Synthesis of α-Chloroarylacetic Acid via Electrochemical Carboxylation of α,α-Dichloroarylmethane Derivatives

The electrocarboxylation of α,α-dichloroarylmethane derivatives in the presence of CO2 was achieved, providing several α-chloroarylacetic acid derivatives with modest yields but high selectivity (chlorinated vs. non-chlorinated or dicarboxylic acid products). The obtained products were then involved in several chemical transformations, underlining their potential as versatile intermediates in synthetic chemistry. A mechanism was also proposed based upon a control experiment and cyclic voltammetry (CV) study.

dichloroarylmethane 4 derivatives in order to provide an efficient and selective synthesis of α-chloroarylacetic acid derivatives 1 (Scheme 1c).Scheme 1.General context of the project.

Influence of Reaction Parameters
In order to investigate this transformation, we first turned our attention to the simple commercially available α,α-dichlorodiphenylmethane 4b as a substrate in order to study the main reaction parameters (For a complete study of the reaction parameters, see Supporting Information).For the purpose of clarity, we have used the selectivity value which consists of the percentage obtained by dividing the NMR yield of the desired α-chlorophenylacetic acid methyl ester 6b (obtained after an esterification step with trimethylsilyl diazomethane ) by the sum of the NMR yields of 6b and the over-reduced product 7b.Starting from the aluminum anode as reported by Silvestri et al. [13] but in acetonitrile as a solvent instead of DMF, we have obtained a 66/34 mixture of α-chlorodiphenylacetic methyl ester 6b and the over-reduced diphenylacetic methyl ester 7b (s = 66%) in low (23% and 12%) NMR yields, respectively (Table 1, entry 1).As we were aware of the crucial role of the consumable anode in this type of reaction [21][22][23][24], we then examined the nature of this metallic electrode (Table 1, entries 2-4).Other metals such Co, Zn or Mg provided lower NMR yields, along with a significant amount of diphenylacetic methyl ester 7b.Whereas in THF no reaction occurred (Table 1, entry 5), other aprotic polar solvents such as DMF and DMA (Table 1, entries 6 and 7) allow the reaction to proceed in 17% and 42% NMR yields, respectively, of 6b, along with non-negligible amounts of 7b (11% and 20% NMR yields, respectively).DMA was thus chosen as the solvent to study the influence of the cathode on the reaction performances (Table 1, entries 8-11).Among all cathodes tested, only stainless steel (SS) improved the NMR yield of 6b to 54%, while reducing the NMR yield of 7b to only 5% (s = 91%).As long as the charge is concerned, a 1.8 F/mol charge resulted in the complete selectivity of 6b (s > 99%, only traces of 7b), but at the Scheme 1.General context of the project.

