Nitrosobenzene: Reagent for the Mitsunobu Esterification Reaction

Abstract

Nitrosobenzene has been demonstrated to participate in the Mitsunobu reaction in an analogous manner to dialkyl azodicarboxylates. The protocol using nitrosobenzene and triphenylphosphine (1:1) under mild conditions (0 °C) provides the ester derivatives of aliphatic and aromatic acids using various alcohols in moderate yield and with good enantioselectivity, giving the desired products predominantly with an inversion of configuration. The proposed mechanism, which is analogous to that observed using dialkyl azodicarboxylates, involves a nitrosobenzene–triphenylphosphine adduct and an alkoxytriphenylphosphonium ion and was supported by density functional theory calculations, ³¹P NMR spectroscopy, and experiments conducted with isotopically labeled substrates.

Introduction

The Mitsunobu reaction (i.e., the reaction of a primary or secondary alcohol with a pronucleophile mediated by the combination of a trialkyl- or triarylphosphine (usually PPh₃) and dialkyl azodicarboxylate (DIAD or DEAD)) has been established as one of the most useful tools in organic synthesis, which allows the effective transformation of the hydroxy group (Scheme 1A).¹ Originally, the method was developed for the reaction with a carboxylic acid to provide the corresponding ester.² Subsequently, many other pronucleophiles such as phenols or imides were shown to react by an analogous procedure.^1b In addition to its mild reaction conditions, the stereoselectivity was found to be another important aspect of the Mitsunobu reaction that made it popular for the total synthesis of biologically active compounds and pharmaceuticals.^1a,3 If a chiral secondary alcohol was used as the substrate, the desired product usually formed with an inversion of configuration. This property is inter alia used for the inversion of the configuration of a stereocenter bearing an OH group, known as “Mitsunobu inversion” (i.e., a reaction sequence consisting of ester formation followed by its hydrolysis).⁴ The dialkyl azodicarboxylate reagent is involved in the formation of Morrison–Brunn–Huisgen⁵ betaine I (Scheme 1B), which is reactive enough to convert the hydroxy group in an alcohol into a good leaving group (triphenylphosphine oxide).⁶ In exceptional cases, the hydroxy group in the carboxylic acid is activated in the same way leading to an S_NAc reaction, which provides a product with retention of configuration.⁷

Scheme 1. Overall Scheme of Mitsunobu Esterification (A) and Modes of Triphenylphosphine Activation with Azo Compounds (B), in the Form of Phosphonium Ylide (C) and by Nitrosoarene (D).

One of the drawbacks of the Mitsunobu reaction that limits its use in large-scale applications is the manipulation of large quantities of a stoichiometric amount of the dialkyl azodicarboxylate reagent.⁸ These compounds are high-energy materials that potentially possess explosive properties. To minimize these problems, few catalytic systems have been developed for in situ azodicarboxylate regeneration usually providing esters with the inversion of configuration with good to high enantioselectivity.^9,10 In addition, azodicarboxylate-free esterification using triphenylphosphine activated via a one-electron oxidation process has been reported.^10e,11 However, these methods mainly lead to carboxylic acid activation and thus to products with retention of configuration. An alternative strategy toward improving the Mitsunobu protocol is to replace the dialkyl azodicarboxylate reagent with a suitable alternative.^1b Many reagents have been investigated for this reason, but most of them are based on an azo-function acting via the betaine I intermediate. To the best of our knowledge, with the exception of employing phosphonium ylide II (Scheme 1C)^12a,12b or analogous cyanomethylenetrialkylphosphorane^12c,12d prepared in advance, there are no other examples of triphenylphosphine or trialkylphosphine activation in the esterification reaction. Herein, we report that nitrosobenzene can act as the Mitsunobu reagent as an alternative to an azo compound to provide esters with similar stereoselectivity (Scheme 1D).

