|Home | About | Journals | Submit | Contact Us | Français|
A general, mild, and efficient method for the hydrolysis of organotrifluoroborates to unveil organoboronic acids using silica gel and H2O was developed. This method proved to be tolerant of a broad range of aryl-, heteroaryl-, alkenyl-, and alkyltrifluoroborates as well as structurally diverse aminomethylated organotrifluoroborates. As anticipated, electron-rich substrates provided the corresponding boronic acids more readily than electron-poor substrates, owing to the resonance stabilized difluoroborane intermediate. The method developed was expanded further for the conversion of organotrifluoroborates to the corresponding boronate esters.
Boronic acids1 are most commonly employed as nucleophilic partners in synthetically valuable organic transformations, particularly in Suzuki-Miyaura cross-coupling reactions.2 To a lesser extent, they have also been utilized as biological inhibitors and sensors as well as drug delivery agents.3 The recent FDA approval of the drug Velcade, a boronic acid-based proteasome inhibitor used in the treatment of multiple myeloma, has initiated immense interest in small molecules containing boronic acids in drug discovery efforts, wherein these agents can be screened for lead-like or drug-like properties (Figure 1).4
Even though boronic acids have been used extensively in the synthesis of organic molecules, they possess one major drawback. The tricoordinate nature of these compounds renders them susceptible to reaction with commonly employed organic reagents including bases, organic acids, nucleophiles, and oxidants, thus limiting the manner in which these valuable reagents are employed in synthetic schemes. Boronic acids are rarely carried intact through several synthetic steps;5 instead they are either purchased or prepared and then immediately transformed. Additionally, the late stage incorporation of boronic acids within densely functionalized molecules can be challenging because the conditions required to install boronic acids might not be compatible with the functional groups present within the molecule. One way to circumvent the problems associated with boronic acids is to utilize a more robust organoboron species that can be carried through several synthetic steps and then deprotected at an appropriate stage to unveil the boronic acid.
Over the last decade, organotrifluoroborates have received considerable attention as useful alternatives to boronic acids.6 Owing to their tetracoordinate nature, these organoboron species do not undergo undesirable side reactions with the commonly employed organic reagents mentioned above. Consequently, the organic substructure of simple organotrifluoroborates can be functionalized to build molecular complexity, while leaving the boron-carbon bond intact (Scheme 1). In this context, trifluoroborates have been successfully used as a protecting group for boronic acids.5c
The hydrolysis of organotrifluoroborates to the corresponding boronic acids using fluorophiles (e.g., SiCl4, TMSCl) has been previously reported.7 However, these fluorophiles are either toxic, difficult to handle, or highly reactive, and therefore not ideal for use in complex molecule synthesis. Recently, Hutton and Yuen published a two-step procedure for the hydrolysis of boronate esters via the intermediate organotrifluoroborate, allowing access to boronic acids. 8 For the hydrolysis of organotrifluoroborates, they employed either LiOH/acetonitrile or TMSCl/H2O. Although these methods proved effective, the study was limited in scope because only aryltrifluoroborate substrates were examined.
Current limitations associated with the hydrolysis of organotrifluoroborates, as well as the emerging importance of boronic acids in drug discovery efforts, prompted us to investigate a general, efficient, mild, and convenient method for the late stage hydrolysis of organotrifluoroborates. Herein, we report the hydrolysis of a broad range of simple aryl-, heteroaryl-, alkenyl-, and alkyltrifluoroborates along with structurally diverse aminomethylated organotrifluoroborates to the corresponding boronic acids using silica gel and H2O. The method developed was also extended to the direct conversion of organotrifluoroborates to the boronate esters.
Perrin and coworkers have reported that the rate of organotrifluoroborate solvolysis is governed by substituent groups.9 They pointed out that loss of a fluoride ion leads to a vacancy of the p-orbital on boron, providing a difluoroborane intermediate that should be stabilized by electron-rich substituent groups (Scheme 2) or destabilized by electron-withdrawing groups. Consequently, electron-rich organotrifluoroborates undergo solvolysis faster than those with electron-withdrawing groups.
Because substituent groups play a key role in the hydrolysis, we independently investigated conditions for hydrolysis of the electron-rich potassium 4-methoxyphenyltrifluoroborate and the electron-poor potassium benzothiophen-2-yltrifluoroborate substrates. Silica gel5e,10 was employed as a convenient, inexpensive and readily available fluorophile. Various solvents were examined including acetonitrile, acetone, DMF, DMSO, and H2O. The reactions were monitored via 11B NMR spectroscopy. Of the solvents screened, H2O proved to be the best solvent for the hydrolysis of potassium 4-methoxyphenyltrifluoroborate and benzothiophen-2-yltrifluoroborate, as the reactions were complete in 1 and 3 h, respectively. The efficacy of H2O in these transformations can be attributed to the enhanced solubility of 4-methoxyphenyltrifluoroborate and benzothiophen-2-yltrifluoroborate in H2O compared to the solvents mentioned above. Because silica gel proved to be a suitable fluorophile for this study, additional screening of fluorophiles was not necessary. The most efficient hydrolysis conditions employed 1 equiv of silica gel and a substrate concentration of 0.33 M in H2O.
