Remote substitution on N-heterocyclic carbenes heightens the catalytic reactivity of their palladium complexes

Article abstract of DOI:10.1021/om200349y

A new series of exceptionally bulky N-heterocyclic carbene (NHC) ligands is based on 1,3-bis-2′,6′-diisopropylphenylimidazol-2-ylidene (IPr), through substitution of large groups at the para (4′) position of the aryl rings. π-Cinnamylpalladium chloride complexes with these ligands are precatalysts for Suzuki-Miyaura coupling. Large triarylmethyl substituents at the 4′ position of the NHC give significantly greater catalyst activity. Although the %V_bur parameter has been proposed as a general measure of the size of bulky ligands, the sterically induced enhancement of the catalytic rate does not correlate with the %V_bur parameter; instead, solid angles are more useful in quantifying the differences between ligands because they account for bulk more distant from the metal. Overall, these results show that the use of distant steric hindrance can have a distinct positive influence on catalytic ability.

Full text of DOI:10.1021/om200349y

ARTICLE
pubs.acs.org/Organometallics
Remote Substitution on N-Heterocyclic Carbenes Heightens the
Catalytic Reactivity of Their Palladium Complexes
Benjamin R. Dible, Ryan E. Cowley, and Patrick L. Holland*
Department of Chemistry, University of Rochester, Rochester, New York 14627, United States
S Supporting Information
b
ABSTRACT: A new series of exceptionally bulky N-heterocyclic carbene
(NHC) ligands is based on 1,3-bis-20,60-diisopropylphenylimidazol-2-ylidene
(IPr), through substitution of large groups at the para (40) position of the aryl
rings. π-Cinnamylpalladium chloride complexes with these ligands are pre-
catalysts for SuzukiꢀMiyaura coupling. Large triarylmethyl substituents at the
40 position of the NHC give significantly greater catalyst activity. Although the
%Vbur parameter has been proposed as a general measure of the size of bulky
ligands, the sterically induced enhancement of the catalytic rate does not
correlate with the %Vbur parameter; instead, solid angles are more useful in
quantifying the differences between ligands because they account for bulk
more distant from the metal. Overall, these results show that the use of distant
steric hindrance can have a distinct positive influence on catalytic ability.
’ INTRODUCTION
discussion of NHC steric effects, the most commonly discussed
steric parameter is the percent buried volume (%Vbur), which
defines a sphere around the inner part of a complex, and measures
the fraction of the volume of this sphere occupied by the ligand.10
The use of %Vbur in quantifying the size of NHC ligands has been
established by a body of work by Cavallo and Nolan.11
The N-heterocyclic carbene (NHC) ligands have been used
widely in catalysis and in coordination chemistry, where they
serve as extremely strong σ-donors that modify the steric and
electronic environment of the metal center. The study of NHC
complexes has become a rich area that has attracted considerable
interest, as evidenced by the large number of publications and
reviews relating to NHC complexes.1ꢀ5
To better describe and predict the properties of NHC com-
plexes, chemists have developed several methods to quantify
characteristics in terms of both sterics and electronics.6 Work by
Nolan7 and by Crabtree8 used the Tolman electronic parameter
(TEP) to describe the σ-donating abilities of NHC ligands. The
TEP method compares the CO stretching frequencies of tetra-
hedral (NHC)Ni(CO)3 complexes to those for (R3P)Ni(CO)3
complexes.9 However, this method of determining TEP values
was limited, because large NHC ligands, such as ItBu and IAd,
give planar (NHC)Ni(CO)2 complexes that are not directly
comparable to the reference tetrahedral tricarbonylnickel com-
plexes. To avoid this problem, Crabtree and co-workers devel-
oped a method for calculating TEP values based on the carbonyl
stretching frequencies for iridium complexes.8 This development
is advantageous because it facilitates TEP determination in a
wider range of NHCs and because it circumvents the use of
hazardous Ni(CO)4.
We were interested in developing NHC ligands with the intent
of not only shielding the close environment of a metal center but
also projecting far from the metal center to prevent dimerization
and/or aggregation of complexes. To accomplish this goal in the
case of the useful NHC ligands, we designed a series of new
ligands with large remote substituents, which are described in this
paper. We show below that the size of these ligands is not
predicted by the %Vbur parameter, and use another steric measure
that better quantifies the steric variation. To demonstrate the
impact of the remote substituents on catalyst activity with a well-
known catalyst system, we also describe the influence of the
remote substitution on the rate and conversion of catalytic
SuzukiꢀMiyaura coupling by (NHC)(π-cinnamyl)PdCl com-
plexes incorporating the new ligands.12
’ RESULTS AND DISCUSSION
Ligand Synthesis and Palladium Complex Formation. To
evaluate the effects of remote, sterically demanding groups on
ligand properties, we chose IPr as the parent ligand upon which
to build. The synthesis begins with triarylmethanols 1bꢀ1d as
protoelectrophiles to derivatize 2,6-diisopropylaniline 2a at the
4-position via electrophilic aromatic substitution. This approach
Although TEP values are generally accepted as the best method
for quantifying electronic effects of ligands, the description of
ligandsteric effects hasbeen morecontroversial. The Tolmancone
angle9 is a starting point for describing ligand sterics. The Tolman
cone angle is obtained by treating the ligand as a cone with the
metal center acting as the vertex and measuring the resulting angle,
or weighted average of angles for nonsymmetrical ligands. In the
Received: April 24, 2011
Published: September 08, 2011
r
2011 American Chemical Society
5123
dx.doi.org/10.1021/om200349y Organometallics 2011, 30, 5123–5132
|

Organometallics
ARTICLE
is advantageous because the triarylmethanols are simple to
prepare. Furthermore, the use of triarylmethanols as electrophile
precursors for electrophilic aromatic substitution reactions with
anilines is well-precedented.13,14 Thus, treatment of known
triarylmethanols 1bꢀ1d with an excess of aniline 2a in the
presence of HCl produces trityl anilines 2bꢀ2d cleanly and in
high yield (Scheme 1), as long as the reaction is performed under
an inert atmosphere. A basic workup is required to avoid product
contamination with anilinium salts.
given here for convenience, but it must be noted that these
compounds are typically obtained in only 90ꢀ95% purity.
Condensation of α-diimines 3bꢀ3e with paraformaldehyde
and HBF4 Et2O provides imidazolium tetrafluoroborates
3
4bꢀ4e, in direct analogy to the literature preparation of IPr
3
HBF4 salt 4a.16 However, the workup is modified significantly
for the para-substituted anilines. IPr HBF4 salt 4a is typically
3
purified via filtration, followed by washing the crude solid with
THF.16 Unlike IPr HBF4 salt 4a, imidazolium salts 4bꢀ4d are
3
With trityl anilines 2bꢀ2d and 4-(10-adamantyl)aniline 2e in
hand, the remainder of the ligand synthesis may be addressed.
The procedures developed by Nolan for the synthesis of
soluble in THF. Furthermore, the major contaminant in crude
salts 4bꢀ4d appears to be the HBF4 salts of the parent diimines
3bꢀ3d. Diimines 3aꢀ3e undergo a color change from yellow to
IPr HCl15 and by Herrmann for the synthesis of IPr HBF4
red upon the addition of HBF4 Et2O in toluene. In the reaction
16
3
3
3
required several modifications to produce the new ligands, due to
solubility and stability differences. The first step in the synthesis
of IPr is the condensation of aniline 2a with glyoxal in ethanol at
room temperature, which produces α-diimine 3a as a yellow
precipitate. Diimine 3a is then isolated by filtration and washed
with methanol, presumably in order to remove unreacted 2a.
However, unlike 2a, anilines 2bꢀ2e are not very soluble in
ethanol at room temperature, so it is necessary to perform the
condensation reactions at lower concentrations and at an ele-
vated temperature of 80 °C. Under these modified conditions,
α-diimines 3bꢀ3e form readily. Purification of α-diimines
3bꢀ3e is problematic, so the crude products are carried on to
the next step as long as the ratio of product 3 to aniline 2 is g9:1.
