Controlling Homogeneous Chemistry in ... - ACS Publications

---

5962

Ind. Eng. Chem. Res. 2009, 48, 5962–5968

Controlling Homogeneous Chemistry in Homogeneous-Heterogeneous Reactors: Application to Propane Combustion Georgios D. Stefanidis and Dionisios G. Vlachos* Department of Chemical Engineering and Center for Catalytic Science and Technology (CCST), UniVersity of Delaware, 150 Academy Street, Newark, Delaware 19716

Operation strategies for controlling the extent of homogeneous chemistry in homogeneous-heterogeneous (HH) reactors were developed for a catalytic plate microreactor using a two-dimensional computational fluid dynamics model for propane combustion on Pt as our prototype system. The effect of the reactor gap size (distance between plates) was analyzed. We found that homogeneous chemistry is sustained for gaps well below the quenching diameter as a result of enhanced catalyst-induced heating. This finding has important ramifications for catalyst lifetime and safety and could be used to produce chemicals, for example, in oxidative dehydrogenation and oxidative coupling reactions. The homogeneous chemistry contribution decreases with decreasing gap size. Catalytic chemistry alone can occur under suitable flow rates, compositions, and heat loss/heat exchange rates. The synergism or competition between homogeneous and catalytic chemistries is delineated. Introduction Homogeneous-heterogeneous (HH) reactors cover a fairly broad range of processes. Examples include chemical vapor deposition,1 combustion,2-4 partial oxidation (e.g., methane to syngas),5,6 oxidative dehydrogenation (e.g., ethane to ethylene),7-9 oxidative coupling of methane, and more recently introduced processes such as the catalytic fast pyrolysis of biomass.10 The degree of homogeneous chemistry in such systems varies with application and conditions. In general, homogeneous chemistry can be detrimental or desirable. For example, in the case of catalytic combustion, the onset of gas-phase chemistry leads to flames, with severe implications for catalyst lifetime, device failure, and safety. On the other hand, homogeneous chemistry is key when making chemicals. Although HH processes have been studied for years, tuning these reactors to render them heat or chemical machines is nontrivial. Our objective here is to provide insights into the operation of such HH reactors, with a focus on microscales. Small scales become increasingly important for process intensification, for the generation of power and chemicals for the portable and transportation sectors, and for distributed grids.11 Our specific application example is microburners, which will be at the heart of future portable and distributed power generation and compact plants producing chemicals, such as hydrogen for fuel cells.12-15 Homogeneous and catalytic combustion in microburners have each been extensively studied but usually separately.16,18-27 In catalytic microreactors, it is tacitly assumed that homogeneous chemistry is unimportant, given the large surface-to-volume ratio, i.e., the fact that the characteristic dimension is below the quenching diameter (minimum length to propagate a flame). In contrast to this general belief, flames have been found experimentally in catalyst-coated gaps of ∼1 mm.17 As we show here, the sustainability of homogeneous chemistry in these systems does not follow the expected quenching diameter ideas. The interplay of combustion modes has been a long-studied problem, starting as early as in 192729 (see ref 30 for mechanistic details on coupling mechanisms). Earlier fundamental studies of HH systems31-38 focused on large-scale systems, wires, and * To whom correspondence should be addressed. E-mail: vlachos@ udel.edu. Tel.: (302) 831-2830. Fax: (302) 831-1048.

foils, which are not easily related to operation at small scales or are not of practical interest, or the radical coupling mechanism. Karagiannidis et al. recently studied HH reactions in methane/air microcombustion via computational fluid dynamics (CFD) simulations.39 In addition, HH hydrogen combustion was studied inside a microtube by CFD40 and isothermal modeling.41 In this work, we exploit the interaction of HH chemistries in propane combustion in a catalytic plate (Pt) microreactor as a prototype HH reacting system. The effects of the reactor gap size and mixture composition as functions of the inlet velocity are studied. A comparison with homogeneous combustion alone is also made. The effect of heat losses is also discussed to implicitly account for the scale-out of such devices. Finally, general guidelines are suggested in order to manipulate homogeneous chemistry and the desirable temperature for any application. Model, Governing Equations, and Numerics The reactor consisted of two infinite parallel plates, 5 cm long, each of thickness (dw) 0.2 mm, separated by a distance d (hereafter called the gap), as shown in Figure 1. Premixed propane/air mixtures of equivalence ratio φ were fed at an inlet temperature of 300 K. The CFD code FLUENT42 was used. The 2D steady-state model consisted of the continuity, momentum, heat, and species transport equations in the gas phase and the 2D energy equation inside the wall. At the inlet, a fixed, flat velocity profile was used. Atmospheric pressure was specified at the outlet. No-slip boundary conditions were applied at the walls. Newton’s law of cooling was applied at the exterior horizontal surface with an exterior effective heat-loss coefficient h, whose nominal value was 25

Figure 1. Schematic of the computational domain (not to scale).

