Catalytic Partial Oxidation of Methane in Supercritical Water: A

May 12, 2017 - Center for Biofuel and Biochemical Research, Chemical Engineering Department, Universiti Teknologi PETRONAS,. Bandar Seri Iskandar ...
0 downloads 0 Views 932KB Size
Subscriber access provided by University of Florida | Smathers Libraries

Article

Catalytic Partial Oxidation of Methane in Supercritical Water: A Domain in CH4/H2O – O2/CH4 Parameter Space that shows Significant Methane Coupling Muzamil Abdalla Hassan, and Masaharu Komiyama Ind. Eng. Chem. Res., Just Accepted Manuscript • Publication Date (Web): 12 May 2017 Downloaded from http://pubs.acs.org on May 15, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Industrial & Engineering Chemistry Research is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

Catalytic Partial Oxidation of Methane in Supercritical Water: A Domain in CH4/H2O – O2/CH4 Parameter Space that shows Significant Methane Coupling Muzamil A. Hassan† and Masaharu Komiyama∗,‡ Clean Energy Research Center, University of Yamanashi, 4-3-11 Takeda, Kofu 400-8511, Japan Center for Biofuel and Biochemical Research, Chemical Engineering Department, Universiti Teknologi PETRONAS, 32610 Bandar Seri Iskandar, Perak Darul Ridzuan, Malaysia

Abstract Partial oxidation of methane under supercritical water environment at 658 K and 26 MPa using a batch reactor was examined in the presence of metal oxide catalysts and H2O2 as a molecular oxygen source. Within a parameter space of feed concentrations (CH4/H2O and O2/CH4 ratios), there existed a domain that gave high C2+ hydrocarbon yield reaching to ca. 4 %. Obtained products were a mixture of hydrocarbons up to C4 with ethane as a major component, accompanied with CO, CO2, H2 and methanol. The domain that gave high hydrocarbon yield did not coincide with that which gave high methanol yield. Catalyst survey indicated that transition metal oxides are effective for the production of C2+ hydrocarbons under the present conditions. In view of the oxidative coupling of methane in all reaction modes explored previously, presently reported level of hydrocarbon yield is unprecedented at this low reaction temperature.



To whom correspondence should be addressed. E-mail: [email protected]. University of Yamanashi ‡ University of Yamanashi and Universiti Teknologi PETRONAS †

1

ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Keywords: Methane, oxidative coupling, supercritical water, hydrocarbon production, transition metal oxide catalysts, Fe2O3.

1. Introduction Partial oxidation of methane (POM), including oxidative coupling of methane (OCM), to produce oxygenates and/or higher (C2+) hydrocarbons has gained attention past few decades.1,2 Direct POM apparently possesses process advantages compared to indirect methanol synthesis or Fischer-Tropsch hydrocarbon synthesis, both of which goes through energy-intensive synthesis gas (CO + H2) production route.3 Notwithstanding, direct POM is a challenging reaction, due to its required high reaction temperature and high probabilities for secondary reactions, resulting in a tendency toward complete oxidation of reactant methane and its products to CO2. The best hydrocarbon yield obtained so far is ca. 26 %, with O2 or air as an oxidant and Mn/Na2WO4/SiO2 as a catalyst at the temperature range exceeding 1100 K.4 In order to mitigate the complete oxidation in the POM process, by lowering the reaction temperature and diluting the reactant-product mixture, the use of supercritical water (SCW) as a reaction media has been attempted.5-11 Table 1 summarizes the reaction conditions examined for SCW-POM so far. Abraham and co-workers5,6 tested Cr2O3 and MnO2 catalysts at a temperature range of 673–748 K using air or O2 as an oxidant, and obtained up to 4 % yield of methanol in a batch reactor,5 and 1.69 % with a flow reactor.6 Oxygenates like methanol and formic acid were the only SCW-POM products identified in their product analyses, with no mention on higher (C2+) hydrocarbon production. Hirth and Franck7 performed homogeneous (non-catalytic) SCW-POM in a flow reactor at temperatures of 653 and 713 K using O2 as an oxidant and reported a methanol yield of ca. 22%, with a minor amount of formaldehyde and no 2

