Technical feasibility of carvone production from limonene or carveol using the catalytic system FePcCl16-SBA15/TBHP

Abstract

ABSTRACT: Carvone is a highly cost ketone (119 USD/kg) used in pharmaceutical, flavor and fragrance industries. Its production at industrial scale has been related to the extraction and purification from the essential oils from caraway, dill or spearmint seeds, or to ecologically improper processes. Hence, the study of alternative carvone production routes is an interesting research field. Limonene has attached the attention as raw material for carvone production, since it is an affordable (44 USD/kg) and available substrate derived from agro-industrial wastes (orange peel oil) and it can undergo allylic oxidation under mild conditions and with environmentally friendly oxidants like tert-butyl hydroperoxide (TBHP), where the formation of carveol as intermediate product for carvone synthesis is reported. Among the catalysts reported, Fe phthalocyanine complexes (FePcs) have been shown good activity for allylic oxidation with non-aggressive oxidants. In this study, it was analyzed the feasibility of carvone production from limonene or carveol, using hexadecachlorinated iron phthalocyanine catalyst immobilized on mesostructured silica SBA-15 (FePcCl16-NH2-SBA-15) using TBHP under mild conditions. The catalyst synthesis, activity, and stability, as well as possible pathways of reactions are reported here. According with the results, the FePcCl16-NH2-SBA-15 was active for limonene and carveol oxidation with TBHP at ambient pressure and 40 °C using acetone as solvent, with maximum yields to carvone near to 6 % (3 hour reaction) and 29 % (1 hour reaction), respectively. Under the reaction conditions, the catalyst showed stability in at least three uses. Slight decrease in the activity of FePcCl16-NH2-SBA-15 was observed in the fourth re-use, which could be due to adsorption of reactants and products on the catalyst blocking its active sites.

In the mechanistic analysis, the oxo metal species (FeIV = O) probably involved in carveol oxidation, seems to be formed by the homolytic cleavage of the TBHP; while for limonene this oxo metal species are probably involved in the mechanism together with radical species from TBHP. According with the results, a probably competition between substrate and TBHP to be adsorbed on active sites is observed during the reactions. Several mechanisms with their respective rate expressions were proposed for the carveol reaction, finding that carveol rate expressions derived from pseudo-homogeneous mechanism better represents mathematically the experimental results. However, this expression does not represent completely the reaction mechanism, since it does not consider the parallel heterogeneous behavior due to reactants adsorption and the possible effect of carvone concentration, byproducts and side reactions on the reaction rate. On the other hand the rate expression obtained from pseudo-homogeneous mechanism of limonene shows even better high goodness-of-fit, which could be because the free radicals involvement in reactions, being higher for limonene, due to the free radical participation in allylic hydrogen abstraction and limonene hydroperoxide formation. Finally, according to the results obtained and some Aspen® simulation, the technical feasibility analysis was carried out, concluding that despite the low cost of limonene and the better performance of catalytic system for carveol, the processes have not higher feasibility with the assumptions and conditions used in this work, due to the following drawbacks found: Limonene process has a low yield, thus it requires high amounts of raw materials and high reactor capacity, carveol process presents problems for product purification and high cost of raw material (carveol) and both process require high oxidant excess and have a raw material total cost higher than total income from carvone sales at the price of this in the market. These results are the base for needed improvements to complete design and technical analysis for a possible scaling up of the process