10.1021/ie801480m CCC: $40.75  2009 American Chemical Society Published on Web 01/15/2009

Ind. Eng. Chem. Res., Vol. 48, No. 13, 2009

5963

2

W/(m K); heat losses due to radiation from the exterior horizontal wall were lumped into this coefficient. The ambient temperature was 300 K. The axial boundaries (left and right wall surfaces) were considered insulated, consistent with previous work,16,18 because in experiments, radiative shields are usually employed (e.g., blank monoliths) in order to minimize heat losses. The wall conductivity (ks) was 15 W/(m K), representing low- to moderate-conductivity materials (e.g., stainless steel). Uniform mesh spacing was employed, with 250 nodes in the axial direction and 25 and 10 transverse nodes in the gas and the wall, respectively. (This meshing is adequate because no significant transverse temperature gradients occur in the wall.16,18) Piecewise temperature polynomials were used for the calculation of the species-specific heat capacities, which were then used to calculate the mixture heat capacity via a mixing law. The calculation of thermal conductivity, viscosity, and species mass diffusivities was based on the kinetic theory.42 The fluid density was calculated using the ideal-gas law. Similarly to previous work,18,19 we employed the Westbrook and Dryer one-step propane/air combustion kinetics for the homogeneous chemistry43 and a one-step rate expression for propane combustion on Pt.44 The latter is the product of a posteriori systematic reduction methodology of a detailed surface reaction model44 and is in good agreement with the full model, as well as with experimental data (both catalyst ignition and autothermal fixed-bed data) under fuel-lean conditions. In ref 21, this one-step rate expression was used in a pseudo-2D reactor model to simulate a reactor similar to the one described here; the coupled flow-combustion model gave satisfactory predictions of the axial temperature profiles compared to experimental data for different effective wall conductivities. Consequently, our model captures a great deal of physics. Nonetheless, because of simplifications of the gaseous chemistry model imposed by computational demand, the results should be viewed as a guide to the trends of such systems. Simulations were performed for various modes: homogeneous chemistry alone, heterogeneous chemistry alone, and combined HH chemistries. In the latter two cases, the catalyst surface area was taken to be equal to the geometric surface area. The pure heterogeneous chemistry solutions were obtained by turning off the homogeneous chemistry. To find extinction and blowout limits, natural parameter continuation was performed for each mode. Because multiplicity can occur with catalytic and combined HH branches (see, for example, ref 18), we proposed an approach within FLUENT to obtain such solutions. Specifically, HH solutions were also obtained from converged catalytic chemistry solutions by turning on the homogeneous chemistry. If the catalytic solution converges to the HH one, then only the HH solution is stable, and the catalytic solution is of numerical merit only; then mode multiplicity does not exist. For our conditions, multiple-mode solutions were not found. Results and Discussion Heat-Transfer Analysis. Here, a heat-transfer analysis and the interplay of catalytic and homogeneous chemistries are presented. Figure 2 shows temperature profiles along the symmetry plane and the channel wall for two inlet velocities for homogeneous and HH chemistries. The overall picture is the same irrespective of the combustion mode: a preheating zone is followed by a thin combustion zone and a postcombustion zone. In general, the widths of these zones depend on the operating conditions, and the zones are not always precisely distinguished. In the preheating zone, the temperature increase

Figure 2. Temperature at the symmetry plane and the wall vs reactor length for HH and homogeneous chemistries for two inlet velocities. Gap size d ) 1 mm; equivalence ratio φ ) 0.85.

in the gas mixture is milder compared to that in the combustion zone. In the combustion zone, the gas temperature rises sharply and reaches a maximum. The maximum solid temperature occurs at approximately the same axial location as the maximum gas temperature. In the postcombustion zone, both wall and gas temperatures decrease as a result of heat losses to the surroundings. The prevailing heat-transfer mode in the axial direction of the gas is convective heat transfer as time-scale analysis indicates: the convection time scale (l/uj) is on the order of milliseconds, whereas the conduction time scale (l2/ajg) on the order of seconds. Here, l is the reactor length, uj is an average velocity at the symmetry plane, and ajg is the average gas thermal diffusivity. As a result of convection of the cold incoming gases, the gas temperature in the preheating zone is lower than the wall temperature, i.e., the wall acts as a heat source. In the transverse direction, the dominant heat-transfer mechanism is conduction with a typical time scale (d2/4ajg) shorter than that of axial convection. Consequently, the combustible mixture is brought to ignition via the heat flux from the wall. After ignition, the hot flue gases and heat losses result in gas temperatures higher than the wall temperatures. Therefore, unlike in the preheating zone, the wall acts as a heat sink. Farther downstream in the combustion zone, the high transverse heat-transfer rates between the gas and the solid phase (at the microscale) result in a nearly uniform temperature field. Figure 2 shows that, in HH mode, the upstream wall temperature is higher than that for homogeneous chemistry alone because of the heat release from the catalytic reaction. This enhances the importance of the wall as a heat source for the combustible mixture. In this context, the catalytic chemistry promotes homogeneous chemistry. On the other hand, reactant depletion on the catalytic wall renders gas ignition more difficult (see ref 30 and references therein for details on the coupling of HH chemistries). In this context, the catalytic chemistry inhibits

