Replacing sulfuric acid with water in electrochemical metal-free mild aromatic C–H amidation: a direct route to N-phenylamides

Abstract

In this work, we describe a mild electro-oxidative metal-, oxidant- and acid-free direct amidation of aromatic C–H bonds using nitrile solvents as a source of amide and amine moieties. We show that carrying out the reaction in a divided cell makes it possible to eliminate the use of an acid used in previous attempts, enabling at the same time this reaction for easily reducible substrates and paired electrosynthesis. This electrochemical approach was demonstrated on 60 examples of N-phenylamides, including propanil, which has not been obtained under electrochemical conditions so far, and 12 examples of benzoxazoles; it is also amenable to gram-scale synthesis. Mechanistic studies (voltammetry, EPR, and quantum chemical calculations) revealed that the process is initiated by the formation of hydroxyl radicals from residual water molecules on the anode, which attack the nitrile solvent, and the resulting amide radicals add to the arene.

---

Introduction

An efficient formation of amide bonds is important in organic and medicinal chemistry due to the presence of amide moieties in a wide range of natural products, pharmaceutical agents, and functional materials.¹ The amide bond is one of the most essential fragments that forms the backbone of proteins, peptides and many other biologically significant compounds. Currently, approximately 25% of all commercial drugs and two-thirds of all drug candidates contain at least one amide moiety.² Moreover, agrochemicals with herbicidal and antifungal activity (e.g. propanil and boscalid) and a lot of marketed drugs (paracetamol, niclosamide, imatinib, etc.) often contain the phenylamide framework (Fig. 1).³

Fig. 1 Selected agrochemicals and drugs containing the phenylamide framework.

Propanil is a contact herbicide commonly used in the US to control grass and broadleaf weeds in rice fields. In the factory, propanil is obtained by the following process: nitration of 1,2-dichlorobenzene produces 1,2-dichloro-4-nitrobenzene, and subsequent hydrogenation of the nitro group in the presence of a nickel or palladium catalyst forms 3,4-dichloroaniline. Then, acylation of the amino group with propionyl chloride gives propanil (Scheme 1).⁴

Scheme 1 Commercial routes to acetanilide and propanil production, recent advances in Ritter-type electrochemical reactions to produce N-arylamides, and our green and wide-scope approach to them.

Another practically important phenylamide-containing compound is acetanilide, which is a starting compound in the synthesis of dyes and drugs (e.g. sulfa drugs) and is used as a fuel additive.⁵ It is also used to stabilize hydrogen peroxide and as a plasticizer for cellulose nitrates. Acetanilide is commercially obtained by acylation of aniline with acetic acid or acetic anhydride at 120 °C followed by recrystallization from water (Scheme 1).

Thus, the importance of N-phenylamides is obvious, and the search for new ways to obtain them directly in one step under milder conditions still remains a challenge for chemists.

One of the most well-known methods for the preparation of N-substituted amides is the Ritter reaction.⁶ To date, various variations of the Ritter-type reaction of carboxylic acid,⁷ haloalkynes,⁸ aldehydes,⁹ triazenes,¹⁰ arenediazonium tetrafluoroborates,¹¹ and cyclic diaryl iodonium salts¹² are known which allow one to obtain various N-substituted amides, in particular, N-benzylamides and N-arylamides. However, many of the established protocols require elevated temperatures, prefunctionalized aryl (or amide) halide precursors, expensive catalysts or the use of stoichiometric oxidants.¹³

Electrochemistry is a powerful and green strategy compared to traditional organic reactions and allows one to access valuable N-containing compounds in an atom- and step-economical manner under mild conditions.¹⁴ The electrochemical C(sp³)–H Ritter-type amidation reaction of alkanes,¹⁵ esters,¹⁶ and ketones¹⁷ has been well documented, as early as the 1970s. Since then, the study of Ritter-type electrochemical reactions has continued to develop rapidly. In the last few years, the world's leading scientists have proposed new electrochemical protocols for Ritter type C(sp³)–H amidation of benzylic sites (Scheme 1, ex. 1) and Ritter type C(sp²)–H amidation of aromatic sites (Scheme 1, ex. 2–4).¹⁸

Barba^18jet al. proposed an anodic oxidation of 1-(trifluoromethyl)benzene in dry acetonitrile/Bu₄NBF₄ under constant potential conditions (Scheme 1, ex. 2). 2-(Trifluoromethyl)acetanilide was obtained as the main product in this reaction, but the scope consisted of only one reaction.

Banerjee^18k and co-workers reported a regioselective electrochemical Ritter-type reaction at the C(sp²)–H of phenol derivatives toward an environmentally benign and direct synthesis of different N-arylacetamides, in particular paracetamol (Scheme 1, ex. 3). The reaction proceeds under exogenous oxidant- and catalyst-free conditions, but was limited to phenol substrates.

In 2023, Ye's group^18l reported a general approach for amidation of aromatic C(sp²)–H bonds (Scheme 1, ex. 4). It has a good substrate scope but takes 10 hours and requires the use of excess sulfuric acid (4 equiv.), which makes it environmentally unfriendly.

Strikingly, also in 2023, some of us^18m achieved good yields in a very similar electrochemical amidation in the absence of sulfuric acid or any other reagent except water (Scheme 1, ex. 5). In this work, we compare our conditions to those of Ye's group^18l and show that getting rid of sulfuric acid becomes possible simply by the use of a divided cell. We optimize the conditions for this amidation and extend its scope by synthesizing 60 distinct N-arylamides (vs. 22 in our previous article), including propanil, which has not been obtained under electrochemical conditions previously, as well as 12 benzoxazoles. We also show that nitriles can be a source of not only amides, but also the corresponding amines.