The requisite potassium organotrifluoroborates were readily prepared by previously reported procedures.11 Initially, a broad range of electron-rich and electron-neutral aryltrifluoroborates was investigated (Table 1). In all cases, the hydrolysis was complete in 1 h, except for 1b and 1i (Table 1, entries 2 and 9), which required reaction times of 3 and 4 h, respectively. The reactivity of para-, meta-, and ortho-methoxyphenyltrifluoroborate and para-, meta-, and ortho-4-methylphenyltrifluoroborate were also examined. In both substrates, para-derivatives provided the desired products 1a and 1d in higher yields (Table 1, entries 1 and 4). Also, sterically hindered derivatives afforded the products 1c, 1f, and 1g in good yields (Table 1, entries 3, 6, and 7). We demonstrated that the reaction could be scaled to 5 mmol, providing product 1a in 83% yield (Table 1, entry 1).
Next, attention was turned to electron-poor aryltrifluoroborates. The conditions developed worked equally well for these substrates, providing the arylboronic acids in good yields (Table 2). As expected, aldehyde-, nitrile-, and nitro-containing phenyltrifluoroborates required a longer time (24 h) for complete hydrolysis (Table 2, entries 1–4). As mentioned above, the slow hydrolysis is attributed to the destabilized difluoroborane intermediate. Of note, heating the reactions at 50 °C led to the formation of protodeboronation products (boric acid was detected by 11B NMR at ~18 ppm). The reactivity of 4-halophenyltrifluoroborates was investigated, and it was determined that 4-fluorophenyltrifluoroborate required 4 h for full conversion, compared to only 1 h for 4-chloro- and 4-bromophenyltrifluoroborate (Table 2, entries 5–7). However, 4-fluorophenyltrifluoroborate provided the desired product 2e in higher yield (Table 2, entry 5).
To expand the utility of the developed conditions further, heteroaryl systems were investigated including thiophenyl-, pyrimidinyl-, pyridinyl-, benzothioyl-, and indolyl derivatives (Table 3). The heteroaryls afforded the desired boronic acids 3a–f in moderate to good yields, except for the electron-deficient 3-pyridyl- and 4-pyridyltrifluoroborates (data not shown). A few heteroaryls required a longer reaction time (24 h) for complete hydrolysis (Table 3, entries 2–5). Interestingly, thiophene-2-yltrifluoroborate converted to the corresponding boronic acid 3a in 3 h, while thiophene-3-yltrifluoroborate required 24 h for full conversion (Table 3, entries 1 and 2).
The scope of the general reaction conditions was extended to alkyl- and alkenyltrifluoroborates (Table 4). We successfully obtained both the 2-isobutylboronic acid (4a) and octylboronic acid (4b) in 73 and 67% yields, respectively (Table 4, entries 1 and 2). Next, the hydrolysis of alkenyltrifluoroborates was examined, and the conditions developed proved to be effective, providing the alkenylboronic acids 4c–4g in moderate to good yields (Table 4, entries 3–7). Also, both alkyl- and alkenyltrifluoroborates were deprotected in 1 h.
Of special note, during the study we observed that arylboronic acids were more stable than the corresponding heteroaryl-, alkyl-, and alkenylboronic acid counterparts. In our hands, many of the heteroaryl-, alkyl-, and alkenylboronic acids decomposed readily (as observed by the emergence of byproducts appearing at ~18 ppm in the 11B NMR spectrum), even when stored at low temperatures. To avoid extensive decomposition prior to recording of various spectra, these boronic acids were quickly isolated and dried in vacuo for several minutes. Immediate characterization was necessary to avoid contamination by various byproducts. Of the boronic acids examined, 5-indoleboronic acid (3g) was the most problematic substrate as it decomposed within minutes upon drying in vacuo. This study further verifies the advantages of robust potassium organotrifluoroborates, as these species could be kept indefinitely without significant decomposition.
The aminomethyl linkage is found in natural products and pharmaceutically active compounds. Recently, our group showed that aminomethyl-containing organotrifluoroborates could be accessed via the reductive amination of formyl-substituted organotrifluoroborates with a variety of amines.12 For the reductive amination various reducing agents including potassium formate/Pd(OAc)2, sodium triacetoxyborohydride, and pyridine-borane were used. These reducing agents were either used in excess or in stoichiometric amounts. Slightly modifying the previous procedure and utilizing the more stable and less expensive 5-ethyl-2-methylpyridine borane (PEMB)13 as a reducing agent, the desired reductive amination products were obtained (Table 5). Additionally, compared to the former reducing agents, only 0.5 equiv of the PEMB was needed. Potassium para- and meta-formylphenyltrifluoroborate underwent the reductive amination successfully with 4-aminotoluene-, morpholine, and piperazine, providing the desired products 5a–f in modest to excellent yields (Table 5).