In cases where the ratio is lower than 9:1, the mixture is subjected
to the condensation conditions a second time in order to enrich
the material. Partial characterization for α-diimines 3bꢀ3e is
of diimine 3a to produce IPr HBF4, the reaction mixture
3
eventually turns brown and attempts to recover unreacted 3a
were not successful. However, in the reactions of 3bꢀ3e under
analogous conditions, the red color persists until the reaction
mixture is treated with aqueous bicarbonate, at which point the
yellow color characteristic of α-diimines 3aꢀ3e returns. Further-
more, an attempt to prepare NHC salt 4d proceeded with less
than 15% yield, and 58% of α-diimine 3d is recovered. Though
attempts to fully characterize this red protonated intermediate
have not been successful, a broad peak at δ 8.2 ppm and a
significantly downfield shifted septet at δ 3.2 ppm are observed in
the crude spectrum of NHC salt 4d prior to basic workup,
consistent with the presence of a protonated intermediate.
These observations suggest that the decomposition of 3 under
acidic conditions might be the yield-limiting factor in the
synthesis of IPr salts. If so, then the yields of NHC salts
4bꢀ4e should be significantly higher than for 4a. However,
the yields are similar to those reported for IPr HCl and IPr
3
3
Scheme 1. Preparation of Trityl-Substituted Anilines
HBF4 (Scheme 2). Another potential explanation for the low
yields of NHC salts is that diimine protonation inhibits cycliza-
tion. In a test of this hypothesis, the formation of 4d using an
excess of paraformaldehyde and only one equivalent of acid,
results in an improved yield of 61%. However, applying these
conditions to the synthesis of IPr 4a gives an unimproved yield of
35%, suggesting that the yield-limiting factors for IPr 4a are not
the same as those for the other NHCs. Finally, adding 0.25 equiv
of iPr2NEt as a potential buffer has deleterious results on the
yield. Either iPr2NEt does not have the correct pKa to buffer this
Scheme 2. Preparation of NHC HBF4 Salts via Sequential Condensations
3
5124
dx.doi.org/10.1021/om200349y |Organometallics 2011, 30, 5123–5132

Organometallics
ARTICLE
Scheme 3. Preparation of NHC-π-CinnamylPdCl Complexes
reaction or the problem is more complicated than diimine
protonation-inhibited cyclization.
Imidazolium salts 4bꢀ4e may be deprotonated with potas-
sium tert-butoxide under N2 to generate NHCs 5bꢀ5e.
Although it is common to generate and use NHCs in situ when
preparing π-allylpalladium complexes, isolating the NHCs prior
to ligation allows for a final purity check prior to the use of
precious metals. Here, each free NHC has been characterized
using NMR spectroscopy and elemental analysis.
The final step in the synthesis is to break up the chloro-bridged
dimer [(π-cinnamyl)PdCl]2 by treatment with the NHC. In the
process of performing these reactions, we obtained an X-ray quality
crystal of [(π-cinnamyl)PdCl]2, which is presented in the Support-
ing Information. Treatmentof the palladium dimer with each NHC
5bꢀ5e readily produces the desired palladium complexes 6bꢀ6e
(Scheme 3). Complexes 6c and 6d can be purified in the air via flash
chromatography using ether/hexanes as the eluent.
Figure 1. ORTEP diagram of complex 6b, using 50% thermal ellipsoids.
Hydrogen atoms are omitted for clarity.
Analysis of complexes 6aꢀ6e by 1H NMR spectroscopy gives
information about the solution structure. On the NHC ligand,
the backbone protons on C4 and C5 of the imidazol-2-ylidene
are equivalent, indicating either the presence of a mirror plane
bisecting the NHC or a mechanism of exchange that is rapid on
the NMR time scale. There are two septets for the methine
protons of the isopropyl groups, indicating loss of the mirror
plane that contains the imidazolylidene ring. Consistent with this
loss of symmetry, there are three signals for the methyl protons
on the isopropyl groups, with an intensity ratio of 6:6:12. These
features are consistent with a chiral square-planar palladium
complex with rapid rotation about the palladiumꢀNHC bond.
They also indicate that there is no rotation about the nitrogen
ꢀaryl bonds, the Pdꢀcinnamyl bonds, or the arylꢀisopropyl
bonds on the NMR time scale. The NMR spectra of 6bꢀ6e are
very similar to those reported previously for 6a.12 The added
trityl substituents in 6bꢀ6d are spinning rapidly around the
C(2,6-diisopropylphenyl)ꢀC(trityl) bond on the NMR time
scale, as demonstrated by the equivalence of the protons on each
aryl ring of the trityl substituents.
Figure 2. ORTEP diagram of complex 6c, using 50% thermal ellipsoids.
Hydrogen atoms are omitted for clarity.
molecules in the asymmetric unit.12 In all three structures, the
π-allyl group lies in the plane of the other coordinating atoms,
taking up two coordination positions in the square-planar
palladium(II) complex. The NHC ring is perpendicular to the
Pd square plane, consistent with the proton NMR data presented
above. Table 1 shows a comparison of the distances from
palladium to the carbon atom of the NHC ligand, and also to
the terminal (C14) and internal (C34) atoms of the allyl group
within the cinnamyl ligand. None of these measures display any
substantial differences between the different crystal structures
6aꢀ6c. Marion et al. used the difference between the PdꢀC(allyl)
X-ray Crystal Structures. To accurately gauge the structural
effects of the substituents on the NHC aryl rings, we report X-ray
crystal structures of complexes 6b (Figure 1) and 6c (Figure 2).
We have been unable to generate satisfactory crystals of com-
plexes 6d or 6e. As substitutes, we describe the crystal structures
of NHC HBF4 salts 4d (Figure 3) and 4e (Figure 4).
3
The cores of the complexes 6b and 6c closely resemble that
of 6a, which contained two crystallographically inequivalent
5125
dx.doi.org/10.1021/om200349y |Organometallics 2011, 30, 5123–5132

Organometallics
ARTICLE
Table 2. Calculated %Vbur Values for IPr-Derived NHCs
compound
%Vbur values at distance
2.00 Å
2.10 Å
2.28 Å
6a#1
34.9
36.7
35.8
36.9
40.0
47.3
43.6
48.4
46.4
46.6
47.2
46.9
37.6
33.0
34.9
34.0
35.1
38.2
45.3
41.7
46.4
44.5
44.7
45.3
45.0
35.7
29.6
31.6
30.6
31.8
35.0
41.8
38.3
42.8
41.0
41.1
41.7
41.4
32.3
6a#2
6a (av)
6b
6c
IPrH+Clꢀ
IPrH+OTfꢀ
IPrH+AlCl4
IPrH+ (av)
4d#1
ꢀ
Figure 3. ORTEP diagram of imidazolium salt 4d, using 50% thermal
ellipsoids. Hydrogen atoms (except the imidazolium CꢀH) are omitted
for clarity.
4d#2
4d (av)
4e
the crystal structures, calculated using the online tool SambVca.18
This program normalizes the MꢀC distance to facilitate com-
parison of ligands, and there is no obvious choice of MꢀC
distance to use. Thus, our calculations use three different MꢀC
bond distances: 2.00, 2.10, and 2.28 Å. In cases where two
crystallographically nonequivalent complexes or salts are present
in the unit cell, results are reported for each independent
molecule, and also as an average. The results are shown in
Table 2. There are no consistent trends in the values according to
the size of the substituent. In general, one also needs to bear in
mind the low precision of calculated %Vbur values. For example,
the range of %Vbur values for different complexes of the same IPr
ligand ranges from 31.0 to 47.6 at 2.00 Å and from 26.0 to 42.1 at
2.28 Å,11 and all of our new ligands fell within that range.
To test the influence of the distant substituent upon the
calculated %Vbur values, we repeated the %Vbur calculations on
truncated versions of complexes 6b and 6c and salts 4d and 4e,
using identical crystallographic coordinates, but with the substit-
uents at the 40 positions of the 2,6-diisopropylphenyl groups
manually deleted. For each ligand, the %Vbur values for the
truncated ligand are identical to those for the nontruncated
ligand, irrespective of the MꢀC bond distance chosen. This is
not surprising, as the %Vbur measure considers only the buried
volume within 3.5 Å of the metal, and the para substituents on the
aryl groups are at least 6 Å from the metal. In sum, the differences
in %Vbur values between the IPr variants presented here do not
accurately reflect the sizes of the ligands (and, as we shall show
below, the difference in catalytic reactivity).
Figure 4. ORTEP diagram of imidazolium salt 4e, using 50% thermal
ellipsoids. Hydrogen atoms (except the imidazolium CꢀH) are omitted
for clarity.
