Thermally stable chromium(III) η3-allyl complexes. Direct structural comparison of neutral chromium(II) and cationic chromium(III) η3-allyl redox isomers

Article abstract of DOI:10.1021/om050415z

The first thermally stable chromium(III) η³-allyl complexes have been prepared. Oxidation of neutral cyclopentadienylchromium(II) dicarbonyl η³-allyl, η³-crotyl, and η³- cyclohexenyl complexes 1-3, respectively, with NOPF₆, provides the corresponding cationic chromium-(III) analogues 4-6. The structure of the cationic complex 4 was confirmed in the solid state by X-ray crystallography and compared to the crystal structure of neutral Cr(II) complex 1, which bears an identical ligand set. The cationic chromium(III) allyl complexes are thermally stable, but disproportionate at high concentration in donor solvents or in the presence of donor ligands: substitution of the carbonyl ligands with concomitant retention of the η³-allyl moiety is unsuccessful under a range of reaction conditions. The reaction between cationic allyl complex 4 and the N-heterocyclic carbene, 1,3-bis(2,4,6,-trimethylphenyl)-imidazole-2-ylidene (IMes), leads to the formation of IMesHPF₆ and intractable chromium-containing products, while the reaction between the neutral complex 1 and IMes results exclusively in nucleophilic alkylation of the allyl ligand. Treatment of complex 4 with tertiary phosphines provides low yields of diamagnetic bis(phosphine)chromium(II) complexes.

Full text of DOI:10.1021/om050415z

Organometallics 2005, 24, 4461-4467
4461
Thermally Stable Chromium(III) η3-Allyl Complexes.
Direct Structural Comparison of Neutral Chromium(II)
and Cationic Chromium(III) η3-Allyl Redox Isomers
David W. Norman, Robert McDonald,† and Jeffrey M. Stryker*
Department of Chemistry, University of Alberta, Edmonton, Alberta, T6G 2G2, Canada
Received May 25, 2005
The first thermally stable chromium(III) η3-allyl complexes have been prepared. Oxidation
of neutral cyclopentadienylchromium(II) dicarbonyl η3-allyl, η3-crotyl, and η3-cyclohexenyl
complexes 1-3, respectively, with NOPF6, provides the corresponding cationic chromium-
(III) analogues 4-6. The structure of the cationic complex 4 was confirmed in the solid state
by X-ray crystallography and compared to the crystal structure of neutral Cr(II) complex 1,
which bears an identical ligand set. The cationic chromium(III) allyl complexes are thermally
stable, but disproportionate at high concentration in donor solvents or in the presence of
donor ligands: substitution of the carbonyl ligands with concomitant retention of the
η3-allyl moiety is unsuccessful under a range of reaction conditions. The reaction between
cationic allyl complex 4 and the N-heterocyclic carbene, 1,3-bis(2,4,6,-trimethylphenyl)-
imidazole-2-ylidene (IMes), leads to the formation of IMesHPF6 and intractable chromium-
containing products, while the reaction between the neutral complex 1 and IMes results
exclusively in nucleophilic alkylation of the allyl ligand. Treatment of complex 4 with tertiary
phosphines provides low yields of diamagnetic bis(phosphine)chromium(II) complexes.
Introduction
characterization of several thermally stable cationic
chromium(III) η3-allyl complexes and provide a direct
structural comparison of the unsubstituted cationic
complex with the corresponding neutral chromium(II)
η3-allyl analogue.
