Room Temperature Reduction of Titanium Tetrachloride-Activated Nitriles to Primary Amines with Ammonia-Borane

The reduction of a variety of aromatic and aliphatic nitriles, activated by a molar equivalent of titanium tetrachloride, has been achieved at room temperature using ammonia borane as a safe reductant. The corresponding methanamines were isolated in good to excellent yields following a simple acid-base workup.


Introduction
The amine moiety in organic molecules is considered extremely important due to their multifaceted functions, especially in life sciences [1] and industrial chemistry [2,3]. Their applications encompass agro, materials, dye, textile, pharma, surfactant, plastic, and paper industries, to name a few. Accordingly, their syntheses have been the subject of intense research for organic chemists [4,5]. Primary amines function as intermediates or end-products in organic synthesis and have received their due attention. While reductive amination using ammonia or ammonium salts can be envisioned for their synthesis, it is often very challenging to perform [6][7][8]. Another simple process for primary amines is readily achieved via the reduction of organonitriles (Scheme 1) [9].
Most of these procedures have several serious drawbacks, such as air-and moisturesensitivity of the reagents, expensive nature of the reagents or catalysts, the formation of dialkylamines, etc. Efficient procedures for the reduction of nitriles are still being sought actively. As part of our program on amine-boranes, we recently described the reduction of ketones [43] as well as carboxylic acids [44] using ammonia borane (AB, 1a) in diethyl ether (Et 2 O), in the presence of sub-stoichiometric titanium tetrachloride as an activator (Scheme 2). While comparing the competitive reduction of a carboxylic acid and an organonitrile, under standard conditions, the latter was observed to be unreactive. Due to the importance of primary amines, we were eager to learn whether an organonitrile will succumb to the reaction under modified conditions. Accordingly, a systematic examination was initiated and reported herein is the facile conversion of nitriles to primary amines, at room temperature, with ammonia borane in the presence of a molar equivalent of titanium tetrachloride as the activator. Both aliphatic and aromatic nitriles underwent reduction, and the products were isolated, in good to excellent yields, using a simple acid-base workup. tions (TH) using amine-boranes, including ammonia-boran amine have also been reported recently (Scheme 1vi) [29-34 generated via the dehydrogenation of the corresponding am to be effective for the reduction of nitriles to amines [35] (Sc also permeated with reports on the use of silane derivatives for nitrile reduction [36][37][38][39][40][41][42] (Scheme 1viii).
Most of these procedures have several serious drawback sensitivity of the reagents, expensive nature of the reagents o dialkylamines, etc. Efficient procedures for the reduction of n actively. As part of our program on amine-boranes, we recen of ketones [43] as well as carboxylic acids [44] using ammon ether (Et2O), in the presence of sub-stoichiometric titanium (Scheme 2). While comparing the competitive reduction of a nonitrile, under standard conditions, the latter was observed importance of primary amines, we were eager to learn wheth cumb to the reaction under modified conditions. Accordingl was initiated and reported herein is the facile conversion of n room temperature, with ammonia borane in the presence o nium tetrachloride as the activator. Both aliphatic and aromat tion, and the products were isolated, in good to excellent yiel workup. Scheme 2. Reduction of carbonyls activated by TiCl4 using ammoni

Results and Discussion
The successful hydroboration of alkenes [45] and alkyn tetrahydrofuran (THF), as opposed to the lack of any reactio led us to attempt the hydroboration (reduction) of a repres (2a) with 1a in refluxing THF. Surprisingly, even after 20 h, complete with one equiv. and ~60% complete with two equ (entries 2 and 3, Table 1). This prompted a logical modificatio was conducted, under catalyzed conditions, in diethyl ether a the sub-stoichiometric (10%) catalyst loading reported for th acids [43,44], the reaction was now performed in the presen the activator titanium tetrachloride as well as differing stoi agent, 1a. The reaction was followed by thin layer chromatog Scheme 2. Reduction of carbonyls activated by TiCl 4 using ammonia borane.

