Improved Schmidt Conversion of Aldehydes to Nitriles Using Azidotrimethylsilane in 1,1,1,3,3,3-Hexafluoro-2-propanol

The Schmidt reaction of aromatic aldehydes using a substoichiometric amount (40 mol %) of triflic acid is described. Low catalyst loading was enabled by a strong hydrogen-bond-donating solvent hexafluoro-2-propanol (HFIP). This improved protocol tolerates a broad scope of aldehydes with diverse functional groups and the corresponding nitriles were obtained in good to high yields without the need for aqueous work up.

An attractive alternative to the above methods is the Schmidt reaction of aromatic aldehydes with hydrazoic acid as in principle it can deliver the nitriles in one straightforward step [44]. However, historically this reaction has provided a mixture of nitriles and formylanilides (Scheme 1a), thus limiting its utility [45]. Recently, Prabhu and co-workers demonstrated that the Schmidt reaction of aldehydes with sodium azide (NaN 3 ) in the presence of triflic acid (TfOH) as a catalyst and acetonitrile (ACN, CH 3 CN) as solvent exclusively affords the corresponding nitriles (Scheme 1b) [46]. In order to achieve complete conversions, 3 equiv of TfOH was minimally required for high yields of the aromatic nitriles. For example, only 6% conversion was observed when 1.5 equiv of TfOH was used during their optimization studies [46]. Similarly, good results can be obtained using a catalyst in an ionic liquid medium [47]. A one-pot sequential Schmidt/Ritter reactions in the presence of 4 equiv of HBF 4¨O Et 2 (2 equiv for each reaction) was also reported for the synthesis of N-tert-butylbenzamides from benzaldehydes [48]. We recently reported an efficient substoichiometric catalytic version of another type of Schmidt reaction, specifically the intramolecular Schmidt reaction of ketones with alkyl azides. In that chemistry, using 1,1,1,3,3,3-hexafluoro-2-propanol (HFIP, (CF 3 ) 2 CHOH) was key to high yields using low loadings of HCl generated in situ from dissolving acetyl chloride in the solvent [49]. These results prompted us to investigate the strong hydrogen bond donor ability of HFIP in the intermolecular Schmidt reaction of aromatic aldehydes.
An attractive alternative to the above methods is the Schmidt reaction of aromatic aldehydes with hydrazoic acid as in principle it can deliver the nitriles in one straightforward step [44]. However, historically this reaction has provided a mixture of nitriles and formylanilides (Scheme 1a), thus limiting its utility [45]. Recently, Prabhu and co-workers demonstrated that the Schmidt reaction of aldehydes with sodium azide (NaN3) in the presence of triflic acid (TfOH) as a catalyst and acetonitrile (ACN, CH3CN) as solvent exclusively affords the corresponding nitriles (Scheme 1b) [46]. In order to achieve complete conversions, 3 equiv of TfOH was minimally required for high yields of the aromatic nitriles. For example, only 6% conversion was observed when 1.5 equiv of TfOH was used during their optimization studies [46]. Similarly, good results can be obtained using a catalyst in an ionic liquid medium [47]. A one-pot sequential Schmidt/Ritter reactions in the presence of 4 equiv of HBF4·OEt2 (2 equiv for each reaction) was also reported for the synthesis of N-tert-butylbenzamides from benzaldehydes [48]. We recently reported an efficient substoichiometric catalytic version of another type of Schmidt reaction, specifically the intramolecular Schmidt reaction of ketones with alkyl azides. In that chemistry, using 1,1,1,3,3,3-hexafluoro-2-propanol (HFIP, (CF3)2CHOH) was key to high yields using low loadings of HCl generated in situ from dissolving acetyl chloride in the solvent [49]. These results prompted us to investigate the strong hydrogen bond donor ability of HFIP in the intermolecular Schmidt reaction of aromatic aldehydes.  [45]); (b) Chemoselective Schmidt reaction of aldehydes to nitriles (Prabhu; [46]).

Optimization of Reaction Conditions
As reported by Prabhu [46], we began our studies on the reaction of 4-nitrobenzaldehyde 1a with NaN 3 and TfOH, replacing ACN as reported by Prabhu with HFIP (Table 1, entry 1). Low conversions of 2a with 50 mol % TfOH (entry 1) and 80 mol % AcCl (entry 2) were obtained from these experiments, likely resulting from the low solubility of NaN 3 in HFIP. Changing to azidotrimethylsilane (TMSN 3 ) as a soluble azide source drastically improved the yield with 25 mol % of acid catalysts (entries 3 and 4). However, incomplete reactions accompanied by polar byproducts were still observed (TLC) despite long periods of stirring. Both AcCl and TiCl 4 are converted to HCl when dissolved in HFIP, so the comparable results seen in entries 2 and 3 make sense taking into account the fact that TiCl 4 provides fourfold more acid than AcCl. We therefore returned to using triflic acid with TMSN 3 as the azide source. Even though the reaction with 30 mol % TfOH offered complete conversion in 2 h, only a modest yield of nitrile was obtained, again with unidentified byproducts (entry 5). Gratifyingly, a 1:1 solvent combination of HFIP and ACN significantly increased the yield but complete conversion was not achieved even after 4 h (entry 6). Finally, the reaction of 1a with 40 mol % TfOH in HFIP/ACN (1:1) mixture proved optimal, providing a slightly better yield of 2a along with a much shorter reaction time (entry 7).  45 3 likely resulting from the low solubility of NaN3 in HFIP. Changing to azidotrimethylsilane (TMSN3) as  a soluble azide source drastically improved the yield with 25 mol % of acid catalysts (entries 3 and 4). However, incomplete reactions accompanied by polar byproducts were still observed (TLC) despite long periods of stirring. Both AcCl and TiCl4 are converted to HCl when dissolved in HFIP, so the comparable results seen in entries 2 and 3 make sense taking into account the fact that TiCl4 provides fourfold more acid than AcCl. We therefore returned to using triflic acid with TMSN3 as the azide source. Even though the reaction with 30 mol % TfOH offered complete conversion in 2 h, only a modest yield of nitrile was obtained, again with unidentified byproducts (entry 5). Gratifyingly, a 1:1 solvent combination of HFIP and ACN significantly increased the yield but complete conversion was not achieved even after 4 h (entry 6). Finally, the reaction of 1a with 40 mol % TfOH in HFIP/ACN (1:1) mixture proved optimal, providing a slightly better yield of 2a along with a much shorter reaction time (entry 7).

