Precise control of post-treatment significantly increases hydrothermal stability of in-situ synthesized cu-zeolites for NH3-SCR reaction

Precise control of post-treatment significantly increases hydrothermal stability of in-situ synthesized cu-zeolites for NH3-SCR reaction

Journal Pre-proof Precise control of post-treatment significantly increases hydrothermal stability of in-situ synthesized Cu-zeolites for NH3 -SCR reac...

5MB Sizes 0 Downloads 4 Views

Journal Pre-proof Precise control of post-treatment significantly increases hydrothermal stability of in-situ synthesized Cu-zeolites for NH3 -SCR reaction Yulong Shan, Jinpeng Du, Yunbo Yu, Wenpo Shan, Xiaoyan Shi, Hong He

PII:

S0926-3373(20)30070-9

DOI:

https://doi.org/10.1016/j.apcatb.2020.118655

Reference:

APCATB 118655

To appear in:

Applied Catalysis B: Environmental

Received Date:

11 June 2019

Revised Date:

12 January 2020

Accepted Date:

18 January 2020

Please cite this article as: Shan Y, Du J, Yu Y, Shan W, Shi X, He H, Precise control of post-treatment significantly increases hydrothermal stability of in-situ synthesized Cu-zeolites for NH3 -SCR reaction, Applied Catalysis B: Environmental (2020), doi: https://doi.org/10.1016/j.apcatb.2020.118655

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2020 Published by Elsevier.

Precise control of post-treatment significantly increases hydrothermal stability of in-situ synthesized Cu-zeolites for NH3-SCR reaction

Yulong Shana, Jinpeng Dua,b, Yunbo Yua,b,c, Wenpo Shanc, Xiaoyan Shia,b*

ro of

Hong Hea,b,c,*

a State Key Joint Laboratory of Environment Simulation and Pollution Control,

Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences,

-p

Beijing 100085, China.

re

b University of Chinese Academy of Sciences, Beijing 100049, China. c Center for Excellence in Regional Atmospheric Environment, Institute of Urban

na

*Corresponding authors

lP

Environment, Chinese Academy of Sciences, Xiamen 361021, China.

ur

E-mail: [email protected] (Xiaoyan Shi), [email protected] (Hong He).

AUTHOR INFORMATION

Jo

Corresponding Author

*Fax: +86 10 62841040; Tel: +86 10 62841040, Email: [email protected] *Fax: +86 10 62849123; Tel: +86 10 62849123, Email: [email protected] Graphical abstract

ro of -p

The post-treatment can adjust zeolite crystallinity and Cu distribution of in-situ synthesized

lP



re

Highlights

Cu-zeolites.

The in-situ synthesized Cu-zeolites after post-treatment shows excellent hydrothermal stability.

na



Cu2+ ions inhibit dealumination of the SSZ-13 zeolite structure during hydrothermal aging.



Excessive quantities of Cu2+ easily accumulate to form CuOx clusters, leading to collapse of

ur



Jo

long-range order in the zeolite structure during hydrothermal aging.

ABSTRACT

The origin of the beneficial effects of post-treatment processing with HNO3 and

NH4NO3 solutions on the catalytic performance of copper-amine templated zeolites used in the NH3-SCR reaction was investigated. By careful design and optimization of

the dual-treatment procedure (HNO3-NH4NO3), Cu-SSZ-13 catalysts were obtained using Cu-TEPA as a template that exhibited NOx conversion above 90% in a wide temperature window from 250 to 450 ℃ with GHSV of ~ 400,000 h-1, even after hydrothermal aging at 750 and 800 ℃. The results of XRD and H2-TPR indicated that the HNO3 post-treatment adjusts the zeolite crystallinity and optimizes the copper species distribution in Cu-SSZ-13 catalysts. Further treatment with NH4NO3 reduces

ro of

the Cu/Al ratios effectively and avoids the accumulation of Cu2+ ions to form CuOx clusters during hydrothermal aging. We investigated the influence of Cu2+ on the

hydrothermal stability of Cu-SSZ-13, and two opposing effects were found. Cu2+ ions

-p

inhibit dealumination of the SSZ-13 zeolite structure, while excessive quantities of

re

them easily accumulate to form CuOx clusters, leading to collapse of long-range order in the zeolite structure during hydrothermal aging. Before the zeolite structure

lP

collapses, Cu-SSZ-13 catalysts with high Cu loading exhibit higher NH3-SCR activity

na

due to the preservation of more active Cu2+. However, Cu-SSZ-13 catalysts totally lose deNOx activity in NH3-SCR once the zeolite structure collapses. To guarantee

ur

both the activity for NOx reduction and hydrothermal tolerance, the optimal Cu loading (Cu/Al=0.22~0.31, here) and structural stability of in-situ synthesized Cu-

Jo

SSZ-13 should be carefully tuned by HNO3 and NH4NO3 treatment. Introduction

To meet strict emission standards, numerous efforts have been made toward NOx elimination from diesel engine exhaust in recent years. To this aim, selective catalytic reduction of NOx with NH3 (NH3-SCR) is an efficient technique. For the NH3-SCR

process, Cu-zeolite catalysts with different topologies have been developed [1-6], among which Cu-SSZ-13 has been extensively studied experimentally and theoretically in the past few years due to its excellent catalytic performance and hydrothermal stability.[7-13] Generally, Cu-SSZ-13 samples are prepared by an ion-exchange method, with the SSZ-13 substrate synthesized using N,N,N-trimethyl-1-adamantammonium hydroxide

ro of

(TMAdaOH) as a template [14]. Considering the high price of TMAdaOH, many new methods of synthesizing the SSZ-13 substrate zeolite have been developed [13, 15-

17]. For example, Zhao et al. [13] synthesized an Al-rich SSZ-13 zeolite without an

-p

organo-template. Wang et al [16] synthesized a high-silica SSZ-13 zeolite using a

re

solvent-free route with an economical template. After multiple steps such as ionexchange with NH4NO3 solution and Cu2+ salt solutions, the Cu-SSZ-13 catalysts

lP

were obtained and showed good NOx reduction activity and/or hydrothermal stability.

na

Remarkably, by using inexpensive Cu-tetraethylenepentamine (Cu-TEPA) as a template, an in-situ methodology to directly synthesize Al-rich Cu-SSZ-13 with Si/Al

ur

of ~5 was developed by Ren et al.[18]. The utilization of a copper-amine complex as a template reduced the required manufacturing steps due to the introduction of active

Jo

Cu2+ species. Some other small-pore zeolites, such as Cu-SAPO-34 and Cu-SSZ-39, could also be in-situ synthesized using Cu-TEPA as a template [19, 20]. However, the Cu contents in the in-situ synthesized zeolites were too high, especially when using Cu-TEPA as the sole template, which resulted in the formation of CuOx clusters after calcination, finally relating to low hydrothermal stability and NOx conversion at high

temperatures (> 400 ℃) in the NH3-SCR reaction [21]. To increase the SCR activity and hydrothermal stability of in-situ synthesized Cu-zeolites, therefore, post-treatment steps are required. For example, our group proposed post-treatments using NH4NO3 and HNO3 solutions separately to remove excess Cu efficiently, and obtained CuSSZ-13 catalysts with high deNOx activity in a wide temperature window [21, 22]. However, the mechanism of this post-treatment, such as the roles of NH4NO3 and

ro of

HNO3, was vague. Although HNO3 treatment resulted in relatively good hydrothermal stability (750 ℃, 16 h), elevated temperature (800 ℃) caused destruction of the zeolite framework structure and total loss of NH3-SCR performance.

-p

The hydrothermal degradation mechanism of Cu-SSZ-13 has been studied by many

re

researchers [8, 10, 23-25]. As revealed by Kim et al [8], the deactivation of Cu-SSZ13 during hydrothermal aging is related to the distribution of Cu2+ ion species and the

lP

Si/Al ratio. By using EPR and H2-TPR, it was found that Cu2+ ions in the d6r cage

na

(Cu2+-2Z balanced by paired framework charges) are hydrothermally more stable than Cu2+ ions in the cha cage ([Cu(OH)]+-Z balanced by a single framework charge), and

ur

the hydrothermal stability decreases with the increase of the Cu/Al ratio in the CuSSZ-13 catalyst. Recently, Song et al.[10] uncovered the changes occurring in the

Jo

zeolite support and the Cu active sites to gain an atomic-level understanding of hydrothermal stability. The Cu2+-2Z species were demonstrated to be hydrothermally more stable than the [Cu(OH)]+-Z species, while [Cu(OH)]+-Z gradually transformed to Cu2+-2Z and CuOx clusters during hydrothermal aging, therefore resulting in a decrease in deNOx activity.