Organochromium chemistry has received considerable
attention over the past two decades.1 Notable among
many interesting developments in the area of allylchro-
mium chemistry, chromium(III) η3-allyl complexes have
been used as precursors in the preparation of olefin
polymerization catalysts2 and are putative intermedi-
ates in the stereoselective synthesis of homoallylic
alcohols (i.e., the Nozaki-Hiyama-Kishi reaction).1b-e,3
Despite such attractive applications, thermally stable
chromium(III) complexes of this type are unreported.4,5
Here, however, we report the preparation, isolation, and
Results and Discussion
We recently reported a series of thermally stable
chromium(II) η3-allyl dicarbonyl complexes bearing a
range of cyclopentadienyl and allyl ligands.6 Reactivity
studies reveal the compounds to be stable in toluene at
reflux, unreactive toward trimethylamine N-oxide, and
“terminally sensitive” to photolytic conditions. Treat-
ment of chromium(II) allyl complex 1 with ferricinium,
silver, or nitrosonium salts, however, provides crude
reaction mixtures that show spectroscopic evidence for
oxidation of the metal from chromium(II) to chromium-
(III). As monitored by solution infrared spectroscopy, the
carbonyl absorptions for the neutral complex 1 in THF
(1939 and 1869 cm-1) shift to markedly higher fre-
quency in the oxidized product (2070 and 2032 cm-1),
clearly indicative of stronger C-O bonds and weaker
chromium-CO dfπ* back-bonding interactions. Unfor-
tunately, the oxidized product does not persist in THF,
acetonitrile, or acetone solutions: the higher energy
carbonyl vibrations disappear completely after 3 h. To
* Corresponding author. E-mail: jeff.stryker@ualberta.ca.
† X-ray Crystallography Laboratory.
(1) (a) Jolly, P. W. Acc. Chem. Res. 1996, 29, 544-551. (b) Wessjo-
hann, L. A.; Schied, G. Synthesis 1999, 1-36. (c) Fu¨rstner, A. Chem.
Rev. 1999, 99, 991-1045. (d) Avalos, M. Babiano, R.; Cintas, P.;
Jimenez, J. L. Palacios, J. C. Chem. Soc. Rev. 1999, 28, 169-177.
(e) Takai, K.; Nozaki, H. Proc. Jpn. Acad., Ser. B 2000, 76B, 123-
131.
(2) Bade, O. M.; Blom, R.; Ystenes, M. Organometallics 1998, 17,
2524-2533.
(3) Selected recent references: (a) Inoue, M.; Suzuki, T.; Nakada,
M. J. Am. Chem. Soc. 2003, 125, 1140-1141. (b) Fu¨rstner, A.; Shi, N.
J. Am. Chem. Soc. 1996, 118, 12349-12357. (c) Nowotny, S.; Tucker,
C.; Jubert, C.; Knochel, P. J. Org. Chem. 1995, 60, 2762-2772. See
also, references therein.
(4) All previously reported chromium(III) η3-allyl complexes are
unstable at room temperature. (a) Cr(C3H5)3: Wilke, G.; Bogdanovic,
B.; Hardt, P.; Heimbach, P.; Keim, W.; Kro¨ner, M.; Oberkirch, W.;
Tanaka, K.; Steinru¨cke, E.; Walter, D.; Zimmermann, H. Angew.
Chem., Int. Ed. Engl. 1966, 5, 151-166. O’Brien, S.; Fishwick, M.;
McDermott, B.; Wallbridge, M. G. H.; Wright, G. A. Inorg. Synth. 1972,
13, 73-79. (b) CpCr(C3H5)2: Nieman, J.; Pattiasina, J. W.; Teuben, J.
H. J. Organomet. Chem. 1984, 262, 157-169. Angermund, K.; Do¨hring,
A.; Jolly, P. W.; Kru¨ger, C.; Roma˜o, C. C. Organometallics 1986, 5,
1268-1269. (c) Cp′Cr(C3H5)2, Cr(C3H5)2Cl: Betz, P.; Do¨hring, A.;
Emrich, R.; Goddard, R.; Jolly, P. W.; Kru¨ger, C.; Roma˜o, C.; Scho¨n-
felder, K. U.; Tsay, Y.-H. Polyhedron 1993, 12, 2651-2662.
(5) Tentatively assigned chromium(III) η3-allyl complexes have also
been proposed as reactive intermediates. (a) (Cp-ansa-amino)Cr-
(C3H5)2: Do¨hring, A.; Go¨hre, J.; Jolly, P. W.; Kryger, B.; Rust, J.;
Verhovnik, G. P. J. Organometallics 2000, 19, 388-402. (b) CpCr-
(benzamidinato)-(C3H5): Gallant, A. J.; Smith, K. M.; Patrick, B. O.
Chem. Commun. 2002, 2914-2915.
(6) Norman, D. W.; Ferguson, M. J.; Stryker, J. M. Organometallics
2004, 23, 2015-2019.
