Report

Activation of methane to CH3+: A selective industrial route to methanesulfonic acid

See allHide authors and affiliations

Science  22 Mar 2019:
Vol. 363, Issue 6433, pp. 1326-1329
DOI: 10.1126/science.aav0177

Methane oxidation on the plus side

Industrial conversion of methane to alcohol derivatives involves a circuitous route that starts with overoxidation to carbon monoxide. More direct approaches in highly acidic media have shown promise at small scale but are not cost-effective. Díaz-Urrutia and Ott describe a reaction at pilot-plant scale that combines methane and sulfur trioxide directly in sulfuric acid to form methanesulfonic acid with no by-products (see the Perspective by Schüth). The reaction appears to proceed via a cationic chain mechanism initiated by a low concentration of added sulfonyl peroxide and propagated by CH3+.

Science, this issue p. 1326; see also p. 1282

Abstract

Direct methane functionalization to value-added products remains a challenge because of the propensity for overoxidation in many reaction environments. Sulfonation has emerged as an attractive approach for achieving the necessary selectivity. Here, we report a practical process for the production of methanesulfonic acid (MSA) from only two reactants: methane and sulfur trioxide. We have achieved >99% selectivity and yield of MSA. The electrophilic initiator based on a sulfonyl peroxide derivative is protonated under superacidic conditions, producing a highly electrophilic oxygen atom capable of activating a C–H bond of methane. Mechanistic studies support the formation of CH3+ as a key intermediate. This method is readily scalable with reactors connected in series for prospective production of up to 20 metric tons per year of MSA.

Widespread application of fracking techniques and biogas production has provided access to large quantities of inexpensive methane. Methane’s largest chemical transformations remain confined to the highly energy demanding steam-reforming and Fischer-Tropsch processes (Eqs. 1 and 2). However, CH4 functionalization to more complex molecules is limited because of overoxidationCH4(g) + H2O(g) → CO(g) + 3H2(g)(1)(2n + 1)H2(g) + nCO(g) → CnH(2n + 2)(g) + nH2O(g)(2)and higher reactivity of the potential products than that of the starting materials (1, 2). Major efforts to overcome this challenge have been made (1, 314); however, the industrial applicability of these processes is restricted by economic constraints, scalability challenges, and low selectivity. The only industrial process for the direct valorization of CH4 to high–value-added chemicals is the oxidative coupling of CH4 using heterogeneous metal catalysts (14). More recently, environmental concerns have boosted the search for new applications of CH4 beyond its current use as a heating and hydrogen source (15). In this context, sulfonation of CH4 to methanesulfonic acid (MSA) has received substantial attention owing to the abundance of both raw materials and the potential for rapid integration into current industrial chemical processes (4). MSA is biodegradable and nonoxidizing, with current uses in the pharmaceutical and electroplating industries (50 metric kilotons/year), as well as prospective applications in metal recycling, energy storage, and biodiesel (5). Early work by Snyder et al. demonstrated oxidation of CH4 to mixtures of oxygenated and sulfonated products (6), employing fuming sulfuric acid (oleum) at high temperatures (325°C) with HgSO4 as a catalyst. Later, Periana et al. further developed the Hg-mediated electrophilic C–H activation of CH4 in H2SO4 at 180°C, producing SO2 and methyl bisulfate (CH3OSO3H; MBS) (7). Later work by Periana and Schüth and colleagues using Pt complexes (8) and Pt salts (9) in H2SO4 and oleum also produced high yields of MBS and SO2, respectively, at high temperatures. Bell (10, 11) and Sen (12) and colleagues focused on methanesulfonation in oleum using peroxo salts and different metal and nonmetal additives as radical initiators. The sulfonation is initiated in this case by the thermal decomposition of peroxosulfates, generating sulfate radical anions (16, 17). Nevertheless, these reactions suffered from low yields and conversions as a result of free-radical recombination (e.g., ethane formation) and undesired side reactions (1013), rendering them unsuitable for large-scale production. In this regard, superacid chemistry (e.g., Hammett acidity function H0 < −12) provides the balance between reactivity and selectivity that an industrial process requires (18). We report here that treatment of oleum (e.g., 20 to 60% SO3) with CH4 at ~100 bar (50°C) using <1 mol % of electrophilic initiator affords MSA (Eq. 3) in 99% yield with 99% selectivity