Influence of Reaction Parameters
In order to investigate this transformation, we first turned our attention to the simple commercially available α,α-dichlorodiphenylmethane 4b as a substrate in order to study the main reaction parameters (For a complete study of the reaction parameters, see Supporting Information).For the purpose of clarity, we have used the selectivity value which consists of the percentage obtained by dividing the NMR yield of the desired α-chlorophenylacetic acid methyl ester 6b (obtained after an esterification step with trimethylsilyl diazomethane) by the sum of the NMR yields of 6b and the over-reduced product 7b.Starting from the aluminum anode as reported by Silvestri et al. [13] but in acetonitrile as a solvent instead of DMF, we have obtained a 66/34 mixture of α-chlorodiphenylacetic methyl ester 6b and the over-reduced diphenylacetic methyl ester 7b (s = 66%) in low (23% and 12%) NMR yields, respectively (Table 1, entry 1).As we were aware of the crucial role of the consumable anode in this type of reaction [21][22][23][24], we then examined the nature of this metallic electrode (Table 1, entries 2-4).Other metals such Co, Zn or Mg provided lower NMR yields, along with a significant amount of diphenylacetic methyl ester 7b.Whereas in THF no reaction occurred (Table 1, entry 5), other aprotic polar solvents such as DMF and DMA (Table 1, entries 6 and 7) allow the reaction to proceed in 17% and 42% NMR yields, respectively, of 6b, along with non-negligible amounts of 7b (11% and 20% NMR yields, respectively).DMA was thus chosen as the solvent to study the influence of the cathode on the reaction performances (Table 1, entries 8-11).Among all cathodes tested, only stainless steel (SS) improved the NMR yield of 6b to 54%, while reducing the NMR yield of 7b to only 5% (s = 91%).As long as the charge is concerned, a 1.8 F/mol charge resulted in the complete selectivity of 6b (s > 99%, only traces of 7b), but at the expense of a lower, 33%, NMR yield (Table 1, entry 12), whereas a 3.2 F/mol charge provided a higher amount of 7b (11% NMR yield) without improving the NMR yield of 6b (54%, Table 1, entry 13).Regarding the intensity of the electrolysis, higher or lower intensities (10 or 30 mA instead of 20 mA) did not improve the NMR yield of 6b (39% and 50%, respectively, Table 1, entries 14-15).Finally, the reduction in the concentration of 4b from 83 mM to 35 mM had a positive impact on both the NMR yield of 6b (55%) and the selectivity, delivering only a 5% NMR yield of 7b (s > 92%).Under these conditions, a Faraday efficiency of 36.2% was calculated (see Supplementary Materials for calculations detail).In conclusion, after a careful study of the reaction parameters, we have found that the nature of the cathode, the solvent and the concentration has a major influence on the reaction, improving the results obtained by Silvestri [13] for compound 1a in terms of the yield and selectivity.).Regarding the intensity of the electrolysis, higher or lower intensities (10 or 30 mA instead of 20 mA) did not improve the NMR yield of 6b (39% and 50%, respectively, Table 1, entries 14-15).Finally, the reduction in the concentration of 4b from 83 mM to 35 mM had a positive impact on both the NMR yield of 6b (55%) and the selectivity, delivering only a 5% NMR yield of 7b (s > 92%).Under these conditions, a Faraday efficiency of 36.2% was calculated (see Supplementary Materials for calculations detail).
In conclusion, after a careful study of the reaction parameters, we have found that the nature of the cathode, the solvent and the concentration has a major influence on the reaction, improving the results obtained by Silvestri [13] for compound 1a in terms of the yield and selectivity. 1NMR yields were determined on 1 H NMR spectra of the crude product (after an esterification step with TMS diazomethane) in the presence of dimethyl terephthalate as an internal standard. 2Selectivity (s) determined on crude mixture by 1 H NMR using an internal standard, as follows: (NMR yield of 6b/(NMR yield of 6b + NMR yield 7b)) × 100. 3 Isolated yield.SS: stainless steel.nr: no reaction.Bold text indicates the parameter which has been modified.

Scope and Limitations
With these optimal conditions in hand, we then moved to the study of the scope and limitations of our methodology.Since only a few α,α-dichloroarylmethane derivatives 4 are commercially available, we put some efforts in their synthesis.By using conditions reported by Hegarty et al. starting from commercially available aromatic aldehydes 8 (PCl5, toluene, rt; see Section 3 below and Supplementary Materials for more details) [25], 1 NMR yields were determined on 1 H NMR spectra of the crude product (after an esterification step with TMS diazomethane) in the presence of dimethyl terephthalate as an internal standard. 2Selectivity (s) determined on crude mixture by 1 H NMR using an internal standard, as follows: (NMR yield of 6b/(NMR yield of 6b + NMR yield 7b)) × 100. 3 Isolated yield.SS: stainless steel.nr: no reaction.Bold text indicates the parameter which has been modified.

