Solvent-controlled, chemodivergent oxidative anionic Fries rearrangement of O-aryl carbamates under aerobic conditions

Abstract

We disclose herein a mild and efficient organolithium-mediated protocol which enables the chemodivergent transformation of ortho-cresol-derived O-aryl carbamates into diverse molecular structures by simply changing the nature of the reaction medium, working under air and at room temperature. The use of the biobased 2-MeTHF as solvent allows the chemoselective preparation of α-hydroxy arylacetamides in a single synthetic operation with a remarkable functional group tolerance. Our strategy, which exploits the presence of molecular oxygen arising from the use of bench-type aerobic conditions, relies on a one-pot anionic homo-Fries rearrangement/amide enolate autoxidation sequence with two consecutive C–C/C–O bond formation events occurring at the same carbon atom. Furthermore, we also describe the successful use of protic and bioinspired Deep Eutectic Solvents (DESs) as an effective tool to tune the chemoselectivity of the proposed transformation. The fast internal protonolysis of the anion solution, operated by the protic reaction medium, results in an interrupted metalation/rearrangement sequence, enabling the chemoselective preparation of arylacetamides under bench-type aerobic conditions owing to the efficient suppression of the oxidation step.

---

Introduction

The α-hydroxy amide function is of utmost importance in several areas of organic chemistry owing to its widespread occurrence in agrochemicals and pharmaceuticals with diversified biological targets.¹ In this framework, a recurrent motif in bioactive molecules is the α-hydroxy arylacetamide skeleton, the amide derivative of the naturally occurring mandelic acid, which represents a key building block in both the development of novel drug candidates and the production of fine chemicals (Fig. 1A).² Hence, the development of new bench-stable reagents and effective methodologies for the chemoselective preparation of α-hydroxy arylacetamide derivatives using cheap and readily available synthons is of remarkable synthetic value. Besides the traditional approaches based on the amidation of α-hydroxy acids³ or the reduction of α-keto amides,⁴ several de novo synthetic methods relying on the assembly of novel C–C bonds have been developed (Fig. 1B), including the nucleophilic homologation of aldehydes by carbamoyl anions,⁵ multicomponent approaches with isocyanides⁶ or difluorocarbenes,⁷ and the base-promoted anionic [1,2]-rearrangement of O-benzyl carbamates.⁸ A more powerful approach towards mandelic acid amides relies on the α-hydroxylation of arylacetamides by means of a selective C(sp³)–O bond formation. Excellent indirect strategies have been developed, including the chemoselective α-oxyamination of electrophilically activated arylacetamides promoted by TEMPO⁹ and the anionic [2,3]-rearrangement of O-functionalized hydroxamic acids.¹⁰ Direct methods based on the aerobic oxidation of C(sp³)–H bonds are of particular interest, as molecular oxygen is an inexpensive and ubiquitous reagent, and its exploitation in synthetic protocols (rather than its depletion) is of high value also from a sustainability point of view.¹¹ In this context, the direct aerobic α-hydroxylation of arylacetamides (or α-aryllactams) has been accomplished by means of transition metal-promoted,¹² organocatalytic,¹³ and base-promoted¹⁴ approaches (Fig. 1B). However, oxygen-mediated processes frequently suffer from both low selectivity and undesired over-oxidation or degradation pathways.¹⁵ In this context, opportunities for the preparation of α-hydroxy arylacetamides might also arise from the oxidation of nucleophilic polar s-block organometallic reagents, a well-established reaction typically promoted by metal oxenoids (hydroperoxide metal salts)¹⁶ or occurring by autoxidation of the organometallic species in the presence of triplet molecular oxygen¹⁷ (Fig. 1C). Despite the enormous progress in this area, especially in the oxidation of aryllithium compounds¹⁸ and enolates,¹⁹ the use of molecular oxygen for the α-hydroxylation of alkali-metal amide enolates has been barely investigated.²⁰ Recent advancements in the chemistry of s-block organometallic compounds have clearly established that transformations involving organolithium, Grignard, and organosodium reagents could be efficiently performed under “bench-type” conditions, working at room temperature in the presence of air, using non-dried sustainable reaction media.²¹

Fig. 1 (A and B) State of the art in the synthesis of α-hydroxy arylacetamides. (C) General overview of the oxidation of organometallic reagents and (D) aim of this work.

Supported by our recent findings on the reactivity of polar organometallic reagents under non-conventional conditions,²² we envisaged the possibility of designing a streamlined organolithium-mediated strategy to access α-hydroxy arylacetamides which exploits the presence of molecular oxygen in the reaction mixture, arising from the use of bench-type aerobic conditions. We thus herein report a systematic study on the synergistic combination of lithium amides and molecular oxygen to promote the transformation of ortho-cresol derived O-aryl carbamates into α-hydroxy arylacetamides using a one-pot, tandem anionic 1,4-O–C Fries (homo-Fries)²³ rearrangement/oxidation sequence which involves two bond-making events (one C–C and one C–O bond) occurring at the same carbon atom, working under air and at room temperature, using 2-MeTHF as solvent (Fig. 1D). Mechanistic insights revealed that the oxidation of the intermediate lithium amide enolate triggered by molecular oxygen follows a classic autoxidation pathway with a S_N2-type cleavage process. Furthermore, we also describe the successful use of bioinspired Deep Eutectic Solvents (DESs) as an effective tool to tune the chemoselectivity of the proposed transformation. The use of a protic heterogeneous solvent mixture allows the suppression of the oxidation step owing to the fast protonolysis of the organolithium species (Fig. 1D), enabling the chemoselective preparation of arylacetamides²⁴ under bench-type aerobic conditions.

