Abstract
Unlike the “parent” carborane anion CHB11H11−, halogenated carborane anions such as CHB11H5Br6− can be readily C-functionalized in high yield and purity, enhancing their utility as weakly coordinating anions.
B-halogenated icosahedral carborane anions such as CHB11H5X6− and CHB11X11− (X = halogen) (Figure 1) are particularly useful members of a class of exceptionally inert, weakly coordinating anions1–5 whose versatility might be further tailored by suitable C-derivatization chemistry. Long-chain hydrocarbon “tails” formed by C-alkylation should lower the lattice energies of salts, increase their solubilities in low dielectric solvents and allow better exploitation of reactive cations in catalysis. Applications in surfactant chemistry can also be envisioned. Similarly, C-fluorocarbon tails should improve the solubility of carborane ion pairs in fluorocarbon solvents, where catalytic applications have been reported.6 Attachment of a carborane anion to a polymer has allowed exploitation of immobilization chemistry in cation-selective sensor technology.7 C-arylation takes advantage of the unique scaffold of a carborane anion in rigid rod supramolecular chemistry.8,9
Figure 1.
Carborane anions CHB11H5X6− and CHB11X11− (gray = C, white = H, red = H or halogen, green = halogen).
Despite these promising applications, the C-derivatization chemistry of carborane anions has progressed rather slowly. Most work has been performed on the “parent” carborane anion, CHB11H11−, but is frequently hampered by modest yields and difficult separations from starting material. Being ionic rather than neutral, the chromatographic separation of different carborane anions is not trivial on a synthetic scale. The activation of CHB11H11− via C-lithiation with butyl lithium appears to be essentially quantitative10 but the partial regeneration of starting material during subsequent reactions with electrophiles is common, despite careful control of conditions. Alkylation of 1-Li-CB11H11− with alkyl halides gives mixed results. While the yields for methylations are frequently quite high,2,11–14 they drop to 63% for ethylation of CHB11H11− and are even lower for most other alkylations, silylations, phosphinations and metallations.2,10 The yields of C-monohalogenated products, 1-X-CB11H11−, have been raised to 81–96% by careful attention to conditions but chromatography purification is still recommended for most derivatives.11
We now report that when these C-functionalization reactions are performed on an already halogenated carborane anion such as CHB11H5Br6−, rather than on the unfunctionalized parent CHB11H11−, isolated yields are generally excellent and compound purity is sufficiently high that chromatographic purification is unnecessary. High yield C-cyanation of undecahalogenated carboranes has very recently been reported.15 These findings makes sense within the context of known carborane reactivity patterns. 2 C-lithiation is a deprotonation reaction and the C-H bond in B-halogenated carborane anions is more acidic than in the parent CHB11H11−. This is borne out by DFT theory at the B3LYP/6-311+G(d,p) level using dimethylsulfoxide in the IEFPCM solvation model. A 16 kcal.mole−1 decrease in acidity of the C-H bond between CHB11H5Br6− and CHB11H11− is calculated. Similarly, the electron withdrawing effect of B-halogen substituents will more favorably polarize the Li-C bond in 1-Li-CB11H5Br6− towards clean replacement of Li+ by an electrophile.
Scheme 1 summarizes the representative C-functionalization chemistry reported in the present work.
Scheme 1.
C-functionalization of the 1-Li–CB11H5Br6− anion (o denotes boron vertex).
For convenience and solubility reasons, the preferred starting material is the dilithio derivative Li[1-Li-CB11H5Br6], conveniently prepared from the Me3NH+ salt and 2 equiv. of butyl lithium.16 Trialkylsilyl, allyl, benzyl (including Merrifield resin), perfluoroaryl and perfluoroallenyl electrophiles, supplied as halides, illustrate the scope of the reactivity. When performed in a careful manner, isolated yields were >90%. Products were isolated by straightforward solvent extractions, avoiding tedious HPLC protocols. Excellent product purity was indicated both by clean NMR and IR spectroscopies as well as the absence of starting material in anionic mode electrospray mass spectrometry.
There is every reason to believe that the same efficient C-functionalization chemistry can be applied to other halogenated carborane anions.