5964

Ind. Eng. Chem. Res., Vol. 48, No. 13, 2009

Figure 3. Maximum temperature at the symmetry plane vs inlet velocity for different gap sizes for catalytic, HH, and homogeneous modes. The right panel is an enlargement at low velocities. Vertical arrows indicate (a) blowouts and (b) extinctions. Equivalence ratio φ ) 0.85.

Figure 4. Maximum wall temperature vs inlet velocity for different gap sizes for catalytic, HH, and homogeneous modes. The right panel is an enlargement at low velocities. Vertical arrows indicate (a) blowouts and (b) extinctions. Equivalence ratio φ ) 0.85.

homogeneous chemistry. These couplings are consistent with our qualitative understanding from older literature.2 In summary, within the velocity range where homogeneous chemistry alone is sustained, catalytic chemistry plays two roles: it promotes homogeneous chemistry, by heating the unignited mixture, and it inhibits homogeneous chemistry, by consuming reactants and thus lowering the peak gas temperature. The outcome of such competition is less obvious. Figure 2a shows that, for a low inlet velocity of 0.4 m/s (close to the extinction limit), inhibition dominates, and the catalyst causes a slight downstream movement of the flame, along with a considerable decrease in the flame temperature. Rather, at the “high” velocity of 1.1 m/s (Figure 2b), which is close to the blowout limit, promotion dominates, and the catalyst causes the flame to move upstream. At higher velocities, insufficient backward heat transfer through the walls causes blowout. For these conditions, the microflame remains anchored close to the entrance. Inlet Velocity and Effect of Gap Size. Figure 3a presents the maximum temperature at the symmetry plane as a function of the inlet velocity for three gap sizes. Figure 3b is an enlargement at low velocities. The solid lines represent simulation results for catalytic chemistry alone (homogeneous chemistry is turned off). The dashed lines represent HH chemistry data. The solid lines with symbols represent homogeneous chemistry alone (gaseous burners). The two ends of all curves are approximate quenching points (extinction at the left-most part and blowout at the right-most part of each curve), beyond which the burner loses stability. The extinction limits in all cases and the blowout limits for the gaseous burners were calculated

with an accuracy of 0.05 m/s. The blowout limits for the catalytic and HH chemistries, which occur at high velocities, were obtained with an accuracy of 0.5 m/s. Figures 4 and 5a,b show the corresponding wall temperatures, the contribution of homogeneous chemistry to the overall propane conversion (defined as the fraction of propane combusted via homogeneous chemistry), and the position of the maximum symmetry plane temperature, respectively, for the HH mode. Comparison of the HH and homogeneous modes in Figure 3 reveals that extinction occurs at lower inlet velocities in catalytic burners than in gaseous burners. In other words, catalytic burners exhibit slightly better stability toward extinction. This is in stark contrast to large-scale devices, where catalytic combustion is limited by transport and gas-phase combustion is preferred. The difference is more pronounced at lower gap sizes (d ) 0.5 mm), where catalytic chemistry becomes more important and homogeneous chemistry is more difficult to sustain (pure homogeneous combustion simulations for d ) 0.2 mm have not been done). At low inlet velocities or, equivalently, low fuel input (close to extinction; left-most part of Figure 3b), the gas temperatures are too low to ignite the combustible mixture. Therefore, only catalytic chemistry is sustainable close to extinction, and the contribution of homogeneous chemistry to conversion (Figure 5a) is negligible. The residence time is sufficiently long to lead to propane conversion above 97% (not shown) via catalysis. As the inlet velocity is increased, the fuel input increases, resulting in higher heat release at the wall, higher wall temperatures (Figure 4), and higher gas temperatures (Figure

Ind. Eng. Chem. Res., Vol. 48, No. 13, 2009

Figure 5. (a) Homogeneous chemistry contribution (%) and (b) position of maximum temperature at the symmetry plane vs inlet velocity for different gap sizes for the HH mode. Equivalence ratio φ ) 0.85.

