Ortho-Phosphinoarenesulfonamide-Mediated Staudinger Reduction of Aryl and Alkyl Azides

Conventional Staudinger reductions of organic azides are sluggish with aryl or bulky aliphatic azides. In addition, Staudinger reduction usually requires a large excess of water to promote the decomposition of the aza-ylide intermediate into phosphine oxide and amine products. To overcome the challenges above, we designed a novel triaryl phosphine reagent 2c with an ortho-SO2NH2 substituent. Herein, we report that such phosphine reagents are able to mediate the Staudinger reduction of both aryl and alkyl azides in either anhydrous or wet solvents. Good to excellent yields were obtained in all cases (even at a diluted concentration of 0.01 M). The formation of B-TAP, a cyclic aza-ylide, instead of phosphine oxide, eliminates the requirement of water in the Staudinger reduction. In addition, computational studies disclose that the intramolecular protonation of the aza-ylide by the ortho-SO2NH2 group is kinetically favorable and responsible for the acceleration of Staudinger reduction of the aryl azides.


Introduction
Staudinger reduction [1][2][3] is one of the most common transformations used to prepare amino compounds in organic synthesis. Recently, the biorthogonal property [4][5][6][7][8][9][10][11] of organic azides [12] has significantly enhanced the visibility of the Staudinger reaction in the context of chemical biology and medicinal chemistry. The classic Staudinger protocol employs trivalent phosphines, predominately triphenylphosphine, to mediate the reduction of organic azides under aqueous conditions via an aza-ylide (or iminophosphorane) intermediate [13,14] (Scheme 1A). Although thermodynamically the formation of triphenylphosphine oxide provides a sufficient driving force for the Staudinger reduction, reactions are sluggish when the iminophosphorane intermediates are kinetically stable [15][16][17]. For example, iminophosphorane intermediates formed from aryl azides can be quite stable under neutral conditions, thus requiring either acids or bases to mediate their hydrolysis [18,19]. In addition, the amount of water in the system has a significant impact on the rate of Staudinger reduction, particularly for highly hydrophobic substrates. Recently, Ito, Abe, and coworkers [20] reported that o-(diphenylphosphino) benzamide accelerated the reduction of aryl and alkyl azides towards the corresponding amines (Scheme 1B). In their report, aqueous conditions were still required for the reduction of aromatic azides. The authors indicated that a significant amount of by-product was formed without the addition of water. It therefore remains highly desirable to design novel phosphorus reagents that are able to mediate a broad-spectrum Staudinger reduction without the restriction of an aqueous solvent. Very recently, our group prepared a novel class of cyclic iminophosphoranes, namely (benzo[d])-1,2,3-thiazaphosphole (B-TAP) [21]. The improved stability of the B-TAP heterocycle towards hydrolysis suggests that the stronger P=N of B-TAP bond may drive a Staudinger reduction as the P=O of phosphine oxide did. Herein, we report the use of novel class of cyclic iminophosphoranes, namely (benzo[d])-1,2,3-thiazaphosphole TAP) [21]. The improved stability of the B-TAP heterocycle towards hydrolysis sugge that the stronger P=N of B-TAP bond may drive a Staudinger reduction as the P=O phosphine oxide did. Herein, we report the use of ortho-phosphinoarenesulfonamides novel reagents to mediate the Staudinger reduction of both aryl and alkyl azides und either anhydrous or aqueous conditions (Scheme 1C). Scheme 1. Phosphine-mediated Staudinger reductions of organic azides.

