Search

In many laboratory catalytic reactors light-off curves describing the exit conversion versus inlet gas temperature do not follow the same path during ramp up and ramp down and this leads to dynamic hysteresis. This dynamic hysteresis, which may occur even under ideal conditions (e.g. plug flow and negligible heat effects) is fundamentally different from the steady-state hysteresis which is due to heat effects and thermal feedback. We present expressions for the width of the dynamic hysteresis loop as a function of the ramp rate, solid to gas heat capacity ratio, space time and the heat Peclet number, for the limiting cases of pseudo-homogeneous and two phase plug flow models with negligible heat effects ( Δ T ad = 0 ). We review the conditions under which steady-state hysteresis can exist for a finite Δ T ad and demonstrate the combined impact of steady-state and dynamic hysteresis effects. We study the light-off behavior and dynamic hysteresis during the co-oxidation of CO + C 2 H 6 mixture on Pt / Al 2 O 3 . It is found that 1% of CO alone ignites at 473 K whereas 500 ppm of C 2 H 6 alone ignites at 723 K. The exotherm generated by the CO oxidation can reduce the ignition temperature of C 2 H 6 during the co-oxidation of CO and C 2 H 6 . Bifurcation diagrams of solid exit temperature versus fluid inlet temperature for different inlet compositions illustrate the simultaneous as well as separate ignitions for CO and C 2 H 6 . Each 1% increase in the CO feed mole percentage decreases C 2 H 6 ignition temperature by about 80 °C. A feed of 3% CO is required for simultaneous ignition of a mixture of CO and 500 ppm of C 2 H 6 . [ABSTRACT FROM AUTHOR]

Isothermal steady state NO x storage and reduction on Pt/BaO/Al 2 O 3 by H 2 /CO/C 3 H 6 reactant mixtures is studied using, 1 + 1 dimensional two-phase model that accounts for washcoat diffusional limitations. It predicts the storage and reduction performance over a wide range of inlet feed concentrations and temperatures. A major limitation of global kinetic models is not being able to predict accurately the performance of the catalyst with changing feed composition. Micro-kinetic models elucidating the role of individual reductants can account for such variations. The temperature dependence of the rate constants has been quantified so that the surface chemistry is consistent with experimental observations. Inclusion of washcoat diffusion enables the model to accurately predict the NO x conversion and scale-up performance of the catalyst over a wide temperature range. The sensitivity of the estimated kinetic parameters to the effective diffusivities in the washcoat is analyzed. Results indicate that the estimated micro-kinetic parameters are influenced by the washcoat pore structure as well as the properties of the catalytic and storage components. [ABSTRACT FROM AUTHOR]

Abstract: Diesel particulate filters (DPF) are regenerated by combustion of the accumulated particulate matter (PM). This can lead to high peak temperature which can damage the DPF. We present guidance about how this peak temperature depends on the DPF design and operating conditions. Simple criteria are developed for predicting the peak temperature of two limiting regeneration modes. One is the transverse regeneration mode during which the PM is consumed uniformly along the DPF and the axial temperature is close to uniform. The second is the axial regeneration mode under which a negligible temperature difference exists between the solid phase and the gas, the filtration velocity is highly non-uniform and a sharp temperature front forms during the PM combustion. Most DPFs operate in the mixed regeneration mode, and their peak temperature is bounded between those predicted by the two limiting modes. The predictions of the behavior of the two limiting models provide useful guidance and bounds on any DPF design and operating conditions which will lower the peak regeneration temperature. For example, choices that widen the width of the moving temperature front decrease the maximum regeneration temperature under stationary feed conditions. A major technological challenge in the regeneration of the ceramic cordierite filter is that a sudden decrease of the engine load, referred to as Drop to Idle (DTI), may create a transient temperature peak much higher than under either the initial or final stationary feed conditions. The results for the stationary feed conditions suggest selections that decrease the peak temperature under stationary operation will also decrease the peak DTI temperature rise. [Copyright &y& Elsevier]

