Evaluation of Catalyst Support Effects during Rhodium-Catalyzed Hydroformylation in Supercritical CO 2 Greg Snyder, Andrew Tadd, and Martin A. Abraham* Department of Chemical Engineering, University of Toledo, Toledo, Ohio 43606 Hydroformylation chemistry is commercially practiced for the production of aldehyde compounds, which are used as precursors to surfactants and plasticizers. Current technology using the aqueous-phase process is limited by the low solubility of the olefin in water. New techniques using supercritical fluid solvents are in development but rely on the modification of a homogeneous catalyst to increase its solubility in supercritical carbon dioxide (scCO 2 ). An alternative approach for the use of scCO 2 is to use the solvent only as a means of bringing all of the reactants into a single fluid phase, combined with a heterogeneous catalyst. The current paper reports on the development of a heterogeneous catalyst for hydroformylation in scCO 2 , wherein the solid catalyst has been specifically designed to take advantage of the unique properties of this benign solvent. We demonstrate that catalyst tailoring can be achieved by promoting specific fluid-solid interactions that impact the rate of the reaction. This development may allow us to develop new heterogeneous catalysts that further target the fluid-solid interactions to control issues of selectivity and leaching that have been problematical in the development of commercially viable heterogeneous catalysts for the hydroformylation reaction. Introduction Hydroformylation, or “oxo synthesis”, was first dis- covered by Otto Roelen in 1938. In this reaction alde- hydes are produced from olefins, carbon monoxide, and hydrogen, with isomeric aldehydes forming for C 3 and higher olefins. Roelen observed the formation of propi- onaldehyde when ethylene was reacted with CO and H 2 using a cobalt-thorium catalyst at elevated pressures and temperatures. Currently, nominal production capacities for hydro- formylation are more than 7 million tons/year. 1 Butanal accounts for about 73% of aldehyde production; about 75% of the n-butanal is converted into 2-ethylhexanol, which is converted to phthalate ester (a plasticizer) for poly(vinyl chloride) production. All commercial hydroformylation processes in opera- tion today use a homogeneous catalyst. The catalyst is commonly dissolved in an organic solvent, such as toluene or an alkane, or sometimes it is dissolved in the reaction products themselves, usually the higher con- densation products. Cobalt and rhodium are the metals used almost exclusively, although the latest hydro- formylation processes are generally based on rhodium. The addition of phosphine ligands to the homogeneous catalyst provides an increase in reactivity, stability, and selectivity. Homogeneous catalysts are used in hydro- formylation because they provide better activity and selectivity when compared to heterogeneous catalysts. However, homogeneous catalysts are typically soluble transition-metal salts or complexes that are dissolved in a suitable organic solvent that must be eliminated as a waste from the process. The Pollution Prevention Act of 1990 states that the option of first choice is to prevent the formation of wastes at the source. 2 The development of a water- soluble catalyst by Rhone-Poulenc has led to a more efficient and environmentally benign process 3 and im- proved catalyst recovery. However, the reactor contains both an organic and an aqueous liquid phase, with the butanals dissolved in the organic phase and the catalyst dissolved in the aqueous phase. In addition, the solubil- ity of the olefin becomes increasingly small at higher molecular weights, until insufficient olefin can be dis- solved into the aqueous phase to make this process commercially viable. In addition, the multiphase nature of the reaction system leads to inherent difficulties due to mass transfer and/or solubility of the reactants and the products in the catalyst-containing liquid phases. Several researchers have attempted to develop a heterogeneous catalyst for vapor-phase hydroformyla- tion. For example, Rode et al. 4 studied rhodium sup- ported on zeolites for use in vapor-phase propylene hydroformylation and achieved a regioselectivity (n- aldehyde/isoaldehyde) of about 2. Many investigators have used silica as the support, to capitalize on its acidity. Huang et al. 5 impregnated SiO 2 with solutions of RhCl 3 ‚nH 2 O, Co(NO 3 ) 2 ‚6H 2 O, Co 2 (CO) 8 , Rh 4 (CO) 12 , or RhCo 3 (CO) 12 and found that rhodium catalysts gave the highest activities (approximately 3.3 mol of propanal/ mol of catalyst/min) and selectivity toward oxygenates. Chuang and co-workers 6-8 and Brundage et al. 9 have used in situ IR spectroscopy and determined that linearly adsorbed CO on Rh + sites is more active than linearly adsorbed CO on Rh 0 sites. Fukuoka et al. 10 used various bimetallic catalysts for vapor-phase ethylene hydroformylation and CO hydrogenation. Reaction rates were improved by factors of 62-110 on catalysts derived from rhodium-iron clusters with the metal composition of FeRh 5 , FeRh 4 , and Fe 3 Rh 2 as compared to Fe-free Rh 4 /SiO 2 . Polymer-supported rhodium catalysts have * Correspondingauthor.E-mail: martin.abraham@utoledo.edu. 5317 Ind. Eng. Chem. Res. 2001, 40, 5317-5325 10.1021/ie0010722 CCC: $20.00 © 2001 American Chemical Society Published on Web 05/19/2001