Substrate Scope
A series of aromatic aldehydes was examined under the optimized reaction conditions (Table 2). A wide array of functional groups on the aldehydes was well tolerated and the corresponding nitriles were obtained in good to excellent yields. Benzaldehydes containing electron-withdrawing substituents at the para position gave the corresponding nitriles in good yields (entries 1-5). Benzaldehyde 1e required a slightly higher catalyst loading (60 mol %) to achieve a good conversion of the nitrile 2e (entry 5). Electron-rich substrates with a broad range of functional groups such as hydroxyl, O-allyl, and O-propargyl at the para position underwent facile conversion (entries [6][7][8][9][10][11][12][13][14]. Due to the presence of a basic amine, the substrate with a morpholine substituent needed 1.4 equiv of triflic acid, where 1.0 equiv of acid probably ended up in the amine salt (entry 13). Biphenyl substrate 1o afforded nitrile 2o in 80% yield (entry 15). The resulting nitriles were obtained in slightly lower yields for the meta-and ortho-substituted benzaldehydes (entries [16][17][18]. Disubstituted benzaldehydes were also efficiently converted to the desired nitriles in good to high yields (entries 19-25). 2-Naphthonitrile 2z was readily prepared in 77% yield from 2-naphthaldehyde 1z (entry 26). The scope could be extended to heteroaromatic aldehydes affording the representative nitriles in good yields (entries 27 and 28). Throughout, we found that the position of the substituents on the phenyl ring had a relatively minimal influence on the reaction outcome.

Substrate Scope
A series of aromatic aldehydes was examined under the optimized reaction conditions (Table 2). A wide array of functional groups on the aldehydes was well tolerated and the corresponding nitriles were obtained in good to excellent yields. Benzaldehydes containing electron-withdrawing substituents at the para position gave the corresponding nitriles in good yields (entries 1-5). Benzaldehyde 1e required a slightly higher catalyst loading (60 mol %) to achieve a good conversion of the nitrile 2e (entry 5). Electron-rich substrates with a broad range of functional groups such as hydroxyl, O-allyl, and O-propargyl at the para position underwent facile conversion (entries [6][7][8][9][10][11][12][13][14]. Due to the presence of a basic amine, the substrate with a morpholine substituent needed 1.4 equiv of triflic acid, where 1.0 equiv of acid probably ended up in the amine salt (entry 13). Biphenyl substrate 1o afforded nitrile 2o in 80% yield (entry 15). The resulting nitriles were obtained in slightly lower yields for the metaand ortho-substituted benzaldehydes (entries [16][17][18]. Disubstituted benzaldehydes were also efficiently converted to the desired nitriles in good to high yields (entries 19-25). 2-Naphthonitrile 2z was readily prepared in 77% yield from 2-naphthaldehyde 1z (entry 26). The scope could be extended to heteroaromatic aldehydes affording the representative nitriles in good yields (entries 27 and 28). Throughout, we found that the position of the substituents on the phenyl ring had a relatively minimal influence on the reaction outcome.                                         Table 3; see the Experimental Section for details). e TfOH (60 mol %) was used. f TfOH (1.4 equiv) was used. g Commercially used 1af was ca. 77% pure. h TfOH (25 mol %) and TMSN3 (3.0 equiv) was used; see the Experimental Section for details. i Yield of 2af was not corrected w.r.t. 77% purity of 1af.
Cinnamaldehydes 1ac and 1ad lacking double bond substitution reacted smoothly to afford the resultant cinnamonitriles in excellent yields (entries 29-30) whereas α-methyl substituted cinnamaldehyde 1ae provided the nitrile 2ae in only 53% yield (entry 31). We would have been pleased if this method were extendable to aliphatic ketones, which have proved problematic in previous methods as well. Unfortunately, reaction of an aliphatic aldehyde, hydrocinnamaldehyde 1af, with 3 equiv of TMSN3 in the presence of 25 mol % TfOH resulted in a complex mixture from which 3-phenylpropionitrile 2af was isolated in low yield (entry 32). Accordingly, additional aliphatic aldehydes were not explored.
This seemingly simple transformation raises a number of interesting mechanistic questions (Scheme 2). Most workers have adopted some variation of the mechanism originally suggested by P. A. S. Smith [50], in which an initially formed azidohydrin adduct A loses water to afford a pair of equilibrating diazoiminium ions, which can undergo migration of the phenyl group leading to phenylformamide after re-hydration and tautomerization (Scheme 2b). Alternatively, hydride migration followed by deprotonation would similarly afford nitrile; a variation that involves the same intermediate would entail an E2-style elimination of a proton and nitrogen gas, although this is rarely proposed. Confining oneself to the Smith manifold in Scheme 2b, it is hard to justify why a change in solvent would effect the essentially exclusive formation of nitrile since that would most likely be a matter of either intrinsic migration potential between a phenyl vs. hydride or differences in the ratio of the acyliminium ion stereoisomers shown in brackets (in general, the barrier for the interconversion between these is thought to be high) [51]. On the other hand, Ostrovskii et al. have suggested that the Smith dehydration mechanism, leading to nitrile, is in competition with a direct rearrangement   Table 3; see the Experimental Section for details). e TfOH (60 mol %) was used. f TfOH (1.4 equiv) was used. g Commercially used 1af was ca. 77% pure. h TfOH (25 mol %) and TMSN3 (3.0 equiv) was used; see the Experimental Section for details. i Yield of 2af was not corrected w.r.t. 77% purity of 1af.
Cinnamaldehydes 1ac and 1ad lacking double bond substitution reacted smoothly to afford the resultant cinnamonitriles in excellent yields (entries 29-30) whereas α-methyl substituted cinnamaldehyde 1ae provided the nitrile 2ae in only 53% yield (entry 31). We would have been pleased if this method were extendable to aliphatic ketones, which have proved problematic in previous methods as well. Unfortunately, reaction of an aliphatic aldehyde, hydrocinnamaldehyde 1af, with 3 equiv of TMSN3 in the presence of 25 mol % TfOH resulted in a complex mixture from which 3-phenylpropionitrile 2af was isolated in low yield (entry 32). Accordingly, additional aliphatic aldehydes were not explored.
This seemingly simple transformation raises a number of interesting mechanistic questions (Scheme 2). Most workers have adopted some variation of the mechanism originally suggested by P. A. S. Smith [50], in which an initially formed azidohydrin adduct A loses water to afford a pair of equilibrating diazoiminium ions, which can undergo migration of the phenyl group leading to phenylformamide after re-hydration and tautomerization (Scheme 2b). Alternatively, hydride migration followed by deprotonation would similarly afford nitrile; a variation that involves the same intermediate would entail an E2-style elimination of a proton and nitrogen gas, although this is rarely proposed. Confining oneself to the Smith manifold in Scheme 2b, it is hard to justify why a change in solvent would effect the essentially exclusive formation of nitrile since that would most likely be a matter of either intrinsic migration potential between a phenyl vs. hydride or differences in the ratio of the acyliminium ion stereoisomers shown in brackets (in general, the barrier for the interconversion between these is thought to be high) [51]. On the other hand, Ostrovskii et al. have suggested that the Smith dehydration mechanism, leading to nitrile, is in competition with a direct rearrangement  Cinnamaldehydes 1ac and 1ad lacking double bond substitution reacted smoothly to afford the resultant cinnamonitriles in excellent yields (entries 29-30) whereas α-methyl substituted cinnamaldehyde 1ae provided the nitrile 2ae in only 53% yield (entry 31). We would have been pleased if this method were extendable to aliphatic ketones, which have proved problematic in previous methods as well. Unfortunately, reaction of an aliphatic aldehyde, hydrocinnamaldehyde 1af, with 3 equiv of TMSN3 in the presence of 25 mol % TfOH resulted in a complex mixture from which 3-phenylpropionitrile 2af was isolated in low yield (entry 32). Accordingly, additional aliphatic aldehydes were not explored.