Results and Discussion

Investigating various esterification procedures with model substrates 1a and 2a, we observed a small amount of ester 3a, which was formed after 20 h in the presence of triphenylphosphine and nitrosobenzene (Table 1, entry 1). This result inspired us to explore the reaction in more detail. A preliminary study showed that the most reproducible results were obtained when the reaction was started by the addition of a concentrated solution of triphenylphosphine into a stirred tempered solution of alcohol 1a, acid 2a, and nitrosobenzene in acetonitrile. The esterification reaction was observed to be rapid; the composition of the reaction mixture did not change after a time period of 30 min, and after this time, all triphenylphosphine was found to be converted to triphenylphosphine oxide either by the productive (esterification) or nonproductive pathway (see mechanistic studies below). Thus, the reaction time was reduced to 30 min for the subsequent experiments. Using Job’s plot, we found that the reaction achieved the highest conversion of 1a to 3a using a Ph₃P/PhNO ratio of 1:1 (see entries 2–4 and Job’s plot found in the Supporting Information, Figure S2). The efficiency of the esterification reaction was found to increase upon diluting the reaction mixture by 1 order of magnitude (cf. entries 3 and 5). A further decrease in the reaction concentration caused a significant reduction in the amount of product 3a (entry 6; see the Supporting Information for the plot of conversion vs concentration, Figure S1). An excess of both reagents relative to alcohol 1a was found to be beneficial to the reaction efficiency (cf. entries 5, 7, and 8), but the effect was rather small when using >2.5 equiv. Finally, using the optimal reagent ratio and reaction concentration, we found that the reaction was better when conducted at 0 °C when compared to +25 or −25 °C (cf. entries 5, 9, and 10). Solvent alternations (dichloromethane or toluene) did not improve the reaction toward ester (see the Supporting Information, Table S2).

Table 1. Optimizing Conditions for the Mitsunobu Reaction with Nitrosobenzene^a.

entry	PhNO (x equiv)	Ph3P (y equiv)	c (1a) (mM)	T (°C)	conv.b (%)
1c	1	2	75	25	7
2	1	4	75	25	25
3	2.5	2.5	75	25	43
4	3	2	75	25	23
5	2.5	2.5	9.4	25	67
6	2.5	2.5	2.4	25	10
7	2	2	9.4	25	58
8	3	3	9.4	25	72
9	2.5	2.5	9.4	0	82
10	2.5	2.5	9.4	–25 °C	52

^aConditions: 1a (0.15 mmol), 2a (0.18 mmol), PhNO (x equiv relative to 1a), Ph₃P (y equiv relative to 1a) added in the form of solution in acetonitrile (V = 2–64 mL, to adjust the concentration); reaction time: 30 min; see the Supporting Information for details.

^bDetermined by ¹H NMR.

^cPh₃P added as a solid. Reaction time: 20 h.

Interestingly, we found that replacing nitrosobenzene with an electron-rich (1-methoxy-4-nitrosobenzene) or electron-poor aromatic (1-nitroso-3-(trifluoromethyl)benzene) or aliphatic (2-methyl-2-nitrosopropane) nitroso compound decreased the reaction efficiency. In addition, using a substituted or aliphatic analog of triphenylphosphine caused a decrease in the reaction conversion toward ester 3a (see the Supporting Information, Figure S3). It should be noted that the reaction conversion to the ester in the absence of either PhNO or Ph₃P (blank experiments, see the Supporting Information, Table S1) was <2%. The reaction in the absence of the acid and alcohol afforded Ph₃P=O almost quantitatively, confirming that an undesired nitrosobenzene deoxygenation reaction occurs under our optimized reaction conditions, too.¹³

With the optimized reaction conditions in hand, we explored the substrate scope of the Mitsunobu esterification reaction using nitrosobenzene (Table 2). Several types of esters were observed to be formed in moderate to good conversions and isolated in moderate yields. The substrate scope includes aromatic and aliphatic acids and benzyl and aliphatic alcohols. In some cases, esters were obtained in significantly lower yields relative to conversions (e.g., ester 3e) because of workup. The esterification reaction tolerates also the nitro group (in ester 3i), which was described to be deoxygenated by phosphines in some cases.¹⁴ The corresponding benzylether was formed in very small conversions, and a significant amount of ether was obtained only when the more acidic 4-nitrophenol was used instead of phenol as the pronucleophile (see Table 2, 3k). Phthalimide provided only a small amount (3%) of alkylated product (not isolated) by the reaction with benzyl alcohol.

Table 2. Substrate Scope for Mitsunobu Esterification Mediated by PhNO and Ph₃P^a.