Table 1. Bond Distances (Å) in Literature Complex 6a and
New Complexes
compound
PdꢀNHC
PdꢀC14
PdꢀC34
6a
2.025(7)
2.064(8)
2.029(1)
2.024(1)
2.125(7)
2.136(10)
2.08(1)
2.250(9)
2.279(10)
2.28(1)
6b
6c
2.11(2)
2.26(2)
distances as a measure of the trans-influence of the NHC, from
which one could infer the relative ability to extrude the allyl group
during activation of the catalyst.12 Both 6b and 6c had disorder in
the cinnamyl ligands, so the uncertainty in the PdꢀC(allyl)
distances is high and small differences might be obscured. How-
ever, one may still conclude that there is no evidence of significant
conformational or electronic differences upon remote substitution
of the NHC aryl groups.
Quantitative Steric Parameters. Assuming that the para-
alkyl substituent on the aryl group of an NHC exerts a negligible
electronic influence, the only significant source of differences
comes from steric effects. To compare the ligand sterics to those
in literature NHC ligands, it is necessary to calculate quantitative
steric parameters %Vbur and G (defined below), which use the
X-ray crystal structures of 6a versus 6b and 6c. Because 6d and
6e were not crystallized, we compare HBF4 salts 4d and 4e to
Recent papers have introduced the use of solid angles as a
quantitative steric measure.19 These can be visualized by en-
visioning a light source at the metal, and a circumscribing sphere
of radius r upon which this light may fall unless it is blocked by the
ligand (Figure 5). The ligand casts a shadow upon an area A of
the sphere, and the solid angle distended by the ligand is Ω = A/r2.
(This differs from a cone angle in that the area A is not necessarily
circular.) Because solid angle values are in the inconvenient units
of steradians, it is preferable to divide Ω by 4π steradians (the
solid angle of a complete sphere) to give a unitless value called
percent coverage (G).19 G represents the fraction of a hollow
coordination sphere that is blocked by the ligand. For our
calculations of G parameters for NHC ligands, MꢀC bond
known IPr HX salts17 in order to gauge the influence of para
3
substituents for these ligands.
The percent buried volume (%Vbur) is defined as the percent
of the total volume of a sphere of radius 3.5 Å that is occupied by
the ligand. Table 2 gives %Vbur values for the new ligands from
5126
dx.doi.org/10.1021/om200349y |Organometallics 2011, 30, 5123–5132

Organometallics
ARTICLE
Scheme 4. SuzukiꢀMiyaura Reaction under Soluble
Conditions
Figure 5. The G parameter for a ligand describes the fraction of a
circumscribing sphere in the shadow of a ligand. Unlike a cone angle, the
shadowed area is not necessarily circular, and therefore, “gearing” of
ligands is taken into account.
Table 3. Calculated G Valuesa for IPr-Derived NHCs
compound
G (%) at distance d
2.00 Å
2.10 Å
2.28 Å
6a#1
42.3
43.4
49.2
42.5
56.6
45.3
48.1
46.9
48.6
58.7
47.1
57.6
48.1
47.4
43.9
41.2
42.2
48.0
41.1
55.6
44.0
46.7
45.5
47.2
57.4
45.6
56.5
46.8
46.3
42.7
39.1
40.1
46.1
39.1
53.9
41.8
44.4
43.2
44.7
55.3
43.1
54.5
44.3
44.3
40.6
6a#2
6b
6b (trunc.)b
6c
6c (trunc.)b
[IPrH]Cl17
[IPrH]OTf21
Figure 6. Plot of conversion versus time for the cross-coupling of
p-chloro(trifluoromethyl)benzene with phenylboronic acid at 60 °C
using a 100 ppm catalyst loading in iPrOH. Conversions measured using
gas chromatography (see the Experimental Section).
22
[IPrH]AlCl4
4d#1
4d#1 (trunc.)b
4d#2
4d#2 (trunc.)b
ligands for palladium in SuzukiꢀMiyaura couplings,2,24 IPr
complex 6a has been shown to be very active and robust as a
precatalyst for this reaction, with catalyst loadings as low as
50 ppm, using technical grade isopropanol as a solvent.12,25 It
might seem difficult to improve upon this reaction, but, as shown
below, the supersized versions of the catalyst are even more
active.
4e
4e (trunc.)b
a G is the fraction of the coordination sphere overshadowed by the
ligand. See text and ref 19. b Truncated by deleting atoms from the trityl
groups from the input file.
We chose the reaction of p-chloro(trifluoromethyl)benzene
and phenylboronic acid as a test reaction for study (Scheme 4).
The conditions used by Nolan and co-workers with catalyst 6a12
provided an excellent starting point for a qualitative comparison
to 6bꢀ6e. For an initial experiment, we compared the turnover
number (TON) and turnover frequency (TOF) of one trityl-
substituted catalyst (6c) to the parent catalyst 6a at Nolan’s
conditions (iPrOH, 60 °C, 100 ppm Pd), but at a halved substrate
concentration to help improve solubility. Under these condi-
tions, both catalysts 6a and 6c showed induction periods of ca.
20 min (Figure 6), and the trityl-substituted catalyst 6c showed a
slightly larger TON (19 000 in 85 min) and maximum TOF
(36 000 hꢀ1 measured between the 20 and 35 min time points)
than parent catalyst 6a (15 000 in 85 min, and 21 000 hꢀ1).
It is important to note that these conditions are not ideal for a
fair comparison of catalysts. For example, the reaction is hetero-
geneous under the reported conditions (1 mmol of aryl halide in
1 mL of isopropanol), and it is important that differences in
distances are again standardized for consistency, and G para-
meters using several MꢀC distances are shown in Table 3.20
Values are also given for the PdꢀNHC structures modified by
artificial truncation at the 40 position, as described above in the
discussion on %Vbur values.
The G values reflect the increase of size due to the para sub-
stituents on the NHC aryl groups. The G values decrease
markedly when the ligands are truncated, in contrast to the
situation described above for the %Vbur values. This decrease
upon truncation is about 4% for adamantyl-substituted 4e, about
7% for the trityl-substituted 6b, and 10ꢀ12% for larger trityl
groups in 6c and 4d. This shows that the para substituents
overshadow a part of the metal’s coordination sphere that was
not covered by the IPr ligand alone.
SuzukiꢀMiyaura Reactions. The SuzukiꢀMiyaura reaction
is a powerful method for the formation of carbonꢀcarbon
bonds.23 Although phosphines have typically been used as supporting
5127
dx.doi.org/10.1021/om200349y |Organometallics 2011, 30, 5123–5132

Organometallics
ARTICLE
the observed rate increases. For example, the bulky substituents
may inhibit aggregate formation. The differences in reactivity are
probably not due to differences in the stability of the catalyst,
because the overall final conversion of the SuzukiꢀMiyaura
reaction is similar with all catalysts, including the less-active IPr
complex 6a.
’ CONCLUSIONS
The synthesis of substituted IPr derivatives and the derived π-
cinnamylpalladium complexes reveals several interesting obser-
vations. The synthesis of 4-trityl derivatives of 2,6-diisopropyla-
niline is straightforward, as long as changes are implemented in
the synthesis to account for changes in solubility. These excep-
tionally large NHC ligands are special additions to the growing
number of bulky NHC ligands, which have many applications in
catalysis. Using SuzukiꢀMiyaura coupling conditions that are
modified for purposes of ligand comparison, it is clear that trityl-
subsitituted NHC complexes 6bꢀ6d produce palladium cata-
lysts that are significantly more active than the parent NHC
(“IPr”). Though %Vbur is popular for measuring steric effects as a
predictor of reactivity, it does not predict these changes in ligand
sterics within this family of IPr derivatives. This is because %Vbur
is influenced only by bulk close to a metal, though more distant
substituents can also have an important influence. G parameters
derived from solid angles are more effective in predicting the
changes in size and roughly correlate with the increase in catalytic
activity.