This seemingly simple transformation raises a number of interesting mechanistic questions (Scheme 2). Most workers have adopted some variation of the mechanism originally suggested by P. A. S. Smith [50], in which an initially formed azidohydrin adduct A loses water to afford a pair of equilibrating diazoiminium ions, which can undergo migration of the phenyl group leading to phenylformamide after re-hydration and tautomerization (Scheme 2b). Alternatively, hydride migration followed by deprotonation would similarly afford nitrile; a variation that involves the same intermediate would entail an E2-style elimination of a proton and nitrogen gas, although this is rarely proposed. Confining oneself to the Smith manifold in Scheme 2b, it is hard to justify why a change in solvent would effect the essentially exclusive formation of nitrile since that would most likely be a matter of either intrinsic migration potential between a phenyl vs. hydride or differences in the ratio of the acyliminium ion stereoisomers shown in brackets (in general, the barrier for the interconversion between these is thought to be high) [51]. On the other hand, Ostrovskii et al. have suggested that the Smith dehydration mechanism, leading to nitrile, is in competition with a direct rearrangement  Cinnamaldehydes 1ac and 1ad lacking double bond substitution reacted smoothly to afford the resultant cinnamonitriles in excellent yields (entries 29-30) whereas α-methyl substituted cinnamaldehyde 1ae provided the nitrile 2ae in only 53% yield (entry 31). We would have been pleased if this method were extendable to aliphatic ketones, which have proved problematic in previous methods as well. Unfortunately, reaction of an aliphatic aldehyde, hydrocinnamaldehyde 1af, with 3 equiv of TMSN3 in the presence of 25 mol % TfOH resulted in a complex mixture from which 3-phenylpropionitrile 2af was isolated in low yield (entry 32). Accordingly, additional aliphatic aldehydes were not explored.
This seemingly simple transformation raises a number of interesting mechanistic questions (Scheme 2). Most workers have adopted some variation of the mechanism originally suggested by P. A. S. Smith [50], in which an initially formed azidohydrin adduct A loses water to afford a pair of equilibrating diazoiminium ions, which can undergo migration of the phenyl group leading to phenylformamide after re-hydration and tautomerization (Scheme 2b). Alternatively, hydride migration followed by deprotonation would similarly afford nitrile; a variation that involves the same intermediate would entail an E2-style elimination of a proton and nitrogen gas, although this is rarely proposed. Confining oneself to the Smith manifold in Scheme 2b, it is hard to justify why a change in solvent would effect the essentially exclusive formation of nitrile since that would most likely be a matter of either intrinsic migration potential between a phenyl vs. hydride or differences in the ratio of the acyliminium ion stereoisomers shown in brackets (in general, the barrier for the interconversion between these is thought to be high) [51]. On the other hand, Ostrovskii et al. have suggested that the Smith dehydration mechanism, leading to nitrile, is in competition with a direct rearrangement  Cinnamaldehydes 1ac and 1ad lacking double bond substitution reacted smoothly to afford the resultant cinnamonitriles in excellent yields (entries 29-30) whereas α-methyl substituted cinnamaldehyde 1ae provided the nitrile 2ae in only 53% yield (entry 31). We would have been pleased if this method were extendable to aliphatic ketones, which have proved problematic in previous methods as well. Unfortunately, reaction of an aliphatic aldehyde, hydrocinnamaldehyde 1af, with 3 equiv of TMSN3 in the presence of 25 mol % TfOH resulted in a complex mixture from which 3-phenylpropionitrile 2af was isolated in low yield (entry 32). Accordingly, additional aliphatic aldehydes were not explored.
This seemingly simple transformation raises a number of interesting mechanistic questions (Scheme 2). Most workers have adopted some variation of the mechanism originally suggested by P. A. S. Smith [50], in which an initially formed azidohydrin adduct A loses water to afford a pair of equilibrating diazoiminium ions, which can undergo migration of the phenyl group leading to phenylformamide after re-hydration and tautomerization (Scheme 2b). Alternatively, hydride migration followed by deprotonation would similarly afford nitrile; a variation that involves the same intermediate would entail an E2-style elimination of a proton and nitrogen gas, although this is rarely proposed. Confining oneself to the Smith manifold in Scheme 2b, it is hard to justify why a change in solvent would effect the essentially exclusive formation of nitrile since that would most likely be a matter of either intrinsic migration potential between a phenyl vs. hydride or differences in the ratio of the acyliminium ion stereoisomers shown in brackets (in general, the barrier for the interconversion between these is thought to be high) [51]. On the other hand, Ostrovskii et al. have suggested that the Smith dehydration mechanism, leading to nitrile, is in competition with a direct rearrangement  Cinnamaldehydes 1ac and 1ad lacking double bond substitution reacted smoothly to afford the resultant cinnamonitriles in excellent yields (entries 29-30) whereas α-methyl substituted cinnamaldehyde 1ae provided the nitrile 2ae in only 53% yield (entry 31). We would have been pleased if this method were extendable to aliphatic ketones, which have proved problematic in previous methods as well. Unfortunately, reaction of an aliphatic aldehyde, hydrocinnamaldehyde 1af, with 3 equiv of TMSN3 in the presence of 25 mol % TfOH resulted in a complex mixture from which 3-phenylpropionitrile 2af was isolated in low yield (entry 32). Accordingly, additional aliphatic aldehydes were not explored.
This seemingly simple transformation raises a number of interesting mechanistic questions (Scheme 2). Most workers have adopted some variation of the mechanism originally suggested by P. A. S. Smith [50], in which an initially formed azidohydrin adduct A loses water to afford a pair of equilibrating diazoiminium ions, which can undergo migration of the phenyl group leading to phenylformamide after re-hydration and tautomerization (Scheme 2b). Alternatively, hydride migration followed by deprotonation would similarly afford nitrile; a variation that involves the same intermediate would entail an E2-style elimination of a proton and nitrogen gas, although this is rarely proposed. Confining oneself to the Smith manifold in Scheme 2b, it is hard to justify why a change in solvent would effect the essentially exclusive formation of nitrile since that would most likely be a matter of either intrinsic migration potential between a phenyl vs. hydride or differences in the ratio of the acyliminium ion stereoisomers shown in brackets (in general, the barrier for the interconversion between these is thought to be high) [51]. On the other hand, Ostrovskii et al. have suggested that the Smith dehydration mechanism, leading to nitrile, is in competition with a direct rearrangement  Cinnamaldehydes 1ac and 1ad lacking double bond substitution reacted smoothly to afford the resultant cinnamonitriles in excellent yields (entries 29-30) whereas α-methyl substituted cinnamaldehyde 1ae provided the nitrile 2ae in only 53% yield (entry 31). We would have been pleased if this method were extendable to aliphatic ketones, which have proved problematic in previous methods as well. Unfortunately, reaction of an aliphatic aldehyde, hydrocinnamaldehyde 1af, with 3 equiv of TMSN3 in the presence of 25 mol % TfOH resulted in a complex mixture from which 3-phenylpropionitrile 2af was isolated in low yield (entry 32). Accordingly, additional aliphatic aldehydes were not explored.