^aConditions: 1a (0.60 mmol), 2a (0.72 mmol), PhNO (1.5 mmol), Ph₃P (1.5 mmol), acetonitrile (64 mL); reaction time: 30 min; 0 °C.

^bDetermined by ¹H NMR. Average of two experiments.

^cInversion/retention of configuration; determined by high-performance liquid chromatography.

The stereochemical outcome of the Mitsunobu reaction is essential. Thus, we explored our method using chiral, nonracemic substrates (Table 2). With (R)-ethyl lactate ((R)-1f), 3-trifluoromethylbenzoic acid (2c) gave ester 3n exclusively with an inversion of configuration. Similarly, when using 1-phenylethanol ((S)-1e or (R)-1e), esters 3l and 3m were formed upon reaction with both 2c and sterically hindered 2-methyl-2-phenylpropanoic acid (2f) with a predominant inversion of configuration. The enantioselectivity of 3l was close to that obtained using the conventional Mitsunobu reaction (Scheme 2).

Scheme 2. Comparison of Stereoselective Esterification under Classical Mitsunobu Reaction Conditions (with DIAD) and that with PhNO.

Conditions: 2c (0.60 mmol), 1e (0.72 mmol), PhNO (1.5 mmol), Ph₃P (1.5 mmol), acetonitrile (64 mL); 0 °C.

Conditions: 2c (0.60 mmol), 1e (0.72 mmol), DIAD (0.90 mmol), Ph₃P (0.90 mmol), acetonitrile (8 mL); 0 °C.

The observed inversion of configuration was in accordance with the proposed mechanism shown in Scheme 3A, which involves the activation of the alcohol by its transformation to alkoxyphosphonium species IV and an S_N2 substitution reaction (Scheme 3A). This mechanism was also supported by the esterification reaction observed using either ¹⁸O-labeled benzoic acid or benzyl alcohol. The ester containing ¹⁸O was only formed when starting from the ¹⁸O-labeled acid (see the Supporting Information, Table S9). Moreover, we observed the formation of alkoxyphosphonium species IV when the reaction was monitored using ³¹P NMR spectroscopy recorded at −50 °C (see the Supporting Information, Figure S12).

Scheme 3. Proposed Mechanism of Esterification Mediated by PhNO (A); Interconversion of Various Forms of the Me₃P Adduct with Substituted Nitrosobenzene Described in Ref (15) (in Purple) and Calculated by us (in Green) (B); Relative Energies of Structures from (B) at B3PW91/def2-SVPD in Vacuo (red), Single-Point Solvation (Blue), and Optimized in Solvent (Green) (C).

Gibbs free energies in kcal/mol relative to Me₃P + ArNO are shown.

Nitrene pathway is shaded in gray.

The Mitsunobu reaction with azodicarboxylates has been previously studied in detail using density functional theory (DFT) calculations.^5a,15 In addition, the mechanism of interaction of trimethylphosphine (Me₃P) with 2-nitrosostyrene has been investigated and proposed as follows: Me₃P attacks the nitrogen atom of nitrosobenzene forming betaine intermediate IIIa. The Me₃P can migrate to the oxygen atom (forming IIIc) via cyclic oxazaphosphiridine IIIb, the relative energies of structures being 4.7, 3.4, and −3.4 kcal/mol with respect to reactants¹⁶ (see Scheme 3B). Intermediate IIIc is then reported to undergo deoxygenation to nitrene.¹⁶

We followed the reaction path using exactly the same procedure and found nitrosobenzene without the vinyl group giving qualitatively different energetics of the reaction. All three states including the oxazaphosphiridine ring are of close-lying energies, between −9.1 and −11.8 kcal/mol (Scheme 3C, MeCN single point). The effect of solvent on the relative stability is enormous (Scheme 3C, MeCN optimization), as can be expected for the species with significantly different dipole moments. Optimization of geometry in implicit solvent destabilizes oxazaphosphiridine, leading to spontaneous opening to the P–O–N structure. Thus, rapid equilibration of all of the structures IIIa, IIIb, and IIIc is proposed.

The productive step of the Mitsunobu reaction is the interaction of III with methanol. The product VI is composed of alkoxyphosphonium ions (IV) and O-deprotonated phenylhydroxylamine, creating either a tight ion pair in vacuo or a separated ion pair when optimized under implicit solvent conditions. The product VI is energetically comparable to phenylnitrene, in line with the experimental observation. We did not attempt to locate the transition state for the reaction, as the role of acid present in the reaction media could not be modeled without invoking the explicit solvent models and molecular dynamics simulations.

