × 10–5. Decrement of H2 concentration in the carrier gas resulted in low yields of HD and CHD3 and a high yield of C2D6. We consider that a low concentration of H2 caused low H–In coverage on the liquid-metal indium surface; therefore, the probability of coupling of CD3–In species to C2D6 increased while that of association of CD3–In and H–In species to CHD3 decreased. These results also indicated the existence of eqs 5–7 and CH3–In on the liquid-metal indium surface.

3. Conclusions

We investigated liquid-metal indium catalysis for the DCM reaction in detail. It is clear that indium can cleave a C–H bond of CH4, although indium cannot activate C2H6 and C2H4. The unique catalysis of indium to activate only methane results in high selectivity to hydrocarbons in the DCM reaction. Our previous first-principles MD calculations showed that the liquid indium surface is significantly disordered under operating conditions.29 The result inspired the catalytic role of the small In clusters which appear as transient species on the surface. The result of DFT calculations showed that the C–H activation energy of In2 was the smallest among the clusters.29 From the results of DFT and MD calculations, it was observed that the thermally excited surface of liquid-metal indium, binuclear indium cluster (In2) as a suitable model,29 is presumably the active species for methane activation. CH3 species and H species are produced on the surface of liquid-metal indium. First-principles MD calculations were performed using the Blue moon ensemble method to evaluate the activation free energy for methane activation, its reverse reaction, CH3–CH3 coupling via LH mechanism, ER mechanism, and some of other side reactions (see Figure 7). Coupling of CH3 species to C2H6 is via the LH mechanism. Though H species can spread into the liquid-metal indium, CH3 species freely move on the liquid surface and couple to ethane. H species moves in three dimensions and CH3 species moves in two dimensions, which enhance the probability of coupling CH3 species to C2H6. The liquid state of indium plays an important role for CH4 activation and coupling of CH3 species to C2H6. This mechanism is compared with the previous one in much detail. The isotope experimental data directly proved the cleavage of a C–H (C–D) bond of methane to the adsorbed species of CH3 (CD3) and H (D) on the surface of liquid In and formation of ethane (CH3CD3) by coupling of two methyl species. It is suggested that the catalytic performance of indium might be improved by modification of the catalyst design to suppress the side reaction of recombination of CH3 and H species to CH4. Although this mechanistic insight was derived with our best effort in computation, the mechanistic interpretation could be at a qualitative level because of the limitation in computational resources. In this sense, mechanistic understanding of the liquid metal catalysis at high temperatures would be still at the very forefront of the computational approach.

4. Experimental Section

4.1. Catalyst Preparation

Two kinds of SiO2 (CARiACT Q-3, Fuji Silysia Chemical, and Admafine SO-E6, Admatechs) for a support, In(NO3)·nH2O (99.99%, Sigma-Aldrich), were used for catalyst preparation. CARiACT Q-3 and Admafine SO-E6 were calcined in air at 1273 K before the preparation. The In/SiO2 catalyst was prepared by a conventional impregnation method. In(NO3)·nH2O was added to deionized water, and then the SiO2 support was added to the solution. The mixture was dried up at 353 K under stirring. The powder of the catalyst precursor was dried for 2 h at 393 K and calcined in air for 3 h at 773 K. The calcined powder was ground in an agate mortar and reduced with H2 for 3 h at 873 K. The loading of In0 on the In/SiO2 catalyst was 10 wt %. Micrometer-size droplets of liquid-metal indium were on the SiO2 support and the catalyst from was powder under the reaction condition.27 A Ni/SiO2 catalyst as a reference was prepared by a conventional impregnation method from Ni(NO3)2·6H2O (99.9%, Wako Pure Chemical Industries).

4.2. Catalytic Reaction Procedures

Catalytic activity measurements were carried out in a fixed-bed quartz reactor (i.d. 12 mm) with 100 mg of the catalyst. The reactor was heated to reaction temperature in Ar flow (10 mL min–1) with 25 K min–1 after the air in the reactor was displaced by Ar flow (100 mL min–1) for 30 min. Reaction tests were performed in atmospheric pressure. H2 was analyzed by a gas chromatograph with a thermal conductivity detector (TCD) (GC-8A-TCD, Shimadzu) equipped with an activated carbon column (3φ, 2 m). CH4, ethane (C2H6), ethylene (C2H4), acetylene (C2H2), and propylene (C3H6) were analyzed by a gas chromatograph with a frame ionization detector (FID) (GC-8A-FID, Shimadzu) coupled with an Unibeads 1S column (3φ, 2 m). Aromatics (benzene, toluene, o-xylene, styrene, indene, naphthalene, biphenyl, anthracene, and phenanthrene) were analyzed by an on-line capillary-GC-FID (GC-2025, Shimadzu) equipped with an HR-1 capillary column (30 m).

