Electrochemical alcohols oxidation mediated by N-hydroxyphthalimide on nickel foam surface

Alcohol to aldehyde conversion is a critical reaction in the industry. Herein, a new electrochemical method is introduced that converts 1 mmol of alcohols to aldehydes and ketones in the presence of N-hydroxyphthalimide (NHPI, 20 mol%) as a mediator; this conversion is achieved after 8.5 h at room temperature using a piece of Ni foam (1.0 cm2) and without adding an extra-base or a need for high temperature. Using this method, 10 mmol (1.08 g) of benzyl alcohol was also successfully oxidized to benzaldehyde (91%) without any by-products. This method was also used to oxidize other alcohols with high yield and selectivity. In the absence of a mediator, the surface of the nickel foam provided oxidation products at the lower yield. After the reaction was complete, nickel foam (anode) was characterized by a combination of scanning electron microscopy (SEM), energy-dispersive X-ray spectroscopy (EDX), X-ray diffraction (XRD), X-ray absorption spectroscopy (XAS), X-ray photoelectron spectroscopy (XPS), and spectroelectrochemistry, which pointed to the formation of nickel oxide on the surface of the electrode. On the other hand, using other electrodes such as Pt, Cu, Fe, and graphite resulted in a low yield for the alcohol to aldehyde conversion.


Scientific Reports
| (2020) 10:19378 | https://doi.org/10.1038/s41598-020-75397-8 www.nature.com/scientificreports/ the related carboxylic acids and ketones 13 . The alkali metal nitrates are the next category of mediators for oxidation of aliphatic alcohols and various derivatives of benzyl alcohol 14,15 . N-oxyl compounds such as 2,2,6,6-tetramethylpiperidine N-oxyl (TEMPO) and phthalimide N-oxyl (PINO) could be used as catalysts or mediators for the selective oxidation of organic molecules [16][17][18] . TEMPO and N-hydroxyphthalimide (NHPI) are significantly different with respect to stability, price, and synthesis process. NHPI can be synthesized from phthalic anhydride and hydroxylamine through a simple method; also, NHPI (1 g≈3.26 Eur) is inexpensive compared to TEMPO (1 g≈ 23.9 Eur) and other nitroxyl radicals. However, NHPI must be converted to PINO to become reactive, but TEMPO is stable under ambient temperature 19 . In some electrochemical methods, reactive oxidants such as oxoammonium and imidoxyl species are produced from aminoxyls and N-hydroxyimides. In such methods, the mediator is first oxidized on the surface of the electrode, and then it oxidizes the organic materials in the solution. It has been shown that imidoxyl radicals can also be efficient mediators for hydrogen-atom transfer (HAT) from weak C-H bonds [20][21][22][23] .
The first electrochemical oxidation with TEMPO and other nitroxyl radicals was reported in 1983 24 . So far, there have been many reports of various nitroxyl radicals being used in oxidation reactions 25 . For instance, Azabicyclo-N-oxyl and TEMPO were investigated as catalysts for the electrochemical oxidation of sterically hindered secondary alcohols 26 . However, TEMPO and other nitroxyl groups have a very low solubility in polar environments. To solve this problem, a double system of nitroxides and halide has been applied 27 . Tanakaʼs group reported that the addition of NaBr in presence of nitroxyl mediators as a double mediatory system was useful in resolving secondary benzyl alcohols in biphasic media 28 . In 2016, a copper/TEMPO catalytic system was designed with alcohol oxidation at a lower potential and higher reaction rate than the TEMPO-alone system 29 .
To the best of our knowledge, there are few studies that have made use of nickel foam electrodes for alcohol oxidation; however, the application of nickel oxides in aqueous alkaline media is a very innovative and reliable method for the oxidation of alcohols to the related carboxylic acids [30][31][32] . The basis of these heterogeneous reactions is the transfer of electrons from the surface of nickel oxide to the substrate. During this reaction, a black layer of nickel(lll)(hydr)oxide is formed on the surface of the electrode 33 . In most of the approaches to preparing nickel oxide on nickel foam, a hydrothermal method is used that requires a special nickel salt source and is performed at high temperatures 34 . Another method is using alkaline media, in which the presence of an external base is essential and stopping the oxidation in the aldehyde stage is impossible [30][31][32] .