To increase the hydrothermal stability of Cu-SSZ-13, therefore, its intrinsic properties including copper species distribution, Si/Al ratio, and zeolite framework structure stability should be carefully tuned. The synthesis of Cu-SSZ-13 catalysts by the in-situ method using Cu-TEPA as sole template always results in Al-rich zeolites with a low Si/Al ratio of ~5, which makes them more susceptible to dealumination. Although Corma et al prepared high-silica Cu-SSZ-13 by using a combination of Cu-

ro of

TEPA and other templates as OSDA, the activity and hydrothermal stability of the

prepared Cu-SSZ-13 catalysts were limited. [20, 26] In fact, the Al-rich Cu-SSZ-13

catalysts provided a large ion exchange capacity and showed outstanding NH3-SCR

-p

performance due to the large amount of Brønsted acid sites [13, 21, 27]. Given these

re

advantages, increasing the hydrothermal stability of Al-rich Cu-SSZ-13 catalysts is important. So, this work proposed a strategy of precisely controlling the post-

lP

treatments to adjust the stability of the skeleton structure and optimize the copper

na

species distributions in copper-amine templated zeolites. In detail, the in-situ synthesized Cu-SSZ-13 samples were post-treated sequentially by HNO3 and

ur

NH4NO3 solutions with varying concentration, the optimization of which enabled CuSSZ-13 catalysts with high hydrothermal stability to be obtained. Furthermore, the

Jo

zeolite structure and copper species changes induced by post-treatment were elucidated precisely by detailed characterizations such as XRD, NMR, and H2-TPR measurements. The principles obtained are expected to guide the rational synthesis and post-treatment of in-situ synthesized zeolite catalysts using Cu-TEPA as a template or co-template, with high deNOx activity and hydrothermal stability in the

NH3-SCR reaction. 2. Experimental 2.1. Catalyst preparation The in-situ synthesized Cu-SSZ-13 samples were prepared using the method published by Ren et al.[18] For post-treatment, the as-synthesized Cu-SSZ-13 (no previous calcination) was added to HNO3 solutions with concentration ranging from

ro of

0.005 to 1.0 M at 80 ℃ for 12 h, after which the samples were filtered, washed, and dried overnight prior to calcination at 600 ℃ for 6 h. The obtained catalysts were

denoted as Cux-SSZ-13-Ⅰ, where x and Ⅰrepresent the Cu loading (wt.%) and the

-p

first post-treatment in HNO3 solution, respectively. Then the optimal Cu4.8-SSZ-13-I

re

(0.1 M HNO3 post-treatment) was stirred in NH4NO3 solution (0.01 ~ 0.1 M) at 40 ℃ for 5 h for the second post-treatment process, followed by the same procedure of

lP

filtration, washing, drying, and calcination at 600 ℃ for 6 h. The obtained catalysts

na

after the dual post-treatment process with HNO3 and NH4NO3 were denoted as CuxSSZ-13-Ⅱ, where x and Ⅱ represent the Cu loading (wt.%) and the second post-

ur

treatment in NH4NO3 solution after treatment in HNO3 solution. For the hydrothermal stability studies, the Cu-SSZ-13 catalysts were hydrothermally aged in 10 % H2O/air

Jo

at 750 ℃ for 16 h or 800 ℃ for 5 h. The in-situ synthesized Cu-SAPO-34 catalyst was also prepared with the gel molar

composition: 1AlOOH: 40H2O: 1H3PO4: 0.25SiO2: 0.06 (0.07) CuSO4: 0.072 (0.084) TEPA: 1.6 MOR (morpholine). The sources of aluminum, phosphoric acid and silicon were pseudoboehmite, phosphoric acid and fumed silica, respectively. The gel was

transferred to a Teflon-lined stainless-steel autoclave for hydrothermal crystallization at 190 ℃ for 72 h. After washing and drying, the product was calcined in air at 700 ℃ for 6 h. The prepared Cux-SAPO-34 (x represents Cu contents) was then treated by 0.03 mol/L HNO3 for 3 h, and the obtained catalyst was denoted as Cux-SAPO-34HNO3. For the hydrothermal stability studies, the Cu-SAPO-34 catalysts were hydrothermally aged in 10 % H2O/air at 800 ℃ for 16 h.

ro of

2.2. NH3-SCR reaction measurement

Standard NH3-SCR reaction measurements were conducted in a fixed-bed flow

reactor system with an online Nicolet Is10 spectrometer, which was used to analyze

-p

the gas content of NO, NO2, N2O, and NH3. The composition of the reaction gas

re

included 500 ppm NO, 500 ppm NH3, 5% O2, 5% H2O and balance N2, with a total flow rate of 500 mL/min. For activity testing, ~ 50 mg of catalyst (40-60 mesh) was

lP

used, resulting in a gas hourly space velocity (GHSV) of ~ 400,000 h-1. Different

na

GHSVs were achieved by tuning the weight of samples. The NOx (NO and NO2) conversion was calculated based on the inlet and outlet NOx concentrations at steady

ur

state:

𝑁𝑂𝑥 𝑐𝑜𝑛𝑣𝑒𝑟𝑠𝑖𝑜𝑛 = (1 −

[𝑁𝑂𝑥]𝑜𝑢𝑡 [𝑁𝑂𝑥]𝑖𝑛

) × 100%

Jo

2.3 Catalyst characterization Inductively coupled plasma Atomic Emission Spectroscopy (ICP-AES) was used to

measure the elemental contents of the catalysts. Based on the results, the Cu loadings, Al contents, Si/Al, Cu/Al and ion exchange degree as a function of HNO3 and NH4NO3 concentration were calculated and listed in Table 1. Powder X-ray

diffraction (PXRD) analysis was conducted on a computerized Bruker D8 Advance diffractometer with Cu Kα (λ= 0.15406 nm) radiation at room temperature, during which data were collected with 2θ ranging from 5o to 40o with the step size of 0.02o. Solid state 27Al MAS NMR spectra were collected on a Bruker AVANCE Ⅲ 400 WB spectrometer using a 4 mm standard bore CP MAS probe. The NH3-adsorption diffuse reflectance infrared Fourier transform spectroscopy

ro of

(NH3-DRIFTS) experiments were conducted on a Nicolet Is10 spectrometer with an

Omega programmable temperature controller. The sample was pretreated in air for 30 min at 500 ℃ before testing. The background was collected in pure N2 and subtracted

-p

when the sample spectrum was collected. The probe molecule NH3 (500 ppm

re

NH3/N2) was introduced to the catalyst and then swept out with N2. All spectra were recorded by accumulating 100 scans with a resolution of 4 cm-1. A Micromeritics

lP

AutoChem 2920 chemisorption analyzer was used to carry out H2 temperatureprogrammed reduction (H2-TPR) experiments. In this case, about ~50 mg catalyst was

na

used and pretreated in 20% O2/N2 at 500 ℃. After cooling down to 35 ℃, 10% H2/Ar

ur

was introduced to reduce the samples with the temperature increasing from 35 to 1000 ℃ at a rate of 10 ℃/min. The electron paramagnetic resonance (EPR) of the

Jo

samples was recorded at -120 ℃ on a Bruker E500 X-band spectrometer. All catalysts were placed into quartz tubes for the measurement. 3. Results 3.1 Treatment in HNO3 solution. The ICP-AES measurement was carried out to analyze the elemental composition

of the Cu-SSZ-13 catalysts after HNO3 post-treatment. As shown in Table 1, when the HNO3 concentration increased from 0 to 0.03 M, the Cu loadings significantly decreased from 10.0 to 4.7 wt.% (Cu10.0-SSZ-13-initial to Cu4.7-SSZ-13-I). As the HNO3 concentration increased from 0.03 to 0.3 M, interestingly, the Cu loadings (4.54.8 wt.%) as well as Cu/Al ratios (0.39-0.42) hardly changed, which was regarded as a stable region in the HNO3 post-treatments. When the HNO3 concentration continued

ro of

increasing from 0.3 to 1.0 M, however, dealumination occurred, with the Al contents decreasing from 4.6 to 2.2 wt.%, which resulted in an increase of Si/Al (from 5.3 to

13.9). With the process of dealumination, simultaneously, the Cu/Al increased (0.42-

-p

0.49) and leveled off with the saturated ion exchange (96-98%). For all of the HNO3

re

treatment processes, it should be noted that a high ion exchange level above 78% was maintained for the Cu-SSZ-13 catalysts regardless of the HNO3 concentration,

lP

resulting in the facile accumulation of Cu2+ ions during hydrothermal aging [8].

na

XRD results of the Cux-SSZ-13-I samples, which were post-treated in HNO3 solution with different concentrations, are shown in Fig. 1. For all the samples, the

ur

typical diffraction peaks of CHA were observed regardless of the HNO3 solution concentration. From the magnified region, it was observed that the peak intensity of

Jo

CuOx (2 theta = 35.6o and 38.8o) decreased with increasing HNO3 solution concentration, indicating that HNO3 treatment is efficient for CuOx removal. To further reveal the influence of HNO3 treatment, the relative CuO peak areas and crystallinity levels were calculated and shown in Fig. 2. The samples treated by HNO3 solutions with concentrations above 0.03 M (Cu4.7-SSZ-13-Ⅰ) showed a sharp

decrease in CuOx peaks, although a small peak of CuOx could be still observed for these samples except for Cu2.8-SSZ-13-Ⅰ (treated by 0.7 M HNO3) and Cu2.5-SSZ-13Ⅰ(treated by 1.0 M HNO3). This indicates that the formation of CuOx can be basically avoided for the calcined samples with Cu content below ~4.7 wt.%. As shown in Fig. 2b, when as-synthesized Cu-SSZ-13 was treated in HNO3 solution with concentration lower than 0.3 M (Cu4.6-SSZ-13-Ⅰ), meanwhile, the relative

ro of

crystallinity of the Cux-SSZ-13-I samples noticeably increased. However, further

elevation in HNO3 concentration decreased the relative crystallinity of the Cux-SSZ13-I samples, which could probably be attributed to the occurrence of dealumination

