Electrochemically Initiated Synthesis of Methanesulfonic Acid

Abstract

The direct sulfonation of methane to methanesulfonic acid was achieved in an electrochemical reactor without adding peroxide initiators. The synthesis proceeds only from oleum and methane. This is possible due to in situ formation of an initiating species from the electrolyte at a boron‐doped diamond anode. Elevated pressure, moderate temperature and suitable current density are beneficial to reach high concentration at outstanding selectivity. The highest concentration of 3.7 M (approximately 62 % yield) at 97 % selectivity was reached with a stepped electric current program at 6.25–12.5 mA cm⁻², 70 °C and 90 bar methane pressure in 22 hours. We present a novel, electrochemical method to produce methanesulfonic acid, propose a reaction mechanism and show general dependencies between parameters and yields for methanesulfonic acid.

The synthesis of methanesulfonic acid was achieved from only methane and oleum in an electrochemical reactor. At the boron‐doped diamond anode the species is formed, which initiates the catalytic cycle where methane is selectively sulfonated to the product. A concentration of 3.7 M (approximately 62 % yield) was achieved at 70 °C and 90 bar methane pressure in 22 hours.

Introduction

Selective functionalization of methane is still one of the major challenges of chemical research due to its low reactivity compared to its oxidation products. ^[1] However, being abundant, inexpensive and even a renewable resource, it could be a great feedstock for a number of useful products that can either serve as fuels or represent central commodities in the chemical industry. ^[2]

In order to avoid the standard conversion route via syngas formation, various concepts, ^[3] including direct oxy‐functionalization ^[4] or coupling, ^[5] are under extensive investigation. The main research focus is on alternative pathways to produce methanol. Several direct oxidation methods, e.g. on zeolites, ^[6] or oxidation to methyl esters, ^[7] e.g. the Periana‐system, ^[8] were developed and optimized. Most of the very innovative developments, however, can still not meet the demands for an industrial application, especially high TOF at high selectivity. Interestingly, the two most recently developed processes of methane functionalization that are promising for industry have reaction products other than methanol. Those are the oxidative methane coupling as developed at Linde and Siluria ^[9] as well as a process developed by Díaz‐Urrutia and Ott for the production of methanesulfonic acid (MSA), ^[10] which is also the product of our process.

Due to its unique properties, MSA is already used in some processes, and the demand is expected to further increase in the near future. ^[11] The numerous advantages of MSA are: strongly acidic without being oxidizing, low vapor pressure and odorless, low toxicological risk, splendid solubility of its salts, high chemical stability, and additionally it is biodegradable. ^[12] All these characteristics make MSA an attractive agent for a variety of chemical reactions. Some areas wherein MSA is already used are e.g. the electronics industry for electroplating, cleaning processes, metal recycling or in ionic liquids for a number of other processes.[ ¹¹ , ¹² ]

The industrial production of MSA is so far usually achieved by the oxidation of thiols or disulfides. ^[13] As these two types of molecules have to be synthesized first, those are multi‐step processes. In addition, by‐products can be generated which is in general unfavorable. A more attractive pathway is the direct sulfonation of methane, which is known for more than two decades. ^[14] In most of the published methods, the sulfonation takes place in oleum and is initiated by a metal peroxo‐ or peroxosulfate species. The most recent development, the mentioned method by Díaz‐Urrutia et al., is carried out with the initiator monomethylsulfonylperoxide sulfuric acid, which can be synthesized from hydrogen peroxide and MSA. ^[10] The process works at moderate temperature, i.e. 50 °C, and at high methane pressure, i.e. 100 bar. The sulfur trioxide concentration of the used oleum is variable and it even works in pure SO₃.

Over the past two decades, electro‐organic synthesis gained tremendous interest in technical and academic research and led to several applications. ^[15] The replacement of stoichiometric oxidizers or reducing agents enables more sustainable routes for redox reactions. ^[16] Beside the inherent safety of such conversions they can pay off economically. ^[17]

In recent studies on electrochemical methyl bisulfate synthesis, we also found MSA as a side product. ^[18] The reaction was carried out at high temperature in concentrated sulfuric acid using a boron‐doped diamond (BDD) anode. This innovative and durable electrode material is commonly used for the generation of oxyl species at very positive potentials. ^[19] The electrolysis of sulfates in aqueous solution on these anode types is a common way to synthesize peroxosulfates,[ ^19a , ²⁰ ] which can serve as initiators for the direct methane sulfonation. Thus, we assume that an initiator is formed in situ, which will induce the MSA formation reaction. Based on those prior observations, we investigated the influence of different conditions on this reaction and eventually developed a method to directly synthesize MSA in an electrochemical reactor. In our system, only methane and fuming sulfuric acid are required as starting materials without addition of an initiator or other additives. After extended reaction times, we were able to obtain high concentrations of MSA with almost exclusive selectivity.