Results and discussion

Reaction development

We started our investigation on the anionic homo-Fries rearrangement/oxidation sequence of O-2-tolyl carbamates under bench-type aerobic conditions using the sterically hindered N,N-diisopropyl carbamate 1a as a model substrate (Table 1). In a preliminary experiment, a solution of compound 1a (0.2 mmol, 0.2 M) in cyclopentyl methyl ether (CPME) was reacted with a freshly prepared solution of LiTMP (1 M in CPME, 2.0 equiv.) in the absence of an external electrophile, working at room temperature and under air (Table 1, entry 1).^22d Aqueous quench of the anion solution after 30 min afforded a mixture of unreacted carbamate 1a (64%) and the arylacetamide 3a hydroxylated at the benzylic position in 34% yield. The structure was confirmed by NMR spectroscopy and X-ray analysis of 3a, which crystallised by slow, room temperature evaporation of a chloroform solution into the monoclinic P2₁/n space group.²⁵ Performing the metalation of 1a using 2-MeTHF as reaction medium under the same reaction conditions led to a significant increase of the reaction yield of 3a (54%), alongside the formation of unidentified byproducts (entry 2). Pleasingly, increasing the amount of metalating agent (3 equiv.) resulted in the chemoselective formation of the desired α-hydroxy arylacetamide 3a in a satisfactory 70% yield (entry 3). The use of higher amounts of LiTMP or, alternatively, longer reaction times had a detrimental effect on the reaction yield due to the partial over-oxidation of 3a to the corresponding α-ketoamide byproduct (see the SI for details). Similar results were obtained when the reaction was performed using either the emerging eco-friendly alternative 4-MeTHP (entry 4) or other ethereal solvents with high coordinating ability for organolithiums, such as 1,2-dimethoxyethane (DME, entry 5) and tetrahydrofuran (THF, entry 6). While the use of less coordinating solvents had no effect on the anionic rearrangement, a significant impact on the oxidation step was observed as demonstrated by the considerable amount of unoxidized 2a (17–35%) detected in the reaction mixture (entries 7–9). By contrast, treatment of carbamate 1a with the less basic and sterically hindered lithium amide LDA (entry 10) resulted in a significant decrease of the reaction yield (34%). Analogously, the use of common alkyllithium compounds was less effective, resulting in a loss of selectivity due to the formation of ortho-Fries (s-BuLi)^22d or several unidentified byproducts (t-BuLi) in non-negligible amounts (entries 11 and 12). Noteworthy, the stability of the organolithium species in 2-MeTHF is remarkable under these bench-type aerobic conditions. Performing the reaction on 1a in 2-MeTHF either under a dry Ar/O₂ atmosphere or in the presence of humidified air (82% RH) provided comparable yields of 3a (71% and 69%, respectively) to those obtained with a standard bench-type (open air) experimental setup (39% RH, see the SI for details). These results clearly indicate that almost no protonolysis of the organolithium species occurs under air over the reaction time, owing to the beneficial effect of the highly hydrophobic 2-MeTHF on the stability of lithium amides.^22b,26 Consequently, the reaction could be performed even at high moisture levels without significantly affecting the reaction yield.

Table 1 Oxidative anionic homo-Fries rearrangement of carbamate 1a under bench-type aerobic conditions^a

Entry	RLi (eq.)	Solvent	2a ^b (%)	3a ^b (%)
a Reaction conditions: 1a (0.2 mmol), solvent (1 mL), RLi, room temperature (RT), under air (relative humidity = 39%). CPME = cyclopentyl methyl ether, 2-MeTHF = 2-methyltetrahydrofuran, 4-MeTHP = 4-methyltetrahydropyran, DME = 1,2-dimethoxyethane, MTBE = methyl t-butyl ether). LiTMP (1 M in 2-MeTHF), LDA (1 M in 2-MeTHF), t-BuLi (1.7 M in pentane), s-BuLi (1.4 M in cyclohexane).b Determined by ^1H NMR using n-heptane as the internal standard.c Unreacted 1a (64%) was recovered.d Yield of isolated 3a = 68%.e The corresponding N,N-diisopropyl-2-oxoacetamide was also detected in the reaction mixture. See the SI for details.f The ortho-rearranged salicylamide (45%) was observed in the crude reaction mixture. See ref. 22d.
1	LiTMP (2)	CPME	0	34^c
2	LiTMP (2)	2-MeTHF	0	54
3	LiTMP (3)	2-MeTHF	0	70^d
4	LiTMP (3)	4-MeTHP	0	53^e
5	LiTMP (3)	DME	0	62^e
6	LiTMP (3)	THF	0	54^e
7	LiTMP (3)	Et2O	35	50
8	LiTMP (3)	MTBE	17	48
9	LiTMP (3)	n-hexane	32	57
10	LDA (3)	2-MeTHF	0	34
11	t-BuLi (3)	2-MeTHF	0	32
12	s-BuLi (3)	2-MeTHF	10	36^e^,^f

---

Scope of the reaction

With optimized reaction conditions in hand (Table 1, entry 3), the scope and limitations of this transformation were evaluated for a series of functionalized O-aryl carbamates 1 bearing different substituents at both the O-aryl ring and the amine moiety, using the commodity metalating agent LiTMP in 2-MeTHF under air. The anionic Fries rearrangement of O-aryl N,N-diisopropylcarbamates 1a–q proceeded smoothly en route to a variety of substituted α-hydroxy arylacetamides bearing neutral (3a–b), electron-donating (3f–i and 3o), halogenated (3c and 3e) groups at the aromatic moiety, and polyaromatic rings (3j–k) without formation of byproducts arising from competitive directed ortho-, remote²⁷ or peri-²⁸ metalation processes (Scheme 1). Remarkably, the synthesis of compound 3a has been easily scaled up to 4.0 mmol of carbamate 1a (1 g) with comparable efficiency in terms of yield and selectivity (66% versus 68% on a small scale, see the SI). Pleasingly, our methodology also tolerates the presence of common organolithium-sensitive functional groups, allowing the chemoselective preparation of several polyfunctionalized α-hydroxy arylacetamides without competitive pathways, such as X/Li exchange (3d), transmetalation (3l), carbolithiation (3m–n) or α-lithiation (3h and 3p) reactions. Treatment of carbamate 1q, prepared from the ortho-methyl derivative of eugenol, with LiTMP gave access to the corresponding polysubstituted α-hydroxy arylacetamide 3q, albeit in a modest 34% yield. Other sterically hindered O-tolyl carbamates derived from secondary (cyclo)alkyl (1s–1aa) and aryl (1ab–ac) amines performed as well, thereby delivering a series of α-hydroxylated tertiary acetamides 3s–ad in good yields (21–85%). By contrast, when the reaction was performed on carbamate 1r the sole [1,2]-carbamoyl migration product, arising from the competitive α-lithiation at the O-benzylic position,^8b was recovered in 88% yield (see the SI). Similarly, attempts to extend our methodology to 2-ethylphenol derivatives were unsuccessful. Treatment of the N,N-diisopropyl carbamate of 2-ethylphenol under optimized reaction conditions resulted in a partial anionic homo-Fries rearrangement²⁹ which afforded the corresponding non-oxidized arylacetamide in 36% yield (alongside 52% of unreacted starting material), whereas no oxidation products were detected in the reaction mixture (see the SI).

