Effect of gas compositions on SO₂ poisoning over Cu/SSZ-13 used for NH₃-SCR

Highlights

• Poisoning of Cu/SSZ-13 in SO₂ + O₂ resulted in weakly bound SO₂ and copper sulfate.

• Significantly larger amount of copper sulfates was formed in presence of H₂O.

• NH₃ during the poisoning resulted in the formation of ammonium sulfate species.

• Heating in O₂ + H₂O transformed ammonium sulfates into copper sulfates with high stability.

• Hydrothermal aging of Cu/SSZ-13 resulted in copper sulfates already at 200 °C.

Abstract

This study focuses on the effect of gas composition during SO₂ poisoning over Cu/SSZ-13 for NH₃-SCR application and was performed by conducting SO₂-TPD experiments in a variety of lean gas compositions. In addition, the poisoned monoliths were characterized in detail using ICP-SFMS, UV–vis and XPS. During SO₂ poisoning under dry and lean conditions, two different sulfur species were found, which were assigned to weakly bound SO₂ and copper sulfate like species. Moreover, a significantly larger amount of copper sulfates was present in humid environment. The presence of NH₃ during the poisoning resulted in the formation of ammonium sulfate species which were decomposed at the same temperature independently if the poisoning with SO₂ was conducted in ammonia oxidation conditions or under standard or fast SCR conditions. Moreover, if the temperature ramp was conducted with O₂ and H₂O compared to Ar alone, more stable sulfate species were formed. In addition, SO₂ poisoning under standard SCR conditions resulted in mostly ammonium sulfate formation at 200 °C, whereas copper sulfates were predominant after poisoning at 400 °C. After hydrothermal aging at 800 °C, more reducible copper species were noticeable and UV–vis showed that copper oxides had been formed. Sulfur poisoning of the hydrothermally aged sample resulted in the additional formation of copper sulfates during poisoning at 200 °C, which was not the case for poisoning of the fresh catalyst. Thus, the copper oxide species enhanced the copper sulfate formation.

Introduction

Ammonia selective catalytic reduction (SCR), in which aqueous urea is used as a reductant, is a commonly practiced technology to reduce NO_x in diesel vehicles. For this application, the popularity of vanadia supported on a titania catalyst has been substituted by metal-exchanged zeolites because of its sensitivity towards hydrothermal aging [1]. Among metal-exchanged zeolites, Fe-zeolites are more active in the high temperature region whereas copper zeolites offer high activity at lower temperature [2]. In the last few years, focus of the research has turned to copper chabazite catalysts (Cu/CHA) [3], such as Cu/SSZ-13 and Cu/SAPO-34 because of their resistance towards hydrothermal aging [4], [5], [6] and hydrocarbon poisoning [7], [8]. The activities of the Cu/CHA catalysts rely on the copper sites which are suggested to occupy two different cationic positions [9]. Distribution of the copper sites in these positions, i.e. inside the six-membered ring (6R) and in the large zeolite cages (in the 8MR), is determined primarily by the copper loading [9], [10]. Understanding the nature and behavior of the active copper sites can be useful to design improved NH₃-SCR system that has minimal effect of sulfur poisoning.

Studying the impact of sulfur poisoning on the catalyst system is crucial because of the unavoidable presence of sulfur in the fuel. The copper-based catalyst has shown to be more prone to poisoning by sulfur compared to Fe-zeolite [11], [12], [13]. Several studies [12], [13], [14], [15], [16], [17] have been published regarding this subject and several factors are proven to be important in the deactivation of the copper zeolites caused by SO₂. Decreasing activity of the catalysts is mainly observed at low temperatures [12], [13], [14], [15], [16] where the Cu-zeolite SCR catalysts have performance advantage over other SCR catalysts [12]. The degree of impact of SO₂ poisoning on the different reactions is not the same, e.g. the standard SCR reaction is more affected by the poisoning than the fast SCR reaction [11], [14], [15]. When the ratio of NO₂/NO_x in the SCR is higher than 0.5, the impact is even less than in the fast SCR case [14], [15].

