Skip to content
BY-NC-ND 3.0 license Open Access Published by De Gruyter November 29, 2012

Plasma enhanced C1-chemistry: towards greener methane conversion

Tomohiro Nozaki and Ken Okazaki

Abstract

Direct conversion of methane to methanol is considered as a promising next-generation green technology, because it would eliminate energy intense, high temperature syngas production. Before 2000, various catalysts and thermochemical reaction systems were investigated towards converting direct methane to methanol; however, one-pass yield of methanol was <5%. More recently, bioreaction and photochemical synthesis have attracted keen attention, because these processes use renewable solar energy. However, the yield and productivity are still the main issues. This paper presents a low temperature (<600 K), direct conversion of methane to methanol/syngas via non-thermal plasma technology, which enables one-pass liquid yield of 20% (sum of CH3OH, HCHO and HCOOH), with selectivity between 40 and 60%. First, it emphasizes the impact of plasma catalysis in the future sustainable energy system. Second, the principle of micro-channel plasma chemical reactor is presented, then experimental results are overviewed based on our work. Finally, concluding remarks are provided.

1 Introduction

Direct conversion of methane to methanol has been considered as an energy saving reaction pathway, compared to energy intense, high temperature syngas production [13]. Figure 1 shows the Gibbs free energy of representative reactions for methane to methanol conversion. Direct methane conversion to methanol using H2O (R1) does not occur simply because the Gibbs free energy for this reaction takes a positive and almost constant value in terms of temperature [Figure 1(A)]. Direct methane conversion (R1) is therefore thermodynamically not favored.

Figure 1 Methane conversion routes. (A) The Gibbs free energy for R1~R3 and (B) R4.

Figure 1

Methane conversion routes. (A) The Gibbs free energy for R1~R3 and (B) R4.

In contrast, direct methane oxidation to methanol is thermodynamically favored, because the Gibbs free energy takes a negative value [Figure 1(B)].

The idea of “reaction splitting”, such as R1=R2+R3, is generally taken into account. Although indirect methane to methanol conversion via syngas is widely used in industry, Figure 1(A) poses two major problems that current C1-chemistry faces. First, syngas production needs a high temperature thermal energy to occur. Second, syngas to methanol conversion is possible at a much lower temperature than syngas production: there is a large temperature mismatch between R2 and R3. According to the exergy analysis of methane conversion [4], production of high temperature thermal energy, or combustion of initial feed, is the main exergy loss process. Heat exchange between high temperature exhaust and low-temperature feed is the second largest exergy loss process.

R4 is possible without a high temperature thermal energy. However, R4 is kinetically hard to control and the results are a combination of full combustion of methane followed by ordinary methane steam/dry reforming [5]. Although a tremendous effort has been made on direct methane partial oxidation in the homogeneous gas phase reaction and over solid catalysts [69], the yield of desired products has been below economic values [10, 11]. We have been working on the direct conversion of methane to methanol using a microchemical non-thermal plasma reactor at low temperatures (<300°C) [1214]. One-pass liquid yield of 20%, with selectivity between 40 and 60%, was demonstrated experimentally. In addition to methanol, plasma catalysis produces syngas directly, with approximately 40% selectivity and H2/CO=1. We would like to emphasize that direct conversion of methane to syngas at a low temperature is also a fascinating reaction route, because it would eliminate high temperature syngas manufacturing. It is noteworthy that methane-to-syngas is thermodynamically desirable compared to methane-to-methanol [Figure 1(B)].

Generally, methane partial oxidation needs pure oxygen (not air); we propose water electrolysis to supply pure oxygen, which is possible at near room temperature (RT) operation (R5). Pure oxygen is subsequently used for plasma catalysis of direct methane to syngas (R6). Syngas is then converted to methanol using low temperature thermal energy (R7).