3), which, in turn, cause ignition of the combustible mixture homogeneously. The contribution of flame chemistry increases, and the propane conversion goes to 100% for all gap sizes. Upon gaseous ignition, the maximum temperature along the symmetry plane is the maximum temperature in the burner. The contribution of homogeneous chemistry first increases with increasing inlet velocity, then reaches a maximum that depends on the gap size, and eventually decreases at high inlet velocities. This decrease is moderate, and no evidence of gaseous chemistry quenching occurs. For moderate gap sizes (∼1 mm), upon ignition of the gas-phase chemistry, the majority of the fuel is burned homogeneously (Figure 5a). The heat released via homogeneous combustion increases the wall temperature (Figure 4) over a range of velocities, a signature of gas-promoting catalytic chemistry. This promotion, however, is not as significant, given masstransfer limitations of surface processes, but indicates the inherent complexity of the two-way coupling of chemistries. Upon increasing the inlet velocity, the homogeneous chemistry ignition might move first upstream (for d ) 1 mm) because of the increased power input that enhances combustion (Figure 5b). Subsequently, it is pushed downstream as a result of the fast advection of the cold inlet gases, which prolongs the time for the heat released at the wall to raise the bulk gas to the ignition temperature, and the increased heat losses from the extended preheating/combustion zone, which should be compensated by additional fuel consumption at the catalytic wall. All things considered, the former effect dominates on the left of the maximum points in Figure 5a, where homogeneous chemistry is promoted; the latter dominates on the right of the maximum points in Figure 5a, where homogeneous chemistry is slightly inhibited. When blowout occurs (rightmost arrows in Figures 3a and 4), both chemistries die simultaneously (lack of hysteresis). What

5965

is more remarkable is that, even though homogeneous chemistry cannot be sustained without a catalyst at small scales above some critical (blowout) velocity, because of heat loss to the wall and then to the environment in conjunction with insufficient backward heat transfer through the wall, gas chemistry is, in fact, sustained in the HH mode over a much wider (high) inlet velocity range (compare the end points of the dashed and solid lines with symbols) at gaps considerably lower than the quenching diameter. What one experiences at high velocities is a catalyst-induced stabilization of the gas-phase chemistry. Although the concept of thermally stabilized catalytic combustion is not new, the efficient heat-transfer mechanism along the wall and then from the wall to the gas results in flame stabilization over a substantial velocity range. In turn, the blowout limit of the HH mode is shifted to higher velocities than that of the heterogeneous process alone; i.e., the increased conversion of the fuel via gas chemistry renders the burner more stable toward blowout than a (hypothetical purely) catalytic one. At these high velocities, the two chemistries work in synergism. The effect of gap size on flame stabilization (Figures 4 and 5a) is as expected: the smaller the gap size, the lower the contribution of the flame chemistry. The mechanism, however, is initially less clear. As the gap size is decreased, the gas-tosolid mass-transfer coefficient increases, favoring catalytic chemistry outside the kinetically controlled regime; as shown in recent work of ours,45,46 transport effects are important for propane combustion at the microscale. In parallel, the solid-togas heat-transfer rate in the preheating zone increases, which inhibits catalytic chemistry alone (through the lowering of the surface temperature in a pure catalytic mode)21 but promotes homogeneous chemistry. Smaller gaps provide less volume for gas chemistry, and for the same flow velocity, the volumetric flow rate and thus the heat input are lower. Both of these factors work against homogeneous chemistry. Finally, the increased gasto-solid heat-transfer coefficient in the postcombustion zone could affect the heat losses that are generally detrimental for flame stability.19 For example, for d ) 0.2 mm and an inlet velocity of 7 m/s, the heat-transfer resistances determining heat loss are (m2 K/W): Rgas ) (d/2)/kg ) 2.2 × 10-3, Rwall ) (dw/ ks) ) 1.3 × 10-5, and Rext ) 1/h ) 4 × 10-2. The largest resistance is the exterior one, Rext, for our exterior heat-loss coefficient, h. This means that the transverse heat transfer is relatively insensitive to the wall material and the gap size for our conditions: the increased mass-transfer coefficient with decreasing gap size dominates performance. This is indeed confirmed from the higher operating temperatures and blowout velocities that can be achieved with decreasing the gap size (Figure 4). In recent work,39 it was noted that the flame moves upstream at lower gap sizes as a result of higher surface temperatures near the inlet, attributed to the decreased radiative heat losses due to the reduction of the geometrical exchange view factors with the colder inlet. This finding is confirmed from our results in Figure 5b, which show that the position of the maximum temperature at the symmetry plane shifts upstream as the gap size decreases. In addition, the earlier work reported that the homogeneous combustion intensity decreased with increasing confinement. This is also confirmed from our simulations. In fact, smaller gap sizes promote catalytic chemistry through enhanced transverse mass transfer but also drive the flame upstream through enhanced transverse heat transfer. The latter inhibits catalytic chemistry because the fuel is consumed faster in the gas phase. Overall, the enhanced transverse mass-transfer

