Direct Conversion of Methane into Methanol and Formaldehyde in an RF Plasma Environment II : Effects of Experimental Parameters

To provide information on the conversion of methane in a non-catalyzed radio-frequency (RF) plasma system into valuable chemicals, such as HCHO, CH3OH, C2H6, C2H4 and C2H2, an experiment was conducted to convert methane directly into methanol and formaldehyde. The effects of experimental parameters—specific energy (power or flow variation) and CH4 and O2 feeding concentrations—were examined. Carbon-based by-products generated in a CH4/O2/Ar RF plasma system included CO, CO2, HCHO, CH3OH, C2H6, C2H4 and C2H2. The methane conversion ratio increased as the feeding concentration of O2 and the specific energy was increased, but decreased with feeding of CH4. Increasing power in the RF plasma system did not favor the partial oxidation of CH4 toward CH3OH and HCHO, but did favor the production of C2-hydrocarbons (C2H6, C2H4 and C2H2), CO, and CO2. The CH4 feeding concentration of under 15% or an O2 content of under 10% favored the formation of CH3OH and HCHO. CO concentration decreased as the feeding concentration of CH4 increased, and increased as the feeding concentration of O2 increased. The yield of CH3OH was less than 1%. No carbon black or deposition was observed. Further research will seek to increase the yield of CH3OH and HCHO by adding catalysts to the system. Corresponding author. Tel: +886-3-2654904 ; Fax: +886-3-2654949 E-mail address: yfwang@cycu.edu.tw Wang et al., Aerosol and Air Quality Research, Vol. 5, No. 2, pp. 211-224, 2005 212


INTRODUCTION
Due to the cost and risk of producing, handling, shipping, and storing hazardous chemicals the chemical industry has avoided using them as raw materials in chemical synthesis.Some alternatives to their use also are major generators of environmentally harmful waste, which raises serious safety concerns (Shreiber et al., 1996(Shreiber et al., , 1999)).A new approach to the problem of using hazardous and reactive chemicals is in-situ generation; moreover, C 2 -hydrocarbon by-products (C 2 H 6 , C 2 H 4 and C 2 H 2 ) are industrially important and are useable as raw materials or as fuels (Okumoto and Mizuno, 2001;Okazaki et al., 1997).Regarded as a highly desirable fuel and alternative chemical feedstock, natural gas supplies more energy per CO 2 molecule created than oil does; thus helping to mitigate global warming, particularly when the methane in natural gas is converted directly into methanol (Lange, 1997).Among the carbon-based by-products generated in this study-CO, CO 2 , HCHO, CH 3 OH, C 2 H 6 , C 2 H 4 and C 2 H 2 -CH 3 OH, which is a promising clean fuel for the future, and syngas (CO+H 2 ), as a feedstock, can be converted into a liquid fuel by the Fischer-Tropsch process, or converted to methane, methanol, ammonia and other chemicals.HCHO is a versatile intermediate and a basic building block of several chemicals with various end uses, including wood products, coatings and resins.
Traditionally, the production of methanol from methane proceeds in two steps-the formation of syngas (CO + H 2 ), then subsequent reactions, such as F-T synthesis, to form methanol (Tsang et al. 1995).
In this study the characteristics of the conversion of methane using the RF (radio frequency) plasma method were measured in order to offer a better understanding of the parameters that affect homogeneous gas-phase reaction.Those parameters include specific energy, and CH 4 and O 2 feeding concentration, all of which impact the input of total carbon converted into CH 3 OH (F CH3OH ), HCHO (F HCHO ) and C 2 H 2 + C 2 H 4 + C 2 H 6 (F C2 ).The carbon balance in all experiments was between 0.87 and 1.24.Notably, no carbon black or deposition was observed during the process.The one-step oxidation of methane into methanol, formaldehyde, ethane, ethylene, and acetylene markedly reduces the costs of the process and provides an alternative to reducing the concentration of the global warming gas, methane, in the atmosphere.