This seemingly simple transformation raises a number of interesting mechanistic questions (Scheme 2). Most workers have adopted some variation of the mechanism originally suggested by P. A. S. Smith [50], in which an initially formed azidohydrin adduct A loses water to afford a pair of equilibrating diazoiminium ions, which can undergo migration of the phenyl group leading to phenylformamide after re-hydration and tautomerization (Scheme 2b). Alternatively, hydride migration followed by deprotonation would similarly afford nitrile; a variation that involves the same intermediate would entail an E2-style elimination of a proton and nitrogen gas, although this is rarely proposed. Confining oneself to the Smith manifold in Scheme 2b, it is hard to justify why a change in solvent would effect the essentially exclusive formation of nitrile since that would most likely be a matter of either intrinsic migration potential between a phenyl vs. hydride or differences in the ratio of the acyliminium ion stereoisomers shown in brackets (in general, the barrier for the interconversion between these is thought to be high) [51]. On the other hand, Ostrovskii et al. have suggested that the Smith dehydration mechanism, leading to nitrile, is in competition with a direct rearrangement  Cinnamaldehydes 1ac and 1ad lacking double bond substitution reacted smoothly to afford the resultant cinnamonitriles in excellent yields (entries 29-30) whereas α-methyl substituted cinnamaldehyde 1ae provided the nitrile 2ae in only 53% yield (entry 31). We would have been pleased if this method were extendable to aliphatic ketones, which have proved problematic in previous methods as well. Unfortunately, reaction of an aliphatic aldehyde, hydrocinnamaldehyde 1af, with 3 equiv of TMSN3 in the presence of 25 mol % TfOH resulted in a complex mixture from which 3-phenylpropionitrile 2af was isolated in low yield (entry 32). Accordingly, additional aliphatic aldehydes were not explored.
This seemingly simple transformation raises a number of interesting mechanistic questions (Scheme 2). Most workers have adopted some variation of the mechanism originally suggested by P. A. S. Smith [50], in which an initially formed azidohydrin adduct A loses water to afford a pair of equilibrating diazoiminium ions, which can undergo migration of the phenyl group leading to phenylformamide after re-hydration and tautomerization (Scheme 2b). Alternatively, hydride migration followed by deprotonation would similarly afford nitrile; a variation that involves the same intermediate would entail an E2-style elimination of a proton and nitrogen gas, although this is rarely proposed. Confining oneself to the Smith manifold in Scheme 2b, it is hard to justify why a change in solvent would effect the essentially exclusive formation of nitrile since that would most likely be a matter of either intrinsic migration potential between a phenyl vs. hydride or differences in the ratio of the acyliminium ion stereoisomers shown in brackets (in general, the barrier for the interconversion between these is thought to be high) [51]. On the other hand, Ostrovskii et al. have suggested that the Smith dehydration mechanism, leading to nitrile, is in competition with a direct rearrangement  Cinnamaldehydes 1ac and 1ad lacking double bond substitution reacted smoothly to afford the resultant cinnamonitriles in excellent yields (entries 29-30) whereas α-methyl substituted cinnamaldehyde 1ae provided the nitrile 2ae in only 53% yield (entry 31). We would have been pleased if this method were extendable to aliphatic ketones, which have proved problematic in previous methods as well. Unfortunately, reaction of an aliphatic aldehyde, hydrocinnamaldehyde 1af, with 3 equiv of TMSN3 in the presence of 25 mol % TfOH resulted in a complex mixture from which 3-phenylpropionitrile 2af was isolated in low yield (entry 32). Accordingly, additional aliphatic aldehydes were not explored.
This seemingly simple transformation raises a number of interesting mechanistic questions (Scheme 2). Most workers have adopted some variation of the mechanism originally suggested by P. A. S. Smith [50], in which an initially formed azidohydrin adduct A loses water to afford a pair of equilibrating diazoiminium ions, which can undergo migration of the phenyl group leading to phenylformamide after re-hydration and tautomerization (Scheme 2b). Alternatively, hydride migration followed by deprotonation would similarly afford nitrile; a variation that involves the same intermediate would entail an E2-style elimination of a proton and nitrogen gas, although this is rarely proposed. Confining oneself to the Smith manifold in Scheme 2b, it is hard to justify why a change in solvent would effect the essentially exclusive formation of nitrile since that would most likely be a matter of either intrinsic migration potential between a phenyl vs. hydride or differences in the ratio of the acyliminium ion stereoisomers shown in brackets (in general, the barrier for the interconversion between these is thought to be high) [51]. On the other hand, Ostrovskii et al. have suggested that the Smith dehydration mechanism, leading to nitrile, is in competition with a direct rearrangement  Cinnamaldehydes 1ac and 1ad lacking double bond substitution reacted smoothly to afford the resultant cinnamonitriles in excellent yields (entries 29-30) whereas α-methyl substituted cinnamaldehyde 1ae provided the nitrile 2ae in only 53% yield (entry 31). We would have been pleased if this method were extendable to aliphatic ketones, which have proved problematic in previous methods as well. Unfortunately, reaction of an aliphatic aldehyde, hydrocinnamaldehyde 1af, with 3 equiv of TMSN3 in the presence of 25 mol % TfOH resulted in a complex mixture from which 3-phenylpropionitrile 2af was isolated in low yield (entry 32). Accordingly, additional aliphatic aldehydes were not explored.