5966

Ind. Eng. Chem. Res., Vol. 48, No. 13, 2009

Figure 6. Maximum temperature at the symmetry plane for HH and catalytic chemistries (left axis) and homogeneous chemistry contribution for the HH mode (right axis) vs equivalence ratio. The inlet velocity is 0.5 m/s, and the gap size d is 1 mm.

rates dominate, and the intensity of homogeneous chemistry decreases with decreasing gap size (Figure 5a). Effect of Composition. The effect of mixture composition on the interplay between homogeneous and catalytic chemistries is discussed in this section. Figure 6 shows the maximum temperature at the symmetry plane as a function of the equivalence ratio (φ) for HH mode and catalytic chemistry alone, as well as the homogeneous chemistry contribution as a function of the equivalence ratio for HH mode (conditions in caption). The temperature decrease with decreasing equivalence ratio occurs not only because of the reduction in fuel content but also because of a decrease in the contribution of homogeneous chemistry to the total fuel conversion. For sufficiently fuel-lean compositions and low velocities, shown in Figure 6, very little homogeneous chemistry actually occurs. Below a critical composition (∼0.62), extinction of the catalyst follows (end left point). This equivalence ratio is in reasonable agreement with experimental data obtained under similar conditions.17 Figure 7a shows the position of the maximum temperature at the symmetry plane as a function of the inlet velocity for three equivalence ratios, as indicated. The two ends of all lines are turning points beyond which the burner quenches. It is found that, at low and moderate velocities, the maximum temperature and, thus, the ignition point of the gas mixture move downstream as the equivalence ratio decreases. This is because leaner mixtures produce less heat. Consequently, a larger portion of the fuel reacts catalytically before the temperature is adequate for the gas mixture to ignite. Figure 7b,c shows the maximum temperature at the symmetry plane and the homogeneous chemistry contribution, respectively, as functions of the inlet velocity for three equivalence ratios at a fixed burner gap. Turning points (extinction and blowout), designated by vertical arrows, are observed in Figure 7b as in Figure 3. Several important remarks can be made. The lower the equivalence ratio, the lower the temperature (Figure 7b), and the lower the contribution of homogeneous chemistry (Figure 7c). For mixtures closer to the stoichiometric point, catalytic chemistry occurs near extinction (low velocities) without significant homogeneous chemistry; the latter is promoted with increasing inlet velocity (Figure 7c). The trend follows that of d ) 1 mm in Figures 3-5. Finally, Figure 7b shows that increasing the equivalence ratio results in a dramatic extension of the blowout limit toward higher velocities and only a marginal extension of the extinction limit toward lower velocities. In summary, our analysis suggests that the extent of homogeneous chemistry and, consequently, the desirable temperature can be maintained through adjustment of the inlet composition

Figure 7. (a) Position of maximum temperature at the symmetry plane, (b) maximum temperature at the symmetry plane, and (c) homogeneous chemistry contributions vs inlet velocity for three different equivalence ratios for the HH mode. The insets in b and c are enlargements at low velocities. d ) 1 mm.

and/or flow velocity. The choice of inlet composition significantly affects the stability limits of the process. Effect of Heat Losses. The effective heat-loss coefficient depends on the application. It is different in a stand-alone microreactor, in a stack of microreactors, and in a monolith and can range anywhere between nearly zero (adiabatic) and >60 W/(m2 K). In multifunctional devices, h can refer to the heat exchange from the exothermic to the endothermic channel. In Figure 8a, the maximum temperature at the symmetry plane is plotted as a function of the external effective heat-loss coefficient for the three designated inlet velocities for the HH mode. The heat-loss coefficient gradually increased from the nominal value of 25 W/(m2 K) to the extinction or blowout stability limit (designated by the arrows), which is given with an accuracy of 1 W/(m2 K). At low inlet velocities, high heat-loss coefficients cause extinction because the heat release cannot overcome the heat losses from the reactor (note the very low maximum gas temperature at the stability limit for Vin ) 0.5 m/s). At high inlet velocities, high heat-loss coefficients cause blowout because

Ind. Eng. Chem. Res., Vol. 48, No. 13, 2009

Figure 8. (a) Maximum temperature at the symmetry plane and (b) homogeneous chemistry contribution vs heat-loss coefficient for three different inlet velocities for the HH mode. Gap size d ) 1 mm; equivalence ratio φ ) 0.85.