Experimental
All experimental manipulations were carried out using standard drybox or Schlenk tube inert atmospheric techniques. Spectrochemical grade tetrahydrofuran (THF), acetonitrile and diethylether (Fischer) were distilled from sodium/benzophenone. Cs(CHB11H5Br6) was prepared as previously described. 1 Other commercially available chemicals were reagent grade and were degassed and dried over 4A molecular sieves for 1 week prior to use. All solvent evaporations were performed at reduced pressure on a rotary evaporator. 11B, 1H, 19F and 31P NMR spectra were obtained on a Bruker Avance 300 MHz spectrometer. 11B chemical shifts were referenced to BF3-OEt2 as an external reference in a capillary within the sample tube. 1H chemical shifts were measured relative to internal residual protons from the lock solvent and referenced to (CH3)4Si (0.0 ppm). 19F chemical shifts were referenced to CFCl3 (0.0 ppm). 31P chemical shifts were referenced to 85% aqueous H3PO4 (0.0 ppm) in a capillary. Electrospray ionization mass spectrometry data were obtained on an Agilent 6200 LC TOF in anionic mode. IR data were recorded on a Perkin Elmer Spectrum 100 Series as KBr pellets.
DFT calculations were carried out on CB11H12− and CHB11H5Br6− at B3LYP/6-311+G(d,p) level with the Gaussian 03 system of programs. The intrinsic gas phase acidities of the carborane anions were calculated according to the following thermodynamic heterolytic equilibrium, analogous to work17,18 on neutral species:
The absence of imaginary frequencies was taken as the criterion of a true minimum. The gas phase values of ΔGacidwere then recalculated using the IEFPCM solvation model with DMSO as a solvent. The resultant ΔGacid values were: CB11H12− = 325.75 and CHB11H5Br6− = 309.58 kcal.mol−1.
[Me3NH][closo-1-H-CB11H5Br6]
[Cs][CHB11H5Br6] (8.27 g, 11 mmole) was suspended in deionized water (325 mL) and the suspension heated to boiling to effect dissolution. The hot solution was filtered through a medium porosity frit and the filtrate reheated to boiling. [Me3NH]Cl (3.15 g, 33 mmole) was added, resulting in a thick white precipitate. The solution was allowed to stir at a gentle boil for 5 min. before being removed from heat. After cooling to room temperature, the suspension was further cooled in an ice bath. After 30 min., the suspension was filtered onto a 30 mL fine porosity frit and washed with aliquots of deionized water (300 mL). After air drying overnight, the product was transferred to a glass vial, dried overnight in a convection oven at 95° C (7.24 g (97%) and stored in a drybox.
Li[closo-1-Li-CB11H5Br6]
In a typical reaction, [Me3NH][closo-1-H-CB11H5Br6] (0.1–3 g) was dissolved in sufficient THF (~10 mL/g) to give a clear solution. Butyl lithium in hexanes (1.6 M, 2.1 molar equivalents) was then added dropwise over the course of minutes, maintaining a clear solution. If a white precipitate formed, additional THF was added to effect dissolution. The solvent volume was then reduced under vacuum by ca. 50% to remove butane and trimethylamine. THF was added to redissolve any white precipitate and the clear solution was immediately used for subsequent reactions.
[Me4N][closo-1-Me3Si-CB11H5Br6]
To a THF solution of Li[1-Li-CB11H5Br6] generated from [Me3NH][CHB11H5Br6] (0.1251g, 0.185 mmol), Me3SiCl (~.5 mL) was added dropwise over 5 min. After stirring for 24 hrs, the flask was removed from the inert atmosphere and the solvent evaporated. The resulting white solid was dissolved in water (45 mL) transferred to a beaker. A pellet of sodium hydroxide was added to neutralize any remaining Me3SiCl and acidic byproducts. The solution was heated to boiling and filtered while hot using a fine frit. The resulting clear solution was reduced in volume to 20 mL by boiling and Me4NCl (200 mg, 1.8 mmol) was added. A white precipitate formed which was isolated by filtration onto a fine frit, washed with water and vacuum dried (0.135 g, 88%). 1H NMR (d6-acetone): δ 3.46 [s, 12H, Me4N], 0.096 ppm [s, 9H, Me3Si]. 11B NMR: +0.49 [s, 1B, B(12)], −7.52 [s, 5B, B(7–11)], − 17.49 ppm, [d, 5B, B(2–6), JBH 503.2 Hz]. M/z: calcd. for Me3Si-CB11H5Br6− 688.7000, found 688.6996.