The reaction scheme (R5–R7) is similar to R2–R3. R1 is split into three stages (R5–R7) by “electron-driven” chemistry, enabling methane to methanol conversion without high temperature thermal energy. R5–R7 is schematically described in Figure 2. Figure 2 highlights several other advantages. First, electrolysis produces pure hydrogen, which can be used for driving fuel cell as necessary. Excess hydrogen can be used to enhance carbon capture and utilization (CCU) where CO2 is converted into syngas. The H2/CO ratio is tuned by various routes to produce the desired synthetic fuels through catalytic reforming. The electrical energy necessary for electrolysis and plasma catalysis would be supplied through a renewable energy source, where energy conversion efficiency, as well as the energy distribution grid, are rapidly growing worldwide. Renewable energy is eventually converted into chemical energy of carbon-containing liquid fuels. The main part of electrical-to-chemical energy conversion is owned by electrolysis, because R5 is an endothermic reaction. In contrast, plasma catalysis of methane partial oxidation is exothermic and the external energy supply, in principle, is minimal. In this sense, existing C1-chemistry is enhanced by plasma catalysis, which is highlighted in the dotted square in Figure 2. The energy density of liquid fuels is 100 times greater than solid state secondary batteries, thus the transport and storage capability of renewable energy is greatly improved [15].

Figure 2 Concept of plasma enhanced C1-chemistry that integrates renewable energy supply, greenhouse gas conversion (CH4, CO2), carbon capture and utilization (CCU), and a catalytic fuel conversion system (material and energy balance is not considered in this schematic diagram).

Figure 2

Concept of plasma enhanced C1-chemistry that integrates renewable energy supply, greenhouse gas conversion (CH4, CO2), carbon capture and utilization (CCU), and a catalytic fuel conversion system (material and energy balance is not considered in this schematic diagram).

Non-thermal plasma catalysis of methane to syngas (R6) at low temperatures reduces the capital and operating costs of high temperature, energy intensive syngas production. Also, this technology enables integration of renewable energy (mostly electricity) and chemical energy of CH4, H2, and carbon-containing synthetic fuels, which will eventually provide viable solutions in future energy and material use. The next section is an overview of the plasma conversion of methane to synthetic fuels, based on our work.

2 Experimental

Figure 3 shows a micro-plasma reactor used for the experiment. The reactor consists of a thin quartz tube (inner diameter 1.5 mm, outer diameter 2.7 mm) with stainless steel wire (0.2 mm) at the center. Aluminum foil was wrapped around the glass tube over a 200 mm length. A high-voltage power supply (±6 kV, 10 kHz) was connected between the wire and the foil, generating dielectric barrier discharge (DBD) in the glass tube. The micro-channel configuration is essential to enhance the heat transfer property of the reactor, which removes heat generated by methane partial oxidation efficiently. Moreover, the micro-channel suppresses flame propagation and thus full combustion of methane is avoided effectively.

Figure 3 Schematic diagram of a micro-channel non-thermal plasma reactor.

Figure 3

Schematic diagram of a micro-channel non-thermal plasma reactor.

A mixture of CH4 and O2 was fed into the reactor. The total flow rate was varied from 40 to 200 cm3/min. Distilled water was also supplied continuously at 0.1 cm3/min, in order to wash out liquid components condensed on the reactor wall. Incomplete oxidation products, such as CH3OH and HCHO, are relatively stable in an aqueous solution; however, they are readily decomposed or polymerized in the dry environment. In this sense, water injection stabilizes fragile organic compounds during plasma catalysis. Condensable components were collected by the cold trap (5°C). Organic compounds (CH3OH, CH2O, HCOOH, CH3OOH) and H2O2 were collected in the form of an aqueous solution; hereafter, this liquid is called the plasma solution. Unreacted methane and gaseous products were quantitatively analyzed by gas chromatography (see [13] for detailed information).

