Steric effects in the aerobic oxidation of π-Allylnickel(II) complexes with N-heterocyclic carbenes

Article abstract of DOI:10.1021/ic0612451

π-Allylchloro(NHC)nickel(II) complexes were synthesized and their reactions with O₂ were studied. Ligand steric effects were found to determine the difference between rapid oxidation of the allyl group to produce bis-μ-hydroxonickel complexes and no observable reaction. The ability of the metal-NHC bond to rotate correlates with the ability of the complex to react with O₂. In the limiting cases, conformationally restricted complexes are stable to O₂ and complexes with rapid Ni-NHC bond rotation react rapidly with O₂. Complexes with intermediate conformational flexibility were found to exhibit lesser reactivity with O₂. On the basis of the observed inertness of complexes with saddle-shaped ligands to O₂, we propose the adoption of a nonplanar geometry upon reaction with O₂ to be required. The issue of conformational flexibility versus rigidity is expected to directly impact the catalytic behavior of metal-NHC complexes.

Full text of DOI:10.1021/ic0612451

Inorg. Chem. 2006, 45, 8430−8441
Steric Effects in the Aerobic Oxidation of
with N-Heterocyclic Carbenes
π-Allylnickel(II) Complexes
Benjamin R. Dible and Matthew S. Sigman*
Department of Chemistry, UniVersity of Utah, Salt Lake City, Utah 84112
Received July 5, 2006
π
-Allylchloro(NHC)nickel(II) complexes were synthesized and their reactions with O2 were studied. Ligand steric
effects were found to determine the difference between rapid oxidation of the allyl group to produce bis- -hydroxonickel
complexes and no observable reaction. The ability of the metal NHC bond to rotate correlates with the ability of
the complex to react with O2. In the limiting cases, conformationally restricted complexes are stable to O2 and
complexes with rapid Ni NHC bond rotation react rapidly with O2. Complexes with intermediate conformational
µ
−
−
flexibility were found to exhibit lesser reactivity with O2. On the basis of the observed inertness of complexes with
saddle-shaped ligands to O2, we propose the adoption of a nonplanar geometry upon reaction with O2 to be
required. The issue of conformational flexibility versus rigidity is expected to directly impact the catalytic behavior
of metal−NHC complexes.
Table 1. Tolman Electronic Parameters (TEPs) for NHCs and Selected
Phosphines
Introduction
The emergence of N-heterocyclic carbenes (NHCs, see
Figure 1) since Arduengo’s seminal publication in 19911 has
invigorated the catalysis field by providing a powerful
complement to phosphine/arsine ligands for a wide array of
reactions from olefin metathesis2 to cross-coupling.3,4 Among
the useful properties of NHCs is a pronounced stability to
O2, which makes them potent ligands for transition-metal-
mediated aerobic oxidations.5-9
ligand
v (cm-1
)
ligand
v (cm-1
)
ICya
2049.6
2050
2051
2050.7
2051.5
2051.5
2052.2
IBioxMe4
2054
c
tmiyb
biyb
P(tBu)3
2056.1
2060.3
2064.1
2066.6
2068.9
d
d
PBu3
IMesa
SIMesa
IiPra
PMe3
d
d
P(o-Tol)3
d
PPh3
SliPra
a Nolan.12 b Crabtree.11 c Glorius.13 d Tolman.10
It is widely believed that NHCs are stronger σ-donors than
phosphines. Measurements of Tolman electronic parameters10
(TEP) by Crabtree11 and Nolan12 strongly support this
viewpoint (Table 1). What is striking about these measure-
ments is how small the differences in TEP values are among
reported NHCs, essentially falling within a 3 cm-1 range
(4.5 cm-1 if IBioxMe4 is included).13 On the other hand, TEP
values for phosphines vary by more than 12 cm-1. It should
be noted that the TEP values reported by Crabtree11 and
Glorius13 were determined using iridium complexes. The
relatively narrow range of TEP values for NHCs as compared
to phosphines suggests that differences in the properties of
metal-NHC complexes with varying NHCs are more likely
to stem from steric effects.
* To whom correspondence should be addressed. E-mail: sigman@
chem.utah.edu.
(1) Arduengo, A. J., III; Harlow, R. L.; Kline, M. J. Am. Chem. Soc.
1991, 113, 361-363.
(2) Trnka, T. M.; Grubbs, R. H. Acc. Chem. Res. 2001, 34, 18-29.
(3) Herrmann, W. A. Angew. Chem., Int. Ed. 2002, 41, 1290-1309.
(4) Navarro, O.; Kelly, R. A., III; Nolan, S. P. J. Am. Chem. Soc. 2003,
125, 16194-16195.
(5) Jensen, D. R.; Schultz, M. J.; Mueller, J. A.; Sigman, M. S. Angew.
Chem., Int. Ed. 2003, 42, 3810-3813.
(6) Jensen, D. R.; Sigman, M. S. Org. Lett. 2003, 5, 63-65.
(7) Cornell, C. N.; Sigman, M. S. J. Am. Chem. Soc. 2005, 127, 2796-
2797.
(8) Schultz, M. J.; Hamilton, S. S.; Jensen, D. R.; Sigman, M. S. J. Org.
Chem. 2005, 70, 3343-3352.
(9) Sigman, M. S.; Jensen, D. R. Acc. Chem. Res. 2006, 39, 221-229.
(10) Tolman, C. A. Chem. ReV. 1977, 77, 313-348.
(11) Chianese, A. R.; Li, X.; Janzen, M. C.; Faller, J. W.; Crabtree, R. H.
Organometallics 2003, 22, 1663-1667.
(12) Dorta, R.; Stevens, E. D.; Scott, N. M.; Costabile, C.; Cavallo, L.;
Hoff, C. D.; Nolan, S. P. J. Am. Chem. Soc. 2005, 127, 2485-2495.
To describe the steric effects of NHCs on nickel com-
plexes, Nolan recently proposed to consider the fraction of
the metal sphere occupied by the ligand: %Vbur.12 Essentially,
the larger the %Vbur value, the more steric pressure exerted
(13) Altenhoff, G.; Goddard, R.; Lehmann, C. W.; Glorius, F. J. Am. Chem.
Soc. 2004, 126, 15195-15201.
8430 Inorganic Chemistry, Vol. 45, No. 20, 2006
10.1021/ic0612451 CCC: $33.50
© 2006 American Chemical Society
Published on Web 08/25/2006

Steric Effects in the Aerobic Oxidation of π-Allylnickel(II)
proposed to be a reversible reaction.17 In contrast, the reaction
of ItBu with Ni(cod)2 produces several fascinating products,
but not Ni(ItBu)2 (eq 2).18 In this case, %Vbur correlates with
the results: the larger ligand fails to form a stable NiL2
complex. The reaction of IiPr with Ni(CO)4 produces
saturated, tetrahedral Ni(CO)3(IiPr) (eq 3) whereas the
reaction of ItBu with Ni(CO)4 produces unsaturated, trigonal
Ni(CO)2(ItBu) (eq 4).12 This result is consistent with %Vbur
in that the larger ligand leads to the formation of a less
coordinated product. Since this report on %Vbur and NHC-
nickel complexes appeared, our lab has reported a discrep-
ancy between IiPr and ItBu in the attempted synthesis of
bis-µ-chloronickel(I) complexes. When a 1:1 mixture of Ni-
(cod)2 and Ni(dme)Cl2 is treated with 2 equiv of IiPr in
toluene, the desired bis-µ-chloronickel(I) complex is pro-
duced (eq 5). Treating the same mixture of nickel sources
with 2 equiv of ItBu results in the formation of a bis-µ-
ethoxonickel(II) complex, apparently due to an ethanol
impurity in the Ni(dme)Cl2 used (eq 6). It should be noted
that the same bottle of Ni(dme)Cl2 source was used for the
reactions in eqs 5 and 6 and that the use of ethanol-free Ni-
(dme)Cl2 prevented formation of the bis-µ-ethoxonickel(II)
complex shown in eq 6.19 In this case, the ligand with the
larger %Vbur value leads to the formation of a more-
coordinated product. This result is not consistent with %Vbur.
Figure 1. N-Heterocyclic Carbenes (NHCs) discussed in this manuscript.
Figure 2. (a) Depiction of %Vbur, where the darkened area depicts the
space filled about the nickel center by the ligand. (b) Values of %Vbur
reported by Nolan.12
by the ligand. An illustration of %Vbur and the reported values
for nickel are shown in Figure 2. Nolan has also proposed
the use of %Vbur for palladium14 and ruthenium15 complexes
with NHCs.
In the context of nickel-NHC systems, there are several
documented examples of discrepancies in reactivity between
IiPr and ItBu. For example, the respective reactions of IiPr
and ItBu with various nickel sources have drastically different
outcomes. The reaction of IiPr with Ni(cod)2 is reported to
form Ni(IiPr)2 cleanly (eq 1),16 although it has also been
To study this discrepancy in more detail, we harnessed
the O2 reactivity of π-allylchloro(NHC)nickel(II) complexes
previously reported from our lab.20 A number of additional
complexes were prepared in order to evaluate steric effects
(14) Viciu, M. S.; Navarro, O.; Germaneau, R. F.; Kelly, R. A., III; Sommer,
W.; Marion, N.; Stevens, E. D.; Cavallo, L.; Nolan, S. P. Organo-
metallics 2004, 23, 1629-1635.
(15) Hillier, A. C.; Sommer, W. J.; Yong, B. S.; Petersen, J. L.; Cavallo,
L.; Nolan, S. P. Organometallics 2003, 22, 4322-4326.
(16) Bo¨hm, V. P. W.; Gsto¨ttmayr, C. W. K.; Weskamp, T.; Herrmann, W.
A. Angew. Chem., Int. Ed. 2001, 40, 3387-3389.
(17) Louie, J.; Gibby, J. E.; Farnworth, M. V.; Tekavec, T. N. J. Am. Chem.
Soc. 2002, 124, 15188-15189.
(18) Caddick, S.; Cloke, F. G. N.; Hitchcock, P. B.; Lewis, A. K. d. K.
Angew. Chem., Int. Ed. 2004, 43, 5824-5827.
(19) Dible, B. R.; Sigman, M. S.; Arif, A. M. Inorg. Chem. 2005, 44, 3774-
3776.
(20) Dible, B. R.; Sigman, M. S. J. Am. Chem. Soc. 2003, 125, 872-873.
Inorganic Chemistry, Vol. 45, No. 20, 2006 8431

