A greener solution

In a 'biphasic' system, a catalyst dissolved in a liquid phase can be recycled if the reaction products are captured in an adjacent phase. A catalyst-storing liquid that is plentiful and environmentally friendly has been found.

Catalysis is the key to sustainable synthetic chemistry. Homogeneous catalysts, dispersed in a solution of reactants, have many potential advantages over solid-phase, heterogeneous ones: the molecular nature of the catalyst means that it can be designed rationally to be selective and highly active. The difficulty is in separating out the products and recycling the catalytic material. An elegant solution to this problem, however, is 'multiphase catalysis'1: the catalyst resides in one liquid phase and the reactants and products are dissolved in a second phase. Supercritical carbon dioxide2 — carbon dioxide under pressure, in a phase neither truly liquid nor gas — has been found to be a useful reactant/product phase in such systems3,4. In the Journal of the American Chemical Society, Heldebrandt and Jessop5 introduce an attractive system using poly(ethylene glycol), a material that is benign and readily available, to immobilize a homogeneous catalyst while an adjacent phase of supercritical CO2 (scCO2) ensures selective extraction of the product.

Classical multiphase catalysis relies on two immiscible liquids, one dissolving the catalyst and the other containing reactants and products1. Water is often used as the catalyst phase in combination with organic solvents such as diethyl ether, benzene or methylene chloride. Although this approach is generally referred to as 'biphasic catalysis', a reaction system that involves gaseous products will in fact have three phases (two liquids and the gas). In contrast, the system of liquid and supercritical phase, as used by Heldebrandt and Jessop5, is truly biphasic (Fig. 1), as any gaseous reagents are miscible in the supercritical medium. Using scCO2 also eliminates any toxicological or ecological hazards that might be associated with the use of organic solvents.

Figure 1: Biphasic catalysis.

Heldebrandt and Jessop5 have devised a catalytic system comprising two phases, one liquid and the other a supercritical fluid. The reactants for the hydrogenation of styrene are dissolved in supercritical carbon dioxide; a liquid phase of poly(ethylene glycol) holds the catalyst for the reaction, which is a rhodium compound known as Wilkinson's complex. The two phases are immiscible, but mass transport between them is sufficient to aid the catalysed production of ethylbenzene. 'Biphasic catalysis' is not new, but this demonstration of poly(ethylene glycol) as a recyclable liquid catalyst phase could be a step towards 'green' catalysis on an industrial scale.

The choice of the liquid phase that will hold the catalyst is crucial. It should have a minimum solubility in scCO2, but still allow efficient mass transfer with the reactant/product phase to catalyse the desired reaction effectively. Water and ionic liquids have been used successfully, but both combinations have their limitations3,4. In the water–scCO2 system, all reaction components must be stable against acidic conditions (carbonic acid solutions have a pH of around 3) and many catalysts must be modified to be sufficiently soluble in the water phase. Ionic liquids are molten salts consisting of large, cationic organic molecules and suitable anions that are liquid in the typical temperature range of organic reactions (0–100 °C). Held together by electrostatic forces, typical ionic liquids are insoluble in scCO2. But carbon dioxide does have a very high affinity for ionic liquids, which ensures rapid mass transport between the two phases6. For practical applications, however, high material costs and the current lack of toxicological data are drawbacks — although rapid progress on these points can be expected7,8.

The Jessop group has been actively involved in developing biphasic catalysis with both water–scCO2 and ionic-liquid–scCO2 combinations9,10. Consequently, the new approach5 is largely complementary to these systems. Heldebrandt and Jessop's choice of liquid catalyst phase is poly(ethylene glycol), or PEG. This material is widely used in food and beverages and for medical purposes, which is a testament to its benign character. At the same time, liquid PEG can dissolve many organometallic compounds that would typically be used as homogeneous catalysts without any need for their structure to be modified.