Results and Discussions
The B-TAP heterocycle was originally synthesized in our laboratory by the annu tion of ortho-phosphinoarenesulfonyl fluorides with commercially available trimethyls azide [21]. Further studies indicated that B-TAP can also be prepared by the oxidat condensation of ortho-phosphinoarenesulfonamide. We speculated that the formation B-TAP could be achieved by the use of an organic azide as an oxidant. This hypothesis us to explore ortho-phosphinoarenesulfonamide-mediated Staudinger reductions. proposed that the introduction of an acidic sulfonamide group at the ortho position wou facilitate the decomposition step of the original aza-ylide and thus the addition of wa would not be necessary. In this way, the formation of a B-TAP heterocycle with a m stable P=N bond would provide the driving force for Staudinger reduction.
We selected methyl 4-azidobenzoate (1a) as a model substrate ( Table 1). As expect when 2a was applied the major product was the triphenylphosphine-derived aza-yl while the characteristic signals of methyl 4-aminobenzoate (3a) were not observed in 1 H NMR (entry 1, Table 1). Interestingly, ortho-phosphinoarenesulfonamide (2b and 2 mediated reductions were both complete within 3 h. Reductions at lower concentrat (0.01 M) with shorter reaction times indicated that 2c gave faster rates than 2b (entries 5, Table 1) and therefore 2c was chosen for further investigation. The 2c-media Staudinger reduction was found compatible with a broad range of organic solvents ( tries 8-12, Table 1). The rate of 2c-mediated Staudinger reduction in aqueous solvent ( try 13, Table 1) was similar to anhydrous solvents. This suggested that the addition water did not affect the Staudinger reduction. This is in sharp contrast to other kno staudinger reduction processes where water accelerates the hydrolysis step. Notably, majority of B-TAP 4c (>90%) survived in aqueous THF and was not hydrolyzed to corresponding phosphine oxide.