3 The principle of microchemical non-thermal plasma

The principle of generation of non-thermal plasma at atmospheric pressure is similar to DBD. A peculiarity of DBD is the presence of a dielectric insulator on one or both metallic electrodes, which suppresses the formation of arc discharge and a large number of filamentary micro-scale discharges are temporarily generated with 1–10 ns duration. The formation of nanosecond current pulses suppresses gas heating; a highly reactive and non-equilibrium condition is readily formed at ambient conditions. DBD has a long and proven history for ozone generation [16, 17]. More recently, this technology has been frequently used for gas cleaning, lighting devices, biomedical application and so on [17, 18]. DBD has been used for large scale water treatment at water purifying plants, indicating that this technology is capable of being used for industry scale applications. An industry scale DBD reactor consists of a number of co-axial tubular reactors, with a relatively narrow discharge gap (2–5 mm). A number of co-axial reactors are integrated to satisfy demanded gas treatment [17]. In the case of methane partial oxidation, the discharge gap must be much narrower than the conventional ozone generator, in order to satisfy micro-channel configuration; detailed analysis is beyond the present work.

Figure 4(A) shows an image of DBD without water injection. Although DBD consists of a number of filamentary discharges, a time- and space-averaged image is fairly uniform. Electron impact activation of methane is possible almost independently of the reaction temperature. Methane partial oxidation is initiated, even at RT, and full combustion of methane is avoided due to low temperature oxidation. Non-thermal plasma minimizes the temperature mismatch between the initiation reaction and the oxidative chain propagation necessary for direct methane oxidation. For this reason, atmospheric pressure non-thermal plasma [1923] and its combination with catalysts [24, 25] are highlighted as a promising new technology that potentially enables one-pass methanol yields beyond economically attractive values. However, results to date are still unsatisfying, due to the paradoxical condition between methane conversion and methanol selectivity: methanol is much more reactive than methane and methanol selectivity is inevitably low, as methane conversion increases. In contrast, syngas (CO/H2) selectivity increases continuously with methane conversion. Figures 4(B) and 4(C) show DBD with water injection. Methanol is synthesized in the gas phase and condensed onto the water film. CH3OH is separated from the reactive plasma zone before it is further decomposed. Product separation is necessary to achieve high methanol yields. Interestingly, DBD is no longer uniform, as a result of the complex interaction between the DBD and the water film.

Figure 4 Image of dielectric barrier discharge (DBD) produced in a thin glass tube. Discharge power=30 W; CH4=O2=20 cm3/min: (A) without water injection; (B) and (C) with water injection.

Figure 4

Image of dielectric barrier discharge (DBD) produced in a thin glass tube. Discharge power=30 W; CH4=O2=20 cm3/min: (A) without water injection; (B) and (C) with water injection.

4 Reaction scheme

Plasma-induced methane partial oxidation is expressed from R8 to R13 [26, 27]. CH4 is dissociated into CH3· and H· by electron impact, then CH3· recombines with O2 to produce CH3OO·. If CH3OO· abstracts hydrogen from CH4 (R10), CH3· is regenerated as a chain propagator and CH4 is spontaneously oxidized into methyl hydroperoxide (CH3OOH). CH3OOH is subsequently decomposed to yield CH3OH and HCHO (R11) [28]. Therefore, CH3OOH is the important intermediate of CH3OH, and is a good indicator of the radical chain reaction (R9, R10). Nevertheless, few reports have demonstrated the presence of CH3OOH experimentally and its role on methane partial oxidation has not been discussed.

Figure 5 shows the 1H-NMR spectrum of plasma solution synthesized at 5°C (Bruker AVANCE III 400, Yokohama, Kanagawa, Japan). CH3OH, HCHO, HCOOH, and CH3OOH are clearly recognized in Figure 5(A). HCHO is highly reactive and forms methylene glycol (HOCH2OH) and polyoxymethylene glycols [H(OCH2)nOH] in an aqueous solution [29, 30]. The reactor temperature was elevated to 300°C without external water cooling. At 300°C, faint peaks associated with CH3OOH and HOCH2OH were observed, but all other peaks almost disappeared [Figure 5(B)]. Interestingly, an intense CH3OH peak is visible at 300°C. The presence of CH3OOH is strong evidence that low temperature, non-thermal plasma catalysis of methane is a viable approach to direct methane conversion to liquid oxygenates at high yields. Atomic oxygen is also produced by electron impact (R12), which is considered as the key species initiate methane oxidation (R13). However, kinetic analysis by CHEMKIN 4.0 (Reaction Design, Chuo-ku, Tokyo, Japan) predicted that atomic oxygen is not able to abstract hydrogen from CH4 at room temperature [31]. Also, radical chain reaction (R10) is not activated when the temperature is too low.