4.3. Reaction Conditions of Various Studies

To reveal the catalysis of the In/SiO2 catalyst, various experiments were done, as described below.

4.3.1. Stability Test of the In/SiO2 Catalyst

Pure CH4 (10 mL min–1) was introduced at 1173 K for 24 h. This corresponds to GHSV: 3500 h–1 and W/F: 6000 mL gcat–1 h–1.

4.3.2. Effects of Partial Pressure of CH4

CH4 diluted by Ar was allowed to flow with a total flow rate of 10 mL min–1 at 1173 K.

4.3.3. Effects of Contact Time of CH4

Pure CH4 was introduced at 1073 K. The flow rate of CH4 was varied from 5 to 30 mL min–1 with a fixed catalyst weight (100 or 300 mg). The catalyst weight was varied from 50 to 300 mg with a fixed flow rate of CH4 (20 mL min–1).

4.3.4. Activation Energy of the DCM Reaction

Pure CH4 (20 mL min–1) was introduced at constant temperature from 1023 to 1188 K. When we conducted the experiments at different temperatures, we reset the apparatus and used the fresh In/SiO2 catalyst.

4.3.5. Reactivity Tests of the In/SiO2 Catalyst for C2H6

2% C2H6/Ar (20 mL min–1) and 0.5% C2H6/H2 (20 mL min–1) were introduced at 1073 K.

4.3.6. TPR Study

Pure CH4 (10 mL min–1), 5% C2H6/Ar (30 mL min–1), or 5% C2H4/Ar (30 mL min–1) were introduced with a ramping rate of 4 K min–1. Products were detected by an on-line mass spectrometer (M-200QA-M, Canon Anelva).

4.3.7. Isotope Study

2% H2/N2 (20 mL min–1) was introduced as a carrier gas. CD4 (2.5 mL) was injected with a six-way valve at (i) 298, (ii) 1073, and (iii) 298 K in turn. It was pulsed every four times at each temperature. The products were detected by an on-line mass spectrometer (OmniStar GSD 320 O2, Pfeiffer Vacuum).

4.3.8. Conversion and Selectivity Calculations

The CH4 conversions and product selectivities for the DCM reaction are calculated from eqs 13–15. An accurate carbon balance could not be obtained because quantitative analysis of carbon deposition on a reactor wall was impossible. However, an accurate hydrogen balance could be obtained by gas chromatography (GC) analyses of hydrocarbon products and H2. When an excess H2 production was observed in a difference between the amount of hydrogen in hydrocarbons and an amount of gaseous H2, the excess H2 corresponded to the C deposition from CH4 decomposition. C2H6 conversion and product selectivity in the reactivity tests for C2H6 were calculated from eqs 16–18. Selectivity to carbon was defined as the rest of the sum of selectivity to hydrocarbons in all calculations.

4.3.8. 13
4.3.8. 14
4.3.8. 15
4.3.8. 16
4.3.8. 17
4.3.8. 18

x: a quantified component (H2 or hydrocarbons: CiHj), N(x): the flow rate of component x at the moment [mol min–1], or sum of component x during the reaction [mol], in: input, out: output, HN(x): hydrogen number of component x (e.g., HN(H2) = 2, HN(CiHj) = j), CN(x): hydrogen number of component x (e.g., CN(H2) = 0, and CN(CiHj) = i).

4.3.9. Nonperiodic DFT Calculations

Gaussian 09 program package was used.42 Both geometry optimizations and energy calculations were performed at the ωB97XD functional level.43 We employed the Stuttgart/Dresden effective core potential (ECP60MWB)44 for the indium atom. The 6-31G(d) basis sets45,46 were used for the other atoms.