Scheme 1 Scope of the oxidative homo-Fries rearrangement of O-aryl carbamates 1. Reaction conditions: 1 (0.2 mmol, 0.2 M in 2-MeTHF), LiTMP (1 M in 2-MeTHF, 3 equiv.), room temperature (RT), under air. Reported yields refer to isolated products.

Mechanistic investigations

To gain more mechanistic insights into the anionic homo-Fries rearrangement/oxidation sequence, additional control experiments were performed (Scheme 2A). Carbamate 1a was treated with LiTMP (3 equiv.) under an inert atmosphere (Ar), using dry and thoroughly degassed 2-MeTHF as solvent. Under these conditions, the sole arylacetamide 2a arising from the benzylic metalation/anionic Fries rearrangement was isolated in 73% yield, confirming the effective role of molecular oxygen as the oxidant in the reaction sequence.³⁰ Additionally, electrophilic quenching with deuterium of the anion solution generated by metalation of 1a with LiTMP under air yielded compound 2a-D with good deuterium incorporation (73% D) at the benzylic position after only 10 s of metalation, resulting from a fast enolization process occurring after the carbamoyl migration step. Taken together, these results confirmed that our developed protocol, which allows the chemoselective conversion of O-aryl carbamates to α-hydroxy arylacetamides in a single synthetic operation, exploits both a cascade of LiTMP-mediated transformations and the presence of molecular oxygen arising from the use of bench-type conditions.

Scheme 2 (A) Control experiments on O-aryl carbamate 1a. Reaction conditions: 1a (0.2 mmol, 0.2 M in 2-MeTHF), LiTMP (1 M in 2-MeTHF, 3 equiv.), room temperature (RT). (i) Reaction was performed under an Ar atmosphere. (ii) Reaction was performed under air. CD₃OD (5 eq.) was added after 10 s. Reported yields refer to isolated products. D incorporations are based on ¹H NMR integration and confirmed with ²H NMR. (B) Mechanistic insights into the oxidation step. Reaction conditions: 2a or 4a (0.2 mmol, 0.2 M in 2-MeTHF), LiTMP (1 M in 2-MeTHF), room temperature (RT), under air. ^a Determined by ¹H NMR using n-heptane as the internal standard.

Further mechanistic investigations were subsequently devoted to the oxidation step. The nature of the species involved in the aerobic oxidation of organolithium compounds by molecular oxygen has been clearly established, and oxenoid-type compounds were often proposed as intermediates. The cleavage of the intermediate lithium peroxide is often rationalized by the fast interaction of the electrophilic lithium peroxide with a second molecule of non-oxidized organolithium to form the corresponding lithium alkoxide (autoxidation process).³¹ In our proposed anionic homo-Fries rearrangement/oxidation sequence, we speculated that the cleavage of the intermediate lithium peroxide 2-OOLi could be performed either by the non-oxidized amide enolate 2-Li (autoxidation) or, alternatively, by the excess of lithium amide LiTMP (Scheme 2B). The latter should account for the required three equivalents of metalating agent (one for each metalation step and an additional equivalent for the cleavage of 2-OOLi species) to achieve a satisfactory yield of hydroxylated acetamide 3, as emerged from the optimization studies (see Table 1, entry 3). Hence, additional experiments were performed on arylacetamide 2a to elucidate the nature of the species involved in the nucleophilic cleavage of the intermediate peroxide 2-OOLi generated during the oxidation step (Scheme 2B). As expected, treatment of 2a with a stoichiometric amount of LiTMP (1 equiv.) was ineffective in promoting the oxidation due to the protonolysis of the metalating agent by the more acidic phenolic hydroxy group, leading to the exclusive formation of the corresponding lithium phenoxide and the consequent recovery of unreacted 2a upon acidic quench (Scheme 2B, entry 1). By contrast, the use of a two-fold amount of LiTMP resulted in the complete α-hydroxylation of 2a owing the effective aerobic oxidation of amide enolate 2a-Li, generated in turn by the additional equivalent of metalating agent (entry 2). These findings strongly suggest a classic autoxidation pathway promoted by the non-oxidized 2a-Li, rather than a LiTMP-induced cleavage of the intermediate oxenoid-type species which should theoretically afford an equimolar mixture of 2a:3a. Interestingly, a similar result was observed upon metalation of the anisole-derived acetamide 4a with a stoichiometric amount of LiTMP (1 equiv.), which afforded an almost 1 : 1 mixture of unreacted 4a and α-hydroxy arylacetamide 5a (entry 3). By contrast, complete conversion of 4a was solely observed in the presence of a two-fold excess of metalating agent (entry 4), suggesting in this case a competitive behaviour of LiTMP as nucleophile versus 4a-Li in the cleavage of the intermediate lithium peroxide.