Results and Discussion

The synthesis of methanesulfonic acid from methane and fuming sulfuric can be carried out under moderate reaction conditions using a BDD anode. The reactor used is an adapted version of a pressurized electrolysis cell that was described in a recent publication on electrochemically initiated methyl bisulfate formation. ^[18] Simplified, an H‐cell‐type glass vessel is placed inside an autoclave that has two electrodes attached to the cap. The glass inlet is filled with electrolyte prior to an experiment and electrodes are immersed while closing the autoclave. The geometrical surface area of the used working anode is 1.6 cm² in this reactor setup. Literature data from classical thermochemical methane sulfonation suggests that high methane pressure is beneficial, which was also confirmed in our experiments (see Figure S2).[ ¹⁰ , ^14a , ^14b , ^14c , ^14d ] Additionally, in literature moderate temperatures are suggested. Hence, the initial experiments were carried out at 90 bar methane pressure and 70 °C. Since initial voltammograms did not reveal any significant oxidation waves, which could be used to adjust a specific voltage for the experiments (see Figure S3) a constant current density of 1.25 mA cm⁻² was chosen. The cell voltage did adjust accordingly during the course of the experiments (see Figure S4). In Figure 1, the product distribution after different reaction times of the initial MSA synthesis experiment is displayed.

Figure 1.

Concentration of methanesulfonic acid (MSA) and methyl bisulfate (MBS) (left axis) and amount of MSA produced per electron passed during electrolysis (right axis). Conditions: 1.25 mA cm⁻² at 1.6 cm² BDD‐anode, 70 °C, 90 bar methane, fuming sulfuric acid with 20–30 wt % free SO₃.

At the chosen reaction conditions, the concentration of MSA steadily increases over the reaction time. The only detected side product in the liquid phase is methyl bisulfate (MBS), which even after longer reaction times does not reach significant concentrations. The selectivity towards MSA in this experiment is thus 97 % or higher. An additional important observation is the high amount of MSA that is formed per electron flowing in the cell. As depicted in Figure 1 on the right axis, this value is substantially higher than one for most data points, which means that the formation of MSA is an electrochemically initiated, thermocatalytic reaction. Since there is zero background activity without current (see Figure S5), an electrochemical step is required to initiate the reaction.

In classical methane sulfonation an initiator molecule is added to start the reaction.[ ¹⁰ , ^14a , ^14b , ^14c , ^14d ] As can be noted from the figure, only after a reaction time of one hour the concentration of MSA starts to increase significantly. This period could be necessary to form the initiator at the anode. The synthesis in this system is therefore proposed to consist of two parts: firstly the anodic formation of an initiator and secondly the actual catalytic cycle to form MSA.

One key element in the first part of the synthesis, the electrochemical initiation of MSA formation, is the use of a BDD‐anode. (Please see Figure S6 for a comparative experiment with a Pt/Ir‐anode). It is known that in sulfuric acid species can be formed ^[21] on BDD that will activate methane as is shown in Scheme 1 on the left. Please note that the presence of a free methyl radical is neither proven nor required in this reaction; however, for simplicity it is shown as such in the figure. The following catalytic cycle (see Scheme 1, right), which is the second part of the synthesis, starts by the reaction of the radical species with sulfur trioxide, as it was suggested by Li et al. ^[22] Additionally, further reactions are indicated in the figure which are the known anodic formation of peroxosulfuric acid, ^[23] the possible oxidation or decomposition of MSA, and the formation of methyl peroxosulfate species. All intermediates or side products can possibly decompose at the anode or in solution over time.

Scheme 1.

