Key points are not available for this paper at this time.
Glycerol is an attractive renewable building block for the synthesis of di- and triglycerols, which have numerous applications in the cosmetic and pharmaceutical industries. In this work, the selective etherification of glycerol to di- and triglycerol was studied in the presence of alkaline earth metal oxides and the data are compared with those obtained with Na2CO3 as a homogeneous catalyst. It was found that glycerol conversion increased with increasing catalyst basicity; that is, the conversion increases in the order: MgO90 % at 60 % conversion) are obtained over CaO, SrO, and BaO. For these catalysts no substantial acrolein formation was observed. Furthermore, at the start of the reaction mainly linear diglycerol was produced, whereas at higher conversion degrees branched diglycerol started to form. In another series of experiments different types of CaO materials were prepared. It was found that these CaO-based materials not only differed in their surface area and number of basic sites, but also in their Lewis acid strength. Within this series the CaO material possessing the strongest Lewis acid sites had the highest catalytic activity, comparable to that of BaO, pointing towards the important role of Lewis acidity for this etherification reaction. Based on these observations a plausible alternative reaction scheme for glycerol etherification is presented, which considers the facilitation of the hydroxyl leaving process. Finally, the stability of the catalytic solids under study was investigated and it was found that colloidal CaO particles of about 50-100 nm can be spontaneously generated during reaction. Catalytic testing of these CaO colloids, after isolation from the reaction medium, revealed a very high etherification activity. Understanding the nature of these Ca-based colloids opens new opportunities for investigating supported colloidal particle catalysts to take advantage of both their hetero- and homogeneous nature.
Ruppert et al. (Tue,) studied this question.