Scope and Limitations
With these optimal conditions in hand, we then moved to the study of the scope and limitations of our methodology.Since only a few α,α-dichloroarylmethane derivatives 4 are commercially available, we put some efforts in their synthesis.By using conditions reported by Hegarty et al. starting from commercially available aromatic aldehydes 8 (PCl 5 , toluene, rt; see Section 3 below and Supplementary Materials for more details) [25], we were able to obtain several α,α-dichloroarylmethane derivatives 4. Next, the scope and limitations of the electrochemical carboxylation reaction were studied (Scheme 2).
we were able to obtain several α,α-dichloroarylmethane derivatives 4. Next, the scope and limitations of the electrochemical carboxylation reaction were studied (Scheme 2).Generally speaking, even if a complete conversion was observed, modest yields of around 50% were obtained, but a high selectivity (Cl (1 or 6) vs. H (5 or 7) and/or dicarboxylated product (8 or 9)) were observed (s = 60-96%).Regarding the modulation of the Generally speaking, even if a complete conversion was observed, modest yields of around 50% were obtained, but a high selectivity (Cl (1 or 6) vs. H (5 or 7) and/or dicarboxylated product (8 or 9)) were observed (s = 60-96%).Regarding the modulation of the Ar part, electron-withdrawing or -donating groups can be introduced, with electrodonating groups providing higher isolated yields (1e, 37% vs. 1i, 55%) and selectivities (1e, s = 60% vs. 1i, s = 78%).Fluoride and bromine at the para position are also welltolerated (1d, 53% and 1g, 48%, respectively).The introduction of a methyl substituent at the para or ortho position afforded the desired product in acceptable yields (1c, 54% and 1j, 48%), whereas a methyl at the meta position resulted in a drop of the yield to 36% for compound 1h.As far as the methyl substituent is concerned, an ortho substitution gave a lower selectivity than a meta and para substitution (s = 63% for 1j vs. s = 90 and 86% for 1c and 1h, respectively).The phenyl group at the para position provided the desired product in a low yield and modest selectivity (1f, 32%, s = 67%).It is worthy of note that during the silica gel purification of 1l, the chlorine atom was substituted by a methoxy group due to the presence of methanol in the mobile phase, affording compound 10 in a 24% isolated yield.We have also tried to vary the R 1 substituent.Accordingly, a phenyl can be introduced, affording a decent 48% yield and excellent selectivity (s = 96%) for compound 6b.Starting from α,α,α-trichlorotoluene, methyl ester 6k bearing a gem-dichloro group was successfully obtained in a modest 23% yield but a good level of selectivity (s = 81%).However, some products remain problematic with otherwise improved methodology.In the case of thiophene derivatives, a complex mixture was obtained, whereas the formation of the product with an N-Ts indole moiety was not observed.Finally, compound 1m was formed but as an inseparable mixture with 5m (25% NMR yield, s = 45%).

Versatile Transformations of the α-Chlorophenylacetic Acid 1a
In order to demonstrate the synthetic utility of the α-chloroarylacetic acids, we have turned our attention to several chemical transformations (Scheme 3).

28, x FOR PEER REVIEW 5 of 12
Ar part, electron-withdrawing or -donating groups can be introduced, with electro-donating groups providing higher isolated yields (1e, 37% vs. 1i, 55%) and selectivities (1e, s = 60% vs. 1i, s = 78%).Fluoride and bromine at the para position are also well-tolerated (1d, 53% and 1g, 48%, respectively).The introduction of a methyl substituent at the para or ortho position afforded the desired product in acceptable yields (1c, 54% and 1j, 48%), whereas a methyl at the meta position resulted in a drop of the yield to 36% for compound 1h.As far as the methyl substituent is concerned, an ortho substitution gave a lower selectivity than a meta and para substitution (s = 63% for 1j vs. s = 90 and 86% for 1c and 1h, respectively).The phenyl group at the para position provided the desired product in a low yield and modest selectivity (1f, 32%, s = 67%).It is worthy of note that during the silica gel purification of 1l, the chlorine atom was substituted by a methoxy group due to the presence of methanol in the mobile phase, affording compound 10 in a 24% isolated yield.We have also tried to vary the R 1 substituent.Accordingly, a phenyl can be introduced, affording a decent 48% yield and excellent selectivity (s = 96%) for compound 6b.
Starting from α,α,α-trichlorotoluene, methyl ester 6k bearing a gem-dichloro group was successfully obtained in a modest 23% yield but a good level of selectivity (s = 81%).However, some products remain problematic with otherwise improved methodology.In the case of thiophene derivatives, a complex mixture was obtained, whereas the formation of the product with an N-Ts indole moiety was not observed.Finally, compound 1m was formed but as an inseparable mixture with 5m (25% NMR yield, s = 45%).