Figure 5 1H-NMR spectrum of the plasma solution synthesized at (A) 5°C and (B) 300°C. Conditions: CH4=O2; 20 cm3/min, distilled water injection; 0.1 cm3/min. (A) peak-1 and peak-2 indicate C2H5OH and CH3OCOH, respectively.

Figure 5

1H-NMR spectrum of the plasma solution synthesized at (A) 5°C and (B) 300°C. Conditions: CH4=O2; 20 cm3/min, distilled water injection; 0.1 cm3/min. (A) peak-1 and peak-2 indicate C2H5OH and CH3OCOH, respectively.

In addition to organic components, H2O2 is produced and trapped in the plasma solution. HCHO is oxidized by H2O2 in the liquid phase (R14), producing HCOOH via an equilibrium limited reaction [32, 33]. Equilibrium composition depends not only on species concentration, but also on the pH value of the solution.

HCOOH is eventually oxidized into CO2 (R15). We speculate that HCOOH is synthesized in the liquid phase rather than the gas phase methane oxidation. CH3OOH would be further oxidized by H2O2 in the plasma solution, because the H1-NMR signal decreases over time. In addition to gas phase plasma chemistry, in-liquid oxidation of organic components by H2O2 must be well understood. Combination of an appropriate post-plasma process should increase the overall methane conversion efficiency.

5 Plasma catalysis versus thermochemical catalysis

Figure 6 shows the selectivity of liquid components, with respect to methane conversion, obtained for a wide spectrum of operating conditions [13]. The results are compared with three literatures that used thermochemical methods [68]. In conventional methods, a relatively high temperature thermal energy is added to initiate methane oxidation, while small amounts of oxygen, typically 10% at most, are introduced, so that the successive oxidation of methanol is suppressed. As a result, methane conversion beyond 10% was hardly achieved. Although methane conversion increased by increasing either the oxygen content or the reaction temperature, selectivity for useful oxygenates dropped sharply. There is a clear trade-off relationship between CH3OH selectivity and CH4 conversion. Rasmussen and Glarborg investigated methane partial oxidation numerically, assuming fuel-rich, high-pressure (30–100 bar), and relatively low-temperature conditions (550–800 K). They suggest that the highest one-pass methanol yield of 4.2% would be feasible at optimum conditions of 97.4 bar, 643.3 K, and the initial CH4/O2 ratio of 23.63 [34]. Their prediction explains well the experimental results reported so far, implying that the conventional thermochemical method may not be a promising approach for direct methane conversion to methanol at high yields. In contrast, liquid selectivity for the micro-channel plasma reactor is slightly decreased with methane conversion (●), implying that liquid product separation is effective in the given plasma reactor. Although one-pass liquid yield with 20% is obtainable, liquid selectivity was below 60% (■). Assuming post dimethyl ether (CH3OCH3) synthesis with the syngas (H2/CO~1), the overall liquid selectivity could be enhanced to 80% (■). Note, Figure 6 does not count CH3OOH as a useful product.

Figure 6 Liquid selectivity vs. methane conversion [13]. ●, sum of CH3OH, HCHO, HCOOH; ■, syngas is also counted for liquid selectivity assuming syngas to DME. Three dotted curves represent constant yield lines. Hollow symbols express cited literature; □ Bjorklund [6], ○ Feng [7], ◊ Yarlagadda [8].

Figure 6

Liquid selectivity vs. methane conversion [13]. ●, sum of CH3OH, HCHO, HCOOH; ■, syngas is also counted for liquid selectivity assuming syngas to DME. Three dotted curves represent constant yield lines. Hollow symbols express cited literature; □ Bjorklund [6], ○ Feng [7], ◊ Yarlagadda [8].

