Developing a thermally stable Co/Ce-Sn catalyst via adding Sn for soot and CO oxidation

Summary

The thermal stability of the catalysts is of particular importance but still a big challenge for working under harsh conditions at high temperature. In this study, we report a strategy to improve the thermal stability of the ceria-based catalyst via introducing Sn. XRD, Rietveld refinement, and other characterizations results indicated that the formation of Sn-Co solid solution plays a key role in the thermal stability of the catalyst. The developed ternary 3%Co/Ce_0.5Sn_0.5O₂ catalyst not only exhibits outstanding thermal stability and resistance to SO₂ and H₂O for soot oxidation from diesel vehicle exhaust but also remains extraordinary thermal stability for CO oxidation. Remarkably, even after thermal aging at 1000°C, it still possessed high catalytic activity similar to that of the fresh catalyst.

Graphical abstract

Highlights

• •The developed 3%Co/Ce-Sn catalyst processes extraordinary thermal stability
• •The Sn species could restrain the aggregation of Co active component
• •The Sn-Co solid solution plays a key role in improving the thermal stability
• •The 3%Co/Ce-Sn catalyst exhibited perfect and stable resistance to H₂O and SO₂

Chemistry; Catalysis; Thermal property; Materials characterization

Introduction

With the rapid development of catalytic science and technology, catalytic materials are applied widely in environment, devices, biomedicine, and so on. During these applications, the catalytic materials are placed under all kinds of harsh conditions, such as high temperature environment and so forth. Therefore, thermal stability is often a crucial feature determining the practical application of catalysts (Xu et al., 2018; Yang et al., 2019; Liu et al., 2018; Yan et al., 2020; Wang et al., 2020; Zhang et al., 2017).

To improve the thermal stability of the catalysts, great efforts have been devoted. For example, Xu et al. reported a novel entropy-driven strategy to stabilize Pd single atom on the high-entropy fluorite oxides, which exhibited excellent resistance to thermal and hydrothermal degradation (Xu et al., 2020). In addition, modifying with dopants is also applied to resist the sintering of active components. For the three-way catalysts (TWCs) applied in the gasoline engine aftertreatment, Zr is added into CeO₂ to resist the aggregation of active components (Pt, Pd, and Rh) and stabilize the ability of oxygen storage (Farrauto et al., 2019; Monte et al., 2004). Compared with gasoline engines, diesel engines are widely used owing to high fuel efficiency, durability, and excellent dynamic performance. Yet, the development of aftertreatment catalysts with excellent thermal stability still remains a huge challenge in diesel emission control. When the DPF is regenerated to remove the collected particulate matter, the peak temperature inside the DPF can even reach 1000°C (Guo and Zhang, 2007; Yu et al., 2013a, 2013b). Thus, not only high catalytic activity but also good thermal stability is essential for the aftertreatment catalysts, such as CO oxidation and soot oxidation catalysts. For instance, a MnO_x-CeO₂-Al₂O₃ catalyst for soot oxidation reported by Wu et al. (2011) exhibited good thermal stability, due to the introduction of stabilized Al₂O₃ and its maximum soot oxidation only shifted upward by 53°C after aging treatment at 800°C for 20 h. In addition, Zr doping has been reported to be able to improve the thermal stability of CeO₂ calcined at 1000°C, but the T₅₀ of soot oxidation for the obtained Ce-Zr mixed oxide was only about 20°C lower than that without a catalyst (Atribak et al., 2008).

Above all, the development of the catalysts with excellent thermal stability is still urgently needed for industrial catalytic application, especially for those involving high-temperature environment. However, the previously reported aftertreatment catalysts usually cannot hold their activity after thermal aging even at 800°C. Therefore, the thermal stability of this type of catalysts remains a formidable challenge.

