Effect of framework Al pairing on NO storage properties of Pd-CHA passive NOx adsorbers

https://doi.org/10.1016/j.apcatb.2022.122074Get rights and content

Highlights

  • The degree of Al pairing in Pd/H-CHA on passive NOx adsorption was examined.

  • Pd/H-CHA was prepared containing 53.0 %, 10.8 % and 6.5 % paired Al sites at fixed Si/Al ratio.

  • Pd was present as isolated cations in the two samples with the most paired Al sites.

  • A high concentration of paired Al sites was beneficial for NOx storage capacity.

  • High Al pairing was also beneficial for minimizing deactivation in the presence of CO/H2.

Abstract

Three Pd/H-CHA samples were prepared containing 53.0 %, 10.8 % and 6.5 % paired Al sites at near fixed Si/Al ratio and similar Pd loading. According to H2 temperature-programmed reduction, Pd was present almost exclusively as isolated cations in the two samples containing the higher concentrations of paired Al sites, whereas in the other sample PdO was also present. Simulated lean cold start tests on the fresh samples conducted in a microflow reactor showed that the sample containing PdO stored the lowest amount of NOx. When tested with CO/H2, the sample containing 53.0 % paired Al sites showed significantly better storage capacity than the other samples and deactivated less rapidly upon sequential tests. Experiments using lean gasoline engine exhaust revealed similar trends. This study showed that a high concentration of paired Al sites in Pd/H-CHA is beneficial for NOx storage capacity, thermal durability, and minimizing deactivation in the presence of CO/H2.

Introduction

The low exhaust temperatures that result from improvements to engine efficiency represent a challenge to the control of pollutant emissions. Standard aftertreatment technologies such as three-way catalysts (TWCs) and selective catalytic reduction (SCR) catalysts fail to function efficiently at low temperatures, from which it follows that high efficiency internal combustion engines require new or improved aftertreatment technologies that specifically address this issue. Passive NOx adsorbers (PNAs), also known as low temperature NOx adsorbers (LTNAs), represent a possible strategy to mitigate these emissions, with palladium-loaded zeolites being of particular interest [1], [2], [3]. The intended function of these materials is to adsorb NOx and hydrocarbons at low temperatures (∼25–200 °C) and then to desorb them once the downstream catalytic converter has reached operational temperature (>200 °C) [4]. Several characteristics are required for a material to serve as a PNA, including adsorption of large quantities of NOx (and particularly NO) at near ambient temperature and complete desorption of the stored NOx species at temperatures that are within the range of exhaust temperatures on the vehicle during normal customer usage (typically no more than 300–350 °C for diesel vehicles). These materials must also resist the detrimental effects of all typical exhaust components (e.g., H2O, CO2, CO, H2, HC, SO2, etc.) over a wide range of temperature for the useful life of the vehicle.

Recent studies have shown that under working conditions, the NO adsorption sites in Pd zeolites correspond to isolated Pd cations, including Pd2+ and Pd+ [5], [6], [7], [8], [9], [10]. Depending on a number of factors related to material structure and preparation, including the zeolite framework topology, synthesis and treatment conditions, and the framework Al content and extent of Al pairing in the framework, other Pd species may be present in as-prepared Pd zeolite samples in the form of metallic Pd and PdO (in nanoparticulate or bulk form) [11], [12]. NO can also adsorb at acid sites (H+) in the zeolite framework that are generated upon isomorphous substitution of framework Si with Al, although such adsorption is suppressed in the presence of water vapor [1], [3], [13]. The presence of Pd2+ ions in these materials has been substantiated by a variety of methods, including X-ray photoelectron spectroscopy, diffuse reflectance UV–vis spectroscopy, X-ray absorption spectroscopy, and H2 temperature programmed-reduction (TPR) [11], [14]. Spectroscopic evidence for the presence of Pd+ exists in the form of IR spectra collected using CO [15], [16] and NO [17], [18], [19] as probe molecules, while electron paramagnetic resonance measurements have indicated the presence of Pd+ in Pd zeolites exposed to reducing conditions [20], [21]. Indeed, the formation of Pd+ from Pd2+ has been reported to occur upon NO adsorption in Pd/H-ZSM-5 [18], [19], with the Pd2+ species undergoing reduction being assigned as either isolated Pd2+ ions or Pd2+ hydroxyl complexes (i.e., [Pd(OH)x]2-x+). Moreover, recent density functional theory (DFT) calculations by Van der Mynsbrugge et al. for Pd/H-CHA have inferred that Pd+ ions are thermodynamically preferred over Pd2+ in some configurations [22]; for example, next-nearest neighbor (NNN) Al pairs (separated by a single Si atom) in the CHA 6-ring cannot provide the ideal square planar coordination to stabilize Pd2+, such that the relative stability of Pd+ and Pd2+ cations at these Al pairs depends on the temperature and partial pressure of water [22].