Figure 7. Plot of conversion versus time for the cross-coupling of p-
chloro(trifluoromethyl)benzene with phenylboronic acid at 60 °C using
i
a 0.02% catalyst loading in 9:1 PrOH/THF. Conversions measured
using gas chromatography (see the Experimental Section).
solubility should not influence the comparison of catalyst per-
formance. Therefore, in subsequent experiments, the concentra-
tions of aryl halide and boronic acid are reduced to 0.10 M and
the solvent system is a 9:1 mixture of isopropanol and THF. A
catalyst loading of 0.02% and a reaction temperature of 60 °C
gives a rate at which complexes 6aꢀ6e may be more reliably
compared (Figure 7). We reiterate that these conditions are
optimized for catalyst comparison, not for conversion.
Under the modified conditions, it is clear that the catalysts
6bꢀ6e, which have NHCs with remote substitution, give a
significant improvement in catalyst activity over the control
catalyst 6a. The rates roughly correlate with increased size as
measured with the G value, which indicates that the adamantyl-
substituted ligand covers a fraction of the coordination sphere
that is intermediate between the aryl with no para substituent and
the aryls with para-trityl groups. However, the predictive ability
of the solid angle is limited: though the solid angles are larger for
6c and 6d due to the additional groups on the trityl substituents,
the catalytic activity is not improved over the trityl-substituted
6b. It is possible that the G values will have a greater ability to
predict reactivity if they take ligand dynamics into account, as was
done in recent reports of NHC complexes with flexible steric
bulk.26,27
The new results reported here build on literature reports
showing that bulky NHC ligands perform exceptionally well
in SuzukiꢀMiyaura reactions.28 The beneficial effects of larger
supporting NHC ligands on the rate of the SuzukiꢀMiyaura
reaction are typically understood as the ability of the bulkier
ligand to stabilize monoligated Pd(0) intermediates.12,24,29 How-
ever, previous studies have focused on introducing substituents
very close to the metal center. The results described here show
that substituents distant from the metal can also have a marked
influence on catalytic rate. One potential explanation for the
higher activity of the trityl-substituted NHC ligands is that the
coordination of multiple NHC ligands (which would inhibit the
reaction) is prevented by steric interactions between the trityl
groups. Other influences of ligand size might also contribute to
’ EXPERIMENTAL SECTION
General Considerations. Unless otherwise noted, reactions were
performed under nitrogen. Tris(4-methylphenyl)methanol 1c,30 tris(4-tert-
butylphenyl)methanol 1d,31 4-(10-adamantyl)-2,6-diisopropylaniline 2e,32
[(π-cinnamyl)PdCl]2,12 and Pd complex 6a12 were prepared according to
published procedures. Potassium tert-butoxide was purchased from
Aldrich and purified via sublimation. 2,6-Diisopropylaniline was pur-
chased from TCI or Acros and purified by vacuum distillation prior to
use. p-Chloro(trifluoromethyl)benzene was purified by distillation
under nitrogen. Phenylboronic acid was purified via flash chromatogra-
phy using 2:1 ethyl acetate/hexanes.33,34 It is critical to purify the
phenylboronic acid prior to use in order to obtain reproducible solubility
in SuzukiꢀMiyaura coupling reactions. Isopropanol and THF were
degassed via nitrogen sparge for a minimum of 30 min before use in
SuzukiꢀMiyaura coupling experiments.
Aniline 2b. To a 500 mL Schlenk flask equipped with a stir bar was
added glacial acetic acid (150 mL), followed by triphenylmethanol 1b
(26.0 g, 100 mmol). The flask was then fitted with a rubber septum, and
the contents were sparged with nitrogen for ca. 15 min. 2,6-Diisopro-
pylaniline 2a (28 mL, 150 mmol) was then added via a syringe, followed
by concentrated HCl (9.1 mL, 110 mmol). The flask was then heated
in an 80 °C oil bath overnight, under an inert atmosphere. The reaction
mixture was then allowed to cool to room temperature. The resulting
solid was isolated via filtration and washed with water, followed by
methanol, and dried under vacuum. Crystallization from hot ethanol
1
yielded 2b as a white solid (31.9 g, 76% yield). HNMR (400 MHz,
CDCl3): δ 1.09 (d, J = 6.9 Hz, 12H, CHMe2), 2.87 (septet, J = 6.9 Hz,
2H, CHMe2), 3.65 (br s, 2H, NH2), 6.83 (s, 2H, ArH), 7.15ꢀ7.26
(m, 15H, ArH) ppm. 13C{1H} NMR (101.6 MHz, CDCl3): δ 22.46,
27.97, 64.80, 125.55, 126.05, 127.15, 131.18, 136.42, 137.69, 147.55
ppm. Anal. Calcd for C31H33N: C, 88.73; H, 7.93; N, 3.34. Found: C,
88.70; H, 8.07; N, 3.68.
5128
dx.doi.org/10.1021/om200349y |Organometallics 2011, 30, 5123–5132

Organometallics
ARTICLE
Aniline 2c. To a 500 mL flask equipped with a stir bar was added
glacial acetic acid (190 mL), followed by tris-(4-methylphenyl)methanol
1c (19.62 g, 64.9 mmol). The flask was then fitted with a rubber septum
and the contents sparged with nitrogen for ca. 15 min. 2,6-Diisopropy-
laniline 2a (18.4 mL, 97.3 mmol) was then added via a syringe, followed
by concentrated HCl (5.9 mL, 71 mmol). The flask was then heated in
an 80 °C oil bath for 8 h, under an inert atmosphere. The reaction
mixture was then allowed to cool to room temperature. A 2 M solution of
aqueous KOH (ca. 75 mL) was then slowly added to the reaction
mixture. The resulting solid was isolated via filtration and washed with
water, followed by methanol, and dried under vacuum to yield 2c as a
white solid (27.8 g, 92.7% yield). 1HNMR (400 MHz, CDCl3): δ 1.11
(d, J = 6.5 Hz, 12H, CHMe2), 2.31 (s, 9H, ArMe), 2.88 (septet, J =
6.8 Hz, 2H, CHMe2), 3.65 (br s, 2H, NH2), 6.87 (s, 2H, ArH), 7.02
(d, 6H, J = 8.0 Hz, ArH), 7.12 (d, 6H, J = 8.0 Hz, ArH) ppm. 13C{1H}
NMR (101.6 MHz, CDCl3): δ 20.85, 22.46, 27.97, 63.74, 125.88,
127.79, 130.87, 131.14, 134.66, 136.93, 137.35, 144.93 ppm. Anal. Calcd
for C34H39N: C, 88.45; H, 8.51; N, 3.03. Found: C, 88.19; H, 8.57;
N, 3.10.
Aniline 2d. To a 500 mL Schlenk flask equipped with a stir bar was
added glacial acetic acid (220 mL), followed by tris-(4-tert-butylphenyl)-
methanol 1d (15.58 g, 36.3 mmol). The flask was then fitted with a
rubber septum and the contents sparged with nitrogen for ca. 15 min.
2,6-Diisopropylaniline 2a (10.3 mL, 54.4 mmol) was then added via a
syringe, followed by concentrated HCl (3.3 mL, 40 mmol). The flask
was then heated in an 80 °C oil bath for 8 h, under an inert atmosphere.
The reaction mixture was then allowed to cool to room temperature,
then 2 M aqueous KOH (ca. 40 mL) was slowly added to the reaction
mixture. The resulting solid was isolated via filtration and washed with
water, followed by methanol, and dried under vacuum to yield 2d as a
white solid (20.6 g, 96.7% yield). 1H NMR (400 MHz, CDCl3): δ 1.08
(d, J = 6.9 Hz, 12H, CHMe2), 1.30 (s, 27H, CMe3), 2.86 (septet, J = 6.9
Hz, 2H, CHMe2), 3.62 (br s, 2H, NH2), 6.74 (s, 2H, ArH), 7.11 (d, 6H,
J = 8.5 Hz, ArH), 7.22 (d, 6H, J = 8.5 Hz, ArH) ppm. 13C{1H} NMR
(101.6 MHz, CDCl3): δ 22.42, 27.97, 31.40, 34.25, 63.52, 123.74,
126.18, 130.86, 137.17, 137.45, 144.60, 148.04 ppm. Anal. Calcd for
C43H57N: C, 87.85; H, 9.77; N, 2.38. Found: C, 87.83; H, 9.85; N, 2.42.
α-Diimine 3b. To a 500 mL flask equipped with a stir bar was added
aniline 2b (13.1 g, 31.2 mmol), followed by ethanol (225 mL), in air.