Combining our findings with the previously reported energetics for 2-vinylnitrosobenzene,¹⁶ the effect of substituents on nitrosobenzene seems crucial for the course of the reaction. The vinyl group is neither a significant electron-donating nor a significant electron-withdrawing group;¹⁷ nevertheless, its presence substantially affects the calculated energies. The steric effect plays probably a role. It could be speculated that a well-chosen substituent in a suitable position on the nitrosobenzene auxiliary would alter the course of the reaction even more toward the preferred Mitsunobu channel and disfavor the nitrene formation.

Nitrene is known to react smoothly with nitrosobenzene to form azoxybenzene or it can polymerize.^13,18 Indeed, besides triphenylphosphine oxide, we found azoxybenzene after the reaction of Ph₃P with PhNO (see the Supporting Information, Table S7). We did not detect N-phenylhydroxylamine (V) in the reaction mixture after the esterification reaction, which can be attributed to its decomposition under our reaction conditions. We proved this hypothesis in an independent experiment, which showed that N-phenylhydroxylamine completely disappeared upon stirring with nitrosobenzene and Ph₃P (see the Supporting Information, Table S8). We also checked the N-phenylhydroxylamine byproduct as a potential substrate for the reaction with benzoic acid, but we observed the formation of an O-benzyl derivative in an amount below 1% under our esterification condition.

One can speculate on the participation of the above-mentioned byproducts in the esterification reaction. Thus, we performed the reaction between 1a and 2a in the presence of azoxybenzene, azobenzene, or phenylnitrene (generated from azidobenzene using irradiation with visible light), but we did not observe the formation of 3a in any of these experiments.

Conclusions

In summary, we have demonstrated that a nitroso compound can be used as an alternative to a dialkyl azodicarboxylate reagent in the Mitsunobu esterification reaction. Esterification with nitrosobenzene and triphenylphosphine occurs under very mild reaction conditions and provides the desired products in moderate yields with good to excellent enantioselectivity. Although there is a drawback in our protocol because of the undesired deoxygenation of the reagent (nitrosobenzene), our system has potential to be optimized by suppressing the side reaction, e.g., by tuning the nitroso compound structure. Generally, our results show that compounds containing N=O bonds can be involved in the activation of triphenylphosphine in addition to the conventional azo-type compounds. Thus, this study may act as an inspiration and stimulus toward new alternative Mitsunobu reaction protocols.

Experimental Section

General

NMR spectra were recorded on Varian Mercury Plus 300 (299.97 MHz for ¹H, 75.44 MHz for ¹³C, and 282.23 MHz for ¹⁹F) or Agilent 400-MR DDR2 (399.94 MHz for ¹H, 100.58 MHz for ¹³C, and 376.29 MHz for ¹⁹F) at 298 K unless otherwise indicated. Chemical shifts δ are given in parts per million, using residual solvent as an internal standard. ¹⁹F NMR and ³¹P NMR chemical shifts were measured relative to CCl₃F and H₃PO₄, respectively. Coupling constants J are reported in hertz. High-resolution mass spectra were obtained on Q-Tof Micro (Waters), equipped with a quadrupole, TOF analyzers, and an MCP detector or LTQ Orbitrap Velos (Thermo Fisher Scientific). Gas chromatography–mass spectrometry (GC–MS) spectra were obtained on GC–MS DSQ II (Thermo). Thin layer chromatography analyses were carried out on DC Alufolien Kieselgel 60 F254 (Merck). Preparative column chromatography separations were performed on silica gel Kieselgel 60 of 0.040–0.063 mm (Merck). Melting points were measured on a Boetius melting point apparatus and are uncorrected. Starting materials, reagents, and substrates were obtained from commercial suppliers and used without further purification. The solvents were purified and dried using standard procedures. For synthesis of reagents, see the Supporting Information.