Proposed reaction pathways including part of the possible side reactions; mainly based on literature reports on the sulfonation of methane and persulfate production at BDD anodes.[ ²¹ , ²² , ²³ ]

Some of the proposed reactions could be identified by experiments that are outlined below. Due to the harsh and challenging reaction conditions and the experimental setup, it is, however, not possible to directly identify the proposed intermediates using operando spectroscopic techniques.

As stated, methane can be activated by radical species that are anodically formed from the electrolyte. If these species do not react with methane, they can form peroxosulfate species (see Scheme 1, left). In order to obtain insight into the mechanism, the procedure was separated in the two putative parts of the overall reaction, i.e. the electrolysis part, assumed to provide the initiator, and the pressurization of the reactor with methane to start the catalytic cycle. The fuming sulfuric acid was first electrolyzed for one hour at 10 mA in absence of methane (Q=36 C, V=3 mL); after disconnecting the potentiostat, the reactor was pressurized with methane. In order to determine the lifetime of potentially formed initiator species, different intervals between the end of electrolysis and the pressurization were chosen. During the delay time, the pre‐electrolyzed mixture was just stirred at reaction temperature. Additionally, the two steps were investigated separately: the products of the electrolysis step were analyzed by redox titration to identify potentially oxidizing species (which should be peroxides), and the reaction of methane was simulated by carrying out the sulfonation using potassium peroxodisulfate as an initiator.

The results, shown in Figure 2, match the mechanistic assumptions, which are the presence of peroxo species as active initiators and the decomposition of intermediates. On the left side, the dark blue colored columns display the obtained MSA concentration when the autoclave is pressurized only after the electrolysis. When the reactor is pressurized immediately after the electrolysis, the highest MSA concentration is obtained. MSA concentration is lower, if the system is pressurized with methane 30 min after the end of the electrolysis, and after a 60 min interval it has decreased significantly stronger. This indicates that the anodically formed species slowly decomposes. The first light blue column shows the MSA concentration reached in an experiment where potassium peroxodisulfate was used as an initiator. In this experiment, the initial concentration was chosen to correspond to the maximum persulfate concentration, which could have formed in the electrolysis experiments (left side), if all electrons passed had formed persistent persulfate species. However, the MSA concentration reached is substantially higher than after the electrolysis experiment. This suggests, that the persulfate species decompose to some extent. In order to check this, the electrolyzed mixture immediately after electrolysis was analyzed for persulfate concentration by redox titration. A persulfate concentration of only 6 mM was determined, i.e. compared to the theoretical value (i.e. at full Faradaic Efficiency) roughly ten percent of peroxo species were found in the mixture. This is due to two effects: the initiator decomposes over time; this means that possibly overall a higher FE is achieved but then immediately lowered due to decomposition of the species after formation. Secondly, a possible anodic side reaction that directly competes and lowers the FE is the oxygen evolution reaction (OER). The experiment with potassium persulfate as initiator was thus repeated with equally low potassium persulfate concentration (second light blue column). The obtained MSA concentration is then close to the one of the first electrolysis experiment, in which oleum was electrolyzed before adding methane. This supports the hypothesis that electrolysis leads to formation of a persulfate initiator, which serves as the starting point of the catalytic cycle for MSA formation.

Figure 2.

Dark blue: MSA concentration after reaction of methane with a pre‐electrolyzed mixture with different time intervals between electrolysis and pressurization. Light blue: MSA concentration after methane sulfonation initiated by K₂S₂O₈. Conditions: 70 °C, fuming sulfuric acid with 20–30 wt % free SO₃; Electrolysis: 6.25 mA cm⁻² at 1.6 cm² BDD‐anode, 1 h; Reaction of methane: 90 bar, 2 h.

In summary, two mechanistic elements can be inferred from this set of experiments: the reaction is initiated by anodically formed species, and these species are highly reactive and decompose over time, which is unavoidable, if no methane is present.

Since MSA may not be stable under highly oxidizing conditions at the anode in fuming sulfuric acid, the decomposition of MSA at different current density was studied. MSA was added to fuming sulfuric acid, which was then electrolyzed for seven hours in the absence of methane, so that decomposition of the product under reaction conditions would be detected as a loss in MSA. The results are depicted in Figure 3.

Figure 3.

Relative loss of MSA after 7 hours electrolysis time at different current density at 1.6 cm² BDD‐anode. Conditions: 70 °C, fuming sulfuric acid with 20–30 wt % free SO₃, initial MSA concentration: 575–600 mM.