Under ambient conditions, only PEG with a fairly low relative molecular mass forms liquid phases — but then its solubility in scCO2 is so high that the combination is useless for biphasic catalysis. Heavier PEG materials are waxy solids, but pressurizing them with CO2 lowers their melting points enough to render the PEG phase liquid at reasonable reaction temperatures. (For example, PEG with an average molecular mass of 1,500 has a melting point of 48–51 °C under ambient conditions, but it can be used as a liquid catalyst phase at 40 °C under CO2 pressures higher than 90 bar.)

Heldebrandt and Jessop5 used a rhodium compound known as Wilkinson's complex — (Ph3P)3RhCl — as a prototypical homogeneous catalyst to demonstrate the feasibility of their approach. This versatile compound can be used to catalyse a wide range of chemical transformations. In this study, the hydrogenation of styrene to ethylbenzene was the chosen test reaction (Fig. 1). Although this reaction is not synthetically important (styrene is in fact produced by the reverse reaction), it is often used to benchmark new multiphase catalyst systems. The authors show that, on extraction of the ethylbenzene product from the scCO2 phase, the catalyst remained stable in the PEG environment and could be recycled four times without noticeable loss of activity. Rhodium contamination in the product was below the detection limit (less than one part per million). Using PEG with an average molecular mass of 900 as the liquid catalyst phase, measurable amounts of PEG were detected among the extracted products; but for PEG with a mass of 1,500, the contamination was found to be as low as 0.1% by weight.

Heldebrandt and Jessop's approach to homogeneous catalysis is an attractive addition to the concept of multiphase catalysis with scCO2. The PEG catalyst phase is non-toxic and readily available, and, as has been demonstrated for ionic-liquid–scCO2 systems, liquid–scCO2 catalysis can operate efficiently under continuous-flow conditions11,12. Engineering for this system would be largely identical to that of heterogeneous catalysis with supercritical fluids, for which a commercial-scale plant has just been opened in Britain13. Even though important data — such as long-term stability, reaction rates and leaching under continuous-flow conditions — have yet to be collected, the PEG–scCO2 system has clear potential for practical application in 'green' catalytic chemistry.


  1. 1

    Cornils, B. & Herrmann, W. A. (eds) Aqueous Phase Organometallic Catalysts (Wiley-VCH, Weinheim, 1998).

    Google Scholar 

  2. 2

    Jessop, P. G. & Leitner, W. (eds) Chemical Synthesis Using Supercritical Fluids (Wiley-VCH, Weinheim, 1999).

    Google Scholar 

  3. 3

    Leitner, W. Acc. Chem. Res. 35, 746–756 (2002).

    CAS  Article  Google Scholar 

  4. 4

    Cole-Hamilton, D. J. Science 299, 1702–1706 (2003).

    ADS  CAS  Article  Google Scholar 

  5. 5

    Heldebrandt, D. J. & Jessop, P. G. J. Am. Chem. Soc. 125, 5600–5601 (2003).

    Article  Google Scholar 

  6. 6

    Blanchard, L. A., Hâncu, D., Beckman, E. J. & Brennecke, J. F. Nature 399, 28–29 (1999).

    ADS  Article  Google Scholar 

  7. 7

    Jastorff, B. et al. Green Chem. 5, 136–142 (2003).

    CAS  Article  Google Scholar 

  8. 8

  9. 9

    Bonilla, R. J., James, B. R. & Jessop, P. G. Chem. Commun. 941 (2000).

  10. 10

    Brown, R. A. et al. J. Am. Chem. Soc. 123, 1254–1255 (2001).

    CAS  Article  Google Scholar 

  11. 11

    Sellin, M. F., Webb, P. B. & Cole-Hamilton, D. J. Chem. Commun. 781 (2001).

  12. 12

    Bösmann, A. et al. Angew. Chem. Int. Edn 40, 2697–2699 (2001).

    Article  Google Scholar 

  13. 13

Download references

Author information



Corresponding author

Correspondence to Walter Leitner.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Leitner, W. A greener solution. Nature 423, 930–931 (2003).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Sign up for the Nature Briefing newsletter for a daily update on COVID-19 science.
Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing