Methane is an underused resource, in part because technologies for converting it into more-valuable chemicals are highly underdeveloped, presenting a major challenge for chemists1,2. On page 605, Shan et al.3 address this challenge by reporting rhodium catalysts that promote the oxidation of methane to either methanol (CH3OH) or acetic acid (CH3COOH). The optimized catalytic systems provide mechanistic insight into these reactions, which might inform the development of industrial processes for preparing useful products from methane.
The availability of natural-gas resources from fracking (hydraulic fracturing) has transformed the US chemical industry. Most of the investment and growth4–6 has been in the production of ethylene (CH2=CH2), which is widely used as a chemical building block in the industrial synthesis of other compounds, particularly polythene (polyethylene). Ethylene was conventionally produced by processes broadly known as cracking, in which the components of naphtha fractions derived from crude oil are broken down into smaller molecules. However, many cracker facilities have now been retrofitted to use cheap natural gas as a feedstock, greatly reducing production costs. This has encouraged petrochemical companies to invest in new projects in the United States, or to expand their capacities5, thus generating new jobs.
A side effect of this feedstock shift is that higher olefins — compounds such as propylene (CH3CH=CH2) and isomers of butylene (C4H8), which were also formed by cracking — are now produced globally in lower quantities than before1,2,4. Moreover, there is an abundance of methane (CH4, the main component of natural gas), which was already an underused resource before fracking started. The development of ‘on-purpose’ methods for making olefins from the corresponding small alkanes, such as propane (CH3CH2CH3) and isomers of butane (C4H10), found in natural gas, has therefore become a hot research topic, and the number of emerging technologies in this field has grown7–10. But the methods available for ‘upgrading’ methane by converting it into more-valuable chemicals have been limited.
One currently available methane-upgrading process converts the gas into carbon monoxide and hydrogen, a mixture known as synthesis gas (Fig. 1). This mixture can then be converted into methanol or hydrocarbons using the Fischer–Tropsch process11. But methane upgrading via synthesis gas is generally energy-intensive and can be expensive (although the costs are currently offset by the low cost of methane). The direct conversion of methane into value-added chemicals in economically viable processes therefore remains an outstanding goal of the chemical industry, and one that many researchers have pursued for decades. The required capital investments, and the fact that the synthesis-gas approach is currently relatively cheap, means that direct methane-upgrading technology will have to be substantially more economical than other established processes if it is to be adopted.
Shan and colleagues’ catalytic systems not only enable direct upgrading of methane, but also offer a choice of products. The catalysts consist of rhodium (Rh) dispersed on either a zeolite (ZSM-5, a crystalline, porous solid) or titanium dioxide (TiO2). In the presence of carbon monoxide and oxygen, the catalysts convert methane to either acetic acid or methanol, depending on the acidity of the supporting material (Fig. 1). Acetic acid is currently largely produced by the reaction of methanol with carbon monoxide in the rhodium-catalysed Monsanto process or the iridium-catalysed Cativa process12.
The authors’ optimized Rh-ZSM-5 catalyst produces about 0.4 kilograms of acetic acid per kilogram of catalyst per hour, using methane, carbon monoxide and oxygen at a total pressure of less than 30 bar, and water as a solvent. Although this formation rate is probably too low to make the reaction economically viable, it is a remarkable step forward. By exchanging the hydrogen ions in the porous zeolite for sodium ions, the authors could eliminate the acidity of the catalyst, thus tuning it to make methanol rather than acetic acid. Similarly, the Rh/TiO2 catalyst produces methanol because titanium dioxide isn’t acidic. These results open the door to a more-extensive investigation of the effects of the supports in the future. Interestingly, the methanol-forming reaction requires carbon monoxide to be present, even though this gas is not a reactant.
Using a variety of spectroscopic and microscopic characterization techniques, Shan et al. conclude that individual rhodium species isolated inside pores in the supports are responsible for catalysis. The authors also found that the way in which the catalyst is prepared is crucial for controlling the distribution and form of rhodium species: heating the catalyst in air favours the undesirable formation of nanoparticles of rhodium oxide, whereas heat treatment under hydrogen gas maximizes the amount of isolated rhodium species predominantly in the +1 oxidation state.
Shan and colleagues propose an organometallic mechanism for the reaction, on the basis of the observed correlation between catalytic activity and the presence of isolated rhodium(i) species. In the first step, a stable intermediate containing a Rh–CH3 group forms when the metal comes into contact with methane and oxygen. This intermediate then reacts with carbon monoxide to form acetic acid — presumably, because the carbon monoxide inserts itself into the Rh–CH3 bond to form an acetyl intermediate (Rh–COCH3), which is then hydrolysed to yield acetic acid.
On the basis of the observation that methanol, rather than acetic acid, forms when hydrogen ions are eliminated from the catalyst support, the authors hypothesize that the ions facilitate the insertion of carbon monoxide into Rh–CH3. In the absence of those ions, an oxygen atom instead inserts into the Rh–CH3 intermediate to form a Rh–OCH3 group, but only in the presence of carbon monoxide molecules; hydrolysis of the Rh–OCH3 group then generates methanol. Carbon monoxide therefore acts as a co-catalyst in the conversion of methane to methanol.
The proposed mechanism lays the groundwork for further studies of the versatility of the catalytic system. For instance, understanding how solvents affect the reaction might provide a viable method for making other value-added products from methane. Such versatility would be a good starting point for the development of inexpensive, efficient technologies for upgrading methane from natural-gas sources.
Perhaps the most exciting aspect of this work is the discovery that the catalytically active sites are dispersed rhodium complexes, rather than rhodium nanoparticles, as many chemists might have expected. Shan and colleagues’ work therefore links homogeneous organometallic chemistry — which typically involves the reactivity of individual metal complexes — with solid-phase (heterogeneous) catalysis, and illustrates the importance of understanding catalysts at the atomic scale. The findings therefore open up interesting opportunities for exploring the field of heterogeneous catalysis.
Nature 551, 575-576 (2017)