Dible and Sigman
Table 2. Preparation of π-Allylchloro(NHC)Nickel(II) Complexes from
on this reaction. Herein, we present the results of this study,
which strongly implicate sterics as the dominant factor
affecting the O2 reactivity of π-allylnickel complexes with
NHCs and possibly the dominant factor affecting other
nickel-NHC systems as well.
Ni(cod)2
NHC
IiPr
SIiPr
IMes
ItBu
lAd
IBioxMe4
ICy
solvent
yield (%)
Results and Discussion
la
lb
1c
ld
le
PhMe
PhMe
PhMe
PhMe
PhMe
thf
99a
99a
Synthesis of π-Allylchloro(NHC)nickel(II) Complexes.
π-Allylchloro(NHC)nickel(II) complexes were synthesized
via the one-pot procedure we reported previously for IiPr.20
In this reaction, Ni(cod)2 is oxidized by allyl chloride using
1,5-cyclooctadiene (cod) as the solvent. A solution of NHC
in toluene is added to the resulting π-allylnickel(II) inter-
mediate in a dropwise fashion to generate the desired
π-allylchloro(NHC)nickel(II) product. This reaction proceeds
in good to excellent yields, as shown in Table 2. The NHCs
were isolated prior to use following the procedure developed
by Arduengo,21 except in cases where the NHCs were not
amenable to isolation. In those cases, the NHC was generated
in situ and then added to the π-allylnickel(II) intermediate
and the yields were substantially lower. As noted by
Arduengo, the key to isolating NHCs is the avoidance of
water, where sensitive NHC’s seem to originate from more
hygroscopic imidazolium salts. It is noteworthy that drying
the precursor imidazolium salts via azeotropic distillation
with toluene using a Dean-Stark trap is especially beneficial
for IMes, ITol, and IMe.21
98a
>99a
98a
If
74b
lg
1h
1i
Ij
thf
51b
ITol
PhMe
PhMe
thf
63a
MeltBu
MeliPr
IMe
56b
47b
1k
thf
53b,c
a Isolated NHC used. b NHC generated in situ. c Isolated (C3H5NiCI)2
used.
Figure 3. Reactions of π-allylchloro(NHC)nickel(II) complexes 1a-1e
with O2.
Reactions of π-Allylchloro(NHC)nickel(II) Complexes
with O2. In an earlier communication, we reported on the
unusual reaction of π-allylchloro(IiPr)nickel(II) with O2 (eqs
7 and 8).12 In this reaction, a bis-µ-hydroxochloro(IiPr)nickel-
the λmax value of 500 nm previously reported for bis-µ-
hydroxonickel(II) complex 2a.20 Attempts to isolate the
putative bis-µ-hydroxonickel(II) products of the oxidation
of 1b or 1c were unsuccessful because of decomposition of
the products in solution. Even the previously reported,
crystallographically characterized bis-µ-hydroxonickel(II)
complex 2a decomposes in solution. For this reason, discus-
sion of the O2 stability of π-allylchloro(NHC)nickel(II)
complexes is based on consumption of the π-allylnickel
complexes.
We next examined the ItBu (1d) and IAd (1e) analogues
and were greatly surprised to find that they were stable to
O2, even when refluxed under an O2 atmosphere in d6-
1
benzene. No new species were observed by H NMR, nor
(II) dimer is produced and the allyl groups are oxidized and
liberated as R,â-unsaturated carbonyl compounds. IiPr
complex 1a was found to react rapidly with oxygen in
hexanes, benzene, toluene, and thf, with rates increasing in
that series from hexanes to thf. However, complex 1a is
kinetically stable to O2 in methylene chloride and 1,2-
dichloroethane, with reaction times greater than 36 h. It
should be noted that even the addition of small amounts of
CH2Cl2 to benzene or THF (1:10 mixtures) prohibits oxida-
tion, which may be due to competitive inhibition of O2
binding by chlorinated solvents. The analogous SIiPr (1b)
and IMes (1c) complexes behaved in a similar manner
(Figure 3), with formation of purple products with λmax(1b)
) 492 nm and λmax(1c) ) 497 nm. These are consistent with
was the formation of insoluble products observed. This
contrasts markedly with the aforementioned complexes 1a-
1c, which oxidize in a matter of seconds at room temperature.
This discrepancy was initially of unknown origin. The O2-
stable complexes of ItBu (1d) and IAd (1e) each have N,N′-
dialkyl substituents, whereas the reactive complexes of IiPr
(1a), SIiPr (1b), and IMes (1c) all have N,N′-diaryl substit-
uents. It would be tempting to call this discrepancy an
electronic effect, but the TEP value for IMes is 2050.7 and
the TEP values for N,N′-dialkyl substituted NHC ligands ICy,
tmiy, and biy range from 2049.6 to 2051. These values
strongly suggest that electronic effects are not the determin-
ing factor. Furthermore, one would expect that stronger
σ-donors, ligands with smaller TEP values, would be more
susceptible to oxidation if electronic effects were the sole
determining factor. However, to disprove electronic effects
(21) Arduengo, A. J., III; Dias, H. V. R.; Harlow, R. L.; Kline, M. J. Am.
Chem. Soc. 1992, 114, 5530-5534.
8432 Inorganic Chemistry, Vol. 45, No. 20, 2006