This seemingly simple transformation raises a number of interesting mechanistic questions (Scheme 2). Most workers have adopted some variation of the mechanism originally suggested by P. A. S. Smith [50], in which an initially formed azidohydrin adduct A loses water to afford a pair of equilibrating diazoiminium ions, which can undergo migration of the phenyl group leading to phenylformamide after re-hydration and tautomerization (Scheme 2b). Alternatively, hydride migration followed by deprotonation would similarly afford nitrile; a variation that involves the same intermediate would entail an E2-style elimination of a proton and nitrogen gas, although this is rarely proposed. Confining oneself to the Smith manifold in Scheme 2b, it is hard to justify why a change in solvent would effect the essentially exclusive formation of nitrile since that would most likely be a matter of either intrinsic migration potential between a phenyl vs. hydride or differences in the ratio of the acyliminium ion stereoisomers shown in brackets (in general, the barrier for the interconversion between these is thought to be high) [51]. On the other hand, Ostrovskii et al. have suggested that the Smith dehydration mechanism, leading to nitrile, is in competition with a direct rearrangement  Cinnamaldehydes 1ac and 1ad lacking double bond substitution reacted smoothly to afford the resultant cinnamonitriles in excellent yields (entries 29-30) whereas α-methyl substituted cinnamaldehyde 1ae provided the nitrile 2ae in only 53% yield (entry 31). We would have been pleased if this method were extendable to aliphatic ketones, which have proved problematic in previous methods as well. Unfortunately, reaction of an aliphatic aldehyde, hydrocinnamaldehyde 1af, with 3 equiv of TMSN3 in the presence of 25 mol % TfOH resulted in a complex mixture from which 3-phenylpropionitrile 2af was isolated in low yield (entry 32). Accordingly, additional aliphatic aldehydes were not explored.
This seemingly simple transformation raises a number of interesting mechanistic questions (Scheme 2). Most workers have adopted some variation of the mechanism originally suggested by P. A. S. Smith [50], in which an initially formed azidohydrin adduct A loses water to afford a pair of equilibrating diazoiminium ions, which can undergo migration of the phenyl group leading to phenylformamide after re-hydration and tautomerization (Scheme 2b). Alternatively, hydride migration followed by deprotonation would similarly afford nitrile; a variation that involves the same intermediate would entail an E2-style elimination of a proton and nitrogen gas, although this is rarely proposed. Confining oneself to the Smith manifold in Scheme 2b, it is hard to justify why a change in solvent would effect the essentially exclusive formation of nitrile since that would most likely be a matter of either intrinsic migration potential between a phenyl vs. hydride or differences in the ratio of the acyliminium ion stereoisomers shown in brackets (in general, the barrier for the interconversion between these is thought to be high) [51]. On the other hand, Ostrovskii et al. have suggested that the Smith dehydration mechanism, leading to nitrile, is in competition with a direct rearrangement  Cinnamaldehydes 1ac and 1ad lacking double bond substitution reacted smoothly to afford the resultant cinnamonitriles in excellent yields (entries 29-30) whereas α-methyl substituted cinnamaldehyde 1ae provided the nitrile 2ae in only 53% yield (entry 31). We would have been pleased if this method were extendable to aliphatic ketones, which have proved problematic in previous methods as well. Unfortunately, reaction of an aliphatic aldehyde, hydrocinnamaldehyde 1af, with 3 equiv of TMSN3 in the presence of 25 mol % TfOH resulted in a complex mixture from which 3-phenylpropionitrile 2af was isolated in low yield (entry 32). Accordingly, additional aliphatic aldehydes were not explored.
This seemingly simple transformation raises a number of interesting mechanistic questions (Scheme 2). Most workers have adopted some variation of the mechanism originally suggested by P. A. S. Smith [50], in which an initially formed azidohydrin adduct A loses water to afford a pair of equilibrating diazoiminium ions, which can undergo migration of the phenyl group leading to phenylformamide after re-hydration and tautomerization (Scheme 2b). Alternatively, hydride migration followed by deprotonation would similarly afford nitrile; a variation that involves the same intermediate would entail an E2-style elimination of a proton and nitrogen gas, although this is rarely proposed. Confining oneself to the Smith manifold in Scheme 2b, it is hard to justify why a change in solvent would effect the essentially exclusive formation of nitrile since that would most likely be a matter of either intrinsic migration potential between a phenyl vs. hydride or differences in the ratio of the acyliminium ion stereoisomers shown in brackets (in general, the barrier for the interconversion between these is thought to be high) [51]. On the other hand, Ostrovskii et al. have suggested that the Smith dehydration mechanism, leading to nitrile, is in competition with a direct rearrangement  Cinnamaldehydes 1ac and 1ad lacking double bond substitution reacted smoothly to afford the resultant cinnamonitriles in excellent yields (entries 29-30) whereas α-methyl substituted cinnamaldehyde 1ae provided the nitrile 2ae in only 53% yield (entry 31). We would have been pleased if this method were extendable to aliphatic ketones, which have proved problematic in previous methods as well. Unfortunately, reaction of an aliphatic aldehyde, hydrocinnamaldehyde 1af, with 3 equiv of TMSN3 in the presence of 25 mol % TfOH resulted in a complex mixture from which 3-phenylpropionitrile 2af was isolated in low yield (entry 32). Accordingly, additional aliphatic aldehydes were not explored.
This seemingly simple transformation raises a number of interesting mechanistic questions (Scheme 2). Most workers have adopted some variation of the mechanism originally suggested by P. A. S. Smith [50], in which an initially formed azidohydrin adduct A loses water to afford a pair of equilibrating diazoiminium ions, which can undergo migration of the phenyl group leading to phenylformamide after re-hydration and tautomerization (Scheme 2b). Alternatively, hydride migration followed by deprotonation would similarly afford nitrile; a variation that involves the same intermediate would entail an E2-style elimination of a proton and nitrogen gas, although this is rarely proposed. Confining oneself to the Smith manifold in Scheme 2b, it is hard to justify why a change in solvent would effect the essentially exclusive formation of nitrile since that would most likely be a matter of either intrinsic migration potential between a phenyl vs. hydride or differences in the ratio of the acyliminium ion stereoisomers shown in brackets (in general, the barrier for the interconversion between these is thought to be high) [51]. On the other hand, Ostrovskii et al. have suggested that the Smith dehydration mechanism, leading to nitrile, is in competition with a direct rearrangement  Cinnamaldehydes 1ac and 1ad lacking double bond substitution reacted smoothly to afford the resultant cinnamonitriles in excellent yields (entries 29-30) whereas α-methyl substituted cinnamaldehyde 1ae provided the nitrile 2ae in only 53% yield (entry 31). We would have been pleased if this method were extendable to aliphatic ketones, which have proved problematic in previous methods as well. Unfortunately, reaction of an aliphatic aldehyde, hydrocinnamaldehyde 1af, with 3 equiv of TMSN3 in the presence of 25 mol % TfOH resulted in a complex mixture from which 3-phenylpropionitrile 2af was isolated in low yield (entry 32). Accordingly, additional aliphatic aldehydes were not explored.