Results and discussion

Although in some studies^18l H₂SO₄ is considered an indispensable reagent, which forms persulphates, thus initiating the reaction, examples 2, 3 and 5 in Scheme 1 suggest that electrochemical amidation of arenes by nitriles can proceed in its absence. We have hypothesized that the presence of an acid simply alleviated the reduction of easily reducible substrates¹⁹ and products on the cathode when the reaction is carried out in an undivided cell. Then it should be easy to get rid of the acid by using a divided cell. This would not only “greenify” the process by removing the hazardous reagent, but also make it more economical by allowing pairing with some cathode process (see below for an example).

We chose 1,4-bis(trifluoromethyl)benzene (1a) as a model substrate for optimizing the conditions of acid-free divided-cell electrolysis, since it is very difficult to oxidize and CF₃-bearing arylamides are potential drug candidates.²⁰

Initially, we explored the oxidation of a solution of 1a in acetonitrile (20 mL) in a divided cell equipped with Pt–Pt electrodes at a constant current of 35 mA at room temperature for 2.5 hours under air; the salt PyHBF₄ (3 equiv.) as a supporting electrolyte was placed only in the cathodic compartment (Table 1, entry 1). As a result, N-(2,5-bis(trifluoromethyl)phenyl)acetamide (2a) was obtained in 63% yield with no detection of other products. We hypothesized that the amount of water present in acetonitrile (∼0.02%, Merck) was limiting the reaction yield, since it is the only available source of oxygen for the amide group. Indeed, adding 0.8 equiv. of water allowed us to obtain 2a in 97% yield under the same conditions (Table 1, entry 2). However, the reaction turned out to be water-sensitive, with excess water negatively affecting the formation of 2a (Table 1, entries 3 and 4). Expectedly, no product was observed when the reaction was carried out in freshly dried MeCN under an inert atmosphere and using molecular sieves (Table 1, entry 5), or without electricity (Table 1, entry 6).

Table 1 Electrochemical oxidation of 1,4-bis(trifluoromethyl)benzene, optimization of reaction conditions^a

Entry	Variation from the standard conditions	Yield of 2a,^b %
a General conditions: 1a (1.2 mmol), PyHBF4 (3 equiv.) only in the cathodic compartment, MeCN (20 mL), divided cell, Pt anode (20 cm^2), Pt cathode (10 cm^2), rt, 35 mA for 2.5 hours (2.4 F mol^−1). b Yield determined by ^1H NMR using trimethoxybenzene as an internal standard. c Isolated yield. d The solution changes its color, but no 2a could be detected. e Molecular sieves. n.r. = no reaction.
1	Without water (≪0.8 equiv.)	63
2	None	97 (87)^c
3	H2O (1/1.5/3/10 equiv.)	92/88/81/32
4	MeCN : H2O (3 : 1), or ∼232 equiv. of H2O	0^d
5	Under N2, freshly dried MeCN, M.S.^e	0^d
6	No cell potential	n.r.
7	Et4NBF4 instead of PyHBF4	94
8	Et4NCl instead of PyHBF4	38
9	Et4NBr instead of PyHBF4	29
10	GC(+)–Pt(−)	74
11	C(+)–Pt(−)	43
12	DCM solvent, MeCN (1–10 equiv.)	0^d
13	DCM solvent, MeCN (30–100 equiv.)	<42
14	DCM : MeCN (1 : 1)	57
15	25 mA for 3.5 hours	76
16	50 mA for 1.5 hours	87
17	Undivided cell	27
18	Undivided cell, H2SO4 (4 equiv.)	59
19	Heating 50 °C	53

---

Replacing PyHBF₄ with Et₄NBF₄ had little effect on the yield of the resulting 2a (Table 1, entry 7); however, Et₄NBF₄ must be added to both compartments (cathodic and anodic) to improve electrical conductivity, whereas PyHBF₄ needs to be added only to the cathodic compartment due to its easier diffusion through the membrane and it can be easily recovered from the reaction mixture after electrolysis (see the ESI† for details). The electrolytes Et₄NBr and Et₄NCl proved to be less efficient for this reaction (Table 1, entries 8 and 9) due to their hygroscopicity and lower oxidation potentials (+1.40 and +1.43 vs. Ag/AgCl, respectively). These salts can also act as nucleophiles, promoting unwanted halogenation processes.

Replacing the platinum anode with a glassy carbon (GC) or graphite anode (C) led to a decrease in the yield of 2a (Table 1, entries 10 and 11), which is probably associated with higher overvoltage that can cause higher adsorption of aromatics on the electrode, leading to the inhibition of the target process. The platinum electrode is optimal since it is more reliable and stable in most electrochemical reactions.

Carrying out the reaction in dichloromethane (DCM) using a 1a : MeCN ratio of 1 : 1–1 : 10 resulted in no products (Table 1, entry 12). Product 2a was formed in moderate yields when an excess of nitrile relative to the substrate was used (Table 1, entries 13 and 14). This is likely linked to the necessity for the formed OH-radical to encounter a MeCN molecule before it encounters an aryl (see below for mechanism investigation).

A constant current of 35 mA turned out to be optimal. A decrease in the current led to slower and incomplete conversion of the starting substrate and longer synthesis time, whereas its increase somewhat lowered the product yield (Table 1, entries 15 and 16).

The yield of the model reaction expectedly decreased when it was carried out in an undivided cell (Table 1, entry 17) due to the higher propensity of 1a to undergo reduction than to undergo oxidation (see below). But addition of sulfuric acid, indeed, inhibited reduction, somewhat improving the product yield (Table 1, entry 18). Heating the reaction mixture to 50 °C led to a decrease in the yield of the target product 2a (Table 1, entry 19), since in this case the resulting amide decomposes into an amine.