stability is limited by the residence time and not the power input (note the high maximum gas temperature at the stability limit for Vin ) 3.0 m/s). Operation at intermediate inlet velocities (e.g., 1 m/s) results in higher stability against heat losses as compared to operation at low and high inlet velocities. This is in agreement with the bell-shaped curve shown in a relevant study of Karagiannidis et al.39 in the critical heat-loss coefficient vs inlet velocity graph therein. Figure 8b shows that homogeneous chemistry is very sensitive to the heat losses for all inlet velocities. For example, for Vin ) 0.5 m/s, an increase in the heat-loss coefficient by a factor of ∼2 causes a decrease in the homogeneous chemistry contribution from ∼60% to ∼2% at the extinction limit, where conversion has dropped to ∼90% (not shown). In conclusion, near the extinction limit essentially only catalytic chemistry occurs, and the fuel conversion is incomplete. In contrast, at high inlet velocities (e.g., 3 m/s), gas chemistry is unavoidable over the entire stable operating regime, and complete conversion is always attained (not shown). Heat loss is a crucial factor. In stand-alone microburners that have large heat-loss coefficients or in burners coupled with reformers, the possibility of flames is low. In contrast, for larger systems (e.g., large-diameter monoliths), the possibility of gas-phase chemistry is high under many operating conditions. Conclusions A 2D CFD model has been employed to study homogeneous chemistry, catalytic chemistry, and combined homogeneous-heterogeneous (HH) chemistries in parallel-plate microreactors for propane/air combustion on Pt with submillimeter gap sizes. Our focus was on providing design guidelines (microburner size and operating conditions) for

5967

different applications for which homogeneous chemistry might be desirable. The results show that, if only catalytic chemistry is desired, the microreactor should be operated at very low inlet velocities (near extinction) for all gap sizes. Combustible mixtures with compositions away from the stoichiometric point decrease the homogeneous chemistry contribution and the operating temperature but also narrow the window of operating velocities for which combustion is sustainable. Unexpectedly, homogeneous chemistry is sustained for very low channel gap sizes (as low as 200 µm studied here) because of efficient heat transfer from the catalyst (enhanced thermally stabilized catalytic combustion). The latter finding is attributed to enhanced gas-to-solid mass transport, which assists catalytic chemistry, and enhanced solid-to-gas heat transfer, which promotes homogeneous chemistry. The presence of homogeneous chemistry in catalytic microburners moderately extends the blowout limit toward higher velocities. As a result, at high velocities, there is a synergism between the HH reactions toward stabilizing operation. Heat losses have a large quantitative impact on homogeneous chemistry and reactor temperatures. At low velocities and high heat losses, only catalytic chemistry can be sustained. The contribution of homogeneous chemistry decreases with decreasing gap size, and this provides a reactor design variable for control. The overarching strategy emerging from all of these simulations is fairly simple and intuitively obvious: to minimize or avoid gas chemistry, one should keep device temperatures low. This can be achieved by varying composition (fuel-lean or -rich), flow rate (low), and heat loss/exchange rate (high). The last factor brings in the “scale” effect of the number of such microburners put together in an application. We believe that several additional topics should be studied in future work. One example is the effect of pressure (the interested reader is referred to ref 47 concerning the effect of pressure on the pure heterogeneous and the coupled HH combustion of fuel-lean propane/air mixtures over platinum in mesoscale reactors). Other interesting topics include how the choice of the fuel affects ignition of the homogeneous chemistry and the self-sustained combustion envelope through the change in reactivity and the Lewis number effect, as well as the synergism of homogeneous and heterogeneous chemistries under conditions relevant to partial oxidation. Acknowledgment This work was supported in part by the NSF (CBET0729701). This article was written in honor of the 70th birthday of D.G.V.’s inspiring colleague, Prof. T. W. Fraser Russell. Literature Cited (1) Mountziaris, T. J.; Jensen, K. F. Gas-phase and surface reaction mechanisms in MOCVD of GaAs with trimethyl-gallium and arsine. J. Electrochem. Soc. 1991, 138, 2426. (2) Pfefferle, W. C.; Pfefferle, L. D. Catalytically stabilized combustion. Prog. Energy Combust. Sci. 1986, 12, 25. (3) Pfefferle, L. D.; Pfefferle, W. C. Catalysis in combustion. Catal. ReV.-Sci. Eng. 1987, 29, 219. (4) Karim, A. M.; Federici, J. A.; Vlachos, D. G. Portable power production from methanol in an integrated thermoeletric/microreactor system. J. Power Sources 2008, 179 (1), 113. (5) Zhu, H.; Kee, R. J.; Engel, J. R.; Wickham, D. T. Catalytic partial oxidation of methane using RhSr- and Ni-substituted hexaaluminates. Proc. Combust. Inst. 2007, 31, 1965.