Steric Effects in the Aerobic Oxidation of π-Allylnickel(II)
as the determining factor, we would require either a reactive
non-aryl-substituted NHC or an inert aryl-substituted NHC.
The recent publications by Glorius on IBiox ligands13
inspired an attempt to prepare a complex with a constrained
ItBu analogue. An examination of the crystal structure of
IBioxMe4-HOTf suggested that IBioxMe4 would be ideal
for this study. IBioxMe4 complex 1f was prepared and
characterized crystallographically. As expected, IBioxMe4
is less hindering than ItBu. Furthermore, IBioxMe4 complex
1f was found to oxidize rapidly when treated with O2 in thf
and slowly in benzene. The TEP value for IBioxMe4 is 2054
cm-1, which is slightly larger than those measured for N,N′-
dialkyl NHC ligands. The observed instability of 1f to O2
suggested that the tipping point between O2 stability and
instability was close, which inspired the study of NHC
ligands with smaller alkyl substituents or with fewer aryl
substituents.
ICy complex (1g) and ITol complex (1h) were prepared
for study with O2. ICy complex 1g was completely consumed
within 5 s in oxygenated thf at 0 °C with concomitant
formation of a purple species with λmax ) 483 nm. This
successfully disproves that dialkyl-substituted NHCs always
form O2-stable complexes, which in turn refutes electronic
effects as the sole determining factor. ITol complex 1h was
consumed in ca. 30 min under these conditions. It should be
noted that IiPr complex 1a, SIiPr complex 1b, and IMes
complex 1c were completely consumed within 5 s under
these conditions.
When oxygenated benzene was used as the reaction solvent
at room temperature instead of thf at 0 °C, ICy complex 1g
was consumed within 5 min. Under these conditions, IiPr
complex 1a, SIiPr complex 1b, and IMes complex 1c were
still consumed within 5 s. In stark contrast, ITol complex
1h was not observed to form a bis-µ-hydroxonickel(II)
complex under these conditions, instead undergoing a side
reaction to produce trans-[ITol2NiCl2] (3h). Control experi-
ments revealed that this reaction does not require O2.
Furthermore, this reaction has not been observed for any
other complex presented herein.
Dynamic NMR Study of NHC Complexes. The com-
plexes evaluated above can be divided into three groups: O2
reactive (1a, 1b, and 1c), moderately O2 reactive (1g and
1h), and O2 stable (1d and 1e). Ligands for the O2-reactive
complexes have %Vbur values ranging from 26 to 30. Ligands
for the O2-stable complexes each have %Vbur values of 37.
ICy, the ligand for moderately reactive complex 1g, has a
%Vbur value of 23. There is no simple correlation between
O2 reactivity and %Vbur values among complexes 1a-1h.
However, a lack of correlation with %Vbur does not disprove
steric effects as the determining factor. %Vbur calculations
use DFT-optimized geometries of the free ligands and then
place the NHC ligands 2 Å from the metal center. The metal
is then considered to be a sphere with a 3 Å radius, and
%Vbur is defined as the percentage of that sphere occupied
by the ligand. The inherent assumption in this model is that
the ligand has a static conformation, but the metal-ligand
bond freely rotates. Otherwise, treating the non-NHC portion
of the complex as a sphere would be an invalid assumption.
Figure 4. Selected NHC complexes reported to have hindered metal-
NHC bond rotation.
If these conditions are not met, then the model should fail.
To consider this possibility, we examined complexes 1a-
1
1h by dynamic H NMR.
Crabtree reported and discussed several NHC complexes
of rhodium and iridium that exhibited hindered rotation about
the metal-carbene bond.11 To summarize, square-planar
iridium and rhodium complexes of NHCs exhibited fluxional
or even fixed metal-NHC bond rotation on the NMR time
scale at room temperature (Figure 4). In these cases, inhibited
rotations were largely observable because of AB or ABXY
splitting on the R-methylene protons on the N-substituents.
Unfortunately, many of the NHCs and complexes used in
catalysis lack spectroscopic handles to probe for inhibited
metal-NHC rotation. As a consequence, the ability/inability
of an NHC to rotate in a given complex seems to be given
little consideration in terms of implications for catalysis. In
our case, the planar chirality of η3-allyl complexes facilitates
such observations.
The protons located on C4 and C5 of imidazol-2-ylidenes
(backbone protons) are useful for purposes of examining
metal-NHC rotation. In complexes that have slow metal-
NHC rotation on the NMR time scale, these protons are
nonequivalent and resolve as a pair of doublets. In complexes
with freely rotating metal-NHC bonds, they are equivalent
and appear as a singlet. Furthermore, these protons essentially
account for the only signals observed between 5.5 and 7 ppm
in η3-allylnickel(II) complexes. We have examined Ni-NHC
rotations in η3-allylnickel(II) complexes by variable-tem-
perature NMR, and the results are quite striking (Figure 5).
IiPr complex 1a, SIiPr complex 1b, and IMes complex
1c all exhibit rapid rotation on the NMR time scale. ItBu
complex 1d and IAd complex 1e each exhibit no exchange
on the NMR time scale, even at 80 °C. ICy complex 1g and
ITol complex 1h each exhibit intermediate exchange on the
NMR time scale. Backbone proton coalescence was observed
at 45 °C for ICy complex 1g and at 35 °C for ITol complex
1h. From the coalescence data, it was possible to calculate
∆Ge‡ values of 15.7 kcal/mol for ICy complex 1g and 15.3
kcal/mol for ITol complex 1h. To summarize: O2-reactive
complexes exhibit rapid Ni-NHC bond rotation, moderately
O2-reactive complexes exhibit slower Ni-NHC bond rota-
tion, and O2-stable complexes do not exhibit observable Ni-
NHC bond rotation. It should be noted that the observed
relative rates of rotation do not correlate with %Vbur. ItBu
(%Vbur ) 37) complex 1d and IAd (%Vbur ) 37) complex
1e do not exhibit Ni-NHC bond rotation. IiPr (%Vbur ) 29)
complex 1a, SIiPr (%Vbur ) 30) complex 1b, and IMes
Inorganic Chemistry, Vol. 45, No. 20, 2006 8433

Dible and Sigman
1
Figure 5. Comparison of nickel-NHC bond rotations as observed by dynamic H NMR (300 MHz) with O2 stability.
(%Vbur ) 26) complex 1c each exhibit rapid Ni-NHC bond
rotation. On the other hand, ICy (%Vbur ) 23) complex 1g
exhibits intermediate Ni-NHC bond rotation, yet has a
smaller %Vbur value. While in review, a manuscript detailing
a study on dynamic NMR of π-allylnickel(II) complexes with
IME was published, see: Silva, L. C.; Gomes, P. T.; Veiros,
L. F.; Pascu, S. I.; Duarte, M. T.; Namorado, S.; Ascenso,
J. R.; Dias, A. R. Organometallics 2006, 25, 4391-4403.
It is worth noting that there is some evidence of inhibited
rotation in related π-allylchloro(NHC)palladium(II) com-
plexes. The NHC backbone protons in the ItBu complex are
reported as a pair of doublets in CDCl3 by Faller and
Sarantopoulos.22 However, the ItBu complex is reported to
have a single backbone peak in d6-benzene by Viciu, Nolan,
et al. The IAd complex is reported to have a doublet of
doublets for the backbone peak in d6-benzene in the same
report.14 It appears that metal-NHC bond rotation is less
sensitive to the nature of the NHC for palladium, possibly
because of the slightly longer palladium(II)-NHC bonds of
2.03-2.06 Å14 compared to 1.90-1.93 Å for nickel(II).
O2 reactivity (Tables 3 and 4). Notably, the steric bulk of
the t-butyl groups in ItBu is concentrated in the empty axial
sites above and below the square plane of complex 1d (Figure
6). On the other hand, the steric bulk of the isopropyl groups
in IiPr is more distributed in complex 1a. These structural
observations are consistent with the lack of observable Ni-
ItBu bond rotation in that the localized out of plane steric
bulk of the t-butyl groups would be expected to cause a
prohibitive activation barrier to rotation about that bond.
Incidentally, the blockage of the empty axial sites in ItBu
complex 1d may be the source of O2 stability. Prior
mechanistic studies on the reaction of 1a with O2 indicate
saturation kinetics with respect to 1a and O2.20 The saturation
behavior is consistent with the reversible formation of a
reactive intermediate from 1a and/or O2, followed by rate-
limiting decomposition of that intermediate. If the reactive
intermediate in this oxidation is a nonplanar O2 adduct, such
as a superoxonickel(III) complex, then an available axial site
would be required for oxidation to occur. Such intermediates
have been proposed for observed aerobic ligand oxidations
in nickel(II) complexes with dioxopentaazamacrocycles24-26
and with cyclotetrapeptides.27 If this is the case, then Ni-
Crystallographic Study. The crystal structures of 1a20
and 1d12,23 were compared in order to develop a hypothesis
as to why 1a oxidizes and 1d does not and to rationalize the
observed correlation between Ni-NHC bond rotation and
(24) Chen, D.; Motekaitis, R. J.; Martell, A. E. Inorg. Chem. 1991, 30,
1396-1402.
(25) Chen, D.; Martell, A. E. J. Am. Chem. Soc. 1990, 112, 9411-9412.
(26) Cheng, C.-C.; Gulia, J.; Rokita, S. E.; Burrows, C. J. J. Mol. Catal.
A: Chem. 1996, 113, 379-391.
(22) Faller, J. W.; Sarantopoulos, N. Organometallics 2004, 23, 2179-
2185.
(23) We independently crystallized complex 1d and present our data here.
(27) Haas, K.; Dialer, H.; Piotrowski, H.; Schapp, J.; Beck, W. Angew.
Chem., Int. Ed. 2002, 41, 1879-1881.
8434 Inorganic Chemistry, Vol. 45, No. 20, 2006