The relationship between fuel penalty and precious metal loading in coupled LNTSCR systems for NOx reduction at low temperatures is determined. Simulations of adiabatic, aerobic NOx reduction determine the impact of catalyst architecture, precious metal loading, cycle time, catalyst length and the support material. These revealed that (a) high NOx conversion can be achieved for a given precious metal loading at low temperatures (510 K) with about 0.3% fuel penalty, (b) shortened cycle time and increased pulse duty reduce the overall NOx slip from the coupled catalyst and lead to significant improvement in the NOx conversion, (c) alternate arrangement of the LNT/SCR catalysts increases the NOx conversion, (d) a monotonic decreasing PGM loading in LNT leads to higher NOx conversion compared to uniform loading, (e) for a fixed catalyst volume, there is an optimal aspect ratio that maximizes the nonisothermal effect, and (f) metal based monolith supports improve NOx conversion. © 2013 American Institute of Chemical Engineers AIChE J, 59: 3421-3431, 2013 [ABSTRACT FROM AUTHOR]

Highlights: [•] A criterion is developed for obtaining uniform filtration in a DPF. [•] Formulas are presented for estimating the width and speed of temperature fronts. [•] Formulas are presented for estimating the heat-up time and pressure drop. [•] A long DPF leads to non-uniform filtration and sharp temperature front. [•] A short DPF leads to uniform filtration and temperature. [ABSTRACT FROM AUTHOR]

The particulate matter (PM) emitted by a diesel engineis collectedand then burned in a diesel particulate filter (DPF). A major technologicalchallenge in the operation of the ceramic, often cordierite, filteris that a rapid shift to idle may create a local hot region with atemperature much higher than under stationary feed conditions. Thisexcessive transient temperature rise may cause local melting or crackingof the ceramic filter. Almost all previous studies of temperatureexcursions during the DPF regeneration (combustion of the depositedPM) were of cases in which equal exhaust flow rate was fed to allthe parallel inlet channels. The diesel engine exhaust pipe is sometimesconnected to the DPF by a wide-angled cone (diffuser). This leadsto a mal-distribution of the flow rate to the inlet channels and ofthe deposited PM. Simulations revealed that following a rapid shiftto idle the highest regeneration temperature in a DPF fed by a coneexceeded that in one not fed by a cone and it may exceed the cordieriteDPF melting temperature (1200 °C). Moreover, it may generatetransient radial and axial temperature gradients several times higherthan under stationary regeneration that may crack the cordierite DPF.The increase in the temperature gradient is especially large in theaxial direction. One of the surprising findings is that the highesttemperature attained following a step change to idle is not a monotonicfunction of the initial PM loading. [ABSTRACT FROM AUTHOR]

Several experimental studies have been conducted to determinethe NOxreduction by a series of LNT (leanNOxtrap) and SCR (selective catalyticreduction) catalytic bricks. An important goal is to minimize therequired precious metal loading in the LNT while keeping the NOxemission below a specified level. We presenta mathematical model of this system using hydrogen as the reductant.Simulations are used to determine the influence of the architectureof the LNT–SCR bricks, nonuniform precious metal loading inthe LNT bricks, and the cycle time at temperatures in the range of200–350 °C. The simulations lead to the following observations:(a) Low temperature reduction is the limiting step in the optimizationof precious group metal (PGM) loading in LNT. (b) The NOxconversion increases as the number of the sequentialbricks (with total length fixed) increase and reaches an asymptoticlimit. From a practical point of view, there is little incentive inusing more than two sequential pairs. (c) Nonuniform precious metalloading of the LNT bricks results in only a minor improvement in thedeNOxperformance. (d) The cycle timehas a significant impact on the NOxconversion.In the simulated example, the NOxconversionat low temperatures is increased by about 15–20% by reducingthe cycle time by a factor of 2. (e) Even at low temperature operation,diffusional limitations in the washcoat are most likely to be importantin the LNT but not in the SCR operation. The NOxconversion and ammonia selectivity are reduced when washcoatdiffusion is dominant in the LNT. [ABSTRACT FROM AUTHOR]

Experiments were conducted to determine the impact of key operating variables (ceria loading, space velocity, cycle time, and rich pulse intensity) on the oxygen storage and release process of a Pt/Rh/CeO 2 /BaO monolithic catalyst during periodic lean (oxygen)–rich (propylene) operation. The concentrations were measured by spatially-resolved capillary inlet mass spectrometry (SpaciMS), and the temperature profile was measured by coherent optical frequency domain reflectometry (c-OFDR), providing detailed insight into the spatio-temporal features of the reaction system. The experiments revealed that the addition of ceria increased the breakthrough time of the propylene during a lean-to-rich transition due to an increased oxidation rate. Hydrogen was formed in the upstream and was consumed in the downstream section of the catalyst by reaction with ceria. Increasing the space velocity increased the upstream hydrogen formation rate and decreased its downstream oxidation rate. Increasing the lean and rich durations (while keeping their duration ratio constant) increased the oxygen uptake but resulted in propylene breakthrough and decreased the solid temperature. Increasing the rich period (at a fixed rich feed and total cycle time) allowed the catalyst to release more oxygen. A comparison of the ceria-containing catalyst to one without ceria revealed a more efficient complete oxidation for the former, while a comparison of the periodic to stationary operation revealed that less than half the propylene was converted to CO 2 by the former. [ABSTRACT FROM AUTHOR]

Highlights: [•] New method of measuring internal catalytic reactor temperature. [•] Method achieves±3mm precision and sampling frequency of 0.9Hz. [•] Method applied to transient exothermic catalytic reaction in a monolith channel. [•] Method demonstrates wrong-way temperature rise for Pt-catalyzed NH3 oxidation. [•] Temperature rise due to feed temperature decrease exceeds that due to flow increase. [ABSTRACT FROM AUTHOR]

Abstract: Lean reduction of NO x (NO and NO2) was studied using monolith-supported catalysts consisting of a layer of a metal-exchanged (Fe, Cu) zeolite (ZSM-5) selective catalytic reduction (SCR) catalyst deposited on top of a Pt/Rh/BaO/CeO2 lean NO x trap (LNT) catalyst. During periodic switching between lean and rich feeds, the LNT layer reduces NO x to N2 and NH3. The SCR layer traps the latter, resulting in additional NO x reduction. The dual-layer catalysts exhibited high N2 selectivity and low NH3 selectivity over the temperature range of 150–300°C. The NO x conversion was incomplete due to undesired side reactions in the LNT layer, such as NH3 oxidation to N2O at low temperature and NO x at high temperature. The NO x conversion and N2 selectivity of the Cu-exchanged ZSM-5 was higher than that of the Fe-exchanged ZSM-5. This was due to a higher low temperature SCR activity and a higher NH3 storage capacity on the Cu-zeolite. The dual-layer catalyst had a higher NO x conversion than the LNT catalyst below 300°C and a higher N2 selectivity over the entire temperature range when H2O and CO2 were present in the feed. The NO x storage capacity and NH3 generation increased upon addition of CeO2 to the LNT layer below 250°C. It also led to increased NH3 oxidation at high temperatures. The addition of ceria mitigated the undesirable migration of Pt between the LNT and SCR layers. Hydrothermal aging had a smaller effect on dual-layer catalysts containing ceria. [Copyright &y& Elsevier]

Abstract: Infrared measurements reveal that cyclic shifts between lean and rich (H2 and/or CO) feeds to a Pt/CeO2/Al2O3 monolith catalyst generate complex, spatio-temporal temperature features. A sharp temperature rise occurs in the upstream of the monolith shortly after the cyclic introduction of either H2/CO to a pre-oxidized catalyst or O2 to a pre-reduced catalyst. This initial upstream temperature rise following the reduction of oxygen stored on a pre-oxidized catalyst is higher than following the oxidation of either H2 or CO (or their mixture) stored on the monolith. The upstream hot zone temperature decreases with time without forming a downstream moving temperature front. The intricate transient temperature gradients are caused by a competition between the chemical and transport rate process. However, the effluent concentrations do not reflect these complex interactions. Only about 20% of the total oxygen trapped during the pre-oxidation with a 5% O2/N2 mixture at 350°C is strongly bound or chemisorbed. Most of the oxygen that reacts at high temperature is loosely-bound. The introduction of a nitrogen sweep flow between the lean and rich feeds removes a significant amount of the loosely held oxygen, leading to a much more uniform reduction. Inadequate resolution of the spatio-temporal phenomena may lead to a misinterpretation of the apparent kinetics. The spatial features of the thermal fronts of the two reductants (CO or H2) are similar. The amplitude of the hot spot of the two reactants differ due to differences in the temperature dependencies of their oxidation rates. The transient oxidation of CO/H2 mixtures reveal H2 enhanced CO oxidation is likely due to a kinetic interaction as seen in previous steady-state studies. [Copyright &y& Elsevier]

Highlights: [•] Adding ceria to LNT layer enhances the low-temperature deNO x efficiency. [•] SCR diffusion layer in a dual-layer catalyst degrades high-temperature performance. [•] Zoned LNT and SCR catalyst leads to high NO x conversion over wide temperature range. [•] Optimized zoned dual-layer catalyst enables reduction of needed PGM loading. [ABSTRACT FROM AUTHOR]