ACS Paragon Plus Environment

Page 2 of 23

Page 3 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

C2+ hydrocarbons in the products. Using a batch reactor, Savage et al.8 designed 32 experiments to determine the influence of five different variables (temperature, time, water density, CH4/H2O ratio and CH4/O2 ratio) on non-catalytic SCW-POM. The highest methanol yield observed was 0.7 %, with no mention on the C2+ hydrocarbons analysis. Lee and Foster9 used a laminar flow reactor to examine non-catalytic SCW-POM with O2 as an oxidant and reported that CO, CO2, H2 and methanol were the main products. The highest methanol yield obtained was approximately 1 %, with no mention on the C2+ hydrocarbons production. More recently, Bröll et al.10 examined non-catalytic and catalytic SCW-POM in a flow reactor using H2O2 as a molecular oxygen source and found that the yield of methanol and formaldehyde to be below 1 %. Higher hydrocarbon analysis was not mentioned, except that they confirmed the production of no ethylene (thus using C2H4 as an added internal standard for gas analysis). Sato et al.11 analyzed the density effect on non-catalytic SCW-POM with a flow reactor. The main products were CO, methanol, formaldehyde and a small amount of CO2 and hydrogen. The highest methanol and formaldehyde yields were 0.50% and 0.88%, respectively. Higher hydrocarbon analysis was not mentioned. Two points became apparent from the above short review on the preceding SCW-POM studies: (1) It may not be obvious in the above discussions, but as Fig. 1 which plots the feed concentrations listed in the literature reveals, initial CH4/H2O and O2/CH4 feed ratios examined were limited in the parameter space of either low CH4/H2O (< 0.01) or low O2/CH4 (< 0.1) or both. This may be due to the experimental difficulties inherent to each reactor setup employed, such as available cylinder pressures in case of batch reactions, or low CH4 or O2 solubility in water in a flow system in which those reactants were fed into the system dissolved in water. (2) Maybe due to this 3

ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

limited feed concentration parameter space explored, as it will become apparent later in this paper, almost all the previous studies did not consider the formation of higher hydrocarbons in their analytical procedures,5,6,8,9,11 or, when they did, arrived to the conclusion that C2+ hydrocarbon production (OCM) is not at all significant.7,10 In the present paper, catalytic SCW-POM using a batch reactor was performed, within a parameter space of feed concentrations (CH4/H2O and O2/CH4 ratios) that has not been explored previously. A domain that indicates high OCM activity was found, with unpreceded C2+ hydrocarbon yield of ca. 4 % at a low temperature of 658 K. This domain did not coincide with that which gave high methanol yield (up to 0.8 %). Preliminary catalyst search indicated that first transition metal oxides are effective components for the production of C2+ hydrocarbons under the present conditions, while the catalysts did not appear to accelerate methanol production at the same reaction conditions.

2. Experimental 2.1.

Equipment and Reaction Procedures

Reaction experiments were conducted using a stainless steel tube (SUS316, 1/2 in. O.D.) with 8.80 mL total inner volume as a batch reactor at 658 K and 26 MPa. The reaction procedure follows our previous biomass SCW gasification work.12-14 In a typical experiment, 0.1g of catalyst was added to the reactor (except for non-catalytic reactions) followed by addition of predetermined amount of distilled water and the reactor is connected to a high pressure valve. A predetermined amount of H2O2 aqueous solution (Wako Pure Chemical Industries, Ltd., 39.1 % v/v as determined by decomposing it with Fe2O3 catalyst at room temperature) used as a molecular O2 source 4

ACS Paragon Plus Environment

Page 4 of 23

Page 5 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

was charged in an external pipe that is connected outside of the high pressure valve, to avoid its decomposition by contacting the catalyst before reaction starts. The assembly is then connected to a methane gas cylinder (99.9%, Tokyo Gas Chemical Co. Ltd., gas chromatograph analyses with thermal conductivity and flame ionization detectors indicated no components other than methane to the detection limit of respective detector) and desired amount of methane was introduced to the reactor within a very short time, during which procedure H2O2 solution is also pushed into the reactor, and the high pressure valve closed. Reaction was commenced by immersing the charged reactor into a molten salt bath kept at 673 K, and a stopwatch was started to measure the reaction time. Figure 2 shows the time course of the temperature inside the reactor during the heat-up process measured separately for pure water (3 mL, circles), a typical reactant mixture (0.3 mL H2O2, 1.7 mL H2O and 3.75 MPa CH4, triangles) and the same reactant mixture with Fe2O3 catalyst (squares). In the case of pure water, the temperature inside the reactor monotonously increased to 659 K in 2 min, reached 664 K asymptotically within 4 min and stayed at that temperature thereafter. With a typical reactant mixture, temperature increase in the first 0.5 min is faster than in the case of pure water, most likely due to the H2O2 decomposition to give molecular oxygen in the early stage of this heating process. The temperatures in the 0.5-3.0 min range for the actual reactant mixture is lower compared to pure water case, obviously due to the presence of methane whose thermal conductivity is almost one order of magnitude lower than that of water in the present temperature and pressure ranges. After 3.0 min the temperature approached that of pure water, with the stable temperature being 661 ±1 K. When a catalyst (0.1 g of Fe2O3) co-exist with the reactant mixture, the temperature profile is almost the same 5

ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 23

with that of the reactant mixture without catalyst up to 2.0 min, but during the next 2 min the temperature is lower than the case of reactants-only. Apparently, this could be attributed to the reaction occurring due to the presence of catalyst, and it seems that the overall reaction at this stage is endothermic. After 4 min the temperature inside the reactor approaches to that of pure water as well as that of reactant mixture only, to the asymptotic value of 658 K, indicating that major part of the reaction is completed within 4 min (for the reaction products at this reaction conditions, see Figure 4). After designated reaction time (typically 5 min from the time of reactor drop into the molten salt bath), the reactor was withdrawn from the molten salt bath and immediately quenched to room temperature under running water. Product gas was retrieved by using a gas bag and its volume was measured by the water displacement method. The reactor was then opened, and the liquid product was collected and filtered with a microfilter. The product gas and liquid were analyzed using two types of gas chromatography: one with a thermal conductivity detector (TCD) for CH4, CO2, H2, CO, and air quantification, and the other with a flame ionization detector (FID) for C2+ hydrocarbons (ethane, ethylene, propane, propylene, iso-butane and n-butane) and oxygenate

compounds

(formaldehyde,

acetaldehyde,

methanol

and

ethanol)

quantification. Conversions, yields and selectivity are all calculated on a carbon basis (except H2) using the following equations: CH4 conversion % =

Yield % =

Selectivity % =

Mol of C2 + hydrocarbons, oxygenates, CO and CO2 × 100 Total mol of C analyzed Mol of carbon in the respective product × 100 Total mol of C analyzed

Mol of carbon in each product × 100 Total mol of carbon reacted

6

ACS Paragon Plus Environment

Page 7 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

The carbon balance for the experiments listed in Fig. 4 spanned from 82 % to 132 %, with average of 108.2 % and σ = 14.0. Multiple experiments were performed for some catalysts, and their error bars are shown in the figure. The carbon balance fluctuations mainly come from the uncertainty in the charged methane pressure (typically at around 4 MPa), which was read by a Bourdon gauge on a pressure regulator that had ca. ±0.38 MPa absolute reading error. In order to avoid this error to come into the reaction analysis, conversion, yield and selectivity calculations were based on analysis after the reaction, as shown in the above equations. 2.2. Catalyst Preparation Catalysts of selected first transition metals (Ti, Cr, Mn, Fe, Cu and Zn) were prepared. The choice of the catalysts is to replicate the preceding SCW-POM studies, and is not intended for exhaustive catalyst survey. Commercial Fe2O3 and MnO2 catalysts, obtained from Süd Chemie Catalysts Japan, Inc., were used as received. The preparation of Mn/Na2WO4/SiO2 catalyst followed a description in the literature.15 Mixed metal oxides of Fe-Cr,16 Fe-Ti16 and Cu-Zn-Cr17 were also prepared following the preparation procedures described in the respective literature. Catalysts in powder form were compressed at 3 ton/cm2 and then crushed to a narrower size range (0.25-0.35 mm) to use it in granular form. Physical parameters of the catalysts employed are listed in Table 2. Metal oxide particle size was obtained from X-ray diffraction (SmartLab, Rigaku, Japan) and the BET surface areas were determined by the single point N2 adsorption method (BELCAT, Microtrack Bell, Japan).

