Reactivity of a frustrated Lewis pair and small-molecule activation by an isolable arduengo carbene-B{3,5-(CF3)2C6H 3}3 complex

Article abstract of DOI:10.1002/chem.201202840

Tris[3,5-bis(trifluoromethyl)phenyl]borane reacts with the sterically demanding Arduengo carbenes 1,3-di-tert-butylimidazolin-2-ylidene and 1,3-bis(2,6-diisopropylphenyl)imidazolin-2-ylidene to form isolable normal adducts. In the case of 1,3-di-tert-butylimidazolin-2-ylidene, the adduct exhibits dynamic behaviour in solution and frustrated-Lewis-pair (FLP) reactivity. Fast cleavage of dihydrogen and THF, the C-H activation of phenylacetylene, and carbon dioxide fixation were achieved by using solutions of this adduct in benzene. This adduct is stable at room temperature in the absence of suitable substrates; however, thermal rearrangement into an abnormal carbene-borane adduct can be observed. In contrast, the 1,3-bis(2,6- diisopropylphenyl)imidazolin-2-ylidene adduct exhibits no evidence of FLP reactivity or of dissociation in solution. DFT calculations confirmed the experimental behaviour and stability of these carbene-borane adducts. Frustrated but not frustrating: Arduengo carbene 1,3-di-tert-butylimidazolin-2-ylidene and tris[3,5-bis(trifluoromethyl)phenyl]borane form an isolable adduct (see scheme) that shows frustrated-Lewis-pair (FLP) reactivity towards H₂, CO₂, and other small molecules. Copyright

Full text of DOI:10.1002/chem.201202840

DOI: 10.1002/chem.201202840
Reactivity of a Frustrated Lewis Pair and Small-Molecule Activation by an
À
Isolable Arduengo Carbene B
{3,5-
N
Eugene L. Kolychev, Thomas Bannenberg, Matthias Freytag, Constantin G. Daniliuc,
Peter G. Jones, and Matthias Tamm*[a]
Dedicated to Professor Anthony J. Arduengo III on the occasion of his 60th birthday
À
Abstract: Tris
phenyl]borane reacts with the sterically
demanding Arduengo carbenes 1,3-di-
[3,5-bis(trifluoromethyl)-
the C H activation of phenylacetylene,
and carbon dioxide fixation were
achieved by using solutions of this
adduct in benzene. This adduct is
stable at room temperature in the ab-
sence of suitable substrates; however,
thermal rearrangement into an abnor-
mal carbene–borane adduct can be ob-
served. In contrast, the 1,3-bis(2,6-di-
tert-butylACHTUNGTRENNUNGimidazolin-2-ylidene and 1,3-
bis(2,6-diisopropylphenyl)imidazolin-2-
ylidene to form isolable normal ad-
ducts. In the case of 1,3-di-tert-butyl-
ACHTUNGTRENUNiNG sopropylphenyl)imidazolin-2-ylidene
adduct exhibits no evidence of FLP re-
activity or of dissociation in solution.
DFT calculations confirmed the experi-
mental behaviour and stability of these
carbene–borane adducts.
Keywords: boranes
· carbenes ·
ACHTUNGTRENNUNGimidazolin-2-ylidene, the adduct exhib-
frustrated Lewis pairs · noncovalent
interactions · small-molecule activa-
tion
its dynamic behaviour in solution and
frustrated-Lewis-pair (FLP) reactivity.
Fast cleavage of dihydrogen and THF,
Introduction
FLP to promote H2 splitting can be ascribed to an enhanced
cumulative acid–base strength.[13] However, in the absence
of reactive substrates, self-deactivation and the irreversible
formation of the abnormal carbene–borane adduct (3) were
observed,[8] whereas the normal adduct (2a) could not be
isolated (Scheme 1). In contrast, sterically less-demanding
carbene 1b (R=2,6-diisopropylphenyl, Dipp)[14] gave the
isolable normal adduct 2b with no indication of FLP reactiv-
Since the first report of its preparation in 1963,[1] tris(penta-
fluorophenyl)borane, BACHTUNRGTNE(NUG C6F5)3, has emerged from relative
“obscurity to an industrially important commodity”,[2] in
particular because of its extensive use as a co-catalyst in ho-
mogeneous Ziegler–Natta olefin-polymerisation reactions.[3]
Nowadays, BACHTUNGTRENNUNG(C6F5)3 has become firmly established as a
“special boron Lewis acid for special reactions”,[4] that is, as
a catalyst or as a stoichiometric reagent for numerous or-
ganic and organometallic transformations.[2–5] More recently,
this borane has found additional widespread application as
the most prominent Lewis acid component in so-called frus-
trated Lewis pairs (FLPs), which are prevented from form-
ing normal adducts by steric factors and, therefore, are capa-
ble of mutually activating small molecules, such as dihydro-
gen.[6,7]
ity.[9] The different behaviour of these pairs, 1a/B
ACTHNGUTERN(NUG C6F5)3 and
Thus, the heterolytic cleavage of dihydrogen,[8,9] as well as
the activation of other small molecules,[10,11] have been ac-
complished by a combination of BACHTNUTRGNE(UNG C6F5)3 with bulky N-het-
erocyclic carbene (“Arduengo carbene”, NHC) 1a (R=
tBu),[12] whereby the particularly strong propensity of this
[a] Dr. E. L. Kolychev, Dr. T. Bannenberg, Dr. M. Freytag,
Dr. C. G. Daniliuc, Prof. Dr. P. G. Jones, Prof. Dr. M. Tamm
Institut fꢀr Anorganische und Analytische Chemie
Technische Universitꢁt Braunschweig
Hagenring 30, 38106 Braunschweig (Germany)
Fax : (+49)531-391-5309
E-mail: m.tamm@tu-bs.de
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201202840.
Scheme 1. The formation of normal and abnormal carbene–borane ad-
ducts with B(C6F5)3; Dipp=2,6-diisopropylphenyl, tBu=tert-butyl.
AHCTUNGTRENNUNG
16938
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16938 – 16946

FULL PAPER
1b/BACHTUNGTRENNUNG(C6F5)3, can be conveniently rationalized by comparing
the stability of their normal adducts, which indicated a much
higher stability of the isolable 2b (DE=À51.5 kcalmolÀ1)
compared to elusive adduct 2a (DE=À23.8 kcalmolÀ1; see
Table 1 and discussion below).[15] Whereas the abnormal
Table 1. Energies [kcalmolÀ1] for the formation of normal and abnormal
carbene–borane adducts.[a]
Compound
(R)
Adduct
type
Borane
DEM05À2X DEM06À2X DEB97ÀD
2a (tBu)
2b (Dipp)
3 (tBu)
4a (tBu)
4b (Dipp)
5 (tBu)
normal
normal
abnormal
normal
normal
abnormal
B
B
B
N
À23.8[b]
À51.5[b]
À44.2[b]
À28.4
À59.0
À46.2
À29.4
À61.9
À44.2
À24.8
À56.7
À43.8
À36.7
À69.4
À47.6
B
BACHTUNGTRENNUNG
G
BACHTUNGTRENNUNG
[a] DE=uncorrected zero-point M05-2X, M06-2X, and B97-D/6-311G-
ACHTUNGTRENNUNG(d,p) energies; for full thermodynamic data, including DE0K, DH298K, and
DG298K, see the Supporting Information. [b] These values have previously
Scheme 2. The formation of normal and abnormal carbene–borane ad-
ducts with BACTHNUGTRNEUNG(m-XyF6)3; Dipp=2,6-diisopropylphenyl, tBu=tert-butyl.
been reported in refs. [8] and [15].
from C6F5MgBr and boron trifluoride etherate,[1c] turned out
to be quite similar to that reported by Ashley and co-work-
ers.[21] Hence, the Grignard reagent,[23] which was prepared
according to the Knochel method from 3,5-bis(trifluoro-
adduct (3) is significantly more stable (DE=À44.2 kcal
molÀ1),[8] we anticipated that adducts that are similar to
À
structure 2a should still become isolable if C H activation
at the 4,5-positions of the imidazole ring and rearrangement
into an abnormal carbene–borane complex are blocked.