This seemingly simple transformation raises a number of interesting mechanistic questions (Scheme 2). Most workers have adopted some variation of the mechanism originally suggested by P. A. S. Smith [50], in which an initially formed azidohydrin adduct A loses water to afford a pair of equilibrating diazoiminium ions, which can undergo migration of the phenyl group leading to phenylformamide after re-hydration and tautomerization (Scheme 2b). Alternatively, hydride migration followed by deprotonation would similarly afford nitrile; a variation that involves the same intermediate would entail an E2-style elimination of a proton and nitrogen gas, although this is rarely proposed. Confining oneself to the Smith manifold in Scheme 2b, it is hard to justify why a change in solvent would effect the essentially exclusive formation of nitrile since that would most likely be a matter of either intrinsic migration potential between a phenyl vs. hydride or differences in the ratio of the acyliminium ion stereoisomers shown in brackets (in general, the barrier for the interconversion between these is thought to be high) [51]. On the other hand, Ostrovskii et al. have suggested that the Smith dehydration mechanism, leading to nitrile, is in competition with a direct rearrangement    Table 3; see the Experimental Section for details). e TfOH (60 mol %) was used. f TfOH (1.4 equiv) was used. g Commercially used 1af was ca. 77% pure. h TfOH (25 mol %) and TMSN3 (3.0 equiv) was used; see the Experimental Section for details. i Yield of 2af was not corrected w.r.t. 77% purity of 1af.
Cinnamaldehydes 1ac and 1ad lacking double bond substitution reacted smoothly to afford the resultant cinnamonitriles in excellent yields (entries 29-30) whereas α-methyl substituted cinnamaldehyde 1ae provided the nitrile 2ae in only 53% yield (entry 31). We would have been pleased if this method were extendable to aliphatic ketones, which have proved problematic in previous methods as well. Unfortunately, reaction of an aliphatic aldehyde, hydrocinnamaldehyde 1af, with 3 equiv of TMSN3 in the presence of 25 mol % TfOH resulted in a complex mixture from which 3-phenylpropionitrile 2af was isolated in low yield (entry 32). Accordingly, additional aliphatic aldehydes were not explored.
This seemingly simple transformation raises a number of interesting mechanistic questions (Scheme 2). Most workers have adopted some variation of the mechanism originally suggested by P. A. S. Smith [50], in which an initially formed azidohydrin adduct A loses water to afford a pair of equilibrating diazoiminium ions, which can undergo migration of the phenyl group leading to phenylformamide after re-hydration and tautomerization (Scheme 2b). Alternatively, hydride migration followed by deprotonation would similarly afford nitrile; a variation that involves the same intermediate would entail an E2-style elimination of a proton and nitrogen gas, although this is rarely proposed. Confining oneself to the Smith manifold in Scheme 2b, it is hard to justify why a change in solvent would effect the essentially exclusive formation of nitrile since that would most likely be a matter of either intrinsic migration potential between a phenyl vs. hydride or differences in the ratio of the acyliminium ion stereoisomers shown in brackets (in general, the barrier for the interconversion between these is thought to be high) [51]. On the other hand, Ostrovskii et al. have suggested that the Smith dehydration mechanism, leading to nitrile, is in competition with a direct rearrangement    Table 3; see the Experimental Section for details). e TfOH (60 mol %) was used. f TfOH (1.4 equiv) was used. g Commercially used 1af was ca. 77% pure. h TfOH (25 mol %) and TMSN3 (3.0 equiv) was used; see the Experimental Section for details. i Yield of 2af was not corrected w.r.t. 77% purity of 1af.
Cinnamaldehydes 1ac and 1ad lacking double bond substitution reacted smoothly to afford the resultant cinnamonitriles in excellent yields (entries 29-30) whereas α-methyl substituted cinnamaldehyde 1ae provided the nitrile 2ae in only 53% yield (entry 31). We would have been pleased if this method were extendable to aliphatic ketones, which have proved problematic in previous methods as well. Unfortunately, reaction of an aliphatic aldehyde, hydrocinnamaldehyde 1af, with 3 equiv of TMSN3 in the presence of 25 mol % TfOH resulted in a complex mixture from which 3-phenylpropionitrile 2af was isolated in low yield (entry 32). Accordingly, additional aliphatic aldehydes were not explored.
This seemingly simple transformation raises a number of interesting mechanistic questions (Scheme 2). Most workers have adopted some variation of the mechanism originally suggested by P. A. S. Smith [50], in which an initially formed azidohydrin adduct A loses water to afford a pair of equilibrating diazoiminium ions, which can undergo migration of the phenyl group leading to phenylformamide after re-hydration and tautomerization (Scheme 2b). Alternatively, hydride migration followed by deprotonation would similarly afford nitrile; a variation that involves the same intermediate would entail an E2-style elimination of a proton and nitrogen gas, although this is rarely proposed. Confining oneself to the Smith manifold in Scheme 2b, it is hard to justify why a change in solvent would effect the essentially exclusive formation of nitrile since that would most likely be a matter of either intrinsic migration potential between a phenyl vs. hydride or differences in the ratio of the acyliminium ion stereoisomers shown in brackets (in general, the barrier for the interconversion between these is thought to be high) [51]. On the other hand, Ostrovskii et al. have suggested that the Smith dehydration mechanism, leading to nitrile, is in competition with a direct rearrangement    Table 3; see the Experimental Section for details). e TfOH (60 mol %) was used. f TfOH (1.4 equiv) was used. g Commercially used 1af was ca. 77% pure. h TfOH (25 mol %) and TMSN3 (3.0 equiv) was used; see the Experimental Section for details. i Yield of 2af was not corrected w.r.t. 77% purity of 1af.
Cinnamaldehydes 1ac and 1ad lacking double bond substitution reacted smoothly to afford the resultant cinnamonitriles in excellent yields (entries 29-30) whereas α-methyl substituted cinnamaldehyde 1ae provided the nitrile 2ae in only 53% yield (entry 31). We would have been pleased if this method were extendable to aliphatic ketones, which have proved problematic in previous methods as well. Unfortunately, reaction of an aliphatic aldehyde, hydrocinnamaldehyde 1af, with 3 equiv of TMSN3 in the presence of 25 mol % TfOH resulted in a complex mixture from which 3-phenylpropionitrile 2af was isolated in low yield (entry 32). Accordingly, additional aliphatic aldehydes were not explored.
This seemingly simple transformation raises a number of interesting mechanistic questions (Scheme 2). Most workers have adopted some variation of the mechanism originally suggested by P. A. S. Smith [50], in which an initially formed azidohydrin adduct A loses water to afford a pair of equilibrating diazoiminium ions, which can undergo migration of the phenyl group leading to phenylformamide after re-hydration and tautomerization (Scheme 2b). Alternatively, hydride migration followed by deprotonation would similarly afford nitrile; a variation that involves the same intermediate would entail an E2-style elimination of a proton and nitrogen gas, although this is rarely proposed. Confining oneself to the Smith manifold in Scheme 2b, it is hard to justify why a change in solvent would effect the essentially exclusive formation of nitrile since that would most likely be a matter of either intrinsic migration potential between a phenyl vs. hydride or differences in the ratio of the acyliminium ion stereoisomers shown in brackets (in general, the barrier for the interconversion between these is thought to be high) [51]. On the other hand, Ostrovskii et al. have suggested that the Smith dehydration mechanism, leading to nitrile, is in competition with a direct rearrangement  Cinnamaldehydes 1ac and 1ad lacking double bond substitution reacted smoothly to afford the resultant cinnamonitriles in excellent yields (entries 29-30) whereas α-methyl substituted cinnamaldehyde 1ae provided the nitrile 2ae in only 53% yield (entry 31). We would have been pleased if this method were extendable to aliphatic ketones, which have proved problematic in previous methods as well. Unfortunately, reaction of an aliphatic aldehyde, hydrocinnamaldehyde 1af, with 3 equiv of TMSN3 in the presence of 25 mol % TfOH resulted in a complex mixture from which 3-phenylpropionitrile 2af was isolated in low yield (entry 32). Accordingly, additional aliphatic aldehydes were not explored.