CH4(g) + SO3(l) → CH3SO3H(l)(3)

We studied the reaction in a batch system to optimize the reaction conditions and gain further insight into the reaction mechanism. The electrophilic initiator contains various sulfonyl peroxide derivatives prepared as a mixture of H2O2, MSA, and H2SO4 (19); monomethylsulfonylperoxide sulfuric acid [HOS(O)2OOS(O)2CH3; 1 (MMSP)] is the reactive species (fig. S1). Previously, we reported that employing bis(methylsulfonyl) peroxide [H3CS(O)2OOS(O)2CH3; 2 (DMSP)] as the initiator led to substantial yields of MSA (20, 21); however, this work demonstrates that MMSP outperforms DMSP in terms of rates and technical feasibility. We observed lower productivities when using the 400-ml reactor compared with the 4-liter reactor, presumably due to the larger quantity of CH4 in the headspace of the larger reactor (Table 1, entries 1 and 2). Maintaining a constant amount of CH4 throughout the reaction, we obtained higher yields of MSA (entry 3) than those obtained via regular batch experiments (entry 2). The solubility of CH4 increases with increasing pressure, enabling higher concentrations of CH4 in the liquid phase (fig. S8). We then varied the reaction temperature from 25° to 85°C. More than 99.9% selectivity toward MSA was achieved at 50°C, whereas a more complex mixture of products (MBS, SO2, and methanesulfonic acid anhydride) typical of a radical pathway was obtained at 85°C (entry 4), presumably due to the thermal decomposition of the sulfonyl peroxide (16, 17). Consistently, low-temperature experiments (entry 5) afforded high conversions and selectivity toward MSA; however, long reaction times were needed (e.g., 70.2% yield at 25°C for 720 hours; fig. S5).

Table 1 Sulfonation of methane to MSA in batch reactors using MMSP prepared in situ.

PCH4 refers to initial CH4 pressure. Conversion (Conv.) was calculated in the batch experiments by using the volume (vol.) of the reactor headspace. Percentage yield of MSA (analyzed by ion chromatography) was based on initial amount of SO3. Selectivity (Sel.) was calculated based on moles of MBS. n.a., not applied; n.d., not detected.

View this table:

Experimentally, four major insights supporting a nonradical mechanism are evident from the data in Table 1. First, high yields and selectivities for MSA (entry 1) indicate a nonradical selectivity pattern without the formation of higher alkanes (e.g., radical recombination) or other sulfonated hydrocarbons in the liquid or gas phase, as observed by gas chromatography analysis (fig. S10). In contrast, previous studies showed that radical reactions tend to have lower yields and selectivities, generating a mixture of sulfonated alkanes with concomitant evolution of CO2 (1012). Second, ultraviolet (UV) light irradiation of the reaction mixture containing the electrophilic initiator is known to prompt the homolytic decomposition of the –O–O– bond. Indeed, sulfonylperoxides are well known to easily photodecompose to sulfate radical anions (e.g., S2O82−/SO52− → SO4•−) (16). After the reaction mixture was exposed to UV light (~190 nm) for 8 hours, no changes in the pressure of CH4 were observed, indicating the inability of the initiator (–O–O–) to form radicals under these conditions (entry 6). Third, the use of O2 as a deactivating agent does not affect the rates to the degree expected for a radical reaction (12). We observed similar deactivation effects with O2, SO2, and ethane when using electrophilic initiator 1 (fig. S4). The addition of 0.07 to 0.16 weight % (wt %) O2, SO2, ethane, or propane led to ~10% deactivation, presumably owing to the formation of weakly electrophilic species. The addition of 0.33 wt % SO2 decreased the yield of MSA to 21.5% (entry 7), compared with 58.7% yield under normal conditions (fig. S3). Fourth, increasing the amount of the electrophilic initiator mixture led to a notable increase in the reaction rate constant: 0.197 and 0.359 hours−1 for one- and twofold electrophilic initiators, respectively (entry 9). In contrast, radical initiators typically afford (under specific conditions) similar rates of product formation regardless of a change in concentration: 0.1 or 0.01 mmol of K2S2O8 produced 7.6 or 7.2 mmol of MSA, respectively (12).