[Me4N][closo-1-iPr3Si-CB11H5Br6]
This was prepared from [Me3NH][CHB11H5Br6] (0.1467g, 0.217 mmol) and iPr3SiCl in a similar manner to [Me4N][closo-1-Me3Si-CB11H5Br6] above except that the crude product was extracted into diethylether (45 mL) for the initial filtration before dissolving in water (30 mL). Yield: 0.173g, 94%). 1H NMR (d6-acetone): δ 3.46 [s, 12H, Me4N], 1.17 [m, 21H, i-Pr3Si]. 11B NMR: 1.21 [s, 1B, B(12)], −7.70 [s. 5B, B(7–11)], −16.82 [d, 5B, B(2–6), JBH 498.3 Hz]. M/z: calcd. for (iPr)3Si-CB11H5Br6− 772.7939, found 772.7930.
[Me4N][closo-1-C3H5-CB11H5Br6]
To THF solution of Li[1-Li-CB11H5Br6] generated from [Me3NH][CHB11H5Br6] (0.1525g, 0.225 mmol) was added allyl iodide (~1mL) dropwise over the course of 5 mins. Light was excluded by wrapping the flask in Al foil and the reaction stirred for 24 hours before exposing to air and evaporating the solvent to dryness. The white solid was dissolved in water (45mL) yielding a clear solution which was transferred to a beaker, boiled and filtered hot through a fine frit. The resulting solution was reduced in volume to 20 mL before adding Me4NCl (200 mg, 1.8 mmol) which yielded a white precipitate which was isolated on a fine frit, washed with water and vacuum dried (0.152 g, 92%). 1H NMR (d6-acetone): δ 3.46 [s, 12H, Me4N], 2.57 [d, 2H, methylene, JH 7.2 Hz], 4.99 [m, 2H, vinyl], 5.63 [m, 1H, vinyl]. 11B NMR: −2.29 [s, 1B, B(12)], −8.85 [s. 5B, B(7–11)], −16.97 [d, 5B, B(2–6), JBH 522 Hz ]. M/z: calcd. for C3H5-CB11H5Br6− 656.6917, found 656.6913.
[Me4N][closo-1-CH2(C6H5)-CB11H5Br6]
This was prepared from Li[1-Li-CB11H5Br6] generated from [Me3NH][CHB11H5Br6] (0.1401g, 0.207 mmol) and benzyl bromide (~0.25 mL) in a similar manner to [Me4N][closo-1-C3H5-CB11H5Br6] above (0.243g, 90%). 1H NMR (d6-acetone): δ 3.46 [s, 12H, Me4N], 3.15 [s, 2H, methylene], 7.12 [m, 2H, phenyl], 7.27 [m, 3H, phenyl]. 11B NMR: −2.29 [s, 1B, B(12)], −8.79 [s. 5B, B(7–11)], −16.84 [d, 5B, B(2–6), JBH 501.8 Hz]. M/z: calcd. for C7H7-CB11H5Br6− 706.7074, found 706.7061.
Li[closo-1-(t-butyl)2P-CB11H5Br6]
This was prepared from Li[1-Li-CB11H5Br6] generated from [Me3NH][CHB11H5Br6] (0.2350g, 0.347 mmol) and di-t-butylchlorophosphine (.065 mL). The reaction was allowed to stir for 1 hour at room temperature followed by removal of ca. 85% of the solvent. Only NMR data were collected due to the extreme sensitivity of the product to oxygen. 1H NMR (d3-acetonitrile): δ 1.18 [d, 18 H, t-butyl, JPH 32.2 Hz]. 31P NMR: −151.40 [s, 1P]. 11B NMR: −1.13 [s, 1B, B(12)], −8.69 [s. 5B, B(7–11)], −15.20 [d, 5B, B(2–6), JBH 551.9 Hz].