Results and Discussions
The B-TAP heterocycle was originally synthesized in our laboratory by the annulation of ortho-phosphinoarenesulfonyl fluorides with commercially available trimethylsilyl azide [21]. Further studies indicated that B-TAP can also be prepared by the oxidative condensation of ortho-phosphinoarenesulfonamide. We speculated that the formation of B-TAP could be achieved by the use of an organic azide as an oxidant. This hypothesis led us to explore ortho-phosphinoarenesulfonamide-mediated Staudinger reductions. We proposed that the introduction of an acidic sulfonamide group at the ortho position would facilitate the decomposition step of the original aza-ylide and thus the addition of water would not be necessary. In this way, the formation of a B-TAP heterocycle with a more stable P=N bond would provide the driving force for Staudinger reduction.
We selected methyl 4-azidobenzoate (1a) as a model substrate ( Table 1). As expected, when 2a was applied the major product was the triphenylphosphine-derived aza-ylide while the characteristic signals of methyl 4-aminobenzoate (3a) were not observed in the 1 H NMR (entry 1, Table 1). Interestingly, ortho-phosphinoarenesulfonamide (2b and 2c)-mediated reductions were both complete within 3 h. Reductions at lower concentration (0.01 M) with shorter reaction times indicated that 2c gave faster rates than 2b (entries 4-5, Table 1) and therefore 2c was chosen for further investigation. The 2c-mediated Staudinger reduction was found compatible with a broad range of organic solvents (entries 8-12, Table 1). The rate of 2c-mediated Staudinger reduction in aqueous solvent (entry 13, Table 1) was similar to anhydrous solvents. This suggested that the addition of water did not affect the Staudinger reduction. This is in sharp contrast to other known staudinger reduction processes where water accelerates the hydrolysis step. Notably, the majority of B-TAP 4c (>90%) survived in aqueous THF and was not hydrolyzed to the corresponding phosphine oxide.  Next, we explored the substrate scope of the 2c-mediated Staudinger reduction (Scheme 2). Applying our standardized conditions, all aromatic azides evaluated gave excellent yields (1a-1m, Scheme 2). Reductions of aliphatic azides were relatively slow at 30 °C and required elevated temperatures and longer reaction times for reaction completion, likely attributed to the decreased electrophilicity of alkyl azides. Gratifyingly, the reduction of aliphatic substrates 1n-1v proceeded smoothly under the modified conditions, with satisfactory yields being obtained in all cases (Scheme 2). Indeed, 2c-mediated Staudinger reduction tolerated a broad range of functional groups and various substitution patterns. For instance, substrates with electron-withdrawing groups (1a-1c, 1h, and 1i), electron-donating groups (1d and 1f) or reactive functionalities (1e, 1g, and 1j) are all fully compatible with the current protocol. Moreover, sterically-encumbered amines such as 3k, 3l, 3n, and 3o were obtained in high yields when ortho-substituted or branched substrates were used. Next, we explored the substrate scope of the 2c-mediated Staudinger reduction (Scheme 2). Applying our standardized conditions, all aromatic azides evaluated gave excellent yields (1a-1m, Scheme 2). Reductions of aliphatic azides were relatively slow at 30 • C and required elevated temperatures and longer reaction times for reaction completion, likely attributed to the decreased electrophilicity of alkyl azides. Gratifyingly, the reduction of aliphatic substrates 1n-1v proceeded smoothly under the modified conditions, with satisfactory yields being obtained in all cases (Scheme 2). Indeed, 2c-mediated Staudinger reduction tolerated a broad range of functional groups and various substitution patterns. For instance, substrates with electron-withdrawing groups (1a-1c, 1h, and 1i), electron-donating groups (1d and 1f) or reactive functionalities (1e, 1g, and 1j) are all fully compatible with the current protocol. Moreover, sterically-encumbered amines such as 3k, 3l, 3n, and 3o were obtained in high yields when ortho-substituted or branched substrates were used.
To demonstrate the synthetic potential of this methodology in a more complex setting, we explored the synthesis of a highly functional and sterically-hindered cinchonine-derived amine (3w). Staudinger reductions of cinchona alkaloid-derived azides have been reported to be problematic in some cases [22]. Their corresponding aza-ylide intermediates are reluctant to hydrolyze and are even able to be isolated by flash column chromatography. We proposed that accelerated hydrolysis could occur for 2c-mediated Staudinger reduction since the ortho-SO 2 NH 2 may facilitate the decomposition of the aza-ylide through intramolecular protonation. To our delight, the above proposal was verified by experiments with 2c promoting an efficient Staudinger reduction of azide 1w. The reduction was conducted on a 5.00-mmol scale and 1.21 g of product 3w was isolated in 84% yield (Scheme 3).
To elucidate the mechanism of the 2c-mediated Staudinger reduction, we examined both the 2c-and PPh 3 -mediated Staudinger reductions by DFT calculation (Figure 1) [23]. Both reductions share a three-step reaction pathway. In the initial step, the nucleophilic attack of either of the phosphine reagents on the azide generates a betaine intermediate. Then, this intermediate loses one equivalent of nitrogen gas to form an aza-ylide intermediate. Finally, the decomposition of the aza-ylide yields the amine product. In the conventional PPh 3 -mediated Staudinger reduction, the rate-determining step was the step from Int-2 to TS-3 . The high activation barrier (33.4 kcal/mol) was consistent with the fact that Int-2 was resistant to hydrolysis. In the 2c-mediated Staudinger reduction, the rate-determining step was the step from Int-1 to TS-2 with a moderate activation barrier (24.7 kcal/mol). In addition, several mechanistic insights have been disclosed by comparing both reaction pathways. The activation barrier for the formation of aza-ylide in the 2c-mediated process (24.7 kcal/mol) was slightly higher than the one in PPh 3 -mediated processes (21.7 kcal/mol). However, the free energy barrier for the decomposition of aza-ylide Int-2 (18.8 kcal/mol) was significantly lower than the one for the decomposition of aza-ylide Int-2 (33.4 kcal/mol). Notably, the difference in the enthalpy barriers between both processes was much smaller (18.5 kcal/mol vs. 21.1 kcal/mol). Apparently, the proximal SO 2 NH 2 group in 2c plays an important role in the protonation of aza-ylide Int-2. Its significant contribution to entropy greatly lowers the free energy barrier. We also note that the PPh 3 -mediated reduction is more thermodynamically favorable than the 2c-mediated Staudinger reduction. Therefore, it is most likely that the 2c-mediated process is largely controlled by favorable kinetics. The rate acceleration in the 2c-mediated reduction probably originated from the rapid intramolecular proton transfer in the aza-ylide decomposition step. To demonstrate the synthetic potential of this methodology in a more complex setting, we explored the synthesis of a highly functional and sterically-hindered cinchoninederived amine (3w). Staudinger reductions of cinchona alkaloid-derived azides have been reported to be problematic in some cases [22]. Their corresponding aza-ylide intermediates are reluctant to hydrolyze and are even able to be isolated by flash column chromatography. We proposed that accelerated hydrolysis could occur for 2c-mediated Staudinger Scheme 2. Substrate scope of 2c-mediated Staudinger reduction a . a Reactions were run on a 1.00 mmol scale in 10 mL of anhydrous THF under a N 2 atmosphere. All yields were isolated yields. b Reactions were run at 60 • C for 15 h instead. c Reactions were run on a 0.50 mmol scale instead. are reluctant to hydrolyze and are even able to be isolated by flash column chromatography. We proposed that accelerated hydrolysis could occur for 2c-mediated Staudinger reduction since the ortho-SO2NH2 may facilitate the decomposition of the aza-ylide through intramolecular protonation. To our delight, the above proposal was verified by experiments with 2c promoting an efficient Staudinger reduction of azide 1w. The reduction was conducted on a 5.00-mmol scale and 1.21 g of product 3w was isolated in 84% yield (Scheme 3).