The loss of product correlates with the current density. The only detected reaction product is CO₂ in the gas phase (see Figure S7). These results show that in the chosen reaction environment MSA is significantly decomposed only at high current density over longer reactions times. MSA is thus stable against anodic decomposition to an acceptable extent.

After gaining a basic understanding of the essential elements of the overall reaction, in the following the influence of reaction parameters will be discussed, starting with the influence of current. Normally in electrochemistry the potential is the most important control parameter. However, under the conditions used (elevated temperature, oleum as reaction medium) we were not able—in spite of various different attempts—to identify a suitable reference electrode for adequately determining the potential at the working electrode, and thus only the cell voltage could be measured in a two electrode arrangement. Therefore, the current flowing through the cell during electrolysis was taken as control parameter, keeping in mind that for different currents the cell voltage will be different. In Figure 4 the respective concentration of MSA after three hours reaction time at different current densities is shown.

Figure 4.

MSA concentration and MSA molecules formed per electron passed at different current density at 1.6 cm² BDD‐anode. The connecting line is just a guide to the eye. Conditions: 70 °C, 90 bar, 3 h, oleum with 20–30 wt % free SO₃.

After increasing strongly at lower current densities, the product concentration increases slower when current densities higher than 3.13 mA cm⁻² are applied, which goes along with a significantly decreased number of MSA formed per passed electron with increasing current density. While these results show that a certain current density is needed to produce sufficient initiator species, they also reveal that at higher current undesired reactions become more significant. Results for MSA concentrations reached for different current densities over time are shown in Figure 5.

Figure 5.

MSA concentration at different current density at 1.6 cm² BDD after different reaction times. Values in the upper right corner give the approximate yield of MSA with respect to SO₃ after 16 hours. The connecting lines are just a guide to the eye. Conditions: 70 °C, 90 bar, oleum with 20–30 wt % free SO₃.

The concentration of MSA increases over time at all applied current densities, eventually leading to high concentrations in the molar range. Whereas at 3.13 mA cm⁻² and 6.25 mA cm⁻² the product concentration is similar in the first few hours, 1.25 mA cm⁻² leads to a lower concentration at any point of the reaction. After a longer reaction time 6.25 mA cm⁻² leads to the highest MSA concentration and approximate yield with respect to SO₃. Increase of current density beyond 6.25 mA cm⁻² does not seem to bring an advantage: in the beginning MSA concentration seems to be a bit higher, but after three hours and beyond the product concentration lies in a similar range as with a current density of 6.25 mA cm⁻². This is due to competing reactions, i.e. product decomposition reactions or the oxygen evolution reaction, which becomes more significant at the higher cell voltage required to reach 15.63 mA cm⁻². When the experiment is run with a constant current density, 6.25 mA cm⁻² gives thus the best balance between a rapid formation of product and avoiding detrimental side reactions.

The rate of product formation is not only expected to depend on the formation of active initiator but depends also on the concentration of SO₃ (shown below). Thus, the decrease of the apparent product formation rate at higher currents and longer reaction times could additionally be related to a decreasing SO₃ concentration due to its consumption.

The results indicate that at short times high currents are beneficial for MSA formation. However, total oxidation leads to accelerated degradation by total oxidation at higher current density. Thus, medium‐high current density seems to be the the best compromise, as stated above. However, if one does not operate at constant current density, but with a stepped current approach, further optimization could be possible. The results of exploratory experiments in this direction are shown in Figure 6. For three different reaction times the MSA concentrations of a stepped current experiment are compared to a constant current experiment using 6.25 mA cm⁻².

Figure 6.

Product concentrations after applying a constant current density of 6.25 mA cm⁻² or a stepped current starting from 6.25 mA cm⁻² for the same total reaction time at 1.6 cm² BDD‐anode; shown for seven, sixteen and twenty‐two hours. For readability the current in the figure is shown as total current; 10 mA are 6.25 mA cm⁻², 25 mA are 15.63 mA cm⁻² The approximate according yield of MSA with respect to SO₃ of each concentration can be obtained from the right axis. Conditions: 70 °C, 90 bar, oleum with 20–30 wt % free SO₃.

