2020, Vol.49, No.10

1171-1173

Electrochemical reduction of samarium triiodide (SmI3) into samarium diiodide (SmI2) is investigated as a model reaction to reuse samarium compounds in catalytic ammonia production under ambient reaction conditions. Potentiostatic electrolysis of SmI3 with carbon electrode as a cathode in the presence of bis(trifluoromethylsulfonyl)imide anion based ionic liquids as electrolytes in tetrahydrofuran gives SmI2 in high yields with a high Faradaic efficiency.

Development of an ammonia synthetic method that proceeds under mild reaction conditions, to replace the Haber-Bosch process as an industrial ammonia production method, is one of the most important research issues in chemistry. In 2003, Schrock and a co-worker reported the first successful example of the transition metal-catalyzed reduction of dinitrogen into ammonia under ambient reaction conditions.1 In this reaction system, molybdenum–dinitrogen complexes bearing a triamide-monoamine ligand worked as effective catalysts in the presence of decamethylchromocene (CrCp*2: Cp* = η5-C5Me5) as a reductant and 2,6-lutidinium salt as a proton source.1,2 Since this epoch-making work, our3 and other research groups47 have developed catalytic dinitrogen fixation systems using various transition metal complexes as catalysts with various combination of reducing agents and proton sources.

In 2019, we found molybdenum-catalyzed reduction of dinitrogen into ammonia using samarium diiodide (SmI2) as a reductant and alcohols or water (H2O) as a proton source under ambient reaction conditions, where up to 4350 equiv of ammonia were formed per Mo atom of the catalyst and 113 equiv of ammonia were every minute produced per Mo atom of the catalyst.8 The catalytic activity was extraordinarily high compared to that of the previously reported catalytic systems. This unprecedented catalytic activity was accomplished by the combination of SmI2-alcohol or -H2O via a proton-coupled electron transfer (PCET) process.9 This successful ammonia production requires a stoichiometric amount of SmI2 as a single electron transfer reductant. In order to achieve practical use of this reaction system, it is necessary to reduce the amount of SmI2 in the catalytic formation of ammonia. Toward this goal, we believe that development of electrochemical reduction of Sm(III) compounds to regenerate SmI2 under mild reaction conditions provides one of the most promising approaches.

In contrast to many reports on reduction of Sm(III) compounds with chemical reductants such as magnesium1012 (Mg) and zinc amalgam13 (Zn-Hg) to regenerate Sm(II) species, electrochemical reduction of Sm(III) compounds has been quite limited to only two reaction systems. Little and a co-worker reported electrochemical reduction of Sm(OTf)3 with mercury pool cathode in the presence of nBu4NI as an electrolyte (Scheme 1a).14 In this reaction system, samarium triiodide (SmI3) generated in situ from the reaction of Sm(OTf)3 with nBu4NI was reduced to SmI2 under constant potential electrolysis at −1.8 V (vs Ag/Ag+). Recently, Mellah and co-workers reported electrochemical formation of SmI2 from the reaction of SmI3 with samarium metal as cathode under galvanostatic electrolysis (Scheme 1b).15a In this reaction system, we cannot avoid the possibility that SmI3 reacted with samarium metal as a chemical reductant to afford SmI2. In fact, Mellah and co-workers previously reported the preparation of SmI2 from electrochemical oxidation of samarium metal as an anode in the presence of nBu4NI.15,16 In both reaction systems reported so-far, the formation of SmI2 with electrochemical reduction was confirmed only qualitatively. As a result, to the best of our knowledge, there is no reliable synthetic method for electrochemical reduction of Sm(III) compounds to regenerate SmI2 under mild reaction conditions quantitatively. Herein, we report the first successful example of simple and environmentally friendly electrochemical reduction of SmI3 into SmI2 under ambient reaction conditions quantitatively (Scheme 1c).

We first confirmed the electronic properties of SmI3 by linear sweep voltammetry using glassy carbon as working electrode in THF (0.1 M nBu4NPF6 as electrolyte). The reduction potential (SmIII/SmII) was observed at −2.0 V (vs Ag/Ag+, all potentials are given with respect to this reference, Figure S1). This experimental result indicates that chemical reduction of SmI3 with decamethylcobaltocene (CoCp*2; E1/2 = −1.78 V in MeCN)17 may proceed under ambient reaction conditions. In fact, the reaction of SmI3 with 1 equiv of CoCp*2 in THF at room temperature for 3 h gave the formation of SmI2 (Scheme 2). We can determine the amount of SmI2 by the dark blue solution attributed to SmI2 with UV-vis spectroscopy. The UV-vis spectrum of the solution after the reaction with CoCp*2 is shown in Figure 1. The yield of SmI2 was estimated to be 65% yield by UV-vis spectroscopy. The standard solution of SmI2 was separately prepared from the reaction of Sm with 1 equiv of ICH2CH2I in THF, which is known as a typical synthetic method for SmI2 (Figure S2).18