5968

Ind. Eng. Chem. Res., Vol. 48, No. 13, 2009

(6) Williams, K. A.; Horn, R.; Schmidt, L. D. Performance of mechanisms and reactor models for methane oxidation on Rh. AIChE J. 2007, 53 (8), 2097. (7) Yang, B.; Yuschak, T.; Mazanec, T.; Tonkovich, A. L.; Perry, S. Multi-scale modeling of microstructured reactors for the oxidative dehydrogenation of ethane to ethylene. Chem. Eng. J. 2008, 135S, S147. (8) Donsi, F.; Caputo, T.; Russo, G.; Di Benedetto, A.; Pirone, R. Modeling ethane oxy-dehydrogenation over monolithic combustion catalysts. AIChE J. 2004, 50 (9), 2233. (9) Henning, D. A.; Schmidt, L. D. Oxidative dehydrogenation of ethane at short contact times: Species and temperature profiles within and after the catalyst. Chem. Eng. Sci. 2002, 57 (14), 2615. (10) Carlson, T. R.; Vispute, T. P.; Huber, G. W. Green gasoline by catalytic fast pyrolysis of solid biomass derived compounds. ChemSusChem 2008, 1 (5), 397. (11) Vlachos, D. G. Microreaction Engineering: Processes, Detailed Design and Modeling. In Microfabricated Power Generation DeVices: Design and Technology; Barton, P. I., Mitsos, A., Eds.; Wiley-VCH: New York, 2009. (12) Holladay, J. D.; Jones, E. O.; Phelps, M.; Hu, J. Microfuel processor for use in a miniature power supply. J. Power Sources 2002, 108 (1-2), 21. (13) Deshmukh, S. R.; Vlachos, D. G. Effect of flow configuration on the operation of coupled combustor/reformer microdevices for hydrogen production. Chem. Eng. Sci. 2005, 60, 5718. (14) Kolios, G.; Gritsch, A.; Morillo, A.; Tuttlies, U.; Bernnat, J.; Opferkuch, F.; Eigenberger, G. Heat-integrated reactor concepts for catalytic reforming and automotive exhaust purification. Appl. Catal. B: EnViron. 2007, 70, 16. (15) Mills, P. L.; Quiram, D. J.; Ryley, J. F. Microreactor technology and process miniaturization for catalytic reactionssA perspective on recent developments and emerging technologies. Chem. Eng. Sci. 2007, 62 (24), 6992. (16) Norton, D. G.; Vlachos, D. G. Combustion characteristics and flame stability at the microscale: A CFD study of premixed methane/air mixtures. Chem. Eng. Sci. 2003, 58 (21), 4871. (17) Norton, D. G.; Vlachos, D. G. Hydrogen assisted self-ignition of propane/air mixtures in catalytic microburners. Proc. Combust. Inst. 2005, 30, 2473. (18) Norton, D. G.; Vlachos, D. G. A CFD study of propane/air microflame stability. Combust. Flame 2004, 138 (1-2), 97. (19) Kaisare, N. S.; Vlachos, D. G. Optimal reactor dimensions for homogeneous combustion in small channels. Catal. Today 2007, 120 (1), 96–106. (20) Kaisare, N. S.; Vlachos, D. G. Extending the region of stable homogeneous micro-combustion through forced unsteady operation. Proc. Combust. Inst. 2007, 31 (2), 3293. (21) Kaisare, N. S.; Deshmukh, S. R.; Vlachos, D. G. Stability and performance of catalytic microreactors: Simulations of propane catalytic combustion on Pt. Chem. Eng. Sci. 2008, 63 (4), 1098–1116. (22) Miesse, C. M.; Masel, R. I.; Short, M.; Shannon, M. A. Experimental observations of methane-oxygen diffusion flame structure in a submillimetre microburner. Combust. Theory Model. 2005, 9 (1), 77. (23) Leach, T. T.; Cadou, C. P.; Jackson, G. Effect of Structural Heat Conduction and Heat Loss on Combustion in Micro-channels. Combust. Theory Model. 2006, 10 (1), 85. (24) Maruta, K.; Takeda, K.; Ahn, J.; Borer, K.; Sitzki, L.; Ronney, P. D.; Deutschmann, O. Extinction limits of catalytic combustion in microchannels. Proc. Combust. Inst. 2003, 29, 957. (25) Stefanidis, G. D.; Kaisare, N. S.; Vlachos, D. G. Modeling ignition in catalytic microreactors. Chem. Eng. Technol. 2008, 31 (8), 1170. (26) Raimondeau, S.; Norton, D. G.; Vlachos, D. G.; Masel, R. I. Modeling of high temperature microburners. Proc. Combust. Inst. 2003, 29, 901. (27) Ronney, P. D. Analysis of non-adiabatic heat-recirculating combustors. Combust. Flame 2003, 135 (4), 421.