A
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
Consequently, B
and 4,5-dihydro congeners of compound 1a, but the expect-
ed normal adducts avoided isolation through C F activation
G
white crystalline solid in 62–79% yield after evaporation,
twofold sublimation, and recrystallization from hot tol-
uene.[25] This procedure could be repeated several times,
thus reliably producing 10–15 g of the product. At room
À
or C H activation, respectively.
temperature, BACTHNUTRGNEUNG(m-XyF6)3 is insoluble in nonpolar hydrocar-
bon solvents and is only slightly soluble in aromatic and
chlorinated solvents. Dissolution in donor solvents leads to
the formation of solvate complexes, as evidenced by
11B NMR spectroscopy (d=11.7 ppm in [D8]THF and d=
10.9 ppm in [D6]acetone).
Results and Discussion
It has also been concluded that the reactivity and resulting
self-destruction of carbene–borane Lewis pairs could not
only be moderated by variation in the carbene,[15] but also
by employing other Lewis acids.[17] In this regard, we identi-
[3,5-bis(trifluoromethyl)phenyl]borane, B
(m-XyF6 =hexafluoro-meta-xylyl, 3,5-(CF3)2C6H3), as an in-
In contrast to the observed reactivity of compounds 1a
and 1b towards B
carbenes with B
ACHTUNGTRENNUNG
fied tris
G
E
ACHTUNGTRENNUNG
adducts (4a and 4b) in high yield. Their 11B and 19F NMR
spectra (in C6D6) exhibit one resonance each, at d=À5.9/
À62.8 ppm (4a) and at d=À8.0/À62.2 ppm (4b), thus indi-
teresting candidate, even though, considering the promi-
nence of the corresponding borate ion, [BACTHNUTRGNEUNG
(m-XyF6)4]À
(BARF), as a weakly coordinating anion,[18] surprisingly
little reference has been made to this borane, apart from in
patent literature (Scheme 2).[19] Until very recently, the only
cating free rotation around the C B bonds at room temper-
À
ature on the NMR timescale.[26] The formation of normal
carbene–borane adducts was unambiguously confirmed by
X-ray diffraction analysis and their molecular structures
reported preparation of BACHTNUTRGNE(UNG m-XyF6)3 occurred accidentally
À
by the degradation of the BARF ion in the presence of a
cationic platinum(II) complex, whereby its molecular struc-
ture could be established by X-ray diffraction.[20] However,
much to our surprise, Ashley and co-workers have now in-
(Figure 1) show B C1 bond lengths of 1.675(4) ꢃ (4a) and
1.662(3) ꢃ (4b). These distances are similar to the corre-
À
sponding B C bond lengths that were established for com-
pound 2b in two different crystal modifications (1.663(5)
dependently reported a preparative procedure for B
A
and 1.684(2) ꢃ).[9,15] However, these bond lengths are longer
XyF6)3 and its use as an FLP component together with
2,2,6,6-tetramethyl-piperidine (TMP).[21] Following our first
than that observed in the BACTHNGUTREN(NUG C6F5)3 adduct of the sterically
less encumbered NHC 1,3,4,5-tetramethylimidazolin-2-yli-
unsuccessful attempts to prepare B
(C6F5)3 from 3,5-bis(trifluoromethyl)phenyl bromide by re-
action with nBuLi and BCl3,[22] our final procedure, which
was based on an early report on the synthesis of B(C6F5)3
ACHTUNGRTEN(NUNG m-XyF6)3 in analogy to
dene (1.641(2) ꢃ).[27]
B
A
In contrast to this structure, steric congestion in com-
pounds 4a and 4b is indicated by a significant tilting of the
À
A
imidazole plane with respect to the B C1 axis, whereby the
Chem. Eur. J. 2012, 18, 16938 – 16946
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16939

M. Tamm et al.
and 1b (R=Dipp) is nonetheless somewhat puzzling, even
in view of the fact that NHC 1a is clearly a bigger ligand
than compound 1b when referring, for instance, to the “per-
cent buried volume”, as introduced by the groups of Nolan
and Cavallo, to measure the steric bulk of a NHC.[31] Apart
from bulkiness, the clear difference between the two car-
benes is that compound 1a contains N-alkyl groups, whereas
compound 1b represents an aromatic N-substituted NHC
ligand. For transition-metal complexes of this latter class, it
has recently been established that, in addition to the con-
ventional picture of metal–NHC bonding,[32] a new mode of
interaction occurs through p-face donation from the aromat-
ic substituents, that involves, in particular, the ipso-carbon
atoms.[33,34] Calculations revealed the operation of two dif-
Figure 1. ORTEPs of compounds 4a (left) and 4b (right); thermal ellip-
soids are set at 50% probability. Hydrogen and fluorine atoms are omit-
ferent mechanisms: 1) direct donation to empty metal orACHTUNGTRENNUNGbit-
AHCTUNGTREGaNNNU ls, and 2) donation to another arene moiety that is bound
À
ted for clarity. Selected bond lengths [ꢃ] and angles [8]: 4a: B C1
to the metal atom,[34] as evidenced, for instance, by p p in-
teractions between mesitylene and benzylidene groups in
the solid-state structures of second-generation Grubbs-type
catalysts.[35] Evaluation of compound 4b for similar interac-
À
À
À
À
À
À À
1.675(4), N1 C1 1.375(3), N2 C1 1.386(3); N1 C1 N2 104.9(2), C1 B
À À
À À
À
C12 107.80(18), C1 B C20 109.0(2), C1 B C28 111.9(2). 4b: B C1
À
À
À
À
À
1.662(3), N1 C1 1.368(3), N2 C1 1.387(3); N1 C1 N2 104.20(17), C1
À À À À
À
B C28 105.36(16), C1 B C36 112.74(17), C1 B C44 114.05(17).
tions reveals several very short intramolecular C···C disACHTUNGTRENNUNGtanc-
À
boron atoms lie 0.47 ꢃ (4a) and 0.39 ꢃ (4b) out of the C1
ACHTUNGTRENeNGNU s that are well below the sum of their van der Waals radii
N1 C2 C3 N2 plane. In addition, the N-bonded substitu-
ents in both molecules are twisted towards one side and the
CMe3 carbon atoms in compound 4a are significantly more
(rvdW(C)=1.70 ꢃ);[36] the shortest contacts involve the ipso-
carbon atoms of the Dipp substituents and the ipso- and
À
À
À
À
ortho-carbon atoms of the XyF6 rings, namely C4 C28
À
À
À
strongly displaced (0.49/0.45 ꢃ) with respect to the im
azole plane than the ipso-carbon atoms in compound 4b
(0.38/0.19 ꢃ).
id-
3.23 ꢃ, C4 C33 3.16 ꢃ, C16 C44 3.16 ꢃ, and C16 C45
3.20 ꢃ. A re-inspection of the crystal structure of compound
2b unveils even shorter distances of 2.97 and 3.01 ꢃ for the
ACHTUNGTRENNUNG
Dipp Cipso atoms[15] and, therefore, we propose that p-face
donation in compounds 2b and 4b contributes significantly
to the overall stability of these adducts and to their resulting
lack of FLP reactivity (see below).[37]
À
Based on their solid-state structures, compound 4a is ap-
parently more strongly distorted than compound 4b and,
therefore, might exhibit a higher degree of “frustration” and
reactivity. As we have recently demonstrated, the thermody-
namics of carbene–borane-adduct formation can be used as
a good measure to rationalize and predict their FLP behav-
Because the DFT calculations indicate that adducts 4a
and 4b have significantly different stabilities, we set out to
probe their behaviour in solution further by using 2D ex-
change NMR spectroscopy (2D EXSY). At room tempera-
iour;[15] therefore, the energies for the formation of B
and B(m-XyF6)3 adducts 2a/2b and 4a/4b were calculated
ACHTNUGRTEN(NGNU C6F5)3
ACHTUNGTRENNUNG
1
by using DFT methods. We employed the M05-2X, M06-2X,
and B97-D functionals that were developed by Zhao and
Truhlar[28] and by Grimme[29] to conveniently describe the
noncovalent and long-term dispersion interactions that were
expected to contribute significantly to the overall binding in
adducts 2 and 4 (Table 1).[30] In agreement with the previ-
ously observed reactivity, FLP 2a is significantly less stable
than compound 2b and the energy difference (DDE
ꢀ30 kcalmolÀ1) is well-reproduced with all three function-
als. The relative stabilities of compounds 4a and 4b are very
similar (DDE=27.0–32.7 kcalmolÀ1) and the energies of for-
mation decrease in the order M05-2X>M06-2X>B97-D. It
is noteworthy that only the latter functional affords marked-
ture, H,1H EXSY spectra of mixtures of compounds 1a/4a
and 4a/BACTHUNGTRNEUNG(m-XyF6)3 in [D8]toluene showed exchange peaks
between the resonances that were assigned to compound 4a
and to the uncoordinated species (for a detailed representa-
tion of selected spectra, see the Supporting Information).