Results from several research studies [14], [15], [18], [19], [20], [21], [22], [23], [24] have highlighted important mechanistic information regarding sulfur poisoning of copper zeolite catalysts in the NH₃-SCR system. Since SO₂ is expected to have higher affinity to the copper species than to the strong acid sites of the zeolite [18], the deactivation caused by SO₂ is believed to be correlated with the sulfur species bound to copper sites. This fact is in accordance with our previous finding [15] which indicates that the reducibility of copper in Cu/SAPO-34 decreased during an H₂-TPR experiment after it had been exposed to SO₂ in the presence of O₂ and H₂O. In addition, Cheng et al. [19] observed that sulfate species were formed after the Cu/BEA sample had been poisoned with SO₂. Moreover, Hamada et al. [20] found similar result using Cu/ZSM-5 catalyst and they suggested that the electronic and local structure of the Cu species of sulfur poisoned sample changed, possibly caused by the presence of sulfate ions in the surroundings of Cu ions, which then blocked the NO adsorption. This claim is also supported by Hass and Schneider [21], who performed DFT calculations showing that complexes, such as ZCu-SO₄ or ZCu–SO₄–CuZ, were formed during SO₂ exposure under O₂ excess conditions and these complexes were significantly different compared to bulk CuSO₄.

It has been observed that the presence of other gases during SO₂ poisoning influences the type of sulfur species formed; for example, during sulfur poisoning in the presence of NH₃, as investigated by Zhang et al. [22], ammonium sulfate species were observed over Cu/SAPO-34 and the formed ammonium sulfate species then poisoned the active sites and blocked the zeolite pores. The pore blocking mechanism has also been observed in our previous studies [14], [15] denoted by a decreasing surface area of the sulfur poisoned catalysts. In a recent study, Wang et al. [23] characterized different species present during SO₂ poisoning over Cu/SAPO-34 in NO oxidation and SCR conditions at various temperatures. Based on their experiments, it was found that copper sulfate was present in all sulfated samples, whereas ammonia sulfate was only found after the samples poisoned with SO₂ in SCR environment at 250 °C (but not at 350 °C). Moreover, it has been found that the activity of the catalyst can be regained by heating up the poisoned sample [19], [25], [26] even though a temperature of 500 °C is insufficient to fully recover the catalytic activity [15]. The recoverability also depends on the form of sulfur in the feed; the presence of SO₃ at high temperature results in greater difficulties in regenerating the catalyst and SO₃ also causes a more severe effect on the catalytic activity than SO₂ [26]. Moreover, Kumar et al. [27] found that SCR conditions might enhance sulfur regeneration.

There are several studies examining the sulfur poisoning over Cu/SAPO-34 [15], [17], [18], [22], [23], [24], however there are only few studies focusing on sulfur poisoning of Cu/SSZ-13 [14], [16], [28], [29]. Both Cu/SAPO-34 and Cu/SSZ-13 have CHA structure, but there are several differences between these two materials. For example, Cu/SAPO-34 produces less N₂O and forms more stable ammonium nitrates compared to Cu/SSZ-13 [31]. In a recent study by Su et al. [28], they found during sulfur TPD experiments, that the sulfur release (from 200 to 1000 °C) was about double for Cu/SSZ-13 compared to Cu/SAPO-34, which demonstrates that poisoning of Cu/SSZ-13 is different from Cu/SAPO-34 and it is therefore critical to also study sulfur poisoning of Cu/SSZ-13 in detail.

In our previous investigation [14], we showed that SO₂ poisoning under SCR conditions over Cu/SSZ-13 results in more severe poisoning than sulfur treatment with O₂ and H₂O alone; however, the question of what specifically occurs during SO₂ deactivation in different gas compositions requires further study. To our knowledge, there are no studies available that have used different surface characterization techniques to examine the sulfur species when poisoning the Cu/SSZ-13 catalyst using various gas compositions, which is the objective of the current study. In this work, we characterized the formed sulfur species over Cu/SSZ-13 during different conditions using UV–vis and XPS and examined the stability using the TPD technique. Dry and humid conditions are used, as well as under ammonia oxidation, standard SCR and fast SCR conditions. In addition, the effect of sulfur poisoning in combination with hydrothermal aging was also investigated in this study.