(28) Ju, Y.; Choi, C. W. An analysis of sub-limit flame dynamics using opposite propagating flames in mesoscale channels. Combust. Flame 2003, 133, 483. (29) Coward, H. F.; Guest, P. G. Ignition of natural gas-air mixtures by heated metal bars. J. Am. Chem. Soc. 1927, 49, 2479. (30) Bui, P. A.; Vlachos, D. G.; Westmoreland, P. R. Homogeneous ignition of hydrogen/air mixtures over platinum. In Proceedings of the Twenty-Sixth Symposium (International) on Combustion; The Combustion Institute: Pittsburgh, PA, 1996; p 1763. (31) Griffin, T. A.; Pfefferle, L. D. Gas phase and catalytic ignition of methane and ethane in air over platinum. AIChE J. 1990, 36, 861. (32) Pfefferle, L. D.; Griffin, T. A.; Winter, M.; Crosley, D. R.; Dyer, M. J. The influence of catalytic activity on the ignition of boundary layer flows. Part I: Hydroxyl radical measurements. Combust. Flame 1989, 76, 325. (33) Williams, W. R.; Zhao, J.; Schmidt, L. D. Ignition and extinction of surface and homogeneous oxidation of NH3 and CH4. AIChE J. 1991, 37, 641. (34) Song, X.; Williams, W. R.; Schmidt, L. D.; Aris, R. Bifurcation behavior in homogeneous-heterogeneous combustion: II. Computations for stagnation-point flow. Combust. Flame 1991, 84, 292. (35) Dupont, V.; Zhang, S. H.; Williams, A. High-temperature catalytic combustion and its inhibition of gas-phase ignition. Energy Fuels 2002, 16 (6), 1576. (36) Reinke, M.; Mantzaras, J.; Bombach, R.; Schenker, S.; Inauen, A. Gas phase chemistry in catalytic combustion of methane/air mixtures over platinum at pressures of 1 to 16 bar. Combust. Flame 2005, 141 (4), 448. (37) Reinke, M.; Mantzaras, J.; Schaeren, R.; Bombach, R.; Inauen, A.; Schenker, S. Homogeneous ignition of CH4/air and H2O and CO2-diluted CH4/O(2) mixtures over Pt: An experimental and numerical investigation at pressures up to 16 bar. Proc. Combust. Inst. 2005, 30, 2519. (38) Goralski, C. T.; Schmidt, L. D. Modeling heterogeneous and homogeneous reactions in the high-temperature catalytic combustion of methane. Chem. Eng. Sci. 1999, 54 (24), 5791. (39) Karagiannidis, S.; Mantzaras, J.; Jackson, G.; Boulouchos, K. Hetero-/homogeneous combustion and stability maps in methane-fueled catalytic microreactors. Proc. Combust. Inst. 2007, 31 (2), 3309. (40) Chen, G. B.; Chen, C. P.; Wu, C. Y.; Chao, Y. C. Effects of catalytic walls on hydrogen/air combustion inside a micro-tube. Appl. Catal. A: Gen. 2007, 332 (1), 89. (41) Chattopadhyay, S.; Veser, G. Heterogeneous-homogeneous interactions in catalytic microchannel reactors. AIChE J. 2006, 52 (6), 2217. (42) FLUENT; Fluent Inc.: Lebanon, NH, 2004. (43) Westbrook, C. K.; Dryer, F. L. Simplified reaction mechanisms for the oxidation of hydrocarbon fuels in flames. Combust. Sci. Technol. 1981, 27, 31. (44) Deshmukh, S. R.; Vlachos, D. G. A reduced mechanism for methane and one-step rate expressions for fuel-lean catalytic combustion of small alkanes on noble metals. Combust. Flame 2007, 149 (4), 366. (45) Stefanidis, G. D.; Vlachos, D. G. Millisecond methane steam reforming via process and catalyst intensification. Chem. Eng. Technol. 2008, 31 (8), 1201. (46) Stefanidis, G. D.; Kaisare, N. S.; Maestri, M.; Vlachos, D. G. Methane steam reforming at microscales: Operation strategies for variable power output at millisecond contact times. AIChE J. 2009, 55 (1), 181. (47) Karagiannidis, S.; Mantzaras, J.; Bombach, R.; Schenker, S.; Boulouchos, K. Experimental and numerical investigation of the hetero-/ homogeneous combustion of lean propane/air mixtures over platinum. Proc. Combust. Inst. 2009, 32, 1947–1955.

ReceiVed for reView October 1, 2008 ReVised manuscript receiVed December 15, 2008 Accepted December 16, 2008 IE801480M