The flask was then fitted with a condenser and heated with an 80 °C oil
bath. After most of the material had dissolved, aqueous glyoxal (1.8 mL,
16 mmol) was added to the flask, followed by formic acid (2 drops). The
reaction mixture immediately began to develop a yellow color, followed
by the precipitation of a yellow solid. The flask was heated at 80 °C for 4
h and then allowed to cool to room temperature. The resulting solid was
isolated via filtration, washed with methanol, and dried under vacuum to
yield impure 3b as a yellow powder (12.0 g, 89.2% yield). 1HNMR (400
MHz, CDCl3): δ 1.03 (d, J = 6.8 Hz, 24H, CHMe2), 2.89 (septet, J = 6.8
Hz, 4H, CHMe2), 7.00 (s, 4H, ArH), 7.15ꢀ7.29 (m, 30H, ArH), 8.12 (s,
2H, ArNdCH) ppm.
α-Diimine 3d. To a 100 mL flask equipped with a stir bar was added
aniline 2d (3.02 g, 5.13 mmol), followed by ethanol (45 mL), in air.
The flask was then fitted with a condenser and heated with an 80 °C
oil bath. After most of the material dissolved, aqueous glyoxal (0.29 mL,
2.6 mmol) was added to the flask, followed by formic acid (1 drop). The
reaction mixture immediately began to develop a yellow color, followed
by the precipitation of a yellow solid. The flask was heated at 80 °C for
4 h and then allowed to cool to room temperature. The resulting solid
was isolated via filtration and washed with methanol and dried under
vacuum to yield impure 3d as a yellow powder (2.78 g, 90.3% yield).
1H NMR (400 MHz, CDCl3): δ 1.00 (d, J = 6.8 Hz, 24H, CHMe2), 1.31
(s, 54H, CMe3), 2.87 (septet, J = 6.9 Hz, 4H, CHMe2), 6.88 (s, 2H,
ArH), 7.11 (d, 12H, J = 8.5 Hz, ArH), 7.25 (d, 12H, J = 8.5 Hz, ArH),
8.10 (s, 2H, ArNdCH) ppm.
α-Diimine 3e. To a 250 mL flask equipped with a stir bar was added
aniline 2e (7.29 g, 23.4 mmol), followed by ethanol (100 mL), under air.
The flask was then heated with a heat gun until all solids dissolved.
Aqueous glyoxal (1.3 mL, 12 mmol) was then added to the flask,
followed by formic acid (2 drops). The reaction mixture immediately
began to develop a yellow color, followed by the precipitation of a yellow
solid. The mixture was stirred at room temperature for 10 h. The
resulting solid was isolated via filtration, washed with methanol, and
dried under vacuum to yield impure 3e as a yellow powder (5.56 g,
73.6% yield). 1H NMR (400 MHz, CDCl3): δ 1.21 (d, J = 7.0 Hz, 24H,
CHMe2), 1.80 (br s, 12H, AdH), 1.96 (br s, 12H, AdH), 2.12 (br s, 6H,
AdH), 2.96 (septet, J = 6.7 Hz, 4H, CHMe2), 7.17 (s, 2H, ArH), 8.11 (s,
2H, ArNdCH) ppm.
Imidazolium Tetrafluoroborate 4b. Yellow solid 3b (12.0 g,
13.9 mmol) was added to a 250 mL flask with a stir bar, followed by
paraformaldehyde powder (417 mg, 13.9 mmol), followed by toluene
(100 mL), under air. The flask was then briefly heated to dissolve most of
the solids and allowed to cool back to room temperature. After cooling,
HBF4 Et2O (2.8 mL, 21 mmol) was slowly added to the reaction
3
mixture, allowing each drop of acid to fully disperse before adding
another drop, with a total addition time of ca. 1 h. The reaction mixture
was stirred at room temperature overnight. The resulting solid was
isolated by filtration, washed with ether, and was purified by recrystalli-
zation from hot benzene to yield imidazolium tetrafluoroborate 4b as a
white solid (4.67 g, 35.0% yield). 1H NMR (400 MHz, CDCl3): δ 1.06
(d, J = 6.7 Hz, 12H, CHMe2), 1.10 (d, J = 7.0 Hz, 12H, CHMe2),
2.87 (m, 4H, CHMe2), 7.22ꢀ7.35 (m, 30H, ArH),7.88 (s, 2H,
RNCHCHNR), 8.71 (s, 1H, RNCHNR) ppm. 13C{1H} NMR (101.6
MHz, CDCl3): δ 23.53, 24.18, 28.98, 65.32, 126.23, 126.73, 127.18,
127.67, 127.73, 130.89, 136.83, 143.58, 145.92, 151.27 ppm. Anal. Calcd
for C65H65BF4N2: C, 81.24; H, 6.82; N, 2.91. Found: C, 81.17; H, 6.74;
N, 2.96.
Imidazolium Tetrafluoroborate 4c. Yellow solid 3c (4.69 g,
4.96 mmol) was added to a 100 mL flask with a stir bar, followed by
paraformaldehyde powder (190 mg, 6.2 mmol), followed by toluene
(50 mL), under air. The flask was then briefly heated with a heat gun to
dissolve most of the solids and allowed to cool back to room tempera-
α-Diimine 3c. To a 250 mL flask equipped with a stir bar was added
aniline 2c (4.62 g, 10.0 mmol), followed by ethanol (90 mL), in air. The
flask was then fitted with a condenser and heated with an 80 °C oil bath.
After most of the material dissolved, aqueous glyoxal (0.57 mL, 5.0 mmol)
was added to the flask, followed by formic acid (1 drop). The reaction
mixture immediately began to develop a yellow color, followed by the
precipitation of a yellow solid. The flask was heated at 80 °C for 4 h and
thenallowed tocoolto room temperature. The resulting solid was isolated
via filtration, washed with methanol, and dried under vacuum to yield
impure 3c as a yellow powder (4.69 g, 99.1% yield). 1H NMR (400 MHz,
CDCl3): δ 1.27 (d, J = 6.8 Hz, 24H, CHMe2), 2.56 (s, 18H, ArMe),
3.12 (septet, J = 6.8 Hz, 4H, CHMe2), 7.29 (d, 12H, J = 8.8 Hz, ArH),
7.37 (d, 12H, J = 8.8 Hz, ArH), 7.51 (s, 4H, ArH), 8.35 (s, 2H,
ArNdCH) ppm.
ture. After cooling, HBF4 Et2O (1.0 mL, 7.4 mmol) was slowly added to
3
the reaction mixture, allowing each drop of acid to fully disperse before
adding another drop, with a total addition time of ca. 30 min. The
reaction stirred at room temperature overnight. The reaction was
then quenched by the dropwise addition of aqueous sodium bicarbonate
(ca. 5 mL). The resulting solid was isolated by filtration and was purified
by recrystallization from hot benzene to yield imidazolium tetrafluor-
oborate 4c as a white solid (1.45 g, 28.0% yield). 1HNMR (400 MHz,
CDCl3): δ 1.00 (d, J = 6.8 Hz, 12H, CHMe2), 1.10 (d, J = 6.8 Hz, 12H,
CHMe2), 2.29ꢀ2.37 (m, 22H, ArMe and CHMe2), 7.07 (d, J = 8.8 Hz,
12H, ArH), 7.12 (d, J = 8.8 Hz, 12H, ArH), 7.25 (s, 4H, ArH), 7.94
(s, 2H, J = 8.0 Hz, RNCHCHNR), 8.36 (s, 1H, RNCHNR) ppm.
13C{1H} NMR (101.6 MHz, CDCl3): δ 20.85, 23.55, 24.31, 28.98,
5129
dx.doi.org/10.1021/om200349y |Organometallics 2011, 30, 5123–5132

Organometallics
ARTICLE
64.36, 126.84, 127.11, 127.57, 128.38, 130.58, 130.67, 135.40, 135.58,
136.13, 143.30, 151.87 ppm. Anal. Calcd for C71H77BF4N2: C, 81.59; H,
7.43; N, 2.68. Found: C, 81.82; H, 7.41; N, 2.79.