To compare our optimal conditions with those proposed in ref. 18l, we carried out the oxidation of 1a and 1-chloro-4-nitrobenzene (1o), one of the most easily reducible substrates from their article (Fig. 2). We considered the reaction in an undivided cell with 4 equiv. of sulfuric acid (“Ye's conditions” (A), as close as possible to the conditions from ref. 18l), the same conditions but without the acid (“acid-free Ye's conditions” (B)), and our optimized conditions (“our conditions” (C)). The formation of products was monitored by TLC, GC-MS and ¹H NMR. See Scheme 2 for the results.

Fig. 2 Cyclic voltammograms (CVs): 1a (purple), 1o (green) and sulfuric acid (orange). Conditions: 5 mM, MeCN, 0.1 M Bu₄NBF₄. GC – working electrode, Ag/AgCl – reference electrode, scan rate −100 mV s⁻¹.

Scheme 2 Electro-oxidation of arenes. Ye's conditions (A): arene (1.2 mmol), H₂SO₄ (4.8 mmol), MeCN (15 mL), Pt–Pt (14 mm × 32 mm × 1 mm both electrodes), undivided cell, nBu₄NBF₄ background salt. Acid-free Ye's conditions (B): arene (1.2 mmol), H₂O (0.8 equiv.), MeCN (15 mL), Pt–Pt (14 mm × 32 mm × 1 mm both electrodes), undivided cell, nBu₄NBF₄ background salt. Our conditions (C): arene (1.2 mmol), H₂O (0.8 equiv.), MeCN (20 mL), Pt anode (20 cm²), Pt cathode (10 cm²), divided cell, PyHBF₄ background salt. Yield was determined by ¹H NMR.

Oxidation of 1a under acid-free Ye's conditions (B) produced 2a in only 27% yield, and 21% of the starting substrate remained in the reaction mixture unreacted (Scheme 2). The low product yield was due to gum formation during electrosynthesis due to the reduction of acetonitrile at the cathode to 3-aminocrotononitrile and trimers,²² which can polymerize quite easily, forming a film on the electrodes. Adding sulfuric acid (Ye's conditions, A) indeed improved the product yield to 59% and significantly reduced gum formation (Scheme 2). However, 19% of the starting material was still present, and the formation of 8% of a dimer was observed. Our conditions (C), on the other hand, allowed us to obtain the desired product 2a in an excellent yield and with no gum formation or dimerization.

However, as widely known in the machine learning field,²¹ it is incorrect to compare different conditions on a reaction that was used to optimize one of them but not others. So, we compared these conditions on 1o, which was used in ref. 18l. Oxidation of 1o under conditions B provided 2o in a very low yield of 7%, while 88% of the starting substrate remained in the reaction mixture (Scheme 2). No gum formation was observed, however. This behavior can be explained by the ability of nitroarenes to easily reduce, but then oxidize back under electrochemical conditions (Fig. 2).

Expectedly, addition of 4 equivalents of sulfuric acid increased the product yield, but only to 14% (Scheme 2). Strikingly, the yield of the dimer was twice as large (49%), and 32% of the starting 1o remained unreacted. Our greener conditions (C) allowed us to obtain 2o in 87% yield with no traces of starting materials or dimers (Scheme 2). Thus, our conditions are clearly eco-friendlier and more efficient.

With the optimized conditions in hand, the scope of aromatic derivatives with electron-donating and electron-withdrawing functional groups was expanded to 60 entities (Table 2). The data on the redox properties of all arenes in the C–H amidation are summarized in Table S2 (see the ESI†).

Table 2 Substrate scope in Ritter-type electro-oxidative aromatic C–H amidation of arenes^a

a General conditions: arene (1.2 mmol), RCN (20 mL), PyHBF₄ (3 equiv., 3.6 mmol) in the cathodic compartment, Pt anode (20 cm²), Pt cathode (10 cm²), divided cell, rt, constant current I = 35 mA. Isolated yield.

---

Electrochemical oxidation of 1,3-bis(trifluoromethyl)benzene (1b) in MeCN, EtCN and n-PrCN led to anilides 2b, 3b and 4b as the main isomers in 81%, 75% and 74% yields, respectively, with trace formation of other regioisomers. The product yield decreased in the case of 1,3,5-tris(trifluoromethyl)benzene (1c), probably due to the steric factor. Oxidation of trifluoromethylbenzene (1d) gave ortho- and meta-substituted anilides 2d-a, 3d-a, 2d-b, and 3d-b as the major regioisomers with an ortho/meta/para ratio of 3 : 3 : 1 (2d) and 3 : 3 : 2 (3d). The structure of 2d-a was also confirmed by single-crystal X-ray analysis (CCDC 2336800†).

Oxidation of fluorobenzene (1f) and chlorobenzene (1g) in MeCN and EtCN resulted only in para-substituted anilides (2f, 2g, 3f and 3g), while oxidation of bromobenzene and iodobenzene failed to give any amidation product. Oxidation of 1g led to the formation of a dimer in addition to the anilide, which was confirmed by mass spectrometry and ¹H NMR.

Oxidation of 1,2-dichlorobenzene (1h) in propionitrile and acetonitrile resulted in the formation of N-(3,4-dichlorophenyl)propionamide (propanil, 3h-a) and N-(3,4-dichlorophenyl)acetamide (2h-a) as the main products.