Versatile Transformations of the α-Chlorophenylacetic Acid 1a
In order to demonstrate the synthetic utility of the α-chloroarylacetic acids, we have turned our attention to several chemical transformations (Scheme 3).First, the acid moiety was submitted to an amidation reaction via the corresponding acid chloride in the presence of benzyl amide, affording the amide derivatives 11 in 59%.An esterification was also possible by using trimethylsilyldazomethane in a mixture of toluene and methanol to give the chlorinated ester 6a in almost quantitative yields.More interestingly, the chlorine atom of 6a can be submitted to nucleophilic substitution in the presence of sodium azide under liquid/liquid phase-transfer conditions, leading to 12 in an excellent yield.The product 12 is of particular interest as it paves the way to the access of non-proteogenic α-amino acid derivatives [26,27].

Mechanistic Considerations
Finally, control experiments and CV analyses were performed to gain insight into the mechanism (Scheme 4).We first checked whether the reaction proceeded exclusively under electrochemical conditions by performing the reaction without a current for 1.5 h.No product was detected by an NMR analysis of the crude mixture (Scheme 4a).Then, the First, the acid moiety was submitted to an amidation reaction via the corresponding acid chloride in the presence of benzyl amide, affording the amide derivatives 11 in 59%.An esterification was also possible by using trimethylsilyldazomethane in a mixture of toluene and methanol to give the chlorinated ester 6a in almost quantitative yields.More interestingly, the chlorine atom of 6a can be submitted to nucleophilic substitution in the presence of sodium azide under liquid/liquid phase-transfer conditions, leading to 12 in an excellent yield.The product 12 is of particular interest as it paves the way to the access of non-proteogenic α-amino acid derivatives [26,27].

