A set of standardized and realistic aftertreatment catalyst test protocols have been developed by the Advanced Combustion and Emission Control Technical Team in support of the U.S. DRIVE Partnership. The protocols are intended to accelerate the pace of aftertreatment catalyst innovation by enabling the accurate evaluation and comparison of aftertreatment catalyst performance data from various testing and research facilities to maximize the impact of discovery-phase research occurring across the nation. The protocols address a need identified by the Partnership’s industry partners for consistent and accurate metrics for aftertreatment catalyst evaluation and comparison. The protocols consist of a set of standardized requirements and test procedures that sufficiently capture the performance capability of a catalyst technology in a manner that is adaptable in various laboratories. The protocols provide a detailed description of the necessary reactor system, the steps for achieving a desired aged state of the catalyst, all necessary sample pretreatments to be performed prior to testing, and realistic test conditions for evaluating performance. This article details four low-temperature catalyst test protocols applicable to (1) oxidation catalysts, (2) passive storage (and release) catalysts, (3) three-way catalysts, and (4) NH 3 -SCR catalysts. The catalyst test protocol descriptions are presented in five (5) sections: protocol general guidelines and the four individual catalyst test protocol descriptions. The general guidelines plus the individual protocol description forms the complete low-temperature catalyst test protocol for the application.

Combined NOx storage and reduction (NSR) and selective catalytic reduction (SCR) were conducted in a bench flow reactor comprising a Pt/Rh/BaO/Al2O3 lean NOx trap (LNT) catalyst and either Fe-ZSM-5 or Cu-SSZ-13 SCR catalyst. Simulated exhaust gas containing C3H6 or CO reductant was used to evaluate catalyst performance in terms of cycle-averaged NOx conversion and product selectivities over a range of feed concentrations and temperatures, and gas hourly space velocities (GHSV). Instantaneous concentration and temperature measurements reveal strong coupling of mass and energy between the LNT and SCR. The data show the importance of NH3 generation by the LNT for effective use of the downstream SCR. Operating conditions are identified that maximize utilization of the downstream SCR. A non-NH3 SCR pathway is found to be significant for Cu-SSZ-13 at intermediate temperature and high space velocity. The post-LNT NH3 to NOx ratio (ANR) correlates with the incremental NOx conversion achieved for Fe-ZSM-5, noting that ANR similar to 1 favors NH3-based standard SCR at high conversion. A reactor system productivity metric for N-2 generation (space-time yield analog) is used to rank-order the LNT + SCR systems and together with the propylene conversion are used to identify conditions giving the best performance. The combination of an overall GHSV of 135 k h(-1) and 350 degrees C feed temperature generates a SCR feed from the LNT that results in the highest incremental cycle-averaged NOx conversion in the SCR (similar to 30% with Cu-SSZ-13) and high cycle-averaged conversions of NOx (similar to 90%) and propylene (similar to 80%). The Cu-SSZ-13 out-performs the Fe-ZSM-5 catalyst under most conditions. (C) 2015 Elsevier B.V. All rights reserved.

Combined NOx storage and reduction (NSR) and selective catalytic reduction (SCR) were conducted in a sequential reactor system containing a Pt/Rh/BaO/Al2O3 Lean NOx Trap (LNT) catalyst and Cu-SSZ-13 SCR catalyst. Spatially-resolved mass spectrometry (SpaciMS) was used to construct temporal concentration profiles spanning the two monolith catalysts. The effects of feed gas temperature, gas hourly space velocity (GHSV) and carrier gas water were examined with propylene as the reductant. The working concept of the sequential LNT + SCR is evident in both the transient and cycle-averaged concentration profiles. During the rich phase NH3 is generated in the upstream LNT and trapped in the downstream SCR where it reacts with NOx that slips from the LNT during the subsequent lean phase. The instantaneous profiles provide insight into the storage and reduction dynamics and the mass coupling between the LNT and SCR catalysts. Axial gradients in the NOx storage and release during the lean and rich phases confirm classical LNT cyclic behavior. The spatio-temporal temperature measurements reveal a large exotherm caused by the propylene oxidation, manifested as a propagating temperature front. The cycle-averaged concentration profiles help to pinpoint the LNT length that gives a product mixture having a NH3/NOx ratio approaching unity, the desired stoichiometry for promoting NOx reduction in the SCR. The generation of NH3 and conversion of NOx is enhanced by water, suggesting an important role of the water gas shift chemistry. Propylene consumption and breakthrough from the LNT reveals its role in contributing to the overall NOx reduction. A non-NH3 SCR reaction pathway is identified that has an increasing contribution down the length of Cu-SSZ-13 SCR catalyst. The generation of formaldehyde over the Cu-SSZ-13 SCR catalyst suggests a pathway resulting from breakthrough of propylene from the LNT, followed by its oxidation to acrolein and followed in turn by reverse aldol condensation. (C) 2016 Elsevier B.V. All rights reserved.