In our previous development of the catalysts for selective catalytic reduction of NO_x with NH₃ (NH₃-SCR), it was found that SnO₂ can remarkably promote a Ce-W mixed oxides catalyst for sintering resistance at high temperatures, which was related to the preservation of highly dispersed Ce-W species induced by Sn (Liu et al., 2021). In addition, it was reported that doping Sn could also improve the oxygen storage capacity (OSC) of CeO₂-based materials because the reversible Sn⁴⁺ ⇋ Sn²⁺ redox process involves two electrons transfer (Liu et al., 2015; Mukherjee et al., 2018). For instance, Chang et al. found Sn-modified MnO_x-CeO₂ catalysts presented nearly 100% NO_x conversion at 110°C–230°C, which is associated with that doping of Sn can promote the formation of oxygen vacancies and facilitate NO oxidation to NO₂ (Chang et al., 2013). In Sasikala’s work, it was found that a Ce-Sn mixed oxide exhibited better catalytic activity for CO oxidation reaction as comparing with their pure oxides due to a synergetic effect between Ce and Sn (Sasikala et al., 2001). Therefore, Sn is very attractive for the development of the catalysts with excellent activity and thermal stability. Herein, we report a facile strategy to develop outstanding thermally stable non-noble catalyst via introducing Sn. A series of Ce_1-xSn_xO₂ supports with different Ce/Sn molar ratios were prepared by a co-precipitation method, and then Co was loaded on the optimal Ce_0.5Sn_0.5O₂ support (Figure S1) by the impregnation method. All catalysts were calcined at 700°C for 3 h. Meanwhile, the optimal 3%Co/Ce_0.5Sn_0.5O₂ catalyst (Figure S2) was calcined at 500, 700, 900, and 1000°C for 3 h for the thermal stability tests.

Results and discussion

Figure 1 displays a series of elemental mapping images of the catalysts aging at different temperatures. As shown in Figure 1A, with the increase of aging temperature, Co₃O₄ phase over the 3%Co/CeO₂ catalyst was aggregated, indicating that aging treatment has a negative effect on the 3%Co/CeO₂. Yet, the introduction of Sn could restrain the aggregation of Co₃O₄, as confirmed by the elemental mapping and TEM imagines in Figures 1B, S3, and S4. In addition, the surface areas values of the as-prepared catalysts are presented in Table S1. The results showed that the specific surface areas of all the catalysts decreased with the increase of aging temperatures. Yet, after thermal aging at high temperatures (900 and 1000°C), the developed ternary 3%Co/Ce_0.5Sn_0.5O₂ catalyst exhibited larger surface area than 3%Co/CeO₂, indicating that the addition of Sn had a positive effect on the specific surface area.

Figure 1.

Element distributions measurements

EDS elemental mapping images of the (A) 3%Co/CeO₂ and (B) 3%Co/Ce_0.5Sn_0.5O₂ catalysts at different aging temperatures.

The crystal structures of the 3%Co/CeO₂ and 3%Co/Ce_0.5Sn_0.5O₂ catalysts subjected to different aging temperatures were examined by XRD (Figures 2, S5, and S6). For the 3%Co/CeO₂ catalyst calcined at 500°C, it exhibited the structures of cubic fluorite CeO₂ (PDF#43-1002), and no peaks of any Co species were observed. Figure 2A shows that there is no shift for the peak of (220) crystal face of CeO₂ in the 3%Co/CeO₂ catalyst, in comparation with the pure CeO₂ sample. These results indicated that Co species were highly dispersed on the CeO₂ surface rather than existing as Ce-Co solid solutions (He et al., 2018; Venkataswamy et al., 2015). Moreover, when the aging temperature increased to 700, 900, and 1000°C, cubic Co₃O₄ phase was detected, and the crystallite size of Co₃O₄ calculated by the Scherrer formula (in Table S1) also increased gradually for the 3%Co/CeO₂ catalyst (shown in Figure 2A), which can be attributed to the sintering of the active Co component.

Figure 2.

XRD analysis

XRD patterns of (A) 3%Co/CeO₂ and (B) 3%Co/Ce_0.5Sn_0.5O₂ catalysts at different aging temperatures.