Oxidation of 4-chloro-1-(trifluoromethyl)benzene (1e) in MeCN, EtCN and PrCN afforded anilides 2e, 3e and 4e with the amide group at the ortho position to chlorine in high yields of 89%, 87% and 85%, respectively. Upon oxidation of 1-chloro-4-fluorobenzene (1i) and 1-bromo-4-fluorobenzene (1j) under selected conditions, fluorine was replaced by a hydroxyl group, resulting in the formation of N-arylamides 2i, 3i, 4i, 2j, 3j, and 4j in low yields since benzoxazoles 2ii, 2jj, 3ii, 3jj, 4ii, and 4jj were formed predominantly (Scheme 3). Electro-oxidation of ester 1r in both MeCN and EtCN led to ortho/meta/para regioisomers with the formation of the ortho- and meta-substituted anilides (2r-a, 3r-a, 2r-b, and 3r-b) as the main products. Oxidation of dimethyl isophthalate (1s) in MeCN led to the formation of 2s with a yield of 75% and an a : b isomer ratio of 1.3 : 1. The oxidation reaction of 1s in EtCN resulted in the formation of 3s in an a : b ratio of 1.5 : 1, but with a lower yield of 64%.

Scheme 3 Formation of benzoxazoles and N-phenylamides during electro-oxidation of arenes. ^a General conditions: arene (1.2 mmol), RCN (20 mL), PyHBF₄ (3 equiv.) in the cathodic compartment, divided cell, Pt anode (20 cm²), Pt cathode (10 cm²), rt, 35 mA. Isolated yield.

Phenols reacted solely at the para position to the hydroxyl group to afford phenylamides 2t–2v and 3t–3v. Acetaminophen or paracetamol (2t) was formed in 74% yield directly from phenol under mild conditions. The para-protected phenol did not give any anilide product, as was shown for 2,4-dichlorophenol.

Oxidation of methyl salicylate (1w) and ethyl salicylate (1x) in aceto-, propio- and butyronitrile led to the formation of two products: anilides 2w, 3w, 4w, 2x, 3x, and 4x and benzoxazoles 2ww, 3ww, 4ww, 2xx, 3xx, and 4xx (Scheme 3). On passing more electricity (4.8 F mol⁻¹), benzoxazoles were predominantly formed rather than anilides. The formation of amidation products with two or more amide groups was not detected in each case by ¹H NMR and HRMS.

In the case of nitrobenzene (1m) oxidation, the addition of the amide fragment occurred predominantly at the ortho position to the nitro group. N-(2-Nitrophenyl)acetamide (2m) and N-(2-nitrophenyl)propionamide (3m) were obtained in moderate yields of 39% and 42%, respectively, with incomplete conversion of 1m even after passing 7.2 F mol⁻¹ of electricity. Addition of the amide fragment to the ortho position relative to the nitro group probably occurs due to the formation of a more stable intramolecular hydrogen bond of the N–H⋯O type which is the driving force for the regioselective formation of the ortho isomer. Oxidation of 1-chloro-4-nitrobenzene (1o) in EtCN and n-PrCN afforded N-(2-chloro-5-nitrophenyl)propionamide (3o) and N-(2-chloro-5-nitrophenyl)butyramide (4o) as the only isomers in 85% and 82% yields while oxidation of 1o in MeCN gave two regioisomers (2o-a and 2o-b) with an a : b ratio of 7 : 1 and a total yield of 88%. The structure of 3o was confirmed by single-crystal X-ray analysis (CCDC 2336823†). Complete conversion of 1o occurred after passing 4.8 F mol⁻¹ of electricity while 1-chloro-2,4-dinitrobenzene failed to afford any amidation products.

1-Fluoro-4-nitrobenzene (1p) in MeCN, EtCN and n-PrCN gave N-phenylamides 2p, 3p and 4p in 68%, 72% and 70% yields, respectively. Oxidation of 1-bromo-4-nitrobenzene (1q) in nitrile solvents led to the formation of anilides 2q and 3q in lower yields (64% and 57%). Anilide formation with an amide bond predominantly at the ortho position to the halogen (Cl, F, Br) was probably due to the fact that the positive mesomeric effect of the halogen outweighs the formation of an intramolecular hydrogen bond between the amide radical and nitro group in the transition state (which probably directs the attack on the ortho-position in the case of nitrobenzene).

Electro-oxidation of a number of arenes that are quite easy to oxidize (below +2 V), such as methoxybenzene, naphthalene, 2,6-dimethylnaphthalene, 1,3,5-trimethoxybenzene, and anthracene, or arenes with bromine or iodine substituents under similar conditions resulted solely in a C–C coupling.²³

When the reaction mixture after electrolysis was heated at 60 °C for half an hour, the resulting amides were converted into amines due to hydrolysis, as shown for a number of the obtained N-phenylamides (Scheme 4). Heating of p-amidophenols 2t–2v did not lead to the formation of amines.

Scheme 4 Formation of amines from the obtained amides during hydrolysis and heating. Isolated yield according to arenes.

To demonstrate the capabilities of the proposed approach, a paired electrosynthesis experiment in MeCN was carried out using 1o as a model substrate (Scheme 5). Oxidation of 1o resulted in 2o-a with 74% yield, while reduction of 1o gave N-(4-chlorophenyl)hydroxylamine (1oa) in 83% yield after passing 4.8 F mol⁻¹ of electricity (see section 2.4 of the ESI†). Note that the substrates in the anode and cathode areas do not need to be identical.

Scheme 5 Paired electrosynthesis: oxidation of 1o in MeCN (left), reduction of 1o in MeCN (right). Isolated yield.

To understand the underlying mechanism of this process, a detailed investigation involving electron paramagnetic resonance spectroscopy (EPR) and quantum chemical calculations was performed.

Electron paramagnetic resonance spectroscopy (EPR)

The formation of reactive radicals during electrochemical oxidation of arenes in nitriles was monitored by electron paramagnetic resonance (EPR) spectroscopy. We began our research with the oxidation of a solution of acetonitrile (2 ml) in water (100 μl) in the presence of N-tert-butyl-α-phenylnitrone (PBN) as a spin-trap (C = 0.5 mg ml⁻¹). The EPR spectrum of PBN–OH˙ spin adduct A was recorded (Fig. 3), which indicated that the process can begin with the formation of a hydroxyl anion radical that can further react with nitrile or arene.

Fig. 3 EPR spectrum of PBN–OH˙ recorded during oxidation of a solution of MeCN (2 ml) in water (100 μl) at 293 K with simulations. g = 2.001, a_N = 13.9 G, a_H = 2.3 G. C (PBN) = 0.5 mg ml⁻¹.

To reveal the formation of other spin adducts, an EPR study was further carried out on the oxidation of a solution of benzene in MeCN and EtCN. Benzene was taken as a model arene because its oxidation potential is slightly lower (+2.56 V) than the oxidation potentials of other aromatic compounds (Table S1†). Oxidation of a benzene solution in MeCN and EtCN containing 0.8 equiv. of water in the presence of a PBN spin-trap provided the EPR spectra as shown in Fig. 4 and 5, which are notably distinct from each other. The EPR spectrum obtained during the oxidation in EtCN gave a clearer line splitting than the spectrum obtained in acetonitrile. The obtained values for the number of lines, constants and g-factor in both cases indicate a structure containing two nitrogen atoms and two protons, which may correspond to intermediates B and C (Fig. 4 and 5).

Fig. 4 EPR spectrum of spin-adduct B recorded during oxidation of a benzene solution in MeCN with water (0.8 equiv.) at 293 K with simulations. g = 2.009, a_N1 = 5.2 G, a_N2 = 4.4 G, a_H1 = 5.3 G, a_H2 = 9.2 G. C (PBN) = 1.5 mg ml⁻¹.

Fig. 5 EPR spectrum of spin-adduct C recorded during oxidation of benzene solution in EtCN with water (0.8 equiv.) at 293 K with simulations. g = 2.006, a_N1 = 5.24 G, a_N2 = 4.32 G, a_H1 = 5.15 G, a_H2 = 9.24 G. C (PBN) = 1.5 mg ml⁻¹.

Unfortunately, we did not find the EPR spectra of similar structures in the literature. The high-resolution mass spectrum of an oxidized solution of benzene in acetonitrile with water (0.8 equiv.) in the presence of the spin-trap PBN also indicates the most likely formation of intermediate B (molecular peak 235.1441) (Fig. 4).

As we did not detect any aromatic radicals with EPR, but detected products of OH-radical and RC(OH)N radical additions to PBN, we can conclude that the reaction starts with the oxidation of a water molecule on the anode (at least in the case of the platinum anode, which is known to adsorb OH radicals).

After the formation of an OH radical, the reaction can follow one of the four pathways shown in Scheme 6. However, conventional quantum chemistry methods are not well suited to modeling heterogeneous processes, or the processes directly occurring in the near-electrode layers of the solution, so we cannot reliably quantify their plausibilities via calculations.

Scheme 6 Possible pathways involving the OH radical after its formation on the Pt anode.

Previous studies by Kissner et al. suggest²⁴ that the OH radical is capable of attacking acetonitrile (first stage of Pathways 1 and 2). On the other hand, Pathway 2 is, in fact, a formal SET reaction of the OH radical with acetonitrile, and this process can be expected to be thermodynamically unfavorable – nitriles have higher oxidation potentials compared to water.²⁵ Pathway 3 can be ruled out by the EPR studies of the reaction mixture, which have not shown any signs of aromatic radicals.

On the other hand, hydroxylated aromatics are widely recognized intermediates in such processes and have been extensively studied in the anodic oxidation of organic pollutants (Pathway 4).^26,27 However, this particular pathway results in the formation of an Ar–O bond, which only forms as a replacement of the Ar–F bond in our above examples (Scheme 3), but never of Ar–H (Table 2). So, we consider this mechanism unfeasible.

Thus, only Pathway 1 is in full agreement with our experimental observations.

Quantum chemical calculations

Molecular modeling involving concurrent quantum chemical methods can be used to establish the feasibility of the suggested mechanism (Pathway 1).

We have computed all steps of Pathway 1 at the DLPNO-CCSD(T)²⁸/def2-TZVPP²⁹/CPCM(CH₃CN)//B3LYP^30,31-D3BJ^32,33/def2-TZVP/CPCM(CH₃CN)³⁴ level of theory. Transition states were located using conformational searches in CREST,³⁵ followed by pre-optimization³⁶ and, finally, optimization in Orca (see the ESI†).³⁷ The computed free energy profile for Pathway 1 using benzene as a model aromatic compound is shown in Fig. 6.

Fig. 6 Relative free energies along the reaction coordinate, calculated at the DLPNO-CCSD(T)/def2-TZVPP/CPCM(CH₃CN)//B3LYP-D3BJ/def2-TZVP/CPCM(CH₃CN) level of theory. Only the minimum energy structures are shown; different elements are shown in different colors: red – oxygen, grey – carbon, white – hydrogen and blue – nitrogen. The bottom section shows the modeled mechanism.

Pathway 1 further splits into two possible routes with the attacking radical particle being T1 or T2 (Fig. 6). According to quantum chemical calculations, in both cases, the addition of T1/T2 to benzene constitutes the rate-limiting step. The first route has a lower activation energy of ∼21 kcal mol⁻¹, which is feasible for a reaction proceeding at room temperature.

Conversion to the final reaction product can proceed in two ways: through oxidation of the intermediate radical I1 to A1 at the anode and subsequent deprotonation, or via direct abstraction of the hydrogen atom by an OH radical (Fig. 6). To confirm that the second process is not rate limiting, we performed a relaxed PES scan that showed that it is barrierless (Fig. S16†).

Conclusions

Phenylamides are an important class of substances with practically significant properties, so the search for new sustainable approaches to obtain them is undoubtedly relevant. Organic electrosynthesis in which electrons are utilized as a traceless agent is of increasing interest in dehydrogenative C–H/N–H cross-coupling reactions. In this article, we have shown that the use of excess sulfuric acid in electrochemical C(sp²)–H amidation can be replaced simply by the use of a membrane (i.e., use of a divided cell instead of an undivided cell) and shown that these conditions are highly universal by synthesizing 60 N-phenylamides and 12 benzoxazoles directly from arenes using nitriles as a source of an amide or amine moiety. The only requirement is for the aryl to have a higher oxidation potential than that of water; otherwise, dimers are formed.²³ The established protocol features excellent yields of up to 89% under mild reaction conditions, allows functionalization of difficult-to-oxidize arenes, and is amenable to gram-scale synthesis and paired electrosynthesis. The reaction proceeds more selectively and with a higher product yield when difficult-to-oxidize (more positive than +3.2 V) arenes are used. Based on the data of preparative electrosynthesis, cyclic voltammetry, EPR studies in the presence of a spin-trap and quantum chemical calculations, it was shown that the process starts with the formation of hydroxyl radicals, which react with nitrile and then attach to arenes to afford desirable N-phenylamides.

Author contributions

Conceptualization: SOS, AIK, MGM. Methodology: AIK, SOS, VIM, MGM, EVK, OBB, VMB, KAI. Investigation: AIK, SOS, VIM, KVB, VIT. Data curation: AIK, SOS, VIM, KVB, VIT, MGM. Writing – original draft: SOS, AIK. Writing – reviewing and editing: SOS, AIK, MGM, YHB. Visualization: SOS, MGM. Supervision: SOS, AIK, MGM, YHB.

Data availability

The data supporting this article have been included as part of the ESI.†

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This research was supported by the grant from the Russian Science Foundation, No. 22-13-00017.

The measurements were carried out using the equipment of the Distributed Spectral-Analytical Center of Shared Facilities for Study of Structure, Composition and Properties of Substances and Materials of the FRC Kazan Scientific Center of RAS.

The Siberian Branch of the Russian Academy of Sciences (SB RAS) Siberian Supercomputer Center is gratefully acknowledged for providing supercomputer facilities. This work has been carried out using computing resources of the federal collective usage center Complex for Simulation and Data Processing for Mega-science Facilities at NRC “Kurchatov Institute”, https://ckp.nrcki.ru/. This research was supported in part through computational resources of HPC facilities at NRU HSE. The research is carried out using the equipment of the shared research facilities of HPC computing resources at Lomonosov Moscow State University.³⁸