These signals could be observed even after short mixing
times, thus confirming a fast exchange between uncoordinat-
ed and coordinated carbene and borane species in solution.
In contrast, mixtures of compounds 1b/4b and 4b/BACHTUNGTRENNUNG(m-
XyF6)3 did not afford spectra with any indication of ex-
change in solution at room temperature, which was in full
agreement with the significantly higher calculated stability
of compound 4b compared to compound 4a (DDEꢀ30 kcal
molÀ1; Table 1). In line with these findings, the addition of
one equivalent of compound 1b to a solution of compound
4a in C6D6 led to complete carbene exchange and to the
quantitative formation of a mixture of compounds 1a/4b, as
indicated by NMR spectroscopy (Scheme 3).
ly different stabilities for B
N
=
À24.8 kcalmolÀ1) and its
BACHTUNGTRENNUNG
DEB97ÀD =À36.7 kcalmolÀ1), which accounts for the elusive-
ness of adduct 2a in contrast to the successful isolation of
compound 4a (see above).
Although it is in complete agreement with the experimen-
tal findings, this vast calculated difference in stability be-
tween the borane adducts that contain NHCs 1a (R=tBu)
In light of the dynamic behaviour of compound 4a in sol-
ution, we set out to experimentally investigate this adduct
for potential FLP reactivity. Compound 4a is indefinitely
16940
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16938 – 16946

À
Isolable Arduengo Carbene BACHTUNGTRENN{GU 3,5-CAHTUTGNNER(NNGU CF3)2C6H3}3 Complex
FULL PAPER
The ring-opening of THF is another characteristic reac-
tion of carbene–borane and phosphane–borane FLP sys-
tems.[8,15,16] Therefore, compound 4a was dissolved in THF
and stirred for 24 h. Evaporation of the solvent afforded a
white solid, NMR characterisation of which indicated the
presence of the expected (CH2)4O-bridged product (6) to-
Scheme 3. Solution-phase behaviour of normal carbene–borane adduct
gether with the abnormal adduct (5) in
a 3:2 ratio
4a.
(Scheme 4). This reaction was not optimised further, but the
formation of both products indicated FLP behaviour.
stable in benzene at room temperature and also for at least
one hour under reflux conditions (T=808C), whereas heat-
ing in toluene at 1108C resulted in fast rearrangement into
abnormal adduct 5 within 15 min. On cooling a hot solution
of compound 4a in toluene, adduct 5 was isolated as a col-
ourless crystalline solid in 92% yield. Single resonances are
found in the 11B and 19F NMR spectra (in [D6]acetone) at
d=À7.4 and À62.3 ppm and the observation of two doublets
at d=9.10 and 6.84 ppm for the imidazole CH hydrogen
atoms and of two singlets at d=1.62 and 1.28 ppm for the
tBu methyl groups indicate the formation of adduct 5 with
an asymmetrically bound abnormal NHC ligand. X-ray dif-
À
fraction analysis revealed a B C2 distance of 1.661(2) ꢃ,
which is very close to the values for compounds 4a and 4b.
In contrast to the corresponding normal adduct 4a, the ab-
sence of any unusual structural distortions indicates the
presence of an unstrained adduct (Figure 2). DFT calcula-
Scheme 4. FLP reactivity of normal carbene–borane adduct 4a.
Another important issue in FLP chemistry is CO2 fixa-
tion;[38] consequently, the reactivity of compound 4a towards
carbon dioxide was studied. When CO2 was bubbled
through a clear solution of compound 4a in hot benzene
(T=608C), a white precipitate formed instantaneously and
the CO2 adduct (7) could be isolated in 66% yield by filtra-
tion and crystallisation from CH2Cl2/n-hexane (Figure 3).
Alternatively, the exposure of a suspension of compound 4a
in benzene to CO2 at room temperature for 24 h afforded
1
compound 7 in higher yield (86%) and purity. The H and
19F NMR spectra of compound 7 are very similar to those of
compound 4a, whilst its 13C NMR spectrum exhibits an ad-
ditional resonance at d=157.7 ppm for the CO2 carbon
atom; the 11B NMR resonance is found at d=4.7 ppm. X-
ray diffraction analysis confirmed the formation of a CO2-
Figure 2. ORTEP of compound 5; thermal ellipsoids are set at 50% prob-
ability. Hydrogen and fluorine atoms are omitted for clarity. Selected
À
À
À
bond lengths [ꢃ] and angles [8]: N1 C1 1.338(2), N1 C2 1.428(2), N1
À
À
À
À
C4 1.521(2), N2 C1 1.324(2), N2 C3 1.383(2), B C2 1.661(2), B C12
À
À
bridged zwitterion with C1 C12 and B O2 distances of
1.5289(19) and 1.5910(18) ꢃ, respectively. The CO2 frag-
ment is planar (angular sum of 359.988) and displays two
À
À
À
À
À
1.663(2), B C20 1.639(2), B C28 1.639(2); N1 C1 N2 110.05(14), N1
C2 B 133.21(13), C3 C2 B 123.43(14), C12 B C2 106.50(12), C20 B
C2 110.77(13), C28 B C2 113.19(13).
À
À
À
À À
À À
À À
À
À
distinctly different C O bond lengths of 1.2141(18) ꢃ (C12
À
O1) and 1.2930(18) ꢃ (C12 O2).
tions confirm the expected higher stability of adduct 5 rela-
tive to compound 4a; comparison with the corresponding
couple 3/2a reveals a smaller difference in energy, which
varies with the applied functionals (Table 1). Again, the
B97-D functional affords the highest stabilities for com-
pounds 4a and 5, whilst producing the smallest difference
(DDE=11.0 kcalmolÀ1).
The activation of phenylacetylene had previously been
observed for the FLP 1a/BACHTNUTGRNEUNG
(C6F5)3;[11] similarly, the addition
of phenylacetylene to a suspension of compound 4a in ben-
zene afforded a yellow oil, which was separated after stirring
the reaction mixture for 20 min. Cooling a hot solution of
this oil in C6H6/CH2Cl2 afforded colourless crystals of com-
pound 8 in 91% yield. 11B and 19F NMR analysis of these
Chem. Eur. J. 2012, 18, 16938 – 16946
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16941

M. Tamm et al.
Figure 3. ORTEP of compound 7; thermal ellipsoids are set at 50% prob-
ability. Hydrogen and fluorine atoms are omitted for clarity. Selected
À
À
bond lengths [ꢃ] and angles [8]: N1 C1 1.3507(19), N2 C1 1.3812(19),
À
À
À
À
C1 C12 1.5289(19), O1 C12 1.2141(18), O2 C12 1.2930(18), B O2
À
À
À
À
1.5910(18), B C1 1.675(4), B C13 1.627(2), B C21 1.634(2), B C29
Figure 4. ORTEP of one independent ion pair of compound 8; thermal
ellipsoids are set at 50% probability. Hydrogen and fluorine atoms are
À
À
À
À
À
À
1.627(2); N1 C1 N2 108.42(12), N1 C1 C12 126.63(13), N2 C1 C12
À
À
À
À
À
À
124.17(13), O1 C12 C1 116.98(13), O2 C12 C1 115.09(12), O1 C12
À
omitted for clarity. Selected bond lengths [ꢃ] and angles [8]: B C12
À À
À À
À À
O2 127.92(13), O2 B C13 103.95(11), O2 B C29 109.69(11), O2 B
C21 105.97(11).
