Carbon dioxide is an abundant resource, but difficult for industry to use effectively. A simple reaction might allow it to be used to make commercial products more sustainably than with current processes. See Letter p.215
As raw materials go, there is a lot to like about carbon dioxide — it is available everywhere, inexpensive, non-flammable and less toxic than most of the chemicals widely used in industrial processes. But it is relatively unreactive, making it difficult to activate so that it can be transformed into desirable compounds. Nevertheless, in nature, many plants have evolved molecular machinery that overcomes the inherent stability of CO2 to use it to make biological building blocks (sugars) and materials (polysaccharides). Inspired by the carbon–carbon bond-formation processes used by plants, Banerjee and colleagues1 (page 215) have identified a synthetic route that not only uses CO2 to make useful compounds, but also involves tractable processing conditions. Their route is simple, is potentially more sustainable and economical than the one it is designed to replace, and could be applicable to a variety of product types.
CO2 has been used as a raw material by the chemical industry in the past2,3, albeit rather sparingly, to make urea (a fertilizer and building block for the chemical industry) and cyclic carbonate (a solvent). The processes were commercialized not because they were more sustainable than other routes, but because the chemistry was available to make these valuable products economically. Scientists have been interested in expanding the role of CO2 as a raw material for many years, but, for the most part, either the compounds generated from it were not sufficiently useful to merit industrial production, or the processes involved were too energy-intensive or inefficient to warrant further development.
The processes that have been successfully adopted to make commercial products from CO2 were typically preceded by breakthroughs in chemistry and/or catalysis. For example, groundbreaking work on catalyst design4 allowed CO2 to be polymerized with another compound, propylene oxide, to create polycarbonate polyols — important building blocks for polyurethanes, and saleable products in their own right. This work was scaled up and commercialized by Novomer, a chemistry-technology company in Ithaca, New York; the international chemical company Bayer has also pursued this role of CO2 using their own catalysts5. Banerjee and colleagues now report new chemistry to make another valuable molecular building block from CO2.
The authors used caesium carbonate, a simple salt, to activate organic substrates that could then be reacted with CO2. Their key finding is that CO2 can be reacted with 2-furan carboxylate (FC; Fig. 1) to form furan-2,5-dicarboxylic acid (FDCA). This is notable because FC is readily derived from biomass waste material, such as maize (corn) stover and sawdust. Furthermore, FDCA is one of the monomers used to generate polyethylene furandicarboxylate (PEF) — a plant-based polyester that is being commercialized6 to compete with the widely used plastic polyethylene terephthalate (PET), which is derived from petrochemicals.
Banerjee et al. show that the caesium carbonate can be recycled, and that the product can be separated easily from the reaction mixture. Both of these features will aid in scaling up the reaction. Production of PEF results in fewer carbon emissions than production of PET (ref. 6), but the authors' route to FDCA should reduce the overall carbon footprint still further. Once scaled up, the new route might be less wasteful — needing fewer raw materials and less energy — than the conventional industrial synthesis of FDCA, which uses fructose as a starting material.
Synthetic processes involving CO2 as a raw material can be considered more sustainable than existing processes only if the chemistry involved reduces environmental impacts over the entire life cycle of the process. Carbon footprint is only one of several metrics7 used to gauge the environmental impact of a product; other considerations include the potential to increase acidification (acid rain) or to trigger photochemical oxidation (smog). Even though Banerjee and co-workers' process seems to be much less wasteful than the fructose route to FDCA, a comparison of the life-cycle impacts of the two routes will need to be performed to ensure that it is truly more sustainable.
The authors also show that benzene can be reacted with CO2 and caesium carbonate to form benzoic acid in a single step (see Fig. 3c of the paper1). This is intriguing because it has been known8 since the 1950s that benzoic acid can be transformed into terephthalic acid, one of the monomers used to make PET. Although the initial reaction yields reported by Banerjee and colleagues are low, the finding raises the possibility that PET, like PEF, could be made using CO2. If the yields can be improved, then this chemistry would be a marked improvement on the current multistep route used by industry to make terephthalic acid.
More than 45 million tonnes of PET are produced annually9, making it one of the largest potential synthetic 'sinks' for CO2. That said, no synthesis that uses CO2 will lead to sizeable reductions in atmospheric concentrations of the gas. Nevertheless, finding sustainable uses for abundant resources such as CO2 as alternatives to non-renewable resources remains a worthy goal. More broadly, Banerjee and co-workers' results suggest that the molecular machinery devised by chemists will follow the example of plants, by evolving to use CO2 efficiently to create the feedstocks and materials that we need.