Results and Discussion
The successful hydroboration of alkenes [45] and alkynes [46] with 1a in refluxing tetrahydrofuran (THF), as opposed to the lack of any reaction at room temperature [47], led us to attempt the hydroboration (reduction) of a representative nitrile, benzonitrile (2a) with 1a in refluxing THF. Surprisingly, even after 20 h, the reaction was only~24% complete with one equiv. and~60% complete with two equiv. of the reducing agent 1a (entries 2 and 3, Table 1). This prompted a logical modification of the reaction of 2a, which was conducted, under catalyzed conditions, in diethyl ether at room temperature. Unlike the sub-stoichiometric (10%) catalyst loading reported for the reduction of ketones and acids [43,44], the reaction was now performed in the presence of increasing amounts of the activator titanium tetrachloride as well as differing stoichiometries of the reducing agent, 1a. The reaction was followed by thin layer chromatography for the disappearance of 2a.
After several attempts, we were delighted to observe that using a molar ratio of 1:2:0.7 for 2a:1a: activator led to complete conversion of 2a to benzylamine (3a) in 77% isolated yield within 3 h (entry 4, Table 1). No conversion of benzonitrile was observed in the absence of the catalyst, confirming the crucial necessity of TiCl 4 for this reduction (entry 1, Table 1). Remarkably, increasing the stoichiometry of TiCl 4 to 1 equiv. and of 1a to 2 equiv. provided an increase in yield (95%), while decreasing the reaction time to an hour (entry 5, Table 1). Decreasing the reagent load of 1a to 1.5 equiv., however, resulted in an inefficient reaction and the yield decreased to 71% (entry 6, Table 1). Increasing the reaction time up to 24 h did not have any effect on the yield. The conversions and reaction rates for amine formation from nitriles strongly depended on the reaction parameters, such as solvent, Lewis acid and its equivalences, as well as the amine-borane used. The effect of the solvent was exemplified by replacing Et 2 O, with dichloromethane (CH 2 Cl 2 ), THF, and pentane under similar conditions. These observations confirmed that Et 2 O is the best solvent to effect the transformation effortlessly (entries 7-9 in Table 1).
Molecules 2023, 28, x FOR PEER REVIEW 4 of 11 meta-, and para-positions was evaluated. Thus, ortho-chlorobenzonitrile (2b), meta-fluorobenzonitrile (2c), and para-fluoro-(2d), -chloro-(2e), and -bromo-(2f) benzonitriles were converted to the corresponding amines (3b-3f) in 70%-72% yields, respectively. Additionally, 2,4-difluorobenzonitrile (2g) provided the desired benzylamine product 3g in 77% yield. No dehalogenation product was observed in all these cases. Reductions of benzonitrile with an electron-deficient group on the aromatic ring, for example, 4-trifluoromethylbenzonitrile (2h) underwent the reduction efficiently to the corresponding amine 3h in almost quantitative yield 97%, indicating that weak electronwithdrawing groups have no impact on this transformation. The electron-donating 2-and 4-methyl (2i-2j) did not inhibit the formation of amines (3i-3j), isolated in 99% and 86% yields in 3 h, respectively, although a slightly higher molar equivalent of 1a (2.5 equiv.) was necessary for complete reduction. Furthermore, the reaction of increased electrondonating 4-methoxybenzonitrile was converted to the desired product methanamine 3k in good yield (86%). It should be noted that higher temperatures were required when diisopropylaminoborane reagent was used for reduction of 2k [35]. However, para-N, Ndimethylaminobenzonitrile (2l) provided the corresponding aminobenzylamine (3l), al- Reductions of benzonitrile with an electron-deficient group on the aromatic ring, for example, 4-trifluoromethylbenzonitrile (2h) underwent the reduction efficiently to the corresponding amine 3h in almost quantitative yield 97%, indicating that weak electronwithdrawing groups have no impact on this transformation. The electron-donating 2-and 4-methyl (2i-2j) did not inhibit the formation of amines (3i-3j), isolated in 99% and 86% yields in 3 h, respectively, although a slightly higher molar equivalent of 1a (2.5 equiv.) was necessary for complete reduction. Furthermore, the reaction of increased electron-donating 4-methoxybenzonitrile was converted to the desired product methanamine 3k in good yield (86%). It should be noted that higher temperatures were required when diisopropylaminoborane reagent was used for reduction of 2k [35]. However, para-N, N-dimethylaminobenzonitrile (2l) provided the corresponding aminobenzylamine (3l), albeit in diminished yield (51%) even when the reaction was extended to 24 h. The sluggish reactivity was attributed to the deleterious effect of the dimethylamino group, which might be exchanging borane with ammonia and rendering the reduction ineffective or due to the deactivation of the catalyst by complexation with the non-bonding electrons on nitrogen.
As a representative of a bulky aryl nitrile, 1-cyanonaphthalene (2m) was subjected to the new ammonia borane reduction under the optimized conditions when the corresponding 1-naphthylmethanamine (3m) was isolated in 83% yield. In addition, a representative heteroaromatic nitrile, 1-thiophenenitrile (2n) also proved highly amenable to the reaction conditions and afforded thiophenylmethanamine (3n) in 73% yield.
Reduction of alkyl nitriles is considered a challenge and numerous methodologies have failed to reduce aliphatic nitriles to primary amines, mainly due to a competitive deprotonation of the acidic α-proton prior to the reduction of the nitrile moiety. Accordingly, a series of straight chain and branched aliphatic nitriles were also included in the study. We were pleased to observe that the ammonia-borane/TiCl 4 reducing system is effective for the reduction of these nitriles as well (Figure 3). Our catalytic system does not induce deprotonation and, indeed, all the aliphatic nitriles were reduced, within 3 h, to their corresponding amines at room temperature in excellent yields. For example, acyclic octane-(2o) and dodecanenitrile (2p) were reduced to the amines 3o and 3p, respectively, in 89% and 92% yields. A branched nitrile, cyclohexanenitrile (2q) was also reduced, albeit, in a decreased 63% yield, to the corresponding methanamine 3q. Additionally, substituted, and unsubstituted 2-phenylethanenitriles were examined and all of them provided the corresponding amines in >90% yields. Thus, the parent 2-phenylethanenitrile (2r), 2-(4-fluorophenyl)ethanenitrile (2s), 2-(4-methoxyphenyl)ethanenitrile (2t) and 2-(2 ,4 -difluorophenyl)ethanenitrile (2u) with electron-neutral,-poor, and -rich substituents were converted to the corresponding amines 3r-3t in 90%-98% yields. Gratifyingly, even a highly hindered ethanenitrile derivative, such as α,α-diphenylethanenitrile (2v), was reduced to the corresponding β,β-diphenylethylamine (3v) in quantitative yield 95%.