Imidazolium Tetrafluoroborate 4d. Yellow solid 4d (2.71 g,
2.26 mmol) was added to a 50 mL flask with a stir bar, followed by
paraformaldehyde powder (68.0 mg, 2.26 mmol), followed by toluene
(25 mL), under air. The flask was then briefly heated with a heat gun to
dissolve most of the solids and allowed to cool back to room tempera-
The resulting solid was extracted with a 1:4 mixture of toluene and
benzene (ca. 5 mL) and then filtered through Celite. The solids were
washed with 2 ꢁ 1 mL portions of benzene, and the combined filtrate
was evaporated under vacuum to yield NHC 5c as a white solid (480 mg,
>99% yield). 1H NMR (400 MHz, C6D6): δ 1.10 (d, J = 6.9 Hz, 12H,
CHMe2), 1.29 (d, J = 6.9 Hz, 12H, CHMe2), 2.07 (s, 18H, ArMe),
2.98 (septet, J = 6.9 Hz, 4H, CHMe2), 6.54 (s, 2H, RNCHCHNR), 6.97
(d, J = 8.0 Hz, 12H, ArH), 7.51 (d, J = 8.0 Hz, 12H, ArH), 7.58 (s, 4H,
ArH) ppm. 13C{1H} NMR (101.6 MHz, C6D6): δ 20.55, 23.26, 24.49,
28.65, 64.81, 121.29, 125.38, 126.63, 129.01, 131.21, 135.18, 136.35,
144.72, 147.65, 220.34 ppm. Anal. Calcd for C71H76N2: C, 89.07; H,
8.00; N, 2.93. Found: C, 87.58; H, 8.28; N, 3.14.
NHC 5d. In the glovebox, a scintillation vial was charged with a stir
bar, 4d (324 mg, 0.250 mmol), and potassium tert-butoxide (31.0 mg,
0.275 mmol), followed by toluene (ca. 3 mL). The resulting mixture was
stirred overnight. The volatile materials were removed under vacuum.
The resulting solid was extracted with a 1:4 mixture of toluene and
benzene (ca. 5 mL) and then filtered through Celite. The solids
were washed with 2 ꢁ 1 mL portions of benzene, and the combined
filtrate was concentrated under vacuum to yield NHC 5d as a white solid
(300 mg, >99% yield). 1H NMR (400 MHz, C6D6): δ 1.08 (d, J = 7.0 Hz,
12H, CHMe2), 1.20 (s, 54H, CMe3), 1.28 (d, J = 6.5 Hz, 12H, CHMe2),
2.99 (septet, J = 6.8 Hz, 4H, CHMe2), 6.56 (s, 2H, RNCHCHNR), 7.21
(d, 12H, J = 8.5 Hz, ArH), 7.50 (s, 4H, ArH), 7.53 (d, 12H, J = 8.5 Hz,
ArH) ppm. 13C{1H} NMR (101.6 MHz, C6D6): δ 23.14, 24.48, 28.64,
31.14, 34.05, 64.56, 121.28, 124.40, 125.38, 131.26, 136.43, 144.55,
147.69, 148.44, 220.35 ppm. Anal. Calcd for C89H112N2: C, 88.35; H,
9.33; N, 2.32. Found: C, 87.34; H, 8.96; N, 2.28.
ture. After cooling, HBF4 Et2O (0.5 mL, 3 mmol) was slowly added to
3
the reaction mixture, allowing each drop of acid to fully disperse before
adding another drop, with a total addition time of ca. 30 min. The
reaction stirred at room temperature overnight. The reaction was then
quenched by the dropwise addition of aqueous sodium bicarbonate (ca.
3 mL). The resulting solid was isolated by filtration and was purified by
recrystallization from hot benzene to yield imidazolium tetrafluorobo-
rate 4d as a white solid (1.04 g, 35% yield). 1HNMR (400 MHz, CDCl3):
δ 0.97 (d, J = 6.8 Hz, 12H, CHMe2), 1.08 (d, J = 6.8 Hz, 12H, CHMe2),
1.30 (s, 54H, CMe3), 2.34 (septet, J = 6.8 Hz, 4H, CHMe2), 7.09
(d, 12H, J = 8.8 Hz, ArH), 7.12 (s, 4H, ArH), 7.28 (d, 12H, J = 8.8 Hz,
ArH), 8.00 (s, 2H, RNCHCHNR), 8.28 (s, 1H, RNCHNR) ppm.
13C{1H} NMR (101.6 MHz, CDCl3): δ 23.44, 24.29, 28.90, 31.27,
34.31, 64.03, 123.92, 124.30, 127.87, 128.15, 130.57, 141.22, 142.86,
143.17, 148.95, 149.40 ppm. Anal. Calcd for C89H113BF4N2: C, 82.37;
H, 8.78; N, 2.16. Found: C, 82.39; H, 8.70; N, 2.08.
Imidazolium Tetrafluoroborate 4e. Yellow solid 3e (5.55 g,
8.61 mmol) was added to a 50 mL flask with a stir bar, followed by
paraformaldehyde powder (259 mg, 8.61 mmol), followed by toluene
(20 mL). The flask was then briefly heated with a heat gun to dissolve
most of the solids and allowed to cool back to room temperature.
NHC 5e. In the glovebox, a scintillation vial equipped with a stir bar
was charged with 4e (0.37 g, 0.50 mmol) and potassium tert-butoxide
(56 mg, 0.50 mmol), followed by toluene (ca. 5 mL). The resulting
mixture was stirred overnight. The volatile materials were removed
under vacuum. The resulting solid was extracted with a 1:4 mixture of
toluene and benzene (ca. 5 mL) and then filtered through Celite. The
solids were washed with 2 ꢁ 1 mL portions of benzene, and the
combined filtrate was evaporated under vacuum to yield NHC 5e as
a white solid (258 mg, 78% yield). 1H NMR (400 MHz, C6D6): δ 1.32
(d, J = 7.0 Hz, 12H, CHMe2), 1.39 (d, J = 7.0 Hz, 12H, CHMe2), 1.73
(m, 12H, AdH), 2.04 (m, 18H, AdH), 3.08 (septet, J = 7.0 Hz, 4H,
CHMe2), 6.67 (s, 2H, RNCHCHNR), 7.45 (s, 4H, ArH) ppm. 13C{1H}
NMR (101.6 MHz, C6D6): δ 23.56, 24.72, 28.83, 29.16, 36.57, 36.80,
43.39, 119.76, 121.38, 136.57, 145.32, 151.25, 220.27 ppm. Anal. Calcd for
C47H64N2: C, 85.92; H, 9.82; N, 4.26. Found: C, 86.25; H, 9.82; N, 4.56.
Complex 6b. In the glovebox, a scintillation vial was charged
with [(cinnamyl)PdCl]2 (62 mg, 0.12 mmol) and NHC 5b (210 mg,
0.24 mmol), a stir bar, and toluene (ca. 3 mL). The reaction mixture
stirred overnight and was then removed from the glovebox. The solvent
was removed under vacuum, and the resulting solid was extracted with
benzene. The mixture was filtered through Celite and dried under
vacuum to yield 6b as a tan solid (265 mg, 98% yield). 1H NMR (400
MHz, C6D6): δ 0.99 (d, J = 6.5 Hz, 12H, CHMe2), 1.31 (d, J = 6.5 Hz,
6H, CHMe2), 1.37 (d, J = 6.5 Hz, 6H, CHMe2), 1.87 (d, J = 11 Hz,
1H, cinnamylH), 3.01 (d, J = 6.5 Hz, 1H, cinnamylH), 3.13 (sept, J =
6.5 Hz, 2H, CHMe2), 3.26 (septet, J = 6.5 Hz, 2H, CHMe2), 4.46 (d, J =
13 Hz, 1H, cinnamylH), 5.14ꢀ5.21 (m, 1H, cinnamylH), 6.65 (s, 2H,
RNCHCHNR), 6.93ꢀ7.00 (m, 6H, ArH), 7.01ꢀ7.05 (m, 1H, ArH),
7.07ꢀ7.11 (m, 12H, ArH), 7.21ꢀ7.26 (m, 2H, ArH), 7.40ꢀ7.48 (m,
16H, ArH) ppm. 13C{1H} NMR (101.6 MHz, CDCl3): δ 22.61, 25.86,
28.47, 46.63, 65.31, 89.24, 108.97, 124.01, 125.93, 126.55, 126.80,
127.35, 127.47, 128.23, 131.06, 133.53, 138.23, 144.71, 146.62,
147.96, 185.29 ppm. Anal. Calcd for C74H73ClN2Pd: C, 78.50; H,
6.50; N, 2.47. Found: C, 78.43; H, 6.41; N, 2.61.