The aminomethyl-substituted organotrifluoroborates were subsequently subjected to the hydrolysis conditions. For these reactions, using H2O as the solvent resulted in prolonged reaction times, which in turn led to protodeboronation products. This could be attributed to the low solubility of amine-containing organotrifluoroborates in H2O. This problem was circumvented with the use of EtOAc as a co-solvent, [H2O:EtOAc (1:1)]. With these slightly modified conditions in hand, aminomethyl-substituted organotrifluoroborates were deprotected in 3 and 5 h, and the desired products 6a–f were obtained in good yields (Table 6).
Boronate esters1a are more stable alternatives to boronic acids, and these species have found extensive use in organic synthesis.14 Matteson and coworkers showed that organotrifluoroborates can be converted to boronate esters via the intermediate dichloroborane by treatment of organotrifluoroborates with SiCl4 in the presence of MeOH followed by treatment with pinacol.6a Owing to the important complementary physical and chemical properties of organotrifluoroborates, we sought to determine whether the conditions developed herein would be applicable to their direct conversion to boronate esters. Treatment of 4-trifluoroboratoanisole with silica gel, H2O, and pinacol afforded the desired boronate ester 7a in 81% yield (Table 7, entry 1). Encouraged by this result, other diols including chiral dimethyl d(−)- and l(+)-tartrates and neopentyl glycol were examined. In each case the products 7b–d were obtained in good yields (Table 7, entries 2–4). Furthermore, octyltrifluoroborate and benzothiophene-2-yltrifluoroborate were successfully converted to boronate esters 7e and 7f in good yields (Table 7, entries 6 and 7).15
In conclusion, a general, mild and efficient method was developed for the hydrolysis of aryl-, heteroaryl-, alkyl-, and alkenyltrifluoroborates, along with structurally diverse aminomethylated organotrifluoroborates, to the corresponding boronic acids. The rate of hydrolysis was influenced by the substituent groups, wherein electron-rich substrates underwent hydrolysis faster than electron-poor substrates. We demonstrated that the method developed could be extended to the direct formation of boronate esters from organotrifluoroborates.
To a 50 mL round bottom flask containing a mixture of potassium 4-methoxyphenyltrifluoroborate (456 mg, 3.0 mmol) and silica gel (180 mg, 3.0 mmol) under N2 was added H2O (9 mL) in one portion. The reaction was stirred at rt until 11B NMR indicated completion of the reaction (~1 h). The reaction mixture was filtered to remove silica gel, and the filter cake was thoroughly rinsed with ethyl acetate.17 The aqueous and organic layers were separated, and the aqueous layer was extracted with ethyl acetate (2 × 15 mL) (Table 2, electron-deficient arylboronic acids were washed with brine). The combined organic layers were dried (MgSO4), filtered, concentrated, and dried in vacuo overnight to afford the desired pure product in 83% yield (0.38 g, 2.49 mmol) as a white solid. 1H NMR (500 MHz, DMSO-d6) δ 7.82 (d, J = 8.4 Hz, 2H), 6.94 (d, J = 8.5 Hz, 2H), 3.77 (s, 3H). 13C NMR (125.8 MHz, DMSO-d6) δ 161.6, 137.6, 136.0, 114.0, 55.8.
To a vial containing potassium 2-fluoro-5-formylphenyltrifluoroborate (200 mg, 0.87 mmol) and KHF2 (270 mg, 3.48 mmol) in MeOH (4.4 mL) was added 4-aminotoluene (140 mg, 1.3 mmol). The reaction mixture was stirred for 3 h at rt. 5-Ethyl-2-methylpyridine borane (PEMB, 0.06 mL, 0.43 mmol) was then added, and stirring was continued for 3 h. The solvent was removed in vacuo, and the resulting solid was rinsed with hexane (2 × 5 mL). The crude solid was purified by continuous Soxhlet extraction (3 h) with acetone (125 mL). The collected solvent was concentrated and then precipitated with acetone/hexane to afford the desired pure product in 96% yield (279 mg, 0.83 mmol) as a light yellow solid. mp: > 200°C. 1H NMR (360.1 MHz, DMSO-d6) δ 7.31 (s, 1H), 7.00 (s, 1H), 6.82 (d, J = 7.2 Hz, 2H), 6.69 (t, J = 8.6 Hz, 1H), 6.46 (d, J = 8.2 Hz, 2H), 5.78 (brs, 1H), 4.08 (d, J = 5.7 Hz, 2H), 2.10 (s, 3H). 13C NMR (125.8 MHz, DMSO-d6) δ 166.6, 147.6, 134.5, 134.11, 130.1, 126.9, 124.6, 114.3, 113.2, 47.6, 21.0. 19F NMR (470.8 MHz, Acetone-d6) δ-113.1, - 139.8. 11B NMR (128.4 MHz, Acetone-d6) δ 2.1. FT-IR (KBr) 3393, 3020, 1616, 1478, 1274, 1156, 975 cm−1. HRMS (ESI) m/z calcd. for C14H13BF4N (M-K) 282.1077, found 282.1074.