Recently, several studies have reported that NHPI serves as a successful co-catalyst in the aerobic oxidation reaction of organic compounds. Among a number of catalysts used, complexes of Co and Mn, as well as V, Cu, and other metals have shown a positive effect on catalytic activity 35 . Although these processes have high conversion efficiency, there are significant drawbacks such as the toxicity of the homogenized metal catalysts and the need for completing the reaction at high temperatures and under pressure 36 . Therefore, replacing them with environmentally friendly methods is a challenge.
Since there are only a few reports of using NHPI in electrochemical oxidation [37][38][39][40][41][42][43] , and no report of using it on the surface of nickel foam, in this study, a mild procedure to oxidize various alcohols using a mediator and Ni surface is introduced.

Results and discussion
To attempt the oxidation of benzyl alcohol to benzaldehyde at room temperature, an electrochemical procedure was carried out using several electrodes as anodes and cathodes. It was found that the presence of a mediator was critical for the quick transfer of electrons from an electrode to a substrate under the experimental conditions. The inclusion of a mediator was necessary to decrease the potential for the oxidation reaction and increase the selectivity 17 . Three mediators were investigated: TEMPO, N-Hydroxysuccinimide (NHS) and NHPI. Among them, NHPI showed the best results toward alcohol oxidation. Following the selection of the mediator, various electrodes were examined to optimize the conditions for alcohol oxidation.
Although the mechanism of alcohol oxidation was not studied, the cyclic voltammetry (CV) studies were carried out with a conventional three-electrode setup in which Ni foam, Ag|AgCl|KCl sat and Ni foam served as the working, reference and auxiliary electrodes, respectively. CV indicated that after the addition of benzyl alcohol to the acetonitrile, a doubling of the current density occurred, which was attributed to alcohol-oxidation reaction (Fig. 1). After adding NHPI, a further increase in the current density was observed. The metal foams were shown to have a cellular structure with high porosity and a large volume fraction of gas-filled pores. These pores could be sealed, or could form an interconnected network. The cellular structure for the nickel foam is shown in Fig. 2a,b. After the oxidation reaction, significant corrosion was observed on the surface of Ni foam (Fig. 2c,d). The SEM images indicated flake-like morphology and nanoparticles with ca. 20-100 nm on the surface of the electrode were observed. The SEM-EDX images revealed oxygen on the surface of the foam (Fig. 2e,f). FTIR spectroscopy is a helpful method for identification of M-O bonds in the metal oxides 44 . The FTIR spectra of the solids on the surface of the Ni foam after the oxidation reaction are shown in Fig. 3a. The spectra show a stretching vibration of the Ni-O octahedra at 630 and 711 cm -1 . The peaks at 1124-1750 cm -1 correspond to the organic groups on the surface of the Ni foam. FTIR spectra also showed a broad peak at around 3000-3600 cm -1 attributed to anti-symmetric and symmetric O-H stretching modes.
The XRD patterns indicate that only metallic Ni was present before and after the electrochemical reaction (Fig. 3b). We, therefore, conclude that the amount of NiO x on the surface of the foam is small and it is amorphous.