EXPERIMENTAL SECTION
The apparatus used in this experiment is depicted schematically in a preliminary study and elsewhere (Tsai et al. 2003).A diffusion pump (Pfeiffer, DUO 065 DC) was used to maintain the system pressure below 0.0013 mbar to prevent pre-experimental contamination.CH 4 , O 2 and Ar were metered using Brooks type 5850E mass flow controllers at a total flow of 100 cm 3 .min - and introduced separately into the reactor (4.5 cm I.D. × 15 cm height).The inside of the reactor was separated into two sections by a glass plate to separate the input gases, CH 4 and O 2 , before they were recombined in the r e a c t o r ' s effluent.
A plasma generator (PFG 600 RF, Fritz Huttinger Elektronik Gmbh), operating at 13.56 MHz, with a matching unit (Matchbox PFM) produced the RF plasma discharge.The reactants and final mixtures were analyzed by an on-line Fourier transform infrared (FTIR) spectrometer (Thermo Nicolet AVATRA 360), which was used to determine the amounts of CH 4 , CH 3 OH, HCHO, C 2 H 2 , C 2 H 4 , C 2 H 6 , CO and CO 2 .
A gas chromatograph (GC, HP 5890A) equipped with a thermal conductivity and flame photometric detector was also used to identify reactants and products.The accuracy of the results was also determined using a carbon mass balance.Under each experimental condition, the applied power, CH 4 , O 2 and Ar feeding concentrations and operational pressures were measured more than three times to ensure that the system remained in a steady-state.All experiments were performed by setting the operational parameters and design conditions as follows: the feeding concentrations of CH 4 ([CH 4 ] in ) and O 2 ([O 2 ] in ) were kept at 2-30% and 1-40%, respectively, with inlet [CH 4 ]/[O 2 ] ratios of 0.13-5.0 and total flow rates of 20-100 cm 3 .min - ; the RF power levels were applied within the 30-150W range at room temperature (25 o C); and the operational pressure was 13.3 mbar to generate discharge at low effluent gas temperatures, which were measured using a thermocouple at the rear of the discharge zone.

SPECIFIC ENERGY
The specific energy, measured in kWh/Nm 3 of feed gas, was varied by changing either the applied power or the flow rate.Applied power significantly affected the conversion ratio in the plasma reactor, as previous studies have demonstrated (Zhu et al. 1992;Wang et al. 2000).Sustaining the RF plasma required sufficient energy to maintain ionization, which depended on the amount of energy supplied (Rossnagel et al. 1990).Figure 1 illustrates that X CH4 increased with specific energy.The general trends in X CH4 with a specific energy were very similar at various applied powers and total flow rates.The initial reactions in the plasma reactor involved the excitation of Ar as follows: Ar + e -＝ Ar* + e - (Boenig, 1988)  (1) Ar + e -＝ Ar + + 2e - (Boenig, 1988)  (2) In the CH 4 /O 2 /Ar system, O 2 can be dissociated and excited as follows: Boenig, 1988)  ( O 2 + e -＝O 2 + + 2e - (Boenig, 1988)  (4) - (Boenig, 1988) (5) (Boenig, 1988)  (6) Moreover, CH 4 can be directly ionized, attached, or dissociated by electron-impact.
Figure 2 indicates that F CH3OH and F HCHO declined as the specified energy was increased; however, F C2 increased with specific energy.The influence on F CH3OH , F HCHO and F C2 above 12 kWh.Nm -3 -especially on F HCHO -of increasing the applied power was much stronger than that of reducing the total flow rate.
At a given specific energy, a higher applied power did not favor the partial oxidation of CH 4 to CH 3 OH and HCHO in the RF plasma system.A higher applied power inhibited the production or favored the decomposition of CH 3 OH and HCHO, as follows.
CH x + O = CO + xH (Hsieh, 1998) HCO = H + CO (Hsieh, 1998)  (28) HCO + OH = CO + H 2 O (Hsieh, 1998)  (29) HCO + O = CO + OH (Hsieh, 1998)  (30) The increment of F CO for various applied powers exceeded that for various total flow rates.However, this was not the case for F CO2 , which increased by a factor of 4.86 for specific energies from 5.0 to 33.3 kWh.Nm -3 at various total flow rates, and reached a maximum of 8.3 kWh.Nm -3 for various applied powers.The steady state was maintained at around 12% when specific energy exceeded 10.0 kWh.Nm -3 .
The formation and decomposition of CO 2 may have proceeded as follows.