À
À
À
À
1.593(3), B C20 1.644(3), B C28 1.647(3), B C36 1.643(3), C12 C13
À
À
À
À À
1.210(3), C13 C14 1.442(3); N1 C1 N2 110.1(2), C12 B C20
À À
À À
À
À
107.21(18), C12 B C28 110.07(18), C12 B C36 110.70(18), C13 C12 B
À
À
179.3(2), C12 C13 C14 178.2(2).
crystals show signals at d=À8.9 and À62.3 ppm and the
1H NMR spectrum exhibits a characteristic low-field triplet
(d=7.35 ppm) that can be assigned to an imidazolium
À
NC(H)N hydrogen atom, thus indicating that heterolytic C
H bond cleavage and the formation of the imidazolium
phenACHTUNGTRENNUNGylalkynylborate (8) had occurred. Single crystals of
8·C6D6 that were suitable for X-ray diffraction were ob-
tained by the slow evaporation of CD2Cl2 from a solution of
compound 8 in C6D6/CD2Cl2. The asymmetric unit contains
two crystallographically independent cations and anions
Scheme 5. Heterolytic cleavage of dihydrogen by carbene–borane adduct
4a.
with similar structural parameters; Figure 4 shows one imACTHNUTRGNEUNGid-
À
À
azolium phenylalkynylborate ion pair. The B C12 and C12
C13 bond lengths, 1.593(3) and 1.210(3) ꢃ, together with the
to another broad 11B NMR signal at d=À10.0 ppm. Elemen-
À
À
C12 C13 C14 angle, 178.2(2)8, are similar to the reported
tal analysis also confirmed the composition of 9·BACHTNGUTERNNU(G m-XyF6)3.
ꢁ
values for the related [PhC CB
(C6F5)3]À ion.[11]
Evidently, the reaction of compound 4a with H2 only con-
sumed half the amount of carbene 1a that was present in
adduct 4a and, consequently, the other half could be isolat-
ed and spectroscopically characterized from the benzene fil-
trate. Optimisation of the H2-activation procedure was
achieved by exposing a 1:1 mixture of compound 4a and B-
To test whether compound 4a was capable of promoting
the heterolytic cleavage of dihydrogen with the concomitant
formation of the imidazolium hydroborate (9, Scheme 5), a
suspension of compound 4a in benzene was stirred over-
night under a H2 atmosphere (1 atm), thereby affording a
white precipitate. The solid was isolated by filtration and
washed with CH2Cl2; it was appreciably soluble only in
THF. Its 1H NMR spectrum in [D8]THF exhibits a triplet
and a doublet at d=8.78 and 7.90 ppm, respectively, in line
with the formation of an imidazolium salt. In addition,
broad resonances at d=7.87 and 7.68 ppm correspond to the
ortho- and para-hydrogen atoms of the XyF6 groups. How-
ever, their integrals indicate the presence of two borane
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
dine (py) to the [(m-XyF6)3B H B
pyridine adduct [(py)B(m-XyF6)3] and hydroborate ion [H
(m-XyF6)3]À. In our hands, the addition of a few drops of
MeCN to a suspension of 9·B(m-XyF6)3 in CD2Cl2 gave a
(m-XyF6)3]À salt afforded
ACHTUNGTRENNUNG
À
AHCTUNGTRENNUNG
BACHTUNGTRENNUNG
AHCTUNGTRENNUNG
clear solution that contained compound 9; the presence of
moieties and thus suggest the formation of 9·B
(m-XyF6)3,
the [H B
(m-XyF6)3]À ion was unambiguously confirmed by
À
À À
which contains the [(m-XyF6)3B H B
was also observed for the TMP/B
(m-XyF6)3]À anion, as
the observation of a quadruplet at d=3.63 ppm in the
(m-XyF6)3 pair (see
1H NMR spectrum and a doublet at d=À8.7 ppm in the
above).[21] Moreover, this anion gives rise to a broad
11B NMR spectrum with J
A
1
1H NMR resonance at d=3.75 ppm for the BHB moiety and
tional broad signals can be assigned to the acetonitrile
16942
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16938 – 16946

À
Isolable Arduengo Carbene BACHTUNGTRENN{GU 3,5-CAHTUTGNNER(NNGU CF3)2C6H3}3 Complex
FULL PAPER
on Finnigan MAT 95 (EI) and Finnigan MAT 95 XL (ESI) systems, re-
spectively. All solvents were purified by a solvent purification system
(MBraun) and stored over molecular sieves (4 ꢃ) prior to use. In addi-
tion, benzene and THF were purified by distillation over sodium/benzo-
phenone. Unless otherwise indicated, all of the starting materials were
obtained from Aldrich or Acros and were used without further purifica-
tion. Gaseous dihydrogen (99.999%) and carbon dioxide (99.5%) were
purchased from Westfalen AG and passed through a P2O5 column prior
to use. 1,3-Di-tert-butylimidazolin-2-ylidene (1a)[12] and 1,3-bis(2,6-diiso-
propylphenyl)imidazolin-2-ylidene (1b)[14] were prepared according to lit-
erature procedures.
adduct, [(CH3CN)B
plays a significantly higher tendency than BACHTUNGTRENNUNG
G
ACHTUNGRTNEN(UNG m-XyF6)3 dis-
hydrogen-bonded hydrodiborate,[39] which could be ascribed
to its high Lewis acidity along with its lower steric demand.
Finally, it should be emphasized that complexes 2b and
4b, which both contained Dipp-substituted carbene 1b, dis-
played no FLP-type reactivity; they were stable in THF and
solutions of these complexes in toluene did not react with
dihydrogen, even after long periods of time (2 weeks) or at
elevated temperatures. Compound 2b has been reported to
be unstable in solution and, indeed, it slowly decomposed in
refluxing toluene to afford a mixture of products.[9,15] In con-
trast, compound 4b remained stable under these conditions
and its NMR spectra in such solutions showed no signs of
decomposition, which, again, was in agreement with the
higher calculated stability of compound 4b compared to
those of compounds 2b and 4a (Table 1, see above).
X-ray crystallography: Crystallographic data are given in Table 2. Data
collection and reduction: Single crystals were mounted in inert oil and
transferred into a stream of cold gas in an Oxford Diffraction Xcalibur E
diffractometer. Intensities were recorded by using monochromated MoKa
radiation. Absorption corrections were based on multiple scans. Refine-
ment: Structures were refined anisotropically on F2 by using the program
SHELXL-97.[43] Hydrogen atoms were either 1) included as idealized
methyl groups that were allowed to rotate but not tip or 2) placed geo-
metrically and allowed to ride on their attached carbon atoms. Special
features: Except for compounds 5 and 7, the large number of trifluoro-
methyl groups led to problems with disorder (or at least high U values).
Disordered groups were treated by using appropriate systems of re-
straints to improve the stability of the refinement; details are given in
the CIF files. In addition, compound 4b contained disordered toluene
molecules that could not be refined satisfactorily; for this reason, the pro-
gram SQUEEZE[44] was used to mathematically remove the effects of
the solvent. Compound 8 crystallized with two independent molecules of
compound 8 and three independent molecules of deuterated benzene
(two with inversion symmetry) in the asymmetric unit. Short hydrogen
Conclusion
In this contribution, we have introduced a procedure for the
bulk preparation of the new fluorinated borane BACTHNUTRGNE(UNG m-XyF6)3
and have studied its use for the design of frustrated car-
bene–borane Lewis pairs. In combination with 1,3-di-tert-bu-
tylimidazolin-2-ylidene (1a), the isolable Lewis acid–base
adduct 4a “at the boundary of classical and frustrated-
Lewis-pair reactivity” was obtained;[40] this adduct partially
dissociated in solution and, therefore, was able to exhibit
bonds: Compound
7 displayed a short intermolecular interaction,
H2···O1 2.20 ꢃ.
CCDC-883184 (4a), CCDC-883185 (4b·0.5C7H8), CCDC-88318 (5),
CCDC-883187 (7), and CCDC-883188 (8·C6D6) contain the supplemen-
AHCTUNGERTGtNNUN ary crystallographic data for this paper. These data can be obtained free
of charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
À
FLP-type reactivity, such as CO2 fixation, C H activation,
and even dihydrogen splitting. This observed reactivity can
be interpreted in terms of the concept of “thermally induced
frustration”,[41] with the “reactive frustrated Lewis form” of
compound 4a being sufficiently populated at room tempera-
ture, whereas similar reactivity cannot be thermally induced
for the more stable adduct 4b in the studied temperature
range. However, in the absence of substrates, the rearrange-
ment of compound 4a into abnormal carbene–borane
adduct 5 can be achieved under comparatively forcing con-
ditions. Whilst this stable adduct has lost its FLP potential,
it can serve as a source of a new members of the recently es-
tablished class of anionic N-heterocyclic carbenes that con-
tain weakly coordinating borate moieties (WCA-NHC).[42]
BACHTNUGTRENNUG(m-XyF6)3: To a mixture of m-ACHUTNTGREN(NGUN CF3)2C6H3Br (0.08 mol, 23.44 g) and
THF (150 mL) in a 250 mL Schlenk flask under an argon atmosphere at
48C (ice bath) was added a 2.0m solution of iPrMgCl in THF (40 mL).