After cooling, HBF4 Et2O (1.5 mL, 11 mmol) was slowly added to
3
the reaction mixture, allowing each drop of acid to fully disperse before
adding another drop, with a total addition time of ca. 30 min. The
reaction was stirred at room temperature overnight. The resulting solid
was isolated by filtration, washed with ether, and was purified by
recrystallization from acetone/ether to yield imidazolium tetrafluoro-
1
borate 4e as a tan solid (1.30 g, 20.0% yield). H NMR (400 MHz,
CDCl3): δ 1.18 (d, J = 7.0 Hz, 12H, CHMe2), 1.28 (d, J = 6.5 Hz, 12H,
CHMe2), 1.80 (dd, J = 12.5 and 7.0 Hz, 12H, AdH), 1.94 (br s, 12H,
AdH), 2.14 (br s, 6H, AdH), 2.41 (septet, J = 7.0 Hz,4H, CHMe2), 7.30
(s, 4H, ArH), 7.82 (s, 2H, RNCHCHNR), 8.38 (s, 1H, RNCHNR) ppm.
13C{1H} NMR (101.6 MHz, CDCl3): δ 23.82, 24.55, 28.74, 29.14,
36.54, 36.80, 42.91, 121.22, 126.96, 127.11, 136.28, 144.16, 155.37 ppm.
Anal. Calcd for C47H65BF4N2: C, 75.79; H, 8.80; N, 3.76. Found: C,
75.73; H, 8.79; N, 4.06.
NHC 5b. In a N2-filled glovebox, a scintillation vial was charged with
4b (0.48 g, 0.50 mmol), potassium tert-butoxide (56 mg, 0.50 mmol),
and a stir bar, followed by toluene (ca. 5 mL). The resulting mixture was
stirred overnight. The volatile materials were removed under vacuum.
The resulting solid was extracted with a 1:4 mixture of toluene and
benzene (ca. 5 mL) and then filtered through Celite. The solids were
washed with 2 ꢁ 1 mL portions of benzene, and the combined filtrate
was evaporated under vacuum to yield NHC 5b as a white solid (407 mg,
1
93% yield). H NMR (400 MHz, C6D6): δ 1.06 (d, J = 6.5 Hz, 12H,
CHMe2), 1.25 (d, J = 6.5 Hz, 12H, CHMe2), 2.95 (septet, J = 6.9 Hz, 4H,
CHMe2), 6.53 (s, 2H, RNCHCHNR), 6.95ꢀ7.05 (m, 6H, ArH),
7.07ꢀ7.12 (m, 12H, ArH), 7.45 (s, 4H, ArH), 7.46ꢀ7.51 (m, 12H,
ArH) ppm. 13C{1H} NMR (101.6 MHz, C6D6): δ 23.22, 24.40, 28.60,
65.64, 121.28, 126.05, 126.76, 129.02, 131.36, 136.43, 144.80, 147.12,
219.95 ppm. Anal. Calcd for C65H64N2: C, 89.40; H, 7.39; N, 3.21.
Found: C, 88.82; H, 7.48; N, 3.44.
NHC 5c. In the glovebox, a scintillation vial was charged with a
stir bar, 5c (520 mg, 0.50 mmol), and potassium tert-butoxide (62 mg,
0.55 mmol), followed by toluene (ca. 5 mL). The resulting mixture was
stirred overnight. The volatile materials were removed under vacuum.
Complex 6c. In the glovebox, a scintillation vial was charged with
[(cinnamyl)PdCl]2 (130 mg, 0.25 mmol), NHC5c(480 mg, 0.50 mmol),
5130
dx.doi.org/10.1021/om200349y |Organometallics 2011, 30, 5123–5132

Organometallics
ARTICLE
a stir bar, and toluene (ca. 3 mL). The reaction was stirred for 4 h and was
then removed from the glovebox. The reaction mixture was loaded
directly onto a 25 mm column containing ca. 50 mL silica. The column
was initially eluted with hexanes (ca. 75 mL) to immobilize all colored
bands and then eluted with 10% ether in hexanes (v/v) until the
first yellow band was collected. The yellow fractions were combined,
concentrated, and dried under vacuum to yield 6c as a tan solid (401 mg,
nitrogen for ca. 5 min. Mesitylene (139 μL, 1.00 mmol) and p-chloro-
(trifluoromethyl)benzene (267 μL, 2.00 mmol) were then added to
the 1 mL volumetric flask via a syringe. The liquids were then dissolved
in THF, and the solution was diluted to 1.0 mL. Catalyst stock solutions
were prepared at a concentration of 0.40 mM by first preparing
stock solutions at a concentration of 4 mM and then performing 1/10
dilutions.
Each reaction was performed in a 1 dram vial containing a flea stir bar
and capped with a septum cap. Each vial was sparged with nitrogen for
ca. 5 min prior to use. To each vial was added 900 μL of the
phenylboronic acid stock solution and 50 μL of the aryl halide stock
solution. Each vial was then heated using an aluminum block heater set
to 60 °C. After thermal equilibration, each reaction was initiated via the
addition of 50 μL of the appropriate 0.40 mM catalyst solution, for a final
palladium concentration of 20 μM. Aliquots (100 μL) were removed at
reaction times of 80, 220, and 900 min. Aliquots were purified by
filtration through pipet filters containing ca. 1 cm of silica and eluted with
1ꢀ1.2 mL of ethyl acetate directly into GC vials. Conversion was
determined by comparison of the GC responses of product and the
internal mesitylene standard.
1
66% yield). H NMR (400 MHz, C6D6): δ 0.99 (d, J = 6.0 Hz, 12H,
CHMe2), 1.33 (d, J = 6.5 Hz, 6H, CHMe2), 1.39 (d, J = 6.0 Hz, 6H,
CHMe2), 2.07 (s, 18H, ArMe), 2.11 (d, J = 2 Hz, 1H, cinnamylH),
3.01 (d, J = 6.0 Hz, 1H, cinnamylH), 3.14 (m, 2H, CHMe2), 3.26
(m, 2H, CHMe2), 4.41 (d, J = 13 Hz, 1H, cinnamylH), 5.09ꢀ5.17
(m, 1H, cinnamylH), 6.64 (s, 2H, RNCHCHNR), 6.98 (d, J = 8.0 Hz,
12H, ArH), 7.01ꢀ7.25 (m, 5H, ArH), 7.66 (d, J = 8.0 Hz, 12H, ArH),
7.56 (s, 4H, ArH). 13C{1H} NMR (101.6 MHz, CDCl3): δ 20.86, 22.62,
25.88, 28.47, 46.64, 64.34, 89.10, 108.95, 123.98, 126.46, 127.31, 128.02,
128.99, 130.91, 133.37, 135.14, 144.07, 144.52, 148.36, 185.33. Anal.
Calcd for C80H85ClN2Pd: C, 78.99; H, 7.04; N, 2.30. Found: C, 79.24;
H, 7.03; N, 2.21.
Complex 6d. In the glovebox, a scintillation vial was charged with
[(cinnamyl)PdCl]2 (64.8 mg, 0.125 mmol) and NHC 5d (300 mg,
0.250 mmol), a stir bar, and toluene (ca. 2 mL). The reaction was stirred
for 3 h and was then removed from the glovebox. The reaction mixture
was loaded directly onto a 25 mm column containing ca. 40 mL silica.
The column was initially eluted with hexanes (ca. 75 mL) to immobilize
all colored bands and then eluted with 10% ether in hexanes (v/v) until
the first yellow band was collected. The yellow fractions were combined,
concentrated, and dried under vacuum to yield 6d as a tan solid (300 mg,
81.7% yield) 1H NMR (400 MHz, C6D6): δ 0.99 (br s, 12H, CHMe2),
1.19 (s, 54H, CMe3), 1.27ꢀ1.39 (m, 12H, CHMe2), 1.92 (d, J = 10 Hz,
1H, cinnamylH), 3.01 (d, J = 5.5 Hz, 1H, cinnamylH), 2.98ꢀ3.32
(m, 4H, CHMe2), 4.45 (d, J = 13 Hz, 1H, cinnamylH), 5.11ꢀ5.23
(m, 1H, cinnamylH), 6.66 (s, 2H, RNCHCHNR), 7.23 (d, J = 8.5 Hz,
12H, ArH), 7.29 (d, J = 7.0 Hz, 2H, ArH), 7.52 (d, J = 8.5 Hz, 15H, ArH).