Mechanistic Considerations
Finally, control experiments and CV analyses were performed to gain insight into the mechanism (Scheme 4).We first checked whether the reaction proceeded exclusively under electrochemical conditions by performing the reaction without a current for 1.5 h.No product was detected by an NMR analysis of the crude mixture (Scheme 4a).Then, the reaction was performed at 20 mA for 36 min (1.8 F/mol) and, then, the mixture was stirred for one additional hour without a current.The result was almost identical to that obtained when the reaction was stopped after the 1.8 F/mol electrolysis (see also Table 1,entry 12), indicating that the presence of Al 3+ presumably formed at the anode during electrolysis is not involved in the reaction.reaction was performed at 20 mA for 36 min (1.8 F/mol) and, then, the mixture was stirred for one additional hour without a current.The result was almost identical to that obtained when the reaction was stopped after the 1.8 F/mol electrolysis (see also Table 1, entry 12), indicating that the presence of Al 3+ presumably formed at the anode during electrolysis is not involved in the reaction.According to previous reports [13,14] and our own study (see Supplementary Materials for CV experiments) ((a) The reduction potential of PhCHCl2 in DMA (Ered = −2.2V/SCE) was found to be similar to previously reported value in DMF, see ref [28] and supporting information.(b) Our attempts to determine the reduction potential of CO2 in DMA failed (see supporting information).(c) Our attempts to clarify the exact role of the Al 3+ by CV have not been conclusive (see Supporting Information)), we have proposed a plausible mechanism (Scheme 4b).First, the α,α-dichloroarylmethane derivatives PhRCCl2 4 would be reduced at the cathode (Ered = −2.2V/SCE for R = H and Ered = −1.8V/SCE for R = Ph in DMF) [28] thanks to a two electron process to provide the corresponding α-chloro anion A which then would add to CO2 (Ered = −2.2V/SCE in DMF) [29] to give the α-chlorophenylacetic carboxylate B flanked by an aluminum counter-ion.Simultaneously, the aluminum anode plate would be oxidized to afford the corresponding aluminum (III) cation in solution.This aluminum counter-ion is expected to stabilize the carboxylate anion generated during the electrolysis which is generally unstable under the electrochemical conditions [7] and also to avoid its possible transformation into mandelic acid [13].Finally, the expected product 1 would be obtained after aqueous work-up.The by-product 5 could be obtained by the further reduction in B at the cathode followed by aqueous work-up.Nevertheless, due the close proximity between the reduction potential of CO2 and that of some of the α,α-dichloroarylmethane derivatives 4, a mechanism first involving the reduction in CO2 to form the corresponding radical anion followed by the radical coupling with 4 • (resulting from the one electron reduction in 4 at the cathode) cannot totally be ruled out (Scheme 4c).According to previous reports [13,14] and our own study (see Supplementary Materials for CV experiments) ((a) The reduction potential of PhCHCl 2 in DMA (E red = −2.2V/SCE) was found to be similar to previously reported value in DMF, see ref [28] and supporting information.(b) Our attempts to determine the reduction potential of CO 2 in DMA failed (see supporting information).(c) Our attempts to clarify the exact role of the Al 3+ by CV have not been conclusive (see Supporting Information)), we have proposed a plausible mechanism (Scheme 4b).First, the α,α-dichloroarylmethane derivatives PhRCCl 2 4 would be reduced at the cathode (E red = −2.2V/SCE for R = H and E red = −1.8V/SCE for R = Ph in DMF) [28] thanks to a two electron process to provide the corresponding α-chloro anion A which then would add to CO 2 (E red = −2.2V/SCE in DMF) [29] to give the αchlorophenylacetic carboxylate B flanked by an aluminum counter-ion.Simultaneously, the aluminum anode plate would be oxidized to afford the corresponding aluminum (III) cation in solution.This aluminum counter-ion is expected to stabilize the carboxylate anion generated during the electrolysis which is generally unstable under the electrochemical conditions [7] and also to avoid its possible transformation into mandelic acid [13].Finally, the expected product 1 would be obtained after aqueous work-up.The by-product 5 could be obtained by the further reduction in B at the cathode followed by aqueous work-up.Nevertheless, due the close proximity between the reduction potential of CO 2 and that of some of the α,α-dichloroarylmethane derivatives 4, a mechanism first involving the reduction in CO 2 to form the corresponding radical anion followed by the radical coupling with 4 • (resulting from the one electron reduction in 4 at the cathode) cannot totally be ruled out (Scheme 4c).

General Information
Reactions were performed using oven dried glassware under an inert atmosphere of nitrogen.Unless otherwise noted, all reagent-grade chemicals and solvents were obtained from commercial suppliers and were used as received.Toluene was dried over an MBRAUN MB SPS-800 apparatus (MBRAUN inertgas-system gmbh, Garching, Germany).Reactions were monitored by thin-layer chromatography with silica gel 60 F254 pre-coated aluminum plates (0.25 mm).Visualization was performed under UV light, phosphomolybdic acid or KMnO 4 oxidation.Chromatographic purification of compounds was achieved with 60 silica gel (40-63 µm).Melting points were measured on a WME Köfler hot-stage (Stuart SMP3) and are uncorrected (Barloworld Scientific France SAS, Nemours, France).Infrared spectra (IR) were recorded on a PerkinElmer Spectrum 100 Series FT-IR spectrometer.Liquids and solids were applied on the Single Reflection Attenuated Total Reflectance (ATR) Accessories (PerkinElmer, Wellesley, MA, USA).Data are reported in cm −1 . 1 H Spectra (300 MHz) and 13 C NMR spectra (75 MHz) were recorded on a Bruker Advance 300 (Bruker, Billerica, MA, USA).Processing and analysis of the spectra were performed with the Topspin 3.6 software (Bruker, Billerica, MA, USA) on a PC workstation.Data appear in the following order: chemical shifts in ppm which were referenced to the internal solvent signal, number of protons, multiplicity (s, singlet; d, doublet; t, triplet; q, quadruplet; dd, doublet of doublet, ddd, doublet of doublet of doublet, dt, doublet of triplet; ddt, doublet of doublet of triplet, td, triplet of doublet; tdd, triplet of doublet of doublet; m, multiplet, ABq, AB system) and coupling constant J in Hertz.Accurate mass measurements (HRMS) were performed by the Mass Spectrometry Laboratory of the University of Rouen and were recorded with a Waters LCP 1er XR spectrometer (Waters, Milford, CT, USA).The electrosynthesis were carried out by means of IKAElectraSynth ® 2.0 apparatus.Electrodes were all purchased from IKA®(IKA, Staufen, Germany).Cyclic Voltammetry (CV) measurements were carried out with an OrigaFlex®potentiostat/galvanostat by means of three electrodes (Origalys, Rilleux-la-Pape, France).