6 Energy efficiency

For simplicity, the energy efficiency for methane partial oxidation is calculated by Eq. (1):

E0 expresses the energy efficiency for ideal methane partial oxidation (R6) and ΔH represents the lower heating value of CH4, CO and H2. A small part of CH4 chemical energy is lost through partial oxidation, showing 95.6% energy efficiency. Note, E0 takes 96.0% by a higher heating value basis. In the actual case, Eq. (1) is rewritten as Eq. (2):

Here, n and m represent H2/CO and CH3OH/CO ratio, respectively: CO, H2 and CH3OH are counted for major products. Based on experimental work, assume H2/CO=0.8, CH3OH/CO=0.2, CO selectivity=50%, and CH4 conversion=40%, resulting in an energy efficiency of 45%, which is approximately half of the theoretical efficiency (E≈0.5 E0). The energy efficiency of partial oxidation increases by improving syngas (CO, H2) and CH3OH selectivity and minimizing CO2 formation. The heat released by methane partial oxidation is approximately 10 W, while 20 W discharge power was consumed. If a trace amount of plasma-generated reactive species induces a radical chain reaction that promotes hydrogen abstraction from methane efficiently, the discharge power consumption would be minimized. In other words, the energy efficiency of plasma-induced methane partial oxidation is improved.

7 Concluding remarks

Non-thermal plasma catalysis of methane is overviewed based on our work. Low temperature, direct conversion of methane to methanol and syngas has two major impacts. First, plasma-driven low temperature methane conversion would reduce dependency on a high temperature catalytic process, especially energy intense syngas manufacturing. Second, plasma catalysis enhances existing C1-chemistry, if this technology is appropriately combined. Renewable energy could be integrated into hydrogen economy, carbon capture utilization and carbon-containing liquid fuels, through plasma catalysis of methane. Liquid fuels improve the transport and storage capability of renewable energy in the form of carbon resource. Although the energy efficiency of plasma catalysis needs to be improved, low temperature plasma catalysis provides a flexible solution to the future sustainable energy and material use.


Corresponding author: Tomohiro Nozaki, Department of Mechanical Sciences and Engineering, Tokyo Institute of Technology, 2–12–1, Ookayama, Meguro, Tokyo 1528550, Japan

The authors would like to thank former postdoctoral fellow, Dr. Anil Agiral and Dr. Valentin Goujard (JSPS PD fellow) for supporting the project. This project was partially supported by KAKENHI (Innovative Areas: 22110504). The Center for Advanced Materials Analysis, Tokyo Tech, supported NMR analysis.

References

[1] Lunsford JH. Catal. Today 2000, 63, 165–174.Search in Google Scholar

[2] Havran V, Dudukovic MP, Lo CS. Ind. Eng. Chem. Res. 2011, 50, 7089–7100.Search in Google Scholar

[3] Periana RA, Taube DJ, Gamble S, Taube H, Satoh T, Fujii H. Science 1998, 280, 560–564.Search in Google Scholar

[4] Simpson AP, Lutz AE. Int. J. Hydrogen Energy 2007, 32, 4811–4820.Search in Google Scholar

[5] York APE, Xiao T, Green MLH. Top. Catal. 2003, 22, 345–358.Search in Google Scholar

[6] Bjorklund MC, Carr RW. Eng. Chem. Res. 2002, 41, 6528–6836.Search in Google Scholar

[7] Feng W, Knopf FC, Dooley KM. Energ. Fuel. 1994, 8, 815–822.Search in Google Scholar

[8] Yarlagadda PS, Morton LA, Huntera NR, Gesser HD. Ind. Eng. Chem. Res. 1998, 27, 252–256.Search in Google Scholar

[9] Danen WC, Ferris MJ, Lyman JL, Oldenborg RC, Rofer CK, Steit GE. Prepr. - Am. Chem. Soc., Div. Pet. Chem. 1991, 36, 166–167.Search in Google Scholar

[10] Casey PS, McAllister T, Foger K. Ind. Eng. Chem. Res. 1994, 33, 1120–1125.Search in Google Scholar

[11] Ohtsuka K, Wang Y. Appl. Catal. A-Gen. 2001, 222, 145–161.Search in Google Scholar