This seemingly simple transformation raises a number of interesting mechanistic questions (Scheme 2). Most workers have adopted some variation of the mechanism originally suggested by P. A. S. Smith [50], in which an initially formed azidohydrin adduct A loses water to afford a pair of equilibrating diazoiminium ions, which can undergo migration of the phenyl group leading to phenylformamide after re-hydration and tautomerization (Scheme 2b). Alternatively, hydride migration followed by deprotonation would similarly afford nitrile; a variation that involves the same intermediate would entail an E2-style elimination of a proton and nitrogen gas, although this is rarely proposed. Confining oneself to the Smith manifold in Scheme 2b, it is hard to justify why a change in solvent would effect the essentially exclusive formation of nitrile since that would most likely be a matter of either intrinsic migration potential between a phenyl vs. hydride or differences in the ratio of the acyliminium ion stereoisomers shown in brackets (in general, the barrier for the interconversion between these is thought to be high) [51]. On the other hand, Ostrovskii et al. have suggested that the Smith dehydration mechanism, leading to nitrile, is in competition with a direct rearrangement  Cinnamaldehydes 1ac and 1ad lacking double bond substitution reacted smoothly to afford the resultant cinnamonitriles in excellent yields (entries 29-30) whereas α-methyl substituted cinnamaldehyde 1ae provided the nitrile 2ae in only 53% yield (entry 31). We would have been pleased if this method were extendable to aliphatic ketones, which have proved problematic in previous methods as well. Unfortunately, reaction of an aliphatic aldehyde, hydrocinnamaldehyde 1af, with 3 equiv of TMSN3 in the presence of 25 mol % TfOH resulted in a complex mixture from which 3-phenylpropionitrile 2af was isolated in low yield (entry 32). Accordingly, additional aliphatic aldehydes were not explored.
This seemingly simple transformation raises a number of interesting mechanistic questions (Scheme 2). Most workers have adopted some variation of the mechanism originally suggested by P. A. S. Smith [50], in which an initially formed azidohydrin adduct A loses water to afford a pair of equilibrating diazoiminium ions, which can undergo migration of the phenyl group leading to phenylformamide after re-hydration and tautomerization (Scheme 2b). Alternatively, hydride migration followed by deprotonation would similarly afford nitrile; a variation that involves the same intermediate would entail an E2-style elimination of a proton and nitrogen gas, although this is rarely proposed. Confining oneself to the Smith manifold in Scheme 2b, it is hard to justify why a change in solvent would effect the essentially exclusive formation of nitrile since that would most likely be a matter of either intrinsic migration potential between a phenyl vs. hydride or differences in the ratio of the acyliminium ion stereoisomers shown in brackets (in general, the barrier for the interconversion between these is thought to be high) [51]. On the other hand, Ostrovskii et al. have suggested that the Smith dehydration mechanism, leading to nitrile, is in competition with a direct rearrangement  Cinnamaldehydes 1ac and 1ad lacking double bond substitution reacted smoothly to afford the resultant cinnamonitriles in excellent yields (entries 29-30) whereas α-methyl substituted cinnamaldehyde 1ae provided the nitrile 2ae in only 53% yield (entry 31). We would have been pleased if this method were extendable to aliphatic ketones, which have proved problematic in previous methods as well. Unfortunately, reaction of an aliphatic aldehyde, hydrocinnamaldehyde 1af, with 3 equiv of TMSN3 in the presence of 25 mol % TfOH resulted in a complex mixture from which 3-phenylpropionitrile 2af was isolated in low yield (entry 32). Accordingly, additional aliphatic aldehydes were not explored.
This seemingly simple transformation raises a number of interesting mechanistic questions (Scheme 2). Most workers have adopted some variation of the mechanism originally suggested by P. A. S. Smith [50], in which an initially formed azidohydrin adduct A loses water to afford a pair of equilibrating diazoiminium ions, which can undergo migration of the phenyl group leading to phenylformamide after re-hydration and tautomerization (Scheme 2b). Alternatively, hydride migration followed by deprotonation would similarly afford nitrile; a variation that involves the same intermediate would entail an E2-style elimination of a proton and nitrogen gas, although this is rarely proposed. Confining oneself to the Smith manifold in Scheme 2b, it is hard to justify why a change in solvent would effect the essentially exclusive formation of nitrile since that would most likely be a matter of either intrinsic migration potential between a phenyl vs. hydride or differences in the ratio of the acyliminium ion stereoisomers shown in brackets (in general, the barrier for the interconversion between these is thought to be high) [51]. On the other hand, Ostrovskii et al. have suggested that the Smith dehydration mechanism, leading to nitrile, is in competition with a direct rearrangement a To a solution of aldehyde 1 (1.0 equiv) and TMSN 3 (2.0 equiv) in a premixed HFIP/ACN solvent mixture (2.0 mL, 1:1) was added TfOH (40 mol %) and the reaction was allowed to stir at rt for a period of 20-75 min. b Concentration of aldehyde 1 was ca. 0.25 M. c Isolated yields. d Contains ca. 4% of unreacted 1a (see the Experimental Section for details). e TfOH (60 mol %) was used. f TfOH (1.4 equiv) was used. g Commercially used 1af was ca. 77% pure. h TfOH (25 mol %) and TMSN 3 (3.0 equiv) was used; see the Experimental Section for details. i Yield of 2af was not corrected w.r.t. 77% purity of 1af.
Cinnamaldehydes 1ac and 1ad lacking double bond substitution reacted smoothly to afford the resultant cinnamonitriles in excellent yields (entries 29-30) whereas α-methyl substituted cinnamaldehyde 1ae provided the nitrile 2ae in only 53% yield (entry 31). We would have been pleased if this method were extendable to aliphatic ketones, which have proved problematic in previous methods as well. Unfortunately, reaction of an aliphatic aldehyde, hydrocinnamaldehyde 1af, with 3 equiv of TMSN 3 in the presence of 25 mol % TfOH resulted in a complex mixture from which 3-phenylpropionitrile 2af was isolated in low yield (entry 32). Accordingly, additional aliphatic aldehydes were not explored.