-p

(as indicated by Table 1).

re

Fig. 3a shows the H2-TPR results of the series of Cu-SSZ-13-I samples. In this measurement, previous studies [21, 28, 29] revealed that the Cu2+ reduction in Cu-

lP

SSZ-13 catalysts occurred in two steps, involving the reduction of Cu2+ to Cu+ (often occurring below 500 ℃) and Cu+ to Cu0 ( often occurring above 500 ℃), while CuOx

na

reduction to Cu0 always resulted in one reduction peak located below 500 ℃.

ur

Therefore, the presence of CuOx could result in extra H2 consumption below 500 ℃ compared to that above 500 ℃. With this in mind, the peak areas of H2 consumption

Jo

for Cu-SSZ-13-I below and above 500 ℃ were integrated and depicted in Fig. 3b, together with the total H2 consumption. As for the samples with Cu loading above ~4.7 wt.% (HNO3 concentration < 0.03 M), the integrated area of the low-temperature peaks was much greater than that of the high-temperature ones, further confirming the presence of large amounts of CuOx. For the samples with Cu loadings from ~4.7 to

~3.3 wt.% (HNO3 concentration was from 0.03 to 0.5 M), it was found that the H2 consumption at low temperatures was slightly higher than the high-temperature consumption, indicating that only a small amount of CuOx existed in these samples. As for the Cu2.8-SSZ-13-I and Cu2.3-SSZ-13-I samples, however, there was almost no difference in the H2 consumption between the low- and high-temperature regions, suggesting the absence of CuOx. As can be seen, the CuOx identification as analyzed

ro of

by H2 consumption is in good agreement with the results derived from XRD profiles. Fig. 3b also shows that the decrease of Cu loading always results in a decrease of total H2 consumption. As the Cu loading decreased from 10 wt.% to ~ 4.7 wt.%

-p

(HNO3 concentration increased from 0 to 0.03 M), however, it should be noted that

re

the H2 consumption above 500 ℃ (Cu+ to Cu0) increased markedly. This result indicates the increase of active Cu2+ ions, since the Cu+ is derived from the reduction

lP

of isolated Cu2+ ions. This phenomenon indicated that HNO3 post-treatment could

na

rearrange the Cu distribution and expose more active Cu2+ ions, which was beneficial for the NH3-SCR performance of the catalyst. Therefore, the amount of active Cu2+

ur

ions increased, although the HNO3 treatment decreased the Cu loadings of the CuSSZ-13 catalysts.

Jo

Within the low temperature region, furthermore, three reduction peaks were

observed at ~165, ~200, and 350 ℃, respectively. The lowest peak (165 ℃) is assigned to the reduction of CuOx, which resulted from the accumulation of excess copper species during calcination. The appearance of the other two peaks indicated the presence of two types of dehydrated Cu2+ ions species, Cu(OH)+-Z in cha cages (~

200 ℃) and Cu2+-2Z in d6r cages (~350 ℃) [10, 30]. As shown in Fig. 3a, the Cu(OH)+-Z in cha cages (~200 ℃) still remained even though the Cu loading was 2.3 wt.% (HNO3 concentration was 1.0 M). The Cu(OH)+-Z species has outstanding NOx conversion performance, but low hydrothermal stability [8, 10]. Fig. 4a shows the NH3-SCR activities of fresh and hydrothermally aged Cux-SSZ13-I catalysts. The fresh catalysts with Cu loading ranging from 4.5 to 4.8 wt.%

ro of

(treated in HNO3 solution from 0.03 to 0.3 M) showed outstanding NOx conversion in a wide temperature window from 200 to 450 ℃. The catalysts containing Cu content above 7.9 wt.% (treated by low-concentration HNO3 of 0.03 M) showed relatively

-p

lower NOx conversion at high temperatures due to the presence of CuOx, which

re

caused the non-selective oxidation of NH3 [31, 32]. At high temperatures, simultaneously, the Cu3.3-SSZ-13-I, Cu2.8-SSZ-13-I, and Cu2.5-SSZ-13-I catalysts

lP

prepared with the high-concentration HNO3 (0.5~1.0 M) post-treatment also showed a

na

decrease in NOx conversion. This was probably caused by the framework structure damage due to dealumination occurring in the HNO3 post-treatment, as shown in the

ur

XRD (Fig. 1) and ICP (Table 1) results. After hydrothermal aging at 750 ℃, it should be noted that all the Cux-SSZ-13-I catalysts lost their NOx conversion activity

Jo

completely (only Cu9.0-SSZ-13-Ⅰ, Cu4.8-SSZ-13-Ⅰ, and Cu2.8-SSZ-13-Ⅰ are shown in Fig. 4a), the occurrence of which is mainly attributed to the high ion-exchange level of those samples (above 78%).[8] In order to improve the hydrothermal stability, it is necessary to further decrease the Cu/Al ratio. Considering that the treatment of Cu-SSZ-13 by high HNO3 concentration can result in dealumination of the zeolite

framework, NH4NO3 solution was used as a second post-treatment agent to exchange the Cu2+ ions from Cu-SSZ13-I. 3.2 Second treatment in NH4NO3 solution The Cu4.8-SSZ-13-I sample with the highest crystallinity was selected to study the second post-treatment process in NH4NO3 solutions with concentration from 0.01 to 0.1 M. In this case, as expected, a series of Cux-SSZ-13-II with different Cu/Al

ro of

ratio were obtained, while keeping an almost constant Si/Al ratio (Table 1). The

higher the concentration of NH4NO3 solution, the lower the Cu loading and Cu/Al

ratio were. Such decreased Cu content slightly decreased the low-temperature NOx

-p

conversion of fresh Cux-SSZ-13-II (Fig. 4b). It should be noted, however, that all the

re

NH4NO3 post-treated samples, except for Cu1.0-SSZ-13-Ⅱ, still exhibited outstanding deNOx activity, with nearly 100% conversion above 225 ℃. After hydrothermal aging

lP

at 750 ℃ for 16 h (Fig. 4c) and at 800 ℃ for 5 h (Fig. 4d), a decrease of deNOx

na

activity was observed over the whole temperature range when compared to the fresh catalysts. In both cases, the activities of Cu-SSZ-13-II with Cu loading from 1.0 to 2.6

ur

wt.% markedly decreased, while Cu3.8-SSZ-13-II still exhibited relatively high activity, with NOx conversion above 90% at temperatures from 250 to 450 ℃.

Jo

Furthermore, the GHSV was slightly decreased to 200,000 h-1 to evaluate the Cu3.8SSZ-13-II and Cu2.6-SSZ-13-II catalysts in a wider temperature range (125 to 600 ℃). As shown in Fig. S1, both catalysts showed excellent deNOx activity with NOx conversion of ~100% when the temperature was above 175 ℃. After hydrothermal aging, above 80% NOx conversion was still maintained, indicating that the post-

treated in-situ synthesized Cu-SSZ-13 zeolites were outstanding NH3-SCR catalysts. Based on the above results, therefore, decreasing the Cu loadings by NH4NO3 after HNO3 post-treatment created a convenient way to increase the hydrothermal stability and high-temperature deNOx activity of in-situ synthesized Al-rich Cu-SSZ-13 zeolite catalysts. XRD measurements were carried out for these catalysts to analyze the degradation

ro of

of the zeolite structure caused by hydrothermal aging, which is shown in Fig. 5. After hydrothermal aging at 750 ℃, the Cu4.8-SSZ-13-Ⅰ catalyst shows none of the typical diffraction peaks of CHA, indicating the collapse of the zeolite structure, thus

-p

resulting in total loss of deNOx activity in the NH3-SCR reaction. After the NH4NO3

re

post-treatment, however, the obtained samples showed no observable change in XRD peak intensities, indicating that the long-range order of the structure remained intact.

lP

As for the samples post-treated by NH4NO3, such characteristic peaks also remained

na

even after severe hydrothermal aging at 800 ℃, while exhibiting decreased peak intensity. As shown in Fig. 5, furthermore, the decreased peak intensity was more

ur

pronounced for the sample with high Cu loading. This result indicates that the presence of a large amount of Cu2+ ions promoted the destruction of the zeolite

Jo

structure under severe hydrothermal aging conditions. To further detect the change in the Al local environment induced by NH4NO3 post-

treatment and hydrothermal aging, solid state 27Al MAS NMR measurements were conducted. As shown in Fig. 6, the fresh catalysts showed a primary peak at chemical shift of 58 ppm and a slight peak at 0 ppm, which were attributed to tetrahedral

aluminum incorporated into the framework and extra-framework octahedral aluminum, respectively [33]. By comparing peak intensities, it was easily found that there were no clear differences in the Al local environment among the fresh catalysts. After being hydrothermally aged at 750 ℃, all the catalysts exhibited quite different NMR spectra. A serious decay of the peak at 58 ppm was observed for the aged samples with Cu content below 2.6 wt.% and the one without NH4NO3 treatment,

ro of

indicating the occurrence of pronounced dealumination. Such hydrothermal

dealumination was also observed for the samples Cu2.6-SSZ-13-Ⅱ and Cu3.8-SSZ-13Ⅱ, while a strong peak at 58 ppm still remained after hydrothermal aging. These

-p

results clearly suggest that the presence of Cu in suitable amounts protects framework

re

Al and inhibits dealumination during hydrothermal aging. The process of dealumination was also revealed by the appearance of peaks at 48 and ~ 33 ppm,

lP

ascribing to partially framework bonded pentahedral aluminum [33, 34]. The amount of extra-framework Al (peak at ~33 and 0 ppm) became more pronounced after

na

hydrothermal aging at 800 ℃, which was related to CuOx formation. The relationship

ur

between CuOx formation and dealumination will be described in the discussion section.

