Aerobic oxidation of aqueous ethanol using heterogeneous gold catalysts: Efficient routes to acetic acid and ethyl acetate

Abstract

The aerobic oxidation of aqueous ethanol to produce acetic acid and ethyl acetate was studied using heterogeneous gold catalysts. Comparing the performance of Au/MgAl₂O₄ and Au/TiO₂ showed that these two catalysts exhibited similar performance in the reaction. By proper selection of the reaction conditions, yields of 90–95% of acetic acid could be achieved at moderate temperatures and pressures. Based on our findings, a reaction pathway for the catalytic oxidation of ethanol via acetaldehyde to acetic acid is proposed, and the rate-determining step (RDS) in the mechanism is found to be the (possibly oxygen-assisted) dehydrogenation of ethanol to produce acetaldehyde. It also is concluded that most of the CO₂ formed as a byproduct in the reaction results from the absorbed intermediate in the dehydrogenation of ethanol to produce acetaldehyde. By varying the amount of water in the reaction mixture, the possibilities for producing ethyl acetate by the aerobic oxidation of ethanol is also studied. At low ethanol concentrations, the main product is acetic acid; at concentrations $>60\text{ wt}\mathrm{\%}$, it is ethyl acetate.

Introduction

Gold catalysis has attracted significant attention over the last two decades [1], [2], [3], [4]. This increasing interest can be traced back primarily to two pioneering discoveries: the aerobic oxidation of CO and the addition of HCl to acetylene by Haruta et al. [5] and Hutchings et al. [6], respectively. Initially, it was primarily the possibility of performing preferential oxidation of carbon monoxide to carbon dioxide with dioxygen in the presence of dihydrogen that led to the growing interest in the special reactivity of supported nanosized gold particles [7], [8]; however, most recently, the selective oxidation of alcohols to carbonyl compounds with dioxygen/air over gold catalysts has become of interest. Both supported gold particles and polymer-stabilized clusters have been investigated as possible catalysts for the selective oxidation of alcohols [9], [10]. The substrates investigated to date include aromatic, aliphatic, and allylic alcohols; both primary and secondary alcohols; and some polyols. It has been suggested that aliphatic alcohols can be more difficult to oxidize than benzyllic alcohols [11]. For the polyols, gold catalysts have been reported to exhibit a higher chemoselectivity than the analogous supported palladium and ruthenium catalysts [12], [13], [14], thereby making it possible to control the alcohol group in the polyol that is preferentially oxidized. For the oxidation of aliphatic diols containing both a primary and a secondary alcohol group, the regioselectivity is highly shifted toward oxidation of the primary alcohol group [15]. In addition, the effect of the solvent is noteworthy, because in aqueous solution, the carboxylic acid is favored over the aldehyde, but in solvent-free experiments, aldehyde is favored [12]. Moreover, the use of Fe₂O₃ or C as a support material seems, in contrast to CeO₂-, TiO₂-, or SiO₂-supported catalyst, more active toward the formation of esters [16]. The group of Prati and Rossi investigated the oxidation of both simple alcohols and more complex polyols such as glucose [17], [18], [19], [20], [21]. They also studied in more detail the reactivity of various diols, particularly ethylene glycol and phenylethane-1,2-diol [15], [22], [23], [24], [25]. In these studies, it was found that the activity of gold on metal oxides increased with decreasing particle size, whereas for gold on carbon catalysts, the activity apparently reached a maximum at a mean diameter of 7–8 nm [22]. Hutchings et al. worked with both pure supported gold catalysts and Au/Pd alloys for the oxidation of alcohols and aldehydes [15], [26]. Carrettin et al. studied the oxidation of glycerol in aqueous solution and found that oxidation occurred only in the presence of a base but still at moderate temperature (333 K) and with high selectivity [27], [28]. Similarly, others also have conducted oxidation of glycerol with dioxygen using gold catalysts and found that the selectivity varied significantly with conversion [29]. Recently, Corma et al. used supported gold catalysts for the solvent-free oxidation of various alcohols and concluded that nanocrystalline CeO₂ acts as a co-catalyst, improving the catalytic activity of the gold [30]. They demonstrated that the gold catalyst is superior to palladium catalysts in the selective oxidation of secondary alcohols to form ketones [12], and also investigated the oxidation of allylic alcohols in detail [31]. In the case of the allylic alcohols, the gold catalyst was found to have significantly greater chemoselectivity than the Pd catalyst; moreover, gold makes it possible to form mainly the $\alpha ,\beta$-unsaturated carbonyl compound. Idriss et al. investigated the gas-phase oxidation of ethanol over an Au/CeO₂ catalyst in the temperature range of 373–1073 K and found that the product composition changed significantly with temperature [32]; at low temperatures, acetaldehyde was the main product, whereas at higher temperatures, the selectivity switched toward acetone and finally to methane.

Here we report the aerobic oxidation of one of the simplest alcohols, ethanol, using heterogeneous gold catalysts. Previously, we reported that using a Au/MgAl₂O₄ catalyst, high yields of acetic acid can be achieved when ethanol is oxidized in aqueous phase at moderate temperatures and pressures [33]. In this work, the influence of the support was been investigated and the reaction pathway clarified, with the origin of CO₂ as a side product in the reaction and the RDS identified. This could be important in improving the catalyst performance or for identifying completely new catalysts. Furthermore, we found that ethyl acetate also can be formed from ethanol solutions by aerobic oxidation.

The reason for specifically studying the oxidation of ethanol is that ethanol could be one of the future feedstocks of the chemical industry. Ethanol can be easily produced from agricultural products by fermentation and thus can be considered a renewable resource. The annual amount of bioethanol produced currently exceeds 50 million tons per year and is increasing. In comparison, the total amount of acetic acid produced annually is around 10 million tons. Thus, a sufficient amount of bioethanol is produced to have a significant impact, even if only part of it is used as a chemical feedstock rather than a fuel additive [34].

Crude bioethanol contains only 8–10 vol% ethanol [35], and thus purification of ethanol by distillation to produce a useful fuel is a very costly process. Consequently, an investigation aimed at determining whether crude bioethanol can be directly converted to acetic acid or ethyl ester, which could be less expensive to isolate industrially, is of great interest.