UC Berkeley

Automotive exhaust contributes to air pollution through the emission of NOx, CO, and hydrocarbons (HCs). Since the 1970s, catalytic converters, installed downstream of the engine, have significantly reduced the emissions of these harmful pollutants by converting toxic gases such as NOx to N2. However, to do so, catalytic converters need to reach a high enough temperature to efficiently convert a high fraction of the NOx in the engine exhaust, which occurs during a “cold start” of the engine. One way of handling this issue is to install a passive NOx adsorber (PNA), between the catalytic converter and engine. The purpose of the PNA is to adsorb NO at lower temperatures and release it once the catalytic converter is at a sufficiently high temperature. Pd/H-CHA has attracted attention as a possible PNA because of its ability to adsorb NO strongly and its hydrothermal stability. Despite the attention Pd/H-CHA has received, the mechanism by which NO adsorption on this adsorbent occurs has been heavily debated in the literature; most of the discussion focuses on the formation of Pd+ cations, believed to be responsible for storing NO and releasing at appropriately high temperatures. A second issue has centered on the interpretation of peaks observed in infrared spectra of adsorbed NO and their assignment to NO adsorbed on Pd+ vs Pd2+. The roles of CO and ethene in the stabilization of NO adsorption have also been a subject of debate.This dissertation is focused on uncovering the mechanism of NO adsorption on Pd-based zeolites, specifically Pd/H-CHA. The emphasis is on investigating the influence of the various components of exhaust gas (i.e. water, CO, ethene, air, and NO) to see how they affect the NO release temperatures and the infrared spectrum of adsorbed NO. From observing the impact of co-adsorbates, insights into the adsorption mechanism were obtained and used to propose elementary processes by which NO adsorbs and desorbs from Pd/H-CHA. This understanding is essential for determining the effectiveness of this adsorbent as a PNA material in terms of its ability to adsorb NO and desorb it at temperatures above 573 K Chapter 1 gives an overview of the literature concerning Pd/H-CHA and details what is known and what is still controversial regarding the modes of NO adsorption, the oxidation state of Pd, how Pd+ is formed, and the interactions of CO and ethene with adsorbed NO. Pd/H-CHA is well-suited for use in PNAs due to its combination of high capacity for NO storage and its durability to hydrothermal aging and poisoning from other exhaust gas components. Theoretical calculations have proposed that Pd+ binds NO more strongly than other Pd states, such as Pd2+, PdO, and Pd nanoparticles. But observation of this oxidation state has eluded the field, and most evidence of this material is founded on indirect observation through theoretical calculations and, for example, the production of NO2 during the adsorption step. Electron paramagnetic resonance spectroscopy has only identified Pd+ in very restricted conditions outside the operating range of PNAs. Another important factor to consider when studying NO storage on Pd is the impact other co-adsorbates in the exhaust gas will have on total storage capacity. CO, H2O, and C2H4 are of particular interest due to their proposed abilities to increase total NO storage capacity. The mechanism by which these co-adsorbates increase storage is also debated, ranging from partial reduction of Pd2+ to Pd+ to the production of yet to be identified co-adsorbed complexes. The literature surrounding NO adsorption on Pd-based zeolites has a wide variety of hypotheses. To better understand the adsorption mechanism, this study seeks to reduce the complexity of the feed conditions and propose reaction mechanisms, examining the effect of one adsorbate gas at a time. Chapter 2 explores how NO adsorbs on Pd/H-CHA and the support zeolite, H-CHA. It was determined that Pd improves the total adsorption capacity of the base zeolite, and is necessary for high-temperature NO storage. The composition of the feed plays an important role in the adsorption capabilities of Pd/H-CHA: in the presence of air, all of the adsorbed NO is released at low temperatures (<573 K). A similar behavior in the temperature-programmed desorption (TPD) spectra is also seen when NO is adsorbed from He. Water is found to be essential to produce the sites that desorb NO at