This experimental evidence has been further investigated by DFT calculations. The reaction free energies involved in the nucleophilic cleavage of model oxenoid-type intermediates, bearing a lithium alkoxide (2b-OOLi, A) or a methoxy- (4b-OOLi, C) substituent at the aromatic ring, operated by either the parent non-oxidized organolithiums 2b-Li/4b-Li (pathway a) or LiTMP (pathway b) have been estimated (Scheme 3a). Computational results disclosed that the cleavage of hydroperoxide salt 2b-OOLi (A, Scheme 3b) by the non-oxidized amide enolate 2b-Li,³² according to an established autoxidation pathway which generates two molecules of lithium alkoxide 2b-OLi (B) as products, proceeds through a transition state (TS I) in which the key structural element is a six-membered cyclic (–Li–O–)₃ core organized into a chair-like conformation. The significant elongation of the O–O bond (from 1.43 Å in 2b-OOLi to 1.75 Å in TS I), and the almost linear alignment of the former 2b-Li nucleophilic carbon with the electrophilic oxygen in TS I (C_Li–O–O angle = 16.3°), strongly suggest a S_N2-type process involved in the cleavage of hydroperoxide salt 2b-OOLi.³³ The relatively low cleavage barrier of the O–O bond (estimated ΔG^‡ = 23.7 kcal mol⁻¹), and the significant thermodynamic stability of the resulting lithium alkoxide with respect to the starting reactants, further corroborate the hypothesis that a classic autoxidation process of amide enolate 2b-Li occurs in the presence of molecular oxygen. Conversely, the interaction of 2b-OOLi with the lithium amide LiTMP (Scheme 3b), which should produce the lithium alkoxide 2b-OLi (B) and the N-hydroxy TMP lithium salt (TMPO⁻Li⁺) species, was significantly disfavoured due to the high free energy barrier required for the heterolytic O–O bond cleavage (estimated ΔG^‡ = 99.3 kcal mol⁻¹), arising from the formation of a more sterically crowded and less organized transition state (TS II). As emerged from the calculations, the computed geometry of TS II clearly discourages a nucleophilic cleavage pathway on 2b-OOLi operated by LiTMP (O–O = 1.40 Å, N_Li–O–O angle = 101.0°).

Scheme 3 (a) Proposed nucleophilic cleavage pathways involved in the oxidation step. (b and c) Gibbs free energy profiles (kcal mol⁻¹, at 298 K) estimated by density functional theory (DFT) calculations for the nucleophilic cleavage of (b) 2b-OOLi and (c) 4b-OOLi species. Red dotted lines in TS I-IV represent breaking and forming bonds. Calculations were performed at the M06-2X/def2-TZVP//M06-2X/def2-SVP level; see the SI for details.

A different scenario emerged from computational investigations on the oxidation of the metalated anisole-derived arylacetamide 4b (Scheme 3c). In close analogy to the phenoxide-derived intermediate 2b-OOLi, the nucleophilic cleavage of the hydroperoxide salt 4b-OOLi (C) operated by the unoxidized amide enolate 4b-Li proceeds through the formation of an organized heterodimeric transition state (TS III) containing a bent four-membered (–Li–O–)₂ cyclic core (estimated ΔG^‡ = 35.3 kcal mol⁻¹). Interestingly, in this case the exergonic O–O bond cleavage of the electrophilic 4b-OOLi by LiTMP proceeds via the formation of a relatively more stable transition state TS IV (−5.8 kcal mol⁻¹ versus TS III) with a comparable free energy barrier (estimated ΔG^‡ = 38.9 kcal mol⁻¹). The geometries of both TSs, showing comparable O–O bond distances (1.73 Å and 1.74 Å for TS III and TS IV, respectively) and an almost linear orientation of the nucleophile with the peroxide moiety (C_Li–O–O = 19.2°, TS III and N_Li–O–O = 10.5°, TS IV), were in good accordance with those involved in a S_N2 cleavage process. Overall, these simulations disclosed that in the presence of a lithium alkoxide group on the aromatic ring (arising from the preliminary anionic Fries rearrangement step) the oxidation of the lithium amide enolate follows a classic autoxidation pathway, where the intermediate hydroperoxide salt interacts with the non-oxidized enolate to yield two molecules of the corresponding lithium alkoxide.³⁴ In contrast, in the absence of a complementary anionic site at the aromatic ring (e.g. in the case of arylacetamides 4a–b) the analysis of the free energy barriers indicates that the lithium amide strongly competes against the amide enolate (or is privileged) in the nucleophilic cleavage of the electrophilic peroxide. These findings are in good agreement with the experimental results (see Scheme 2B), where the complete oxidation of arylacetamides 2a (R = OLi) and 4a (R = OMe) has been achieved in the presence of one (2a, entry 2, autoxidation pathway) or two (4a, entry 4) equivalents of metalating agent, respectively.

Hence, we are inclined to propose the initial benzylic metalation of carbamate 1, followed by a fast anionic 1,4-O-C Fries rearrangement, to form the corresponding intermediate arylacetamide 2, which is promptly enolized by the excess of metalating agent (Scheme 4).³⁵ The resulting 2-Li species then undergoes a spontaneous autoxidation process to deliver the α-hydroxy arylacetamide 3 upon protonolysis of the corresponding alkoxide 2-OLi, generated in situ from the nucleophilic cleavage of the intermediate oxenoid species 2-OOLi by the unoxidized amide enolate 2-Li. Altogether, four synthetic steps are involved in the aerobic tandem protocol developed herein: the regioselective lateral (benzylic) lithiation, an anionic carbamoyl migration, a chemoselective lithiation α to amide carbonyl, and the autoxidation of the resulting amide enolate promoted by oxygen. This allows, overall, the consecutive formation of a C–C and a C–O bond at the same carbon atom in a single synthetic operation.

Scheme 4 Proposed reaction mechanism based on experimental data and estimated reaction free energies.

Interrupted anionic Fries rearrangement in deep eutectic solvents

Whereas we have shown that the presence of atmospheric molecular oxygen in the reaction mixture, a commonly detrimental species for organolithium-promoted transformations, could be efficiently exploited to assemble hydroxylated acetamide-based structural motifs, final considerations have been devoted to the possibility of suppressing the anion oxidation step while maintaining the use of bench-type aerobic conditions.

Deuterium labelling experiments performed on carbamate 1a (see Scheme 2A) suggest that the first steps of the reaction sequence (anionic Fries rearrangement and enolization) are faster than the oxidation process, as demonstrated by the relatively high deuterium incorporation in the arylacetamide 2a observed after only 10 s. Kinetic studies of the reaction, using the model carbamate 1a (0.2 mmol) as a substrate in 2-MeTHF (1 mL), revealed the classic profile of a sequential transformation (Fig. 2). During the initial stage of the reaction the oxidation product 3a was not formed. Subsequently, a slow consumption of 2a occurred with the concomitant steady production of 3a over 30 min. A comparison of the observed rate constants (k₁ and k₂)³⁶ confirmed that the oxidation of the amide enolate is the rate determining step of the consecutive process.

Fig. 2 Plots of 1a, 2a and 3a concentrations [M] versus time (s) for the metalation of carbamate 1a (0.2 mmol, Am=CON(iPr)₂) with LiTMP (0.6 mmol, 1 M in 2-MeTHF) at 25 °C, under air. The curves represent unweighted least-squares fits. 2-MeTHF (0.125 M) was used as solvent. k₁ = 1.1 ± 0.1 s⁻¹; k₂ = (1.9 ± 0.5) × 10⁻³ s⁻¹. Inset: expansion of the reaction profile after 50 s.

Consequently, the different kinetics of these two steps could be exploited to perform a rapid quench of the reaction mixture before the oxidation occurs. The rapid protonolysis of the anion solution (e.g. by performing an external acidic quench) represents the most intuitive and immediate route to avoid the oxidation of the enolized arylacetamide; however this approach could lack reproducibility due to the short reaction times involved. An alternative and more intriguing possibility is offered by the use of protic Deep Eutectic Solvents (DESs), mixtures of specific hydrogen bond donors (HBDs) and acceptors (HBAs) combined in eutectic molar ratios with peculiar physical properties, high biodegradability and biocompatibility.³⁷ Several studies have disclosed the ability of protic DESs to promote the kinetic activation of highly polar s-block organometallic reagents and, consequently, to improve both the reaction time and the chemoselectivity of several organometallic-mediated transformations without the need for external additives.³⁸ Overall, the activating effect of DESs facilitates the reactivity of air-sensitive RLi and RMgX species in a protic and aerobic environment over the competitive protonolysis process, which usually becomes predominant after short reaction times.³⁹ Hence, we envisaged that a combination of the intrinsic properties of protic DESs (acidity and activating effect on organolithiums) with the fast kinetics of the anionic Fries rearrangement step at room temperature could be efficiently exploited to perform in situ a rapid quench of the organolithium species in solution generated upon the metalation/carbamoyl migration process, namely the excess of LiTMP and the amide enolates 2-Li. In principle, this would suppress the formation of the corresponding α-hydroxylated compounds 3, allowing the preparation of arylacetamides 2 in a chemoselective fashion working under bench-type aerobic conditions (Scheme 5A). A series of carbamates 1 (0.2 mmol, 1 M in 2-MeTHF), differently substituted at both the aromatic ring and the amide nitrogen, were thus treated with LiTMP (3 equiv.) using a choline chloride (ChCl)/glycerol (Gly) 1 : 2 mol mol⁻¹ deep eutectic mixture as the reaction medium (1 g), under air at RT and with vigorous stirring (Scheme 5B).⁴⁰

Scheme 5 (A) Programmed Deep Eutectic Solvent-driven strategy to suppress the oxidation step. (B) Scope of the aerobic homo-Fries rearrangement of O-aryl carbamates 1 in Deep Eutectic Solvent. Reaction conditions: 1 (0.2 mmol, 1 M in 2-MeTHF), ChCl/Gly 1 : 2 mol mol⁻¹ (1 g), LiTMP (1 M in 2-MeTHF, 0.6 mmol), 60 s, room temperature, under air. Reported yields refer to isolated products.

Pleasingly, under heterogeneous conditions the rearrangement products 2 were exclusively formed within only 60 s in good yields (42–90%), and the corresponding oxidation compounds 3 have never been observed even after a longer reaction time (0.5 h) owing to the efficient protonolysis of the organolithium species by the protic deep eutectic mixture. As expected, when carbamate 1a (0.2 mmol, 1 M in 2-MeTHF) was treated with LiTMP (0.6 mmol) under heterogeneous aerobic conditions, using the ChCl/Gly 1 : 2 mol mol⁻¹ deep eutectic mixture as solvent, the kinetic profile of the reaction lacked the formation of the oxidation product, and a fast conversion of 1a into the rearranged arylacetamide 2a was observed (k₁ = 3.0 ± 0.8 s⁻¹, see the SI).

Conclusions

In summary, we have developed a general and efficient organolithium-mediated protocol which exploits the presence of molecular oxygen, a commonly detrimental species in the chemistry of polar s-block organometallic reagents, to build poly-substituted scaffolds of remarkable synthetic value in a single synthetic operation. Our methodology enables the transformation of O-aryl carbamates into α-hydroxy arylacetamides by resorting to a one-pot, tandem sequence of C–C and C–O bond formation events occurring at the same carbon atom, working under air and at room temperature, using the biobased 2-MeTHF as solvent. Our strategy allows a significant increase in the structural complexity of a simple molecular skeleton with remarkable functional group tolerance while suppressing the formation of overoxidation products without the need for external additives. Furthermore, we have shown that the use of a protic heterogeneous solvent mixture represents an effective tool to tune the chemoselectivity of the proposed transformation. The fast internal protonolysis of the anion solution operated by the protic reaction medium results in the efficient suppression of the oxidation step, enabling the chemoselective preparation of arylacetamides in an aerobic environment as a result of an interrupted metalation/rearrangement sequence. Overall, our strategy allows for the generation of diverse molecular structures in a chemodivergent fashion by simply changing the nature of the reaction medium, thus enlarging the portfolio of organolithium-mediated transformations under non-conventional conditions. The development of other consecutive anionic migration strategies under bench-type aerobic conditions are under investigation and will be reported in due course.

Author contributions

Riccardo Gnavi (methodology, investigation, visualization, writing review and editing); Federica De Nardi and Carolina Meazzo (methodology, investigation, writing review and editing); Simone Ghinato (methodology, investigation); Ettore Grimaldi (investigation); Andrea Maranzana (investigation, software, formal analysis, writing review and editing); Cristina Prandi (supervision, funding acquisition, writing review and editing); and Marco Blangetti (conceptualization, funding acquisition, project administration, supervision, visualization, formal analysis, writing – original draft, writing – review and editing). All authors have given approval to the final version of the manuscript.