Examination of the reaction profile reveals an induction period immediately after the addition of the electrophilic initiator (Fig. 1A, i). During the induction period, the amount of MSA produced is proportional to the initial quantity of MMSP, suggesting the initial formation of the active species (methyl cation). During the second period of the reaction profile (Fig. 1A, ii), CH4 consumption follows a pseudo–first-order rate until the critical pressure of CH4 (e.g., 60 bar) is reached, at which point the rate decreases. The solubility of CH4 in the reaction mixture decreases with the pressure; hence, lower yields of MSA are obtained at low pressures (fig. S8). Figure 1B depicts the reaction profile of the sulfonation of CH4 with successive additions of CH4, in which periods ii to v share the same slope value. Under these conditions, the reaction becomes pseudo–zero-order with respect to SO3 and CH4, until CH4 addition is arrested; at this point a decay pattern similar to that in Fig. 1A is observed (Fig. 1B, v).

Fig. 1 Reaction profile for the sulfonation of methane.

Pressure of CH4 is plotted versus time under (A) standard conditions using 0.9 mol % electrophilic initiator (Table 1, entry 2) and (B) successive additions of CH4 (Table 1, entry 3). The inset in (A) shows a zoomed-in view of region i.

The activation energy of this process, determined experimentally by varying the temperature of several batch experiments, was calculated to be 111 ± 1 kJ mol−1 (fig. S6), a value similar in magnitude to those of previously reported electrophilic hydride abstractions (22). Performing the sulfonation using D2SO4 (deuterated oleum: 33%) results in a KSIE (kinetic solvent isotope effect) value of kSolv H/kSolv D = 2.4 ± 0.17. This value, higher than previously reported solvation effect values, is consistent with the stabilization of a transition state with a highly ionic contribution (23). Deuterium incorporation is observed in 1H nuclear magnetic resonance spectra at 4.77 parts per million (CH2DSO3D; coupling constant JH-D = 2.7 Hz) and accounts for <1% of the total MSA. This indicates that H/D scrambling is not occurring by H-atom abstraction from CH4. On the basis of the extensive studies reported above, we propose the general catalytic cycle for the selective formation of MSA under superacidic conditions (Fig. 2A). The electrophilic initiator (1 or 2) is initially protonated to a peroxonium ion [R–OO+(H)–R; 3 (R:HOS(O)2, CH3S(O)2)] (2427), thus creating a superelectrophilic oxygen atom that triggers the oxidation of CH4 by electrophilic hydride abstraction to generate CH3+ (5) and water (27). The electrophilic initiator decomposes to MSA and 4 (MSA or H2SO4). The initial amount of 1 is ~0.9 mol%, based on the total amount of SO3, and hence the amount of CH3+ that enters the productive catalytic cycle is equimolar to 1. Addition of CH3+ across the S=O bond of SO3 generates the highly electrophilic intermediate CH3S(O)2O+ (7). We have ruled out the formation of a sulfite intermediate [CH3OS(O)2+] on the basis of decomposition of alkylsulfites to esters in SO3 and H2SO4. Intermediate 7 abstracts a hydride from CH4 to produce CH3SO3H and regenerates the cationic species CH3+. The cationic pathway highlighted above takes place under very specific conditions (Fig. 2B), wherein high selectivity is achieved through the greater degree of electronic changes in electrophilic substitutions relative to atom abstraction in free-radical reactions. The cationic chain reaction propagated by CH3+ resembles those described in the acid-catalyzed alkylation of hydrocarbons (28). In contrast to previous examples (11, 12), our system employs superacidic conditions at low temperatures and in the absence of additives, providing conditions that favor the protonation of the –O–O– bond and subsequent formation of methyl cations to produce MSA.