CH 4 FEEDING CONCENTRATION
Figure 4 plots the effect of the CH 4 feeding concentration on the CH 4 conversion ratio.X CH4 increased by a factor of 1.20 times as applied power increased from 50 to 100 W at a 2% CH 4 feeding concentration, and increased by a factor of 2.25 as applied power increased from 50 to 100 W at a 30% CH 4 feeding concentration.The decreasing CH 4 conversion ratio was due to the fixed power supply in the system.More competition in the fixed volume reactor resulted in reducing the conversion ratio.
Furthermore, Figure 5(a) shows that F CH3OH reached its maximum, 0.92 and 0.86%, at CH 4 feeding concentrations of 10% and 20% for applied powers of 50 W and 100 W, respectively.Notably, no CH 3 OH was formed at a CH 4 feeding concentration under 4% for an applied power of 50 W.The    Wang et al., Aerosol and Air Quality Research, Vol. 5, No. 2, pp. 211-224, 2005 219 powers of 50 W and 100 W respectively before declining to 11.9 and 5.52% at a CH 4 feeding concentration of 30% at applied powers of 50 W and 100 W, respectively.Increasing the CH 4 feeding concentration resulted in the generation of more CH x radicals and led to the formation of CH 3 O and CH 3 OO, increasing the probability of the formation of CH 3 OH and HCHO.However, introducing excess CH 4 into the system favored the decomposition of CH 3 OH and HCHO: shows that F C2 reached its maximum, 1.93 and 3.75%, at CH 4 feeding concentrations of 10% and 15% at applied powers of 50 W and 100 W, respectively.
Figure 6 shows the effect of CH 4 feeding concentration on F CO and F CO2 .F CO2 decreased to a minimum 12.2% at a CH 4 feeding concentration of 8% and an applied power of 100 W. It declined from 27.8% to 10.5% as the CH 4 feeding concentration increased from 2 to 30% at an applied power of 50 W.The reaction pathway was possibly favored CH x + O 2 = CO 2 + xH and CO + OH = CO 2 + H at 100 W, this increasing F CO2 .
O 2 FEEDING CONCENTRATION Figure 8(a) shows that F CH3OH reached its maximum, 0.79 and 0.26%, at an O 2 feeding concentration of 8% at both 50 W and 100 W, respectively.Figure 8(b) shows that F HCHO reached its maximum, 11.9 and 1.84%, at an O 2 feeding concentration of 8% at both 50 W and 100 W. Interestingly, the yields of CH 3 OH and HCHO were higher at an input power of 50 W than at 100 W. A lower input power favored the formation of CH 3 OH and HCHO over the decomposition of CH 3 OH or HCHO to CO 2 , which was favored at a higher reaction temperature.Figure 8(c) shows that F C2 increased steadily from 1.20 to 1.56% as the O 2 feeding concentration increased from 1 to 8% (O 2 /CH 4 = 0.13 to 1.0); it apparently decreased to 0.18% at an O 2 feeding concentration of 40% (O 2 /CH 4 = 5.0) at an applied power of 50 W.Furthermore, the F C2 declined from 4.35 to 0.26% as the O 2 feeding concentration increased from 1 to 40% at an applied power of 100 W. The raising O radicals were unfavorable to the formation of

POSSIBLE PATHWAYS OF FORMATION OF CH 3 OH
Figure 10 proposes major formation pathways of CH 3 OH.First, RF plasma is used to generate highspeed electrons and excited species via discharge at low pressure.These energetic species impact CH 4 and O 2 to generate free radicals, positive ions, negative ions, and excited molecules or atoms, through electron impact dissociation, electron ionization-dissociation, electron attachment and Penning dissociation processes (Chapman, 1980).Then, large numbers of free radicals, such as H, CH 3 , CH 2 , CH and O, were produced with small amounts of ions.The probability that CH x radicals collide with each other to generate higher hydrocarbons was higher than that associated with the formation of CH 3 O or