Conflicts of interest

There are no conflicts to declare.

Data availability

CCDC deposition number 2475367 contains the supplementary crystallographic data for this paper.²⁵

The data underlying this study are available in the published article and its supplementary information (SI). Supplementary information: general procedures, experimental details, characterization data for both new and known compounds, copies of ¹H, ¹³C, ¹⁹F and ²H NMR spectra, and computational details. See DOI: https://doi.org/10.1039/d5sc09227b.

Acknowledgements

We warmly thank Dr Rosangela Santalucia and Dr Emanuele Priola for HRMS and X-ray analyses, respectively. This publication was produced while Mr Riccardo Gnavi was attending the PhD programme in Sustainable Development And Climate Change at the University School for Advanced Studies IUSS Pavia, Cycle XXXIX, with the support of a scholarship financed by the Ministerial Decree no. 118 of 2nd March 2023, based on the NRRP – funded by the European Union-Next Generation EU – Mission 4 “Education and Research”, Component 1 “Enhancement of the offer of educational services: from nurseries to universities” – Investment 3.4 “Advanced teaching and university skills”. This work was also carried out under the framework of the National PRIN 2022 project “Unlocking Greener Metal-assisted Synthetic Tactics by Sustainable Solvents and Technologies” (SUSMET) (Project no. 20228W9TBL) funded by “Unione Europea–Next Generation EU, Missione 4 Componente 2 (CUP: D53D23010260006)”. The computational study was funded by “Unione Europea–Next Generation EU, Missione 4 Componente 1 CUP D13C22001340001 – CN00000013”, Spoke 7- Materials and Molecular Sciences. This research was also supported by EU funding within the MUR PNRR Extended Partnership initiative on Emerging Infectious Diseases (Project no. PE00000007, INF-ACT). The authors acknowledge support from the Project CH4.0 under the MUR program “Dipartimenti di Eccellenza 2023–2027” (CUP: D13C22003520001).