Next, electrochemical reduction of SmI3 into SmI2 was carried out as a model reduction of Sm(III) compounds into Sm(II) species by potentiostatic electrolysis using carbon paper (TORAY, TGP-H-90) as working electrode and platinum plate as counter electrode at −2.0 V for 1 h in a divided cell under an atmospheric pressure of dinitrogen (Figure 2). Typical results are shown in Table 1. The amount of SmI2 was determined by UV-vis spectroscopy. When the reaction was performed in the presence of nBu4NPF6 as electrolyte, only a small amount of SmI2 (6% yield, 62% Faradaic efficiency (FE)) was observed due to low current (Table 1, run 1). To improve the low current, we selected ionic liquids based on bis(trifluoromethylsulfonyl)imide (NTf2) as electrolytes because NTf2-based ionic liquids typically show a higher electrical conductivity.19 1-Butyl-1-methylpyrrolidinium bis(trifluoromethylsulfonyl)imide (BMPNTf2) worked as an effective electrolyte to afford a 10 times higher electric charge passed and SmI2 in 64% yield with 62% FE (Table 1, run 2). The UV-vis spectrum of the resulting solution in the presence of BMPNTf2 was almost the same as that after the chemical reduction with CoCp*2 (Figure 1). Interestingly, no formation of SmI2 was observed at all when BMPPF6 was used as an electrolyte in place of BMPNTf2 (Table 1, run 3). This result indicates that the use of NTf2 anion is an essential factor to promote the electrochemical reduction of SmI3 into SmI2.

Table
Table 1. Electrochemical reduction of SmI3 to SmI2 by potentiostatic electrolysisa
Table 1. Electrochemical reduction of SmI3 to SmI2 by potentiostatic electrolysisa

Next, we used other electrolytes with NTf2 anion such as 1-hexyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide (E1), 1-butyl-1-methylpiperidinium bis(trifluoromethylsulfonyl)imide (E2), and trihexyltetradecylphosphonium bis(trifluoromethylsulfonyl)imide (E3) under the present reaction conditions. Interestingly, E1 and E2 worked as more effective electrolytes than E3 (Table 1, runs 4–6). Separately, we confirmed detailed reaction conditions. No formation of SmI2 was observed at all when Pt electrode was used as cathode in place of carbon electrode (Table 1, run 7). In this reaction system, the use of a divided cell was necessary. In fact, no SmI2 was detected when an undivided cell was used in place of a divided cell (Table 1, run 8). This result indicated that SmI2 formed at the cathode oxidized to SmI3 at the anode because the cathode and anode are not separated by a filter using an undivided cell.

With the method for the electrochemical reduction of SmI3 to SmI2 in hand, we investigated the effect of applied potentials from −2.0 to −1.4 V. Typical results are shown in Table 2. The FE of the formation of SmI2 at −2.0 V was maintained until −1.5 V (Table 2, runs 1–4). However, the reaction with the constant potential electrolysis at −1.4 V gave the formation of SmI2 in 13% yield with 32% FE (Table 2, run 5). These results indicate that an applied potential over −1.5 V was required for electrochemical reduction of SmI3 into SmI2 under the present reaction system.

Table
Table 2. Effect of applied potential for electrochemical reduction of SmI3a
Table 2. Effect of applied potential for electrochemical reduction of SmI3a

Finally, we carried out the reaction of a larger amount of SmI3 (0.18 mmol) in the presence of BMPNTf2 as electrolyte at −2.0 V (Scheme 3). A yellow suspension of SmI3 changed to a dark blue solution of SmI2 within 3 h, where SmI2 was obtained in 82% yield with 81% FE.

In summary, we have found the first successful example of simple and environmentally friendly electrochemical reduction of SmI3 into SmI2 under ambient reaction conditions quantitatively. In this reaction system in the presence of NTf2-based ionic liquids as electrolytes by potentiostatic electrolysis, SmI2 was obtained in high yields with a high Faradaic efficiency. The present electrochemical reduction described in this paper is considered as a model reaction of electrochemical reduction of Sm(III) compounds into Sm(II) species under mild reaction conditions. Further study on molybdenum-catalyzed ammonia formation with the present electrochemical reduction method is currently in progress.

The present project is supported by CREST, JST (JPMJCR1541). We are grateful for Grants-in-Aid for Scientific Research (Nos. JP17H01201, JP18K19093, and 20H00382) from JSPS and MEXT.

Supporting Information is available on https://doi.org/10.1246/cl.200429.