General Information
Ammonia-borane [50] and other amine-boranes used in this study were prepared according to our earlier published procedures [48,49]. Other reagents and solvents as well

General Information
Ammonia-borane [50] and other amine-boranes used in this study were prepared according to our earlier published procedures [48,49]. Other reagents and solvents as well as the amines or amine-hydrochlorides to prepare the amine-boranes were purchased from Sigma-Aldrich or Oakwood chemicals. The nitriles, amines, sodium bicarbonate, and sodium borohydride were used as received. Anhydrous diethyl ether was prepared by distillation over sodium-benzophenone and anhydrous dichloromethane was prepared by distillation over calcium hydride and stored under nitrogen atmosphere. Thin layer chromatography (TLC) was performed on silica gel F60 plates and visualized under UV light or ceric ammonium molybdate solution. The structures of the product amines were confirmed by nuclear magnetic resonance (NMR) spectroscopy and measured in δ values in parts per million (ppm). 1 H NMR spectra of reduction products were recorded on a Bruker 400 MHz spectrometer at ambient temperature and calibrated against the residual solvent peak of CDCl 3 (δ = 7.26 ppm) as an internal standard. The 13 C NMR spectra were reported at 101MHz (297 K) and calibrated using CDCl 3 (δ = 77.0 ppm) as an internal standard. Coupling constants (J) are given in hertz (Hz), and signal multiplicities are described of NMR data as s = singlet, d = doublet, t = triplet, dd = doublet of doublets, dt = doublet of triplets, qd = quartet of doublets, q = quartet, quint and p = pentet, m = multiplet, and br = broad. 11 B, 1 H (300 MHz), and 13 C NMR (75 MHz) spectra of synthesized amineboranes were recorded at room temperature on a Varian INOVA or MERCURY 300 MHz NMR instrument. 11 B NMR spectra were recorded at 96 MHz and chemical shifts were reported relative to the external standard, BF 3 :OEt 2 (δ = 0 ppm).