This seemingly simple transformation raises a number of interesting mechanistic questions (Scheme 2). Most workers have adopted some variation of the mechanism originally suggested by P. A. S. Smith [50], in which an initially formed azidohydrin adduct A loses water to afford a pair of equilibrating diazoiminium ions, which can undergo migration of the phenyl group leading to phenylformamide after re-hydration and tautomerization (Scheme 2b). Alternatively, hydride migration followed by deprotonation would similarly afford nitrile; a variation that involves the same intermediate would entail an E2-style elimination of a proton and nitrogen gas, although this is rarely proposed. Confining oneself to the Smith manifold in Scheme 2b, it is hard to justify why a change in solvent would effect the essentially exclusive formation of nitrile since that would most likely be a matter of either intrinsic migration potential between a phenyl vs. hydride or differences in the ratio of the acyliminium ion stereoisomers shown in brackets (in general, the barrier for the interconversion between these is thought to be high) [51]. On the other hand, Ostrovskii et al. have suggested that the Smith dehydration mechanism, leading to nitrile, is in competition with a direct rearrangement pathway, leading to phenylformamide (Scheme 2c) [52,53]. Acidic HFIP is a strongly dehydrating medium, which would be consistent with this observation. Finally, it is tempting to speculate that "superelectrophilic" species [54] like the protonated (or hydrogen bonded) diazoiminium ion or nitrilium ions shown in Scheme 2d might also be involved, although this must remain, for the moment, an intriguing conjecture pending further mechanistic work.
Molecules 2016, 21, 45 6 pathway, leading to phenylformamide (Scheme 2c) [52,53]. Acidic HFIP is a strongly dehydrating medium, which would be consistent with this observation. Finally, it is tempting to speculate that "superelectrophilic" species [54] like the protonated (or hydrogen bonded) diazoiminium ion or nitrilium ions shown in Scheme 2d might also be involved, although this must remain, for the moment, an intriguing conjecture pending further mechanistic work.

Scheme 2.
Mechanistic possibilities. In all cases, the SiMe3 group might be replaced by H under the reaction conditions (leading to exactly analogous pathways).

General Information
Reactions were performed in glass sample vial with rubber lined cap. All chemicals were used as received from commercial source, without further purification. Acetonitrile was dried by passage through neutral alumina columns using a commercial solvent purification system prior to use. Thin-layer chromatography (TLC) was performed using commercial glass-backed silica plates (250 microns) with an organic binder. Visualization was accomplished with UV light. Flash chromatography was carried out on a CombiFlash ® purification system using a 4 g normal phase silica flash column. Infrared (IR) spectra were acquired as a solid (Shimadzu FTIR-8400S, Kyoto, Japan). All nuclear magnetic resonance (NMR) spectra ( 1 H, 13 C, APT) were recorded on a 400 MHz instrument (Bruker AV-400, Billerica, MA, USA). NMR spectra were recorded in deuterated chloroform. Chemical shifts are reported in parts per million (ppm) and are referenced to the center line of the solvent (δ 7.26 ppm for 1 H-NMR and δ 77.23 for 13 C-NMR, respectively). Coupling constants are given in Hertz (Hz). Melting points were determined on an automated melting point apparatus and are uncorrected. A sample concentrator using N2 gas was used for the concentration of reaction mixtures. Spectroscopic data for the aromatic nitriles prepared according to the methodology described in this paper matched well with those reported in the literature.

General Information
Reactions were performed in glass sample vial with rubber lined cap. All chemicals were used as received from commercial source, without further purification. Acetonitrile was dried by passage through neutral alumina columns using a commercial solvent purification system prior to use. Thin-layer chromatography (TLC) was performed using commercial glass-backed silica plates (250 microns) with an organic binder. Visualization was accomplished with UV light. Flash chromatography was carried out on a CombiFlash purification system using a 4 g normal phase silica flash column. Infrared (IR) spectra were acquired as a solid (Shimadzu FTIR-8400S, Kyoto, Japan). All nuclear magnetic resonance (NMR) spectra ( 1 H, 13 C, APT) were recorded on a 400 MHz instrument (Bruker AV-400, Billerica, MA, USA). NMR spectra were recorded in deuterated chloroform. Chemical shifts are reported in parts per million (ppm) and are referenced to the center line of the solvent (δ 7.26 ppm for 1 H-NMR and δ 77.23 for 13 C-NMR, respectively). Coupling constants are given in Hertz (Hz). Melting points were determined on an automated melting point apparatus and are uncorrected. A sample concentrator using N 2 gas was used for the concentration of reaction mixtures. Spectroscopic data for the aromatic nitriles prepared according to the methodology described in this paper matched well with those reported in the literature.

General Procedure for the Optimization of Reaction Conditions for the Synthesis of 4-Nitrobenzonitrile 2a
To a solution of 4-nitrobenzaldehyde 1a (0.25 or 0.50 mmol, 1.0 equiv) and NaN 3 or TMSN 3 (1.5-2.0 equiv) in HFIP or HFIP/ACN mixture (0.50, 1.0, or 2.0 mL) was added a catalyst (effervescence due to nitrogen gas evolution was immediately observed). The vial was capped and the reaction mixture was allowed to stir at rt for a specified period (45 min to 24 h). The reaction mixture was concentrated under nitrogen. The residue obtained was diluted with appropriate solvent (CH 2 Cl 2 or EtOAc) and was either filtered through a Pasteur pipette containing a cotton plug to get a crude 1 H-NMR ratio (for entries 1 and 2) or purified using a 4 or 12 g normal phase silica flash column on a CombiFlash purification system with a gradient elution of 0%-10% EtOAc/hexanes (for entries 3-7). Concentration of the appropriate fractions afforded 4-nitrobenzonitrile 2a contaminated with a small amount (ca. 3%-6%) of 1a (except for entry 5, where pure 2a was obtained).

General Procedure A for the Synthesis of Aromatic Nitriles
To a solution of an aromatic aldehyde 1 (0.500 mmol, 1.0 equiv) and TMSN 3 (115 mg, 1.00 mmol, 2.0 equiv) in a premixed HFIP/ACN mixture (2.0 mL, 1:1) in a nitrogen-flushed two dram vial was added triflic acid (TfOH; 17.7 µL, 0.200 mmol, 0.40 equiv) (exotherm and brisk effervescence due to nitrogen gas evolution was immediately observed). The vial was capped and the reaction mixture was allowed to stir at rt for 20-75 min. The reaction mixture was concentrated under nitrogen. The residue obtained was suspended in CH 2 Cl 2 /hexanes mixture and loaded on a silica gel in a 5 g sample cartridge. Purification using a normal phase silica flash column on a CombiFlash purification system afforded a corresponding aromatic nitrile 2 upon concentration of appropriate fractions. mixture was stirred at rt for 30 min. Purification using a 4 g silica flash column on a CombiFlash purification system (0%-10% EtOAc/hexanes over 40 min) afforded 2d (eluted between 2.5%-4.2% EtOAc/hexanes) as a colorless crystalline solid (63.0 mg, 0.391 mmol, 78% yield).