10.1021/om050415z CCC: $30.25 © 2005 American Chemical Society
Publication on Web 07/29/2005

4462 Organometallics, Vol. 24, No. 18, 2005
Norman et al.
Table 1. Synthesis of Cationic η3-Allyl Complexes
4-6 (eq 1)a
plex,6 however, does not provide a precipitate, and no
tractable reaction product can be isolated from solution.
The infrared spectra of complexes 4 and 6 each show
two carbonyl stretching frequencies, suggesting the
presence of a single stereoisomer, whereas the spectrum
of complex 5 displays four high-energy absorptions,
implying the presence of both endo and exo isomers.9
Both chromium(III) η3-allyl complexes 4 and 5 are green
to yellow-green in color; the η3-cyclohexenyl complex 6,
however, is an orange-red powder. This contrast in color
suggests the possibility of unique bonding interactions
for the cyclohexenyl ligand, including the possibility of
an agostic interaction between one methylene C-H
bond of the η3-cyclohexenyl ligand and the now unsatur-
ated 17-electron metal center. Indeed, the X-ray crystal
structure6 of the neutral η3-cyclohexenyl precursor 3
reveals the distance between the metal and one distal
(nonallylic) methylene hydrogen to be just 3.22 Å; upon
one-electron oxidation, this distance may contract suf-
ficiently to create an agostic bonding interaction. Al-
ternatively, the difference in color may simply result
from the unique 1,3-bis(anti) substitution pattern im-
posed by the endocyclic disposition of the η3-allyl moiety.
Single crystals of the unsubstituted cationic allyl
complex 4 were deposited directly from the oxidation
reaction conducted at higher dilution, and the structure
was confirmed by X-ray crystallography (Figure 1).10
While the solid state structure of this complex is in itself
unremarkable, more notable is the opportunity to
compare this cationic chromium(III) complex with the
solid state structure of the neutral but otherwise identi-
cal chromium(II) complex 1 (Table 2).6 The structures,
while superficially similar, differ in a number of sig-
nificant parameters. The chromium responds to the
oxidation by marginally decreasing the metal-carbon
bond distances for the two terminal allyl positions to
2.220(3) Å from 2.231(5) and 2.239(5) Å in complex 1,
but the distance between the allyl central carbon and
the chromium center increases to 2.151(4) Å from
2.108(4) Å. These differences in bond distances alter the
dihedral angle between the planes containing the allyl
ligand and the cyclopentadienyl, respectively, which
increases by more than 5° in the oxidized complex. The
most striking difference between the two η3-allyl com-
plexes, however, is manifest in the dihedral angle
between the plane containing the Cr(CO)2 fragment and
that of the cyclopentadienyl ring: in cationic complex
4, the Cr(CO)2 plane is tilted fully 10° closer to the plane
of the ring than it is in neutral complex 1. This
substantial relaxation of the roughly square pyramidal
coordination sphere may be attributed to the reduction
in dfπ* back-donation from the oxidized metal center
to the ancillary carbonyl and allyl ligand. The location
of the counterion in or just under the cleft formed
between the allyl and carbonyl ligands (Figure 1),
however, suggests that the ancillary ligands shift
toward the cyclopentadienyl ring in order to accom-
modate a closer association of counteranion and the
cationic chromium center. Unsurprisingly, with the
a Reaction conditions: NOPF6, DME, 0 °C RT, 20 min. b Infra-
red spectra recorded as a Nujol mull, cm-1
.
c Isolated yield (%) of
analytically pure material.
stabilize the nitrosonium oxidant7 and to avoid pro-
longed exposure to donor solvent, the oxidation reaction
was conducted in 1,2-dimethoxyethane (DME), a solvent
in which the chromium(III) product is only sparingly
soluble. Thus, treatment of the chromium(II) η3-allyl
complex 1 with nitrosonium hexafluorophosphate in
DME at 0 °C leads to considerable effervescence and
precipitation of an analytically pure paramagnetic green
solid, subsequently identified as cationic chromium(III)
η3-allyl complex 4 (eq 1 and Table 1, entry 1). Infrared
spectroscopy of this solid as a Nujol mull shows carbonyl
stretching frequencies at 2070 and 2032 cm-1, with no
evidence of the neutral η3-allyl dicarbonyl complex. It
is interesting to note that addition of NOPF6 to complex
1 does not lead to a nitrosyl complex, as reported for
the corresponding molybdenum analogue.7b The smaller
ionic radius of chromium apparently permits one-
electron oxidation by nitrosonium ion, but not the
subsequent coordination of the co-generated nitric ox-
ide.8
This methodology is readily extended to the oxidation
of substituted chromium(II) η3-allyl complexes. Thus,
treatment of η3-crotyl and η3-cyclohexenyl complexes 2
and 3 with NOPF6 in DME leads to the oxidized
congeners 5 and 6, respectively (Table 1, entries 2 and
3). Oxidation of the more soluble indenyl η3-allyl com-
(7) Liebeskind7a has established that nitrosonium salts are unstable
in most donor solvents, with the notable exception of DME at 0 °C.