In all three step‐experiments shown in Figure 6, the concentration of MSA increases by 10–25 % compared to the constant current experiment. It can therefore be assumed that after a certain progress of the reaction an increase of the current density is useful to further promote the formation of MSA.

Elevated temperature in general increases rates. However, due to the different side reactions that can occur it does not necessarily lead to higher MSA concentration and yield. Indeed, in previous publications on the purely thermochemical reactions it was reported that for this reaction a temperature optimum exists. The influence of temperature for the studied system is shown in Figure 7.

Figure 7.

MSA and MBS concentrations at different reaction temperatures. Conditions: 90 bar, 3 h, 3.13 mA cm⁻² at 1.6 cm² BDD‐anode, oleum with 20–30 wt % free SO₃.

Similar to the data in literature we observe a steep increase in MSA concentration up to 85 °C for a reaction time of 3 h. This can be attributed to increasing rate with increasing temperature for the main reaction pathway. Further increase leads to a decreased MSA concentration. At these temperatures, unfavorable reactions start to dominate. This could, on the one hand, be due to faster MSA oxidation to CO₂ with increasing temperature. On the other hand, also active intermediates and stable peroxo‐species, which are also active in this reaction,[ ¹⁰ , ^14d ] may decompose more rapidly to non‐active molecules at higher temperature. Additionally, at the two highest temperatures the selectivity towards MSA decreases, since MBS formation becomes significant.

According to the proposed catalytic cycle, the concentration of SO₃ should enter the rate law for the reaction and might become rate limiting under certain conditions. Furthermore, superacidic conditions are beneficial to activate methane. ^[24] SO₃ concentration could also affect the oxygen evolution reaction as a competing process. At higher SO₃ contents oxygen evolution might be suppressed somewhat, since SO₃ changes the species present in the electrolyte.

Since the conductivity of the electrolyte changes strongly with varying SO₃ concentration, ^[25] it was not possible to carry out all experiments at different SO₃ concentration at equal current density. These experiments were hence performed at equal cell voltage. The studied SO₃ concentration ranges from 1 % up to 65 %, and the results are shown in Figure 8; the exact values are given in Table S1.

Figure 8.

Product concentrations (blue/green), MSA molecules formed per electron (orange) and charge passed (red) at different SO₃ concentrations of the electrolyte; concentration ranges of SO₃ are given, because the starting fuming sulfuric acid contains 20–30 wt % free SO₃ according to the manufacturer. Conditions: 70 °C, 90 bar, 3 h, 1.65 V.

As discussed before, one key parameter for the formation rate of methanesulfonic acid is the current density, since the initiator is anodically formed. The total charge passed scales with the current density at the given cell voltage, and it has a maximum using the 20–30 wt % SO₃ electrolyte. This behavior correlates to the literature conductivity data of fuming sulfuric acid. ^[25]

The product concentration, selectivity and amount of MSA obtained per electron increase up to the medium SO₃‐concentration. Since the increase does by far exceed the increase in charge passed, the SO₃ content in the electrolyte is crucial responsible factor. At 44–49 % SO_3, the MSA concentration still increased, despite the sharp drop in charge passed. At 65 % SO₃ the obtained concentration of MSA is significantly decreased. However, also at high SO₃ concentrations SO₃ promotes the MSA formation, as the strongly increased amount of MSA obtained per electron shows. The formation of MSA at the higher SO₃‐concentrations is thus limited by the low conductivity and thus the low charge passed.

It is notable that the selectivity at the two higher SO₃‐concentrations decreases to the benefit of MBS formation, and that at these high SO₃ concentrations the amount of formed MBS does by far exceed the moles of electrons passed to the electrode. This is due to conversion of MSA to MBS at high SO₃ concentration, as proven by a test experiment starting from MSA (with c _0,MSA=575 mM in oleum 65 %, T=70 °C, p _methane=0 bar; c _t=3h,MSA=169 mM, c _t=3h,MBS=399 mM). The primary product of MBS formation was thus also MSA, and the amount of MSA obtained per electron is twice as high as it appears for the 65 wt % SO₃ experiment. This again underlines the beneficial effect of SO₃ on MSA formation.