3. Results and Discussion 7

ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

3.1 Non-catalytic SCW-POM: C2+ Hydrocarbon Production Firstly, a preliminary survey of homogeneous (non-catalytic) SCW-POM was conducted with CH4/H2O and O2/CH4 feed concentrations as variables. Two CH4/H2O ratios, 0.100 and 0.200, were arbitrary chosen, and O2/CH4 ratio was scanned from 0.003 to 0.241, into the region of the parameter space not examined previously. The results are shown in Fig. 3. To our surprise, the production of sizable amount of C2+ hydrocarbons were observed above 0.03 O2/CH4, reaching the yield of 0.46 % at the O2/CH4 ratio of 0.21 and the CH4/H2O ratio of 0.20, where the yield of C2+ hydrocarbons exceeded that of methanol for the same initial reactant mixture. In the liquid phase methanol was the main product, while methane, ethane, propane, iso-butane, H2, CO and CO2 were detected in the gas phase. The result was striking in view of the fact that no preceding works on SCW-POM described the production of C2+ hydrocarbons. It is noted that at the lower end of O2/CH4 feed ratios, where most of the previous SCW-POM study was performed, C2+ hydrocarbon yields are very low compared to that of methanol. This apparently render an argument that the presently observed C2+ hydrocarbon production is a direct consequence of higher O2/CH4 feed ratio employed compared to the preceding works. It is noted that between the O2/CH4 feed ratio of 0.03 and 0.1 methanol production shows maxima, while C2+ hydrocarbon yields seem to level off at a higher O2/CH4 feed ratio. Apparently, for methanol production there seems to be optimal CH4/H2O and O2/CH4 feed concentration combinations exist, while for C2+ hydrocarbon production it is not apparent in Fig. 3. This may indicate the independent nature of methanol production route and hydrocarbon production route to each other (see later discussions). 3.2 Effect of Catalysts on C2+ Hydrocarbon Production by SCW-POM 8

ACS Paragon Plus Environment

Page 8 of 23

Page 9 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

The above observation prompted us to conduct preliminary catalyst survey. At a particular point in CH4/H2O – O2/CH4 domain, 0.100 CH4/H2O and 0.147 O2/CH4 feed ratios, where we observed C2+ hydrocarbon yield of 0.3 % in non-catalytic SCW-POM (cf. Fig. 3), various oxide catalysts listed in Table 2 were examined, and the outcome were summarized in Fig. 4. This quick catalyst survey meant to demonstrate the effect of catalyst per se, and not to give systematic or comprehensive search for the best catalyst. The tested catalysts were chosen rather arbitrary here, and their usage was limited to 0.1 g, with no optimization on any of the reaction parameters. Nevertheless, Fig. 4 provides us with a few interesting observations. Firstly, all the metal oxide catalysts tested were active for the C2+ hydrocarbon production in SCW-POM at this reaction conditions (Fig. 4 (a)). Specifically, Mn/Na2WO4/SiO2, MnO2, Fe-Cr mixed oxide and Fe2O3 catalysts gave almost one order of magnitude higher C2+ hydrocarbon yields (black bars) compared to the homogeneous (non-catalytic) reaction (furthest left). Among the catalysts, Fe2O3 (furthest right) showed highest hydrocarbon + methanol selectivity (gray bars) among the catalysts tested here, almost twice as much compared to the non-catalytic run. It is noted that the well-known high-temperature OCM catalyst,15 Mn/Na2WO4/SiO2 (third from left), is active but not particularly effective for methane coupling under the present SCW conditions. Secondary, methanol production (hatched bars) was not accelerated, but more likely to be hindered, by the presence of these heterogeneous catalysts. This is also true with the standard methanol synthesis catalyst,17 Cu-Zn-Cr mixed oxide (second from left). Breakdown of the POM products (Fig. 4 (b)) shows that the main hydrocarbon component detected in the gas phase is ethane (dark gray bars). In case of MnO2 (forth 9

ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

from left) and Fe2O3 (furthest right) ethylene (light gray bars) comes in second, while with other catalysts C3 component (hatched and white bars) is the second-abundant hydrocarbon component. It should be stressed that the data presented in Fig. 4 are the results of preliminary catalyst testing. Catalyst optimization needs more extensive survey, including catalyst support combinations and crystallographic considerations. Among those parameters is catalyst durability, influenced by the morphological as well as time-dependent catalyst changes during reaction. Due to the preliminary nature of the present catalyst survey, two catalysts were subjected to post-reaction analysis, as found in Table 2, columns 4 and 6. These two catalysts did show metal oxide particle growth, but no structural transformation was observed. This may be due to the fact that the reaction time employed here is short (typically 5 min including the heat-up time) for any major structural transformations. Again, catalyst comparison performed here is rather preliminary, and detailed catalyst study is underway. 3.3 SCW-POM Product Mapping within the CH4/H2O – O2/CH4 Parameter Space Among the catalysts tested here, Fe2O3 exhibited high performance in terms of OCM activity, giving a high C2+ hydrocarbon yield and high hydrocarbon+methanol selectivity. Thus using the Fe2O3 catalyst, product mapping within the CH4/H2O – O2/CH4 parameter space (up to 0.45 CH4/H2O and 0.63 O2/CH4) was carried out, and compared with non-catalytic counterpart. The obtained results are illustrated in Fig. 5, left column ((a) through (e)) for Fe2O3-catalyzed and right column ((f) through (j)) for homogeneous (non-catalytic) reactions. These maps are created by using Origin 2016 software, with smoothing factor of 0.05. The software draws the contour map first by connecting all x-y data points by triangulation, then z data are linearly interpolated and 10

ACS Paragon Plus Environment

Page 10 of 23

Page 11 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

finally contour lines are connected and smoothed. Note that product yield scale for each figure in Fig. 5 is not the same, but changed from figure to figure so that details within each figure are shown more clearly. It is also noted that the discussions given below is limited to the data obtained here, viz., at the conditions of 658 K and 26 MPa. A research is ongoing for temperature and pressure optimizations along with more extensive catalyst survey. Let us first examine the C2+ hydrocarbon production map (top row of Fig. 5, (a) and (f), and note the vertical scale difference between them). In homogeneous reactions (Fig. 5 (f)), high C2+ hydrocarbon production domain is observed in the middle part of this map, indicating that higher O2/CH4 and higher CH4/H2O feed ratios are beneficial for non-catalytic SCW-OCM reaction. Comparison of Fig. 1 and Fig. 5(f) exemplifies the fact that the O2/CH4 and CH4/H2O feed ratios adopted in the previous SCW-POM studies fall into the regions where C2+ hydrocarbon yield is very low. With Fe2O3 catalyst the situation is dramatically changed as found in Fig. 5 (a): almost one order of magnitude higher C2+ hydrocarbon production domain exist in a particular region of this parameter space, specifically in between 0.05 and 0.4 O2/CH4 and 0.02 and 0.4 CH4/H2O. The maximum C2+ hydrocarbon yield observed with this catalyst (0.1 g) is 3.71 % at 0.149 O2/CH4 and 0.214 CH4/H2O. Let us turn to methanol (second row of Fig. 5, (b) and (g)). As we found out in section 3.2, the presence of catalyst did not influence the methanol yield in SCW-POM very much (note that the vertical scales are the same with these two figures). However, its production map in CH4/H2O–O2/CH4 parameter space show some distinct characteristics: (1) Methanol production is limited to lower part of these maps, or lower value of O2/CH4 below 0.3, both for catalytic and non-catalytic reactions. (2) In case of 11

ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

non-catalytic SCW-POM methanol production peaks at about 0.1 CH4/H2O and 0.1 O2/CH4, while with Fe2O3 catalyst there appears another peak at 0.3 CH4/H2O and 0.07 O2/CH4. (3) In either non-catalytic or catalytic cases, methanol production domains do not coincide with C2+ hydrocarbon production domains (cf. Fig. 5 (a) and (f)). These observations lead to the following: (1) Both in non-catalytic and catalytic cases, methanol production and C2+ hydrocarbon production may proceed separately and independently. (2) Fe2O3 catalyst is not necessarily effective in methanol production, but it does seem to create a new domain of methanol production in this parameter space. The present SCW-POM reactions also accompany partial (to CO) to complete (to CO2) oxidation of methane. These products may also be produced by steam reforming (CH4 + H2O → CO + 3H2) and water-gas shift (CO + H2O → CO2 + H2) since certain level of H2 production was observed previously9 as well as in the present work. These product maps are shown in rows 3 to 5 in Fig. 5 ((c) through (e) for catalytic and (h) through (j) for non-catalytic reactions). Close observation of these maps indicate the following: (1) Oxidation to CO and CO2 and H2 production are enhanced by the presence of Fe2O3 catalyst. (2) While those activities span in center part of the map (higher O2/CH4 and higher CH4/H2O feed ratios) in case of non-catalytic reaction (Fig. 5 (h) through (j)), with Fe2O3 catalyst the activity is limited to upper edge of the parameter space examined here. (3) For non-catalytic reactions domains active for CO, CO2 and H2 production mostly include that of C2+ hydrocarbon production, whereas in the Fe2O3-catalyzed reactions active domains for those two sets of products are different. As shown above, in case of homogeneous (non-catalytic) reactions, the region of OCM (to C2+ hydrocarbons) in this parameter space is included within the active area 12