Steric Effects in the Aerobic Oxidation of π-Allylnickel(II)
Table 3. X-ray Diffraction Data Collection Parameters for Crystal Structures Discussed
la
ld
If
1l
empirical formula
fw
C30H41N2Ni
523.81
C14H25CIN2Ni
315.52
C17.5H25ClN2NiO2
389.55
CllH19IN2Ni
364.89
T (K)
150(1)
150(1)
150(1)
150(1)
wavelength (Å)
cryst syst
space group
a (Å)
b (Å)
c (Å)
0.71073
monoclinic
P21/c
12.4456(2)
14.6922(2)
16.5225(3)
103.7751(6)
2934.30(8)
4
0.71073
orthorhombic
Pna2l
10.9968(3)
9.6424(3)
14.3616(2)
90
1522.84(7)
4
1.376
0.71073
monoclinic
P21/c
12.5757(4)
10.2031(2)
15.6864(4)
113.6613(10)
1843.54(8)
4
0.71073
monoclinic
P21/c
8.9469(3)
11.6836(3)
13.5112(4)
107.2586(13)
1348.76(7)
4
â (deg)
V (A3)
Z
F (g/cm3)
µ (mm)
1.186
0.771
1.404
1.208
1.797
3.700
1.435
R1, wR2 (I > 2σ(I))
GOF
0.0338, 0.0792
1.018
0.0240, 0.0521
1.054
0.0398, 0.0971
1.047
0.0375, 0.0957
1.058
Table 4. Selected Distances (Å) and Angles (deg) for π-Allylnickel-NHC Complexes
la
ld
If
11
Ni-C(carbene)
1.9025(16)
2.1862(5)
1.981(2)
1.972(2)
2.056(2)
94.85(5)
98.21(8)
169.50(9)
72.68(10)
1.9290(18)
2.2319(7)
1.998(2)
1.982(2)
2.056(2)
1.902(2)
2.2108(7)
2.004(3)
2.000(4)
2.056(3)
1.909(4)
2.5241(6)
1.994(5)
2.005(6)
2.067(5)
95.31(11)
95.95(19)
168.86(19)
73.2(2)
Ni-CI (Ni-I for 1l)
Ni-C(allyl, cis to carbene)
Ni-C(allyl, internal)
Ni-C(allyl, trans to carbene)
C(carbene)-Ni-CI (-I for 1l)
C(carbene)-Ni-C(allyl, cis)
C(carbene)-Ni-C(allyl, trans)
C(allyl, cis)-Ni-C(allyl, trans)
97.91(6)
98.79(7)
95.81(10)
167.37(11)
72.76(12)
93.85(10)
165.34(13)
72.26(13)
NHC bond rotation is simply a mechanism to open an axial
site for O2 binding. Perhaps other conformational changes
could also permit O2 binding. Both ICy complex 1g and ITol
complex 1h exhibit slower Ni-NHC bond rotations than
complexes 1a-1c as well as reduced rates of oxidation. In
both ICy complex 1g and ITol complex 1h, the barrier to
rotation is between 15 and 16 kcal/mol. If Ni-NHC bond
rotation is rate-limiting, then these two complexes should
oxidize with similar rates. This is not the case. ICy complex
1g reacts with O2 faster than ITol complex 1h does in either
thf or benzene. If axial binding of O2 is operative, then this
discrepancy could stem from a requirement for both cyclo-
hexyl groups to adopt a conformation oriented away from
the square plane for rotation to occur, but with only one
cyclohexyl group having to rotate away for oxidation. ITol
complex 1g, on the other hand, exhibits rapid N-aryl rotation,
as observed by 1H NMR, which would rotate the aryl groups
through a conformation where the p-tolyl group is parallel
to the imidazol-2-ylidene. In this conformation, the axial site
on that face of the square plane would be blocked. It should
be noted that N-aryl rotation was not observed for any other
aryl-substituted ligand discussed herein (Figure 7). This
explanation is internally consistent, but would benefit greatly
from study of a complex that is incapable of nickel-NHC
bond rotation yet has an open axial site. For this reason, an
alternate ligand was devised.
necessary to verify that N-methyl substituents would not
interfere with O2 reactivity. For this reason, MeIiPr complex
1j and IMe complex 1k were prepared and evaluated. In each
case, rapid reaction with O2 was observed, although there
was no spectroscopic evidence for a bis(µ-hydroxo)nickel-
(II) product in the case of IMe complex 1k.
MeItBu complex 1i was synthesized and characterized.
The related iodo complex with MeItBu, complex 1l, was also
prepared and studied via single-crystal X-ray diffraction. In
each complex, Ni-NHC bond rotation was not observed by
NMR, even at 80 °C. Chloro complex 1i was found to react
with O2, albeit at slower rates than complexes 1a-1c. On
the other hand, iodo complex 1l exhibited stability to O2
comparable to that of ItBu complex 1d. This was an
unexpected development, but the substitution of iodine for
another halogen was found to interrupt O2 reactivity in a
palladium(I) system (eqs 9 and 10),28 so this phenomenon
is not unique to nickel-O2 chemistry.
It was envisioned that a nonsymmetrical NHC ligand with
a large group on one side and a small group on the other
could permit O2 binding on one face of the square plane
and block Ni-NHC rotation from steric interactions with
the large group on the other face. A methyl group was
considered to be the ideal small group and a t-butyl group
to be the simplest large group. However, it was first
Control Experiments. A notable alternative hypothesis
for the pathway by which allylic oxidation occurs is via
(28) Dura´-Vila´, V.; Mingos, D. M. P.; Vilar, R.; White, A. J. P.; Williams,
D. J. Chem. Commun. 2000, 1525-1526.
Inorganic Chemistry, Vol. 45, No. 20, 2006 8435

Dible and Sigman
Figure 6. ORTEPs of (a) IiPr complex 1a, (b) ItBu complex 1d, (c) IiPr complex 1a viewed down the Ni-NHC bond, (d) ItBu complex 1d viewed down
the Ni-NHC bond, (e) IBioxMe4 complex 1f, and (f) MeItBu complex 1i.
autocatalysis. In this case, the difference between O2-stable
and O2-reactive complexes could simply arise from the ability
of a given complex to form a competent oxidation catalyst.
This hypothesis was tested by treating a mixed solution of
IiPr complex 1a and ItBu complex 1d in d6-benzene with
O2. This resulted in the rapid, complete conversion of IiPr
complex 1a into bis-µ-hydroxonickel(II) complex 2a and
acrolein with no change to ItBu complex 1d (eq 11). This
result does not disprove the possibility of such an autocata-
lytic mechanism, but it emphasizes that ItBu complex 1d is
essentially inert to aerobic oxidation, regardless of whether
the pathway is intermolecular or intramolecular, autocatalytic
or not.
To probe the possibility of trace impurities promoting the
reaction, we undertook a series of experiments with MeItBu
complex 1i. Complex 1i was selected for study because it is
the most O2-resistant complex for which the formation of a
bis-µ-hydroxonickel(II) species has been confirmed via
visible spectroscopy. Samples of 1i in thf were exposed to
O2 in the presence of the following additives: potassium
tert-butoxide, Ni(cod)2, and bis-η3-allylnickel chloride. Each
of these three compounds is used at some stage within the
synthesis of the complexes. The samples of 1i were found
8436 Inorganic Chemistry, Vol. 45, No. 20, 2006

Steric Effects in the Aerobic Oxidation of π-Allylnickel(II)
Figure 7. Putative nickel-NHC conformation effects on O2 reactivity.
to decompose at the same rate as the additive-free control
sample. A final control experiment was performed to
determine the role of thf quality in O2 reactivity. A sample
of dry, distilled thf was sparged with O2 for 15 min and
subsequently heated to an incipient boil under an O2
atmosphere. After cooling, a sample of 1i in thf was added
to the preoxygenated thf. Once again, no deviation from the
control experiment was observed, casting doubt on the
possiblity that thf-peroxide formation is operative in the
enhanced reactivity in thf.
The ability or inability to adopt a nonplanar geometry is
almost certainly a relevant factor in nickel-NHC catalyzed
processes. The imidazol-2-ylidene ligands that block axial
space above and below the square plane exert the least steric
pressure within the square plane. A recent report on Ni-
NHC catalyzed cyclizations of cyclopropen-ynes by Zuo and
Louie29 is consistent with this quandary. When SIiPr is used
as the ligand, the distribution of products is dependent on
the size of the alkyne substituent (eq 12). When ItBu is used,
the only product observed is the product preferred for small
alkyne substituents in the SIiPr system (eq 13). This could
be attributed to the less-crowded environment within the
square-plane induced by ItBu, if one considers that ItBu is
a saddle-shaped ligand. This would allow for more facile
â-hydride elimination.
Structural Implications
The fact that the more geometrically constrained com-
plexes are not capable of reaction with O2 in the same
manifold suggests that reaction with O2 requires a substantial
geometric change about the metal. It is possible that reaction
with O2 requires the adoption of tetrahedral or a five-
coordinate geometry to proceed, which would be consistent
with the observed correlation between geometric constraint
and inertness to O2 (Figure 8). Formation of a tetrahedral
O2 adduct of IiPr complex 1a, SIiPr complex 1b, or IMes
complex 1c is consistent with observed tetrahedral Ni(CO)3-
(NHC) complexes of IiPr, SIiPr, and IMes.12 However, ICy
also forms a tetrahedral Ni(CO)3(NHC) complex, yet ICy
complex 1g is less reactive with O2. Facile metal-NHC bond
rotation also does not appear to be required in cases where
other mechanisms for opening space about the metal exist,
such as a conformational change in a cyclohexyl group in
ICy complex 1g or the presence of a more open face in
MeItBu complex 1i. With these observations in mind, the
most likely O2 adduct is square pyramidal, although a
bipyrimidal geometry is also feasible.
Conclusions
A series of π-allylchloro(NHC)nickel(II) complexes with
varying NHCs have been prepared. These complexes are
(29) Zuo, G.; Louie, J. J. Am. Chem. Soc. 2005, 127, 5798-5799.
Inorganic Chemistry, Vol. 45, No. 20, 2006 8437