General Method for the Preparative Mitsunobu Reaction

The reactions were performed in a 100 mL round-bottom flask. Alcohol 1 (0.60 mmol), acid 2 (0.72 mmol), and nitrosobenzene (1.5 mmol) were dissolved in dried acetonitrile (54 mL). The flask was closed by septum and cooled in a bath with water and ice to 0 °C. A cold solution of triphenylphosphine (10 mL, 0.15 M, 1.5 mmol) in dried acetonitrile was added. The reaction mixture was stirred at 0 °C for 30 min and then evaporated. The crude product was inspected by ¹H NMR to estimate the conversion, and then, it was loaded on silica gel. The products 3 were purified by column chromatography (hexane–ethyl acetate 10:1) and were obtained as oil substances.

Benzyl Benzoate (3a)

Benzyl benzoate (3a) was prepared from alcohol 1a (62 μL; 0.60 mmol) and acid 2a (88 mg; 0.72 mmol) according to the general procedure. Yield (58 mg, 47%); ¹H NMR (400 MHz, CDCl₃): δ 8.12–8.06 (m, 2H), 7.60–7.53 (m, 1H), 7.49–7.32 (m, 7H), 5.38 (s, 2H); ¹³C NMR (101 MHz, CDCl₃): δ 166.6, 136.2, 133.2, 130.3, 129.8, 128.7, 128.5, 128.4, 128.3, 66.8; HRMS (EI⁺) calculated for C₁₅H₁₁F₃O₂ ([M]⁺): 212.08318, found: 212.08239.

4-Chlorobenzyl Benzoate (3b)

4-Chlorobenzyl benzoate (3b) was prepared from 4-chlorobenzyl alcohol (1b) (86 mg; 0.60 mmol) and acid 2a (88 mg; 0.72 mmol) according to the general procedure. Yield (80 mg; 57%); ¹H NMR (400 MHz, CDCl₃): δ 8.10–8.03 (m, 2H), 7.61–7.52 (m, 3H), 7.49–7.33 (m, 4H), 5.32 (d, J = 5.6, 2H); ¹³C NMR (101 MHz, CDCl₃): δ 166.3, 134.5, 134.1, 133.1, 129.9, 129.7, 129.6, 128.8, 128.4, 65.9; HRMS (EI⁺) calculated for C₁₄H₁₁ClO₂ ([M]⁺): 246.04421, found: 246.04544.

4-Chlorobenzyl-2-phenylacetate (3c)

4-Chlorobenzyl-2-phenylacetate (3c) was prepared from alcohol 1b (86 mg; 0.60 mmol) and 2-phenylacetic acid (2b) (98 mg; 0.72 mmol) according to the general procedure. Yield (83 mg, 53%); ¹H NMR (400 MHz, CDCl₃): δ 7.94–6.94 (m, 9H), 5.10 (s, 2H), 3.67 (s, 2H); ¹³C NMR (101 MHz, CDCl₃): δ 171.4, 134.5, 134.2, 133.9, 129.6, 129.4, 128.9, 128.8, 127.3, 65.9, 41.5; HRMS (APCI) calculated for C₁₅H₁₃ClO₂ ([M]⁺): 260.05986, found: 260.05996.

4-Methoxybenzyl Benzoate (3d)

4-Methoxybenzyl benzoate (3d) was prepared from 4-methoxybenzyl alcohol (1c) (82 mg; 0.60 mmol) and acid 2a (88 mg; 0.72 mmol) according to the general procedure. Yield (23 mg; 16%); ¹H NMR (400 MHz, CDCl₃): δ 8.17–7.98 (m, 1H), 7.63–7.49 (m, 1H), 7.47–7.34 (m, 2H), 6.97–6.85 (m, 1H), 5.30 (s, 1H), 3.88–3.74 (s, 2H); ¹³C NMR (101 MHz, CDCl₃): δ 166.66, 159.79, 133.08, 130.41, 130.21, 129.81, 128.47, 128.31, 114.11, 66.69, 55.45; ¹H HRMS (EI⁺) calculated for C₁₅H₁₄O₃ ([M]⁺): 242.09429, found: 242.0929.

