An Efficient Synthesis of Achiral and Chiral 1,2,4-Triazolium Salts

Abstract

The promising utility of triazolyl N-heterocyclic carbene catalysts in umpolung aldehyde chemistry requires a straightforward reliable synthesis from readily available materials. Herein we describe the synthesis of a variety of triazolyl N-heterocyclic carbene precursors. The reactions commence from commercially available amino acids and proceed in 44-68% overall yields. The N-heterocyclic salts are air stable crystalline solids that can be stored with no special precaution and can generate the active catalyst when treated with an appropriate base.

N-Heterocyclic carbenes have become a notable area of research since the first stable carbene was reported in 1991 by Arduengo and coworkers.¹ The imidazolinylidene carbene scaffold has been extensively used as a ligand in transition metal mediated processes,² and a number of facile procedures are available for its preparation.³ Unlike their imidazolium salt counterparts, the preparation and utility of triazolium salts as precursors for N-heterocyclic carbenes have been less well explored. Noteworthy exceptions are the practical and efficient synthesis reported by Enders and coworkers of an achiral tris-phenyl substituted triazolium salt and its transformation into the free carbene,⁴ and a description of the synthesis of N-admantyl substituted bis-aryl triazolium salts.⁵ Further reports of the syntheses of chiral bicyclic triazolium salts by Leeper⁶ and Enders⁷ have led to a catalyst capable of inducing high enantioselectivity in the benzoin reaction.

In an effort to develop the utility of N-heterocyclic carbenes in asymmetric catalysis, our laboratory has pursued a practical and efficient synthesis of these triazolium salts from readily available materials. Our focus in the early stage of this research pivoted around the incorporation of easily accessible chiral building blocks into a rigid framework that could be manipulated upon further investigation. Catalyst preparation from amino acid derivatives was desirable in order to take advantage of their diverse steric profile and ready availability. Two different chiral bicyclic cores (1 and 2) were envisioned to possess qualities stated above (Figure 1).

(1)

(2)

(3)

(4)

(5)

Figure 1.

Chiral bicyclic triazolium salts.

We have demonstrated the utility of these triazolium salts as precursors for nucleophilic carbenes in a highly enantioselective intramolecular Stetter reaction (eq 1).⁸ We have since illustrated their efficacy in the formation of quaternary stereocenters (eq 2),^9a contiguous stereocenters (eq 3),^9b and their application in a novel internal redox reaction manifold (eq 4) capable of generating high enantiomeric excess in a meso diol desymmetrization (eq 5).^9c Herein we report the synthesis of a variety of chiral and achiral triazolium salt nucleophilic carbene precursors.

The synthesis of enantiopure bicyclic triazolium salt 7a and 7b began with Boc protection of phenylalanine according to Meyers’ procedure (Scheme 1).¹⁰ With an easy route to large amount of N-Boc protected phenylalanine the synthesis of the pyrrolidinone core was realized according to literature precedent.¹¹ Coupling of the Boc-protected amino acid with Meldrum’s acid in the presence of DMAP and DCC affords the desired product that can be used without further purification. Reduction of the ketone with slow addition of 2.5 equivalents of sodium borohydride at 0 °C over 3 hours followed by stirring at 0 °C for 24 hours provides the desired product as yellow oil. Initial attempts at using the resulting yellow oil resulted in a complicated purification of the desired pyrrolidinone 6. However, 5 can be recrystallized from diethyl ether affording an analytically pure white crystalline solid. Cyclization of 5 in toluene at 110 °C followed by removal of the N-Boc protecting group with trifluoroacetic acid gives the desired pyrrolidinone 6 as a yellow solid that can be used without further purification.¹² With the pyrrolidinone in hand, a one-pot modification of the Leeper synthesis⁶ was used for the three-step conversion into triazolium salt. Methylation of 6 with Meerwein’s reagent affords the desired amidate, which was treated in situ with phenylhydrazine to generate a red solution that corresponds to the desired cyclization precursor. Finally, treatment with trimethyl orthoformate in methanol (7:1) at 80 °C affords s triazolium salt 7a. The triazolium salt could be precipitated from ethyl acetate and then recrystallized from hot methanol affording a white crystalline solid. This one pot protocol can be extended to the synthesis of 4-(trifluoromethyl)phenylhydrazine, however, the cyclization step was performed in triethyl orthoformate in methanol (7:1 ) at 110 °C in order to obtain clean formation of catalyst 7b. An X-ray crystal structure of the chloride salt of 7a is shown in Figure 2.

Scheme 1.

Synthesis of Pyrrolidinone-Derived Catalyst Precursors

Figure 2.

X-ray crystal structure of 7a. Counterion and water molecule omitted for clarity.

In efforts to make a des-benzyl analog of this catalyst by a more economical route, amide activating agents other than trimethyloxonium tetrafluoroborate were investigated. For this purpose, 2-pyrrolidinone 8 was implemented to provide the aliphatic part of the bicyclic skeleton (Scheme 2). A variety of attempts to activate the carbonyl through iminoyl chloride intermediates met with no success. Fortunately, refluxing the amide in acetonitrile with dimethyl sulfate provides the desired amidate, which may be treated with phenylhydrazine in situ. Counterion exchange is achieved by liberating the free hydrazino compound with 40% KOH and treating with methanolic HCl to provide the chloride salt. Cyclization in o-dichlorobenzene with trimethyl orthoformate and catalytic HCl provides the achiral catalyst 13.

Scheme 2.