[Me4N][closo-1-C6F5-CB11H5Br6]
This was prepared from Li[1-Li-CB11H5Br6] generated from [Me3NH][CHB11H5Br6] (0.2708 g, 0.400 mmol) and C6F6 (~1 mL) in a similar manner to [Me4N][closo-1-Me3Si-CB11H5Br6] above, except that the crude yellow product was extracted into diethylether (45 mL) for the initial filtration before dissolving in water (45 mL). (If a large excess of C6F6 is not used, multiple substitution occurs). Yield of off-white solid (0.323 g, 94%). 11B NMR: +.049 [s, 1B, B(12)], −8.55 [s. 5B, B(7–11)], −15.85 [d, 5B, B(2–6), JBH 495.9 Hz]. 19F NMR: −133.53 [broad, o-F], −154.70 [tt, p-F, JFF 22.9, 5.1 Hz], −162.9531 [m, m-F]. M/z: calcd. for C6F5-CB11H5Br6− 782.6446, found 782.6459.
[Me4N][closo-1-C6F11-CB11H5Br6]
This was prepared from Li[1-Li-CB11H5Br6] generated from [Me3NH][CHB11H5Br6] (0.3051 g, 0.451 mmol and perfluoro-1-hexene (0.150 g, 0.5 mmol) in a similar manner to [Me4N][closo-1-Me3Si-CB11H5Br6] above except that the crude brown product was extracted into diethylether (45 mL) for the initial filtration before evaporation and dissolution in water (45 mL). An additional extraction of the final product into dichloromethane (10 mL) followed filtration through a fine frit and evaporation of solvent gave an oil that solidified to a brown solid (0.3682 g, 91.1%). 11B NMR: +0.049 [s, 1B, B(12)], −8.55 [s. 5B, B(7–11)], −15.85 [d, 5B, B(2–6), JBH 487.5 Hz]. 19F NMR: −157.73 [d, alkene, JFF 144.5 Hz], −146.70 [d, alkene, JFF 137.37 Hz], - 127.70 [m, CF2], −118.75 [m, CF2], −84.67 [dd, CF2, JFF 31.7, 9.93 Hz], −80.73[t, CF3, JFF 8.6 Hz]. M/z: calcd. for C6F11CB11H5Br6− 896.6351, found 893.6390.
[Me4N][closo-1-C3F5-CB11H5Br6]
Extreme care should be exercised when confining hexafluoropropene (B.Pt. −28° C) to glass Schlenkware. All reactions must be maintained at dry ice temperatures to avoid explosion. In heavy walled Schlenkware, hexafluoropropene (~1mL) was pre-condensed at −78 °C then transferred to a heavy walled reaction vessel at dry ice temperature containing THF (5 mL) and Li[closo-1-Li-CB11H5Br6] generated from [Me3NH][CHB11H5Br6] (0.4510 g, 0.666 mmol). The resulting yellow solution was allowed to stir for 3 hours before gradually removing the excess hexafluoropropene under vacuum. The reaction was then allowed to warm to room temperature and the solvent removed under reduced pressure to give a colorless residue. This was extracted into diethylether (50 mL), filtered through a fine frit and the filtrate evaporated to dryness. The crude product was dissolved in water (35 mL) and Me4NCl (200 mg, 1.8 mmol) was added. The resulting white precipitate was isolated by filtration onto a fine frit, washed with water and vacuum dried (0.4608 g, 92.7%). 11B NMR: +1.82 [s, 1B, B(12)], −8.60 [s. 5B, B(7–11)], −17.52 [d, 5B, B(2–6), JBH 504.2 Hz]. 19F NMR: −67.56 [dd, CF2(sp3), JFF 23.5, 10.7 Hz], −122.60 [dm, CF, JFF 146.1, 23.7 Hz], −160.60 [d, CF2(sp2), JFF 145.3 Hz]. M/z: calcd. for C3F5CB11H5Br6− 746.6446, found 746.6458.
Cs[closo-1-Merrifield Peptide Resin-CB11H5Br6]
To THF solution [Li][1-Li-CB11H5Br6] generated from [Me3NH][CHB11H5Br6] (0.2330 g, 0.344 mmol) was added Merrifield’s Peptide Resin (0.185g, 1.95 mmol Cl/g, 1.05 equiv.) was added. The reaction was allowed to stir for 1 week yielding a yellow powder which was filtered onto a medium frit and was washed thoroughly with 3 aliquots of THF (50 mL). All filtrates were collected and the solvent evaporated. To the residue, deionized water (20 mL) was added. 11B NMR of this solution showed no detectable signals. To the filtrate, 2 drops of HNO3 were added, followed AgNO3 (1 mL, 0.5 M). The resulting white precipitate of AgCl was filtered onto a fine frit and dried (0.048g, 0.99 equiv.). The resin was suspended in water and CsCl (1 g, 6 mmol) added. The suspension was stirred for 1 day before collecting the resin by filtration onto a medium frit and washing with water. The faintly pale product was oven dried for 2 hours at 90° C before obtaining an IR spectrum.