Benzyl 3-(Trifluoromethyl)benzoate (3e)

Benzyl 3-(trifluoromethyl)benzoate (3e) was prepared from alcohol 1a (62 μL; 0.60 mmol) and acid 2c (134 mg; 0.72 mmol) according to the general procedure. Yield (99 mg, 60%); ¹H NMR (400 MHz, CDCl₃): δ 8.35–8.32 (m, 1H), 8.30–8.23 (m, 1H), 7.85–7.79 (m, 1H), 7.63–7.55 (m, 1H), 7.49–7.34 (m, 5H), 5.40 (s, 2H); ¹⁹F NMR (376 MHz, CDCl₃): δ −63.24; ¹³C NMR (101 MHz, CDCl₃): δ 165.3, 135.7, 133.1 (d, J = 1.5), 131.2 (q, J = 33.2), 131.2, 129.7 (q, J = 3.6), 129.2, 128.8, 128.7, 128.5, 126.8 (q, J = 3.9), 123.8 (q, J = 272.5), 67.4; HRMS (EI⁺) calculated for C₁₅H₁₁F₃O₂ ([M]⁺): 280.07057, found: 280.07160.

4-Chlorobenzyl-3-(trifluoromethyl)benzoate (3f)

4-Chlorobenzyl-3-(trifluoromethyl)benzoate (3f) was prepared from alcohol 1b (86 mg; 0.60 mmol) and acid 2c (134 mg; 0.72 mmol) according to the general procedure. Yield (107 mg, 57%); ¹H NMR (400 MHz, CDCl₃): δ 8.34–8.29 (m, 1H), 8.27–8.22 (m, 1H), 7.86–7.80 (m, 1H), 7.63–7.55 (m, 1H), 7.43–7.34 (m, 4H), 5.36 (s, 2H); ¹⁹F NMR (376 MHz, CDCl₃): δ −63.26; ¹³C NMR (101 MHz, CDCl₃): δ 165.2, 134.6, 134.2, 133.0, 131.3 (q, J = 33.1 Hz), 130.9, 129.9, 129.8 (q, J = 3.7 Hz), 129.3, 129.1, 126.8 (q, J = 3.9 Hz), 123.8 (q, J = 272.6 Hz), 66.6; HRMS (EI⁺) calculated for C₁₅H₁₀ClF₃NO₂ ([M]⁺): 314.03159, found: 314.03206.

n-Octyl-3-(trifluoromethyl)benzoate (3g)

n-Octyl-3-(trifluoromethyl)benzoate (3g) was prepared from octyl alcohol (1d) (94 μL; 0.60 mmol) and acid 2c (134 mg; 0.72 mmol) according to the general procedure. Yield (76 mg, 42%); ¹H NMR (400 MHz, CDCl₃): δ 8.32–8.27 (m, 1H), 8.26–8.19 (m, 1H), 7.86–7.76 (m, 1H), 7.63–7.53 (m, 1H), 4.35 (t, J = 6.8, 2.2 Hz, 2H), 1.84–1.70 (m, 2H), 1.48–1.22 (m, 10H), 0.93–0.84 (m, 3H); ¹⁹F NMR (376 MHz, CDCl₃): δ −63.24; ¹³C NMR (101 MHz, CDCl₃): δ 165.3, 132.7, 131.4, 131.0 (q, J = 32.9 Hz), 129.3 (q, J = 3.6 Hz), 128.9, 126.4 (q, J = 3.8 Hz), 123.7 (q, J = 272.6 Hz), 65.7, 31.8, 29.2, 29.2, 28.6, 25.9, 22.6, 14.1; elemental analysis: calculated for C₁₆H₂₁F₃O₂: C 63.56%, H 7.00%, F 18.85%; found: C 63.58%, H 6.86%, F 18.52%; HRMS calculated for C₁₅H₁₃ClO₂ ([M]⁺): 260.05986, found: 260.06006.

4-Chlorobenzyl Hexanoate (3h)

4-Chlorobenzyl hexanoate (3h) was prepared from alcohol 1b (86 mg; 0.60 mmol) and hexanoic acid (2d) (90 μL; 0.72 mmol) according to the general procedure. Yield (56 mg, 40%); ¹H NMR (400 MHz, CDCl₃): δ 7.35–7.31 (m, 2H), 7.31–7.26 (m, 2H), 5.07 (s, 2H), 2.34 (t, J = 7.6 Hz, 2H), 1.78–1.49 (m, 2H), 1.40–1.20 (m, 4H), 0.88 (t, J = 7.0 Hz, 3H); ¹³C NMR (101 MHz, CDCl₃): δ 173.7, 134.8, 134.2, 129.7, 128.9, 65.4, 34.4, 31.4, 24.8, 22.4, 14.0; HRMS (EI⁺) calcd for C₁₃H₁₇ClO₂ ([M]⁺): 240.09171, found: 240.09134.