In contrast, at lower SO₃ concentration, the MBS is formed directly from methane analogous to a previously described process. ^[18] Optimum selectivity is therefore reached at medium SO₃‐content, due to the fast reaction between SO₃ and the methyl radical intermediate species (see Scheme 1) on the one hand and an insufficiently high SO₃ concentration for further conversion of MSA on the other hand. Under the conditions used in this study (i.e. reactor design, temperature, cell voltage, reaction time) the use of oleum with 20–30 wt % SO₃ is the optimum to achieve a high MSA concentration and high selectivity at the same time.

Overall, the series of experiments with changed parameters support the mechanistic proposal outlined in Scheme 1. The catalytic reaction is initiated by anodically formed species that are precursors of persulfates. This starts a cycle, in which SO₃ and methane are the key reagents affecting the overall MSA formation, with reduced conductivity of the electrolyte at high SO₃ concentration as a complicating factor. The rate of MSA formation and the final MSA concentration can decrease due to the decomposition of intermediates and due to MSA decomposition. As a last note, one should also keep in mind the possibility of sulfur dioxide and oxygen formation in this system, both of which can act as inhibitors of the desired reaction. ^[10]

Some exploratory experiments were performed in a bigger electrolyzer, wherein the reaction volume was increased, which also allows active species to migrate further away from the electrode and thus possibly reduce decomposition. In the modified electrolyzer the H‐type cell glass inlet was replaced by a glass cylinder with larger volume, which was filled with 20 ml of fuming sulfuric acid instead of 3 ml (see Figure S9). No product decomposition at the counter electrode was observed in a control experiment (see Figure S8). The purpose of the modified design was an increase in electron yield by preventing direct anodic decomposition of intermediates in close proximity to the anode. In several experiments, MSA concentrations in almost molar range were achieved (see Figures S10 and S11). The absolute amount of MSA formed was significantly higher than in the H‐type cell experiments, equal in both, terminal cell voltage and current. Number of MSA molecules formed per electron could be increased by approximately forty percent. Although reactions in the bigger electrolyzer have not yet been studied systematically beyond the exploratory runs, these first results suggest that a change of reactor design has an influence on the overall performance and potentially opens the possibility to further optimize this promising method.

Conclusion

Methanesulfonic acid can be produced in an electrochemically initiated process from methane and oleum. No external peroxo‐species has to be added as an initiating agent for the reaction. The boron‐doped diamond anode is known to interact with sulfates to form persulfates at very positive potentials. In the system described here, these persulfates or radical precursors for persulfates activate methane, which then leads to the initiation of the catalytic cycle.

Similar to other systems that directly sulfonate methane, it is beneficial to use high methane pressure to accelerate the reaction, and to use moderate temperature to favor high product selectivity. A current density of 6.25 mA cm⁻² in the system was found to be optimal to ensure a rapid product formation and eventually reach a high MSA concentration and selectivity. If the current density is too high, side reactions will occur, which reduce turnover in terms of coulombic efficiency (i.e. MSA formed per electron passed to the anode). Too low current density, on the other hand, will lead to a decreased reaction rate and to a decreased product concentration and yield. The concentration of sulfur trioxide in the used oleum plays a crucial role in several respects. Rate generally increases with increasing SO₃ concentration; however, a too high SO₃ concentration leads to increased conversion of MSA to MBS and a decreased current response due to low electrolyte conductivity, which limits the reaction progress.

Two main undesired reactions were identified to limit the final MSA concentration: the decomposition of active intermediates and the decomposition of MSA itself. These can be reduced to some extent by carefully adjusting the current density (e.g. by changing the current during electrolysis) and the temperature. For further optimization, it is crucial to investigate the side reactions in order to be able to minimize their influence and thus reach complete conversion of SO₃ to the desired product.

In addition to these investigations, it is also of interest to further explore this system using a reactor and electrode of different size. Exploratory experiments have shown that the total amount of product formed and the number of product molecules per electron can be varied just by using a modified reactor design. In future studies a continuous set‐up that allows for parameter changes along the reactor could also be considered.

Conflict of interest

The authors declare no conflict of interest.

Supporting information

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re‐organized for online delivery, but are not copy‐edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

Supporting Information

J. Britschgi, W. Kersten, S. R. Waldvogel, F. Schüth, Angew. Chem. Int. Ed. 2022, 61, e202209591; Angew. Chem. 2022, 134, e202209591.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.