Notes and references

1. (a) F. Depeint, W. R. Bruce, N. Shangari, R. Mehta and P. J. O'Brien, Chem. Biol. Interact., 2006, 163, 94–112; (b) M. Sani, D. Belotti, R. Giavazzi, W. Panzeri, A. Volonterio and M. Zanda, Tetrahedron Lett., 2004, 45, 1611–1615; (c) M. Bassetto, S. Ferla, F. Pertusati, S. Kandil, A. D. Westwell, A. Brancale and C. McGuigan, Eur. J. Med. Chem., 2016, 118, 230–243; (d) S.-L. Zhang, W. Zhang, Z. Yang, X. Hu and K. Y. Tam, Eur. J. Pharm. Sci., 2017, 110, 87–92; (e) M. R. Wood, K. M. Schirripa, J. J. Kim, S. D. Kuduk, R. K. Chang, C. N. Di Marco, R. M. DiPardo, B.-L. Wan, K. L. Murphy, R. W. Ransom, R. S. L. Chang, M. A. Holahan, J. J. Cook, W. Lemaire, S. D. Mosser, R. A. Bednar, C. Tang, T. Prueksaritanont, A. A. Wallace, Q. Mei, J. Yu, D. L. Bohn, F. C. Clayton, E. D. Adarayn, G. R. Sitko, Y. M. Leonard, R. M. Freidinger, D. J. Pettibone and M. G. Bock, Bioorg. Med. Chem. Lett., 2008, 18, 716–720.
2. (a) T. R. Herrin, J. M. Pauvlik, E. V. Schuber and A. O. Geiszler, J. Med. Chem., 1975, 18, 1216–1223; (b) Y. Xie, X.-H. Ruan, H.-Y. Gong, Y.-H. Wang, X.-B. Wang, J.-P. Zhang, Q. Li and W. Xue, J. Heterocycl. Chem., 2017, 54, 2644–2649; (c) J. P. Ley and H.-J. Bertram, Tetrahedron, 2001, 57, 1277–1282; (d) H. A. Schenck, P. W. Lenkowski, I. Choudhury-Mukherjee, S.-H. Ko, J. P. Stables, M. K. Patel and M. L. Brown, Biorg. Med. Chem., 2004, 12, 979–993; (e) G. V. Kaiser, M. Gorman and J. A. Webber, J. Infect. Dis., 1978, 137, S10–S16; (f) J. Verhaegen and L. Verbist, J. Antimicrob. Chemother., 1998, 41, 381–385; (g) H. J. Meulenbeld, R. H. J. Mathijssen, J. Verweij, R. de Wit and M. J. A. de Jonge, Expert Opin. Investig. Drugs, 2012, 21, 383–393.
3. (a) A. R. Katritzky, H.-Y. He and K. Suzuki, J. Org. Chem., 2000, 65, 8210–8213; (b) S. E. Kelly and T. G. LaCour, Synth. Commun., 1992, 22, 859–869.
4. (a) J.-J. Zhang, J.-H. Wang, Y.-H. Sang, H.-J. Qin, C. Chen and W. Liu, Org. Biomol. Chem., 2025, 23, 4922–4926; (b) G. N. Rao and G. Sekar, J. Org. Chem., 2023, 88, 4008–4016; (c) K. Soai and H. Hasegawa, J. Chem. Soc., Perkin Trans., 1985, 1, 769–772; (d) N. C. Mamillapalli and G. Sekar, Chem. Commun., 2014, 50, 7881–7884; (e) S. Golla and H. P. Kokatla, J. Org. Chem., 2022, 87, 9915–9925; (f) N. C. Mamillapalli and G. Sekar, RSC Adv., 2014, 4, 61077–61085.
5. (a) M. A. Ganiek, M. R. Becker, G. Berionni, H. Zipse and P. Knochel, Chem. Eur. J., 2017, 23, 10280–10284; (b) A. Nagaki, Y. Takahashi and J.-i. Yoshida, Angew. Chem., Int. Ed., 2016, 55, 5327–5331; (c) R. F. Cunico, Tetrahedron Lett., 2002, 43, 355–358; (d) Y. Yao, W. Tong and J. Chen, Mendeleev Commun., 2014, 24, 176–177; (e) U. Schöllkopf and H. Beckhaus, Angew. Chem., Int. Ed., 1976, 15, 293.
6. (a) T. Yamada, T. Hirose, S. Ōmura and T. Sunazuka, Eur. J. Org Chem., 2015, 2015, 296–301; (b) L. A. Martinho, T. P. F. Rosalba and C. K. Z. Andrade, Eur. J. Org Chem., 2022, 2022, e202201199; (c) M. Serafini, A. Griglio, E. Oberto, T. Pirali and G. C. Tron, Tetrahedron Lett., 2017, 58, 4786–4789; (d) R. Frey, S. G. Galbraith, S. Guelfi, C. Lamberth and M. Zeller, Synlett, 2003, 2003, 1536–1538; (e) N. Yu, J.-F. Lv, S.-M. He, K.-C. He, W.-H. Zheng, Y.-Q. Zhou, K. Jiang, X.-C. Pan and Y. Wei, Org. Lett., 2024, 26, 1358–1363.
7. A. Liu, S. Sun, Q. Xie, R. Huang, T. Kong, C. Ni and J. Hu, Org. Chem. Front., 2023, 10, 5343–5351.
8. (a) K. Tomooka, H. Shimizu, T. Inoue, H. Shibata and T. Nakai, Chem. Lett., 1999, 28, 759–760; (b) P. Zhang and R. E. Gawley, J. Org. Chem., 1993, 58, 3223–3224; (c) A. K. Ghosh, A. J. Basu, C.-S. Hsu and M. Yadav, Chem.–Eur. J., 2022, 28, e202200941.
9. (a) A. de la Torre, D. Kaiser and N. Maulide, J. Am. Chem. Soc., 2017, 139, 6578–6581; (b) X. Li, F. Lin, K. Huang, J. Wei, X. Li, X. Wang, X. Geng and N. Jiao, Angew. Chem., Int. Ed., 2017, 56, 12307–12311.
10. (a) R. V. Hoffman, N. K. Nayyar and W. Chen, J. Org. Chem., 1992, 57, 5700–5707; (b) A. J. Clark, Y. S. S. Al-Faiyz, M. J. Broadhurst, D. Patel and J. L. Peacock, J. Chem. Soc., Perkin Trans., 2000, 1, 1117–1127.
11. C. Tang, X. Qiu, Z. Cheng and N. Jiao, Chem. Soc. Rev., 2021, 50, 8067–8101.
12. T. Inukai, T. Kano and K. Maruoka, Org. Lett., 2021, 23, 792–796.
13. R. D. Shinde, A. R. Paraskar, J. Kumar, E. Ghosh, T. K. Paine and S. Bhadra, J. Org. Chem., 2024, 89, 9666–9671.
14. (a) G. Y. Zhang Xinming, T. Guide and Z. Wu, Chin. J. Org. Chem., 2017, 37, 2993–2999; (b) X. Zhang, Y. Yu, W. Li, L. Shi and H. Li, J. Org. Chem., 2022, 87, 16263–16275.
15. (a) R. Stuhr, P. Bayer and A. J. von Wangelin, ChemSusChem, 2022, 15, e202201323; (b) C. Xu, X. Li and L. Bai, J. Org. Chem., 2022, 87, 4298–4304; (c) B. Song, S. Wang, C. Sun, H. Deng and B. Xu, Tetrahedron Lett., 2007, 48, 8982–8986.
16. (a) Y. Minko and I. Marek, Org. Biomol. Chem., 2014, 12, 1535–1546; (b) M. Julia, V. Pfeuty-Saint Jalmes, K. Plé, J.-N. Verpeaux and G. Hollingworth, Bull. Soc. Chim. Fr., 1996, 133, 15–24.
17. J. L. Wardell, in Comprehensive Organometallic Chemistry, ed. G. Wilkinson, F. G. A. Stone and E. W. Abel, Pergamon, Oxford, 1982, vol. 1, pp. 46–47.
18. (a) K. A. Parker and K. A. Koziski, J. Org. Chem., 1987, 52, 674–676; (b) M. Julia, V. P. Saint-Jalmes and J.-N. Verpeaux, Synlett, 1993, 1993, 233–234; (c) G. Boche and J. C. Lohrenz, Chem. Rev., 2001, 101, 697–756.
19. (a) F. A. Davis and B. C. Chen, Chem. Rev., 1992, 92, 919–934; (b) B.-C. Chen, P. Zhou, F. A. Davis and E. Ciganek, Org. React., 2004, 1–356.