General Procedure for the Synthesis of α,α-Dichloro Benzyl Derivatives 4 [25]
To a dry 10 mL flask containing PCl 5 (1.3 equiv), dry toluene (5 mL) was added the corresponding aldehyde (1 equiv) and the mixture was then stirred for 16 h.After completion, the reaction was diluted with 20 mL EtOAc and washed two times with 20 mL saturated NaHCO 3 aqueous solution, then once with 10 mL brine.The organic layer was dried over MgSO 4 , filtered, and concentrated at 40 • C under 80 mbar pressure.The residue was purified by silica gel column chromatography (100/0 EP:EtOAc to 95/5) to obtain the corresponding α,α-dichloro aryl compounds 4. As mentioned by the authors and due their high reactivity, products 4 were generally used rapidly after their synthesis; 4a, 4j and 4k were commercially available and were used as received.

General Procedure A for the Synthesis of α-Chloroarylacetic Acid Derivatives 1
In a dry electrochemical cell (10 mL vial from IKA) containing n-Bu 4 NBr (33 mg, 0.1 mmol) DMA (7 mL) was added and α,α-dichloro aryl derivates 4 (0.25 mmol) under nitrogen.The electrodes (aluminum electrode as anode and stainless steel as cathode) were installed, and carbon dioxide was bubbled through the solution for 5 min using a CO 2 balloon.The reaction was carried out at room temperature for 48 min, with a constant current of 20 mA and a charge of 3.2 F.mol −1 .After completion, the electrodes were rinsed subsequently with EtOAc, HCl 1M and then water.The reaction mixture was diluted with 15 mL of EtOAc and washed with 15 mL of HCl 0.5 M. The aqueous phase was washed a second time with 10 mL of EtOAc and then the combined organic extract was washed three times with 10 mL of HCl 0.5 M. The organic layer was dried over MgSO 4 , filtered and concentrated under reduced pressure.The NMR yield was measured by 1 H NMR of crude by means of dimethyl terephthalate as internal standard (0.25 equiv, 12.2 mg).The residue was purified by silica gel column chromatography (5/25/70 MeOH:CH 2 Cl 2 :PE, then 1/20/79 HCOOH:Et 2 O:PE) to obtain the corresponding α-chloroarylacetic acid 1.