13C{1H} NMR (101.6 MHz, CDCl3): δ 22.61, 25.86, 28.39, 31.32,
34.27, 46.76, 64.02, 88.95, 109.20, 124.02, 126.44, 126.95, 127.39,
128.17, 128.98, 130.74, 133.39, 143.69, 144.24, 148.49, 185.41. Anal.
Calcd for C98H121ClN2Pd: C, 80.13; H, 8.30; N, 1.91. Found: C, 80.22;
H, 8.54; N, 2.06.
’ ASSOCIATED CONTENT
S
Supporting Information. Crystallographic details. This
b
material is available free of charge via the Internet at http://pubs.
acs.org.
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: holland@chem.rochester.edu.
’ ACKNOWLEDGMENT
This research was supported by the U.S. Department of Energy,
Office of Basic Energy Sciences, Grant No. DE-FG02-09ER16089.
We thank William Brennessel for assistance with X-ray crystal-
lography.
Complex 6e. In the glovebox, a scintillation vial equipped with a stir
bar was charged with [(cinnamyl)PdCl]2 (41 mg, 0.078 mmol) and
NHC 5e (103 mg, 0.16 mmol), followed by toluene (ca. 2 mL). The
reaction mixture stirred overnight and was then removed from the
glovebox. The solvent was removed under vacuum and the resulting
solid extracted with benzene. The mixture was filtered through Celite
and dried under vacuum to yield 6e as a tan solid (135 mg, 94% yield).
1H NMR (400 MHz, C6D6): δ 1.09ꢀ1.21 (m, 12H), 1.48ꢀ1.70 (m,
24H), 1.89ꢀ2.00 (m, 22H), 3.16 (d, J = 6.5 Hz, 1H, cinnamylH), 3.31
(septet, J = 6.5 Hz, 2H, CHMe2), 3.42 (septet, J = 7.0 Hz, 2H, CHMe2),
4.21 (d, J = 12.5 Hz, 1H, cinnamylH), 4.99ꢀ5.09 (m, 1H, cinnamylH),
6.68 (s, 2H, RNCHCHNR), 6.91ꢀ7.00 (m, 3H, ArH), 7.05ꢀ7.11
(m, 2H, ArH), 7.44 (s, 4H, ArH). 13C{1H} NMR (101.6 MHz, CDCl3):
δ 22.60, 26.24, 28.62, 28.95, 31.54, 36.77, 43.18, 47.49, 87.73, 108.97,
120.04, 124.23, 126.38, 127.25, 128.08, 133.50, 138.43, 145.22, 152.35,
185.45. Anal. Calcd for C56H73ClN2Pd: C, 73.42; H, 8.03; N, 3.06.
Found: C, 73.63; H, 7.70; N, 2.88.
Suzuki Reactions. As noted above, it is critical to purify commer-
cial phenylboronic acid by flash chromatography in order to obtain
reproducible solubility. A 5 mL volumetric flask was charged with
phenylboronic acid (71.1 mg, 0.583 mmol) and potassium tert-butoxide
(68.6 mg, 0.611 mmol). The volumetric flask was then capped with a
rubber septum and sparged with nitrogen for ca. 10 min. The solids were
then dissolved in isopropanol, and the solution was diluted to 5.0 mL. A
1 mL volumetric flask was capped with a rubber septum and sparged with
’ REFERENCES
(1) Trnka, T. M.; Grubbs, R. H. Acc. Chem. Res. 2000, 34, 18.
(2) Martin, R.; Buchwald, S. L. Acc. Chem. Res. 2008, 41, 1461.
(3) Hahn, F. E.; Jahnke, M. C. Angew. Chem., Int. Ed. 2008, 47, 3122.
(4) Díez-Gonzꢀalez, S.; Marion, N.; Nolan, S. P. Chem. Rev. 2009,
109, 3612.
(5) Jones, W. D. J. Am. Chem. Soc. 2009, 131, 15075.
(6) Dr€oge, T.; Glorius, F. Angew. Chem., Int. Ed. 2010, 49, 6940.
(7) Dorta, R.; Stevens, E. D.; Scott, N. M.; Costabile, C.; Cavallo, L.;
Hoff, C. D.; Nolan, S. P. J. Am. Chem. Soc. 2005, 127, 2485.
(8) Chianese, A. R.; Li, X.; Janzen, M. C.; Faller, J. W.; Crabtree,
R. H. Organometallics 2003, 22, 1663.
(9) Tolman, C. A. Chem. Rev. 1977, 77, 313.
(10) Cavallo, L.; Correa, A.; Costabile, C.; Jacobsen, H. J. Organomet.
Chem. 2005, 690, 5407.
(11) Clavier, H.; Nolan, S. P. Chem. Commun. 2010, 46, 841.
(12) Marion, N.; Navarro, O.; Mei, J.; Stevens, E. D.; Scott, N. M.;
Nolan, S. P. J. Am. Chem. Soc. 2006, 128, 4101.
(13) Benkeser, R. A.; Schroeder, W. J. Am. Chem. Soc. 1958, 80, 3314.
(14) Clapp, D. B. J. Am. Chem. Soc. 1939, 61, 523.
(15) Huang, J.; Nolan, S. P. J. Am. Chem. Soc. 1999, 121, 9889.
(16) B€ohm, V. P. W.; Weskamp, T.; Gst€ottmayr, C. W. K.; Herrmann,
W. A. Angew. Chem., Int. Ed. 2000, 39, 1602.
(17) Arduengo, A. J.; Krafczyk, R.; Schmutzler, R.; Craig, H. A.;
Goerlich, J. R.; Marshall, W. J.; Unverzagt, M. Tetrahedron 1999, 55,
14523.
5131
dx.doi.org/10.1021/om200349y |Organometallics 2011, 30, 5123–5132

Organometallics
ARTICLE
(18) Poater, A.; Cosenza, B.; Correa, A.; Giudice, S.; Ragone, F.;
Scarano, V.; Cavallo, L. Eur. J. Inorg. Chem. 2009, 1759.
(19) Guzei, I. A.; Wendt, M. Dalton Trans. 2006, 3991.
(20) Guzei, I. A.; Wendt, M. Solid-G; University of Wisconsin:
Madison, WI, 2004.
(21) Blue, E. D.; Gunnoe, T. B.; Petersen, J. L.; Boyle, P. D.
J. Organomet. Chem. 2006, 691, 5988.
(22) Stasch, A.; Singh, S.; Roesky, H. W.; Noltemeyer, M.; Schmidt,
H.-G. Eur. J. Inorg. Chem. 2004, 4052.
(23) Miyaura, N.; Suzuki, A. Chem. Rev. 1995, 95, 2457.
(24) Barder, T. E.; Walker, S. D.; Martinelli, J. R.; Buchwald, S. L.
J. Am. Chem. Soc. 2005, 127, 4685.
(25) Navarro, O.; Marion, N.; Mei, J.; Nolan, S. P. Chem.—Eur.
J. 2006, 12, 5142.
(26) Altenhoff, G.; Goddard, R.; Lehmann, C. W.; Glorius, F. Angew.
Chem., Int. Ed. 2003, 42, 3690.
(27) Ragone, F.; Poater, A.; Cavallo, L. J. Am. Chem. Soc. 2010, 132,
4249.
(28) W€urtz, S.; Glorius, F. Acc. Chem. Res. 2008, 41, 1523.
(29) Lavallo, V.; Canac, Y.; Pr€asang, C.; Donnadieu, B.; Bertrand, G.
Angew. Chem., Int. Ed. 2005, 44, 5705.
(30) Sakata, T.; Maruyama, S.; Ueda, A.; Otsuka, H.; Miyahara, Y.
Langmuir 2007, 23, 2269.
(31) Zhang, W.; Yamamoto, H. J. Am. Chem. Soc. 2006, 129, 286.
(32) Tang, Y.; Zakharov, L. N.; Rheingold, A. L.; Kemp, R. A. Inorg.
Chim. Acta 2006, 359, 775.
(33) Hong, S. Personal communication.
(34) Genov, M.; Almorín, A.; Espinet, P. Chem.—Eur. J. 2006, 12,
9346.
5132
dx.doi.org/10.1021/om200349y |Organometallics 2011, 30, 5123–5132