20. (a) M. Ohtawa, M. J. Krambis, R. Cerne, J. M. Schkeryantz, J. M. Witkin and R. A. Shenvi, J. Am. Chem. Soc., 2017, 139, 9637–9644; (b) M. Y. Kim, J. E. Starrett Jr. and S. M. Weinreb, J. Org. Chem., 1981, 46, 5383–5389; (c) T. L. LaPorte, M. Hamedi, J. S. DePue, L. Shen, D. Watson and D. Hsieh, Org. Process Res. Dev., 2008, 12, 956–966; (d) S. Hanessian and G. McNaughton-Smith, Bioorg. Med. Chem. Lett., 1996, 6, 1567–1572; (e) H. H. Wasserman and B. H. Lipshutz, Tetrahedron Lett., 1975, 16, 1731–1734; (f) M. Velasco, U. Hernández, J. L. Terán, D. Gnecco, M. L. Orea, D. M. Aparicio, V. Gómez-Calvario, S. Bernès and J. R. Juárez, Tetrahedron Lett., 2019, 60, 820–824; (g) J. C. Estévez, R. J. Estévez and L. Castedo, Tetrahedron, 1995, 51, 10801–10810; (h) H. Lubin, A. Tessier, G. Chaume, J. Pytkowicz and T. Brigaud, Org. Lett., 2010, 12, 1496–1499.
21. M. J. Rodríguez-Álvarez, N. Ríos-Lombardía, S. E. García-Garrido, C. Concellón, V. del Amo, V. Capriati and J. García-Álvarez, Molecules, 2024, 29.
22. (a) F. Marra, F. Morelli, F. De Nardi, E. Priola, I. Coldham and M. Blangetti, Org. Biomol. Chem., 2025, 23, 8029–8038; (b) F. Marra, F. De Nardi, F. Rossi, E. Priola, C. Prandi and M. Blangetti, Eur. J. Org Chem., 2024, 27, e202400313; (c) S. Ghinato, C. Meazzo, F. De Nardi, A. Maranzana, M. Blangetti and C. Prandi, Org. Lett., 2023, 25, 3904–3909; (d) S. Ghinato, F. De Nardi, P. Bolzoni, A. Antenucci, M. Blangetti and C. Prandi, Chem. Eur. J., 2022, 28, e202201154; (e) S. Ghinato, D. Territo, A. Maranzana, V. Capriati, M. Blangetti and C. Prandi, Chem. Eur. J., 2021, 27, 2868–2874.
23. M. Korb and H. Lang, Chem. Soc. Rev., 2019, 48, 2829–2882.
24. P. M. P. Ferreira, K. D. C. Machado, S. N. Lavorato, F. C. E. Oliveira, J. D. N. Silva, A. A. C. D. Almeida, L. D. S. Santos, V. R. Silva, D. P. Bezerra, M. B. P. Soares, C. Pessoa, M. O. D. Moraes Filho, J. R. D. O. Ferreira, J. M. Sousa, V. G. Maltarollo and R. J. Alves, Toxicol. Appl. Pharmacol., 2019, 380, 114692.
25. CCDC 2475367 (compound 3a) Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2p2tk5.
26. F. F. Mulks, L. J. Bole, L. Davin, A. Hernán-Gómez, A. Kennedy, J. García-Álvarez and E. Hevia, Angew. Chem., Int. Ed., 2020, 59, 19021–19026.
27. D. Tilly, J. Magolan and J. Mortier, Chem.–Eur. J., 2012, 18, 3804–3820.
28. J. Clayden, C. S. Frampton, C. McCarthy and N. Westlund, Tetrahedron, 1999, 55, 14161–14184.
29. (a) A. V. Kalinin, M. A. J. Miah, S. Chattopadhyay, M. Tsukazaki, M. Wicki, T. Nguen, A. L. Coelho, M. Kerr and V. Snieckus, Synlett, 1997, 1997, 839–841; (b) J. Fässler, J. A. McCubbin, A. Roglans, T. Kimachi, J. W. Hollett, R. W. Kunz, M. Tinkl, Y. Zhang, R. Wang, M. Campbell and V. Snieckus, J. Org. Chem., 2015, 80, 3368–3386.
30. The hydroxylated compound 3a was not observed under these conditions. When the metalation was performed under Ar for 5 min, and the reaction was then stirred under air for 30 min, compound 3a was produced in 70% yield after 30 min.
31. G. Sosnovsky and J. H. Brown, Chem. Rev., 1966, 66, 529–566.
32. Calculated C_C-Li–C_C=O bond length: 1.39 Å.
33. G. Boche, K. Möbus, K. Harms, J. C. W. Lohrenz and M. Marsch, Chem. Eur. J., 1996, 2, 604–607.
34. Treatment of 2a-Li species with 1 eq. of t-BuOOLi (0.5 M in 2-MeTHF) afforded compound 3a in 43% yield after workup. See SI for details.
35. S. R. Wang, S. An, Y.-X. Xin, Y.-Y. Jiang and W. H. Liu, Estimated ΔG of enolization for compound 2b = –13.1 kcal mol⁻¹ at 298 K (see SI for details). Data are in good accordance with previously reported results, JACS Au, 2024, 4, 4435–4444.
36. Estimated by nonlinear least-squares fit of the experimental kinetic data using the integrated rate laws for a consecutive transformation.
37. B. B. Hansen, S. Spittle, B. Chen, D. Poe, Y. Zhang, J. M. Klein, A. Horton, L. Adhikari, T. Zelovich, B. W. Doherty, B. Gurkan, E. J. Maginn, A. Ragauskas, M. Dadmun, T. A. Zawodzinski, G. A. Baker, M. E. Tuckerman, R. F. Savinell and J. R. Sangoro, Chem. Rev., 2021, 121, 1232–1285.
38. (a) L. Cicco, A. Fombona-Pascual, A. Sánchez-Condado, G. A. Carriedo, F. M. Perna, V. Capriati, A. Presa Soto and J. García-Álvarez, ChemSusChem, 2020, 13, 4967–4973; (b) A. W. J. Platten, I. Manasi, M. Campana, K. J. Edler and E. Hevia, ChemSusChem, 2025, 18, e202402083; (c) C. Vidal, J. García-Álvarez, A. Hernán-Gómez, A. R. Kennedy and E. Hevia, Angew. Chem., Int. Ed., 2014, 53, 5969–5973; (d) G. Dilauro, M. Dell'Aera, P. Vitale, V. Capriati and F. M. Perna, Angew. Chem., Int. Ed., 2017, 56, 10200–10203; (e) A. Sánchez-Condado, G. A. Carriedo, A. Presa Soto, M. J. Rodríguez-Álvarez, J. García-Álvarez and E. Hevia, ChemSusChem, 2019, 12, 3134–3143; (f) G. Dilauro, C. Luccarelli, A. F. Quivelli, P. Vitale, F. M. Perna and V. Capriati, Angew. Chem., Int. Ed., 2023, 62, e202304720; (g) S. Ghinato, G. Dilauro, F. M. Perna, V. Capriati, M. Blangetti and C. Prandi, Chem. Commun., 2019, 55, 7741–7744; (h) D. Arnodo, S. Ghinato, S. Nejrotti, M. Blangetti and C. Prandi, Chem. Commun., 2020, 56, 2391–2394.
39. Half-life times of aryllithiums (6.26 s) and benzyllithiums (6.57 s) in the choline chloride (ChCl)/glycerol (Gly) 1 : 2 mol mol⁻¹ deep eutectic mixture have been previously estimated. See ref. 38g and 38h.
40. In the absence of 2-MeTHF as an additive no lithiation occurred and the unreacted starting materials 1 were quantitatively recovered after workup.

---