To a 20 mL vial containing a mixture of potassium 2-fluoro-5-[(4-tolylamino)methyl]-phenyltrifluoroborate (60 mg, 0.187 mmol) and silicon dioxide (11 mg, 0.187 mmol) was added H2O:ethyl acetate (1.1 mL:1.1 mL) in one portion. The reaction was stirred at rt until 11B NMR indicated completion of the reaction (6 h). The reaction mixture was filtered to remove silica gel, and the filter cake was thoroughly rinsed with ethyl acetate. The aqueous and organic layers were separated, and the aqueous layer was extracted with ethyl acetate (2 × 5 mL), and the combined organic layers were dried (MgSO4), filtered, concentrated, and dried in vacuo to afford the desired pure product in 73% yield (35.3 mg, 0.14 mmol) as a white solid. mp: 96–99 °C. 1H NMR (360 MHz, Acetone-d6) δ 7.76 (m, 1H), 7.46 (m, 1H), 7.05 (s, 2H), 6.99 (t, J = 9.5 Hz, 1H), 6.88 (d, J = 8.3 Hz, 2H), 6.56 (d, J = 8.3 Hz, 2H), 5.65 (brs, 1H), 4.30 (s, 2H), 2.15 (s, 3H). 13C NMR (90.5 MHz, Acetone-d6) δ 136.6, 136.5, 132.9, 132.8, 130.8, 116.2, 115.9, 114.3, 48.5, 21.0. 19F NMR (470.8 MHz, Acetone-d6) δ-110.7. 11B NMR (128.4 MHz, Acetone-d6) δ 27.9. FTIR (neat) 3422, 2977, 1689, 1611, 1423, 1004 cm−1. HRMS (ESI) m/z calcd. for C14H16BFNO2 (MH+) 260.1258, found 260.1248.
To a 50 mL round bottom flask containing a mixture of potassium 4-methoxyphenyltrifluoroborate (456 mg, 3.0 mmol), pinacol (350 mg, 3.0 mmol) and silicon dioxide (180 mg, 3.0 mmol) under N2 was added H2O (9 mL) in one portion. The reaction was stirred at rt until 11B NMR indicated completion of the reaction (~30 min). The reaction mixture was filtered to remove silica gel, and the filter cake was thoroughly rinsed with ethyl acetate. The aqueous and the organic layers were separated, and the aqueous layer was extracted with ethyl acetate (3 × 15 mL), and the combined organic layers were dried (MgSO4), filtered, concentrated, and dried in vacuo overnight to afford the desired pure product in 81% yield (0.57 g, 2.42 mmol) as a colorless oil. 1H NMR (500 MHz, CDCl3) δ 7.82 (d, J = 8.7 Hz, 2H), 6.92 (d, J = 8.7 Hz, 2H), 3.80 (s, 3H), 1.36 (s, 12H). 13C NMR (125.8 MHz, CDCl3) δ 162.2, 136.5, 113.3, 83.4, 54.9, 24.8. 11B NMR (128.4 MHz, CDCl3) δ 29.19. FT-IR 2978, 1606, 1360, 831, 655 cm−1. HRMS (ESI) m/z calcd. for C13H20BO3 (MH+) 235.1506, found 235.1509.
This work was generously supported by the National Institutes of Health (GM086209, GM035249), BASF, a Conselho Nacional de Desenvolvimento Cientitico e Tecnologico (CNPq) Graduate Research Fellowship to L.N.C, and the ACS Division of Organic Chemistry (Summer Undergraduate Research Fellowship, sponsored by Wyeth) to L.E.K. We also acknowledge Aldrich, BoroChem, and Frontier Scientific for their donation of the boronic acids. Dr. Rakesh Kohli (University of Pennsylvania) is acknowledged for obtaining HRMS data. We thank Dr. Noel Ellis and Professor Donna Huryn (University of Pennsylvania) for helpful discussions.
Supporting Information Available: Experimental procedures, spectral characterization, and copies of 1H, 13C, 11B, and 19F spectra for all compounds prepared by the method described. This material is available free of charge via the Internet at http://pubs.acs.org.