The reaction mixture was allowed to warm to RT and stirred for 4 h. The
mixture was cooled to À208C and a solution of freshly distilled Et2O·BF3
(3.79 g) in THF (40 mL) was added dropwise. The mixture was warmed
to RT and then heated at reflux for 2 h. The solvent was evaporated and
the remaining oil was dried in vacuo. The resulting solid was sublimed
twice at 1608C under in vacuum (0.3 mbar). During the first sublimation,
it was necessary to increase the temperature of the sublimation device
slowly to prevent contamination of the product by magnesium salts. The
product was further purified by recrystallisation from hot toluene. Yield:
10.69–13.63 g (62–79%); white crystalline powder; m.p. 217–2198C (slow
sublimation was observed at>1708C); b.p. 316–3208C; 1H NMR
(400 MHz, CD2Cl2, 258C): d=8.23 (br s, 3H; p-C6H HCUTNGTRNE(NUNG CF3)2), 8.02 ppm
3A
(CF3)2); 1H NMR (300 MHz, CDCl3, 258C): d=8.23
(CF3)2), 8.00 ppm (br s, 6H; o-C6H3A
(CF3)2); 1H NMR
(H,F)=0.6 Hz, 3H; p-C6H3-
(H,F)=0.6 Hz, 6H; o-C6H3A
(CF3)2); 13C NMR
(100 MHz, CD2Cl2, 258C): d=142.8 (br s; ipso-C6H3A(CF3)2), 138.5 (br s;
(CF3)2), 127.1 (sept, J-
(C,F)=273.9 Hz; CF3);
(br s, 6H; o-C6H
3A
(br s, 3H; p-C6H
G
CHTUNGTRENNUNG
(300 MHz, C6D6, 258C): d=7.87 (sept, 4J
R
Experimental Section
(CF3)2), 7.68 ppm (m, 4J
A
CHTUNGTRENNUNG
CTHUNGTRENNUNG
Materials and methods: All operations with air- and moisture-sensitive
compounds were performed in a glove box under a dry argon atmosphere
(MBraun 200B) or on a high-vacuum line by using Schlenk techniques.
The 1H, 13C, 11B, and 19F NMR spectra were recorded on Bruker DPX
200 (200 MHz), Bruker AVII 300 (300 MHz), and Bruker DRX 400
(400 MHz) devices. The chemical shifts are expressed in parts per million
(ppm) with tetramethylsilane (TMS) or the residual solvent signal as an
internal standard. Coupling constants (J) are reported in Hertz (Hz) and
splitting patterns are indicated as s (singlet), d (doublet), t (triplet), q
(quartet), m (multiplet), sept (septet), and br (broad). Elemental analysis
was carried out on a Vario Micro Cube System. A Bruker Vertex 70 spec-
trometer was used to record the IR spectra. Mass spectra were recorded
2
3
o-C6H
N
ACHTUNGTRNENUG(C,F)=33.7 Hz; m-C6H3ACHTUGNTRENNNUG
H
(CF3)2), 124.0 ppm (q, 1J
19F NMR (376 MHz, CD2Cl2, 258C): d=À63.3 ppm; MS (EI): m/z (%):
650.05 (100); IR (Neat): n˜ =1606, 1382, 1275, 1224, 1166, 1119, 984, 943,
907, 844, 731, 718, 707, 694, 681, 656 cmÀ1; elemental analysis calcd (%)
for C24H9BF18: C 44.34, H 1.40, N 0.00; found: C 44.55, H 1.52, N 0.00.
BACHTUNGTRENNUNG
A
R
CHTUNGTRENNUNG
4
(sept, J
N
CHTUNGTRENNUNG
3
C
ACHTUNGTRENNUNG
A
E
N
ACHTUNGTRENNUNG
Chem. Eur. J. 2012, 18, 16938 – 16946
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16943

M. Tamm et al.
Table 2. Crystallographic data.
4a
4b·0.5C7H8
5
7
8·C6D6
empirical formula
Mr
T [K]
C35H29BF18N2
830.41
100
C54.5H49BF18N2
1084.77
100
C35H29BF18N2
830.41
100
C36H29BF18N2O2
874.42
100
C49H35D6BF18N2
1016.69
100
l [ꢃ]
0.71073
triclinic
P1
10.811(2)
10.936(2)
17.160(3)
92.861(16)
107.395(18)
110.29(2)
1788.9
0.71073
monoclinic
P21/n
13.9953(2)
25.6426(4)
15.4957(3)
90
103.218(2)
90
5413.7
0.71073
triclinic
P1
0.71073
monoclinic
P21/c
14.0690(3)
21.2587(5)
12.6289(3)
90
100.913(2)
90
3708.8
0.71073
monoclinic
P21/c
18.4322(5)
18.4691(5)
28.2520(7)
90
95.376(3)
90
9575.4
crystal system
space group
a [ꢃ]
b [ꢃ]
c [ꢃ]
¯
¯
9.7033(3)
11.0871(3)
18.6065(6)
77.753(3)
84.421(3)
70.698(3)
1845.40
2
a [8]
b [8]
g [8]
V [ꢃ3]
Z
2
4
4
8
total reflns/2qmax
unique reflns
57140/50
6308
186263/52.6
11062
60680/56.4
8375
149444/52.6
7577
240670/50
16892
(Rint =0.155)
1.542
0.16
(Rint =0.072)
1.331
0.12
(Rint =0.034)
1.494
0.15
(Rint =0.046)
1.566
0.16
(Rint =0.077)
1.410
0.13
1calcd [gcmÀ3
m [mmÀ1
]
]
parameters/restraints
R(Fo), [I>2s(I)]
511/0
0.062
0.173
1.08
715/134
0.068
0.167
1.11
0.49/À0.34
511/0
0.043
0.099
1.03
538/0
0.037
0.087
1.02
1355/96
0.053
0.140
1.03
0.48/À0.47
R
G
]
0.67/À0.50
0.44/À0.36
0.51/À0.43
272.5 Hz; CF3), 121.6 ppm (sept, 3J
11B NMR (96 MHz, [D8]THF, 258C): d=11.7 ppm.
(m-XyF6)3·acetone adduct: 11B NMR (96 MHz, [D6]acetone, 258C): d=
10.9 ppm (decomposition of the solution was observed within a few
days).
(C,F)=3.8 Hz; p-C6H
E
Dipp), 136.3 (br; o-C6H CHTUNGTRENNUNG
(CF3)2), 135.2 (Dipp), 132.1 (Dipp), 129.7 (q, 2J-
3A
3A(CF3)2), 126.6 (Dipp), 125.0 (CH=CH), 124.9 (q,
(C,F)=272.6 Hz; CF3), 120.0 (sept, 3J
(C,F)=4.2 Hz; p-C6H3A(CF3)2),
29.4 (iPr), 26.8(iPr), 21.4 ppm (iPr), the resonances for the ipso-C6H3-
A
CHTUNGTRENNUNG
A
R
CHTUNGTRENNUNG
BACHTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
(CF3)2 and NCN moieties were not detected; 11B NMR (96 MHz, C6D6,
258C): d=À8.0 ppm; 19F (188 MHz, C6D6, 258C): d=À62.2 ppm; ele-
mental analysis calcd (%) for C51H45BF18N2: C 58.97, H 4.37, N 2.70;
found: C 56.96, H 4.12, N 2.57.
Normal adduct 4a: To a mixture of BACHTUNTRGNEUNG(m-XyF6)3 (2.350 g, 3.61 mmol,
1 equiv) and compound 1a (0.651 g, 3.61 mmol, 1 equiv) was added tol-
uene (30 mL) and the reaction was stirred for 24 h. n-Hexane (30 mL)
was added to the reaction mixture and the precipitate was filtered off,
washed with toluene/n-hexane (1:1), and dried in vacuo (2.410 g). A
second crop of the product was obtained by keeping the combined fil-
trate in a freezer at À308C for a few days to afford complete precipita-
tion of the product, which was filtered off, washed with toluene/n-hexane
(1:1), and dried in vacuo (0.433 g). Yield: 2.841 g (95%); white crystalline
powder; 1H NMR (300 MHz, CD3CN, 258C): d=7.77 (br s, 6H; o-C6H3-
Abnormal adduct 5: Compound 4a (0.150 g, 0.18 mmol) was heated at
reflux in toluene (5 mL) for 15 min and then slowly cooled to RT. The re-
sulting large crystals of compound 5 were filtered off, washed with cold
toluene, and dried in vacuo. Yield: 0.138 g (92%); colourless crystals;
1H NMR (300 MHz, [D6]acetone, 258C): d=9.10 (d, 4J
1H; NCHN), 7.82 (br s, 6H; o-C6H3A
(CF3)2), 7.71 (sept, 4J
(H,H)=1.9 Hz, 1H; CH=CH), 1.62 (s, 9H;
tBu), 1.28 ppm (s, 9H; tBu); 13C NMR (75 MHz, [D6]acetone, 258C): d=
160.5 (br; ipso-C6H3A(CF3)2), 136.3 (br; o-C6H3A (C,F)=
(CF3)2), 130.2 (q, 2J
31.6 Hz; m-C6H3A (C,F)=272.0 Hz;
(CF3)2), 126.9 (s; BC=CH), 125.4 (q, 1J
CF3), 119.0 (quin, 3J
(C,F)=4.2 Hz; p-C6H3A(CF3)2), 62.9 (C(CH3)3), 59.6
(C(CH3)3), 31.7 (CH3), 29.6 ppm (CH3), the resonance for the NCN
ACHTUNGTRENNUNG
C
ACHTUNGTRENNUNG
4
3H; p-C6H3ACTHNUGTRENN(GNU CF3)2), 6.84 (d, JCAHTUNGTRENNUNG
ACHTUNGTRENNUNG ACHTUNGTRENNUNG(H,F)=0.8 Hz, 3H; p-C6H3ACHTNGUTRENN(GUN CF3)2), 1.60 ppm (s,
(CF3)2), 7.67 (sept, 4J
C
E
ACHTUNGTRENNUNG
18H; tBu; CH=CH protons were not observed because of fast deutera-
1
C
ACHTUNGTRENNUNG
tion by CD3CN); H NMR (300 MHz, C6D6, 258C): d=8.04 (br s, 6H; o-
A
N
ACHTUNGTRENNUNG
C6H
0.62 ppm (s, 18H; tBu); 13C NMR (75 MHz, CD3CN, 258C): d=165.1
(NCN), 160.0 (br; ipso-C6H3A(CF3)2), 134.6 (br s; o-C6H3A(CF3)2), 130.0 (q,
2J (CF3)2), 125.7 (q, 1J
(C,F)=31.6 Hz; m-C6H3A (C,F)=271.7 Hz; CF3),
120.9 (t, 1J(D,C)=30.0 Hz; CD=CD), 119.3 (sept, 3J
(C,F)=3.9 Hz; p-
C6H3A(CF3)2), 61.2 (C
(CH3)3), 29.8 ppm (CH3); 11B NMR (96 MHz,
3ACHTUNGTRENNUNG(CF3)2), 7.63 (br s, 3H; p-C6H3AHCTUNGTRENN(UNG CF3)2), 6.42 (s, 1H; CH=CH),
AHCTUNGTRENNUNG
moiety was not detected; 11B NMR (96 MHz, [D6]acetone, 258C): d=
À7.4 ppm; 19F (188 MHz, [D6]acetone, 258C): d=À62.3 ppm; elemental
analysis calcd (%) for C35H29BF18N2: C 50.62, H 3.52, N 3.37; found:
C 50.37, H 3.41, N 3.43.
C
CHTUNGTRENNUNG
A
N
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
C
ACHTUNGTRENNUNG
CD3CN, 258C): d=À4.8 ppm; 11B NMR (96 MHz, C6D6, 258C): d=
À5.9 ppm; 19F (188 MHz, C6D6, 258C): d=À62.8 ppm; elemental analysis
calcd (%) for C35H29BF18N2: C 50.62, H 3.52, N 3.37; found: C 50.29,
H 3.68, N 3.36.
CO2-activation product (7): A suspension of compound 4a (0.100 g,
0.120 mmol) in benzene (10 mL) was stirred under CO2 (1 atm) over-
night. The resulting precipitate was filtered off and dried in vacuo. Yield
0.081 g (86%); white powder; 1H NMR (200 MHz, CD2Cl2, 258C): d=
7.90 (br s, 6H; o-C6H
CH=CH), 1.49 ppm (s, 18H; tBu); 13C NMR (75 MHz, CD2Cl2, 258C):
d=157.7 (CO2), 154.3 (br; ipso-C6H3A(CF3)2), 140.8 (NCN), 134.6 (br; o-
C6H3A (C,F)=31.9 Hz; m-C6H3A
(CF3)2), 130.0 (q, 2J (CF3)2), 125.1 (q, 1J-
(C,F)=272.6 Hz; CF3), 120.0 (sept, J(C,F)=3.9 Hz; p-C6H3A(CF3)2), 118.3
(CH=CH), 64.3 (C
(CH3)3), 30.1 ppm (CH3); 11B NMR (96 MHz, CD2Cl2,
3ACHTUNTGERN(NUG CF3)2), 7.70 (br s, 3H; p-C6H3CAHTUNGTERN(NUGN CF3)2), 7.23 (s, 1H;
Normal adduct 4b: To a mixture of BACHTUNTRGNEUNG(m-XyF6)3 (0.325 g, 0.5 mmol,
1 equiv) and compound 1b (0.194 g, 0.5 mmol, 1 equiv) was added tol-
uene (20 mL) and the reaction was stirred for 2 h. The mixture was con-
centrated to approximately 5 mL in vacuo and n-hexane (30 mL) was
added. The resulting precipitate was filtered off, washed with toluene/n-
hexane (1:10), and dried in vacuo. Yield: 0.461 g (89%); white powder;
CTHUNGTRENNUNG
G
A
CHTUNGTRENNUNG
3
G
A
CHTUNGTRENNUNG
AHCTUNGTRENNUNG
258C): d=4.7 ppm (br s); 19F NMR (376 MHz, CD2Cl2, 258C): d=
À62.9 ppm; elemental analysis calcd (%) for C36H30BF18N2O2: C 49.39,
H 3.45, N 3.20; found: C 50.14, H 3.48, N 3.06.
1H NMR (300 MHz, C6D6, 258C): d=7.54 (br s, 3H; p-C6H
(br s, 6H; o-C6H3A (H,H)=7.8 Hz, 2H; pDipp), 6.60 (d,
(CF3)2), 6.89 (t, 3J
3J
(H,H)=8.0 Hz, 4H; mDipp), 6.38 (s, 2H; CH=CH), 2.55 (br s, 4H;
CH
6.6 Hz, 12H; CH3); 13C NMR (75 MHz, C6D6, 258C): d=145.0 (ipso-
3ACHTUNGERTN(NUNG CF3)2), 7.47
C
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
3
3
(CH3)2), 0.85 (d, J
N
A
Phenylacetylene-activation product (8): Compound 4a (0.150 g,
0.180 mmol, 1 equiv) was mixed with benzene (2 mL) and a solution of
16944
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16938 – 16946

À
Isolable Arduengo Carbene BACHTUNGTRENN{GU 3,5-CAHTUTGNNER(NNGU CF3)2C6H3}3 Complex
FULL PAPER
phenylacetylene (0.019 g, 0.180 mmol, 1 equiv) in benzene (0.5 mL) was
added. After stirring for 20 min, a yellow oil separated out and was re-
crystallized from hot benzene/CH2Cl2. The resulting large crystals were
ground into a fine powder and dried in vacuo. Yield: 0.154 g (91%);
white powder; 1H NMR (300 MHz, C6D6+5% CD2Cl2, 258C): d=8.29
Chem. Rev. 2000, 100, 1391; c) M. Bochmann, J. Chem. Soc. Dalton
Trans. 1996, 255.
[4] G. Erker, Dalton Trans. 2005, 1883.
[5] a) K. Ishihara, H. Yamamoto, Eur. J. Org. Chem. 1999, 5272; b) U.
Rosenthal, V. V. Burlakov, P. Arndt, W. Baumann, A. Spannenberg,
V. B. Shur, Eur. J. Inorg. Chem. 2004, 4739; c) W. E. Piers, Adv. Or-
ganomet. Chem. 2004, 52, 1–76; d) T. Beringhelli, D. Donghi, D.
Maggioni, G. D’Alfonso, Coord. Chem. Rev. 2008, 252, 2292;
e) M. A. Brook, J. B. Grande, F. Ganachaud, Adv. Polym. Sci. 2011,
235, 161.
[6] a) G. C. Welch, R. R. San Juan, J. D. Masuda, D. W. Stephan, Science
2006, 314, 1124; b) V. Sumerin, F. Schulz, M. Nieger, M. Leskelꢁ, T.
Repo, B. Rieger, Angew. Chem. 2008, 120, 6090; Angew. Chem. Int.
Ed. 2008, 47, 5861.
[7] For reviews, see: a) D. W. Stephan, Org. Biomol. Chem. 2008, 6,
1535; b) D. W. Stephan, Dalton Trans. 2009, 3129; c) D. W. Stephan,
G. Erker, Angew. Chem. 2010, 122, 50; Angew. Chem. Int. Ed. 2010,
49, 46; d) G. Erker, C. R. Chim. 2011, 14, 831; e) D. W. Stephan,
Org. Biomol. Chem. 2012, 10, 5740–5746.
[8] D. Holschumacher, T. Bannenberg, C. G. Hrib, P. G. Jones, M.
Tamm, Angew. Chem. 2008, 120, 7538; Angew. Chem. Int. Ed. 2008,
47, 7428.
[9] a) P. A. Chase, D. W. Stephan, Angew. Chem. 2008, 120, 7543;
Angew. Chem. Int. Ed. 2008, 47, 7433; b) P. A. Chase, A. L. Gille,
T. M. Gilbert, D. W. Stephan, Dalton Trans. 2009, 7179.
[10] a) D. Holschumacher, T. Bannenberg, K. Ibrom, C. G. Daniliuc,
P. G. Jones, M. Tamm, Dalton Trans. 2010, 39, 10590; b) D. Holschu-
macher, C. G. Daniliuc, P. G. Jones, M. Tamm, Z. Naturforsch. 2011,
66b, 0371.
[11] M. A. Dureen, C. C. Brown, D. W. Stephan, Organometallics 2010,
29, 6594.
[12] A. J. Arduengo III, H. Bock, H. Chen, M. Denk, D. A. Dixon, J. C.
Green, W. A. Herrmann, N. L. Jones, M. Wagner, R. West, J. Am.
Chem. Soc. 1994, 116, 6641.
(br s, 6H; o-C6H
2H; Ph), 7.35 (t, 4J
6.37 (d, 4J
(H,H)=1.8 Hz, 2H; CH=CH), 0.82 ppm (s, 18H; tBu);
1H NMR (300 MHz, CDCl3, 258C): d=7.95 (br s, 6H; o-C6H
3A(CF3)2),
7.87 (t, 4J
(H,H)=1.7 Hz, 1H; NCHN), 7.54 (br s, 3H; p-C6H3A(CF3)2),
7.47–7.43 (m, 2H; Ph), 7.28–7.18 (m, 3H; Ph), 7.00 (d, 4J
(H,H)=1.7 Hz;
CH=CH), 1.35 ppm (s, 18H; tBu); 13C NMR (75 MHz, C6D6+5%
CD2Cl2, 258C): d=135.0 (br; o-C6H3A
(CF3)2), 132.0 (NCN), 129.9 (q, 2J-
(C,F)=30.5 Hz; m-C6H3A(CF3)2), 128.7(Ph), 127.0 (Ph), 124.9 (q, J=
271.9 Hz; CF3), 120.5 (CH=CH), 118.5 (sept, 3J
(C,F)=4.2 Hz; p-C6H3-
(CF3)2), 60.8 (C(CH3)3), 29.1 ppm (CH3), the resonances for the ipso-
3A
3ACHTUNGTREN(UNGN CF3)2), 7.46–7.39 (m,
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
CTHUNGTRENNUNG
A
CHTUNGTRENNUNG
AHCTUNGTRENNUNG
CTHUNGTRENNUNG
A
CHTUNGTRENNUNG
AHCTUNGTRENNUNG
A
ACHTUNGTRENNUNG
ꢁ
C6H
CDCl3, 258C): d=161.0 (br; ipso-C6H
(CF3)2 and C C moieties were not detected; 13C NMR (75 MHz,
3A(CF3)2), 134.0 (br; o-C6H3A(CF3)2),
(C,F)=31.6 Hz; m-C6H3A(CF3)2), 128.3(Ph), 128.2
(Ph), 127.8 (Ph), 127.6 (Ph), 124.7 (q, 1J
(C,F)=272.6 Hz; CF3), 120.4
(CH=CH), 117.8 (sept, 3J
(C,F)=3.9 Hz; p-C6H3A(CF3)2), 60.7 (C(CH3)3),
G
CHTUNGTRENNUNG
2
131.3 (NCN), 128.9 (q, J
G
CHTUNGTRENNUNG
AHCTUNGTRENNUNG
A
N
ACHTUNGTRENNUNG
ꢁ
29.2 ppm (CH3), the resonances for the C C moiety were not detected;
11B NMR (96 MHz, C6D6+5% CD2Cl2, 258C): d=À8.9 ppm; 11B NMR
(96 MHz, CDCl3, 258C): d=À11.9 ppm; 19F (188 MHz, C6D6+5%
CD2Cl2, 258C): d=À62.3 ppm; elemental analysis calcd (%) for
C43H35BF18N2: C 55.38, H 3.78, N 3.00; found: C 56.24, H 3.98, N 2.69.
H2-activation product 9·B
N
A suspension of compound 4a
(0.100 g, 0.120 mmol) and BAHCTNUGTRENNNUG
benzene (10 mL) was stirred under H2 (1 atm) overnight. The resulting
precipitate was filtered off, washed with CH2Cl2, and dried in vacuo.
Yield 0.150 g (84%); white powder; 1H NMR (300 MHz, [D8]THF,
258C): d=8.78 (t, 4J
1.7 Hz, 2H; CH=CH), 7.87 (br s, 12H; o-C6H CTHUNGTRENNUNG
N
ACHTUNGTRENNUNG
À
À
C6H3ACHTUNGTRENNUNG
CTHUNGTRENNUNG
[13] T. A. Rokob, A. Hamza, I. Pꢅpai, J. Am. Chem. Soc. 2009, 131,
10701.
[14] A. J. Arduengo III, R. Krafczyk, R. Schmutzler, Tetrahedron 1999,
55, 14523.
A
E
ACHTUNGTRENNUNG
1
AHCTUNGTRENNUNG
C
ACHTUNGTRENNUNG
CHTUNGTRENNUNG
(CF3)2 moiety was not detected; 11B NMR (96 MHz,
[15] S. Kronig, E. Theuergarten, D. Holschumacher, T. Bannenberg,
C. G. Daniliuc, P. G. Jones, M. Tamm, Inorg. Chem. 2011, 50, 7344.
[16] D. Holschumacher, C. Taouss, T. Bannenberg, C. G. Hrib, C. G.
Daniliuc, P. G. Jones, M. Tamm, Dalton Trans. 2009, 6927.
[17] J. W. Runyon, O. Steinhof, H. V. Rasika Dias, J. C. Calabrese, W. J.
Marshall, A. J. Arduengo III, Aust. J. Chem. 2011, 64, 1165.
[18] For selected reviews, see: a) I. Krossing, I. Raabe, Angew. Chem.
2004, 116, 2116; Angew. Chem. Int. Ed. 2004, 43, 2066; b) A. Mac-
chioni, Chem. Rev. 2005, 105, 2039–2073. See also representative
original papers: c) H. Nishida, N. Takada, M. Yoshimura, T. Sonoda,
H. Kobayashi, Bull. Chem. Soc. Jpn. 1984, 57, 2600; d) M. Brook-
[D8]THF, 258C): d=À10.0 ppm (br; B H B); 19F (188 MHz, [D8]THF,
258C): d=À64.3 (br s), À64.6 ppm (s); MS (ESI(+), MeOH): m/z (%):
À
À
À
181.17 (100) ItBuH; MS (ESI(À), MeOH): m/z (%): 651.17 (100) H B-
À
(m-XyF6)3, 681.17 (30) MeO B
N
for C59H40B2F36N2: C 47.80, H 2.72, N 1.89; found: C 48.32, H 2.39,
N 1.65.
9+
258C): d=8.28 (t, 4J
7.89, 7.63, 7.57, 7.50 (8ꢄbr s, 18H; o-C6H
4J(H,H)=1.7 Hz, 2H; CH=CH), 3.63 (q, 1J
1.97 (s, CH3CN), 1.65 ppm (s, 18H; tBu); 11B NMR (96 MHz,
ACHTUNGTRENNUNG[CH3CN·BACHTUNGTRENNUNG
ACHTUNGTRENNUNG
C
CHTUNGTRENNUNG
À
N
ACHTUNGTRENNUNG
AHCTUNGTREGhNNUN art, B. Grant, A. F. Volpe, Jr., Organometallics 1992, 11, 3920;
e) N. A. Yakelis, R. G. Bergman, Organometallics 2005, 24, 3579.