Dible and Sigman
Figure 8. Newman projections of nonplanar O2 adducts.
from sodium/benzophenone and subsequently degassed under a
vacuum. Deuterated benzene and methylene chloride were dried
using activated alumina and degassed via freeze-pump-thaw prior
to use. Deuterated thf was purchased from Cambridge Isotope
Laboratories and used as received. Ni(cod)2 was purchased from
Strem and used as received. Allyl chloride was purified by
distillation from CaCl2 and degassed via freeze-pump-thaw. Allyl-
(IiPr)NiCl (1a),20 SIiPr, IMes,21 ItBu, IAd, ICy, ITol,21 IMe,21 and
IBioxMe4-HOTf 13 were prepared as described in the literature.
NMR spectra were collected on a Varian VXL-300 or on a Varian
Unity-300. IR spectra were collected with a Mattson Satellite FTIR.
Visible spectra were collected with an SI Photonics 440 series
spectrophotometer equipped with a fiber-optic dip probe. Crystal
structures were solved by Dr. Atta M. Arif. Mass Spectra were
collected with a Finnigan MAT95 mass spectrometer by Dr. Elliot
M. Rachlin. Elemental analyses were performed by Desert Ana-
lytics, Tucson, AZ.
found to have a strong ligand dependence in terms of their
ability to react with O2. The ligand dependence does not
correlate with Tolman electronic parameters;12,13 therefore,
electronic effects are not the determining factor. The ligand
dependence also does not correlate with %Vbur.12 However,
it is notable that complexes resulting from ligands capable
of blocking one or both of the empty axial sites above and
below the square plane are inert to O2. Likewise, the ligand
dependence strongly correlates with conformational freedom
about the nickel-NHC bond: complexes with free rotation
react with O2, complexes with restricted rotation do not react
with O2, and complexes with intermediate conformational
freedom exhibit intermediate reactivity. This correlation is
consistent with a mechanism requiring the adoption of a
nonplanar geometry about the metal as a requirement for
reaction with O2. Furthermore, it has been demonstrated that
an O2-reactive complex will not promote the reaction of an
inert complex with O2.
Conformational restriction about metal-NHC bonds is
relevant to catalysis and should be considered in terms of
catalyst design and in the rationalization of observed patterns
of reactivity. As in the case of cyclopropen-yne cyclization,29
a conformationally restricted ligand may be advantageous.
A monodentate ligand with a lesser volume within the square
plane seems more likely to promote â-hydride elimination.
Whether such a property is advantageous or problematic
depends on the intended outcome, which is ample reason to
consider the desired properties of an NHC-based catalyst
carefully.
Allyl(SIiPr)NiCl (1b). Ni(cod)2 (102 mg, 0.37 mmol) was
suspended in cod (ca. 400 mg) in a scintillation vial. Allyl chloride
(28 mg, 0.37 mmol) was then added dropwise to this suspension
while the vial was swirled. A solution of SIiPr (145 mg, 0.37 mmol)
in toluene (2-mL) was then added. Residual SIiPr was then
transferred with toluene rinses (2 × 1 mL). The vial was swirled
by hand until the mixture became homogeneous (ca. 3 min). The
reaction mixture was then concentrated to dryness under a vacuum
to yield the product as an orange solid (194 mg, 99% yield).
1H NMR (300 MHz, C6D6): δ 0.82 (d, J ) 13 Hz, 1H, allyl-H),
1.14 (t, J ) 7.2 Hz, 12H, CHMe2), 1.48 (d, J ) 6.4 Hz, 6H,
CHMe2), 1.54 (d, J ) 6.1 Hz, 6H, CHMe2), 2.32 (d, J ) 6.1 Hz,
1H, allyl-H), 2.45 (d, J ) 14 Hz, 1H, allyl-H), 3.35 (d, J ) 6.3
Hz, 1H, allyl-H), 3.54-3.44 (br s, 6H, overlapping RNCH2CH2-
NR and CHMe2), 3.76 (m, 2H, CHMe2), 4.45 (sept, J ) 6.8 Hz,
1H, allyl-H), 7.1-7.2 (m, 6H, Ar-H). 13C NMR {1H} (75 MHz,
C6D6): δ 24.04, 24.30, 27.13, 27.215, 29.10, 44.25, 54.21, 71.72,
108.40, 124.94, 129.81, 137.24, 218.41. IR (KBr): 3070(w), 3056-
(w), 2958(s), 2868(s), 1475(m), 1446(s), 1419(s), 1266(s), 1239-
(m), 1054(m), 903(m), 801(m), 756(m), 424(w), 414(w). HRMS
(EI) m/Z: (M)+ calcd, 524.2490; obsd, 524.2484.
Experimental Section
General. Unless otherwise noted, all manipulations were carried
out under an inert atmosphere using Schlenk techniques or in a
VAC M0-40 glovebox. Glassware was oven- or flame-dried prior
to use. Toluene, hexanes, and benzene were purified via passage
through activated alumina columns and subsequently degassed
under a vacuum. Tetrahydrofuran (thf) was purified by distillation
Allyl(IMes)NiCl (1c). Ni(cod)2 (177 mg, 0.64 mmol) was
suspended in cod (ca. 900 mg) in a scintillation vial. Allyl chloride
8438 Inorganic Chemistry, Vol. 45, No. 20, 2006