Jo

Besides the degradation of the zeolite framework structure, changes in Cu2+ ions as

well as acid sites could also lead to a decrease in NOx conversion after hydrothermal aging. To probe the acid sites and Cu2+ ions, DRIFTS characterization was conducted. Fig. 7 presents the NH3-adsorption DRIFTS spectra in the N-H bonding region (17001400 cm-1) and zeolite T-O-T bond vibration region (1080-820 cm-1). The bands at

1632 and 1475 cm-1 are attributed to molecular NH3 absorbed onto exchanged Curelated Lewis acid sites and NH4+ interacting with Brønsted acid sites (Si-OH-Al), respectively [9, 35-38]. For the fresh catalysts, it can be seen that the number of Brønsted acid sites (Si-OH-Al) increased with decreasing Cu loading, accompanied by a decrease in exchanged Cu-related Lewis acid sites, indicating that the NH4+ replaced the Cu2+ ions during the NH4NO3 post-treatment and transformed to

ro of

Brønsted acid sites after calcination. After hydrothermal aging at 750 ℃, the Cu4.8-

SSZ-13-Ⅰsample totally lost its acid sites. The Cux-SSZ-13-Ⅱ catalysts showed a

slight decrease in the Cu-related acid sites (1632 cm-1), indicating that a few active

-p

Cu2+ ions transformed to CuOx. The Cu-SSZ-13 samples with low Cu loadings (2.1-

re

1.0 wt.%), however, showed a significant decrease in Brønsted acid sites, which is consistent with the dealumination results observed by 27Al-NMR. Simultaneously,

lP

considerably more acid sites were lost when the hydrothermal aging temperature

na

increased to 800 ℃, indicating that much larger amounts of CuOx clusters were formed and more severe dealumination occurred.

ur

Furthermore, the change of Cu2+ ions on Cu-SSZ-13 was identified by NH3adsorption DRIFTS in the range 1080-820 cm-1. In this IR range, the bands at 950 and

Jo

912 cm-1 are ascribed to the T-O-T vibration perturbed by [Cu(OH)+]-Z in cha cages and isolated Cu2+-2Z in d6r cages, respectively [10, 12, 30, 35]. After NH3 adsorption on Cu2+ ions, previous studies [9, 35] revealed that the interaction between Cu2+ ions and zeolite would be weakened, thus resulting in the appearance of negative peaks assignable to T-O-T vibrations. As shown in Fig. 7, these two negative peaks induced

by NH3 adsorption were clearly observed for the fresh Cu4.8-SSZ-13-Ⅰ and Cu3.8SSZ-13-Ⅱcatalysts, which in turn, reveals that two differentiable types of Cu2+ ions, Cu(OH)+-Z and Cu2+-2Z species, are present. With this in mind, it is reasonable that decreased Cu loading always resulted in lower intensity for both the negative peaks, particularly for that related to the Cu(OH)+-Z species. As for the samples with Cu loading below 2.1 wt.%, indeed, the high-frequency peak almost disappeared, while

ro of

the low-frequency one was still observed. This result indicates that the Cu(OH)+-Z was ion-exchanged first by NH4NO3 solution and then Cu2+-2Z followed, which

showed the opposite order to that observed when Cu2+ ions were ion-exchanged into

-p

the SSZ-13 substrate [8, 39]. After hydrothermal aging at 750 ℃, the characteristic

re

frequency assignable to T-O-T vibration disappeared for the Cu4.8-SSZ-13-Ⅰ catalyst, further confirming the collapse of the zeolite framework structure (as shown in XRD

lP

profiles). After hydrothermal aging at 750 ℃, however, the negative peaks at 950 and

na

912 cm-1 were clearly observed for the Cu-SSZ-13-Ⅱ catalysts with Cu loadings from 3.8 to 2.1 wt.%, while exhibiting a slight decreased intensity in comparison with the

ur

corresponding fresh samples, indicating that a small amount of CuOx clusters formed during hydrothermal aging at 750 ℃. Similar to the fresh Cu1.7-SSZ-13-II, the

Jo

negative peak attributed to Cu(OH)+-Z was hardly observed for the aged sample, while the peak relating to Cu2+-2Z species was still present, with a decreased intensity. For the aged Cu1.0-SSZ-13-II sample, it should be noted that both the negative peaks disappeared, indicating that all these Cu2+-2Z accumulated to form CuOx clusters. When the samples were subjected to hydrothermal aging at 800 ℃, a similar tendency

was observed, while exhibiting a more significant decrease in the peak intensities at 950 and 912 cm-1, demonstrating that a large amount of CuOx clusters formed. Compared to the Cu4.8-SSZ-13-I and Cux-SSZ-13-II catalysts, it was found that suitable post-treatment led to more active Cu2+ ions, especially Cu2+-2Z, being preserved after hydrothermal aging. To investigate the Cu species, H2-TPR measurements were also performed for fresh

ro of

and hydrothermally aged Cux-SSZ-13-Ⅱcatalysts, with the results shown in Fig. 8. At low temperatures, the fresh Cu4.8-SSZ-13-I catalysts showed three peaks at

~165 ℃, ~200 ℃, and ~340 ℃, which were ascribed to reduction of CuO, Cu(OH)+-Z

-p

in cha cages, and isolated Cu2+-2Z in d6r cages, respectively [10, 12, 30, 39]. As seen

re

in the figure 8, the CuOx reduction peak at ~165 ℃ disappeared after NH4NO3 treatment. Meanwhile, post-treatment of Cu4.8-SSZ-13-I with NH4NO3 sharply and

lP

primarily decreased the intensity of the peak of Cu(OH)+-Z, which was followed by a

na

decrease of Cu2+-2Z species. Indeed, the Cu4.8-SSZ-13-I and Cu3.8-SSZ-13-Ⅱ catalysts showed the characteristic peaks of both Cu(OH)+-Z and Cu2+-2Z species,

ur

while the Cu1.7-SSZ-13 and Cu1.0-SSZ-13 samples only showed the peak associated with Cu2+-2Z, which was in good agreement with the NH3-DRIFTS results.

Jo

Simultaneously, all the catalysts showed a high temperature reduction peak of Cu+ to Cu0 from 800 to 880 ℃, which indicated that the Chabazite structure began to degrade [22, 29]. As revealed by the H2 consumption quantities shown in Fig. S2, all the fresh catalysts showed approximately equal relative H2 consumption intensity below and above 500 ℃, indicating the absence of CuOx after NH4NO3 treatment, thus resulting

in nearly 100% high-temperature NOx conversion as shown in Fig. 4b. After undergoing hydrothermal aging at 750 ℃, almost all of the Cu2+ ions in Cu4.8SSZ-13-Ⅰtransformed to CuOx clusters, as seen by the absence of a peak for hightemperature reduction of Cu+ to Cu0. Therefore, the hydrothermally aged Cu4.8-SSZ13-Ⅰat 750 ℃ lost its activity completely. For the other catalysts, however, a broadening effect was observed for the peak after hydrothermal aging compared with

ro of

the results for the fresh samples, suggesting the presence of a less symmetric

coordination environment for Cu2+, derived from the dealumination process that

occurred during hydrothermal aging. After hydrothermal aging, meanwhile, the peak

-p

representing CuOx clusters at ~295 ℃ increased, especially for the Cu3.8-SSZ-13-Ⅱ,

re

Cu2.6-SSZ-13-Ⅱ and Cu2.1SSZ-13-Ⅱ samples with relatively high Cu loadings. The formation of CuOx clusters could also be identified by calculation of the H2

lP

consumption at low (< 500 ℃) and high (> 500 ℃) temperatures, as shown in Fig. S2.

na

The extra H2 consumption at low temperatures indicated the formation of CuOx clusters for the Cu3.8-SSZ-13-Ⅱ, Cu2.6-SSZ-13-Ⅱ, and Cu2.1SSZ-13-Ⅱ catalysts. In

ur

addition, a shift of the high-temperature reduction peak (Cu+ to Cu0) to lower temperature was observed for the hydrothermally aged samples compared with the