4.3.10. First-Principles MD Calculations

For MD with periodic boundary conditions, CP2K program package was used.47 For DFT calculations, we used the PBE functional48 combined with the SZV-MOLOPT-SR-GTH basis set for C and H atoms and the SZV-MOLOPT-SR-GTH plus GTH-PBE-q13 potential for indium atom. Numerical comparison of this computational setup is shown in the Supporting Information. We note that the basis sets used would be qualitatively correct but not suitable for quantitative discussion. In the MD calculations, the NVT ensemble (T = 1200 K) was used. Time step was 5 fs for systems including only atoms and 1 fs for those including In, C, and H atoms. This time step was verified in model calculations as shown in Figures S1–S4. To determine the cell size, we performed the geometry and cell size optimization of bulk indium. We took 5 × 5 × 2 unit cell with (110) surface and the cell size became 16.227 Å × 16.227 Å × 25.000 Å. We performed MD calculations with this setup and a temperature of 1200 K. The calculated mass density of indium liquid reasonably agreed with the experimental data (see more details in ref (29)).

4.3.11. Blue Moon Ensemble Method

To evaluate the Helmholtz activation free energy, the Blue moon ensemble method49 was applied. For constraining the bond distance, the SHAKE algorithm was used. For the cleavage and formation of C–H of methane, C–C of ethane, H–H of hydrogen molecule, the C–H, C–C, and H–H bond length was used for the reaction coordinate, respectively. For the ER type C–C bond formation, the C–C bond length between CH3 species in the gas phase and in the indium surface was taken for the reaction coordinate. In this constrained MD calculation, however, hydrogen abstraction (methane formation) took place before C–C coupling, and the activation energy for this step was not obtained. For hydrogen abstraction from adsorbed CH3 by adsorbed CH3 (methane formation) via the LH mechanism, the C–H bond length was used for the reaction coordinate. For the C–In dissociation process to generate the CH3 radical in the gas phase, a simple C–In stretching caused another C–In formation in the MD calculation with elongated C–In distances. Therefore, the potential energy difference for the CH3–In cleavage was evaluated with the In14 cluster (see Figure S5). To calculate the mean force, the trajectory of first 2 ps was used for equilibration and that of next 10 ps was used for the sampling (for the convergence of temperature, see Figure S4). For the visualization and analysis of the results, the Visual MD program50 was employed.

Acknowledgments

This work was supported by JST-CREST projects (grant number JPMJCR15P4), Innovative catalysts and creation technologies for the utilization of diverse natural carbon resources, by the Japan Ministry of Education, Culture, Sports, Science and Technology (MEXT). This work was also partly supported by the MEXT project “Integrated Research Consortium on Chemical Sciences” and the Photoexcitonix Project in Hokkaido University. A part of the computations was performed at RCCS (Okazaki, Japan) and ACCMS (Kyoto University).

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.0c03827.

  • Regeneration tests, N2 adsorption isotherms, MD and DFT calculation data, and isotope tests (PDF)

The authors declare no competing financial interest.

Supplementary Material

ao0c03827_si_001.pdf (1.8MB, pdf)