temperature >573 K. Pretreating Pd/H-CHA in water results in two NO desorption peaks during TPD, one of which is observed at higher temperatures suitable for PNA applications. Based on a combination of theoretical calculations and experimental results, it is hypothesized that water reacts with NO and Pd2+ cations to produce the Pd+ cations, and these sites are found to be responsible for high-temperature NO desorption. IR spectroscopy was used to observe the frequency of N-O stretching vibrations. Evidence of NO bound on Pd can be found at 1860 cm-1 and 1810 cm-1. Further calculations revealed multiple possible Pd cation species depending on their location in the CHA, that have overlapping, expected IR stretching frequencies. In contrast to the literature, both experiments and theoretical calculations indicate that the two bands observed cannot be assigned unambiguously to NO adsorbed on Pd2+ and Pd+ cations. Multiple NO-Pd2+ and NO-Pd+ species exist, differentiated by their location within the CHA, and a mix of these complexes makeup the TPD peaks and the IR bands. However, experiments designed to target Pd oxidation states could be used to support the formation of the Pd+ cation. For example, water can be introduced to Pd/H-CHA saturated with NO and used to selectively displace NO that was stored on Pd2+: the remaining NO is bound to only Pd+ cations. Chapter 3 complements the findings from the previous chapter and examines the influence of carbon monoxide on the adsorption of NO. The impact of CO on the NO adsorption abilities of Pd has been heavily debated in the literature. This co-adsorbate was of particular interest because of one popular hypothesis that suggests that CO performs a similar function to that of NO by partially reducing Pd2+ to Pd+. TPD experiments revealed that CO increases the amount of NO desorbing at higher temperatures and decreases the amount of NO desorbing at lower temperatures. The total NO stored during these tests does not change as a function of how CO is introduced, demonstrating that CO does not increase the NO storage capacity, but redistributes the available Pd cations between the two oxidation states. In addition, the lack of CO observed during TPD indicates that CO does not form co-adsorbed complexes and bolsters NO storage by reacting with Pd alone. Repeated NO storage cycles on Pd/H-CHA without any pretreatment or regenerative steps were also studied. It is shown that without water in the pretreatment gas stream, the high-temperature adsorption site, Pd+ is lost after releasing NO. Theoretical calculations were used to hypothesize that traces amounts of O2 were sufficient to re-oxidize Pd+ to Pd2+ without the presence of water. As a result, Pd/H-CHA cannot be used in successive storage cycles without the aid of a pretreatment step to enable partial reduction of Pd2+. This phenomenon also impacts efforts undertaken to identify Pd+ via electron paramagnetic resonance spectroscopy. Since only a minute amount of O2 is necessary the re-oxidize it to Pd2+, it is very difficult to detect Pd+. Chapter 4 focuses on the effect of ethene, as a co-adsorbate. Ethene has been reported to perform a similar function as CO and improve the NO storage capacity of Pd-exchanged zeolites. In contrast to literature, TPD experiments show that ethene actually inhibits NO adsorption, particularly the desorption at high-temperatures. Ethene competes with NO for low-temperature adsorption sites (i.e. Pd2+ sites), and reacts with NO to produce CO and methane. Ethene does not adsorb on Brønsted acid sites to a significant extent. Investigation of ethene storage on Pd/H-CHA shows that ethene is primarily adsorbed but reacts during desorption to form CO. In addition, exposure of the Pd/H-CHA to ethene results in the formation of coke, which inhibits NO storage in subsequent adsorption cycles. Other adsorption methodologies, such as co-adsorption of NO and ethene or sequential adsorption of ethene and NO leads to a reduction in the NO storage capacity. The NO helps oxidize ethene to CO and CO2. IR spectroscopy was used to identify the adsorbed species. Though no CO-Pd bands were bands were observed, bands attributed to ethene adsorbed on Pd are observed. However, Pd/H-CHA suffers from coking due to exposure to ethene that hurts its ability to store NO over several adsorption cycles; therefore, Pd/H-CHA functions as an effective HC trap but at the expense of diminished NO storage.