Jo

corresponding fresh ones, indicating the low stability of Cu+ induced by dealumination during the process of hydrothermal aging. After hydrothermal aging at 800 ℃, quite different Cu2+ reduction behaviors were

observed, with results shown in Fig. 8. On one hand, the low-temperature reduction peak broadened more and the high temperature peak shifted to lower temperature due

to the more severe dealumination. On the other hand, a strong CuOx reduction peak (~295 ℃) was observed, especially for the Cu-SSZ-13-Ⅱcatalysts with high Cu loading. The formation of CuOx clusters induced by hydrothermal aging could also be revealed by comparing the H2 consumption at low and high temperatures (Fig. S2). For these samples with high Cu loading, however, it should be noted that the active Cu2+ ion species were still preserved despite the formation of many CuOx clusters.

ro of

This feature was more pronounced for Cu3.8-SSZ-13-Ⅱ, giving a strong peak

assignable to active Cu(OH)+-Z species (~230 ℃). With this in mind, it is reasonable that Cu3.8-SSZ-13-Ⅱ shows deNOx activity even after hydrothermal aging at 800 ℃,

-p

as shown in Fig. 4.

re

To quantify the relative amounts of active Cu2+ ions, furthermore, electron paramagnetic resonance (EPR) was carried out at -120℃ for fully hydrated samples,

lP

because all active Cu-ions, including [Cu(H2O)6]2+ and [Cu(OH)(H2O)5]+ ions, are

na

EPR active, while CuOx clusters are EPR silent [10, 29]. Fig. 9 presents EPR spectra of the fresh and hydrothermally aged Cu-SSZ-13 samples. Furthermore, the

ur

relationship between the EPR signal intensity of the peak at 2620 G and the Cu loading was plotted and depicted in Fig. S3. As for the series of fresh catalysts, it can

Jo

be seen that the EPR signal intensity increases linearly with the Cu loading, indicating that all Cu species are EPR active Cu2+-ions. It should be noted that Cu4.8-SSZ-13-I included some CuOx although it still showed a linear increase, indicating that only a small amount of CuOx was formed, which can also be observed in the XRD and H2TPR results (Fig. 2 and Fig. 3). After hydrothermal aging at 750 ℃, however, the

EPR signal decreased, since the EPR active Cu2+-ions converted to CuOx clusters, especially for Cu4.8-SSZ-13-Ⅰ. After hydrothermal aging, meanwhile, the peaks at low magnetic field broadened, probably due to greater Cu2+ ion mobility. This enhanced mobility of Cu2+ ions resulted from the weakened limiting effect of framework aluminum [40] due to the dealumination that occurred during hydrothermal aging. On increasing the hydrothermal aging temperature to 800 ℃, the

ro of

EPR signal decreased dramatically, indicating that more CuOx clusters formed. As

shown in Fig. S3, there is a bend in the curves of hydrothermally aged samples at high Cu loadings, which indicates that considerably more CuOx clusters formed on the

-p

high Cu loading samples. It should be noted that, however, the EPR signal increased

re

in all cases with the increase of Cu loading, whether for fresh samples or those hydrothermally aged at 750 and 800 ℃, except for the Cu4.8-SSZ-13-Ⅰ catalyst. This

lP

indicates that high Cu loading samples are beneficial to preserve more active Cu2+

na

ions, although more CuOx clusters are formed on these samples after hydrothermal aging, which is consistent with the H2-TPR results. Therefore, the deNOx conversion

ur

increased with increasing Cu loading whether the catalysts were hydrothermally aged or not, which was due to the greater amount of preserved active Cu2+ ions for high

Jo

Cu-loading Cux-SSZ-13-Ⅱ catalysts after hydrothermal aging. 4. Discussion

4.1 Optimization of post-treatments by HNO3 and NH4NO3. Cu-SSZ-13 with high activity and high hydrothermal stability is a prerequisite for its practical application in diesel engine exhaust purification. The Cu-SSZ-13 catalyst

prepared by the economical and facile in-situ synthetic method provides an opportunity for NOx reduction with high deNOx efficiency due to rich Al presence [13, 27], while its hydrothermal stability is not high enough to withstand hydrothermal aging at 800 ℃. During the process of hydrothermal aging, dealumination from the zeolite framework and transformation of active Cu2+ ions to CuOx are thought to be the main factors contributing to deterioration of the catalytic

ro of

performance of Cu-SSZ-13 for the NH3-SCR reaction [8, 10, 25].

Figures S4-S6 show the HAADF-STEM images of as-synthesized Cu-SSZ-13

zeolites with the corresponding element mapping analysis. As shown in Figure S4,

-p

copper accumulation was clearly observed without the presence of Al, which

re

indicated the presence of CuOx clusters in the as-synthesized Cu10.0-SSZ-13-initial zeolite. However, after 0.1 M HNO3 treatment (Figure S5), the Cu and Al elements

lP

showed uniform distributions, therefore inhibiting the formation of CuOx clusters

na

during subsequent calcination. This is consistent with the XRD (Figure 2) and H2TPR results (Figure 3). Interestingly, after 1.0 M HNO3 treatment (Figure S6), the Cu

ur

and Al were uniformly dispersed in the center of the particles, while the framework Si was uniformly dispersed in the whole particle. This indicated that the dealumination

Jo

occurred from the outside in. After dealumination, the Cu species coordinated with Al are removed by HNO3 solution, therefore resulting in a slight increase of the Cu/Al ratio. Moreover, the Cu-TEPA template was also partially removed by HNO3 treatment when dealumination occurred, as shown in Figure S7. Therefore, Cu-SSZ-13 samples with high crystallinity were achieved by using a

HNO3 post-treatment with suitable concentration (0.05-0.1 M) due to the inhibition of copper accumulation. During hydrothermal aging, it was also observed that Cu2+ ions were easily accumulated in the Cu-SSZ-13 catalysts with a high Cu/Al ratio [8]. Under the premise of achieving good deNOx activity in the NH3-SCR reaction, therefore, decreasing the Cu/Al ratio is an efficient way to avoid the accumulation of Cu2+ ion species. By calculating the H2 consumption (Fig. 3), interestingly, it could be

ro of

seen that the relative amount of active Cu2+ ions species increased, which is due to the

high copper species dispersion (as shown in the HAADF-STEM images), even though the total Cu content was decreased by the HNO3 post-treatment, with concentration

-p

from 0 (Cu10.0-SSZ-13-initial) to 0.03 M (Cu4.7-SSZ-13-Ⅱ). Zhang et al also reported

Cu2+ ions in Cu-SSZ-13 [41].

re

that dilute HNO3 could eliminate excess Cu and optimize the spatial distribution of

lP

In summary, the HNO3 post-treatment has two main functions here: (1) to increase

na

the crystallinity of the zeolite structure; (2) to rearrange the Cu species distribution and increase the number of active Cu2+ ion species. These two enhancements make in-

ur

situ synthesized Cu-zeolites exhibit better activity and hydrothermal stability. We also used HNO3 to post-treat in-situ synthesized Cu-SAPO-34 made by using morpholine

Jo

(MOR) and Cu-TEPA as co-templates. As shown in Fig. 10, both HNO3-treated CuSAPO-34 samples showed higher NOx conversion than the untreated ones, whether they were hydrothermally aged or not. Therefore, the HNO3 treatment is essential to achieve in-situ synthesized Cu-zeolites with high NH3-SCR activity and hydrothermal stability.

In this study, however, a small amount of CuOx accumulated inevitably for the HNO3-treated Cu-SSZ-13 catalysts, before the increasing concentration of HNO3 caused zeolite dealumination (HNO3 concentration > 0.05 M). The HNO3-optimized Cu-SSZ-13-I catalysts showed outstanding deNOx activity in the NH3-SCR process (Fig. 4a), while their high temperature NOx activity and hydrothermal stability were still low due to their relatively high Cu/Al ratios of above 0.39 (Table 1). It was

ro of

apparent that increased HNO3 concentration could decrease the Cu loading at the

expense of dealumination, the occurrence of which resulted in serious damage to the zeolite skeleton. In this case, the Cu/Al ratios of HNO3-treated samples increased,

-p

deteriorating their activity and stability.

re

Therefore, NH4NO3 treatment was used to effectively decrease the Cu loading while avoiding the removal of Al from the zeolite skeleton (Table 1). With increasing

lP

NH4NO3 concentration, as expected, the Cu/Al ratios decreased from 0.41 to 0.08.

na

The formation of CuOx disappeared after NH4NO3 treatment, which increased the high temperature NOx conversion significantly (shown in Fig. 4b). Therefore, such

ur

decrease in Cu/Al ratio distinctly elevated the hydrothermal stability of Cu-SSZ-13 catalysts (750 ℃ and 800 ℃), which can be easily seen in Fig. 4c and Fig. 4d. After

Jo

hydrothermal aging, further measurements of XRD, 27Al-NMR, NH3-adsorption DRIFTS, H2-TPR and EPR revealed that the HNO3-NH4NO3 treated Cu-SSZ-13 catalysts possessed a more hydrothermally stable zeolite structure and Cu2+ ion species. The effect of NH4NO3 treatment is, therefore, to effectively decrease the Cu/Al ratio, eliminate the CuOx and avoid the transformation of active Cu2+ ion

species to CuOx clusters during hydrothermal aging. In another aspect, it was found that the H2 consumption peaks (200 and 340 ℃) over Cu4.8-SSZ-13-Ⅰshifted to intermediate temperatures (230 and 330 ℃), which indicated that the stability of Cu2+ ion species changed after the NH4NO3 treatment. The shift of the two peaks between Cu4.8-SSZ-13-Ⅰ and HNO3-NH4NO3 post-treated Cux-SSZ-13-Ⅱcatalysts demonstrated that a multi-step process occurred during the

ro of

NH4NO3 treatment, including rearrangement or movement of active Cu2+ ions and an ion-exchange process during NH4NO3 solution treatment, rather than a single step.