References

  1. Guo Z.; Liu B.; Zhang Q.; Deng W.; Wang Y.; Yang Y. Recent Advances in Heterogeneous Selective Oxidation Catalysis for Sustainable Chemistry. Chem. Soc. Rev. 2014, 43, 3480–3524. 10.1039/c3cs60282f. [DOI] [PubMed] [Google Scholar]
  2. Horn R.; Schlögl R. Methane Activation by Heterogeneous Catalysis. Catal. Lett. 2015, 145, 23–39. 10.1007/s10562-014-1417-z. [DOI] [Google Scholar]
  3. Schwach P.; Pan X.; Bao X. Direct Conversion of Methane to Value-Added Chemicals over Heterogeneous Catalysts: Challenges and Prospects. Chem. Rev. 2017, 117, 8497–8520. 10.1021/acs.chemrev.6b00715. [DOI] [PubMed] [Google Scholar]
  4. Keller G. E.; Bhasin M. M. Synthesis of Ethylene via Oxidative Coupling of Methane: I Determination of Active Catalysts. J. Catal. 1982, 73, 9–19. 10.1016/0021-9517(82)90075-6. [DOI] [Google Scholar]
  5. Hinsen W.; Baerns M. Oxidative Coupling of Methane to C2 Hydrocarbons in the presence of Different Catalysts. Chem.-Ztg. 1983, 107, 223–226. [Google Scholar]
  6. Ito T.; Lunsford J. H. Synthesis of Ethylene and Ethane by Partial Oxidation of Methane over Lithium-Doped Magnesium Oxide. Nature 1985, 314, 721–722. 10.1038/314721b0. [DOI] [Google Scholar]
  7. Otsuka K.; Jinno K.; Morikawa A. The Catalysts Active and Selective in Oxidative Coupling of Methane. Chem. Lett. 1985, 14, 499–500. 10.1246/cl.1985.499. [DOI] [Google Scholar]
  8. Liu H. F.; Liu R. S.; Liew K. Y.; Johnson R. E.; Lunsford J. H. Partial Oxidation of Methane by Nitrous Oxide over Molybdenum on Silica. J. Am. Chem. Soc. 1984, 106, 4117–4121. 10.1021/ja00327a009. [DOI] [Google Scholar]
  9. Khan M. M.; Somorjai G. A. A Kinetic Study of Partial Oxidation of Methane with Nitrous Oxide on a Molybdena-Silica Catalyst. J. Catal. 1985, 91, 263–271. 10.1016/0021-9517(85)90340-9. [DOI] [Google Scholar]
  10. Otsuka K.; Hatano M. The Catalysts for the Synthesis of Formaldehyde by Partial Oxidation of Methane. J. Catal. 1987, 108, 252–255. 10.1016/0021-9517(87)90172-2. [DOI] [Google Scholar]
  11. Wang L.; Tao L.; Xie M.; Xu G.; Huang J.; Xu Y. Dehydrogenation and Aromatization of Methane under Non-Oxidizing Conditions. Catal. Lett. 1993, 21, 35–41. 10.1007/bf00767368. [DOI] [Google Scholar]
  12. Mochida I.; Aoyagi Y.; Fujitsu H. Pyrolysis of Methane into Higher Hydrocarbons on Carbon Fibers. Chem. Lett. 1990, 19, 1525–1526. 10.1246/cl.1990.1525. [DOI] [Google Scholar]
  13. Hirschon A. S.; Wu H.-J.; Wilson R. B.; Malhotra R. Investigation of Fullerene-Based Catalysts for Methane Activation. J. Phys. Chem. 1995, 99, 17483–17486. 10.1021/j100049a001. [DOI] [Google Scholar]
  14. Yagita H.; Ogata A.; Obuchi A.; Mizuno K.; Maeda T.; Fujimoto K. Catalytic Dehydrogenative Coupling of Methane on Active Carbon. Effect of Metal Supported on Active Carbon. Catal. Today 1996, 29, 433–436. 10.1016/0920-5861(95)00316-9. [DOI] [Google Scholar]
  15. Kurosaka T.; Matsuhashi H.; Arata K. Dehydrogenative Coupling of Methane Catalyzed by Platinum-Added Sulfated Zirconia and Characterization of the Catalyst Surface. J. Catal. 1998, 179, 28–35. 10.1006/jcat.1998.2209. [DOI] [Google Scholar]
  16. Soulivong D.; Norsic S.; Taoufik M.; Coperet C.; Thivolle-Cazat J.; Chakka S.; Basset J.-M. Non-Oxidative Coupling Reaction of Methane to Ethane and Hydrogen Catalyzed by the Silica-Supported Tantalum Hydride: (≡SiO)2Ta-H. J. Am. Chem. Soc. 2008, 130, 5044–5045. 10.1021/ja800863x. [DOI] [PubMed] [Google Scholar]
  17. Li L.; Mu X.; Liu W.; Kong X.; Fan S.; Mi Z.; Li C.-J. Thermal Non-Oxidative Aromatization of Light Alkanes Catalyzed by Gallium Nitride. Angew. Chem., Int. Ed. 2014, 53, 14106–14109. 10.1002/anie.201408754. [DOI] [PubMed] [Google Scholar]
  18. Gerceker D.; Motagamwala A. H.; Rivera-Dones K. R.; Miller J. B.; Huber G. W.; Mavrikakis M.; Dumesic J. A. Methane Conversion to Ethylene and Aromatics on PtSn Catalysts. ACS Catal. 2017, 7, 2088–2100. 10.1021/acscatal.6b02724. [DOI] [Google Scholar]
  19. Xiao Y.; Varma A. Highly Selective Nonoxidative Coupling of Methane over Pt-Bi Bimetallic Catalysts. ACS Catal. 2018, 8, 2735–2740. 10.1021/acscatal.8b00156. [DOI] [Google Scholar]
  20. Guo X.; Fang G.; Li G.; Ma H.; Fan H.; Yu L.; Ma C.; Wu X.; Deng D.; Wei M.; et al. Direct, Nonoxidative Conversion of Methane to Ethylene, Aromatics and Hydrogen. Science 2014, 344, 616–619. 10.1126/science.1253150. [DOI] [PubMed] [Google Scholar]
  21. Han S. J.; Lee S. W.; Kim H. W.; Kim S. K.; Kim Y. T. Nonoxidative Direct Conversion of Methane on Silica-Based Iron Catalysts: Effect of Catalytic Surface. ACS Catal. 2019, 9, 7984–7997. 10.1021/acscatal.9b01643. [DOI] [Google Scholar]
  22. Morejudo S. H.; Zanón R.; Escolástico S.; Yuste-Tirados I.; Malerød-Fjeld H.; Vestre P. K.; Coors W. G.; Martínez A.; Norby T.; Serra J. M.; Kjølseth C. Direct Conversion of Methane to Aromatics in a Catalytic Co-Ionic Membrane Reactor. Science 2016, 353, 563–566. 10.1126/science.aag0274. [DOI] [PubMed] [Google Scholar]
  23. Ishihara T.; Miyashita Y.; Iseda H.; Takita Y. Decomposition of Methane over Ni/SiO2 Catalysts with Membrane Reactor for the Production of Hydrogen. Chem. Lett. 1995, 24, 93–94. 10.1246/cl.1995.93. [DOI] [Google Scholar]
  24. Takenaka S.; Kobayashi S.; Ogihara H.; Otsuka K. Ni/SiO2 Catalyst Effective for Methane Decomposition into Hydrogen and Carbon Nanofiber. J. Catal. 2003, 217, 79–87. 10.1016/S0021-9517(02)00185-9. [DOI] [Google Scholar]
  25. Takenaka S.; Serizawa M.; Otsuka K. Formation of Filamentous Carbons over Supported Fe catalysts through Methane Decomposition. J. Catal. 2004, 222, 520–531. 10.1016/j.jcat.2003.11.017. [DOI] [Google Scholar]
  26. Ogihara H.; Takenaka S.; Yamanaka I.; Tanabe E.; Genseki A.; Otsuka K. Formation of Highly Concentrated Hydrogen through Methane Decomposition over Pd-Based Alloy Catalysts. J. Catal. 2006, 238, 353–360. 10.1016/j.jcat.2005.12.024. [DOI] [Google Scholar]
  27. Nishikawa Y.; Ogihara H.; Yamanaka I. Liquid-Metal Indium Catalysis for Direct Dehydrogenative Conversion of Methane to Higher Hydrocarbons. ChemistrySelect 2017, 2, 4572–4576. 10.1002/slct.201700734. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Kashaboina U.; Nishikawa Y.; Wakisaka Y.; Sirisit N.; Nagamatsu S.-i.; Bao D.; Ariga-Miwa H.; Takakusagi S.; Inami Y.; Kuriyama F.; et al. Metamorphosis-like Transformation during Activation of In/SiO2 Catalyst for Non-oxidative Coupling of Methane: In Situ X-ray Absorption Fine Structure Analysis. Chem. Lett. 2019, 48, 1145–1147. 10.1246/cl.190440. [DOI] [Google Scholar]
  29. Ohtsuka Y.; Nishikawa Y.; Ogihara H.; Yamanaka I.; Ratanasak M.; Nakayama A.; Hasegawa J.-y. Theoretical Study on the C-H Activation of Methane by Liquid-Metal Indium: Catalytic Activity of Small Indium Clusters. J. Phys. Chem. A 2019, 123, 8907–8912. 10.1021/acs.jpca.9b06374. [DOI] [PubMed] [Google Scholar]
  30. Potapova L. L.; Cherches B. K.; Egiazarov Y. G. The kinetics of the pyrolysis of methane–ethane mixtures in the presence of indium-oxide catalyst. Kinet. Catal. 1993, 34, 339–343. [Google Scholar]
  31. Baba T.; Abe Y. Metal cation-acidic proton bifunctional catalyst for methane activation: conversion of 13CH4 in the presence of ethylene over metal cations-loaded H-ZSM-5. Appl. Catal., A 2003, 250, 265–270. 10.1016/s0926-860x(03)00321-1. [DOI] [Google Scholar]
  32. Baba T.; Abe Y.; Nomoto K.; Inazu K.; Echizen T.; Ishikawa A.; Murai K. Catalytic Transformation of Methane over In-Loaded ZSM-5 Zeolite in the Presence of Ethene. J. Phys. Chem. B 2005, 109, 4263–4268. 10.1021/jp045640t. [DOI] [PubMed] [Google Scholar]
  33. Gabrienko A. A.; Arzumanov S. S.; Moroz I. B.; Prosvirin I. P.; Toktarev A. V.; Wang W.; Stepanov A. G. Methane Activation on In-Modified ZSM-5: The State of Indium in the Zeolite and Pathways of Methane Transformation to Surface Species. J. Phys. Chem. C 2014, 118, 8034–8043. 10.1021/jp501314j. [DOI] [Google Scholar]
  34. Stepanov K.; Okuda A.; Ohta K.; Matsubara H.; Okumura K.; Yamamoto K.; Itagaki R.; Suganuma S.; Tsuji E.; Katada N. Direct Methylation of Benzene with Methane Catalyzed by Co/MFI Zeolite. ChemCatChem 2018, 10, 3806–3812. 10.1002/cctc.201800724. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Maeno Z.; Yasumura S.; Liu C.; Toyao T.; Kon K.; Nakayama A.; Hasegawa J.-y.; Shimizu K.-i. Experimental and Theoretical Study of Multinuclear Indium-oxo Clusters in CHA Zeolite for CH4 Activation at Room Temperature. Phys. Chem. Chem. Phys. 2019, 21, 13415–13427. 10.1039/c9cp01873e. [DOI] [PubMed] [Google Scholar]
  36. Kashiwadate K.; Saito Y.; Miyamoto A.; Ogino Y. The Dehydrogenation of Butyl Alcohols by the Molten-metal Catalysts. Bull. Chem. Soc. Jpn. 1971, 44, 3004–3009. 10.1246/bcsj.44.3004. [DOI] [Google Scholar]
  37. Ogino Y.Catalysis and Surface Properties of Liquid Metals and Alloys; Marcel Dekker: New York, 1987. [Google Scholar]
  38. Serban M.; Lewis M. A.; Marshall C. L.; Doctor R. D. Hydrogen Production by Direct Contact Pyrolysis of Natural Gas. Energy Fuels 2003, 17, 705–713. 10.1021/ef020271q. [DOI] [Google Scholar]
  39. Upham D. C.; Agarwal V.; Khechfe A.; Snodgrass Z. R.; Gordon M. J.; Metiu H.; McFarland E. W. Catalytic molten metals for the direct conversion of methane to hydrogen and separable carbon. Science 2017, 358, 917–921. 10.1126/science.aao5023. [DOI] [PubMed] [Google Scholar]
  40. Taccardi N.; Grabau M.; Debuschewitz J.; Distaso M.; Brandl M.; Hock R.; Maier F.; Papp C.; Erhard J.; Neiss C.; et al. Gallium-rich Pd-Ga Phases as Supported Liquid Metal Catalysts. Nat. Chem. 2017, 9, 862–867. 10.1038/nchem.2822. [DOI] [PubMed] [Google Scholar]
  41. Liang S.-T.; Wang H.-Z.; Liu J. Progress, Mechanisms and Applications of Liquid-Metal Catalyst Systems. Chem.—Eur. J. 2018, 24, 17616–17626. 10.1002/chem.201801957. [DOI] [PubMed] [Google Scholar]
  42. Frisch M. J.; Trucks G. W.; Schlegel H. B.; Scuseria G. E.; Robb M. A.; Cheeseman J. R.; Scalmani G.; Barone V.; Mennucci B.; Petersson G. A.; et al. Gaussian 09, revision B.01; Gaussian, Inc.: Wallingford, CT, 2009.
  43. Chai J.-D.; Head-Gordon M. Long-Range Corrected Hybrid Density Functionals with Damped Atom-Atom Dispersion Corrections. Phys. Chem. Chem. Phys. 2008, 10, 6615–6620. 10.1039/b810189b. [DOI] [PubMed] [Google Scholar]
  44. Hehre W. J.; Ditchfield R.; Pople J. A. Self-Consistent Molecular Orbital Methods. XII. Further Extensions of Gaussian-Type Basis Sets for Use in Molecular Orbital Studies of Organic Molecules. J. Chem. Phys. 1972, 56, 2257–2261. 10.1063/1.1677527. [DOI] [Google Scholar]
  45. Hariharan P. C.; Pople J. A. The Influence of Polarization Functions on Molecular Orbital Hydrogenation Energies. Theor. Chim. Acta 1973, 28, 213–222. 10.1007/bf00533485. [DOI] [Google Scholar]