Lean NO (NO + NO2) reduction was carried out using rapid periodic injection of C3H6 over a NO storage and reduction (NSR) monolith catalyst containing Pt/Rh/BaO/CeO2/Al2O3. The effects of injection rate, feed temperature, rich phase composition, lean phase duration, and feed concentration of CO2 were systematically varied to quantify their effects on the cycle-averaged NO and propylene conversions, and product selectivities. A factor of 10 increase in the injection frequency from conventional NSR cycling frequency of 0.014 to 0.14 Hz resulted in much higher NOx conversion at high feed temperatures (above 300 degrees C). Both NO and propylene conversions were higher over the entire range of feed temperatures (150-400 degrees C). High frequency injection with propylene resulted in a nearly constant catalyst temperature, in contrast to large swings in the catalyst temperature during lower frequency injection. NO conversion exceeding 90% was achieved for a feed having a cycle-averaged stoichiometric number (lean to rich ratio) of 6; slower injection required a stoichiometric ratio of 4 to achieve the same NO conversion. The same high frequency operation using H-2 as the reductant not only did not show any enhancement, but resulted in decreased NO conversion. A prolonged approach to the cyclic steady state was observed during high frequency operation. Moreover, the detrimental effect of CO2 on NO conversion was observed to decrease at higher frequencies. These observations collectively suggest that the generation of reactive intermediate species "HCxNyOz", whose lifetime on the catalyst surface exceeds the cycle duration and which reacts with NO to produce N-2, are required to achieve conversion enhancement during high frequency operation. The degree of mixing of the rich and lean feeds upstream of the catalyst was found to be an important reactor design parameter that invites further study. (C) 2014 Elsevier B.V. All rights reserved.

The steady-state and transient features of methane oxidation on a Pt/Pd/Al2O3 washcoated monolith are reported for a wide range of temperatures and feed gas compositions. Above 400 degrees C the methane conversion dependence on the O-2 feed concentration exhibits a multiplicity of states. With increasing temperature the multiplicity, in the form of a clockwise hysteresis in the rich regime, expands over a broader O-2 concentration range. At the highest temperature considered (538 degrees C) extinction occurs near the rich-lean transition (O-2/CH4 = 1.9), while ignition occurs at O-2/CH4 similar to 0.6. A low conversion state exists for all O-2 concentrations exceeding the ignition concentration at this temperature. In the rich regime, multiple high conversion states are encountered, all of which produce a mixture of CO, CO2, H-2, and H2O. Only the lowest conversion branch of this group is a steady-state. Another high conversion branch is encountered for O-2 concentrations spanning a wide lean regime (2 < O-2/CH4 < 125). Spatially resolved capillary-inlet mass spectrometry (SpaciMS) measurements reveal two primary zones within the monolith. In the front zone, complete oxidation of methane to CO2 is dominant until the O-2 concentration decreases downstream below a critical value. Beyond that point both CO and H-2 are produced, revealing the emergence of methane partial oxidation, steam reforming, and water gas shift (WGS) reactions. Transmission electron microscopy and energy dispersive spectroscopy reveal particles having a distribution of sizes and compositions, Pt/Pd ratios ranging from 10 (Pt-rich) to 0.2 (Pd-rich), and particles having an average size of similar to 15-20 nm. Accordingly, the conversion trends and spatiotemporal data are interpreted with a mechanistic model accounting for contributions by Pt, Pd, and PdO phases. For example, methane conversion in the lean regime is likely dominated by an active PdO phase, while the rich oxidation behavior is strongly affected by the oxygen self-poisoning on metallic Pd and Pt sites. The low conversion branch is attributed to inhibition by O-2 adsorption occupying the metallic catalyst sites, although the state is unstable in the lean regime as the more active PdO phase forms. The multiple high conversion state branches in the rich regime and the associated slow transient approach to a single stable high conversion branch are attributed to the supply of oxygen from the underlying bulk of PdO which favors methane partial oxidation. (C) 2014 Elsevier B.V. All rights reserved.

SpaciMS was employed to understand the factors influencing the selectivity of NOx reduction in two fully formulated LNT catalysts, both degreened and thermally aged. Both catalysts contained Pt, Rh, BaO and Al2O3, while one of them also contained La-stabilized CeO2. The amount of reductant required to fully regenerate each catalyst was first determined experimentally based on the OSC of the catalyst and the NOx storage capacity (NSC). In this way a correction was made for the change in catalyst OSC and NSC after aging, thereby eliminating these as factors which could affect catalyst selectivity to NH3. For both catalysts, aging resulted in an elongation of the NOx storage–reduction (NSR) zone due to a decrease in the concentration of NOx storage sites per unit catalyst length. In addition to decreased lean phase NOx storage efficiency, stretching of the NSR zone affected catalyst regeneration. Three main effects were identified, the first being an increase of the NOx “puff” that appeared during the onset of the rich front as it traversed the catalyst. Spatially, NOx release tracked the NSR zone, with the result that the NOx concentration peaked closer to the rear of the aged catalysts. Hence the probability that NOx could re-adsorb downstream of the reduction front and subsequently undergo reduction by NH3 (formed in the reduction front) was diminished, resulting in higher rich phase NOx slip. Second, the stretching of the NSR zone resulted in increased selectivity to NH3 due to the fact that less catalyst (corresponding to the OSC-only zone downstream of the NSR zone) was available to consume NH3 by either the NH3-NOx SCR reaction or the NH3-O2 reaction. Third, the loss of OSC and NOx storage sites, along with the decreased rate of NOx diffusion to Pt/Rh sites (as a result of Pt/Rh–Ba phase segregation), led to an increase in the rate of propagation of the reductant front after aging. This in turn resulted in increased H2:NOx ratios at the Pt/Rh sites and consequently increased selectivity to NH3.

The effect of regeneration conditions on NH3 formation in a fully formulated Pt–Rh/BaO/Al2O3 lean NOx trap catalyst was investigated. Experiments were performed on a bench flow reactor under simulated diesel exhaust conditions, employing NOx storage/reduction cycles. Using CO/H2 as the reductant, the selectivity of NOx reduction to NH3 increased with increasing regeneration time, reductant concentration and space velocity, and decreased with increasing amount of stored NOx and increasing temperature. At a given temperature the effect of these parameters on NH3 selectivity can be interpreted in terms of the local H2:NOx ratio at the precious metal sites and the extent to which NH3 is consumed in the reductant front as it propagates through the catalyst. However, selectivity to NH3 increased with increasing temperature (>300°C) during rich purging using C3H6 as the reducing agent. It was shown that NH3 selectivity was governed by the steam reforming activity of the catalyst, selectivity to NH3 increasing with increasing H2 generation. Experiments using a second catalyst to which ceria had been added as an OSC material confirmed these trends, although the presence of the ceria resulted in lower selectivity to NH3 when using H2 and/or CO as the reductant. After aging, the catalysts displayed increased selectivity to NH3; this is attributed in part to lengthening of the NOx storage-reduction zone, as demonstrated by SpaciMS data, and decreased OSC, resulting in decreased NH3 consumption by NOx and O2 downstream of the reductant front.

In order to elucidate the effect of washcoat composition on lean NOx trap (LNT) aging characteristics, fully formulated monolithic LNT catalysts containing varying amounts of Pt, Rh and BaO were subjected to accelerated aging on a bench reactor. Subsequent catalyst evaluation revealed that in all cases aging resulted in deterioration of the NOx conversion as a consequence of impaired NOx storage and NOx reduction functions, while increased selectivity to NH3 was observed in the temperature range 250–450°C. Elemental analysis, H2 chemisorption and TEM data revealed two main changes which account for the degradation in LNT performance. First, residual sulfur in the catalysts, associated with the Ba phase, decreased catalyst NOx storage capacity. Second, sintering of the precious metals in the washcoat occurred, resulting in decreased contact between the Pt and Ba, and hence in less efficient NOx spillover from Pt to Ba during NOx adsorption, as well as decreased rates of reductant spillover from Pt to Ba and reverse NOx spillover during catalyst regeneration. For the aged catalysts, halving the Pt loading from 100 to 50g/ft3 was found to result in a significant decrease in overall NOx conversion, while for catalysts with the same 100g/ft3 Pt loading, increasing the relative amount of Pt on the NOx storage components (BaO and La-stabilized CeO2), as opposed to an Al2O3 support material (where it was co-located with Rh), was found to be beneficial. The effect of Rh loading on aged catalyst performance was found to be marginal within the range studied (10–20g/ft3), as was the effect of BaO loading in the range 30–45g/L.

In order to study the effect of washcoat composition on lean NOx trap (LNT) aging characteristics, fully formulated monolithic LNT catalysts containing varying amounts of La-stabilized CeO2 (5wt% La2O3) or CeO2-ZrO2 (Ce:Zr=70:30) were subjected to accelerated aging on a bench reactor. Subsequent catalyst evaluation revealed that aging resulted in deterioration of the NOx storage, NOx release and NOx reduction functions, whereas the observation of lean phase NO2 slip for all of the aged catalysts indicated that LNT performance was not limited by the kinetics of NO oxidation. After aging, all of the catalysts showed increased selectivity to NH3 in the temperature range 250–450°C. TEM, H2 chemisorption, XPS and elemental analysis data revealed two main changes which can explain the degradation in LNT performance. First, residual sulfur in the catalysts, present as BaSO4, decreased catalyst NOx storage capacity. Second, sintering of the precious metals in the washcoat was observed, which can be expected to decrease the rate of NOx reduction. Additionally, sintering is hypothesized to result in segregation of the precious metal and Ba phases, resulting in less efficient NOx spillover from Pt to Ba during NOx adsorption, as well as decreased rates of reductant spillover from Pt to Ba and reverse NOx spillover during catalyst regeneration. Spectacular improvement in LNT durability was observed for catalysts containing CeO2 or CeO2-ZrO2 relative to their non-ceria containing analog. This was attributed to (i) the ability of ceria to participate in NOx storage/reduction as a supplement to the main Ba NOx storage component; (ii) the fact that Pt and CeO2(-ZrO2) are not subject to phase segregation; and (iii) the ability of ceria to trap sulfur, resulting in decreased sulfur accumulation on the Ba component.