Separating electrophilicity and lewis acidity: The synthesis, characterization, and electrochemistry of the electron deficient tris (aryl)boranes B(C6F5)3-n(C6Cl 5)n (n = 1-3)

Article abstract of DOI:10.1021/ja205037t

A new family of electron-deficient tris(aryl)boranes, B(C₆F ₅)_3-n(C₆Cl₅)_n (n = 1-3), has been synthesized, permitting an investigation into the steric and electronic effects resulting from the gradual replacement of C₆F₅ with C₆Cl₅ ligands. B(C₆F₅) ₂(C₆Cl₅) (3) is accessed via C ₆Cl₅BBr₂, itself prepared from donor-free Zn(C₆Cl₅)₂ and BBr₃. Reaction of C₆Cl₅Li with BCl₃ in a Et₂O/hexane slurry selectively produced B(C₆Cl₅)₂Cl, which undergoes B-Cl exchange with CuC₆F₅ to afford B(C ₆F₅)(C₆Cl₅)₂ (5). While 3 forms a complex with H₂O, which can be rapidly removed under vacuum or in the presence of molecular sieves, B(C₆Cl₅) ₃ (6) is completely stable to refluxing toluene/H₂O for several days. Compounds 3, 5, and 6 have been structurally characterized using single crystal X-ray diffraction and represent the first structure determinations for compounds featuring B-C₆Cl₅ bonds; each exhibits a trigonal planar geometry about B, despite having different ligand sets. The spectroscopic characterization using ¹¹B, ¹⁹F, and ¹³C NMR indicates that the boron center becomes more electron-deficient as n increases. Optimized structures of B(C₆F ₅)_3-n(C₆Cl₅)_n (n = 0-3) using density functional theory (B3LYP/TZVP) are all fully consistent with the experimental structural data. Computed ¹¹B shielding constants also replicate the experimental trend almost quantitatively, and the computed natural charges on the boron center increase in the order n = 0 (0.81) < n = 1 (0.89) < n = 2 (1.02) < n = 3 (1.16), supporting the hypothesis that electrophilicity increases concomitantly with substitution of C ₆F₅ for C₆Cl₅. The direct solution cyclic voltammetry of B(C₆F₅)₃ has been obtained for the first time and electrochemical measurements upon the entire series B(C₆F₅)_3-n(C₆Cl ₅)_n (n = 0-3) corroborate the spectroscopic data, revealing C₆Cl₅ to be a more electron-withdrawing group than C₆F₅, with a ca. +200 mV shift observed in the reduction potential per C₆F₅ group replaced. Conversely, use of the Guttmann-Beckett and Childs' methods to determine Lewis acidity on B(C₆F₅)₃, 3, and 5 showed this property to diminish with increasing C₆Cl₅ content, which is attributed to the steric effects of the bulky C₆Cl₅ substituents. This conflict is ascribed to the minimal structural reorganization in the radical anions upon reduction during cyclic voltammetric experiments. Reduction of 6 using Na_(s) in THF results in a vivid blue paramagnetic solution of Na⁺ [6]^?-; the EPR signal of Na⁺[6]^?- is centered at g = 2.002 with a( ¹¹B) 10G. Measurements of the exponential decay of the EPR signal (298 K) reveal [6]^?- to be considerably more stable than its perfluoro analogue.

Full text of DOI:10.1021/ja205037t

ARTICLE
pubs.acs.org/JACS
Separating Electrophilicity and Lewis Acidity: The Synthesis,
Characterization, and Electrochemistry of the Electron Deficient
Tris(aryl)boranes B(C6F5)3Àn(C6Cl5)n (n = 1À3)
Andrew E. Ashley,*,†,‡ Thomas J. Herrington,† Gregory G. Wildgoose,§ Hasna Zaher,† Amber L. Thompson,†
Nicholas H. Rees,† Tobias Kr€amer,† and Dermot O’Hare*,†
†Chemistry Research Laboratory, Department of Chemistry, University of Oxford, Mansfield Road, Oxford, OX1 3TA, United Kingdom
§School of Chemistry, University of East Anglia, Norwich, NR4 7TJ, United Kingdom
S Supporting Information
b
ABSTRACT: A new family of electron-deficient tris(aryl)-
boranes, B(C6F5)3Àn(C6Cl5)n (n = 1À3), has been synthesized,
permitting an investigation into the steric and electronic effects
resulting from the gradual replacement of C6F5 with C6Cl5
ligands. B(C6F5)2(C6Cl5) (3) is accessed via C6Cl5BBr2, itself
prepared from donor-free Zn(C6Cl5)2 and BBr3. Reaction of
C6Cl5Li with BCl3 in a Et2O/hexane slurry selectively produced B(C6Cl5)2Cl, which undergoes BÀCl exchange with CuC6F5 to
afford B(C6F5)(C6Cl5)2 (5). While 3 forms a complex with H2O, which can be rapidly removed under vacuum or in the presence of
molecular sieves, B(C6Cl5)3 (6) is completely stable to refluxing toluene/H2O for several days. Compounds 3, 5, and 6 have been
structurally characterized using single crystal X-ray diffraction and represent the first structure determinations for compounds
featuring BÀC6Cl5 bonds; each exhibits a trigonal planar geometry about B, despite having different ligand sets. The spectroscopic
characterization using 11B, 19F, and 13C NMR indicates that the boron center becomes more electron-deficient as n increases.
Optimized structures of B(C6F5)3Àn(C6Cl5)n (n = 0À3) using density functional theory (B3LYP/TZVP) are all fully consistent
with the experimental structural data. Computed 11B shielding constants also replicate the experimental trend almost quantitatively,
and the computed natural charges on the boron center increase in the order n = 0 (0.81) < n = 1 (0.89) < n = 2 (1.02) < n = 3 (1.16),
supporting the hypothesis that electrophilicity increases concomitantly with substitution of C6F5 for C6Cl5. The direct solution
cyclic voltammetry of B(C6F5)3 has been obtained for the first time and electrochemical measurements upon the entire series
B(C6F5)3Àn(C6Cl5)n (n = 0À3) corroborate the spectroscopic data, revealing C6Cl5 to be a more electron-withdrawing group than
C6F5, with a ca. +200 mV shift observed in the reduction potential per C6F5 group replaced. Conversely, use of the Guttmann-
Beckett and Childs’ methods to determine Lewis acidity on B(C6F5)3, 3, and 5 showed this property to diminish with increasing
C6Cl5 content, which is attributed to the steric effects of the bulky C6Cl5 substituents. This conflict is ascribed to the minimal
structural reorganization in the radical anions upon reduction during cyclic voltammetric experiments. Reduction of 6 using Na(s)
in THF results in a vivid blue paramagnetic solution of Na+ [6]•À; the EPR signal of Na+[6]•À is centered at g = 2.002 with a(11B)
10G. Measurements of the exponential decay of the EPR signal (298 K) reveal [6]•À to be considerably more stable than its
perfluoro analogue.
’ INTRODUCTION
even at elevated temperatures, combined with a substantial steric
bulk”.5b A more recent role for this molecule is as a Lewis acid
partner in Frustrated Lewis Pair (or FLP) chemistry6 wherein
steric preclusion from adduct formation with a Lewis base leads
to unusual reactivity such as H2 heterolysis and small molecule
activation with alkenes/alkynes,7 CO28 and N2O.9 The enhanced
reactivity of H2 in the presence of B(C6F5)3-derived FLPs has
been utilized to effect the metal-free hydrogenation of CO2 to
CH3OH10 and of organics such as nitriles/imines to their
corresponding amines.11
In 1963, Massey and co-workers reported the synthesis of
tris(pentafluorophenyl)borane, noting its tendency to form a
variety of strongly bound adducts with phosphines, ammonia and
ethers.1 B(C6F5)3 has since found numerous applications in both
organic chemistry (e.g. silylation of alcohols, hydrosilylation of
ketones and imines, reductive cleavage of alcohols and ethers)2
and inorganic chemistry (e.g. synthesis of weakly coordinating
anions, anion-binding, activator in transition metal-mediated R-
olefin polymerizations).3 These attributes are related to its
powerful Lewis acidity, which has been measured to be inter-
mediate between BF3 and BCl3.4 However, unlike these gaseous
species, it is a thermally robust solid (due to its strong BÀC and
CÀF bonds) and is water-tolerant, lending itself to ease of
handling.5 Indeed, B(C6F5)3 has been described as the “ideal
boron-based Lewis acid, due to its high acid strength and stability,
The strength of a Lewis acid has been shown to correlate
with chemical activity in certain processes, for example, the
rate of epoxide ring-opening.4a,12 Simple steric and electronic
Received: June 1, 2011
Published: July 25, 2011
r
2011 American Chemical Society
14727
dx.doi.org/10.1021/ja205037t J. Am. Chem. Soc. 2011, 133, 14727–14740
|

Journal of the American Chemical Society
ARTICLE
modification of the Lewis acid B(C6F5)3 has resulted in changes
in catalytic reactivity. For instance, employing the bulkier B-
(C6F5)2(Mes) (Mes = 2,4,6-Me3C6H2) has been recently shown
to lead to orthogonal reactivity patterns in the FLP-mediated
reduction of imines,13 and use of B(C6F5)2Ph has altered the
predominant mechanism in the allylstannation of aldehydes,
relative to the perfluorophenyl counterpart.2d The majority of
powerful boron-based Lewis acid systems have been formulated
upon electron-withdrawing fluoroaryl ligands,14 which impart
strong acidity at the acceptor site. However, while the high
electronegativity of F (χPauling = 3.98) ensures potent inductive
withdrawal via the σ-bonds from the boron center, mesomeric
donation from ortho- and para-F lone pairs is particularly
effective (2p-2p overlap); this results in significant back-donation
from the aromatic π-system into the acceptor orbital and can
attenuate Lewis acidity. The degree to which π-electrons from
aryl substituents are involved with the aromatic nucleus may be
quantified using the Hammett equation, which derives a free
energy relationship between reaction rates and equilibrium
constants for various meta- and para- substituted aromatic
compounds;15 more positive values denote increasingly powerful
electron-withdrawing groups, in the absence of steric effects.
While Cl is not as electronegative as F (χPauling = 3.16), its
Hammett parameter (σparaCl = 0.227) is substantially greater
(σparaF = 0.062) as a result of weaker (3p-2p) π-overlap with the
aromatic nucleus. Accordingly, substitution of C6F5 for perchloro-
phenyl groups (C6Cl5) should increase the inherent electron-
withdrawing properties of the ligands in the resultant organo-
boranes, a factor which should result in an increase in Lewis
acidity when considered alone.
Scheme 1. Synthesis of B(C6Cl5)(C6F5)2 (3)
decompose only slowly between 0 and À10 °C, and its stability is
thus appreciably greater than that of C6F5Li.18
Retrosynthetic analysis of the target compounds B(C6F5)n-
(C6Cl5)3Àn (n = 1, 2) reveals two plausible routes to their
formation, depending on which perhaloaryl groups are installed
onto boron first. It was reasoned that BÀX metathesis with the
smaller C6F5 group on a B(C6Cl5)nX3Àn (n = 1, 2; X = Cl, Br)
intermediate should be performed at the final stage of the
synthesis in order to minimize potential side-reactions, that is,
para-F substitution (SNAr) on C6F5 rings, which has been
documented for B(C6F5)3 in reaction with bulky nucleophiles.19
(C6Cl5)BCl2 has been previously synthesized from C6Cl5SnMe3
and BCl3(g) at 120 °C.20 Attempts to perform this reaction using
a less hazardous solution-phase protocol (BCl3 is available
commercially in heptane) surprisingly led to no reaction. An
alternative route, avoiding toxic organotin species, was thus
developed. Realizing the potential of Zn(C6F5)2 to selectively
transfer C6F5 groups to organoboron halides,21 base-free Zn-
(C6Cl5)2 (1) was synthesized in a facile manner from C6Cl5Li
and ZnCl2. 1 is poorly soluble in nondonor solvents yet reaction
with excess BBr3 in toluene leads to transfer of both aryl groups
from Zn, cleanly affording C6Cl5BBr2 (2) in good yield and on a
multigram scale (Scheme 1); using the less vigorous Lewis acid
BCl3 leads to a significantly slower reaction.
In this study, the synthesis of a systematic series of new
(perhaloaryl)borane Lewis acids B(C6F5)3Àn(C6Cl5)n (n = 1À3)
is reported, to comprehensively examine the effects on the
spectroscopic and electrochemical properties upon replacement
of C6F5 with C6Cl5 moieties. In addition, the trends in the Lewis
acidity of these boranes with the parent B(C6F5)3 are examined,
using established donorÀacceptor methods, which are con-
trasted with their electrochemical behavior.
2 is a highly moisture sensitive solid, producing HBr fumes in
air; nonetheless it is more easily handled than C6F5BBr2, which is
an oil. Attempts to react C6F5Li or C6F5MgBr with 2 in Et2O led
to products of solvent cleavage, identified by several quartet
resonances between 3 and 4 ppm and corresponding triplets at
higher field (1H NMR), and by MS which revealed ion peaks
attributable to B(C6Cl5)(C6F5)OEt and B(C6Cl5)(OEt)2; it was
subsequently discovered that 2 reacts with Et2O alone to form
various B-OEt containing species. J€akle et al. have reported the
enhanced selectivity exhibited by CuAr (Ar = C6F5, Mes) in
conjunction with BX3 (X = Cl, Br) to form ArBX2 or Ar2BX, in
comparison with their lithium or Grignard analogues; further-
more, these reactions can be conducted in donor-free solvents
such as CH2Cl2 and aromatics.17c Gratifyingly, metathesis using
two equivalents of CuC6F5 with 2 in toluene afforded B(C6Cl5)-
(C6F5)2 (3) as a white powdery solid, in excellent yield (81%)
after vacuum sublimation. 3 is moderately soluble in aliphatic
hydrocarbons, yet highly so in aromatics and chlorinated sol-
vents. It is also stable to oxygen, but binds H2O forming the
’ RESULTS AND DISCUSSION
Synthesis. Commonly used reagents for the introduction of
aryl substituents onto main group metal halides are organo-
lithium and Grignard reagents.16 For example, the synthesis of
B(C6F5)3 is achieved by treating a boron trihalide (typically
BF3 OEt2 or BCl3) with either C6F5MgBr or C6F5Li;1a the latter
3
requires caution in handling since it is can become explosive
above À30 °C. Use of donor solvents, e.g. Et2O or THF are
usually avoided if the anionic synthon can be prepared in
nonpolar media since, if the borane coordinates to the solvent,
a final sublimation step may be required to isolate the free Lewis
acid, e.g. C6F5MgBr and Et2O BF3 react in Et2O to form diethyl
3
etherate complex, Et2O B(C6F5)3. Unfortunately, highly polar
3
organometallic reagents often exhibit poor selectivity and cannot
be used to synthesize partially perfluoroarylated boranes, e.g.
B(C6F5)2Cl, irrespective of the stoichiometry used. For this
purpose, less reactive and more selective reagents such as
Me2Sn(C6F5)2 and CuC6F5 have been used in conjunction with
BX3 (X = Cl, Br) to exercise greater control in halide metathesis
reactions.17 Perchlorophenyllithium (C6Cl5Li), synthesized in a
dative complex H2O [3] as shown by 19F NMR,22 which is a
3
sensitive tool in determining the coordination environment around
n
facile manner from BuLi and C6Cl6 in Et2O, is reported to
the boron center in C6F5-substituted boranes.23 Interestingly, and
14728
dx.doi.org/10.1021/ja205037t |J. Am. Chem. Soc. 2011, 133, 14727–14740

Journal of the American Chemical Society
ARTICLE
Scheme 4. Synthesis of B(C6Cl5)3 (6)
Figure 1. Reversible coordination of H2O by 3.
in vacuo. 6 is insoluble in aliphatic hydrocarbons, slightly so in
aromatics and moderately in CH2Cl2. It is remarkably robust,
remaining unchanged at temperatures up to 250 °C and does not
sublime, even under high-vacuum (1 Â 10À6 mbar), at these
temperatures. Refluxing 6 in a toluene/H2O mixture (1:1) for
several days led to quantitative reclamation of the compound and
thus demonstrates an impressive hydrolytic stability.
Scheme 2. Synthesis of (C6Cl5)2BCl (4)
Structural Characterization. Crystals of 3 suitable for single
crystal X-ray structure determination were grown through slow-
cooling of a saturated toluene solution to À35 °C,24 while for
Scheme 3. Synthesis of B(C6Cl5)2(C6F5) (5)
6 toluene clear prisms were obtained from a saturated solution
3
in boiling toluene (in air) that was slowly cooled to ambient
temperature. Slow evaporation of a toluene solution of 5 afforded
small pale-yellow blocks. Crystallographic data are enclosed in
Table 1 while the solid-state structures are shown in Figures 2, 3
and 4 for 3, 5 and 6 respectively.
Despite finding widespread use as a Lewis acid in many
chemical applications, for example, activator for metallocene-
mediated olefin polymerizations, no structural data exist for
B(C6F5)3. Compounds B(C6F5)3Àn(C6Cl5)n (n = 1À3) repre-
sent the first structurally characterized compounds featuring the
BÀC6Cl5 motif; all three crystallize in centrosymmetric space
groups so that both left and right-handed “propellers” are
present.25 The coordination environment about B is trigonal
planar as judged by the almost zero deviation of this atom from
the plane of three ipso-C atoms, in spite of 3, 5 and 6 having
different ligand sets.26,27 Table 2 shows selected bond lengths and
torsion angles for all three compounds. The CÀCl bond lengths
vary little and are very similar to those in C6Cl6 (range 1.713-
(2)À1.724(3) Å);28 the longest are found in the ortho position
and are likely to reflect the high steric crowding at these sites.
In 3 the two C6F5 rings are inequivalent, which is also found in
the analogous ArB(C6F5)2 (Ar = C6H5, Mes) species;29 the
torsion angles between the best plane of a C6F5 ring and the
plane of the remaining B-ipso(C6X5)2 (X = Cl, F) fragment (see
Figure 5 for definition) best represent this difference. The lowest
energy conformation (minimizing nonbonding interactions be-
tween ortho-substituents on different rings) would be a ‘propel-
ler’ for steric reasons, for which each ArX∧BAr2 = 60° (ArX =
C6Cl5, C6F5).30 However, π-donation from the C6F5 rings into
the vacant B p-orbital lowers the energy of the molecule and a
compromise is achieved; accordingly the aryl group with the
smallest torsion angle also has the shortest B-ipsoC bond for each
ArB(C6F5)2 (Ar = C6H5, Mes, C6Cl5) examined. By comparison,
the much larger torsion angle for C6Cl5 in 3 is likely to be due to
the poorer π-donor ability of this substituent relative to C6F5, in
addition to its larger size; for 6, the angles now approach D3
symmetry.
in contrast to H2O B(C6F5)3, the H2O molecule can be rapidly
3
and reversibly removed under vacuum, or in the solution phase
upon addition of molecular sieves, forming 3 (Figure 1). This
difference is likely to be the result of increased steric bulk around
the boron center due to the ortho-Cl substituents, as opposed to
electronic effects (vide infra), leading to a longer and thus weaker
BÀO interaction.
The synthesis of B(C6Cl5)2(C6F5) required a previously
unreported (C6Cl5)2BX (X = Cl, Br) reagent, and surprisingly,
it was found that slow addition of hexane to C6Cl5Li (Et2O
solution) resulted in the precipitation of a flocculent solid,
presumably C6Cl5Li (OEt2)n, reaction of which with BCl3
3
furnished base-free (C6Cl5)2BCl (4) in 54% yield (Scheme 2).
In the absence of this heterogenization step significant Et2O-
1
cleavage products, as ascertained by H NMR, were obtained.
Conversely, the fluorinated analogue (C6F5)2BCl cannot be
selectively obtained from C6F5M (M = Li or MgBr); it is
prepared most expediently through Sn-aryl cleavage of Me2Sn-
(C6F5)2 with BCl3 at elevated temperatures.17b It is anticipated
that the greater steric bulk of C6Cl5 compared with C6F5 is likely
to permit greater control in this metathesis reaction. 4 is a pale-
orange crystalline solid which fumes slowly in moist air (releasing
HCl) and demonstrates a moderate solubility in chlorinated
solvents, yet poor in aliphatics and aromatics.
Following the successful use of CuC6F5 in the formation of 3,
its reaction with 4 led to the production of B(C6Cl5)2(C6F5) (5)
in good yield, albeit after 72 h (Scheme 3). The slower rate of this
transformation likely reflects the increased bulk of the starting
haloborane 4, and the replacement of a stronger BÀCl bond as
opposed to BÀBr, in 2.
The synthesis of tris(pentachlorophenyl)borane was achieved
through the stoichiometric addition of BCl3 to C6Cl5Li (1:3),
using hexane cosolvent as in the preparation of 4, which is
presumed to be an intermediate in the reaction. Accordingly
B(C6Cl5)3 (6) was isolated in moderate yield (Scheme 4).
Compound 6 is air-stable, allowing for a facile workup by
quenching unreacted BCl3 and C6Cl5Li with H2O and extracting
with CH2Cl2 using ‘open bench’ techniques. Recrystallization
NMR Spectroscopy. The 19F, 11B and 13C NMR spectral data
for 3, 5 and 6 in the solution phase have been summarized
and compared with B(C6F5)3 (Tables 3 and 4). Compounds 3
and 5 display three resonances in their 19F NMR spectra in an
intensity ratio 2:1:2 for the corresponding ortho, para and meta
fluorine environments (to higher field respectively) for the C6F5
rings. The difference between shifts for the para and meta
environments, Δδm,p, are between 17 and 18.5 ppm and, in
from boiling toluene afforded 6 (toluene) as a pale-yellow
3
microcrystalline solid; the solvent may be removed upon heating
14729
dx.doi.org/10.1021/ja205037t |J. Am. Chem. Soc. 2011, 133, 14727–14740

Journal of the American Chemical Society
ARTICLE
Table 1. Crystallographic Data and Structure Refinement for Compounds 3, 5 and 6 Toluene
3
3
5
6 toluene
3
Empirical formula
Formula weight
Temperature
C
18BCl5F10
C18BCl10F5
676.53
C25H8BCl15
850.94
594.25
150 K
150 K
150 K
Wavelength
0.71073 Å
Monoclinic
P21/n
0.71073 Å
Triclinic
P1
0.71073 Å
Triclinic
P1
Crystal system
Space group
Unit cell dimensions
a = 9.3745(2) Å, R = 90°
b = 14.6816(3) Å, β = 90.3251(10)°
c = 14.5635(4) Å, γ = 90°
2004.38(8) Å3
a = 9.6279(1) Å, R = 69.5030(7)°
b = 11.0451(2) Å, β = 85.0590(7)°
c = 12.3145(3) Å, γ = 70.2181(11)°
1153.47(4) Å3
a = 8.9324(2) Å, R = 96.8119(7)°
b = 13.3079(2) Å, β = 99.4211(8)°
c = 13.4168(3) Å, γ = 90.0255(9)°
1561.94(5) Å3
Volume
Z
4
2
2
Density (calculated)
Absorption coefficient
F(000)
1.969 Mg/m3
1.948 Mg/m3
1.809 Mg/m3
0.822 mmÀ1
1.257 mmÀ1
1.341 mmÀ1
1152
656
836
Crystal size
0.200 Â 0.180 Â 0.060 mm3
5.160 to 27.497°
À12 e h e 12, 0 e k e 19,
0e l e 18
0.150 Â 0.080 Â 0.050 mm3
0.500 Â 0.300 Â 0.040 mm3
θ range for data collection
Index ranges
5.148 to 27.520°
5.118 to 27.472°
À12 e h e 12, À14 e k e 14,
À15e l e 15
À11 e h e 11, À17 e k e 16,
À17e l e 17
No. of reflns collected
No. of indep reflns
30894
20665
27296
4568 [R(int) = 0.029]
99.3%
5265 [R(int) = 0.024]
99.2%
7050 [R(int) = 0.031]
99.0%
Completeness to θ max
Absorption correction
Max. and min transmn
Refinement method
Semiempirical from equivalents
0.95 and 0.79
Full-matrix least-squares on F2
Semiempirical from equivalents
0.94 and 0.88
Full-matrix least-squares on F2
Semiempirical from equivalents
0.95 and 0.78
Full-matrix least-squares on F2
No. of data/restraints/params
Goodness-of-fit on F2
Final R indices [I > 2σ(I)]
R indices (all data)
4567/0/308
5265/0/307
7050/256/435
0.9445
0.9498
0.9641
R1 = 0.0415, wR2 = 0.0934
R1 = 0.0653, wR2 = 0.1140
141(15)
0.52 and À0.57 e/Å3
R1 = 0.0407, wR2 = 0.1011
R1 = 0.0593, wR2 = 0.1251
R1 = 0.0342, wR2 = 0.0748
R1 = 0.0547, wR2 = 0.0944
64(4)
0.67 and À0.67 e/Å3
Extinction coefficient
Largest diff peak and hole
0.61 and À0.69 e/Å3
Figure 2. Structure of the right handed form of 3. Orthogonal (left) and side (right) views of 3 (with respect to BC3 plane); thermal ellipsoids at 50%
probability (C atoms blue, Cl atoms orange, F atoms green and B atoms pink).
conjunction with 11B NMR data, support the premise that the
boron retains the three coordinate geometry in the solution
phase for all three species.31
contribution from each of these terms. In summary, the calcula-
tions reveal that an overall decrease of the total magnetic
shielding constant, and hence an increase in the observed 11B
chemical shift in the order B(C6F5)3 < B(C6F5)2(C6Cl5) <
B(C6F5)(C6Cl5)2 < B(C6Cl5)3, is dominated by changes in the
diamagnetic term. As the diamagnetic component is related
11B NMR chemical shift is determined by both diamag-
netic (σd) and paramagnetic (σp) contributions. The electronic
structure calculations (vide infra) enable us to deconvolute the
14730
dx.doi.org/10.1021/ja205037t |J. Am. Chem. Soc. 2011, 133, 14727–14740

Journal of the American Chemical Society
ARTICLE
Figure 3. Structure of the right handed form of 5. Orthogonal (left) and side (right) views (with respect to BC3 plane); thermal ellipsoids at 50%
probability (C atoms blue, Cl atoms orange, F atoms green and B atoms pink).
Figure 4. Structure of the right handed form of 6. Orthogonal (left) and side (right) views (with respect to BC3 plane); thermal ellipsoids at 50%
probability (C atoms blue, Cl atoms orange and B atoms pink). Disordered toluene molecule in asymmetric unit removed for clarity.
Table 2. Selected Bond Lengths and Angles for B(C6F5)3 and Compounds 3, 5 and 6a
B(C6F5)3Àn(C6Cl5)n; n = 0À3
B(C6F5)3
3
5
6
B1ÀC1 (Å)
[1.57]
[1.57]
[1.57]
[1.33À1.34]
À
1.580(4) [1.59]
1.577(4) [1.57]
1.561(4) [1.57]
1.589(4) [1.59]
1.586(4) [1.59]
1.552(4) [1.57]
1.576(4) [1.59)
1.587(4) [1.59]
1.586(4) [1.59]
À
B1ÀC7 (Å)
B1ÀC13 (Å)
Range CÀF (Å)
Range CÀCl (Å)
ArF∧BAr2 (deg)
ArCl∧BAr2 (deg)
1.333(3)À1.352(4) [1.33À1.34]
1.712(3)À1.732(3) [1.74À1.76]
24.4, 51.9 [24, 52]
1.330(4)À1.344(3) [1.33À1.34]
1.714(3)À1.730(3) [1.74À1.76]
22.3 [36]
1.711(3)À1.732(3) [1.74À1.75]
À
[40, 41, 40]
À
69.7[74]
57.1, 62.0 [57, 58]
54.5, 55.3, 58.1 [53, 53, 53]
a Numbers in parentheses are estimated standard uncertainties (esu). Computed values (B3LYP/TZVP) are shown in square brackets. ArX∧BAr2
(ArCl = C6Cl5; ArF = C6F5) is the torsion angle, as defined in Figure 5.
directly to the ground state electron density, we believe that the
observed 11B chemical shift in these systems is a reasonable probe
of the electron density at the boron nucleus. Thus, as the number
of coordinated C6Cl5 ligands increases, the B center is becoming
more electron deficient, corroborating the hypothesis that a
C6Cl5 substituent is more electronegative than C6F5.
Although the ArF groups are inequivalent in the X-ray crystal
structure for 3, 19F NMR resonances for each C6F5 ring show
averaging of aryl substituents in solution at 298 K. The perturba-
tions in the 19F NMR chemical shifts are most pronounced upon
the introduction of the first C6Cl5 group; thereafter, the effects
diminish, while for the 11B NMR chemical shifts the greatest
14731
dx.doi.org/10.1021/ja205037t |J. Am. Chem. Soc. 2011, 133, 14727–14740

Journal of the American Chemical Society
ARTICLE
change occurs between compounds 3 and 5. On replacement of
one C6F5 for C6Cl5 in B(C6F5)3, a stronger electron-withdraw-
ing effect is experienced by B. In the 13C NMR, we observe the
biggest difference between the ipso-C6F5 13C NMR resonance
for 3 and 5 (Δδi = +2.2), in comparison with B(C6F5)3 and 3
(Δδi = À1.0). The corresponding changes in the 13C NMR shifts
for the remaining C6F5 carbon atoms (Δδo, Δδm and Δδp; ortho,
meta and para respectively) become much smaller as n increases.
Electronic Structure Analysis. To explore the electronic
consequences of successive replacement of C6F5 with C6Cl5,
we have optimized the structures of all four members of the series
B(C6F5)3Àn(C6Cl5)n (n = 0À3) using density functional theory
(B3LYP/TZVP). The bond lengths and angles at the minimum
energy structures (shown in parentheses in Table 2) are all fully
consistent with the experimental data. Most significantly in the
context of the present study, the BÀC6F5 distances across the
series are uniformly 0.02 Å shorter than their BÀC6Cl5 counter-
parts. Computed 11B shielding constants also replicate the
experimental trend almost quantitatively, with shifts (relative
to B(C6F5)3, n = 0) of À0.4, 4.5, and 6.6 ppm (for n = 1, 2 and 3,
respectively) c.f. values of 0.3, 4.9, and 7.1 ppm from experiment
(Table 3). The decomposition of the shielding into paramagnetic
and diamagnetic components (see Supporting Information,
Table S1) shows that the diamagnetic term is, in absolute terms,
larger. The trend toward lower shielding across the series B-
(C6F5)3 > B(C6F5)2(C6Cl5) > B(C6F5)(C6Cl5)2 > B(C6Cl5)3 is
dominated by changes in the diamagnetic component. The
(negative) paramagnetic shielding constants show the opposite
trend, increasing slightly across the series. Thus changes in the
paramagnetic term actually attenuate the changes in the chemical
shift, which would be even more pronounced in their absence.
The encouraging level of agreement with both structural and
spectroscopic observables gives us confidence that the chosen
methodology is capturing the essential variations in electronic
structure across the series. The computed natural charges on the
boron center increase in the order n = 0 (0.81) < n = 1 (0.89) < n
= 2 (1.02) < n = 3 (1.16), supporting the hypothesis that Lewis
acidity increases with increasing substitution of C6F5 with C6Cl5.
Kinetics of H2O Dissociation from H2O [3]. H2O forms a
3
number of aqua complexes with B(C6F5)3, [H2O B(C6F5)3]
3
3
(H2O)n, involving H2O molecules hydrogen-bonded together
beyond theprimary coordinationsphereofthe dative 1:1 adduct.32
Although coordinated water in H2O B(C6F5)3 is tightly bound
3
and difficult to remove under vacuum (e.g., negligible loss at 10À3
mbar) or through heating (exceeding 60 °C results in hydrolysis to
(C6F5)2BOH and C6F5H),33 the kinetics of water dissociation
have been studied by observing degenerate aqua ligand transfer
betweenH2O B(C6F5)3 and free B(C6F5)3 usingVT 19F NMR.34
3
Since 3 exhibits reversible complexation of H2O under similar
conditions it was thought prudent to determine comparable data
for H2O dissociation from H2O [3]; such a property is likely to be
3
of use in true Lewis acid catalysis under aqueous regimes, in
contrast to the Brønsted acidic properties of H2O B(C6F5)3
3
resulting from strong activation of the water molecule.35 Due to
such facile decoordination of H2O, an analogous equilibrium was
achieved by combining 3 with H2O in a 2:1 ratio in C7D8 solution.
At 200 K a sharp 19F NMR spectrum is observed with separate
resonances corresponding to a mixture of H2O [3] and 3,
3
whereas at room temperature dynamic averaging reflects fast
exchange (Figures 6 and 7). Using line-shape analysis of the 19F
NMR spectra as a function of temperature enabled the rate
constants, and subsequent thermodynamic parameters ΔHq and
ΔSq, to be obtained from an Eyring plot (Table 5).
As anticipated, ΔHq for dissociation of H2O from 3 is less than
that for B(C6F5)3, which is consistent with the increased steric
profile of the borane due to the C6Cl5 group, leading to a weaker
BÀO interaction. The considerably greater entropic value for
H2O B(C6F5)3 may be rationalized by hydrogen-bonding be-
3
tween the hydroxyl protons and the ortho-F substituents; this
Table 4. 13C NMR Spectral Data for B(C6F5)3 and Com-
pounds 3 and 5 (C6F5 Ligands Only)
δ(13C NMR)/ppma
B(C6F5)3Àn(C6Cl5)n; n = 0À2
ortho-C
para-C
meta-C
ipso-C
B(C6F5)3
148.7
149.6
149.1
145.5
145.9
145.8
138.0
138.0
138.0
113.3
112.3
114.5
Figure 5. Calculation of the torsion angle for C6F5 ring in 5
(C6F5∧BAr2; Ar = C6Cl5) and the B(C6Cl5)2 unit, defined as the angle
between the planes of blue (BAr2) atoms and red (C6F5 ring) atoms. F
atoms shown in green, Cl atoms have been omitted for clarity.
3
5
a Solvent: CD2Cl2 reference (internal).
Table 3. 19F and 11B NMR Spectral Data for B(C6F5)3 and Compounds 3, 5 and 6
δ(19F NMR)/ppma
δ(11B NMR)/ppmc
b
B(C6F5)3Àn(C6Cl5)n; n = 0À3
ortho-F
para-F
meta-F
Δδm,p
B(C6F5)3
À128.2
À127.1
À127.4
À
À143.9
À142.8
À143.1
À
À161.1
À161.1
À160.6
À
17.2
18.3
17.5
À
61.2
61.5
66.1
68.2
3
5
6
a Solvent: CD2Cl2, CFCl3 reference (external). b Difference between 19F NMR meta and para resonances (ppm). c Solvent: CD2Cl2, BF3 OEt2
reference (external).
3
14732
dx.doi.org/10.1021/ja205037t |J. Am. Chem. Soc. 2011, 133, 14727–14740

Journal of the American Chemical Society
ARTICLE
Figure 6. Dissociative exchange between H2O and 3.
Figure 8. Gutmann and Childs Lewis acidity tests.
NMR chemical shift (Δδ) between free Et3PO and that of the
Et3PO-Lewis acid adduct, and has been subsequently modified
by Beckett.36e The second method developed by Childs is based
upon the downfield shift of the trans-crotonaldeyde H3 reso-
nance upon complexation to the Lewis acid.36a This site is
considered sterically remote from the site of bonding yet
electronically connected via conjugation (Figure 8).
The results obtained are summarized in Table 6; for consis-
tency, the Lewis acidity of B(C6F5)3 has also been determined.
The difference in chemical shift (Δδ 31P NMR) upon reaction of
Et3PO produces a trend that decreases in the order B(C6F5)3 > 3
> 5 and introduction of each C6Cl5 group has a linearly
proportional effect on the measured Lewis acidity (Figure 9).
for 6, no evidence of complexation was observed.
However, the Childs’ method yielded a different set of results
with the threshold for complex formation found to lie after
compound 3, and the upfield shift difference (Δδ) in the adduct
markedly smaller than that seen for B(C6F5)3.
Stephan et al. have recently documented the tuning of Lewis
acidity for a series of phosphine-borane/phosphinium-borane
species R2P(C6F4)B(C6F5)2 and [R3P(C6F4)B(C6F5)2]+, ob-
serving a linear correlation between the two techniques.37 Self-
consistent results are obtained because, in addition to maintain-
ing an environment consisting of only BÀC bonds, the site of
electronic modulation is remote (para-bound P on C6F4 ring)
and the steric factors about the borane center remain essentially
unchanged. Conversely, Britovsek et al. have synthesized the
series B(C6F5)3Àx(OC6F5)x (x = 1À3), where systematic re-
placement of pentafluorophenyl groups by harder pentafluoro-
phenoxy ligands results in an opposing binding preference for
Et3PO over crotonaldehyde.38 This was rationalized using Pear-
son’s HSAB principle39 where the largely covalent and softer
CdO pπ-pπ bond is a preferable donor to B(C6F5)3 compared
to the harder, more ionic pπ-dπ PdO bond in Et3PdO, which is
favored by B(OC6F5)3.
In the latter study, as x increases, a gradual increase in the
accessibility of the Lewis acid site permits, in theory, binding by
both donors surveyed (steric argument alone) thus allowing
discrimination between Lewis bases on electronic factors resul-
tant from replacement of BÀC by BÀO linkages. Examination of
the space-fill diagrams for the solid-state structures of com-
pounds 3, 5, and 6 (Figure 10) shows the enhanced screening
of the boron acceptor site upon replacement of C6F5 groups for
C6Cl5.
Figure 7. Variable-temperature 19F NMR spectra (simulated and
experimental) for exchange of water between H2O [3] and 3 (C7D8
3
solution).
Table 5. Activation Parameters for H2O Dissociation from
H2O [B] ([B] = B(C6F5)3, and 3) as Determined by VT 19
F
3
NMR (C7D8 Solution)a
ΔHq/
kcal molÀ1
ΔSq/
cal KÀ1 molÀ1
ΔGq
kcal molÀ1
/
300
34
H2O B(C6F5)3
19.0(3)
9.9(0.3)
24(1)
11.4(1)
8.9(1)
3
H2O [3]
3.2(1.2)
3
a Errors in parentheses.
effect will be accentuated by the strong polarization of the OÀH
bonds in the H2O molecule from the powerfully Lewis acidic
organoborane fragment, and indeed OÀH F bonding is ob-
3 3 3
served in the solid-state structure of this compound.32b In the
ground state of the complex, such organized bonding would likely
restrict free rotation of bonds within the H2O moiety, thus lowering
the total entropy of the system. However, upon H2O dissociation,
loss of ordered H-bonding leads to an overall greater entropy change
than that expected for a unimolecular to bimolecular conversion
alone. In contrast, for H2O [3], the degree of OÀH F interac-
3
3 3 3
Therefore, in determining the Lewis acidity of compounds on
progressing from B(C6F5)3 to B(C6Cl5)3, we examine a further
permutation of variables, that of an increase in steric crowding
concomitant with inherent electrophilicity of the boron center,
while retaining a BÀC(aryl)3 core in each case. A steric threshold
may be envisaged whereby non-bonded clashing between Lewis
base and ortho-ring substituents prevents a dative interaction and
overrides electronic factors; this is likely to be more prominent
for crotonaldehyde (CdO bond length 1.21 Å)40 than the
tions is anticipated to be smaller due to fewer C6F5 ligands in the
borane and poorer OÀH polarization from weaker H2OÀB
binding; therefore, upon dissociation, the entropy gain significantly
diminishes in relation to that found for H2O B(C6F5)3.
3
Lewis Acidity Measurements of B(C6F5)3Àn(C6Cl5)n. A
number of methods to assess relative Lewis acidity have been
developed and are commonly based on spectroscopic (IR,
NMR) techniques.36 The first uses the Gutmann acceptor
number (AN) which is calculated from the change in the 31P
14733
dx.doi.org/10.1021/ja205037t |J. Am. Chem. Soc. 2011, 133, 14727–14740

Journal of the American Chemical Society
ARTICLE
Table 6. 31P and 1H NMR Spectral Data Derived for Lewis Acidity Measurements for 3, 5, 6, and B(C6F5)3
Et3PO
trans-crotonaldeyde
Lewis acid
31P NMR/ppma
Δδ /ppmb
11B NMR/ppma
ANb
1H NMR/ppma,c
Δδ /ppmc
50.7
77.0
75.8
74.5
50.7
À
À
0
6.88
7.93
7.51
6.88
6.88
À
none
33.7
32.5
31.2
0.0
À2.5
À1.1
0.3
78.1
75.3
72.3
À
1.05
0.63
À
B(C6F5)3
3
5
6
68.2
À
a Solvent: CD2Cl2. b See reference 12 for calculation of acceptor number (AN); Δδ = [Et3PO(coordinated)] À δ[Et3PO(hexane)]. Δδ = δ(H3
c
coordinated) À δ(H3 free).
weakly coordinate THF (a moderately strong donor) it is
plausible that the observation of a reduction wave for these
species is due to high enough concentrations of the free three-
coordinate tris(aryl)borane electron acceptors in solution. How-
ever, the C6Cl5 analogues are anticipated to be substantially more
electron-deficient and indeed 3 strongly coordinates THF;
examination of the reduction of this compound using CV under
comparable conditions resulted in poorly defined voltammo-
grams with no discernible waves. However, conducting experi-
ments in the weakly coordinating solvent CH2Cl2 allowed the
observation of well-defined cyclic voltammograms for all com-
pounds, recorded at various scan rates (50À500 mVsÀ1
;
Figure 11). A plot of the reductive peak current vs the square
root of the voltage scan rate was constructed (Figure 11, insets)
and in all cases a linear relationship was observed, confirming that
the reduction was operating under diffusion control.44
When the potential was scanned initially in a negative direction
a single reduction wave was observed at every scan rate for each
complex. At lower scan rates (e.g., 50 mVsÀ1) when the direction
was subsequently reversed the corresponding oxidation waves were
observed to be rather broad and smaller in height than the
reduction wave. At higher scan rates the oxidative waves became
more pronounced and the ratio of the oxidative to reductive
peak current increased, but was always less than 1:1 even at
scan rates of up to 10 V sÀ1. The observed cyclic voltammetric
behavior is consistent with the reduction corresponding to an
EC mechanism45 where “E” denotes a heterogeneous electron
transfer step and “C” denotes a follow-up homogeneous chemi-
cal step, and is similar to the behavior observed by Cummings
et al. for B(C6F5)3Àn(Mes)n. Upon formation of the radical anion
of the parent complex, the radical anion rapidly undergoes
further homogeneous follow-up chemistry leading to decompo-
sition of the radical anion produced at the electrode. At slow scan
rates the decomposition of the intermediate radical anion is
sufficiently fast compared to the voltammetric time scale so that
its corresponding reoxidation is not observed. As the scan rate is
increased the kinetics of decomposition begin to be outrun and a
correspondingly larger oxidation wave is then observed until the
ratio of ipOx/ipRed approaches unity. The formal reduction
potential of each compound 3, 5 and 6 is calculated using the
midpeak potential, Emid = (EpOx + EpRed)/2, and are listed in
Table 7. A clear trend is observable whereby the reduction peak
potential of each complex was found to shift to increasingly less
negative potentials, and the voltammetry appears to become
more reversible as the number of C6Cl5 substituents attached to
the boron center increases. These findings support the NMR
spectral data that a C6Cl5 group is more electron-withdrawing
than a C6F5 substituent, thus rendering the boron center more
Figure 9. Plot of acceptor number (AN) vs n in B(C6F5)3Àn(C6Cl5)n
(n = 0À2).
phosphine oxide (PdO 1.46(1) Å in Ph3PO, for example)41 at
the locus of complexation. Thus, for Et3PO adduct formation is
accomplishable for n = 0, 1, 2 and a linear relationship exists
throughout the series, whereas crotonaldehyde can only bind up
to n = 1; even here a decrease in Lewis acidity is evident from
B(C6F5)3. Despite the powerfully electron-withdrawing effects
of three C6Cl5 ligands in 6, neither Lewis base can achieve
coordination and the species may be considered to have passed
the steric threshold in both circumstances.
Electrochemical Studies of B(C6F5)3Àn(C6Cl5)n (n = 0À3).
For a more absolute determination of electron density at the
boron acceptor orbital, the reduction potentials for 3, 5 and 6 are
of interest. As shown by Power, reduction of B(Mes)3 and
subsequent X-ray structure determination of the resultant radical
anion reveals that minimal structural reorganization of the
trigonal planar environment of the borane occurs upon electron
transfer;29 thus the potentials may be viewed as an approximate
measure of the electrophilicity of B(C6F5)3Àn(C6Cl5)n (n =
0À3) in the absence of steric effects. In spite of the prevalent
use of B(C6F5)3 as a powerful Lewis acid no report exists to
date that claims to directly observe the voltammetric reduction
of this species, even though evidence documents its use as a
one electron oxidant.42 Cummings et al. studied the cyclic
voltammetry of the series B(C6F5)3Àn(Mes)n (n = 1À3) in
order to estimate the redox potential of B(C6F5)3 (THF,
0.1 M [nBu4N][BF4] electrolyte) via extrapolation.43 Since
these mesityl-substituted boranes are documented to only
14734
dx.doi.org/10.1021/ja205037t |J. Am. Chem. Soc. 2011, 133, 14727–14740

Journal of the American Chemical Society
ARTICLE
Figure 10. Space-fill diagram of (a) 3, (b) 5 and (c) 6; C atoms blue, Cl atoms orange, F atoms green, and B atoms pink.
Figure 11. Overlaid cyclic voltammograms recorded at scan rates of 50 to 500 mVsÀ1 in CH2Cl2 (0.1 M [nBu4N][BF4]) of (a) 3 (10 mM
concentration); (b) 5 (10 mM concentration); (c) 6 (5 mM concentration). (Insets) Respective plots of reductive peak current vs square-root of voltage
scan rate.
oxidizing and correlates well with the more positive Hammett
parameter for aryl-bound Cl vs F. In addition, the larger size of Cl
(especially in the ortho position) induces an increased shielding
of the boron-centerd radical anion upon reduction, inhibiting
bimolecular decomposition pathways and hence increasing the
stability/reversibility of the voltammetry; a similar buttressing
effect is attributed to the stability, and hence persistence, of the
isoelectronic [C(C6Cl5)3]•.46 Assuming a linear relationship
between these potentials and the number of C6Cl5 substituents
attached to each boron center provides an estimate
of the reduction potential for B(C6F5)3 of À1.92 V ( 0.1 V
(vs Cp2Fe0/+).
Previous attempts to observe the direct reduction of B(C6F5)3
have employed either CH2Cl2 or THF solvent (despite the fact
that THF B(C6F5)3 is known to be a strongly bound adduct)47
3
and commonly used electrolytes such as [nBu4N][ClO4] or
14735
dx.doi.org/10.1021/ja205037t |J. Am. Chem. Soc. 2011, 133, 14727–14740

Journal of the American Chemical Society
ARTICLE
Table 7. Average values of Emid measured from CV data for
complexes B(C6F5)3Àn(C6Cl5)n (n = 0-3)
B(C6F5)3Àn(C6Cl5)n
Emid/Va
B(C6F5)3
À1.97 ( 0.10
À1.87 ( 0.05
À1.55 ( 0.05
À1.48 ( 0.02
3
5
6
a Potentials are reported vs Cp2Fe0/+ (CH2Cl2) at a Pt macrodisc
electrode.
Figure 13. A plot showing the Emid potentials of complexes B-
(C6F5)3Àn(C6Cl5)n (n = 0À3; denoted  ) vs the number of substituent
C6Cl5 groups in the complex, and also the Emid values determined for the
series B(C6F5)3Àn(Mes)n (n = 0À3) and their estimated value for the
reduction potential of B(C6F5)3 (blue 2).43 The measured potential,
Emid, for B(C6F5)3 in CH2Cl2 (red b), has also been included.
Figure 12. Overlaid cyclic voltammograms of B(C6F5)3 in CH2Cl2
(5 mM, 0.1 M [nBu4N][BArF24]; 1À5 V sÀ1 scan rate).
[nBu4N][BF4]; at best, ill-defined curves were observed.42,43
B(C6F5)3 has demonstrated a rich oxo-Àanion binding chemistry
which quite possibly extends to ClO4 (reported to be more
coordinating than BF4À),48 and since the Lewis acidity of this
borane has been judged to be similar to that of BF3, it is probable
that FÀ abstraction from BF4À to form [FB(C6F5)3]À occurs;49
in both of these examples, the supporting electrolyte would
quench the acceptor orbital and hence inhibit reduction. With
these potential pitfalls in mind we resorted to the [nBu4N]+ salt
of the weakly coordinating anion [BArF24]À in CH2Cl2 as
electrolyte (BArF24 = B[3,5-(CF3)2C6H3]4); Figure 12 shows
the resulting cyclic voltammetry of B(C6F5)3, obtained success-
fully for the first time. At modest scan rates (<1 V sÀ1), a
reduction wave is observed which corresponds to a one-electron
reduction forming the [B(C6F5)3]•À radical anion, at À1.97 (
0.1 V. The rate of decomposition of B(C6F5)3 continues the
trend observed for B(C6F5)3Àn(C6Cl5)n complexes (n = 1À3)
and is the fastest of the series, such that a back peak correspond-
ing to the reoxidation of the radical anion is only observed at scan
Figure 14. Experimental (black) and simulated (red; isotropic line-
width 6.1 G) X-band EPR spectra of [3]•À in THF at 298 K (11B 80.1%,
I = 3/2; 10B 19.9%, I = 3).
referencing nonaqueous potentials to the Cp2Fe0/+ couple in the
solvent system of choice rather than the practice of referencing
potentials to the aqueous SCE reference electrode.50 The data
show that the estimated reduction potential of B(C6F5)3 in the
work of Cummings et al. is also reassuringly close to our
measured value; the difference in values may be accounted for
by recognizing that the studies have been conducted in different
solvents (CH2Cl2 and THF). The difference in shift of ca. +200
mV per C6Cl5 introduced compared with ca. À500 mV obtained
for the mesityl series may be understood in that for the former
case C6F5 (a σ-acid and π-donor) is replaced by a group of
slightly greater electron-withdrawing ability (predominantly σ-
acid) whereas in the latter it is replaced by a strongly electron-
donating ligand (a σ-donor); the effect on the electrode poten-
tials is therefore appreciably greater.
rates in excess of 1 V sÀ1
.
Figure 13 consolidates the measured Emid potentials of all
complexes B(C6F5)3Àn(C6Cl5)n complexes (n = 0À3) vs the
number of C6Cl5 groups in the molecule; a clear linear trend is
observed and the estimated value of the reduction potential of
B(C6F5)3 is indeed very close to the measured value. It is also
interesting to compare the estimated values from the B(C6F5)3-n
-
(Mes)n series used by Cummings et al. to estimate the potential
of B(C6F5)3 and to compare their estimate with our measured
value.43 To do this, we have followed the IUPAC convention of
14736
dx.doi.org/10.1021/ja205037t |J. Am. Chem. Soc. 2011, 133, 14727–14740

Journal of the American Chemical Society
ARTICLE
EPR Study of [B(C6Cl5)3]•À. The radical anion [B(C6F5)3]•À
has been previously reported via reduction of the parent Lewis
acid with Cp*2Co in THF at À50 °C; the EPR spectrum was
rapidly recorded at this temperature due to the transient nature
of the anion (t1/2 ≈ 2 min, 298 K).51 Electrochemical experi-
ments suggest the reduction product of B(C6Cl5)3 to be
considerably more stable, and accordingly the synthesis of
[B(C6Cl5)3]•À was investigated. Reduction of 6 using Na(s) in
THF was conducted at room temperature, resulting in a vivid
blue solution whereupon the EPR spectrum was obtained.
Figure 14 shows the experimental observations and simulated
spectrum, which confidently support the existence of [6]•À;
furthermore, the multiplicity of the EPR signal precludes dimeric
association of the radicals in solution.
in the presence of H2, are the subject of current investigation and
will be reported in due course.
’ EXPERIMENTAL DETAILS
General. All reactions and compounds were manipulated under N2
using either an MBraun Unilab glovebox or using standard Schlenk line
techniques on a dual manifold vacuum/inert gas line, unless stated
otherwise. For the manipulation of moisture sensitive compounds, all
glassware was heated to 170 °C before use. Solvents and solutions were
transferred using a positive pressure of nitrogen through stainless steel or
Teflon cannulae or via plastic syringes for volumes less than 20 mL.
Filtrations were performed using either glassware containing sintered
glass frits or modified stainless steel cannulae fitted with glass microfiber
filters. Pentane, hexane, toluene and CH2Cl2 were dried using an
MBraun SPS-800 solvent purification system, whereas Et2O was distilled
from purple Na/benzophenone diketyl; all except CH2Cl2 were stored
over K-mirrored ampules. Deuterated NMR solvents were dried and
freezeÀthaw degassed over the appropriate drying agent: C6D6, C7D8
(K); CD2Cl2 (activated 3 Å molecular sieves) and purchased from Goss
Scientific (99.6, 99.6 and 99.8% D respectively). BBr3 (99.9%), BCl3
(1.0 M in heptane), C6Cl6 (99.9%), nBuLi (2.5 M in hexanes), 2,2,6,6-
tetramethylpiperidine (>99%), trans-crotonaldehyde (>99%) and
[nBu4N][BF4] were purchased from Sigma Aldrich; all were used as
received. CuC6F5,52 [nBu4N][B(C6H3(CF3)2)4] and B(C6F5)3 were
prepared according to literature procedures. NMR spectra were re-
The value of hyperfine coupling a(11B) (10.3 G) agrees very
well with those reported for the tris(aryl)borane radical anions
[B(C6F5)3]•À (10.5 G), [B(Mes)3]•À (10.3 G) and [BPh3]•À
(9.8 G), and g (2.002) is very close to the free electron value (ge =
2.0023). Measurements of the exponential decay in the EPR
signal intensity gave a half-life of 115 min at 298 K revealing
[6]•À to be considerably more stable than its perfluoro
analogue.51
’ CONCLUSIONS
1
corded on a 300 MHz Varian VX-Works spectrometer. H and 13C
The complete series of perchloroaryl Lewis Acids B-
(C6F5)3Àn(C6Cl5)n (n = 1À3; 3, 5 and 6) have been successfully
synthesized and comprehensively characterized; perchlorination
of all the aryl substituents confers considerable thermal and
hydrolytic stability to 6. The solid-state structures reveal a
trigonal planar environment for boron in all the compounds,
despite the asymmetry of the ligand set in the complexes 3 and 5.
Solution 19F, 13C and 11B NMR studies reveal a trend of BÀC6F5
resonance interactions being replaced by primarily inductive
effects arising from increasing C6Cl5 incorporation. A decrease
in Lewis acidity has been established upon sequential substitu-
tion of C6F5 with C6Cl5 in B(C6F5)3 for n = 0À2, as demon-
strated by the Gutmann-Beckett method, whereas the Childs
method was only successful for n = 0 and 1; the acceptor
properties of 6 could not be determined by either of these
techniques. Conversely, electrochemical studies show that the
boron center becomes more electron deficient (oxidizing) as the
series is traversed, demonstrating a C6Cl5 substituent to be more
electron withdrawing than C6F5. The optimized structures of all
the Lewis acids, B(C6F5)3Àn(C6Cl5)n (n = 0À3), using density
functional theory (B3LYP/TZVP) are all fully consistent with
the experimental structural data. Computed 11B shielding con-
stants also replicate the experimental trend almost quantitatively,
and the computed natural charges on the boron center increase
in the order n = 0 (0.81) < n = 1 (0.89) < n = 2 (1.02) < n = 3
(1.16), supporting the hypothesis that electrophilicity increases
concomitantly with substitution of C6F5 for C6Cl5. All the results
may be coherently rationalized be realizing that various measure-
ments of Lewis acidity may be dominated by either steric and/or
electronic effects. While electrochemistry provides a physio-
chemical measure of the electron affinity of the B center in these
compounds, it neglects the steric cost of B sp2-sp3 rehybridiza-
tion, which is important for bulky boranes upon coordination of
Lewis bases. However, the Gutmann-Beckett/Childs’ methods
incorporate both factors in their measurement and give a more
reliable indication of “chemical” Lewis acidity. The reactivity of
these new boranes, particularly as Frustrated Lewis Pair partners
chemical shifts are given relative to Me4Si and referenced internally to
the residual proton shift in the deuterated solvent employed. 11B, 19F
and 31P chemical shifts were referenced externally to BF3 OEt2, CFCl3
3
and 85% H3PO4. Assessment of Lewis acidity using the Gutmann-
Beckett method12 followed a modification described by D.W. Stephan
et al.37a which used an excess of Lewis acid to Et3PO (3:1), dissolved in
CD2Cl2. To accurately record Δδ, the solution was placed in an NMR
tube along with a sealed reference capillary containing uncoordinated
phosphine oxide. The 31P NMR shifts were recorded at 298 K. For the
Childs Method was performed as described by Childs et al.36a Lewis acid
and trans-crotonaldehyde were mixed in a 1:1 ratio and placed in an
1
NMR tube. The H NMR chemical shift of the H3 proton of croto-
naldehyde was then recorded. High resolution mass spectrometry
samples (HRMS; EI) were recorded using a Bruker FT-ICR-MS Apex
III spectrometer and IR spectra were recorded on a Nicolet MAGNA-IR
560 FT-IR spectrometer (range 4000À400 cmÀ1, resolution 0.5 cmÀ1).
Crystals suitable for X-ray diffraction were mounted on a glass fiber
either bare, or using perfluoropolyether oil, and mounted in a stream of
N2 at 150 K using an Oxford Cryosystems Cryostream unit.53 Diffrac-
tion data were obtained using graphite monochromated Mo KR radia-
tion on an Nonius KappaCCD diffractomer, and processed using the
DENZO-SMN package.54 The structure was then solved using the direct
methods program SIR92,55 which located all the non-hydrogen atoms.
Subsequent full-matrix least-squares refinement was carried out using
the CRYSTALS program suite.56 Full details are in the Supporting
Information (CIF); crystallographic data (excluding structure factors)
for 3, 5 and 6 have been deposited with the Cambridge Crystallographic
Data Centre and can be obtained via www. ccdc.cam.ac.uk/data_re-
quest/cif. Electrochemical experiments were performed using an Auto-
lab PGSTAT 30 computer-controlled potentiostat. Cyclic voltammetry
(CV) was performed using a three-electrode configuration consisting of
a Pt disk working electrode (GoodFellow, Cambridge, UK 99.99% area
1.7 ( 0.3 Â 10À3 cm2), a Pt gauze counter electrode and a Ag wire
pseudoreference electrode. Pt working electrodes were polished be-
tween experiments using successive grades of alumina slurries from 1.0
to 0.3 μm rinsed in distilled water and subjected to brief ultrasonication
to remove any adhered alumina microparticles. The electrodes were
14737
dx.doi.org/10.1021/ja205037t |J. Am. Chem. Soc. 2011, 133, 14727–14740

Journal of the American Chemical Society
ARTICLE
then dried in an oven at 120 °C to remove any residual traces of water.
For either cell arrangement the potentials of the Ag wire pseudorefer-
ence electrodes were found to drift by as much as 100 mV between
experimental runs and therefore calibrated to the ferrocene/ferrocenium
couple in CH2Cl2 in the absence of any borane complexes at the end of
each run. All electrochemical measurements were performed at ambient
temperatures under an inert N2 atmosphere containing either 0.1 M
[nBu4N][BF4] or 0.1 M [nBu4N][B(C6H3(CF3)2)4] in CH2Cl2.
Computational Details. Calculations were performed at the DFT
level with the B3LYP functional57 and the TZVP basis set58 of Ahlrichs
and co-workers, as implemented in Gaussian03.59 The structures of
stationary points were fully optimized without any symmetry constraints
and confirmed to be minima by the absence of imaginary frequencies.
Where crystallographic data were available, the experimental coordi-
nates were used as the initial guess for the structure. 11B NMR shielding
constants were calculated with the Gauge-Independent Atomic Orbital
(GIAO) method,60 using the geometries obtained at the B3LYP/TZVP
level. These calculations employed the B3LYP functional in conjunction
with a polarizable continuum model (PCM),61 using dichloromethane
(ε = 8.93) as the solvent. The TZVP basis set on boron was replaced by a
basis set optimized for shielding constants, aug-pcS-2(triple-ζ quality),62
while the TZVP basis set was retained for the other atoms (C, F, Cl).
Relative chemical shifts (δcalc) were obtained by referencing the
isotropic nuclear magnetic shielding constant of the probe atom (σX)
against the shielding constant (σref) of the B atom in B(C6F5)3 with δref
(11B NMR) = 61.2 ppm, thus δcalc = σref À σX + δref.
(s), 861 (s), 814 (s), 715 (s). Anal. Calcd. for BC6Cl5Br2: C 17.16.
Found: C 17.23.
B(C6Cl5)(C6F5)2 (3). Toluene (150 mL) was added to a stirred
mixture of B(C6Cl5)Br2 (6.78 g, 16.1 mmol) and CuC6F5 (7.61 g, 33.0
mmol), followed by heating to 60 °C for 4 h. The initially translucent
solution rapidly became cloudy, producing a white precipitate of CuBr.
Upon cooling, the solution was filtered through Celite and the residue
washed with toluene (2 Â 50 mL), before removing the solvent under
vacuum. The resultant off-white solid was then sublimed (125 °C, 0.01
mbar) to produce analytically pure B(C6Cl5)(C6F5)2 (3) as a white
powder (7.80 g, 81%, 13.1 mmol). Crystals suitable for X-ray diffraction
were grown from slow-cooling of a toluene solution to À30 °C. 13C{1H}
NMR (CD2Cl2, 75 MHz): δ 149.7 (dm, 1JCF = 251 Hz, ortho-C6F5); δ
145.9 (dm, 1JCF = 262 Hz, para-C6F5); δ 141.0 (br, ipso-C6Cl5); δ 138.0
1
(dm, JCF = 249.5 Hz, meta-C6F5); δ 135.1 (s, para-C6Cl5); δ 132.3,
131.3 (both s, meta-C6Cl5 and ortho-C6Cl5); δ 112.2 (br, ipso-C6F5). 11
B
NMR (C7D8, 128 MHz): 63.6 (s, br). 19F NMR (C7D8, 282.2 MHz): δ
À127.3 (d, 4F, 3JFF = 22 Hz, ortho-C6F5), δ À141.0 (t, 2F, 3JFF = 23 Hz,
para-C6F5), δ À159.9 (m, 4F, meta-C6F5). HRMS (EI, m/z): for
BC18Cl5F10 Calcd: 591.8378. Found: 591.8376. IR (Nujol, cmÀ1):
1700 (m), 1653 (m), 1646 (m), 1559 (m), 1549 (m), 1521 (s), 1507
(w), 1482 (s), 1437 (m), 1382 (m), 1336 (m), 1322 (m), 1235 (w),
1167 (m), 1142 (w), 1015 (w), 979 (s), 674 (m), 668 (m), 659 (w).
Anal. Calcd. for BC18Cl5F10: C 36.38. Found: C 36.27.
B(C6Cl5)2Cl (4). Hexane (100 mL) was slowly added to a stirred
solution of C6Cl5Li (from 29.0 mmol of C6Cl6) at À78 °C, resulting in
the formation of a precipitate. BCl3 (14 mL, 14.0 mmol, 1.0 M in
heptane) was then syringed into this suspension and the reaction
C6Cl5Li. This is adapted from a literature procedure.18 A 500 mL
Schlenk was charged under a nitrogen flush with C6Cl6 (5.70 g, 20
mmol) and left under vacuum for 20 min to remove any moisture.
Following the addition of 150 mL Et2O, the slurry was cooled to À78 °C
using a CO2(s)/acetone bath. With rapid stirring nBuLi (18.2 mL, 20.4
mmol, 2.5 M in hexanes) in hexane was added by means of a syringe. The
contents were allowed to warm to À10 °C until solid C6Cl6 was no
longer visible, and a translucent amber solution of C6Cl5Li had formed.
The contents were later cooled to À78 °C.
mixture allowed to slowly warm up to room temperature in the CO2(s)
/
acetone cooling bath, followed by further stirring for 12 h. The solvent
was then stripped under vacuum and the orange residue extracted with
CH2Cl2 (2 Â 100 mL) and filtered through Celite. The solution was
then reduced to minimum volume and cooled to À35 °C, affording an
orange powder after washing with cold (À35 °C) CH2Cl2 (20 mL) and
drying under vacuum. Two further crops were isolated from the mother
liquor following the latter procedure. A final recrystallization from slow-
cooling a saturated CH2Cl2 solution to À35 °C gave pale orange needles
of B(C6Cl5)2Cl (4), which were dried in vacuo (4.09 g, 54%, 7.5 mmol).
13C{1H} NMR (CD2Cl2, 75 MHz): δ 136.4 (s, para-C6Cl5); δ 134.7,
133.1 (s, meta-C6Cl5 and ortho-C6Cl5). Resonance for ipso-C6Cl5 not
observed. 11B NMR (CD2Cl2, 128 MHz): δ 62.9 (s, br). HRMS (EI, m/
z): for BC12Cl11 Calcd: 539.6667. Found: 539.6660. IR (Nujol, cmÀ1):
1700 (m), 1684 (m), 1653 (m), 1559 (w), 1540 (w), 1521 (w), 1457 (s),
1377 (s), 1322 (s), 1298 (s), 668 (w). Anal. Calcd. for BC12Cl11: C
26.45. Found: C 26.57.
Zn(C6Cl5)2 (1). C6Cl5Li (from 28 mmol C6Cl6) in Et2O at À78 °C
was rapidly transferred via cannula to a stirred Et2O (100 mL) solution
of ZnCl2 (1.91 g, 14 mmol) at À20 °C. The yellow reaction mixture was
then slowly warmed to room temperature over the course of 3 h
whereupon a precipitate began to form, which was stirred for a further
12 h. Filtration through Celite to remove LiCl, washing with Et2O (2 Â
50 mL), and subsequent removal of the solvent under reduced pressure
produced a pale yellow solid, which was washed with cold (À78 °C)
Et2O (2 Â 50 mL) until the washings became colorless. Drying in vacuo
for 12 h while slowly heating to 60 °C afforded base-free Zn(C6Cl5)2 (1)
as a white powder (6.51 g, 82%, 11.5 mmol). HRMS (EI, m/z): for
ZnC12Cl10 Calcd: 557.6177. Found: 557.6163. IR (Nujol, cmÀ1): 1700
(m), 1653 (m), 1559 (m), 1533 (m), 1465 (m), 1334 (s), 1311 (s), 1230
(m), 1128 (m), 982 (s), 659 (s), 524 (w). Anal. Calcd. for ZnC12Cl10: C
25.55. Found: C 25.43.
B(C6Cl5)Br2 (2). A 250 mL greaseless ampule was charged with a
magnetic stirrer bar, Zn(C6Cl5)2 (6.51 g, 11.5 mmol), toluene
(150 mL), and finally BBr3 (7.20 g, 2.77 mL, 28.8 mmol), before being
sealed and heated (100 °C) with stirring for 12 h. The suspension was
then cooled to room temperature and filtered through Celite, before
stripping the solvent under vacuum to produce a solid. Washing this
residue with pentane (2 Â 50 mL) gave B(C6Cl5)Br2 (2) as an off-white
powder (6.78 g, 70%, 16.1 mmol). 13C{1H} NMR (CD2Cl2, 75 MHz):
δ 135.6 (s, para-C6Cl5); δ 132.8, 130.0 (s, meta-C6Cl5 and ortho-C6Cl5).
Resonance for ipso-C6Cl5 not observed. 11B NMR (CD2Cl2, 128 MHz):
δ 55.8 (s, br). HRMS (EI, m/z): for BC6Cl5Br2 Calcd: 415.6902. Found:
415.6912. IR (Nujol, cmÀ1): 1700 (w), 1653 (w), 1539 (s), 1457 (s),
1377 (w), 1338 (s), 1303 (s), 1235 (s), 1132 (s), 976 (s), 920 (w), 888
B(C6Cl5)2(C6F5) (5). A greaseless glass ampule was charged with a
stirrer bar, B(C6Cl5)2Cl (4.09 g, 7.5 mmol), CuC6F5 (1.82 g, 7.9 mmol)
and toluene (100 mL). The vessel was sealed and heated to 80 °C
(temperatures above this result in decomposition of CuC6F5) with
stirring for 72 h before being cooled and the solvent removed in vacuo.
The compound was extracted using CH2Cl2 (2 Â 50 mL), followed by
filtering through Celite and solvent removal in vacuo. The resultant
residue was recrystallized from toluene/hexane (1:2) at À78 °C,
producing a microcrystalline solid which was washed with cold
(À78 °C) pentane (2 Â 20 mL) and dried under vacuum to give
spectroscopically pure 2 (3.39 g, 70%, 5.0 mmol). Crystals suitable for
X-ray diffraction were grown from slow evaporation of a saturated
toluene solution. 13C{1H} NMR (CD2Cl2, 75 MHz): δ 149.0 (dm, 1JCF
=
253 Hz, ortho-C6F5); δ 145.9 (dm, 1JCF = 261 Hz, para-C6F5); δ 138.0
1
(dm, JCF = 251 Hz, meta-C6F5); δ 139.6 (br, ipso-C6Cl5); 136.6
(s, para-C6Cl5); δ 133.0 (s, meta-C6Cl5 and ortho-C6Cl5); δ 114.5
(br, ipso-C6F5). 11B NMR (C7D8, 128 MHz): 64.1 (s, br). 19F NMR
3
(C7D8, 282.2 MHz): δ À127.2 (d, 2F, JFF = 21 Hz, ortho-C6F5), δ
À141.4 (t, 1F, 3JFF = 21 Hz, para-C6F5), δ À159.7 (m, 2F, meta-C6F5).
14738
dx.doi.org/10.1021/ja205037t |J. Am. Chem. Soc. 2011, 133, 14727–14740