Steric Effects in the Aerobic Oxidation of π-Allylnickel(II)
(49 mg, 0.64 mmol) was then added dropwise to this suspension
while the vial was swirled. A solution of IMes (196 mg, 0.46 mmol)
in toluene (2 mL) was then added. The residual IMes was
transferred with toluene rinses (2 × 1 mL). The vial was swirled
by hand until the mixture became homogeneous (ca. 3 min). The
reaction mixture was then concentrated to dryness under a vacuum
to yield the product as an orange solid (277 mg, 98% yield).
1H NMR (300 MHz, C6D6): δ 1.18 (d, 1H, J ) 13 Hz, allyl-H),
2.09 (s, 7H, Ar-Me), 2.15 (s, 6H, Ar-Me), 2.32 (s, 6H, Ar-Me),
2.46 (d, J ) 14 Hz, 1H, allyl-H), 3.31 (d, J ) 6.8 Hz, 1H, allyl-
H), 4.63 (sept, J ) 6.8 Hz, 1H, allyl-H), 6.16 (s, 2H, RNCHCHNR),
6.76 (s, 1H, Ar-H), 6.83 (s, 1H, Ar-H). 13C NMR {1H} (75 MHz,
C6D6): δ 18.55, 18.66, 21.06, 43.52, 69.64, 107.42, 122.88, 129.33,
129.38, 135.93, 135.96, 136.53, 138.78, 186.19. IR (KBr): 3166-
(m), 3140(m), 3075(w), 2916(s), 2855(m), 1487(s), 1398(s), 1330-
(s), 1268(m), 1230(m), 1076(w), 1038(m), 927(m), 900(m), 851(s),
741(m), 705(m). HRMS (EI) m/Z: (M)+ calcd, 438.1362; obsd,
438.1367.
2922(m), 1748(s), 1458(s), 1411(s), 1335(s), 1255(m), 1208(s), 996-
(s), 935(s), 849(m), 668(w), 494(w) cm-1. HRMS (EI) m/Z: (M)+
calcd, 342.0655; obsd, 342.0650.
Allyl(ICy)NiCl (1g). ICy-HBF4 (106 mg, 0.33 mmol), KOtBu
(39 mg, 0.35 mmol), and thf (5-mL) were added to a scintillation
vial, and the resulting mixture was stirred for 1h. The mixture was
diluted with toluene (2 mL) and concentrated under a vacuum to
ca. 2 mL. The reaction mixture was then diluted with toluene (1
mL) and hexanes (7 mL), mixed by pipet for ca. 2 min, and filtered
over Celite. The solids were washed with hexanes (2 × 1 mL).
The filtrate was combined and concentrated to ca. 1 mL under a
vacuum. Ni(cod)2 (91 mg, 0.33 mmol) and cod (∼400 mg) were
added to a 20 mL scintillation vial. Allyl chloride (25 mg, 0.33
mmol) was then added dropwise to the resulting suspension of Ni-
(cod)2, resulting in the appearance of a red color. After the vial
was swirled until all of the solid Ni(cod)2 disappeared (∼5 min),
the solution of ICy in toluene was added dropwise to the nickel/
cod/allyl chloride mixture; residual ICy was rinsed with toluene (2
× 1 mL). This resulted in the formation of a brown heterogeneous
mixture, which was filtered over Celite to remove a green solid.
The filtrate was diluted with hexanes (9 mL) and cooled to -25
°C overnight. An orange-brown precipitate formed and was isolated
by pipeting off the solvent and drying the solid under a vacuum to
yield the product (62 mg, 51%yield).
Allyl(ItBu)NiCl (1d). Ni(cod)2 (125 mg, 0.46 mmol) was
suspended in cod (ca. 750 mg) in a scintillation vial. Allyl chloride
(35 mg, 0.46 mmol) was then added dropwise to this suspension
while the vial was swirled. A solution of ItBu (82 mg, 0.46 mmol)
in toluene (2 mL) was added. The ItBu was then rinsed with toluene
(2 × 1 mL). The vial was swirled by hand until the mixture became
homogeneous (ca. 3 min). The reaction mixture was then concen-
trated to dryness under a vacuum to yield the product as an orange
solid (144 mg, >99% yield). The product has 1H NMR data
identical to the literature values.12 Crystals suitable for X-ray
analysis were obtained by cooling a solution of 1d in 1:1 toluene:
hexanes to -25 °C.
Allyl(IAd)NiCl (1e). Ni(cod)2 (138 mg, 0.50 mmol) was
suspended in cod (ca. 400 mg) in a scintillation vial. Allyl chloride
(38 mg, 0.50 mmol) was then added dropwise to this suspension
while the vial was swirled. A solution of IAd (168 mg, 0.50 mmol)
in toluene (2 mL) was then added. The IAd was rinsed with toluene
(2 × 1 mL). The vial was swirled by hand until the mixture became
homogeneous (ca. 5 min). The reaction mixture was then concen-
trated to dryness under a vacuum to yield the product as an orange
solid (233 mg, 98% yield). The product has 1H NMR data identical
to the literature values.12
1H NMR (300 MHz, C6D6): δ 0.8-1.0 (m, 2H, cy-H), 1.0-2.0
(m, 17H, cy-H), 1.87 (m, 1H, cy-H), 2.30 (m, 2H, cy-H), 2.61 (d,
J ) 10 Hz, 1H, allyl-H), 3.01 (d, J ) 14 Hz, 1H, allyl-H), 3.82 (d,
J ) 7.6 Hz, 1H, allyl-H), 4.76 (br s, 1H, NCHR2), 5.18 (sept, J )
6.7 Hz, 1H, allyl-H), 5.56 (br s, 1H, NCHR2), 6.34 (s, 1H,
RNCHCHNR), 6.41 (s, 1H, RNCHCHNR). 13C NMR {1H} (75
MHz, C6D6): δ 25.50, 25.55, 25.59, 25.67, 25.91, 34.08, 34.24,
34.43, 42.69, 59.80, 60.39, 68.50, 107.24, 117.51, 180.84. IR
(KBr): 2931(s), 2852(s), 1450(m), 1422(m), 1380(m), 1238(m),
1201(s), 895(w), 866(w), 724(s), 699(m), 424(m), 409(w) cm-1
HRMS (EI) m/Z: (M)+ calcd, 366.1403; obsd, 366.1388.
.
Allyl(ITol)NiCl (1h). Ni(cod)2 (171 mg, 0.62 mmol) was
suspended in cod (ca. 1 g) in a scintillation vial. Allyl chloride (48
mg, 0.62 mmol) was then added dropwise to this suspension while
the vial was swirled. A mixture of ITol (154 mg, 0.62 mmol) in
toluene (3 mL) was then added. The ITol was then rinsed over
with toluene (4 × 1 mL). The mixture was swirled for ca. 5 min
and then concentrated under a vacuum. The product was then
crystallized from toluene/hexanes at -25 °C to yield an orange
solid (149 mg, 63% yield).
1H NMR (300 MHz, C6D6): δ 1.10 (d, J ) 12 Hz, 1H, allyl-H),
1.89 (dt, J ) 6.8 Hz, 2.4 Hz, 1H, allyl-H), 2.05 (s, 6H, Ar-Me),
2.80 (d, J ) 14 Hz, 1H, allyl-H), 3.70 (dd, J ) 2.4 Hz, 7.6 Hz,
1H, allyl-H), 4.76 (sept, J ) 6.8 Hz, 1H, allyl-H), 6.54 (br s, 1H,
RNCHCHNR), 6.61 (br s, 1H, RNCHCHNR), 6.96 (d, J ) 7.8
Hz, 2H, Ar-H), 7.02 (d, J ) 7.8 Hz, 2H, Ar-H), 8.01 (d, J ) 7.8
Hz, 2H, Ar-H), 8.44 (d, J ) 7.6 Hz, 2H, Ar-H). 13C NMR {1H}
(75 MHz, C6D6): δ 20.89, 43.03, 68.23, 107.49, 121.85, 122.14,
124.51, 124.85, 127.81, 127.87, 129.98, 130.03, 137.79, 138.06,
185.22. IR (KBr): 3173(w), 3140(m), 3110(w), 3054(m), 3033-
(m), 2918(m), 1514(s), 1409(s), 1338(m), 1273(s), 942(m), 888-
(m), 821(s), 717(m), 682(m) cm-1; Anal. Calcd for C20H21ClN2-
Ni: C, 62.63; H, 5.52; N, 7.30. Found: C, 62.43; H, 5.69; N, 7.37.
MeItBu-HI. 1-t-Butyl-1H-imidazole30 (544 mg, 4.4 mmol) was
added to a 25 mL round-bottomed flask equipped with a stirbar
and dissolved in thf (5 mL). Iodomethane (301µL, 4.8 mmol) was
added to this solution, and the reaction mixture was stirred
Allyl(IBioxMe4)NiCl (1f). IBioxMe4-HOTf (112 mg, 0.33
mmol) and KOtBu (39 mg, 0.34 mmol) were dissolved in thf (3
mL) in a scintillation vial and stirred for 2 h. Benzene (2 mL) was
then added to the resulting solution to precipitate KOTf. Ni(cod)2
(90 mg, 0.33 mmol) was suspended in 1,5-cod (ca. 400 mg) in a
second scintillation vial. Allyl chloride (26 mg, 0.34 mmol) was
added to the suspension of Ni(cod)2, resulting in a color change
from yellow to red. The thf/benzene mixture containing IBioxMe4
was then pipet-filtered over Celite directly into the Ni/allyl chloride/
1,5-cod mixture, followed by 2 × 1 mL rinses with benzene. The
resulting orange solution was concentrated to dryness under a
vacuum to yield a brown solid. The complex was then dissolved
in thf (3 mL), filtered over Celite into a scintillation vial, layered
with hexane (10 mL), and cooled to -25 °C to yield brown prisms
(83 mg, 74%). Crystals suitable for X-ray analysis were obtained
by cooling a solution of 1f in 1:2 toluene:hexanes to -25 °C.
1H NMR (300 MHz, d8-thf): δ 1.58-1.72 (m, 7H, overlapping
allyl-H and CR3Me), 1.87 (s, 3H, CR3Me), 2.00 (s, 3H, CR3Me),
2.39 (d, J ) 6.3 Hz, 1H, allyl-H), 2.66 (d, J ) 14 Hz, 1H, allyl-
H), 3.30 (d, J ) 6.6 Hz, 1H, allyl-H), 4.51 (dt, J ) 24 Hz, 8.2 Hz,
4H, ROCH2CR2N), 5.24 (sept, J ) 6.6 Hz, 1H, allyl-H). 13C NMR
{1H} (75 MHz, d8-thf): δ 25.85, 26.61, 26.67, 44.80, 61.57, 62.06,
68.38, 88.43, 88.57, 108.01, 126.94, 159.37. IR (KBr): 2964(m),
(30) Liu, J.; Chen, J.; Zhao, J.; Zhao, Y.; Li, L.; Zhang, H. Synthesis 2003,
2661-2666.
Inorganic Chemistry, Vol. 45, No. 20, 2006 8439