[12] Nozaki T, Hattori A, Okazaki K. Catal. Today 2004, 98, 607–616.Search in Google Scholar

[13] Nozaki T, Ağıral A, Yuzawa S, Han Gardeniers JGE, Okazaki K. Chem. Eng. J. 2011, 166, 288–293.Search in Google Scholar

[14] Nozaki T, Okazaki K. J. Jpn. Petrol. Inst. 2011, 54, 146–158.Search in Google Scholar

[15] Hou Y, Vidu R, Stroeve P. Ind. Eng. Chem. Res. 2011, 50, 8954–8964.Search in Google Scholar

[16] Kitayama J, Kuzumoto M. J. Phys. D: Appl. Phys. 1997, 30, 2453–2461.Search in Google Scholar

[17] Kogelschatz U. Plasma Chem. Plasma Process. 2003, 23, 1–46.Search in Google Scholar

[18] Kogoma M, Ed., Generation and applications of atmospheric pressure plasmas, Nova Science Pub Inc: 2011.Search in Google Scholar

[19] Bugaev SP, Kozyrev AV, Kuvshinov VA, Sochugov NS, Khryapov PA. Plasma Chem. Plasma Proc. 1998, 18, 247–261.Search in Google Scholar

[20] Zhou LM, Xue B, Kogelschatz U, Eliasson B. Plasma Chem. Plasma Proc. 1998, 18, 375–393.Search in Google Scholar

[21] Larkin DW, Lobban LL, Mallinson RG. Catal. Today 2001, 71, 199–210.Search in Google Scholar

[22] Okazaki K, Kishida T, Ogawa K, Nozaki T. Energ. Convers. Manage. 2002, 43, 1459–1468.Search in Google Scholar

[23] Aghamir FM, Matin NS, Jalili AH, Esfarayeni MH, Khodagholi MA, Ahmadi R. Plasma Sources Sci. Technol. 2004, 13, 707–711.Search in Google Scholar

[24] Chen L, Zhang X-W, Huang L, Lei L-C. Chem. Eng. Proc. 2009, 48, 1333–1340.Search in Google Scholar

[25] Indarto A, Lee H, Choi J-W, Song HK. Energy Sources: Part A 2008, 30, 1628–1636.Search in Google Scholar

[26] Oshima Y, Saito M, Koda S, Tominaga H. Chem. Lett. 1988, 203–206.Search in Google Scholar

[27] Nozaki T, Goujard V, Yuzawa S, Ağıral A, Okazaki K. J. Phys. D: Appl. Phys. 2011, 44, 274010 (6 pp.).Search in Google Scholar

[28] Nizova GV, Suss-Fink G, Shul’pin GB. Chem. Comm. 1997, 397–398.Search in Google Scholar

[29] Hahnenstein I, Hasse H, Kreiter CG, Maurer G. Ind. Eng. Chem. Res. 1994, 33, 1022–1029.Search in Google Scholar

[30] Maiwald M, Grützner T, Ströfer E, Hasse H. Anal. Bioanal. Chem. 2006, 385, 910–917.Search in Google Scholar

[31] Goujard V, Nozaki T, Yuzawa S, Ağıral A, Okazaki K. J. Phys. D: Appl. Phys. 2011, 44, 274011 (13 pp.).Search in Google Scholar

[32] Charles SN, Robert WE, Rene LM, Robert RC. Anal. Chem. 1957, 26, 1792–1797.Search in Google Scholar

[33] Bruschweiler H, Minkoff GJ. Anal. Chem. Acta 1955, 12, 186–200.Search in Google Scholar

[34] Rasmussen CL, Glarborg P. Ind. Eng. Chem. Res. 2008, 47, 6579–6588.Search in Google Scholar

Received: 2012-10-27
Accepted: 2012-10-30
Published Online: 2012-11-29
Published in Print: 2012-12-01

©2012 by Walter de Gruyter Berlin Boston

This article is distributed under the terms of the Creative Commons Attribution Non-Commercial License, which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.