In the next step, we used in-situ visible spectroscopy, which was measured using a 1.0 cm 2 spectroelectrochemical cell (Fig. S5) and a three-electrode setup. 4.0 V (without ohmic-drop correction) was applied and visible absorption of the electrode was recorded every 10 min. The broad peak at 400-500 nm is related to Ni(III)/ (IV)(hydr) oxide (Fig. 3c). As Ni(III) or (IV) ions act as strong oxidants, we hypothesize that these ions play an important role in the alcohol-oxidation reaction. All these experiments could show that the high-valent Ni(III) Scientific Reports | (2020) 10:19378 | https://doi.org/10.1038/s41598-020-75397-8 www.nature.com/scientificreports/ or (IV) formed under the potential could oxidize NHPI to PINO. In the next step, PINO oxidized alcohol to aldehyde (Fig. 3d). O, C, and Ni were detected on the surface of the Ni foam before the reaction by XPS (Fig. 4a). The peaks for the Ni2p region have significantly split spin-orbit components (18.0 eV). Ni2p 3/2 for the foam showed two peaks at 852.5 and 855.9 eV, which were attributed to metallic Ni and Ni(OH) 2 , respectively ( Fig. 4b) 45 . Ni2p 1/2 for the foam showed a weak peak at 873.5 eV (Fig. 4b). Ni satellites for Ni2p 3/2 and Ni2p 1/2 could also be observed at 861 and 879 eV, respectively, indicating the presence of Ni(II) hydroxide 45 . It seems that the carbon in XPS equipment covers the sample, which was observed at 284.4 eV related to C-C (Fig. 4c). The designated area for O1s showed different peaks attributed to OH 2 , adsorbed OH, and O on the surface of the electrode (Fig. 4d).
On the other hand, O, C, and Ni were detected on the surface of the Ni foam by XPS, following the reaction (Fig. 4e). Ni2p 3/2 for the foam after the reaction showed a broad peak at 856-858 eV, which was attributed to Ni(II) and Ni(III) oxides (Fig. 4f) 45 . The observed carbon area suggests that in addition to carbon from the equipment, other carbon compounds containing C-O and C=O could cover the Ni foam following the reaction (Fig. 4g). The designated area for O1s showed different peaks attributed to OH 2 , adsorbed OH, and O on the surface of the foam after the reaction (Fig. 4h).
X-ray absorption spectroscopy (XAS) was performed to obtain information about the oxidation state and structure of Ni foam after the reaction. XANES spectra show that after the reaction, the foam had a higher oxidation state (2.1) than metallic nickel (Fig. 5a).
The simulation results of Fourier transformed EXAFS-spectra of Ni foam after the reaction show a high main Ni-O peak that can be simulated by two Ni-O shells at 2.04 Å and 2.45 Å (Fig. 5b, Table 1). The spectrum has a Ni-Ni peak corresponding to di-μ-oxo-bridged Ni, which could be simulated by three Ni-Ni shells at 2.77 Å, 2.96 Å and 3.11 Å.
Initially, the reaction was investigated in the presence of two nickel foam electrodes and 99% aldehyde formation was confirmed (Entry 1, Table 2). When the nickel foam electrodes were replaced by platinum electrodes, only 44% aldehyde was obtained (Entry 2). Following these results, other electrodes such as iron foam and graphite produced 2% and 10% products, respectively (Entries 3 and 4). However, the copper electrodes were unstable under the experimental conditions (Entry 5). These experiments revealed that the electrolyte has an essential role in this reaction. Therefore, we examined n-Bu 4 NPF 6 , n-Bu 4 NBr, n-Bu 4 NCl, and LiBr, which gave the product in low to moderate yields (Entries 6-9). A series of solvents that were screened provided unsatisfactory results. (Entries 10-12). We next examined TEMPO and NHS as mediators, but only 15% and 21% yields were obtained, respectively (Entry 13 and 14). To be expected, the absence of NHPI led to a significant drop in the yield (Entry 15). Decreasing NHPI to 10 mol% had minimal harmful effects, and the 1b product was not detected in 5 mA (Entries 16 and 17). The results of the oxidation reaction were the same either in the presence or absence of an external base (Entry 18).
Under the optimized concentrations, we then proceeded to employ these conditions for the oxidation of a broad range of alcohols ( Table 3). The different electron-donating or electron-withdrawing functional groups of ortho, meta and para situations of benzyl alcohol indicated an excellent conversion and a high selectivity in the corresponding aldehyde (Entries 1-6, Table 3).
Comparison of the activity of NHPI on nickel foam with other chemical and electrochemical methods in the oxidation of benzyl alcohol is summarized in Table 4.