[19] R. Santi, A. M. Romano, A. Sommazzi, M. Grande, C. Bianchini, G.
Mantovani, J. Mol. Catal. A 2005, 229, 191; in this publication, the
CD2Cl2+1% MeCN, 258C): d=À4.8 (br; CH3CN·B
ACHTUNGRTEN(NUNG m-XyF6)3),
1
À
À8.7 ppm (d, J
N
commercial availability of B
not been able to confirm.
ACHTUGNTERN(NUNG m-XyF6)3 is claimed, which we have
Acknowledgements
[20] W. V. Konze, B. L. Scott, G. J. Kubas, Chem. Commun. 1999, 1807.
[21] T. J. Herrington, A. J. W. Thom, A. J. P. White, A. E. Ashley, Dalton
Trans. 2012, 41, 9019.
This work was supported by the Deutsche Forschungsgemeinschaft
through grant TA 189/9-1. We thank Dr. Kerstin Ibrom (TU Braunsch-
weig) for her substantial support with the NMR studies.
[22] C. Wang, G. Erker, G. Kehr, K. Wedeking, R. Frçhlich, Organo-
AHCTUNGERTGmNNUN etallics 2005, 24, 4760.
[23] J. L. Leazer, Jr., R. Cvetovich, F.-R. Tsay, U. Dolling, T. Vickery, D.
Bachert, J. Org. Chem. 2003, 68, 3695.
[24] M. Abarbri, F. Dehmel, P. Knochel, Tetrahedron Lett. 1999, 40, 7449.
[25] Additional experimental details, analytical data, and details of the
electronic-structure calculations are provided in the Supporting In-
formation.
[1] a) A. G. Massey, A. J. Park, F. G. A. Stone, Proc. Chem. Soc. 1963,
212; b) A. G. Massey, A. J. Park, J. Organomet. Chem. 1964, 2, 245;
c) J. L. W. Pohlmann, F. E. Brinckmann, Z. Naturforsch. B 1965, 20b,
5.
[26] A variable-temperature NMR study of compound 4a in [D8]toluene
[2] W. E. Piers, T. Chivers, Chem. Soc. Rev. 1997, 26, 345.
[3] For selected reviews, see: a) F. Focante, P. Mercandelli, A. Sironi, L.
Resconi, Coord. Chem. Rev. 2006, 250, 170; b) E. Chen, T. J. Marks,
1
revealed that the H and 19F NMR resonances remained sharp down
À
to À708C, thus indicating free rotation around the B C bonds, even
Chem. Eur. J. 2012, 18, 16938 – 16946
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16945

M. Tamm et al.
at low temperature. Below this temperature, broadening of the sig-
nals was observed because of precipitation of the compound, even
at low concentrations (1 mg in 0.65 mL); the sample froze at
À1048C.
1993, 26, 593. For examples of such interactions involving B
see: c) D. J. Parks, W. E. Piers, M. Parvez, R. Atencio, M. J. Zawo-
rotko, Organometallics 1998, 17, 1369; d) J. M. Blackwell, W. E.
ACHTUNGTRENNUNG(C6F5)3,
AHCTUNGTRENNUNG
Piers, M. Parvez, R. McDonald, Organometallics 2002, 21, 1400.
[38] a) A. E. Ashley, A. L. Thompson, D. OꢇHare, Angew. Chem. 2009,
121, 10023; Angew. Chem. Int. Ed. 2009, 48, 9839; b) C. M. Mçm-
ming, E. Otten, G. Kehr, R. Frçhlich, S. Grimme, D. W. Stephan, G.
Erker, Angew. Chem. 2009, 121, 6770; Angew. Chem. Int. Ed. 2009,
48, 6643; c) A. Berkefeld, W. E. Piers, M. Parvez, J. Am. Chem. Soc.
2010, 132, 10660; d) C. Appelt, H. Westenberg, F. Bertini, A. W.
Ehlers, J. C. Slootweg, K. Lammertsma, W. Uhl, Angew. Chem.
2011, 123, 4011; Angew. Chem. Int. Ed. 2011, 50, 3925; e) I. Peuser,
R. C. Neu, X. Zhao, M. Ulrich, B. Schirmer, J. A. Tannert, G. Kehr,
R. Frçhlich, S. Grimme, G. Erker, D. W. Stephan, Chem. Eur. J.
2011, 17, 9640; f) M. Harhausen, R. Frçhlich, G. Kehr, G. Erker, Or-
ganometallics 2012, 31, 2801.
[39] For another example of the formation of hydrodiborate under simi-
lar conditions, see: C. Jiang, O. Blacque, H. Berke, Chem. Commun.
2009, 5518.
[40] S. J. Geier, D. W. Stephan, J. Am. Chem. Soc. 2009, 131, 3476.
[41] T. A. Rokob, A. Hamza, A. Stirling, I. Pꢅpai, J. Am. Chem. Soc.
2009, 131, 2029.
[27] A. D. Phillips, P. P. Power, Acta Crystallogr. Sect. C 2005, 61, o291.
[28] a) Y. Zhao, D. G. Truhlar, Theor. Chem. Acc. 2008, 120, 215; b) Y.
Zhao, D. G. Truhlar, Acc. Chem. Res. 2008, 41, 157.
[29] a) S. Grimme, J. Comput. Chem. 2006, 27, 1787; b) S. Grimme,
WIREs Comput. Mol. Sci. 2011, 1, 211.
[30] a) S. Grimme, R. Huenerbein, S. Ehrlich, ChemPhysChem 2011, 12,
1258; b) S. Grimme, P. R. Schreiner, Angew. Chem. 2011, 123,
12849; Angew. Chem. Int. Ed. 2011, 50, 12639; c) C. F. R. A. C.
Lima, M. A. A. Rocha, L. R. Gomes, J. N. Low, A. M. S. Silva,
L. M. N. B. F. Santos, Chem. Eur. J. 2012, 18, 8934–8943.
[31] For instance, see: a) H. Clavier, S. P. Nolan, Chem. Commun. 2010,
46, 841; b) A. Poater, B. Cosenza, A. Correa, S. Giudice, F. Ragone,
V. Scarano, L. Cavallo, Eur. J. Inorg. Chem. 2009, 1759; c) S. Dꢆez-
Gonzꢅlez, S. P. Nolan, Coord. Chem. Rev. 2007, 251, 874; d) L. Cav-
allo, A. Correa, C. Costabile, H. Jacobsen, J. Organomet. Chem.
2005, 690, 5407.
[32] H. Jacobsen, A. Correa, A. Poater, C. Costabile, L. Cavallo, Coord.
Chem. Rev. 2009, 253, 687.
[33] a) M. Sꢀßner, H. Plenio, Chem. Commun. 2005, 5417; b) S.
Leuthꢁußer, D. Schwarz, H. Plenio, Chem. Eur. J. 2007, 13, 7195.
[34] R. Credendino, L. Falivene, L. Cavallo, J. Am. Chem. Soc. 2012,
134, 8127–8135.
[35] A. Fꢀrstner, L. Ackermann, B. Gabor, R. Goddard, C. W. Lehmann,
R. Mynott, F. Stelzer, O. R. Thiel, Chem. Eur. J. 2001, 7, 3236.
[36] A. Bondi, J. Phys. Chem. 1964, 68, 441.
[37] Molecular complexes between hydrocarbons and fluorocarbons
have long been known; for instance, see: a) C. R. Patrick, G. S.
Prosser, Nature 1960, 187, 1021; b) J. H. Williams, Acc. Chem. Res.
[42] S. Kronig, E. Theuergarten, C. G. Daniliuc, P. G. Jones, M. Tamm,
Angew. Chem. 2012, 124, 3294; Angew. Chem. Int. Ed. 2012, 51,
3240.
[43] G. M. Sheldrick, SHELXL-97, Program for the Refinement of Crys-
tal Structures, University of Gçttingen, Gçttingen; G. M. Sheldrick,
Acta Crystallogr. Sect. A 2008, 64, 112.
[44] A. L. Spek, “SQUEEZE”, part of the PLATON program package;
University of Utrecht, Netherlands, 2008.
Received: May 24, 2012
Published online: November 13, 2012
16946
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16938 – 16946