References

1. (a) S. Kumari, A. V. Carmona, A. K. Tiwari and P. C. Trippier, Amide bond bioisosteres: Strategies, synthesis, and successes, J. Med. Chem., 2020, 63(21), 12290–12358; (b) S. Mahesh, K. C. Tang and M. Raj, Amide bond activation of biological molecules, Molecules, 2018, 23(10), 2615; (c) C. W. Cheung, N. Shen, S. P. Wang, A. Ullah, X. Hu and J. A. Ma, Manganese-mediated reductive amidation of esters with nitroarenes, Org. Chem. Front., 2019, 6(6), 756–761.
2. X. Wang, Challenges and outlook for catalytic direct amidation reactions, Nat. Catal., 2019, 2, 98–102.
3. (a) A. B. Bedadurge and A. R. Shaikh, Designing hypothesis of 2-substituted-N-[4-(1-methyl-4, 5-diphenyl-1H-imidazole-2-yl) phenyl] acetamide analogs as anticancer agents: QSAR approach, J. Korean Chem. Soc., 2013, 57(6), 744–754; (b) D. Kadadevar, K. C. Chaluvaraju, M. S. Niranjan, C. M. Sultanpur, S. Kumar, M. Nagaraj, M. Hegde, M. Smitha and K. Chakraborty, Synthesis of N-(substituted phenyl)-2 [5-phenyl-2H-1, 2, 4-triazol-3ylamino] acetamide as anticonvulsant, Int. J. ChemTech Res., 2011, 3(3), 1064–1069; (c) K. R. Hayes, Handbook of Pesticide Toxicology, ed. R. Krieger, Academic Press, New York, 3rd edn, 2010, pp. 1753–1769; (d) M. Krátký and J. Vinšová, Antifungal activity of salicylanilides and their esters with 4-(trifluoromethyl) benzoic acid, Molecules, 2012, 17(8), 9426–9442.
4. S. Warren and P. Wyatt, Organic synthesis: the disconnection approach, John Wiley & Sons, 2nd edn, 2008, p. 352.
5. R. K. Singh, A. Kumar and A. K. Mishra, Chemistry and pharmacology of acetanilide derivatives: a mini review, Lett. Org. Chem., 2019, 16(1), 6–15.
6. J. J. Ritter and P. P. Minieri, A new reaction of nitriles. I. amides from Alkenes and Mononitriles, J. Am. Chem. Soc., 1948, 70(12), 4045–4048.
7. K. Kiyokawa, T. Watanabe, L. Fra, T. Kojima and S. Minakata, Hypervalent iodine(III)-mediated decarboxylative Ritter-type amination leading to the production of α-tertiary amine derivatives, J. Org. Chem., 2017, 82, 11711–11720.
8. C. Liu and F. Yang, Au–Catalyzed Stereoselective Ritter Reaction of Haloalkynes with Nitriles for (Z)–β–Halogenated Enamides, Eur. J. Org. Chem., 2019,(40), 6867–6870.
9. H. Bao, B. Zhou, H. Jin and Y. Liu, Copper-catalyzed three-component reaction of N-heteroaryl aldehydes, nitriles, and water, Org. Biomol. Chem., 2019, 17, 5021–5028.
10. N. A. Wippert, N. Jung and S. Bräse, Synthesis of arylamides via Ritter-type cleavage of solid-supported aryltriazenes, ACS Comb. Sci., 2019, 21, 568–572.
11. K. Zhang, X.-H. Xu and F.-L. Qing, Copper–Promoted Ritter–Type Trifluoroethoxylation of (Hetero) arenediazonium Tetrafluoroborates: A Method for the Preparation of Trifluoroethyl Imidates, Eur. J. Org. Chem., 2016,(30), 5088–5090.
12. X. Peng, Z. Sun, P. Kuang, L. Li, J. Chen and J. Chen, Copper-catalyzed selective arylation of nitriles with cyclic diaryl iodonium salts: Direct access to structurally diversified diarylmethane amides with potential neuroprotective and anticancer activities, Org. Lett., 2020, 22, 5789–5795.
13. M. E. Chen, X. W. Chen, Y. H. Hu, R. Ye, J. W. Lv, B. Li and F. M. Zhang, Recent advances of Ritter reaction and its synthetic applications, Org. Chem. Front., 2021, 8(16), 4623–4664.
14. (a) M. D. Kärkäs, Electrochemical strategies for C–H functionalization and C–N bond formation, Chem. Soc. Rev., 2018, 47(15), 5786–5865; (b) J. E. Kim, S. Choi, M. Balamurugan, J. H. Jang and K. T. Nam, Electrochemical C–N bond formation for sustainable amine synthesis, Trends Chem., 2020, 2(11), 1004–1019; (c) Y. Zhang, C. Liu, J. Liu, W. Li and H. Lu, Recent advances in electrochemical CH bond amination, Org. Chem. Front., 2023, 10, 5309–5330.
15. (a) D. B. Clark, M. Fleischmann and D. Pletcher, The partial anodic oxidation of n-alkanes in acetonitrile and trifluoroacetic acid, J. Chem. Soc., Perkin Trans., 1973, 11, 1578–1581; (b) V. R. Koch and L. L. Miller, Anodic chemistry of adamantyl compounds. Scissible carbon, halogen, hydrogen, and oxygen substituents, J. Am. Chem. Soc., 1973, 95(26), 8631–8637.
16. L. L. Miller and V. Ramachandran, Remote anodic acetamidation of esters via carbonium ions, J. Org. Chem., 1974, 39(3), 369–372.
17. J. Y. Becker, L. R. Byrd, L. L. Miller and Y. H. So, Remote anodic substitution of ketones, J. Am. Chem. Soc., 1975, 97(4), 853–856.
18. (a) M. A. Kabeshov, B. Musio and S. V. Ley, Continuous direct anodic flow oxidation of aromatic hydrocarbons to benzyl amides, React. Chem. Eng., 2017, 2(6), 822–825; (b) T. Shen and T. H. Lambert, C–H amination via electrophotocatalytic ritter-type reaction, J. Am. Chem. Soc., 2021, 143(23), 8597–8602; (c) L. Zhang, Y. Fu, Y. Shen, C. Liu, M. Sun, R. Cheng, W. Zhu, X. Qian, Y. Ma and J. Ye, Ritter-type amination of C (sp3)-H bonds enabled by electrochemistry with SO42−, Nat. Commun., 2022, 13(1), 4138; (d) Y. Xu, Q. Li, R. Ye, B. Xu and X. Zhou, Electrochemical Oxidative C–H Amination through a Ritter-Type Reaction, J. Org. Chem., 2023, 88, 9518–9522; (e) S. Strekalova, A. Kononov and Y. Budnikova, Amino acids in electrochemical metal-free benzylic CH amidation, Tetrahedron Lett., 2022, 102, 153917; (f) S. Strekalova, A. Kononov, I. Rizvanov and Y. Budnikova, Acetonitrile and benzonitrile as versatile amino sources in copper-catalyzed mild electrochemical C–H amidation reactions, RSC Adv., 2021, 11, 37540–37543; (g) Y. Liang, X. Zhan, F. Li, H. Bi, W. Fan, S. Zhang and M. B. Li, Using a nitrogen-centered radical as a selective mediator in electrochemical C (sp3)-H amination, Chem. Catal., 2023, 3, 100582; (h) Q. Chu, Z. Feng, S. Zhang, P. Liu and P. Sun, Three-component reaction for the synthesis of imides enabled by electrochemical C(sp3)–H