4-Chlorobenzyl-3-nitrobenzoate (3i)

4-Chlorobenzyl-3-nitrobenzoate (3i) was prepared from alcohol 1b (86 mg; 0.60 mmol) and 3-nitrobenzoic acid (2e) (120 mg; 0.72 mmol) according to the general procedure. Yield (114 mg, 65%); mp: 85–87 °C; ¹H NMR (400 MHz, CDCl₃) δ 8.87 (t, J = 1.8 Hz, 1H), 8.43 (ddd, J = 8.2, 2.2, 0.9 Hz, 1H), 8.40–8.36 (m, 1H), 7.66 (t, J = 8.0 Hz, 1H), 7.45–7.35 (m, 4H), 5.38 (s, 2H). ¹³C NMR (101 MHz, CDCl₃) δ 164.4, 148.4, 135.5, 134.8, 133.9, 131.9, 130.1, 129.9, 129.1, 127.8, 124.8, 67.0. HRMS (APCI) calcd for C₁₄H₁₀ClNO₄ ([M]⁺): 291.03038, found: 291.03055.

1-Chloro-4-[(4-nitrophenoxy)methyl]benzene (3k)

1-Chloro-4-[(4-nitrophenoxy)methyl]benzene (3k) was prepared from alcohol 1b (86 mg; 0.60 mmol) and 4-nitrophenol (4) (100 mg; 0.72 mmol) according to the general procedure. Yield (9 mg, 6%); mp: 105–109 °C; ¹H NMR (400 MHz, chloroform-d) δ 8.21 (d, J = 9.3 Hz, 1H), 7.38 (m, 2H), 7.02 (d, J = 9.2 Hz, 1H), 5.13 (s, 1H). ¹³C NMR (101 MHz, chloroform-d) δ 163.51, 141.97, 134.56, 134.09, 129.16, 128.96, 126.12, 114.96, 70.02. HRMS (EI⁺) calcd for C₁₃H₁₀ClNO₃ ([M]⁺): 263.03492, found: 263,02356.

(R)- and (S)-1-Phenylethyl 3-Trifluoromethylbenzoate ((R)-3l and (S)-3i)

(S)-1-Phenylethanol ((S)-1e) and acid 2c (134 mg, 0.72 mmol) provided product (R)-3i (38 mg; 22%; 80:20 er inv./ret.) and (R)-1e (72 μL, 0.60 mmol) and acid 2c (134 mg, 0.72 mmol) provided product (S)-3i (35 mg; 20%; 80:20 er inv./ret.) according to the general procedure. ¹H NMR (400 MHz, CDCl₃): δ 8.35–8.31 (m, 1H), 8.28–8.24 (m, 1H), 7.84–7.79 (m, 1H), 7.62–7.55 (m, 1H), 7.48–7.43 (m, 2H), 7.42–7.36 (m, 2H), 7.35–7.30 (m, 1H), 6.16 (q, J = 6.6, 1H), 1.70 (d, J = 6.6, 3H); ¹⁹F NMR (376 MHz, CDCl₃): δ −63.22; ¹³C NMR (101 MHz, CDCl₃): δ 164.7, 141.4, 133.0, 131.5, 131.2 (q, J = 32.8), 129.6 (q, J = 3.6), 129.2, 128.8, 128.3, 126.7 (q, J = 4.0), 126.3, 123.8 (q, J = 272.8), 73.9, 22.4; HRMS (EI⁺) calculated for C₁₆H₁₃F₃O₂ ([M]⁺): 294.08622, found: 294.08640.