Fig. 2 Mechanistic proposal.

(A) Proposed ionic reaction mechanism for the C–H activation of CH4 in the selective production of MSA. (B) Advantages of the cationic pathway over the radical pathway. T, temperature.

Inspired by these promising results, we set out to build a pilot plant facility to test the economic and technical viability of the industrial production of MSA. In line with the findings from our laboratory-scale batch reactions, the plant, with a projected capacity of 20 metric tons/year, was constructed to maximize the productivity of MSA while accounting for CH4 solubility and recycling, concentration of SO3, reaction rates, and deactivation agents. The reaction is carried out continuously as a cascade (Fig. 3A), with pure SO3 and CH4 feeding the first reactor containing a mixture of MSA and H2SO4. This configuration allows for a constant increase in the concentration of MSA as the reaction mixture passes through the reactors (Fig. 3B). The excess CH4 dissolved in the liquid phase at the quencher Q is reintroduced to the beginning of the cascade (29). The process is completed by a final distillation step, affording pure MSA. No MBS is observed as a decomposition product under these conditions. After distillation, the remaining solution consists of a mixture of H2SO4 and MSA, which is recycled to the first reactor, where fresh SO3 is added to regenerate the SO3/H2SO4 mixture. Using this configuration (four reactors), 200 kg of pure MSA was produced per week, which amounts to 2.3 metric tons in 80 days. The combination of high selectivity, conversion, and atom economy make our process ideal for large-scale valorization of readily available CH4 and SO3. A range of value-added products derived from methane or higher alkanes can be envisioned using this superacid chemistry route.

Fig. 3 Sulfonation of methane to MSA.

(A) Schematic of our process. The reaction proceeds as a cascade through reactors connected in series. The pilot plant could produce up to 20 metric tons of MSA per year. The excess SO3 is quenched in reactor Q, the CH4 excess stream and the MSA/H2SO4 sump stream are recycled back to reactor 1, and the MSA-enriched mixture is distilled in column D to obtain pure MSA. (B) The concentration of MSA increases as it passes through the reactors. (C) Oblong quartz window reactor with gas impeller, which improves CH4 mixing.

Supplementary Materials

www.sciencemag.org/content/363/6433/1326/suppl/DC1

Materials and Methods

Supplementary Text

Figs. S1 to S10

Tables S1 to S3

References (3032)

References and Notes

  1. Employing H2O2 (60%) in H2SO4 (98%) does not trigger the sulfonation of methane, most likely because of rapid decomposition of H2O2.
  2. The recycling stream of methane does not contain higher alkanes or any other radical recombination product and can be used directly as a feedstock for the cascade reaction (fig. S10).
Acknowledgments: We thank I. Biertümpel, M. Vogt, C. Réthoré, and K. Giffin for helpful discussions and N. Bloch, J. Stölzel, B. Röhricht, and C. Schuster for technical support. Funding: This work was solely supported by Grillo-Werke AG. Author contributions: Both authors contributed to all aspects of this work. Competing interests: C.D.-U., T.O., and Grillo-Werke AG have filed a provisional patent resulting from this work (WO2018146153 A1), in addition to patent WO2015071455 A1 (20). The authors declare no other conflicts of interests. Data and materials availability: All results are reported in the main text and supplementary materials.
View Abstract

Navigate This Article