Moreover, after NH4NO3 treatment, the Cu(OH)+-Z species decreased first, then the

-p

Cu2+-2Z species, indicating that the less hydrothermally stable Cu(OH)+-Z species

re

were preferentially exchanged with NH4+ during the NH4NO3 treatment. The reserved Cu2+-2Z species are stable and coordinated with paired Al, which inhibited the

lP

accumulation of CuOx and dealumination of the zeolite structure during hydrothermal

na

aging.[10, 42] In summary, HNO3 and NH4NO3 can both optimize the Cu2+ ion species distribution and decrease the CuOx formation. Meanwhile, the stability of the

ur

zeolite structure and Cu2+ ion species was increased via HNO3 and NH4NO3 posttreatment, respectively.

Jo

Furthermore, we attempted to obtain optimized Cu-SSZ-13 with only one after-

treatment by using a mixed solution of HNO3 and NH4NO3. After co-treatment (80 ℃5 h) with HNO3 (0.1 M) and NH4NO3 (0.02 M) of in-situ synthesized Cu-SSZ-13 sample, the optimized Cu-SSZ-13 also showed excellent NH3-SCR performance and hydrothermal stability. Even after hydrothermal aging at 750 ℃, the optimized Cu-

SSZ-13 maintained above 90% NO conversion at a wide temperature range between 200 and 500 ℃ (Figure S8). 4.2 New insights into the effect of Cu loading on hydrothermal stability. Nam and co-workers reported that for Cu-SSZ-13 with a suitable Si/Al ratio, the Cu-SSZ-13 catalyst with low Cu/Al ratio exhibited the highest hydrothermal stability [8]. In the current study, however, Cux-SSZ-13-II catalysts with higher Cu/Al ratio

ro of

showed better hydrothermal stability, except for the Cu4.8-SSZ-13 catalyst. We

ascribed this phenomenon to the fact that the high Cu loading samples preserved more active Cu2+ ions after hydrothermal aging, as shown in NH3-DRIFTS, H2-TPR and

-p

EPR results. In detail, although the hydrothermally aged Cu3.8-SSZ-13-Ⅱ at 800 ℃

re

suffered from structure damage more heavily compared to other relatively low-Culoading catalysts, as shown by the XRD results (Fig. 6), the presence of more

lP

preserved active Cu2+ ion species, which was shown in DRIFTS, H2-TPR and EPR

maintained.

na

results, could catalyze the NOx reduction as long as the zeolite structure was

ur

During hydrothermal aging of Cu-SSZ-13 catalysts, in fact, CuOx formation is a key issue resulting in deterioration of the zeolite structure and decrease in the NH3-

Jo

SCR performance. The size of formed CuOx clusters was larger than the primary pore openings of SSZ-13, the diffusion of which could cause structural damage [10]. The CuOx formation was closely related to the Cu loading and the conditions of hydrothermal aging. For example, though hydrothermally aged at 750 ℃, the Cu4.8SSZ-13-Ⅰsample showed a large amount of CuOx cluster formation (shown in H2-

TPR and EPR results), which caused the zeolite structure to collapse (shown in XRD results). However, for the case of Cu-SSZ-13 with low Cu loading, the amount of CuOx formation was not enough to cause the collapse of the zeolite structure, although it increased with increasing Cu loading. These formed CuOx clusters caused degradation of the framework structure, and the long-range order of the structure was destroyed more heavily with increasing Cu loading (shown in XRD results) after

ro of

hydrothermal aging at 800 ℃, but the framework structure remained. Therefore, the preserved zeolite framework structure and the active Cu2+ ion species, which were

proved to be present by H2-TPR and EPR results, could catalyze the NOx reduction

-p

even though the samples were hydrothermally aged at 800 ℃. Nevertheless, Nam and

re

co-workers investigated the hydrothermal stability at 850 ℃, and in this case the formed CuOx clusters almost caused the zeolite structure to collapse for Cu-SSZ-13

lP

with relatively high Cu loading, which nearly lost its deNOx activity totally in the NH3-SCR reaction compared to the hydrothermal aging conditions of 750 ℃ and

na

800 ℃ [8]. Therefore, this indicated that it is the Cu-SSZ-13 zeolite framework

ur

structure that is more prone to collapse with increasing Cu loading. As long as the zeolite framework structure remained after hydrothermal aging, the high Cu loading

Jo

Cu-SSZ-13 catalysts showed better deNOx activity due to having greater amounts of preserved active Cu2+ ions species. However, once the formed CuOx clusters accumulated to the threshold value that leads to zeolite structure collapse, the CuSSZ-13 catalyst lost its NH3-SCR performance entirely. In another aspect, from the viewpoint of 27Al-NMR results (Fig .7), a protective

effect of Cu2+ on framework Al was observed due to the increasing tetrahedral aluminum peak intensity with increasing Cu loadings in the hydrothermally aged samples. Therefore, a mutually-inverse relationship can be deduced for the effects of copper species on the Cu-SSZ-13 catalyst hydrothermal stability (Fig. 11). On the one hand, Cu2+ ions stabilize the framework Al and protect the zeolite structure from dealumination during hydrothermal aging (Fig. 11a); on the other hand, excess Cu2+

ro of

ions easily form CuOx clusters, which destroy the zeolite structure, which benefits

Cu2+ accumulation in turn (Fig. 11c). Therefore, there was an optimal Cu/Al ratio to

avoid dealumination and CuOx formation simultaneously during hydrothermal aging

-p

(Fig. 11b). This work showed that the optimized Cu/Al ratio is ~0.31 and ~0.22 on the

re

basis of Si/Al of ~5 (Cu3.8-SSZ-13-II and Cu2.6-SSZ-13-II as shown in Fig. 5). 5. Conclusions

lP

Post-treatments are important and necessary steps for in-situ synthesized Cu

na

zeolitic catalysts to guarantee their high deNOx activity and hydrothermal stability in the NH3-SCR reaction. Via precise adjustment of the process of post-treatment by

ur

HNO3 and NH4NO3, optimized Cu-SSZ-13 with Cu/Al of ~0.31 and ~0.22 and Si/Al of ~5 was developed with outstanding deNOx performance in NH3-SCR before and

Jo

after hydrothermal aging at 750 and 800 ℃. On the one hand, Cu2+ ions not only serve as active sites for NOx reduction in Cu-SSZ-13 catalysts, but also inhibit the dealumination of the zeolite structure during hydrothermal aging. On the other hand, however, Cu2+ ions transform to CuOx clusters during hydrothermal aging, resulting in destruction of the zeolite skeleton and leading to a breakdown in the long-range

order of the structure. Such damage was serious for the Cu-SSZ-13 catalysts with high Cu loading due to greater CuOx formation. However, unless the typical zeolite structure collapses due to CuOx cluster formation during hydrothermal aging, increasing the Cu loading as high as possible is an efficient way to increase the

Jo

ur

na

lP

re

-p

ro of

hydrothermal stability of in-situ synthesized Cu-SSZ-13.

Author contribution Yulong Shan prepared the catalyst and performed the characterizations, catalytic tests and wrote the paper. Jinpeng Du, Yunbo Yu and Wenpo Shan participated the discussion and analysis of the experimental data. Xiaoyan Shi and Hong He designed this research, analyzed the data.

Notes

ro of

The authors declare no competing financial interest. Declaration of interest statement

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.

-p

ACKNOWLEDGMENTS

re

This work was financially supported by the National Natural Science Foundation of

Jo

ur

na

lP

China (21637005, 21777174, 21906172) and the K. C. Wong Education Foundation.

References [1] D. Jo, T. Ryu, G.T. Park, P.S. Kim, C.H. Kim, I.-S. Nam, S.B. Hong, Synthesis of High-Silica LTA and UFI Zeolites and NH3–SCR Performance of Their Copper-Exchanged Form, ACS Catalysis, 6 (2016) 2443-2447. [2] J.H. Lee, Y.J. Kim, T. Ryu, P.S. Kim, C.H. Kim, S.B. Hong, Synthesis of zeolite UZM-35 and catalytic properties of copper-exchanged UZM-35 for ammonia selective catalytic reduction, Applied Catalysis B: Environmental, 200 (2017) 428-438. [3] T. Ryu, N.H. Ahn, S. Seo, J. Cho, H. Kim, D. Jo, G.T. Park, P.S. Kim, C.H. Kim, E.L. Bruce, P.A. Wright, I.S. Nam, S.B. Hong, Fully Copper-Exchanged High-Silica LTA Zeolites as Unrivaled Hydrothermally Stable NH3-SCR Catalysts, Angewandte Chemie, 56 (2017) 3256-3260.

ro of

[4] Y. Shan, X. Shi, J. Du, Y. Yu, H. He, Cu-exchanged RTH-type zeolites for NH3-selective catalytic reduction of NOx: Cu distribution and hydrothermal stability, Catalysis Science & Technology, 9 (2019) 106-115.