General Procedure for the Preparation of Amines from Nitriles
The preparation of benzylamine from benzonitrile is typical. A 50 mL oven dried round bottom flask was charged with benzonitrile (1 mmol, 1 equiv.) and a magnetic stirring bar. The flask was sealed using a rubber septum. After purging the flask with nitrogen, dry diethyl ether (or other solvents) (3 mL) was added, and the solution was cooled to 0 • C with an ice bath. Subsequently, TiCl 4 (or other Lewis acids) (1 mmol, 1 equiv.) was added to the solution, dropwise via syringe if a liquid or by temporarily removing the septum (under a flow of nitrogen) if a solid. The septum was then carefully opened (under a flow of nitrogen) and ammonia borane (or other solid amine-borane) (2.0 mmol, 2.0 equiv.) was added slowly to the reaction mixture (liquid amine-boranes were added via a syringe). Upon complete addition, the reaction flask was again sealed with a septum. After stirring at 0 • C for 1 min, the reaction mixture was allowed to warm to room temperature, stirred and monitored by TLC for completion (disappearance of the starting nitrile), when the crude mixture was brought to 0 • C using an ice bath and quenched by the slow addition of cold 3 M HCl. The acidic solution was stirred for 30 min, made basic with 3 M NaOH to pH 11, transferred to a separatory funnel and extracted with diethyl ether (2 × 15 mL). The combined organic layers were washed with brine (1 × 3 mL), dried over anhydrous sodium sulfate, filtered through cotton, and concentrated under aspirator vacuum using a rotary evaporator. Any remaining traces of solvent were removed by subjecting to high vacuum for 30 min. The product amines were characterized using 1 H and 13 C NMR spectroscopy and compared with those reported in the literature and their references have been included. The spectra are available in Supporting Information. The results from the optimization experiments are shown in Table 1. Ammonia borane as the reductant and titanium chloride as the Lewis acid in diethyl ether solvent was established as the optimal procedure for subsequent reactions.

Conclusions
In conclusion, we have developed a simple protocol for the reduction of nitriles to afford primary amines using ammonia-borane as the reductant in the presence of one molar equivalent of TiCl 4 in diethyl ether at room temperature. A broad range of aromatic, heteroaromatic, benzylic, and aliphatic nitriles were efficiently reduced under this condition Molecules 2023, 28, 60 9 of 11 in moderate to very high yields. This reducing system affords negligible side products, and the workup of the reaction mixture is very simple. The reaction is believed to progress via the activation of the nitrile by titanium tetrachloride, followed by the hydroboration of the carbon nitrogen triple bond.
Author Contributions: Conceptualization, supervision, resources, project administration, and funding acquisition, P.V.R. Both authors have contributed to methodology, validation, investigation, data curation, writing-original draft preparation, review and editing; visualization, A.A.A. All authors have read and agreed to the published version of the manuscript.
Funding: This research was funded by the Herbert C. Brown Center for Borane Research of Purdue University.
Institutional Review Board Statement: Not applicable.

Informed Consent Statement: Not applicable.
Data Availability Statement: All of the 1 H, 11=9 F, and 13 C NMR spectra are available in the supporting information.