Supplementary Material
Acknowledgement
We thank the National Science Foundation (CHE-0841428 and 0742001) for support. MJN was a MARCU* trainee supported by award T34GM062756 from the National Institute of General Medical Sciences.
Footnotes
Supporting Information Available: NMR and mass spectra for all new compounds (18 pages). This material is available free of charge on the internet at acs.pubs.org.
References
- 1.Reed CA. Accounts Chem. Res. 2009;43:121–128. doi: 10.1021/ar900159e. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Koerbe S, Schreiber PJ, Michl J. Chem. Rev. 2006;106:5208–5249. doi: 10.1021/cr050548u. [DOI] [PubMed] [Google Scholar]
- 3.Krossing I, Raabe I. Angew. Chem., Int. Ed. 2004;43:2066–2090. doi: 10.1002/anie.200300620. [DOI] [PubMed] [Google Scholar]
- 4.Reed CA. Accounts Chem. Res. 1998;31:133–139. [Google Scholar]
- 5.Strauss SH. Chem. Rev. 1993;93:927–942. [Google Scholar]
- 6.Douvris C, Ozerov OV. Science. 2008;321:1188–1190. doi: 10.1126/science.1159979. [DOI] [PubMed] [Google Scholar]
- 7.Qin Y, Bakker E. Anal. Chem. 2003;75:6002–6010. doi: 10.1021/ac034447c. [DOI] [PubMed] [Google Scholar]
- 8.Schwab PFH, Levin MD, Michl J. Chem. Rev. 1999;99:1863–1933. doi: 10.1021/cr970070x. [DOI] [PubMed] [Google Scholar]
- 9.Franken A, Kilner CA, Kennedy JDS. Inorg. Chem. Commun. 2003;6:1104–1108. [Google Scholar]
- 10.Jelínek T, Baldwin P, Scheidt WR, Reed CA. Inorg. Chem. 1993;32:1982–1990. [Google Scholar]
- 11.King BT, Koerbe S, Schreiber PJ, Clayton J, Nemcova A, Havlas Z, Vyakaranam K, Fete MG, Zharov I, Ceremuga J, Michl J. J. Am. Chem. Soc. 2007;129:12960–12980. doi: 10.1021/ja066247z. [DOI] [PubMed] [Google Scholar]
- 12.Ivanov SV, Rockwell JJ, Polyakov OG, Gaudinski CM, Anderson OP, Solntsev KA, Strauss SH. J. Am. Chem. Soc. 1998;120:4224–4225. [Google Scholar]
- 13.Xie Z, Tsang C-W, Sze ET-P, Yang Q, Chan DTW, Mak TCW. Inorg. Chem. 1998;37:6444–6451. doi: 10.1021/ic9805628. [DOI] [PubMed] [Google Scholar]
- 14.Kueppers T, Bernhardt E, Eujen R, Willner H, Lehmann C. Angew. Chem. 2007;46:6346–6349. doi: 10.1002/anie.200701136. [DOI] [PubMed] [Google Scholar]
- 15.Finze M, Sprenger J, Schaack B. Dalton Trans. 2010;39:2708–2716. doi: 10.1039/b922720b. [DOI] [PubMed] [Google Scholar]
- 16.Jelínek T, Plešek J, Heřmánek S, Štibr B. Coll Czech. Chem. Commun. 1986;51:819–829. [Google Scholar]
- 17.Koppel I, Burk P, Koppel I, Leito I, Sonoda T, Mishima M. J. Am. Chem. Soc. 2000;122:5114–5124. [Google Scholar]
- 18.Lipping L, Leito I, Koppel I. J. Phys. Chem. A. 2009;113:12972–12978. doi: 10.1021/jp905449k. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.