[5] M. Moliner, C. Franch, E. Palomares, M. Grill, A. Corma, Cu-SSZ-39, an active and

hydrothermally stable catalyst for the selective catalytic reduction of NOx, Chemical communications,

-p

48 (2012) 8264-8266.

[6] N. Martín, C. Paris, P.N.R. Vennestrøm, J.R. Thøgersen, M. Moliner, A. Corma, Cage-based smallpore catalysts for NH3 -SCR prepared by combining bulky organic structure directing agents with

re

modified zeolites as reagents, Applied Catalysis B: Environmental, 217 (2017) 125-136. [7] J.H. Kwak, R.G. Tonkyn, D.H. Kim, J. Szanyi, C.H.F. Peden, Excellent activity and selectivity of Cu-SSZ-13 in the selective catalytic reduction of NOx with NH3, Journal of Catalysis, 275 (2010) 187-

lP

190.

[8] Y.J. Kim, J.K. Lee, K.M. Min, S.B. Hong, I.-S. Nam, B.K. Cho, Hydrothermal stability of CuSSZ13 for reducing NOx by NH3, Journal of Catalysis, 311 (2014) 447-457. [9] J. Luo, D. Wang, A. Kumar, J. Li, K. Kamasamudram, N. Currier, A. Yezerets, Identification of two

na

types of Cu sites in Cu/SSZ-13 and their unique responses to hydrothermal aging and sulfur poisoning, Catalysis Today, 267 (2016) 3-9.

[10] J. Song, Y. Wang, E.D. Walter, N.M. Washton, D. Mei, L. Kovarik, M.H. Engelhard, S. Prodinger,

ur

Y. Wang, C.H.F. Peden, F. Gao, Toward Rational Design of Cu/SSZ-13 Selective Catalytic Reduction Catalysts: Implications from Atomic-Level Understanding of Hydrothermal Stability, ACS Catalysis, 7 (2017) 8214-8227.

Jo

[11] Y. Shan, X. Shi, G. He, K. Liu, Z. Yan, Y. Yu, H. He, Effects of NO 2 Addition on the NH3-SCR over Small-Pore Cu–SSZ-13 Zeolites with Varying Cu Loadings, The Journal of Physical Chemistry C, 122 (2018) 25948-25953. [12] C. Paolucci, A.A. Parekh, I. Khurana, J.R. Di Iorio, H. Li, J.D. Albarracin Caballero, A.J. Shih, T. Anggara, W.N. Delgass, J.T. Miller, F.H. Ribeiro, R. Gounder, W.F. Schneider, Catalysis in a Cage: Condition-Dependent Speciation and Dynamics of Exchanged Cu Cations in SSZ-13 Zeolites, Journal of the American Chemical Society, 138 (2016) 6028-6048. [13] Z. Zhao, R. Yu, R. Zhao, C. Shi, H. Gies, F.-S. Xiao, D. De Vos, T. Yokoi, X. Bao, U. Kolb, M. Feyen, R. McGuire, S. Maurer, A. Moini, U. Müller, W. Zhang, Cu-exchanged Al-rich SSZ-13 zeolite from organotemplate-free synthesis as NH3 -SCR catalyst: Effects of Na+ ions on the activity and

hydrothermal stability, Applied Catalysis B: Environmental, 217 (2017) 421-428. [14] S.I.Zones, Conversion of Faujasites to High-silica Chabazite SSZ-13 in the Presence of N,N,NTrimethyl-1-adamantammonium Iodide., J.Chem. SOC. FARADAY. TRANS, 22 (1991) 3709-3716. [15] B. Chen, R. Xu, R. Zhang, N. Liu, Economical way to synthesize SSZ-13 with abundant ionexchanged Cu+ for an extraordinary performance in selective catalytic reduction (SCR) of NOx by ammonia, Environmental science & technology, 48 (2014) 13909-13916. [16] X. Wang, Q. Wu, C. Chen, S. Pan, W. Zhang, X. Meng, S. Maurer, M. Feyen, U. Muller, F.S. Xiao, Atom-economical synthesis of a high silica CHA zeolite using a solvent-free route, Chemical communications, 51 (2015) 16920-16923, [17] X. Xiong, D. Yuan, Q. Wu, F. Chen, X. Meng, R. Lv, D. Dai, S. Maurer, R. McGuire, M. Feyen, U. Mueller, W. Zhang, T. Yokoi, X. Bao, H. Gies, B. Marler, D.E. De Vos, U. Kolb, A. Moini, F. Xiao, Efficient and rapid transformation of high silica CHA zeolite from FAU zeolite in the absence of water

ro of

solvent, J. Mater. Chem. A, (2017).

[18] L. Ren, L. Zhu, C. Yang, Y. Chen, Q. Sun, H. Zhang, C. Li, F. Nawaz, X. Meng, F.S. Xiao,

Designed copper-amine complex as an efficient template for one-pot synthesis of Cu-SSZ-13 zeolite

with excellent activity for selective catalytic reduction of NOx by NH 3, Chemical communications, 47 (2011) 9789-9791.

-p

[19] R. Martínez-Franco, M. Moliner, C. Franch, A. Kustov, A. Corma, Rational direct synthesis

methodology of very active and hydrothermally stable Cu-SAPO-34 molecular sieves for the SCR of NOx, Applied Catalysis B: Environmental, 127 (2012) 273-280.

re

[20] N. Martín, M. Moliner, A. Corma, High yield synthesis of high-silica chabazite by combining the role of zeolite precursors and tetraethylammonium: SCR of NOx, Chemical communications, 51 (2015) 9965-9968.

lP

[21] L. Xie, F. Liu, L. Ren, X. Shi, F.S. Xiao, H. He, Excellent performance of one-pot synthesized CuSSZ-13 catalyst for the selective catalytic reduction of NOx with NH3, Environmental science & technology, 48 (2014) 566-572.

[22] L. Xie, F. Liu, X. Shi, F.-S. Xiao, H. He, Effects of post-treatment method and Na co-cation on the

na

hydrothermal stability of Cu–SSZ-13 catalyst for the selective catalytic reduction of NOx with NH 3, Applied Catalysis B: Environmental, 179 (2015) 206-212. [23] D.W. Fickel, E. D’Addio, J.A. Lauterbach, R.F. Lobo, The ammonia selective catalytic reduction 448.

ur

activity of copper-exchanged small-pore zeolites, Applied Catalysis B: Environmental, 102 (2011) 441[24] S.J. Schmieg, S.H. Oh, C.H. Kim, D.B. Brown, J.H. Lee, C.H.F. Peden, D.H. Kim, Thermal

Jo

durability of Cu-CHA NH3-SCR catalysts for diesel NOx reduction, Catalysis Today, 184 (2012) 252261.

[25] J.H. Kwak, D. Tran, S.D. Burton, J. Szanyi, J.H. Lee, C.H.F. Peden, Effects of hydrothermal aging on NH3-SCR reaction over Cu/zeolites, Journal of Catalysis, 287 (2012) 203-209. [26] R. Martínez-Franco, M. Moliner, J.R. Thogersen, A. Corma, Efficient One-Pot Preparation of CuSSZ-13 Materials using Cooperative OSDAs for their Catalytic Application in the SCR of NOx, ChemCatChem, 5 (2013) 3316-3323. [27] F. Gao, N.M. Washton, Y. Wang, M. Kollár, J. Szanyi, C.H.F. Peden, Effects of Si/Al ratio on Cu/SSZ-13 NH3-SCR catalysts: Implications for the active Cu species and the roles of Brønsted acidity, Journal of Catalysis, 331 (2015) 25-38.

[28] J. Kim, S.J. Cho, D.H. Kim, Facile Synthesis of KFI-type Zeolite and Its Application to Selective Catalytic Reduction of NOx with NH3, ACS Catalysis, 7 (2017) 6070-6081. [29] F. Gao, E.D. Walter, E.M. Karp, J. Luo, R.G. Tonkyn, J.H. Kwak, J. Szanyi, C.H.F. Peden, Structure–activity relationships in NH3-SCR over Cu-SSZ-13 as probed by reaction kinetics and EPR studies, Journal of Catalysis, 300 (2013) 20-29. [30] Y. Jangjou, Q. Do, Y. Gu, L.-G. Lim, H. Sun, D. Wang, A. Kumar, J. Li, L.C. Grabow, W.S. Epling, Nature of Cu Active Centers in Cu-SSZ-13 and Their Responses to SO2 Exposure, ACS Catalysis, 8 (2018) 1325-1337. [31] J. Park, H. Park, J. Baik, I. Nam, C. Shin, J. Lee, B. Cho, S. Oh, Hydrothermal stability of CuZSM5 catalyst in reducing NO by NH3 for the urea selective catalytic reduction process, Journal of Catalysis, 240 (2006) 47-57. [32] J.H. Kwak, D. Tran, J. Szanyi, C.H.F. Peden, J.H. Lee, The Effect of Copper Loading on the (2012) 295-301.

ro of

Selective Catalytic Reduction of Nitric Oxide by Ammonia Over Cu-SSZ-13, Catalysis Letters, 142 [33] S. Prodinger, M.A. Derewinski, Y. Wang, N.M. Washton, E.D. Walter, J. Szanyi, F. Gao, Y. Wang,

C.H.F. Peden, Sub-micron Cu/SSZ-13: Synthesis and application as selective catalytic reduction (SCR) catalysts, Applied Catalysis B: Environmental, 201 (2017) 461-469.