ACS Paragon Plus Environment

Page 12 of 23

Page 13 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

of total oxidation (to CO and CO2). Thus it may be difficult to manipulate feed ratios in order to favor OCM. With Fe2O3 catalyst, on the other hand, the active domains in the parameter space for OCM and total oxidation are different, leaving the possibility that with proper choice of catalyst and reaction conditions, it may be able to obtain high yield of POM products such as C2+ hydrocarbons under supercritical water environment. Nevertheless, any level of total oxidation depletes CH4 in the reactor and adversely affect the equilibrium for OCM.

Efforts to mitigate total oxidation in catalytic

SCW-POM may be necessary. 3.4 Mechanistic Considerations It may be premature at this stage to discuss mechanistic aspects of the present finding. Nevertheless, a few discussions may be worthwhile so as to provide working hypothesis for undergoing catalyst search and optimization. For catalytic gas-phase OCM, numbers of mechanistic studies have been reported.1 One of those works,18 which may be of relevance to the present study since it emphasizes the importance of H2O presence in the reaction system for OCM over Mn/Na2WO4/SiO2 catalysts, states that C-H bond activation pathways is mediated by either oxygen species on surfaces or by OH radicals formed via H2O/O2 equilibration on catalyst surfaces. Through kinetic and isotopic experiments, they found that the latter mechanism prevails when H2O is present in the system. The formation of OH radicals has also been reported on basic oxides such as La2O3 and Nd2O3.19 Takanabe and Iglasia18 claim that any prevalent H2O-mediated pathways require O2-derived O* species on catalytic surfaces for OH formation. Whether the mechanism prevalent on gas-phase OCM over Mn/Na2WO4/SiO2 catalysts applies to the present SCW system remains to be seen. Nevertheless, when the 13

ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 23

present reaction was run without added oxygen (except the air remained in the head space of the reactor amounting to ca. 500 ppm of O2), Fe2O3 catalyst was converted to Fe3O4 during the reaction, indicating that SCW with CH4 without added oxygen is reductive toward Fe2O3 catalyst. On the other hand, all the reactions presented here with added oxygen did not bring about major transformation of Fe2O3 catalyst, as found by X-ray diffraction (cf. Table 2). Incidentally, the run without added oxygen converted 0.7 % of CH4 (3.4 times as much of non-catalytic no-O2 run), 54.5 % of which was methanol (non-catalytic run gave 30.3 %).

4. Conclusions Partial oxidation of methane (POM) was performed under supercritical water (SCW) environment at 658 K and 26 MPa using a batch reactor. Exploration in a CH4/H2O – O2/CH4 parameter space that have not been examined before found a domain where significant amount of C2+ hydrocarbons are produced alongside methanol. Up to C4 hydrocarbons were produced with ethane as the major component. The heterogeneous oxide

catalysts

significantly

increased

C2 +

hydrocarbon

selectivity,

giving

unprecedented yield of ca. 4 % with Fe2O3 catalyst at a very low temperature of 658 K. Methanol production was not affected with the presence of the catalysts tested. This is the first report on the production of C2+ hydrocarbons in POM under SCW conditions. Further catalyst search and reaction condition optimizations are in progress.

Acknowledgement The authors are grateful to Prof. A. Khaleel of United Arab Emirates University for his assistance in preparing Fe-Cr and Fe-Ti mixed oxide catalysts.