Synthesis of Organic Azides
Organic azides except 1j, 1s, and 1w were prepared from the literature methods and their analytical data were consistent with the literature data (see Supplementary Materials). Azides 1j, 1s and 1w were prepared by the procedures below.

Synthesis of Amines
General Procedure A (for the substrates 1a-1m): Under a N 2 atmosphere, the corresponding organic azide (1.00 mmol, 1.0 equiv) was dissolved in anhydrous tetrahydrofuran (10 mL), followed by the addition of 2c (1.10 mmol, 1.1 equiv). The mixture was stirred at 30 • C for 3 h. Upon completion, the mixture was concentrated in vacuo and the crude product was purified by flash column chromatography.
General Procedure B (for the substrates 1n-1w): Under a N 2 atmosphere, the corresponding organic azide (1.00 mmol, 1.0 equiv) was dissolved in anhydrous tetrahydrofuran (10 mL), followed by the addition of 2c (1.10 mmol, 1.1 equiv). The mixture was stirred at 60 • C for 15 h. Upon completion, the mixture was concentrated in vacuo and the crude product was purified by flash column chromatography. For 1w, 5.00 mmol of azide was used instead.
All amines except 3j and 3s are known compounds and their analytical data were consistent with the literature data (see Supplementary Materials). Their isolated yields were reported in Scheme 2.

Computational Methods
The calculations were carried out with the Gaussian 09 software package [34]. The structures were optimized by the density functional theory (DFT) [35] with the B3LYP functional [36,37] with basis set 6-31G(d) [38,39] in the gas phase. Frequency analysis was conducted at the same level of theory to verify the stationary points to be real minima or saddle points and to obtain the thermodynamic energy corrections at 298.15 K. Intrinsic reaction coordinate (IRC) [40][41][42] calculations were performed to confirm the connection between two correct minima for a transition state. More accurate electronic energy results were refined by calculating the single-point energy at the B3LYP-D3(BJ) [43]/6-311++G(2df, 2p) [38,39] level of theory with the SMD model [44] (solvent = THF).

Conclusions
In conclusion, we developed the first ortho-phosphinoarenesulfonamide-mediated Staudinger reduction without the need for water. Computational studies suggest that the ortho-SO 2 NH 2 substituent of the phosphine reagent is significant for favorable reaction kinetics. We are currently investigating other aza-ylide-driven Staudinger-type transformations which will be reported in due course.