Journal of the American Chemical Society
ARTICLE
HRMS (EI, m/z): for BC18Cl10F5 Calcd: 675.6899. Found: 675.6774.
IR (Nujol, cmÀ1): 1700 (m), 1653 (m), 1559 (m), 1540 (w), 1521 (m),
1507 (w), 1481 (s), 1465 (s), 1394 (m), 1332 (s), 1313 (s), 1237 (m),
1190 (w), 1147 (m), 1127 (w), 1104 (w), 973 (s), 876 (w), 668 (m),
642 (w). Anal. Calcd. for BC18Cl10F5: C 31.96. Found: C 32.27.
B(C6Cl5)3 (6). Hexane (100 mL) was added to C6Cl5Li (from 31.2
mmol of C6Cl6), as detailed in the synthesis of 4. To this slurry was
added BCl3 (10 mL, 10 mmol, 1.0 M in heptane) via syringe at À78 °C.
The solution was allowed to warm slowly to À10 °C and stirred for an
hour before the cloudy orange suspension was removed from the cooling
bath, and reacted for a further 12 h. After quenching the reaction by
addition of 0.5 mL H2O, the solvent was removed in vacuo and
subsequent workup performed in air. CH2Cl2 (150 mL) was used to
extract the crude product, the slurry being filtered through Celite and the
filter pad washed with further CH2Cl2 (2 Â 50 mL). Solvent was
removed using rotary evaporation, yielding an amber solid. Recrystalli-
zation using a minimum quantity of toluene at 100 °C followed by rapid
filtration through glass wool and slow cooling to room temperature led
(2) (a) Gevorgyan, V.; Rubin, M.; Benson, S.; Liu, J. X.; Yamamoto,
Y. J. Org. Chem. 2000, 65, 6179–6186. (b) Blackwell, J. M.; Sonmor,
E. R.; Scoccitti, T.; Piers, W. E. Org. Lett. 2000, 2, 3921–3923. (c) Parks,
D. J.; Piers, W. E. J. Am. Chem. Soc. 1996, 118, 9440–9441. (d) Morrison,
D. J.; Piers, W. E. Org. Lett. 2003, 5, 2857–2860. (e) Blackwell, J. M.;
Foster, K. L.; Beck, V. H.; Piers, W. E. J. Org. Chem. 1999, 64,
4887–4892. (f) Morrison, D. J.; Blackwell, J. M.; Piers, W. E. Pure Appl.
Chem. 2004, 76, 615–623.
(3) (a) Krossing, I.; Raabe, I. Angew. Chem., Int. Ed. 2004,
43, 2066–2090. (b) Bochmann, M.; Lancaster, S. J.; Hannant, M. D.;
Rodriguez, A.; Schormann, M.; Walker, D. A.; Woodman, T. J. Pure Appl.
Chem. 2003, 75, 1183–1195. (c) Lancaster, S. J.; Walker, D. A.;
Thornton-Pett, M.; Bochmann, M. Chem. Commun. 1999, 1533–1534.
(d) Lancaster, S. J.; Rodriguez, A.; Lara-Sanchez, A.; Hannant, M. D.;
Walker, D. A.; Hughes, D. H.; Bochmann, M. Organometallics 2002,
21, 451–453. (e) LaPointe, R. E.; Roof, G. R.; Abboud, K. A.; Klosin, J.
J. Am. Chem. Soc. 2000, 122, 9560–9561. (f) Yang, X. M.; Stern, C. L.;
Marks, T. J. J. Am. Chem. Soc. 1994, 116, 10015–10031. (g) Bernsdorf,
A.; Brand, H.; Hellmann, R.; Kockerling, M.; Schulz, A.; Villinger, A.;
Voss, K. J. Am. Chem. Soc. 2009, 131, 8958–8970. (h) Chen, E. Y. X.;
Marks, T. J. Chem. Rev. 2000, 100, 1391–1434.
(4) (a) Beckett, M. A.; Brassington, D. S.; Coles, S. J.; Hursthouse,
M. B. Inorg. Chem. Commun. 2000, 3, 530–533. (b) Jacobsen, H.; Berke,
H.; Doring, S.; Kehr, G.; Erker, G.; Frohlich, R.; Meyer, O. Organome-
tallics 1999, 18, 1724–1735.
(5) (a) Erker, G. Dalton Trans. 2005, 1883–1890. (b) Piers, W. E.;
Chivers, T. Chem. Soc. Rev. 1997, 26, 345–354.
(6) (a) Stephan, D. W. Dalton Trans. 2009, 3129–3136. (b) Kenward,
A. L.; Piers, W. E. Angew. Chem., Int. Ed. 2008, 47, 38–41. (c) Stephan,
D. W.; Erker, G. Angew. Chem., Int. Ed. 2010, 49, 46–76.
(7) (a) Dureen, M. A.; Stephan, D. W. J. Am. Chem. Soc. 2009,
131, 8396–8397. (b) Chen, C.; Eweiner, F.; Wibbeling, B.; Froehlich, R.;
Senda, S.; Ohki, Y.; Tatsumi, K.; Grimme, S.; Kehr, G.; Erker, G. Chem.
ÀAsian J. 2010, 5, 2199–2208. (c) Chen, C.; Froehlich, R.; Kehr, G.;
Erker, G. Chem. Commun. 2010, 46, 3580–3582. (d) Dureen, M. A.;
Brown, C. C.; Stephan, D. W. Organometallics 2010, 29, 6422–6432. (e)
Dureen, M. A.; Brown, C. C.; Stephan, D. W. Organometallics 2010,
29, 6594–6607. (f) Jiang, C.; Blacque, O.; Berke, H. Organometallics 2010,
29, 125–133. (g) Moemming, C. M.; Froemel, S.; Kehr, G.; Froehlich, R.;
Grimme, S.; Erker, G. J. Am. Chem. Soc. 2009, 131, 12280–12289. (h)
Sortais, J.-B.; Voss, T.; Kehr, G.; Froehlich, R.; Erker, G. Chem. Commun.
2009, 7417–7418. (i) Voss, T.; Chen, C.; Kehr, G.; Nauha, E.; Erker, G.;
Stephan, D. W. Chem.—Eur. J. 2010, 16, 3005–3008.
to the formation of pale yellow crystals of 3 (toluene). The toluene
3
supernatant was siphoned off and the crystals washed with pentane (2 Â
40 mL), followed by drying overnight under vacuum (1 Â 10À3 mbar) to
remove toluene of crystallization. Yield 3.26 g (42%, 4.3 mmol). X-ray
quality crystals were produced by a second toluene recrystallization.
13C{1H} NMR (CD2Cl2, 75 MHz): δ 140.6 (br, ipso-C6Cl5); 136.7
(s, para-C6Cl5); δ 135.3, 133.0 (both s, meta-C6Cl5 and ortho-C6Cl5).
11B NMR (C7D8, 128 MHz): 65.6 (s,br). HRMS (EI, m/z): for
BC18Cl15 Calcd: 751.5421. Found: 751.5177. IR (Nujol, cmÀ1): 1700
(m), 1684 (m), 1652 (m), 1558 (m), 1540 (m), 1507 (m), 1468 (s),
1334 (m), 1313 (m), 1232 (m), 1130 (w), 991 (w), 668 (m), 636 (w).
Anal. Calcd. for BC18Cl15: C 28.49. Found: C 28.63.
’ ASSOCIATED CONTENT
S
Supporting Information. Table S1. Decomposition of
b
the computed 11B isotropic shielding constants into diamagnetic
and paramagnetic terms for B(C6F5)3, B(C6F5)2(C6Cl5) (3),
B(C6F5)(C6Cl5)2 (5), and B(C6Cl5)3 (6). CIF data for 3, 5 and
6, and complete ref 59. This material is available free of charge via
the Internet at http://pubs.acs.org.
(8) (a) Momming, C. M.; Otten, E.; Kehr, G.; Frohlich, R.; Grimme,
S.; Stephan, D. W.; Erker, G. Angew. Chem., Int. Ed. 2009, 48, 6643–
6646. (b) Zhao, X.; Stephan, D. W. Chem. Commun. 2011, 47, 1833–
1835. (c) Appelt, C.; Westenberg, H.; Bertini, F.; Ehlers, A. W.;
Slootweg, J. C.; Lammertsma, K.; Uhl, W. Angew. Chem., Int. Ed.
2011, 50, 3925–3928. (d) Tran, S. D.; Tronic, T. A.; Kaminsky, W.;
Heinekey, D. M.; Mayer, J. M. Inorg. Chim. Acta 2011, 369, 126–132.
(9) (a) Neu, R. C.; Otten, E.; Lough, A.; Stephan, D. W. Chem. Sci.
2011, 2, 170–176. (b) Otten, E.; Neu, R. C.; Stephan, D. W. J. Am. Chem.
Soc. 2009, 131, 9918–9919.
’ AUTHOR INFORMATION
Corresponding Author
a.ashley@imperial.ac.uk; dermot.ohare@chem.ox.ac.uk
Present Addresses
‡Department of Chemistry, Imperial College London, London
SW7 2AZ
(10) (a) Ashley, A. E.; Thompson, A. L.; O’Hare, D. Angew. Chem.,
Int. Ed. 2009, 48, 9839–9843. (b) Menard, G.; Stephan, D. W. J. Am.
Chem. Soc. 2010, 132, 1796–1797.
(11) (a) Spies, P.; Schwendemann, S.; Lange, S.; Kehr, G.; Froehlich,
R.; Erker, G. Angew. Chem., Int. Ed. 2008, 47, 7543–7546. (b) Chase,
P. A.; Jurca, T.; Stephan, D. W. Chem. Commun. 2008, 1701–1703.
(12) Beckett, M. A.; Strickland, G. C.; Holland, J. R.; Varma, K. S.
Polymer 1996, 37, 4629–4631.
(13) Eros, G.; Mehdi, H.; Papai, I.; Rokob, T. A.; Kiraly, P.; Tarkanyi,
G.; Soos, T. Angew. Chem., Int. Ed. 2010, 49, 6559–6563.
(14) Piers, W. E.; Irvine, G. J.; Williams, V. C. Eur. J. Inorg. Chem.
2000, 2131–2142.
’ ACKNOWLEDGMENT
We thank Balliol College, Oxford for a Junior Research
Fellowship to A.E.A. and the Royal Society for a University
Research Fellowship to G.G.W. We also thank EPSRC and the
University of Oxford Challenge Seed Fund for financial support
and Dr Jeffrey Harmer (CAESR, University of Oxford) for
assistance with the EPR spectroscopy.
’ REFERENCES
(1) (a) Massey, A. G.; Park, A. J.; Stone, F. G. A. Proc. Chem. Soc.,
London 1963, 212. (b) Massey, A. G.; Park, A. J. J. Organomet. Chem.
1964, 2, 245–250. (c) Massey, A. G.; Park, A. J. J. Organomet. Chem.
1966, 5, 218–225.
(15) McDaniel, D. S.; Brown, H. C. J. Org. Chem. 1958, 23, 420–427.
(16) Christoph Elschenbroich, A. S. Organometallics: A Concise
Introduction, 2nd ed.; Wiley VCH: Weinheim, 1992.
14739
dx.doi.org/10.1021/ja205037t |J. Am. Chem. Soc. 2011, 133, 14727–14740