In addition to Pd+ and Pd2+ ions, the presence of [Pd(OH)]+ moieties in Pd zeolites has been inferred [11], [15], [16], [23]. NO adsorption on [Pd(OH)]+ is proposed to occur either via direct NO adsorption at the Pd center or via the reaction NO + 2Z-[Pd(OH)]+ ↔ 2Z-Pd+ + NO2 + H2O, where Z- denotes a cation exchange site in the zeolite framework resulting from isomorphous substitution of Si by Al. In other words, [Pd(OH)]+ is suggested to act as a precursor for the formation of Pd+ ions [23], [24], [25]. However, the role of [Pd(OH)]+ as sites for NO adsorption has been called into question by several groups [22], [26]. DFT calculations indicate that [Pd(OH)]+ is thermodynamically less stable than Pd+/H+ and Pd2+ at paired Al sites, and that under realistic experimental conditions the reaction of [Pd(OH)]+ with adjacent protons to form Pd2+ and water is favorable [22]. The stability of Pd2+ and [Pd(OH)]+ relative to non-exchanged species (solid Pd or PdO) has not been reported but can readily be calculated by combining the stability of exchanged Pd species relative to monoatomic Pd gas calculated by DFT with known free energies of formation for solid Pd and PdO versus Pd gas. For example, in 20 kPa O2 and 5 kPa O2, the free energies of [Pd(H2O)4]2+, Pd+H+(H2O)2, and Pd(OH)+H+ at next-next-nearest neighbor Al pairs in a 6-membered ring are − 351, − 265, and − 177 kJ/mol versus atomic Pd at 300 K [22] compared to − 339 kJ/mol for metallic Pd and − 505 kJ/mol for PdO, indicating bulk PdO is the preferred phase at ambient temperature. At 1050 K, the free energies for Pd2+, Pd+H+, and [Pd(OH)]+H+ at the same site are − 263, − 171, and − 25 kJ/mol relative to atomic Pd, and those of solid Pd and PdO are − 246 and − 249 kJ/mol respectively. High temperatures thus stabilize exchanged Pd at a subset of zeolite sites and lower the free energy penalty for exchange at all sites relative to non-exchanged bulk Pd phases. The free energy of exchanged Pd may be further lowered by as much as 80 kJ/mol upon adsorption of NO [17].

From the foregoing, the available evidence points towards Pd+/H+ and Pd2+ being the main sites involved in NO adsorption; in order to maintain charge balance, such species can form only at paired framework Al sites. In principle, Pd+ or [Pd(OH)]+ ions can additionally exist at isolated Al sites, providing a pathway exists for their formation from the Pd2+ precursor used for catalyst preparation. In this regard, it is of note that Van der Mynsbrugge et al. have reported that the NH3 contained in the typical Pd precursor Pd(NH3)4(NO3)2 should be capable of driving the reduction of Pd2+ to Pd+ upon insertion in H-CHA based on the calculated free energy of reaction [22]. The same authors also reported DFT calculations showing that the formation of Pd+ is thermodynamically preferred to [Pd(OH)]+ at isolated Al sites. Lardinois et al. found that CHA samples synthesized purposefully to contain no paired framework Al sites were able to stabilize 0.08 isolated Pd cations per Al, and proposed that the Pd species was [Pd(OH)]+ based on the presence of an OH stretching feature at 3660 cm−1 in IR spectra and H2 consumption per Pd of approximately 1.0 [11]. Notably, CHA samples containing significant amounts of six-membered ring (6-MR) paired Al sites, as quantified by Co2+ titration, were exchanged with Pd2+ and did not show the 3660 cm−1 IR feature, suggesting the preferential formation of Pd2+ at 6-MR paired Al sites. Moreover, at similar bulk Pd contents, the amount of isolated Pd sites formed on Pd-CHA samples systematically increased with their paired Al content, consistent with thermodynamic calculations that (Z-)2Pd2+ sites are more stable than Z-[PdOH]+ after the high temperature air treatments required for the dispersion of Pd in H-CHA [11], [22].

In practice, it is found that the formation of high dispersions of Pd cations at industrially relevant Pd loadings (≥1 wt%) requires a high framework Al content [27], which on average results in a high concentration of paired Al sites. Increasing the Al content inevitably decreases the hydrothermal stability of the zeolite; hence, the Si/Al ratio should be optimized while maximizing the percentage of paired Al sites for that Al content. The use of high calcination temperatures (≥650 °C) to ensure adequate mobility of the Pd species resulting from thermal decomposition of the Pd precursor is also a pre-requisite [11], [12]. Given that increasing the amount of paired Al sites in the zeolite is beneficial for increasing the Pd dispersion, it follows that it is also beneficial for the NO storage capacity and that the NO/Pd ratio will tend towards unity as the Pd dispersion tends towards 100% [27]. Beyond this, however, it is unclear as to what effect increasing the amount of paired Al sites – at fixed Si/Al ratio – has on the NO adsorption properties of Pd/H-CHA under working conditions. Consequently, this study sought to examine the effect of the degree of Al pairing in Pd/H-CHA at fixed Si/Al ratio and Pd loading on PNA performance. The use of realistic exhaust conditions was emphasized given the known susceptibility of Pd-CHA to deactivate in the presence of CO [28], [29], [30] and the possible influence of Al site pairing on such catalyst deactivation and potential catalyst regeneration by means of high temperature oxidative treatment.