mmol Procedure
In a dry electrochemical cell (20 mL vial from IKA) containing n-Bu 4 NBr (110 mg, 0.34 mmol) DMA was added (16 mL) and α,α-dichlorotoluene 4a (126 mL, 1.00 mmol) under nitrogen.The electrodes (aluminum electrode as anode and stainless steel as cathode) were installed, and carbon dioxide was bubbled through the solution for 5 min using a CO 2 balloon.The reaction was carried out at room temperature for 48 min, with a constant current of 20 mA and a charge of 3.2 F.mol −1 .After completion, the electrodes were rinsed subsequently with EtOAc, HCl 1M and then water.The reaction mixture was diluted with 30 mL of EtOAc and washed with 30 mL of HCl 0.5 M. The aqueous phase was washed a second time with 20 mL of EtOAc and then the combined organic extract was washed three times with 20 mL of HCl 0.5 M. The organic layer was dried over MgSO 4 , filtered and concentrated under reduced pressure.The residue was purified by silica gel column chromatography (5/25/70 MeOH:CH 2 Cl 2 :PE, then 1/20/79 HCOOH:Et 2 O:PE) to obtain the corresponding 2-chloro-2-phenylacetic acid 1a as a white solid (75 mg, 44%).

General Procedure B for the Synthesis of Methyl 2-Chloro-2-phenylacetate Derivatives 6b,k
In a dry electrochemical cell (10 mL vial from IKA) containing n-Bu 4 NBr (33 mg, 0.1 mmol), DMA was added (7 mL) and α,α-dichloro aryl derivates 4 (0.25 mmol) under nitrogen.The electrodes (aluminum electrode as anode and stainless steel as cathode) were installed, and carbon dioxide was bubbled through the solution for 5 min using a CO 2 balloon.The reaction was carried out at room temperature for 48 min, with a constant current of 20 mA and a charge of 3.2 F.mol −1 .After completion, the electrodes were rinsed subsequently with EtOAc, HCl 1M and then water.The reaction mixture was diluted with 15 mL of EtOAc and washed with 15 mL of HCl 0.5 M. The aqueous phase was washed a second time with 10 mL of EtOAc and then the combined organic extract was washed three times with 10 mL of HCl 0.5 M. The organic layer was dried over MgSO 4 , filtered and concentrated under reduced pressure.The crude mixture was then dissolved in toluene/methanol mixture (7:3 v/v) and cooled to 0 • C. TMS diazomethane was then added until a yellow coloration persisted in the solution.The solution was then stirred for 15 min at 0 • C and then 15 min at r.t.The solution was then quenched with acetic acid and extracted with ethyl acetate and washed with saturated aqueous NaHCO 3 solution, HCl 1M and then water.The NMR yield was measured by 1 H NMR of crude by means of dimethyl terephthalate as internal standard (0.25 equiv, 12.2 mg).The residue was purified by silica gel column chromatography (5/25/70 MeOH:CH 2 Cl 2 :PE, then 1/20/79 HCOOH:Et 2 O:PE) to obtain the corresponding methyl 2-chloro-2-phenylacetate derivatives 6.

Procedure for the Synthesis of Methyl 2-Azido-2-phenylacetate 12
The title product was synthesized according to a reported procedure [36], as follows: Methyl 2-chloro-2-phenylacetate 6a (194 mg, 1 mmol) was mixed with NaN 3 (130 mg, 2 mmol) and tetrabutylammonium hydrogensulfate (50 mg, 0.15 equiv) in a mixture of water (1 mL) and chloroform (1 mL).The reaction was allowed to stir at room temperature for 24 h in the dark, after which time the aqueous layer was removed and the organic layer was washed with water (3 × 10 mL) and dried over sodium sulfate.The solvent was removed carefully by rotary evaporation under reduced pressure without heating.The title compound 12 was obtained as a colorless oil (191 mg, 99%).1H NMR (300 MHz, CDCl3)

Scheme 4 .
Scheme 4. Mechanistic considerations: (a) control experiments; (b) plausible mechanism; (c) alternative pathway. 1 1 H NMR yield determined on the crude mixture by using dimethyl terephthalate as an internal standard.

Scheme 4 .
Scheme 4. Mechanistic considerations: (a) control experiments; (b) plausible mechanism; (c) alternative pathway. 1 1 H NMR yield determined on the crude mixture by using dimethyl terephthalate as an internal standard.

Table 1 .
Optimization of the reaction conditions.

Table 1 .
Optimization of the reaction conditions.