Conclusions
In summary, a simple electrochemical method for oxidation of alcohols toward the formation of aldehydes and ketones was developed, using inexpensive, accessible and reusable nickel/nickel oxide surfaces. A variety of alcohols were shown to be compatible with this electrochemical process. The successful oxidation of benzyl alcohol (10.0 mmol) was evaluated. Through the method employed for a larger scale than lab-scale, benzyl alcohol (10 mmol, 1.08 gr) was successfully oxidized to benzaldehyde (91%) without any side or by-products. In the absence of a mediator, nickel/nickel oxide surface produced aldehyde in 50% yield. The nickel/nickel oxide surface was characterized by a number of methods. Especially, X-ray absorption spectroscopy (XAS) showed that after the reaction, the foam had a higher oxidation state than the metallic nickel.
The Fourier transformed EXAFS-spectra of the Ni foam after the reaction showed a high main Ni-O peak that could be simulated by both Ni-O shells at 2.04 Å and 2.45 Å. The spectrum had a Ni-Ni peak corresponding to di-µ-oxo bridged Ni, which could be simulated by three Ni-Ni shells at 2.77 Å, 2.96 Å and 3.11 Å.   Fig. S6. The fit parameters for the simulations are given in Table 1. Table 1. Parameters obtained by the simulation of the k 3 -weighted EXAFS spectra shown in Fig. 5. The energy shift (ΔE 0 , 1.4 eV) and Debye-Waller parameters (σ, 0.078 Å and 0.039 Å for O and Ni, respectively) were determined from this fit. The filtered R-factor which was 13, and the reduced χ 2 value was 5.5. www.nature.com/scientificreports/ The nickel foam (Nanobazar), copper foam (Nanobazar), Fe foil (Suzhou JSD Foam Metal Co., Ltd), and graphite (Nanobazar) were purchased from commercial sources. All reactions were performed under atmospheric oxygen or air in test tube flasks. The reactions were followed by Thin-layer chromatography (TLC) on silica gel plates and Gas chromatography (GC). Preparative thin-layer chromatography (PTLC) separations were carried out on 0.25 or 0.5 mm E. Merck silica gel plates (60F-254). The following abbreviations were used to explain the multiplicities: s = singlet, d = doublet, t = triplet, q = quartet, m = multiple.

Ni-O short Ni-O long Ni-Ni short Ni-Ni Ni-Ni long
Preparation of the electrodes. All of the reactions were carried out in an experimental setup. The different electrodes used in this reaction are shown in Fig. S1.
For each measurement, a 1.0 cm × 1.0 cm piece of foam nickel (10 cm × 10 cm) with a thickness of1.5 mm was cut. A piece of glass was placed between the two electrodes which were then connected with parafilm. The bottom half of the electrodes were immersed to the solution, to avoid collision with the stirring bar. A septum was chosen, above which two small incisions were made. Steel wire was threaded through these incisions and attached to each of the electrodes. The assembled electrodes were used as described (Fig. S2). Table 3. Oxidation of different alcohols using the electrochemical method in mild conditions. For all of these reactions, < 10% maximum error is introduced. Reaction condition: (a) The product yields were determined by GC analysis.  carefully washed with acetone. No attempt was made to remove oxygen or water at any stage. Lithium perchlorate (0.6 mmol, 64.0 mg) and NHPI (0.2 mmol, 32.0 mg) were dissolved in 5.0 mL acetonitrile, and the mixture was stirred for 10 min to form a homogeneous solution. Alcohol (1.0 mmol) was added to the reaction mixture, which was mixed until the solution was homogeneous. The pre-prepared electrodes were partially immersed to the solution. Parafilm was wrapped around the septum entirely to keep the solvent constant. After the connection was complete, a constant voltage (4.0 V) was applied and the current was measured, which showed a relatively constant current of 10 mA. After completion of the reaction, the electrodes were washed with EtOAc (10 mL) and, the combined organic phase was washed with H 2 O (2 × 5 mL) and brine (1 × 5 ml). The organic phase was dried over anhydrous MgSO 4 . The solvent was evaporated and purified by flash column with hexane-EtOAc (9:1) to give a corresponding aldehyde or ketone.
Received: 11 June 2020; Accepted: 15 October 2020 Table 4. Comparison of the efficiency of this method with other existing methods for the oxidation of alcohols.