CONCLUSIONS
This work demonstrates the possibility of directly converting CH 4 into CH 3 OH, HCHO, C 2 H 6 , C 2 H 4 and C 2 H 2 without adding a catalyst to the RF plasma system.Neither carbon black nor a deposit was formed during the experiment.The carbon balance in all experiments was between 0.87 and 1.24.At a given specific energy, a higher applied power did not favor the partial oxidation of CH 4 to CH 3 OH and HCHO.However, a CH 4 feeding concentration of less than 15% or an O 2 content of under 10% favored the formation of CH 3 OH and HCHO.This study also addressed the conversion of methane into valuable chemicals, such as HCHO, CH 3 OH, C 2 H 6 , C 2 H 4 and C 2 H 2 , in the non-catalyzed RF plasma system.The one-step oxidation of methane into methanol, formaldehyde, ethane, ethylene and acetylene considerably reduces the cost of the process and constitutes an alternative method to reducing the concentration of the global warming gas, methane.Increasing the pressure and yield of CH 3 OH by altering the design of the RF plasma system reactor can be explored in future research.
C2 changed only a little after 15 kWh.Nm -3 at various total flow rates.Reducing the total flow rate reduced the concentration of Ar, making the transfer of energy in the RF plasma system less inefficient, and possibly favoring the formation of CH 3 O and CH 3 OO radicals, as in CH 3 + (O, O 2 ) = (CH 3 O, CH 3 OO)-or it led directly to the formation of HCHO, as in (CH 3 , CH 2 , C 2 H 3 ) + O 2 = HCHO + (OH, O, CHO).
increasing CH 4 feeding concentration might favor the reaction of CH 4 + CH 3 O = CH 3 + CH 3 OH and led to the increasing CH 3 OH concentration; however, elevated CH x radicals from the increasing CH 4 feeding concentration might also have elevated the reaction rate of CH 3 OH + CH x = CH 2 OH + CH x+1 and reached a dynamic balance in the RF plasma system.Figure 5(b) indicates that F HCHO initially increased and reached a maximum, 18.5 and 6.28%, at a feeding concentration of CH 4 of 15% and 20% at applied

Figure 7
Figure 7 plots the influence of O 2 feeding concentration on the CH 4 conversion ratio.Increasing O radicals enhanced the reaction rate of CH 4 + O = CH 3 + OH and led to the elevation of CH 4 conversion ratio.Figure 8(a) shows that F CH3OH reached its maximum, 0.79 and 0.26%, at an O 2 feeding CH 3 OH, HCHO and C 2 compounds.This was possibly due to the increasing reaction rate of CH 3 OH + O = CH 2 OH + OH, HCHO + O = HCO + OH, and C x H y + O = C x H y-1 + OH.

Figure 9
Figure9plots the effect of O 2 feeding concentration on F CO and F CO2 .The effluent temperature at the top of the reactor was around 60°C.The selectivity shifts toward CO and CO 2 from CH 3 OH as the O 2 c o n t e n t i n c r e a s e s .T h e r e s u l t s a r e s i mi l a r t o t h o s e o f Z h o u ' s r e s e a r c h(1998), which indicated that the formation of H 2 is hard to detect when the O 2 content in the feed mixture exceeds 15%, suggesting that the oxidation of CH 4 to CO 2 and H 2 O is stronger at higher O 2 concentrations.
CH 3 OO radicals, because the separating plate isolates the excited radicals from CH 4 and O 2 inside the reactor.However, the effluent radicals recombine with each other or attach to O radicals in the rear part of the reactor, to form CH 3 O or CH 3 OO radicals and produce CH 3 OH and HCHO.Consequently, further plasma diagnoses must be performed to verify the details of the reaction mechanisms.

Figure 10 .
Figure 10.Proposed reaction pathways in the CH 4 /O 2 /Ar plasma system.