functionalization, Green Chem., 2023, 25, 6728–6732; (i) P. Qian, D. Zhu, X. Wang, Q. Sun and S. Zhang, Electrochemical Benzylic C(sp3)-H Imidation Enabled by Benzoic Acid Derived Radicals., J. Org. Chem., 2024, 89, 6395–6404; (j) F. Barba, I. Barba and B. Batanero, First aromatic ring acetamidation by anodic oxidation, Electrochem. Commun., 2014, 48, 115–117; (k) I. M. Taily, D. Saha and P. Banerjee, Direct synthesis of paracetamol via site-selective electrochemical Ritter-type C–H amination of phenol, Org. Lett., 2022, 24(12), 2310–2314; (l) Y. Fu, L. Zhang, M. Sun, L. Cao, L. Yang, R. Cheng, Y. Ma and J. Ye, Direct Electrochemical Ritter–Type Amination of Electron–Deficient Arenes, Eur. J. Org. Chem., 2023, e202300553; (m) S. Strekalova, A. Kononov, V. Morozov, O. Babaeva, E. Gavrilova and Y. Budnikova, Electrochemical approach to amide bond formation, Adv. Synth. Catal., 2023, 365, 3375–3381.
19. (a) A. D. Stergiou, D. H. Broadhurst and M. D. Symes, Highly selective electrocatalytic reduction of substituted nitrobenzenes to their aniline derivatives using a polyoxometalate redox mediator, ACS Org. Inorg. Au, 2022, 3(1), 51–58; (b) T. Wirtanen, E. Rodrigo and S. R. Waldvogel, Recent advances in the electrochemical reduction of substrates involving N–O bonds, Adv. Synth. Catal., 2020, 362(11), 2088–2101; (c) M. Balasingam, J. D. Becker, P. Guerrero-Almaraz, H. Sun and M. D. Koppang, Dual Electrode Detection in LC–EC Analysis of Sanger–tagged Amino Acids: Electrochemical Reduction of Aromatic Nitro Groups, Electroanalysis, 2022, 34(12), 1854–1868.
20. (a) T. V. Losev, I. S. Gerasimov, M. V. Panova, A. A. Lisov, Y. R. Abdyusheva, P. V. Rusina, E. Zaletskaya, O. V. Stroganov, M. G. Medvedev and F. N. Novikov, Quantum mechanical-cluster approach to solve the bioisosteric replacement problem in drug design, J. Chem. Inf. Model., 2023, 63(4), 1239–1248; (b) J. Wang, M. Sánchez-Roselló, J. L. Aceña, C. del Pozo, A. E. Sorochinsky, S. Fustero, V. A. Soloshonok and H. Liu, Fluorine in pharmaceutical industry: fluorine-containing drugs introduced to the market in the last decade (2001–2011), Chem. Rev., 2014, 114, 2432–2506.
21. I. S. Gerasimov, T. V. Losev, E. Y. Epifanov, I. Rudenko, I. S. Bushmarinov, A. A. Ryabov, P. A. Zhilyaev and M. G. Medvedev, Comment on “Pushing the frontiers of density functionals by solving the fractional electron problem”, Science, 2022, 377(6606), eabq3385.
22. I. Chiarotto, L. Mattiello and M. Feroci, The electrogenerated cyanomethyl anion: an old base still smart, Acc. Chem. Res., 2019, 52, 3297–3308.
23. A. Kononov, S. Strekalova, E. Kobeleva, G. Savelyev, A. Zlygostev, M. Khvorova, V. Morozov, O. Babaeva and Y. Budnikova, C–C and C–N bond formation in electro-oxidation reactions of aromatic compounds, Curr. Res. Green Sustainable Chem., 2024, 8, 100406.
24. R. Kissner, G. Welti and G. Geier, The hydrolysis of gold(I) in aqueous acetonitrile solutions, J. Chem. Soc., Dalton Trans., 1997, 10, 1773–1778.
25. L. C. Portis, J. C. Roberson and C. K. Mann, Anodic background reaction in moist acetonitrile, Anal. Chem., 1972, 44(2), 294–297.
26. M. Panizza and G. Cerisola, Direct and mediated anodic oxidation of organic pollutants, Chem. Rev., 2009, 109(12), 6541–6569.
27. B. Fleszar and J. Poszyńska, An attempt to define benzene and phenol electrochemical oxidation mechanism, Electrochim. Acta, 1985, 30(1), 31–42.
28. C. Riplinger and F. Neese, An efficient and near linear scaling pair natural orbital based local coupled cluster method, J. Chem. Phys., 2013, 138(3), 034106.
29. F. Weigend and R. Ahlrichs, Balanced basis sets of split valence, triple zeta valence and quadruple zeta valence quality for H to Rn: Design and assessment of accuracy, Phys. Chem. Chem. Phys., 2005, 7(18), 3297–3305.
30. A. D. Becke, Density–functional thermochemistry. III. The role of exact exchange, J. Chem. Phys., 1993, 98(7), 5648–5652.
31. P. J. Stephens, F. J. Devlin, C. F. Chabalowski and M. J. Frisch, Ab initio calculation of vibrational absorption and circular dichroism spectra using density functional force fields, J. Phys. Chem., 1994, 98(45), 11623–11627.
32. S. Grimme, J. Antony, S. Ehrlich and H. Krieg, A consistent and accurate ab initio parametrization of density functional dispersion correction (DFT-D) for the 94 elements H-Pu, J. Chem. Phys., 2010, 132(15), 154104.
33. S. Grimme, S. Ehrlich and L. Goerigk, Effect of the damping function in dispersion corrected density functional theory, J. Comput. Chem., 2011, 32(7), 1456–1465.
34. V. Barone and M. Cossi, Quantum calculation of molecular energies and energy gradients in solution by a conductor solvent model, J. Phys. Chem. A, 1998, 102(11), 1995–2001.
35. S. Grimme, J. Chem. Theory Comput., 2019, 15(5), 2847–2862.
36. M. G. Medvedev, M. V. Panova, G. G. Chilov, I. S. Bushmarinov, F. N. Novikov, O. V. Stroganov, A. A. Zeifman and I. V. Svitanko, Exhaustive conformational search for transition states: the case of catechol O-methyltransferase active site, Mendeleev Commun., 2017, 27, 224–227.
37. F. Neese, F. Wennmohs, U. Becker and C. Riplinger, The ORCA quantum chemistry program package, J. Chem. Phys., 2020, 152(22), 224108.
38. V. Sadovnichy, A. Tikhonravov, V. Voevodin and V. Opanasenko, ‘Lomonosov’: Supercomputing at Moscow State University, in Contemporary high performance computing: From petascale toward exascale, CRC Press, 2013, pp. 283–307.

---

Footnotes

† Electronic supplementary information (ESI) available: PDF with methods and compound characterization, XYZ file with all stationary points located during molecular modeling. CCDC 2336800 and 2336823. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d4qo01296h

‡ These authors contributed equally.

---