Dible and Sigman
magnetically for 2 h. The reaction mixture was then diluted with
ether (10 mL). The mixture was filtered over a medium frit. The
filtrate was then purified by recrystallization from acetonitrile/ether
by vapor diffusion to yield the product as white needles (566 mg,
49%).
1H NMR (300 MHz, d6-DMSO): δ 1.58 (s, 9H, CMe3), 3.84 (s,
3H, N-CH3), 7.76 (d, J ) 1.6 Hz, 1H, RNCHCHNR), 7.99 (d, J
) 1.8 Hz, 1H, RNCHCHNR), 9.24 (s, 1H, RNCHNR). 13C NMR
{1H} (75 MHz, d6-DMSO): δ 29.01, 35.74, 59.29, 120.05, 123.74,
135.00; MS (EI) m/Z: (M)+ calcd, 139; obsd, 139.
Allyl(MeItBu)NiCl (1i). MeItBu-HI (88 mg, 0.33 mmol),
potassium t-butoxide (39 mg, 0.33 mmol), and thf (4 mL) were
added to a scintillation vial. The resulting suspension was stirred
for 3 h and then diluted with toluene (4 mL). The volume of the
mixture was reduced to ca. 3 mL under a vacuum. Hexanes (8 mL)
were added to the reaction mixture, which was then filtered over
Celite. The solids were washed with hexanes (3 × 1 mL), and the
combined filtrate was concentrated to ca. 1 mL under a vacuum.
Ni(cod)2 (91 mg, 0.33 mmol) was suspended in 1,5-cod (ca. 400
mg) in a second 20 mL scintillation vial. Allyl chloride (25 mg,
0.33 mmol) was added to the suspension of Ni(cod)2, resulting in
a color change from yellow to red. The solution of MeItBu in
toluene was then added dropwise to the Ni/allyl chloride/cod
mixture with rapid mixing. The residual MeItBu was transferred
with toluene rinses (2 × 1 mL). After ca. 5 min, the reaction mixture
was filtered over Celite and concentrated to 2 mL under a vacuum.
The solution was then diluted with hexanes (8 mL) and cooled to
-25 °C overnight to yield the product as an orange solid (51 mg,
56% yield). Note: this compound exists as a 3:2 mixture of
atropisomers, as determined by NMR.
Hz, 2H, RCHMe2), 4.36 (s, 3H, N-CH3), 7.22 (t, J ) 1.8 Hz, 1H,
RNCHCHNR), 7.36 (d, J ) 7.8 Hz, 2H, Ar-H), 7.59 (t, J ) 7.8
Hz, 1H, Ar-H), 7.66 (br s, 1H, RNCHCHNR), 9.99 (s, 1H,
RNCHNR). 13C NMR {1H} (75 MHz, CD2Cl2): δ 24.55, 24.66,
29.09, 38.22, 124.80, 125.01, 125.17, 130.41, 132.40, 138.15,
145.94; MS (EI) m/Z: (M)+ calcd, 243; obsd, 243.
Allyl(MeIiPr)NiCl (1j). MeIiPr-HI (122 mg, 0.33 mmol),
potassium t-butoxide (39 mg, 0.33 mmol), and thf (4 mL) were
added to a scintillation vial. The resulting suspension was stirred
for 3 h and then diluted with toluene (4 mL). The volume of the
mixture was reduced to ca. 3 mL under a vacuum. Hexanes (8 mL)
were added to the reaction mixture, which was then filtered over
Celite. The solids were washed with hexanes (3 × 1 mL), and the
combined filtrate was concentrated to ca. 1 mL under a vacuum.
Ni(cod)2 (90 mg, 0.33 mmol) was suspended in 1,5-cod (ca. 400
mg) in a second scintillation vial. Allyl chloride (26 mg, 0.34 mmol)
was added to the suspension of Ni(cod)2, resulting in a color change
from yellow to red. The solution of MeIiPr in toluene was then
added to the Ni/allyl chloride/cod mixture by pipet with rapid
mixing. Residual MeIiPr was transferred by pipet using additional
toluene (2 × 1 mL). The reaction mixture became brown with the
formation of a green precipitate. The mixture was filtered over
Celite, and the precipitate was washed with hexanes (2 × 1 mL).
The filtrate was concentrated to ca. 1 mL under a vacuum, diluted
with hexanes (8 mL), and cooled to -25 °C overnight to precipitate
the product as an orange solid (58 mg, 47% yield).
1H NMR (300 MHz, C6D6): δ 0.92 (d, J ) 6.8 Hz, 3H, CHMe2),
0.98 (d, J ) 6.8 Hz, 3H, CHMe2), 1.11 (d, J ) 13 Hz, 1H, allyl-
H), 1.25 (d, J ) 6.8 Hz, 3H, CHMe2), 1.48 (d, J ) 6.3 Hz, 3H,
CHMe2), 2.16 (d, J ) 6.1 Hz, 1H, allyl-H), 2.65-2.70 (m, 2H,
superimposed allyl-H and ArCHMe2), 3.25 (br s, 1H, ArCHMe2),
3.65 (d, J ) 6.1 Hz, 1H, allyl-H), 3.75 (s, 3H, NMe), 4.77 (sept, J
) 6.7 Hz, 1H, allyl-H), 6.10 (d, J ) 4.6 Hz, 1H, RNCHCHNR),
6.33 (s, 1H, RNCHCHNR), 7.05 (d, J ) 6.8 Hz, 1H, Ar-H), 7.15-
7.22 (m, 2H, Ar-H). 13C NMR {1H} (75 MHz, C6D6): δ 23.23,
23.69, 26.71, 26.80, 28.65, 28.84, 38.37, 43.50, 69.75, 108.29,
122.32, 124.25, 124.58, 130.48, 136.66, 146.87, 147.48, 185.39.
IR (KBr): 3160(m), 3126(m), 3097(m), 2964(s), 2864(w), 1461-
(s), 1399(s), 1387(m), 1360(m), 1285(m), 1218(m), 806(m), 766-
(m), 734(m), 694(w), 470(w), 437(w), 423(w) cm-1. HRMS (EI)
m/Z: (M)+ calcd, 376.1214; obsd, 376.1215.
1H NMR (300 MHz, CD2Cl2): δ 1.45 (br d, J ) 12 Hz, 0.4H,
allyl-H), 1.77 (s, 4.2H, superimposed CMe3 and allyl-H), 2.05 (s,
3.6H, CMe3), 2.36 (dt, J ) 6.7 Hz, 2.5 Hz, 0.6H, allyl-H), 2.49
(dt, J ) 6.8 Hz, 2.6 Hz, 0.4H, allyl-H), 2.69 (d, J ) 14 Hz, 0.6H,
allyl-H), 2.71 (d, J ) 14 Hz, 0.4H, allyl-H), 3.33 (d, J ) 2.6 Hz,
0.4H, allyl-H), 3.36 (d, J ) 2.3 Hz, 0.6H, allyl-H), 3.95 (s, 1.2H,
NMe), 4.40 (s, 1.8H, NMe), 5.16 (sept, J ) 6.8 Hz, 0.6H, allyl-H),
5.48 (sept, J ) 6.8 Hz, 0.4H, allyl-H), 6.88 (d, J ) 2.2 Hz, 0.4H,
RNCHCHNR), 6.98 (d, J ) 1.9 Hz, 0.6H, RNCHCHNR), 7.04 (d,
J ) 1.9 Hz, 0.6H, RNCHCHNR), 7.10 (d, J ) 2.1 Hz, 0.4H,
RNCHCHNR). 13C NMR {1H} (75 MHz, C6D6): δ 31.28, 31.54,
37.58, 38.27, 42.82, 42.94, 57.28, 57.67, 66.57, 66.67, 105.90,
107.05, 118.42, 118.66, 121.52, 121.63. IR (KBr): 3161(m), 3107-
(s), 3053(w), 2978(s), 2937(s), 2872(w), 1610(m), 1574(m), 1468-
(s), 1442(s), 1399(s), 1383(s), 1372(s), 1310(m), 1240(s), 1213(s),
Allyl(IMe)NiCl (1k). IMe-HI (123 mg, 0.55 mmol) and KOtBu
(62 mg, 0.55 mmol) were added to a 20 mL scintillation vial and
suspended in thf (5 mL). The resulting mixture was stirred
magnetically for 4 h. The mixture was then filtered over Celite
into a second vial, rinsed with thf (2 × 1 mL), and cooled to -25
°C in the glovebox refrigerator. Bis-allylnickel chloride (68 mg,
0.25 mmol) was added to a 20 mL scintillation vial and dissolved
in thf (2 mL). The resulting red solution was cooled to -25 °C in
the glovebox refrigerator. When both solutions were cold, the
solution of IMe was added to the solution of bis-allylnickel chloride
at a rate of about 1 drop per 2 s with rapid swirling of the bis-
allylnickel chloride solution. A green solid rapidly precipitated from
solution. The reaction mixture was mixed by hand for ca. 5 min
after the addition of IMe/thf was completed. The green solid was
filtered off over Celite and the filtrate was concentrated under a
vacuum, resulting in the precipitation of additional green solid. The
green solid was extracted with toluene (3 × 3 mL) and the toluene
extracts were filtered over Celite. The toluene filtrate was then
concentrated to ca. 2 mL, diluted with hexanes (8 mL), and cooled
to -25 °C to precipitate a mixture of green solid and orange solid.
The solid was isolated by filtration over Celite and then washed
1082(m), 921(w), 870(m), 758(s), 713(w), 688(m), 415(w) cm-1
MS (EI) m/Z: (M)+ calcd, 272.1; obsd, 272.1.
.
MeIiPr-HI. 1-(2,6-Diisopropylphenyl)-1H-imidazole30,31 (1.14
g, 5 mmol) was added to a 20 mL scintillation vial equipped with
a stirbar and dissolved in thf (5 mL). Iodomethane (342µL, 5.5
mmol) was added to this solution, and the reaction mixture was
allowed to stir for 15 h. The reaction mixture was then diluted with
ether (10 mL). The crude product partitioned from the ether phase
as a viscous oil. The ether layer was removed via decantation. The
viscous oil was then dissolved in methylene chloride (4 mL). The
addition of ether (12 mL) to this solution resulted in the precipitation
of a yellow solid. The solid was isolated via filtration with a medium
frit and purified via recrystallization from acetonitrile/ether by vapor
diffusion to obtain the product as clear blocks (1.11 g, 60%).
1H NMR (300 MHz, CD2Cl2): δ 1.16 (d, J ) 6.8 Hz, 6H,
CHMe2), 1.25 (d, J ) 6.8 Hz, 6H, CHMe2), 2.34 (sept, J ) 6.8
(31) Perry, M. C.; Cui, X.; Powell, M. T.; Hou, D.-R.; Reibenspies, J. H.;
Burgess, K. J. Am. Chem. Soc. 2003, 125, 113-123.
8440 Inorganic Chemistry, Vol. 45, No. 20, 2006