Amazingly, as presented in Figures S2B and S5B, no peaks of the Co₃O₄ phase were found for the 3%Co/Ce_0.5Sn_0.5O₂ catalyst, even after thermal aging at 1000°C. The patterns of 3%Co/Ce_0.5Sn_0.5O₂ catalyst calcined at 700°C (shown in Figure S6A) mainly showed the typical peaks of CeO₂ (PDF#43-1002) and SnO₂ (PDF#41-1445). Figures S6B and S6C depict the enlarged drawing of XRD patterns, corresponding to the (220) facet of CeO₂ and the (211) facet of SnO₂ in the Ce_0.5Sn_0.5O₂ and 3%Co/Ce_0.5Sn_0.5O₂ catalysts. With the addition of Co, the peak of the (211) facet of SnO₂ shifted to higher diffraction angle, indicating the formation of Sn-Co solid solution. Yet, the peak of the (220) facet of CeO₂ shifted to lower diffraction angle, which could be due to the occurrence of Sn phase segregation from Ce-Sn solid solution (see Figures S7 and S8A in supplemental information). Figure 2B showed that the peak of the (220) facet of CeO₂ in aged 3%Co/Ce_0.5Sn_0.5O₂ shifted to lower diffraction, while the peak of the (211) facet of SnO₂ still located at higher diffraction angle than the Ce_0.5Sn_0.5O₂ catalyst, which suggested the still existence of Sn-Co solid solution. In addition, a new phase of Co₂SnO₄ was detected for the 3%Co/Ce_0.5Sn_0.5O₂ catalyst aging at 900°C and 1000°C (Figure S9), which further indicated that Co preferred to interact with Sn species.

Rietveld refinement was performed using Fullprof Suite to obtain precise lattice parameters (Rodriguez-Carvajal, 1993), and the results are described in Figures 3 and S8. Figure 3A shows the refined XRD patterns of the 3%Co/CeO₂ catalysts aging at different temperature and the a, b, and c values of CeO₂ were almost equal to the pure CeO₂ sample (shown in Figure S8A), which further revealed that Co₃O₄ ought to exist mainly as dispersed crystallites rather than Ce-Co solid solutions. As shown in Figure S8B, the c value of SnO₂ in 3%Co/Ce_0.5Sn_0.5O₂ catalyst was smaller than that in Ce_0.5Sn_0.5O₂, while the a, b, and c values of CeO₂ were slightly larger than those of the Ce_0.5Sn_0.5O₂ catalyst. These results indicated that the Co was doped to SnO₂ instead of CeO₂ lattice. Thus, the Sn-Co solid solution formed through the replacement of Sn⁴⁺ (0.071 nm) by smaller Co³⁺ (0.061 nm). The lattice parameters of CeO₂, SnO₂, and Co₂SnO₄ in the aged 3%Co/Ce_0.5Sn_0.5O₂ catalysts obtained by Rietveld refinement are shown in Figure 3B. The c values of SnO₂ in the aged 3%Co/Ce_0.5Sn_0.5O₂ are still smaller than that in the Ce_0.5Sn_0.5O₂ catalyst (shown in Figure S8A). It suggested that Sn-Co solid solution was not destroyed by high-temperature treatment. Additionally, the Co₂SnO₄ phase was also identified by Rietveld refinement, and the results are shown in Figure S8C. The above results indicated that the introduction of Sn could improve the thermal stability via the formation of Sn-Co solid solution to restrain the aggregation of Co₃O₄.

Figure 3.

Rietveld refinement of XRD patterns

Rietveld refined XRD patterns of the (A) 3%Co/CeO₂ and (B) 3%Co/Ce_0.5Sn_0.5O₂ catalysts at different aging temperatures.

X-ray photoelectron spectroscopy (XPS) technology was employed to explore the surficial properties of the as-prepared catalysts, and the results are shown in Figures 4 and S10. As the aging temperature increases, the proportion of Co²⁺/Co³⁺ and Ce³⁺/Ce⁴⁺ on the surface of the 3%Co/CeO₂ catalyst gradually decreased. As is well known, the higher the Ce³⁺ and Co²⁺ concentration in these samples, the more oxygen defects are generated (Zhao et al., 2019; Liu et al., 2009). Therefore, the surface oxygen defects associating with the Co²⁺ and Ce³⁺ of the 3%Co/CeO₂ catalyst gradually disappeared, with the increase of aging temperature (Zhao et al., 2019; Cui et al., 2020; Wang et al., 2018; Ren et al., 2019). Figure S10 showed that the O 1s spectra of all catalysts can be deconvoluted into two kinds of O species, with the peak at higher binding energy being attributed to the surface oxygen species (O_α) and the peak at lower binding energy being ascribed to the lattice oxygen species (O_β) (Xu et al., 2020; Ren et al., 2019; Sellers-Anton et al., 2020). The ratio of O_α/O_β over the 3%Co/CeO₂ catalyst decreased obviously with the increase of aging temperature. Yet, after introducing the Sn, the amount of surface Co²⁺, Ce³⁺, and O_α for the catalysts aged at different temperatures did not change significantly. In addition, the results of H₂-TPR (Figure S11) showed that with the addition of Sn, the H₂ consumption significantly increased, which indicated that Sn addition is beneficial to the formation of active oxygen species.