14

ACS Paragon Plus Environment

Page 15 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

References (1) For instance, Ivars, F.; López Nieto, J. M. Oxidation Catalysts, Ch. 24: Light Alkanes Oxidation: Targets Reached and Current Challenges; Imperial College Press: London, 2014. (2) Keller, G. E.; Bhasin, M. M. Synthesis of ethylene via oxidative coupling of methane: I. Determination of active catalysts. J. Catal. 1982, 73, 9. (3) Lunsford, J. H. Catalytic conversion of methane to more useful chemicals and fuels: a challenge for the 21st century. Catal. Today 2000, 63, 165. (4) Wang, D.; Rosynek, M. P.; Lunsford, J. H. Oxidative Coupling of Methane over Oxide-Supported Sodium-Manganese Catalysts. J. Catal. 1995, 155, 390. (5) Dixon, C. N.; Abraham, M. A. Conversion of methane to methanol by catalytic supercritical water oxidation. J. Supercrit. Fluids, 1992, 5, 269. (6) Aki, S. N. V. K.; Abraham, M. A. Catalytic partial oxidation of methane in supercritical water. J. Supercrit. Fluids, 1994, 7, 259. (7) Hirth, T.; Franck, E. U. Oxidation and hydrothermolysis of hydrocarbons in supercritical water at high pressures. Ber. Bunsenges. Phys. Chem., 1993, 97, 1091. (8) Savage, P. E.; Li, R.; Santini Jr., J. T. Methane to methanol in supercritical water. J. Supercrit. Fluids, 1994, 7, 135. (9) Lee, J. H.; Foster, N. R. Direct partial oxidation of methane to methanol in supercritical water. J. Supercrit. Fluids, 1996, 9, 99. (10) Bröll, D.; Krämer, A.; Vogel H. Heterogeneously catalyzed partial oxidation of methane in supercritical water. Chem. Eng. Technol., 2003, 26, 733. (11) Sato, T.; Watanabe, M.; Smith, R. L.; Adschiri, T.; Arai, K. Analysis of the density effect on partial oxidation of methane in supercritical water, J. Supercrit. Fluids 2004, 28, 69. (12) Nguyen, H. T.; Yoda, E.; Komiyama, M. Catalytic supercritical water gasification of proteinaceous biomass: Catalyst performances in gasification of ethanol fermentation stillage with batch and flow reactors. Chem. Eng. Sci. 2014, 109, 197. (13) Hara, T.; Nguyen, H. T.; Komiyama, M. Facile and Green Decomposition of Dioxane with Catalytic Supercritical Water Gasification. Chem. Lett. 2014, 43, 1628. (14) Tiong, L.; Komiyama, M.; Uemura, Y.; Nguyen, T. T. Catalytic supercritical water gasification of microalgae: Comparison of Chlorella vulgaris and Scenedesmus 15

ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 23

quadricauda. J. Supercrit. Fluids 2016, 107, 408. (15) Pak, S.; Qiu, P.; Lunsford, J. H. Elementary reactions in the oxidative coupling of methane over Mn/Na2WO4/SiO2 and Mn/Na2WO4/MgO catalysts. J. Catal. 1998, 79, 222. (16) Khaleel, A.; Al-Marzouqi A. Alkoxide-free sol-gel synthesis of aerogel Iron-Chromium mixed oxides with unique textural properties, Mater. Lett., 2012, 68, 385. (17) Ogino, Y.; Oba, M.; Uchida, H. Study on Zinc oxide-Chromium oxide catalyst. III. Promoting effect of Chromium oxide on catalytic activities for methanol synthesis and decomposition. Bull. Chem. Soc. Jpn. 1959, 32, 284. (18) Takanabe, K; Iglesia, E. Mechanistic aspects and reaction pathways for oxidative coupling of methane on Mn/Na2WO4/SiO2 catalysts, J. Phys. Chem. C 2009, 113, 10131. (19) Anderson, L. C.; Xu, M.; Mooney, C. E.; Rosynek, M. P.; Lunsford, J. H. Hydroxyl radical formation during the reaction of oxygen with methane or water over basic lanthanide oxide catalysts, J. Am. Chem. Soc. 1993, 115, 6322. Table 1. Summary of preceding work for partial oxidation of methane in supercritical water (SCW-POM) Catalyst

Initial O2/CH4

Initial CH4/H2O

0.07–8.7

T (K)

P (MPa)

