The non-destructive separation of homogeneous catalysts from organic solutions, allowing catalyst recycle, would be of considerable interest in commercial organic synthesis. Many of the catalysts are of high value, and the destructive separation methods currently employed often generate substantial amounts of waste. A study designed to test the potential of solvent resistant nanofiltration to achieve such separations is reported. An investigation was conducted into the nanofiltration of three organometallic catalysts commonly used in commercial organic synthesis (the Jacobsen catalyst, the Wilkinson catalyst and Pd-BINAP) from a selection of organic solvents (ethyl acetate, tetrahydrofuran and dichloromethane) using a range of polymeric solvent resistant membranes (W.R. Grace, STARMEM(TM) series; Osmonics, Desal-5; Koch Membrane Systems, MPF-50). The compatibility of the solvent-membrane combinations (membrane stability in solvent plus non-zero solvent flux at 2.0 MPa) was assessed. The solvent flux and membrane rejection of catalyst was then determined for each compatible catalyst-solvent-membrane combination in a dead-end pressure cell. Good catalyst rejection (>0.95) coupled with good solvent fluxes (>50 1 m(-2) h(-1) at 2.0 MPa) were obtained in the majority of systems tested, indicating that the technology has considerable potential in this area. An extension of the investigation to assess the effect of pressure, temperature, and catalyst concentration was conducted on a selection of the most successful catalyst-solvent-membrane systems. Trends in the variation of solvent flux and catalyst rejection with variation of these parameters were identified. Increasing pressure substantially improved both solvent flux and catalyst rejection, whilst increasing catalyst concentration was found to be beneficial in terms of substantial increases in catalyst rejection without significantly affecting solvent flux. Increasing temperature generally resulted in improved solvent fluxes but lower catalyst rejection.