Figure 4.

Analysis of the surficial properties

The ratios of Co²⁺/Co³⁺, Ce³⁺/Ce⁴⁺, and O_α/O_β ratios of as-prepared catalysts at different aging temperatures.

According to the above characterization results, the addition of Sn can protect the active Co component of the 3%Co/Ce_0.5Sn_0.5O₂ catalyst via the formation of Sn-Co solid solution, and the aged catalysts exhibited similar amounts of surface O_α with the fresh one. Surface-active oxygen species play a significant role in many heterogeneous catalytic reactions, including solid-solid-gas and gas-gas-solid reactions. For solid-solid-gas reaction, soot oxidation is a typical represent and the catalyst would undergo high-temperature oxidation process in the actual application. Thus, we tested the catalytic performance of soot oxidation for the catalysts to verify whether the catalyst with Sn introduction possess stable catalytic activity. As shown in Figure 5, the 3%Co/Ce_0.5Sn_0.5O₂ catalyst indeed exhibited extraordinarily thermal stability, with high soot conversion activity still maintaining even after aging at 1000°C, indicating its high resistance to thermal shock. For a comparative study, we tested the soot oxidation performances of the 3%Co/CeO₂ catalyst aged at different temperatures. The results showed that the soot combustion temperature over the 3%Co/CeO₂ catalyst increased sharply with the increase of aging temperature, while that over 3%Co/Ce_0.5Sn_0.5O₂ did not show a noticeable increase after aging at different temperatures from 500°C to 1000°C. In addition, even after 1000°C aging, the CO₂ selectivity of the 3%Co/Ce_0.5Sn_0.5O₂ catalyst was still nearly 100%, while that of the 3%Co/CeO₂ catalyst decreased remarkably (Figure S12). These results indicate that Sn plays a key role in the thermal stability of the 3%Co/Ce_0.5Sn_0.5O₂ catalyst. As we all know, the presence of sulfur compounds in diesel fuels and water vapor in the exhaust gas is almost unavoidable. Therefore, tolerance of the SO₂ and H₂O in diesel exhaust also plays an important role in the application of a soot oxidation catalyst. Figures S13 and S14 showed that the 3%Co/Ce_0.5Sn_0.5O₂ catalyst possess high and stable resistance to H₂O and SO₂, which was of extraordinary importance for its practical application.

Figure 5.

Soot conversion of the catalysts with different aging temperatures during temperature-programmed oxidation of soot

Reaction conditions: 0.1% NO and 10% O₂ balanced by N₂, under loose contact mode, GHSV = 300,000 mL g⁻¹·h⁻¹, and heating rate = 10°C/min.

Moreover, there are also many gas-gas-solid reactions, in which the catalysts could be exposed to high-temperature conditions, such as catalytic elimination of emissions including CO, NO, HC, and so on from internal combustion engines. Therefore, CO-TPO experiment was also applied to further investigate whether the catalyst with doping Sn exhibited outstanding stability for CO oxidation, and the results are shown in Figure S15. It was found that the activity of 3%Co/CeO₂ decreased gradually with the aging temperature increasing from 500°C to 1000°C. Yet, the 3%Co/Ce_0.5Sn_0.5O₂ catalyst exhibited stable activity for CO oxidation, even after 1000°C aging (Figure S15B), which further confirmed that the introduction of Sn is the key to improve the stability of the catalyst.

Conclusion

In summary, we developed an extraordinary 3%Co/Ce_0.5Sn_0.5O₂ catalyst with outstanding thermal stability. The characterization results indicated that the introduction of Sn species induced the formation of Sn-Co solid solution to restrain the aggregation of the Co active component. Moreover, the results of soot and CO oxidations further illustrate the addition of Sn can remarkably improve the stability of the 3%Co/CeO₂ catalyst. This study offers an efficient strategy toward developing the catalyst with outstanding thermal stability. Further optimization of this catalyst and investigation of the reaction mechanism are underway.

Limitations of the study

In this work, it provides a new and efficient strategy toward developing the catalyst with excellent thermal stability. For the analysis of surficial properties of the as-prepared catalysts, only XPS was used. Based on previous studies, the Co 2p spectra were deconvoluted into Co ²⁺ and Co³⁺, and the O 1s spectra were deconvoluted into surface oxygen species (O_α) and lattice oxygen species (O_β). More evidence of surficial properties, such as in situ XPS, in situ XAFS analysis and so on, is needed to further strengthen the conclusion. In addition, the reaction mechanism of soot and CO oxidation over the developed catalyst should be investigated in future work.

STAR★Methods

Key resources table

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Chemicals, peptides, and recombinant proteins
Ce(NO3)3·6H2O	Tianjin Fuchen Chemical Reagent Factory	CAS 10294-41-4
SnCl4·5H2O	Shanghai Aladdin Biochemical Technology Co., Ltd	CAS 10026-06-9
Co(NO3)2·6H2O	Sinopharm Chemical Reagent Co., Ltd	CAS 10026-22-9

Resource availability

Lead contact

Further information and requests for resources should be directed to and will be fulfilled by the lead contact, Wenpo Shan (wpshan@iue.ac.cn)

Materials availability

Materials generated in this study will be made available on reasonable request, but we may require a payment or a completed Materials Transfer Agreement if there is potential for commercial application.

Method details

Catalyst preparations

The catalyst was synthesized by a co-precipitation method. Ce(NO₃)₃·6H₂O and SnCl₄·5H₂O, with different Ce/Sn molar ratios (Ce : Sn=3:1 of 6 g Ce(NO₃)₃·6H₂O and 1.61 g SnCl₄·5H₂O, Ce : Sn=1:1 of 6 g Ce(NO₃)₃·6H₂O and 4.85 g SnCl₄·5H₂O, Ce : Sn=1:3 of 2.49 g Ce(NO₃)₃·6H₂O and 6 g SnCl₄·5H₂O), were dissolved together in deionized water. Then the above aqueous solution was added dropwise into another mixed solution consisting of H₂O : NH₃·H₂O : H₂O₂ = 4:4:1 (H₂O=80 ml, NH₃·H₂O=80 ml, and H₂O₂=20 ml) under vigorous mixing. Next, the obtained suspension was ultrasonicated for 30 minutes, stirred for 1 hour, and washed to neutral pH. The as-prepared precipitate was dried at 110°C for 12 h, and then calcined at 700°C for 3 h in air. The catalysts were named as Ce_xSn_1-xO₂, which did not mean the Ce and Sn species existing as CeO₂-SnO₂ solid solution. CeO₂ and SnO₂ were also synthesized using the same method as reference samples.

Three different theoretical Co loadings, 1%, 3% and 5% with regards to the weight of the support (2 g), were used to investigate the effect of Co loading on the oxidation reaction. Co loading was performed via wetness impregnation with appropriate amounts of an aqueous solution of Co(NO₃)₂·6H₂O (0.099 g Co(NO₃)₂·6H₂O of 1%, 0.296 g Co(NO₃)₂·6H₂O of 3%Co, and 0.494 g Co(NO₃)₂·6H₂O of 5%). The obtained mixture was ultrasonicated for 30 minutes and stirred for 1 h. Then, the mixture was subjected to a rotary evaporation process to remove water. Finally, the obtained powders were dried at 110°C overnight, then the powders were calcined at 700°C for 3 h in air. For comparison, the 3%Co/CeO₂ catalyst was synthesized using the same methods as above. The obtained catalysts were named as 1%Co/Ce_0.5Sn_0.5O₂, 3%Co/Ce_0.5Sn_0.5O₂, 5%Co/Ce_0.5Sn_0.5O₂, and 3%Co/CeO₂, respectively. In addition, the 3%Co/Ce_0.5Sn_0.5O₂ and 3%Co/CeO₂ catalysts were also calcined at 500, 900, and 1000°C for 3 h, respectively, for the thermal stability tests.

Characterizations of the catalysts

The X-ray diffraction (XRD) patterns were obtained on an X'Pert PRO instrument with Cu K_α radiation (λ= 1.5418) at 40 kV and 40 mA. Data were collected between 2θ=10-90°, with a scan step of 0.07°.

The Brunauer-Emmett-Teller (BET) method was applied to measure the specific surface areas of the samples via N₂ adsorption-desorption isotherms at − 196°C using an ASAP 2020N autoscore surface analyzer. The pore volume and pore-size were obtained by the BJH method using the desorption branch.

The morphologies and element mapping of the as-prepared catalysts were obtained with a high-resolution transmission Tecnai FEI-F20 electron microscope.

X-ray photo electron spectroscopy (XPS) results of the catalysts were obtained on a scanning X-ray micro probe (ESCALAB250I, Thermo Fisher scientific) using Al K α radiation. All of the binding energies were calibrated using the C 1s peak (BE = 284.8 eV) as standard.

H₂-TPR was performed using a Micromeritics Auto Chem II 2920 automatic chemical adsorption analyzer. After pre-treatment under 5% O₂ balanced by He at 500°C for 1 h, the 100 mg sample was heated from room temperature to 900°C at a constant rate (10°C/min) in a U-shaped quartz reactor under 10% H₂/Ar mixture gas (30 ml/min). The hydrogen consumption was monitored with a TCD detector, which was calibrated by the signal generated by the introduction of the known amounts of hydrogen.

Catalytic activity measurements

Printex-U (Degussa) was used as model soot, with particle size and specific surface area of 25 nm and 100 m²/g, respectively. The catalytic activity was evaluated by a temperature programmed oxidation reaction, which was carried out in a continuous flow fixed bed reactor consisting of a quartz tube with 5 mm inner diameter. To make evaluation conditions close to actual practice, the loose contact was employed. Under the loose contact, the catalyst and soot were mixed for 2 minutes with a spatula. The mass ratio of catalyst to soot was 10:1, then 110 mg of the admixture was added to 300 mg of quartz sand with particle size from 40 to 60 mesh, in order to reduce air resistance and guard against reaction run off.

The reaction gas, including 1000 ppm NO (when used), 50 ppm SO₂ (when used), 10% O₂, and 5% H₂O (when used), with N₂ as balanced gas, was flowed into the fixed bed reactor at a total flow rate of 500 ml/min. The reaction temperature was increased from room temperature to 700°C at a heating rate of 10°C/min, which was measured by a K-type thermocouple in the fixed bed reactor. The outlet gases in the soot catalytic oxidation process, such as CO₂, CO, NO and NO₂, were detected by an Antaris IGS (Thermo Fisher) equipped with a heated, low-volume multiple-path gas cell (2 m).

• Soot conversion and CO₂ selectivity were calculated as follows:

$$\text{Soot}\text{Conversion}=\frac{A_i}{A_t}\times 100\%$$ (Equation 1)

The catalytic activity was described by the values of T₁₀, T₅₀, and T₉₀, which were defined as the temperatures at 10%, 50%, and 90% soot conversion, respectively.

$$C{O}_2\text{Selectivity}=\frac{A_{iC{O}_2}}{A_{iC{O}_2}+A_{iCO}}\times 100\%$$ (Equation 2)

A_i is the total peak areas of CO₂ and CO at a given temperature, A_t is the total peak areas of CO₂ and CO overall, A_iCO2 is the total peak area of CO₂ at a given temperature, and A_iCO is the total peak area of CO at a given temperature.

The CO oxidation activities for the prepared catalysts were also evaluated in a continuous flow fixed bed reactor using 100 mg catalysts in a gas mixture of 0.4% CO/10%O₂/N₂ at a flow rate of 500 ml/min. The reaction temperature was increased from room temperature to 700°C at a heating rate of 10°C/min, which was measured by a K-type thermocouple in the fixed bed reactor. The outlet gases, such as CO₂ and CO, were detected by an Antaris IGS (Thermo Fisher) equipped with a heated, low-volume multiple-path gas cell (2 m).

• CO conversion was calculated as follows:

$$\mathit{C}\mathit{O}\text{Conversion}=\frac{C{O}_{in}-C{O}_{out}}{C{O}_{in}}\times 100\%$$ (Equation 3)

Quantification and statistical analysis

Our study doesn’t include quantification and statistical analysis.

Published: April 15, 2022

Data and code availability

Data reported in this paper will be shared by the lead contact upon reasonable request.

This study does not report original code.

Any additional information required to reanalyze the data reported in this paper is available from the Lead Contact upon reasonable request.