  46. Bergner A.; Dolg M.; Küchle W.; Stoll H.; Preuß H. Ab initio Energy-Adjusted Pseudopotentials for Elements of Groups 13-17. Mol. Phys. 1993, 80, 1431–1441. 10.1080/00268979300103121. [DOI] [Google Scholar]
  47. VandeVondele J.; Krack M.; Mohamed F.; Parrinello M.; Chassaing T.; Hutter J. QUICKSTEP: Fast and Accurate Density Functional Calculations Using a Mixed Gaussian and Plane Waves Approach. Comput. Phys. Commun. 2005, 167, 103–128. 10.1016/j.cpc.2004.12.014. [DOI] [Google Scholar]
  48. Perdew J. P.; Burke K.; Ernzerhof M. Generalized Gradient Approximation Made Simple. Phys. Rev. Lett. 1996, 77, 3865–3868. 10.1103/physrevlett.77.3865. [DOI] [PubMed] [Google Scholar]
  49. Carter E. A.; Ciccotti G.; Hynes J. T.; Kapral R. Constrained Reaction Coordinate Dynamics for the Simulation of rare events. Chem. Phys. Lett. 1989, 156, 472–477. 10.1016/s0009-2614(89)87314-2. [DOI] [Google Scholar]
  50. Humphrey W.; Dalke A.; Schulten K. VMD: Visual Molecular Dynamics. J. Mol. Graphics 1996, 14, 33–38. 10.1016/0263-7855(96)00018-5. [DOI] [PubMed] [Google Scholar]
  51. Holmen A.; Olsvik O.; Rokstad O. A. Pyrolysis of natural gas: chemistry and process concepts. Fuel Process. Technol. 1995, 42, 249–267. 10.1016/0378-3820(94)00109-7. [DOI] [Google Scholar]
  52. Dibeler V. H.; Mohler F. L. Mass Spectra of the Deuteromethanes. J. Res. Natl. Bur. Stand. 1950, 45, 441–444. 10.6028/jres.045.048. [DOI] [Google Scholar]
  53. Amenomiya Y.; Pottie R. F. Mass Spectra of Some Deuterated Ethanes. II. An Empirical Method of Calculation of the Spectra. Can. J. Chem. 1968, 46, 1741–1746. 10.1139/v68-289. [DOI] [Google Scholar]
  54. In our previous computational study to investigate catalytic active species for the C–H cleavege,29 MD calculations for liquid indium at 1200 K were performed to observe the surface structure. At this simulation condition, the structure of the gas/liquid interface was disrupted significantly. Transient species which resemble atomic indium and tiny indium cluster appeared and disappeared on the surface in the time course. Based on the observation, we employed Inx (x = 1–4, 14, ∞) models, and calculated activation energy and reaction energy. Interestingly, the calculated activation energy of Inx (x = 2–4) was reasonably small. In particular, In2 showed the lowest barrier at transition state (43 kcal/mol) and slight exothermicity in the reaction energy (−0.3 kcal/mol). Therefore, the disrupted surface, whose partial structure involves small indium cluster such as In2, would be responsible to the CH activation. In this manuscript, we utilize “In2” which represents the most plausible catalytically active species.
  55. The trajectory in Figure 6 is a single trajectory. To ensure the same behavior of CH3 and H groups in other trajectories, additional two MD calculations were performed. One includes two CH3 groups on the liquid indium. The result in Figure S2 shows that the two carbon atoms of the CH3 groups keep staying on the gas/liquid interface. The other MD calculation including 10 hydrogen atoms on and inside liquid indium. As Figure S3 shows, hydrogen atoms go through the liquid phase. These results illustrate different behavior of CH3 and H groups, which corroborates the result shown in Figure 6.

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ao0c03827_si_001.pdf (1.8MB, pdf)

Articles from ACS Omega are provided here courtesy of American Chemical Society

RESOURCES