Journal of the American Chemical Society
ARTICLE
(17) (a) Niedenzu, K. Organomet. Chem. Rev. 1966, 1, 305–329. (b)
Parks, D. J.; Piers, W. E.; Yap, G. P. A. Organometallics 1998, 17,
5492–5503. (c) Sundararaman, A.; Jakle, F. J. Organomet. Chem. 2003,
681, 134–142.
(18) Rausch, M. D.; Tibbetts, F. E.; Gordon, H. B. J. Organomet.
Chem. 1966, 5, 493.
(44) Compton, R. G.; Banks, C. E. Understanding Voltammetry;
World Scientific Publishing: Singapore, 2007.
(45) Testa, A. C.; Reinmuth, W. H. Anal. Chem. 1961, 33, 1320–1324.
(46) (a) Ballester, M.; Riera-Figueras, J.; Castaner, J.; Badfa, C.;
Monso, J. M. J. Am. Chem. Soc. 1971, 93, 2215–2225. (b) Ballester, M.
Acc. Chem. Res. 1985, 18, 380–387.
(19) Welch, G. C.; Juan, R. R. S.; Masuda, J. D.; Stephan, D. W.
(47) Chivers, T.; Schatte, G. Eur. J. Inorg. Chem. 2003, 3314–3317.
Science 2006, 314, 1124–1126.
(48) Strauss, S. H. Chem. Rev. 1993, 93, 927–942.
(20) Chivers, T.; David, B. J. Organomet. Chem. 1968, 13, 177–186.
(21) Williams, V. C.; Piers, W. E.; Clegg, W.; Elsegood, M. R. J.;
Collins, S.; Marder, T. B. J. Am. Chem. Soc. 1999, 121, 3244–3245.
(22) Confirmation of the formation of a four coordinate complex is
revealed by measuring Δδm,p, the difference of the chemical shifts
between the para-F and meta-F resonances, which decrease from 18.9
ppm to ca. 9 ppm upon rehybridization from a 3- to 4-coordinate
complex.
(23) Horton, A. D.; deWith, J. Organometallics 1997, 16, 5424–5436.
(24) Crystals of 3 had to be manipulated at temperatures below 0 °C
to inhibit dissolution in either Paratone or perfluoropolyether oil.
(25) For consistency, the right-handed propeller is shown in all
crystallographic figures.
(26) A common test for planarity is to sum-all-angles around a
central atom and comparing with the ideal 360°; in this case however the
sum of the angles about B is 360°. However, this test this can be
extremely insensitive to planarity and the reader is directed to ref 27 for
examples. Using the method applied in the text, the largest deviation is
seen for 5 (0.0028 Å), confirming a planar assignment for 3, 5 and 6.
(27) Blake, A. J.; Cole, J. M.; Evans, J. S. O.; Main, P.; Parsons, S.;
Watkin, D. J. Crystal Structure Analysis: Principles and Practice, 2nd ed.;
Oxford University Press: Oxford, 2009.
(49) Frohn, H. J.; Jakobs, S. J. Chem. Soc., Chem. Commun.
1989, 625–627.
(50) Gritzner, G.; Kuta, J. Pure Appl. Chem. 1984, 56, 461.
(51) Kwaan, R. J.; Harlan, C. J.; Norton, J. R. Organometallics 2001,
20, 3818–3820.
(52) Cairncross, A.; Sheppard, W. A.; Wonchoba, E. Org. Synth.
1988, 50À9, 875–882.
(53) Cosier, J.; Glazer, A. M. J. Appl. Crystallogr. 1986, 19, 105–107.
(54) Otwinowski, Z. ;Minor, W Processing of X-ray Diffraction Data
Collected in Oscillation Mode In Methods in Enzymology; Carter, C. W.,
Sweet, R. M., Eds.; Academic Press: New York, 1997; p 276.
(55) Altomare, A.; Cascarano, G.; Giacovazzo, C.; Guagliardi, A.;
Burla, M. C.; Polidori, G.; Camalli, M. J. Appl. Crystallogr. 1994, 27, 435.
(56) Betteridge, P. W.; Carruthers, J. R.; Cooper, R. I.; Prout, K.;
Watkin, D. J. J. Appl. Crystallogr. 2003, 36, 1487. Cooper, R. I.;
Thompson, A. L.; Watkin, D. J. J. Appl. Crystallogr. 2010, 43, 1100–1107.
(57) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648–5652. (b) Lee,
C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785–789. (c) Vosko,
S. H.; Wilk, L.; Nusair, M. Can. J. Phys. 1980, 58, 1200–1211. (d)
Stephens, P. J.; Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J. J. Phys.
Chem. 1994, 98, 11623–11627.
(58) Sch€afer, A.; Huber, C.; Ahlrichs, R. J. Chem. Phys. 1994, 100,
(28) Tulinsky, A.; White, J. G. Acta Crystallogr. 1958, 11, 7–14.
(29) Power, P. P.; Olmstead, M. M. J. Am. Chem. Soc. 1986,
108, 4253–4236.
5829–5835.
(59) Frisch, M. J. et al. Gaussian 03, Revision E.01; Gaussian, Inc.:
Wallingford CT, 2003.
(30) Zettler, F.; Hausen, H. D.; Hess, H. J. Organomet. Chem. 1974,
72, 157–162.
(60) (a) Ditchfield, R. Mol. Phys. 1974, 27, 789–807. (b) Wolinski,
K.; Hinton, J. F.; Pulay, P. J. J. Am. Chem. Soc. 1990, 112, 8251–8260.
(61) Tomasi, J.; Mennucci, B.; Cammi, R. Chem. Rev. 2005,
105, 2999–3093.
(31) Massey, A. G.; Park, A. J. J. Organomet. Chem. 1966, 5, 218–225.
(32) (a) Danopoulos, A. A.; Galsworthy, J. R.; Green, M. L. H.;
Cafferkey, S.; Doerrer, L. H.; Hursthouse, M. B. Chem. Commun.
1998, 2529–2530. (b) Doerrer, L. H.; Green, M. L. H. J. Chem. Soc.,
Dalton Trans. 1999, 4325–4329. (c) Beringhelli, T.; Maggioni, D.;
D’Alfonso, G. Organometallics 2001, 20, 4927–4938.
(33) Timoshkin, A. Y.; Frenking, G. Organometallics 2008, 27,
371–380.
(62) Jensen, F. J. Chem. Theory Comput. 2008, 4, 719–727.
(34) Bergquist, C.; Bridgewater, B. M.; Harlan, C. J.; Norton, J. R.;
Friesner, R. A.; Parkin, G. J. Am. Chem. Soc. 2000, 122, 10581–10590.
(35) Bergquist, C.; Parkin, G. J. Am. Chem. Soc. 1999, 121,
6322–6323.
(36) (a) Childs, R. F.; Mulholland, D. L.; Nixon, A. Can. J. Chem.
1982, 60, 801–808. (b) Lappert, M. F. J. Chem Soc. 1961, 817. (c)
Lappert, M. F. J. Chem Soc. 1962, 542. (d) Mayer, U.; Gutmann, V.;
Gerger, W. Monatsh. Chem. 1975, 106, 1235–1257. (e) Gutmann, V.
Coord. Chem. Rev. 1976, 18, 225–255.
(37) (a) Welch, G. C.; Cabrera, L.; Chase, P. A.; Hollink, E.; Masuda,
J. D.; Wei, P. R.; Stephan, D. W. Dalton Trans. 2007, 3407–3414. (b)
Beckett, M. A.; Strickland, G. C.; Holland, J. R.; Varma, K. S. Polymer
1996, 37, 4629–4631.
(38) Britovsek, G. J. P.; Ugolotti, J.; White, A. J. P. Organometallics
2005, 24, 1685–1691.
(39) Pearson, R. G. J. Am. Chem. Soc. 1963, 85, 3533–&.
(40) Mackle, H.; Sutton, L. E. Trans. Faraday Soc. 1951, 47,
691–699.
(41) Bandoli, G.; Bortoloz., G; Clemente, D. A.; Croatto, U.;
Panatton., C J. Chem. Soc. A 1970, 2778–&.
(42) Harlan, C. J.; Hascall, T.; Fujita, E.; Norton, J. R. J. Am. Chem.
Soc. 1999, 121, 7274–7275.
(43) Cummings, S. A.; Iimura, M.; Harlan, C. J.; Kwaan, R. J.; Trieu,
I. V.; Norton, J. R.; Bridgewater, B. M.; Jakle, F.; Sundararaman, A.;
Tilset, M. Organometallics 2006, 25, 1565–1568.
14740
dx.doi.org/10.1021/ja205037t |J. Am. Chem. Soc. 2011, 133, 14727–14740