Residence time

Mode of Ref. reaction

723

11–44

5–40 min

batch

5

673, 748

24

13–31 s

flow

6

653, 713

30–100

2.7–283 min

flow

7

Cr2O3 Cr2O3/Al2O3, MnO2/CeO None

0.005, 0.024 0.025–0.27

0.005–0.036 0.0002, 0.001 0.43

None

0.04–0.1

0.05–0.27

622–754

NA

1–9 min

batch

8

None Ag/Al2O3, Ag, Inconel, Cu, Ag/Cu, Au/Ag None

0.031–0.053

0.003–0.007

673–723

25

7–40 s

flow

9

0.03–0.5

0.001–0.04

648–773

22–35

1–20 s

flow

10

0.03

0.0054

673

20–35

29, 30 s

flow

11

NA: not available

Table 2. Physical characteristics of the catalysts employed Metal oxide crystallite size/ nm Catalyst Mn2O3

BET surface area/ m2g-1

Metal loading –

Fresh

Spent

Fresh

Spent

7.7



139.7



16

ACS Paragon Plus Environment

Page 17 of 23

Fe2O3



2.7

21.2

Cu-Zn-Cr mixed oxide

Cu1.6-Zn10-Cr4

16.5 (Cr2O3)



3.3



Mn/Na2WO4/SiO2

Mn 2 wt %

14.1

21.4

11.7



Fe-Cr mixed oxide

Fe90-Cr10

22.9 (Fe2O3)



41.1



Fe-Ti mixed oxide

Fe90-Ti10





103.0



113.9

58.1

0.8 0.6 O2/CH4

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

0.4 0.2 0.0 0

0.1

0.2

0.3

CH4/H2O Figure 1. Feed gas compositions employed in the preceding SCW-POM studies.5-11 One point from Ref. 5 (CH4/H2O = 0.0356, O2/CH4 = 8.724) is not listed here in order to show details of other data distribution, but the discussion given in the text applies to the unlisted datum point, too.

17

ACS Paragon Plus Environment

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Temperature inside the reactor / K

Industrial & Engineering Chemistry Research

Page 18 of 23

673 573 H2O Actual reactants Actual reactants + 0.1g cat.

473 373 273 0

2

4

6

8

10

12

Immersion time / min Figure 2. Time course of temperature change within the reactor during the heat-up and reaction process in a molten salt bath kept at 673 K, measured for pure water (3 mL, circles), a typical reactant mixture (0.3 mL H2O2, 1.7 mL H2O and 3.75 MPa CH4, triangles) and the same reactant mixture with 0.1 g of Fe2O3 catalyst (squares).

18

ACS Paragon Plus Environment

Page 19 of 23

1.0% 0.8% Methanol

0.6% Yield

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

0.4% 0.2% C2+ hydrocarbons 0.0% 0.00

0.10

0.20

0.30

O2/CH4 molar ratio Figure 3. Oxygen concentration effect on non-catalytic SCW-POM at CH4/H2O ratios of 0.100 (solid squares) and 0.200 (solid circles). Solid line: methanol yield, dotted line: C2+ hydrocarbon yield. Reaction temperature: 658 K, reaction pressure: 26 MPa, reaction time: 5 min.

19

ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research

Yiled (%)

4%

HC Yield MeOH Yield HC+MeOH Selectivity CH4 onversion

(a)

25% 20%

3%

15%

2%

10%

1%

5%

0%

0%

3%

2%

Methanol Ethylene Propylene n-Butane

Ethane Propane iso-Butane

Conversion and Selectivity (%)

5%

Yield (%)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 23

(b)

1%

0%

Figure 4. Non-catalytic and catalytic SCW-POM: (a) methane conversion, methanol yield, and C2+ hydrocarbon yield and selectivity, (b) POM product distribution (CO, CO2 and H2 excluded). Feed ratio: CH4/H2O = 0.100 and O2/CH4 = 0.147, reaction temperature: 658 K, reaction pressure: 26 MPa, reaction time: 5 min. Charged catalyst amount: 0.1 g. 20

ACS Paragon Plus Environment

Page 21 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

Fe2O3-catalyzed

non-catalytic

(a)

(f)

(b)

(g)

(c)

(h)

(d)

(i)

(e)

(j)

Figure 5. Product mapping within a CH4/H2O – O2/CH4 parameter space. Left 21 column: Fe2O3-catalyzed, right column: homogeneous (non-catalytic). Vertical scale for each figure is varied so that details within each figure are shown more clearly. ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

22

ACS Paragon Plus Environment

Page 22 of 23

Page 23 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

82x44mm (300 x 300 DPI)

ACS Paragon Plus Environment