Steric Effects in the Aerobic Oxidation of π-Allylnickel(II)
with toluene (3 × 2 mL). The toluene filtrate was concentrated
under a vacuum to yield the product as an orange solid (61 mg,
53% yield).
IR (KBr): 3141(w), 3133(w), 3048(w), 2919(w), 1515(s), 1475-
(s), 1339(m), 1279(s), 1110(w), 1089(w), 1023(w), 945(m), 841-
(m), 828(m), 819(s), 730(s), 702(s), 563(m), 529(s) cm-1. Anal.
Calcd for C34H32Cl2N4Ni: C, 65.21; H, 5.15; N, 8.95. Found: C,
64.85; H, 4.86; N, 8.58.
Dynamic NMR. All VT-NMR experiments were performed on
a Varian VXR-300 instrument. Values of ∆Ge‡ were determined
at coalescence for complexes 1g and 1h in both benzene and thf.
The ∆Ge‡ value for complex 1f in benzene was determined below
the coalescence point at temperatures between 55 and 80 °C.
Insufficient line-broadening was observed below 62 °C for complex
1f in thf.
Oxidation Studies. C6D6: An NMR sample of each complex
in d6-benzene in a 5 mm thin-walled NMR tube was placed under
a gentle stream of O2 that was dried via passage through a drying
tube. The O2-reactive complexes (1b, 1c, 1f, 1j, 1k) rapidly changed
in color with complete consumption of the starting material. The
O2-inert (1d, 1e, 1l) complexes exhibited no change, as verified
by 1H NMR, even when heated to 80 °C under O2. The intermediate
complexes (1f, 1g, 1h, 1i) gradually disappeared.
THF: A sample of each complex 1a-1l (1-2.5 mg) was placed
in a separate 1 dram vial. Each sample was dissolved in thf (ca. 2
mL). The resulting solutions were sparged with O2, capped, and
allowed to stand on the bench for observation. Complexes 1a, 1b,
1c, 1f, 1g, and 1j immediately changed color from orange to purple
upon the addition of O2. IMe complex 1k immediately became
yellow upon the addition of O2. ITol complex 1h gradually turned
purple over ∼30 min. MeItBu complex 1i was completely
consumed within 90 s. ItBu complex 1d and IAd complex 1e did
not turn purple within 24 h, but were observed to decompose over
48 h, still without any observable bis-µ-hydroxonickel complex
formation.
CH2Cl2: A procedure similar to that for thf was used, substituting
freshly distilled CH2Cl2 for thf. In this case, the only sample to
exhibit any change in less than 5 h was IMe complex 1k, which
became yellow over 1 h with precipitation of a white solid. The
only case where formation of a purple solution was observed was
for IMes complex 1c, which took 24 h. Complexes 1a, 1b, 1f, 1g,
and 1j were all observed to decompose within 36 h, albeit without
any bis-µ-hydroxonickel complex formation observed.
Visible Spectroscopy. Formation of bis-µ-hydroxonickel com-
plexes was verified by visible spectroscopy for samples 1b, 1c, 1f,
1g, 1h, 1i, and 1j. These complexes were not sufficiently stable to
fully characterize. However, oxidation of 1b, 1c, and 1j in toluene
at 0 °C revealed the extremely rapid formation of broad peaks with
λmax values of 492 (1b), 497 (1c), and 496 nm (1j). Oxidation of
1f, 1g, 1h, and 1i at 0 °C in thf led to products with λmax values of
488 (1f), 483 (1g), 498 (1h), and 497 nm (1i). The formation of a
bis-µ-hydroxonickel complex for 1k could not be verified at 0 °C.
Oxidation of 1k at -36 °C also did not reveal a peak at ∼500 nm.
In the oxidation of 1k, the product lacks any peaks in the visible
region, with only a strong UV absorbance with a wide shoulder
into the visible region.
1H NMR (300 MHz, C6D6): δ 1.43 (d, J ) 12.7 Hz, 1H, allyl-
H), 2.06 (dt, J ) 6.6 Hz, 2.4 Hz, 1H, allyl-H), 2.94 (d, J ) 14.2
Hz, 1H, allyl-H), 3.33 (s, 6H, NMe), 3.83 (dd, J ) 7.6 Hz, 2.1 Hz,
1H, allyl-H), 5.04 (sept, J ) 6.9 Hz, 1H, allyl-H), 5.84 (s, 2H,
RNCHCHNR). 13C NMR {1H} (75 MHz, C6D6): δ 14.70, 37.24,
42.65, 69.08, 108.06, 121.94; IR(KBr): 3120(m), 3066(w), 2928-
(w), 1571(w), 1461(m), 1400(s), 1229(s), 1196(w), 1135(w), 1088-
(w), 1013(m), 919(m), (871(m), 747(s), 679(m), 568(w), 525(w),
463(m), 434(w) cm-1. HRMS (EI) m/Z: (M)+ calcd, 230.0118;
obsd, 230.0119.
Allyl(MeItBu)NiI (1l). MeItBu-HI (133 mg, 0.5 mmol) and
KOtBu (59 mg, 0.52 mmol) were added to a 20 mL scintillation
vial and suspended in thf (ca. 4 mL). The resulting mixture was
stirred magnetically for 90 min. In a separate 20 mL vial, Ni(cod)2
(137 mg, 0.5 mmol) was suspended in 1,5-cod (ca 500 mg). To
this suspension was added allyl chloride (37 mg, 0.5 mmol). The
suspension of Ni(cod)2 was then swirled by hand until no crystals
of Ni(cod)2 were visible in the vial. The reaction mixture containing
MeItBu, KI, and tBuOH in thf was then added to the vial containing
π-allylnickel chloride by pipet. Residual material was transferred
using 2 × 1 mL portions of thf. After ca. 5 min, the reaction mixture
was filtered over Celite and concentrated to dryness under a vacuum.
The product was then purified by dissolving the residue in a minimal
volume of toluene (ca. 3 mL), layering the resulting solution with
10 mL hexanes, and cooling the solution to -25 °C. Dark orange
crystals were then isolated by pipeting off the mother liquor and
drying the product under a vacuum to yield 1l (60 mg, 33% yield).
One of these crystals of 1l was used for single-crystal X-ray
diffraction studies. Note: this compound exists as a 2:1 mixture of
1
atropisomers as determined by H NMR.
1H NMR (300 MHz, CD2Cl2): δ 1.73 (s, 6H, CMe3), 1.98 (s,
3H, CMe3), 2.38 (d, J ) 13.9 Hz 0.66H, allyl-H), 2.41 (d, J )
14.2 Hz, 0.33H, allyl-H), 2.94 (dt, J ) 6.8 Hz, 2.4 Hz, 0.66H,
allyl-H), 3.03 (dt, J ) 6.9 Hz, 2.4 Hz, 0.33H, allyl-H), 3.44 (d, J
) 12 Hz, 0.66H, allyl-H), 3.56 (d, J ) 7.6 Hz, 0.33H, allyl-H),
3.87 (s, 1H, NMe), 4.23 (s, 2H, NMe), 4.99 (sept, J ) 6.8 Hz,
0.66H, allyl-H), 5.21-5.37 (superimposed with solvent), 6.90 (d,
J ) 1.9 Hz, 0.33H, RNCHCHNR), 6.99 (d, J ) 1.9 Hz, 0.66H,
RNCHCHNR), 7.07 (d, J ) 1.9 Hz, 0.66H, RNCHCHNR), 7.12
(d, J ) 2.0 Hz, 0.33H, RNCHCHNR). 13C NMR {1H} (75 MHz,
C6D6): δ 31.46, 31.69, 38.42, 39.19, 51.06, 51.37, 63.21, 63.33,
104.74, 106.53, 119.70, 119.79, 122.58, 122.67. IR (KBr): 2979-
(s), 2944(m), 2924(m), 1471(s), 1455(m), 1445(m), 1397(s), 1385-
(s), 1371(s), 1306(m), 1235(s), 1210(s), 1084(m), 1017(w), 915(m),
880(w), 733(s), 687(m) cm-1. Anal. Calcd for C11H19IN2Ni: C,
36.21; H, 5.25; N, 7.68. Found: C, 36.93; H, 5.10; N, 7.65.
trans-(ITol)2NiCl2 (3h). ITol complex 1h (69 mg, 0.18 mmol)
was dissolved in benzene (9 mL) in a 20 mL scintillation vial. This
vial was left standing on the lab bench for 10 days. During this
time, a significant quantity of fine beige powder precipitated from
solution, beginning within the first day. Over the last 2 days,
compact ruby red crystals of complex 3h formed. The beige powder
and mother liquor were removed via pipet. The remainder of the
beige powder was removed by successive washings with hexanes
(8 × 2 mL), although the beige solid is not soluble. Complex 3h
was then dried under a vacuum to yield red crystals suitable for
X-ray analysis (12 mg, 21% yield based on ITol). Complex 3h
was insufficiently soluble in noncoordinating NMR solvents to
obtain acceptable NMR spectra. Complex 3h decomposed in
coordinating NMR solvents.
Acknowledgment. This work was supported by the
University of Utah Research Foundation and partially sup-
ported by the National Institutes of Health (NIGMS RO1
GM3540).
Supporting Information Available: Crystallographic details,
1H NMR spectra, and VT-NMR spectra. This material is available
free of charge via the Internet at http://pubs.acs.org.
IC0612451
Inorganic Chemistry, Vol. 45, No. 20, 2006 8441