Understanding the Difference between Electrochemical and Photochemical Single Electron Transfer: Investigation of Radical Cation [2 + 2] Cycloadditions

Abstract

Chemical synthesis by electrocatalysis and photocatalysis has been developed as a new energy conversion system and it can realize sustainable production processes for various products. Electron transfer is driven by electrical energy and solar energy under ambient temperature and pressure. In the case of electrocatalysis, voltage and current are tunable, and it can realize easy control of reaction conditions. In photocatalysis with semiconductor nanoparticles, one particle can work as both oxidant and reductant, and it would be beneficial toward redox neutral reactions. To understand mechanistic details of both electrocatalysis and photocatalysis is important to develop new processes with high efficiency. We reported oxidative single electron transfer (SET) triggered radical cation cycloaddition reactions in lithium perchlorate (LiClO4)/nitromethane (CH3NO2) by electrolysis and TiO2 semiconductor photocatalysis. These reactions start by substrate activation to form radical cation species, which is derived from anodic oxidation (in electrolysis) or hole oxidation (in photocatalysis). In this research, the [2+2] cycloaddition reactions in LiClO4/CH3NO2 solution were used as model reactions to study mechanistic details of three different SET catalysis, including electrocatalysis TiO2 photocatalysis with and without Pt loading. TiO2 and Pt loaded TiO2 (Pt-TiO2) were selected as photocatalyst in this work. Commercially available TiO2 nanoparticles (P25, 20–40 nm) were used. Pt-TiO2 was synthesized from the TiO2 (P25) and H2PtCl6 as a Pt precursor. H2PtCl6 was reduced by NaBH4 in the presence of TiO2 in H2O to give Pt-TiO2. From TEM observations and EDS measurement, Pt nanoparticles were located on the surface of TiO2 with 3.8 wt% of loading ratio and the size was 2.9 nm. Two types of model [2+2] cycloaddition reactions were used in this research. One is reaction between enol ether and olefin, the other is reaction between enyloxy benzene and olefin. All reactions were carried out in 1 mol/L LiClO4/CH3NO2 solution. For electrocatalysis, constant potential (1.2 V, 1.0 F/mol) was used with carbon felt electrodes as both anode and cathode. For photocatalysis, TiO2 or Pt-TiO2 nanoparticles were dispersed in the solution with 10 mg/mL and the solution was irradiated by 15 W UV lamp. From cyclic voltammetry (CV) of substrate, it was indicated these reactions were driven by SET of enol ether or enyloxy benzene. In the case of enol ether model reaction under air, yield of cyclobutane product was similar to electrocatalysis (78%) and TiO2 photocatalysis (79%), however, Pt-TiO2 photocatalysis showed relatively low yield (46%). Previously, it was proposed that Pt can function as an electron trap to alter the mechanism of reductive SET. It could be explained that the reductive SET from Pt was not involved in the reaction, decreasing the yield of the product. In the case of enyloxy benzene model reaction under air, electrocatalysis realized high yield of cyclobutane product (81%), however, TiO2 and Pt-TiO2 photocatalysis showed low yields (53% and 17%, receptively). We found that O2 bubbling to solution improved the yield of the product when TiO2 photocatalysis (80%) was used, in contrast, it was not the case for the reaction using Pt-TiO2 photocatalysis (trace). It was indicated that O2 can work as a mediator or an oxidant and this role is important for the [2+2] cycloaddition between enyloxy benzene and olefin when TiO2 photocatalysis was used, however, it was not observed under Pt-TiO2 condition. One possibility is that, O2 functions as a reductive redox mediator. Specifically, O2 was reduced by excited electron in TiO2 conduction band via single electron reduction and superoxide species were generated. Since these species have high reducing ability, they can reduce cyclobutane intermediate radical cation to form cyclobutane product. In the case of Pt-TiO2, excited electron could be trapped by Pt, which altered the reductive SET from one-electron mechanism to two- or four-electron pathways that produced H2O2 or H2O from O2. Since these species cannot reduce the radical cation, cyclobutane product was not formed. In addition, H2O2 can change to OH radical species that have high oxidative potential and they potentially decomposed substrate and/or product. In conclusion, we demonstrated the mechanistic differences among electrocatalysis and TiO2 in the presence or absence of Pt by using SET triggered [2+2] cycloaddition reactions as models. Pt-TiO2 might decrease the reduction ability of TiO2. O2 bubbling can increase yield of product in enyloxy benzene reaction because superoxide species might be generated by excited electron in TiO2 and they function as a redox mediator to reduce radical cation intermediate. This O2 mediated system cannot be realized when Pt-TiO2 was used, probably because Pt can reduce O2 into H2O2 or H2O. Figure 1