Section snippets

Catalyst synthesis

CHA zeolites were hydrothermally synthesized following previously reported procedures [31], [32] using Ludox AS-40 as the silica source. The other chemicals used in the synthesis were aluminum hydroxide (98%, SPI Pharma), aluminum isopropoxide (98%, Sigma-Aldrich), trimethyladamantylammonium hydroxide (TMAda+, 25 wt% in water, Sachem), sodium hydroxide (98% Alfa Aesar) and deionized water (18.2 MΩ) (see Table S1 in Supplementary Information for the composition of the synthesis solutions).

Catalyst synthesis and characterization

To facilitate studies assessing the effect of Al site pairing on the NO adsorption properties of Pd-CHA, three CHA samples were prepared containing approximately the same Si/Al ratio but differing fractions of paired Al sites. The preparations followed synthetic strategies previously reported by Di Iorio and Gounder [31], [33], and the synthesis recipes are reported in Table S1 in the Supporting Information. X-ray diffractograms of the resulting samples, denoted as CHA-1, CHA-2 and CHA-3, are

Conclusions

This work sought to assess the effect of Al pairing in Pd/H-CHA on the performance of this material as a passive NOx adsorber. According to the results of H2 TPR, the two Pd/H-CHA samples with the highest concentration of paired Al sites (Pd-CHA-2 and Pd-CHA-3) contained Pd almost exclusively as isolated cations, whereas in Pd-CHA-1 PdO was also present. These results are consistent with the inability of the zeolite framework in Pd-CHA-1 to charge compensate all of the loaded Pd2+ ions. Given

CRediT authorship contribution statement

Joseph Theis: Conceptualization, Methodology, Validation, Investigation, Writing – original draft, Writing – review & editing. Justin Ura: Investigation, Validation. Andrew Bean Getsoian: Conceptualization, Methodology, Investigation, Writing – review & editing. Vitaly Prikhodko: Conceptualization, Methodology, Validation, Investigation, Writing – original draft, Writing – review & editing. Calvin Thomas: Investigation, Validation. Josh Pihl: Resources. Trevor Lardinois: Investigation,

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This work was funded by the Department of Energy Office of Vehicle Technologies award number DEEE0008213. This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or

References (41)

  • M. Ambast et al.

    Passive NOx adsorption on Pd/H-ZSM-5: experiments and modeling

    Appl. Catal. B Environ.

    (2020)
  • Y. Ryou et al.

    Effect of reduction treatments (H2 vs. CO) on the NO adsorption ability and the physicochemical properties of Pd/SSZ-13 passive NOx adsorber for cold start application

    Appl. Catal. A Gen.

    (2019)
  • Y. Gu et al.

    Investigation of an irreversible NOx storage degradation mode on a Pd/BEA passive NOx adsorber

    Appl. Catal. B Environ.

    (2019)
  • J. Theis et al.

    Assessment of zeolite-based low temperature NOx adsorbers: Effect of reductants during multiple sequential cold starts

    Catal. Today

    (2021)
  • M. Ambast et al.

    “NOx adsorption with CO and C2H4 on Pd/SSZ-13: experiments and modeling”

    Appl. Catal. B: Environ.

    (2021)
  • A. Gupta et al.

    NOx uptake and release on Pd/SSZ-13: impact of feed composition and temperature

    Catal. Today

    (2021)
  • K. Khivantsev et al.

    Palladium/Beta zeolite passive NOx adsorbers (PNA): clarification of PNA chemistry and the effects of CO and zeolite crystallite size on PNA performance

    Appl. Catal. A: Gen.

    (2019)
  • Y. Gu et al.

    Investigation of NO adsorption and desorption phenomena on a Pd/ZSM-5 passive NOx adsorber

    Appl. Catal. B: Environ.

    (2021)
  • H.-Y. Chen et al.

    Low temperature NO storage of zeolite supported Pd for low temperature diesel engine emission control

    Catal. Lett.

    (2016)
  • K. Khivantsev et al.

    Molecular level understanding of how oxygen and carbon monoxide improve NOx storage in palladium/SSZ-13 passive NOx adsorbers: the role of NO+ and Pd-(II)(CO)(NO) species

    J. Phys. Chem. C

    (2018)
  • Cited by (12)

    • Constructing the multilayer O-g-C<inf>3</inf>N<inf>4</inf>@W<inf>18</inf>O<inf>49</inf> heterostructure for deeply photocatalytic oxidation NO

      2023, Separation and Purification Technology
      Citation Excerpt :

      Moreover, long-time exposure to contaminated air can cause serious respiratory diseases for humans, therefore, it is undoubtedly urgent to harness nitrogen oxide (NOx) [9–11]. Up to now, the various strategies have been implemented for atmospheric NO removal [12–15]. However, these methods are not economically feasible for NO removal at parts of trace levels (ppb).

    View all citing articles on Scopus
    View full text