(S)-1-Phenylethyl-2-methyl-2-phenylpropanoate ((S)-3m)

(S)-1-Phenylethyl-2-methyl-2-phenylpropanoate ((S)-3m) was prepared from alcohol (R)-1e (72 μL, 0.60 mmol) and 2-methyl-2-phenyl propanoic acid (2f) (118 mg, 0.72 mmol) according to the general procedure. Yield (30 mg; 19%; 80:20 er inv./ret.); ¹H NMR (400 MHz, CDCl₃): δ 7.33–7.12 (m, 10H), 5.86 (q, J = 6.6 Hz, 1H), 1.59 (d, J = 9.2 Hz, 6H), 1.43 (d, J = 6.6 Hz, 3H); ¹³C NMR (101 MHz, CDCl₃): δ 175.7, 144.6, 141.7, 128.3, 128.2, 127.5, 126.5, 125.7, 125.7, 72.5, 46.5, 26.4, 26.3, 22.1; elemental analysis: calculated for C₁₈H₂₀O₂: C 80.56%, H 7.51%; found: C 79.90%, H 7.28%.

(S)-Ethyl-2-((3-(trifluoromethyl)benzoyl)oxy) Propionate ((S)-3n)

(S)-Ethyl-2-((3-(trifluoromethyl)benzoyl)oxy) propionate ((S)-3n) was prepared from (R)-ethyl lactate ((R)-1f) (70 μL, 0.60 mmol) and acid 2c (134 mg, 0.72 mmol) according to the general procedure. Yield (38 mg; 22%; 99:1 er inv./ret.); ¹H NMR (400 MHz, CDCl₃): δ 8.36–8.33 (m, 1H), 8.30–8.25 (m, 1H), 7.87–7.81 (m, 1H), 7.64–7.57 (m, 1H), 5.34 (q, J = 7.1 Hz, 1H), 4.24 (q, J = 7.2 Hz, 2H), 1.66 (d, J = 7.2 Hz, 3H), 1.29 (t, J = 7.1, Hz, 3H); ¹⁹F NMR (376 MHz, CDCl₃): δ −63.26; ¹³C NMR (101 MHz, CDCl₃): δ 170.6, 164.8, 133.2, 131.3 (q, J = 32.9 Hz), 130.5, 129.9 (q, J = 3.7 Hz), 129.3, 126.9 (q, J = 3.9 Hz), 123.8 (q, J = 272.4 Hz), 69.8, 61.7, 17.2, 14.3; HRMS calculated for C₁₃H₁₃F₃O₄ ([M]⁺): 290.07604, found: 290.07586.

Comparative Experiment with the Standard Mitsunobu Protocol

(S)-1-Phenylethyl-3-(trifluoromethyl)benzoate ((S)-3i)

Alcohol (R)-1e (72 μL, 0.60 mmol), acid 2e (134 mg, 0.72 mmol), and DIAD (178 μL, 0.9 mmol) were dissolved in dried acetonitrile (4 mL). Triphenylphosphine (236 mg, 0.9 mmol) was dissolved in dried acetonitrile (4 mL). Both solutions were cooled to 0 °C and mixed. The reaction mixture was stirred at 0 °C for 24 h and evaporated. Crude product 3i was fixed at silica gel and purified by column chromatography (hexane–ethyl acetate 10:1). The product (S)-3i was obtained as an oil substance (125 mg; 71%; 90:10 er inv./ret.). ¹H NMR (400 MHz, CDCl₃): δ 8.35–8.31 (m, 1H), 8.28–8.24 (m, 1H), 7.84–7.79 (m, 1H), 7.62–7.55 (m, 1H), 7.48–7.43 (m, 2H), 7.42–7.36 (m, 2H), 7.35–7.30 (m, 1H), 6.16 (q, J = 6.6, 1H), 1.70 (d, J = 6.6, 3H); ¹⁹F NMR (376 MHz, CDCl₃): δ −63.22; ¹³C NMR (101 MHz, CDCl₃): δ 164.7, 141.4, 133.0, 131.5, 131.2 (q, J = 32.8), 129.6 (q, J = 3.6), 129.2, 128.8, 128.3, 126.7 (q, J = 4.0), 126.3, 123.8 (q, J = 272.8), 73.9, 22.4; HRMS (EI⁺) calculated for C₁₆H₁₃F₃O₂ ([M]⁺): 294.08622, found: 294.08640.

Supporting Information Available

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsomega.8b03551.

• Synthesis of reagents; preliminary experiments and method optimization; mechanistic studies; hardcopy of NMR data of products; DFT calculations including energies and Cartesian coordinates for structures reported in the work (PDF)

Author Contributions

The manuscript was written through contributions of all authors.

The authors declare no competing financial interest.