Synthesis of Achiral Catalyst Precursor 13

In addition to the pyrrolidine framework, the morpholine scaffold appeared attractive since it can be readily prepared from amino alcohols.¹³ The synthesis of a chiral bicyclic benzyl-substituted triazolium chloride has been previously reported by Leeper⁶ and can typically be extended to other alkyl groups on the morpholine ring. However, this synthetic route can be problematic with some side-chains and certain aryl hydrazines.

In our explorations into the asymmetric Stetter reaction, we identified aminoindanol-derived catalyst 18 as possessing advantageous properties for catalysis. Furthermore, it became evident that the aryl substituent, introduced with various hydrazines, significantly affects the reactivity of these catalysts.^8a,9a In attempts to address these concerns in the synthesis of these catalysts, we were intrigued by the possibility of the transformation of 14¹⁴ to 18 in a one pot procedure (Scheme 3). Formation of amidate 15 from Meerwein’s salt in dichloromethane over 12 hours at room temperature, followed by addition of phenylhydrazine and stirring at room temperature for 30 minutes gives presumed intermediate 16. Heating 16 in chlorobenzene with triethyl orthoformate for 12 hours provides catalyst precursor 18. This protocol can be extended to the synthesis of slightly modified salts as well, as illustrated in the synthesis of p-anisyl triazolium salt 19.

Scheme 3.

Synthesis of Aminoindanol-Derived Catalyst Precursors

When implementing considerably more electron-deficient aryl hydrazines, we found that the catalyst synthesis had a tendency to be irreproducible during the final cyclization step often resulting in recovery unreacted starting material 16 or 17. As pentafluorophenyl triazolium salt 21 was recently identified as a highly capable catalyst for the intramolecular Stetter reaction,^9a we needed a reliable method for its production. We have found that minor changes in the cyclization step are essential for a clean formation of this catalyst (Scheme 4).

Scheme 4.

Synthesis of Pentafluorophenyl Triazolium Salt 21

Standard amidate formation with Meerwein’s salt in dichloromethane for 12 hours at room temperature and treatment with pentafluorophenyl hydrazine at room temperature for 2 hours provides hydrazinium tetrafluoroborate 20. Evaporation of the solvent, followed by addition of triethyl orthoformate and chlorobenzene and heating at 110 °C for 12 hours initiates the cyclization. Addition of another 5 equivalents of triethyl orthoformate and heating at 110 °C for another 12 hours allows for clean triazolium salt formation. Cooling of this mixture to room temperature, with addition of an equivalent volume of toluene (to chlorobenzene) provides triazolium tetrafluoroborate 21 as a light tan solid which is washed with toluene to provide pure material.

The achiral pentafluorophenyl catalyst can be prepared in a similar manner (Scheme 5). Addition of pentafluorophenyl hydrazine into the amidate derived from Meerwein’s reagent and 2-pyrrolidinone 8 in dichloromethane, stirring for 4 hours, followed by removal of solvent provides hydrazinium tetrafluoroborate 22. Treatment of this compound with 5 equivalents of triethyl orthoformate and heating to 110 °C for 12 hours yields achiral triazolium salt 23.

Scheme 5.

Achiral Pentafluorophenyl Catalyst Precursor 23

In conclusion, we report an improved procedure for the efficient synthesis of a variety of chiral and achiral triazolium salt N-heterocyclic carbene precursors. These are rapidly prepared in modular fashion from common laboratory materials and serve as highly air- and water-stable sources of nucleophilic carbenes. Efforts at expanding the reactivity of these carbenes are currently underway in our laboratories.

Experimental Section

5-Benzyl-2-phenyl-6,7-dihydro-5H-pyrrolo[2,1-c][1,2,4]triazol-2-ium tetrafluoroborate (7a)

A flame-dried 100 mL round bottom flask was charged with 6 (1.0 g, 5.71 mmol) and CH₂Cl₂ (40 mL). Trimethyloxonium tetrafluoroborate (0.93 g, 6.29 mmol) was added and the reaction mixture was stirred overnight at 23 °C. To the pinkish solution was added phenylhydrazine (0.62 mL, 6.29 mmol) and the reaction was stirred overnight. The solvent was removed in vacuo and the product was used without further purification. Methanol (2 mL) and trimethyl orthoformate (14 mL) was added and the reaction mixture was stirred at 80 °C overnight. The reaction mixture was heated to 80 °C and stirred at this temperature overnight. The solvent was removed in vacuo and the product was precipitated from ethyl acetate to give the desired compound as an off white/yellow powder. Recrystallization from hot MeOH affords 7a (1.04 g, 50%) as a white crystalline solid.

2-Pentafluorophenyl-6,10b-dihydro-4H,5aH-5-oxa-3,10c-diaza-2-azonia-cyclopenta[c]fluorene tetrafluoroborate (21)

A flame-dried 100 mL round bottom flask was charged with morpholinone 14 (1.000 g, 5.29 mmol) and CH₂Cl₂ (25 mL). Trimethyloxonium tetrafluoroborate (0.783 g, 5.29 mmol) was added and the reaction mixture stirred for 12 hours at 23 °C. Pentafluorophenylhydrazine (1.047 g, 5.29 mmol) was then added and allowed to stir for 2 hours at 23 °C. The solvent was removed in vacuo and chlorobenzene (50 mL) was added, followed by triethyl orthoformate (2.20 mL, 13.23 mmol). The resulting solution was stirred at 110 °C for 12 hours. At this time, additional triethyl orthoformate (2.20 mL, 13.23 mmol) was added and heating at 110 °C was continued for 12 hours. Upon cooling, toluene (50 mL) was added and the light tan solid product was collected by filtration. This was rinsed with toluene (3 × 5 mL) and heated to 120 °C for 6 hours under vacuum to remove residual water to provide triazolium salt 21 (1.56 g, 63%) as a light tan solid.