[34] M. Haouas, F. Taulelle, C. Martineau, Recent advances in application of 27Al NMR spectroscopy

-p

to materials science, Prog Nucl Magn Reson Spectrosc, 94-95 (2016) 11-36.

[35] Y. Shan, X. Shi, Z. Yan, J. Liu, Y. Yu, H. He, Deactivation of Cu-SSZ-13 in the presence of SO 2 during hydrothermal aging, Catalysis Today, (2017).

re

[36] D. Wang, L. Zhang, K. Kamasamudram, W.S. Epling, In Situ-DRIFTS Study of Selective Catalytic Reduction of NOxby NH3over Cu-Exchanged SAPO-34, ACS Catalysis, 3 (2013) 871-881. [37] I. Lezcano-Gonzalez, U. Deka, B. Arstad, A. Van Yperen-De Deyne, K. Hemelsoet, M. Waroquier,

lP

V. Van Speybroeck, B.M. Weckhuysen, A.M. Beale, Determining the storage, availability and reactivity of NH3 within Cu-Chabazite-based Ammonia Selective Catalytic Reduction systems, Physical Chemistry Chemical Physics, 16 (2014) 1639-1650.

[38] L. Zhang, D. Wang, Y. Liu, K. Kamasamudram, J. Li, W. Epling, SO 2 poisoning impact on the

na

NH3-SCR reaction over a commercial Cu-SAPO-34 SCR catalyst, Applied Catalysis B: Environmental, 156-157 (2014) 371-377.

[39] J. Hun Kwak, H. Zhu, J.H. Lee, C.H. Peden, J. Szanyi, Two different cationic positions in Cu-

ur

SSZ-13?, Chemical communications, 48 (2012) 4758-4760. [40] C. Paolucci, I. Khurana, A.A. Parekh, S. Li, A.J. Shih, H. Li, J.R.D. Iorio, J.D. AlbarracinCaballero, A. Yezerets, J.T. Miller, W.N. Delgass, F.H. Ribeiro, W.F. Schneider, R. Gounder, Dynamic

Jo

multinuclear sites formed by mobilized copper ions in NOx selective catalytic reduction, Science, (2017).

[41] T. Zhang, H. Chang, Y. You, C. Shi, J. Li, Excellent Activity and Selectivity of One-Pot Synthesized Cu-SSZ-13 Catalyst in the Selective Catalytic Oxidation of Ammonia to Nitrogen, Environmental science & technology, 52 (2018) 4802-4808. [42] Y. Shan, W. Shan, X. Shi, J. Du, Y. Yu, H. He, A comparative study of the activity and hydrothermal stability of Al-rich Cu-SSZ-39 and Cu-SSZ-13, Applied Catalysis B: Environmental, 264 (2020).

ro of

Jo

ur

na

lP

re

-p

Fig.1. XRD patterns of Cu-SSZ-13 catalysts post-treated by different HNO3 concentrations. The HNO3 concentration increases from 0.005 to 1.0 M from bottom up.

Fig. 2. (a) Relative crystallinity and (b) CuO peak areas of Cu-SSZ-13 catalysts post-treated by HNO3 solutions with different concentrations calculated based on XRD profiles (Fig. 1).

ro of

Jo

ur

na

lP

re

-p

Fig. 3. H2-TPR profiles (a) and H2 consumption intensity (b) of Cu-SSZ-13 catalysts post-treated by HNO3 solutions with different concentrations.

Fig. 4. NOx conversion activity in NH3-SCR reaction over the Cu-SSZ-13 catalysts with different treatments. (a) Fresh and 750 ℃ hydrothermally aged Cu-SSZ-13 after HNO3 post-treatment; (b)

ro of

Fresh Cu-SSZ-13 catalysts after HNO3-NH4NO3 post-treatment; (c) 750 ℃ hydrothermally aged Cu-SSZ-13 catalysts after HNO3-NH4NO3 post-treatment; (d) 800 ℃ hydrothermally aged CuSSZ-13 catalysts after HNO3-NH4NO3 post-treatment. GHSV = 400,000 h-1.

ur

na

lP

re

-p

Fig. 5. XRD patterns of fresh, 750 ℃ and 800 ℃ hydrothermally aged Cu4.8-SSZ-13-Ⅰ(a), Cu3.8SSZ-13-Ⅱ(b), Cu2.6-SSZ-13-Ⅱ(c), Cu2.1-SSZ-13-Ⅱ(d), Cu1.7-SSZ-13-Ⅱ(e), Cu1.0-SSZ-13-Ⅱ(f).

Jo

Fig. 6. 27Al MAS NMR spectra of fresh, 750 ℃ and 800 ℃ hydrothermally aged Cu-SSZ-13 catalysts

ro of -p re lP na

Jo

ur

Fig. 7. DRIFTS spectra after NH3 adsorption on Fresh, 750 ℃ and 800 ℃ hydrothermally aged Cu4.8-SSZ-13-Ⅰ(a), Cu3.8-SSZ-13-Ⅱ(b), Cu2.6-SSZ-13-Ⅱ(c), Cu2.1-SSZ-13-Ⅱ(d), Cu1.7-SSZ- 13Ⅱ(e), Cu1.0-SSZ-13-Ⅱ(f) at 1700 – 1400 cm-1 and 1080-820 cm-1. Conditions: 500 ppm NH3/N2, absorption temperature: 35 ℃

ro of

ur

na

lP

re

-p

Fig. 8. H2-TPR profiles of fresh, 750 ℃ and 800 ℃ hydrothermally aged Cu4.8-SSZ-13-Ⅰ(a), Cu3.8-SSZ-13-Ⅱ(b), Cu2.6-SSZ-13-Ⅱ(c), Cu2.1-SSZ-13-Ⅱ(d), Cu1.7-SSZ-13-Ⅱ(e), Cu1.0-SSZ-13Ⅱ(f).

Jo

Fig. 9. EPR profiles of fresh, 750 ℃, and 800 ℃ hydrothermally aged Cu4.8-SSZ-13-Ⅰ (a), Cu3.8SSZ-13-Ⅱ(b), Cu2.6-SSZ-13-Ⅱ(c), Cu2.1-SSZ-13-Ⅱ(d), Cu1.7-SSZ-13-Ⅱ(e), Cu1.0-SSZ-13-Ⅱ(f).

Jo

ur

na

lP

re

-p

ro of

Fig. 10. NOx conversion in NH3-SCR reaction over fresh (a) and hydrothermally aged (b) CuSAPO-34 catalysts prepared by in-situ synthetic method.

Fig. 11. The deactivation mechanism of hydrothermal aging of Cu-SSZ-13 with different Cu contents

Table 1. ICP-AES results of Cu-SSZ-13 series catalysts. HNO3

NH4NO3

concentration

concentration

(mol/L)

(mol/L)

166





0.77

154

0.005



4.5

0.67

134

0.01



5.1

4.5

0.39

78

0.03



4.9

4.9

0.39

78

0.05



4.7

5.0

5.1

0.40

80

0.07



Cu4.8-SSZ-13-Ⅰ

4.8

5.0

5.3

0.41

82

0.1



Cu4.6-SSZ-13-Ⅰ

4.6

4.6

5.3

0.42

84

Cu3.3-SSZ-13-Ⅰ

3.3

3.1

9.5

0.46

92

Cu2.8-SSZ-13-Ⅰ

2.8

2.5

11.2

0.48

96

Cu2.5-SSZ-13-Ⅰ

2.5

2.2

13.9

0.49

98

Cu3.8-SSZ-13-Ⅱ

3.8

5.2

5.1

0.31

62

Cu2.6-SSZ-13-Ⅱ

2.6

5.0

5.5

0.22

44

Cu2.1-SSZ-13-Ⅱ

2.1

5.2

5.0

0.17

Cu1.7-SSZ-13-Ⅱ

1.7

5.1

5.2

0.14

Cu1.0-SSZ-13-Ⅱ

1.0

5.3

5.0

0.08

(wt.%)

(wt.%)

Cu10.0-SSZ-13-initial

10.0

5.1

4.5

0.83

Cu9.0-SSZ-13-Ⅰ

9.0

4.9

4.4

Cu7.9-SSZ-13-Ⅰ

7.9

5.0

Cu4.7-SSZ-13-Ⅰ

4.7

Cu4.5-SSZ-13-Ⅰ

4.5

Cu4.7-SSZ-13-Ⅰ

Cu/Al

na ur Jo

exchange a

0.3



0.5



0.7



1.0



0.1

0.01

0.1

0.02

-p

(%)

34

0.1

0.03

28

0.1

0.05

16

0.1

0.1

re

lP

Ion exchange = Cu/Al * 200%

Si/Al

ro of

Al

Samples

a

Ion

Cu