Indeed, nitrosonium salts in DME are persistent enough to react with
cyclopentadienylmolybdenum η3-allyl dicarbonyl compounds to provide
the corresponding η3-allyl nitrosylcarbonyl complexes in much higher
yields than originally reported by Faller.7b (a) Cosford, N. D. P.;
Liebeskind, L. S. Organometallics 1994, 13, 1498-1503. (b) Faller, J.
W.; Rosan, A. M. J. Am. Chem. Soc. 1976, 98, 3388-3389.
(8) Similar reactivity has been observed for other chromium(II)
complexes upon nitrosonium ion oxidation to the chromium(III)
analogues: (a) Shen, J. K.; Freeman, J. W.; Hallinan, N. C.; Rheingold,
A. L.; Arif, A. M.; Ernst, R. D.; Basolo, F. Organometallics 1992, 11,
3215-3224. (b) Connelly, N. G.; Johnson, G. A. J. Organomet. Chem.
1974, 77, 341-344.
(9) Elemental analysis of complex 5 confirms the compositional
homogeneity, supporting the tentative assignment of an endo/exo
mixture of allyl stereoisomers.
(10) Crystal data for complex
-80 °C); orthorhombic, Cmcm (No. 63), a ) 10.4586(10) Å, b ) 11.5544-
(11) Å, c ) 26.008(3) Å, V ) 3142.9(5) Å3, Z ) 8, Fcalcd ) 1.709 g cm-3
µ ) 0.902 mm-1, R1 ) 0.0448 (Fo2 g 2σ(Fo2)), wR2 ) 0.1189 (all data).
4 (C10H10CrF6O2P‚0.5C4H10O2,
,

Thermally Stable Cr(III) η3-Allyl Complexes
Organometallics, Vol. 24, No. 18, 2005 4463
Table 2. Selected Bond Lengths (Å), Angles (deg),
and Dihedral Angles between Planes (deg) for
η3-Allyl Complexes 1 (neutral) and 4 (cation)a
Figure 1. Solid state molecular structure of complex 4
showing the disordered PF6 counterion; 0.5 equiv of inter-
stitial DME is omitted. Non-hydrogen atoms are repre-
sented by Gaussian ellipsoids at the 20% probability level.
Hydrogen atoms are shown with arbitrarily small thermal
parameters. See Table 2 for selected bond lengths and
angles of the cation. Complete listings of bond lengths and
angles are available as Supporting Information.
novel dicationic cyclopentadienylchromium(III) tris-
(acetonitrile) complex 7 in high yield (eq 2). The pres-
ence of 1,5-hexadiene (unquantified) in the crude reac-
tion mixture was confirmed by gas chromatography,
presumably arising from homolytic cleavage of the allyl
ligand. The mechanism of this reaction is presumed to
involve equilibrium dissociation of the carbonyl ligands
to give the cationic bis(acetonitrile) complex I (Scheme
1). This relatively electron rich intermediate can then
reduce the remaining dicarbonyl cation 4, ultimately
producing a 1:1 mixture of the neutral η3-allyl complex
1 and the thermally unstable dicationic chromium(IV)
allyl intermediate II. Subsequent association of aceto-
nitrile followed by homolysis of the consequent η1-allyl
ligand leads to the formation of 1,5-hexadiene and the
observed dicationic tris(acetonitrile) complex 7. Consis-
tent with this mechanism, the rate of the dispropor-
tionation is reduced at higher dilution. The structural
assignment is supported by independent synthesis:
treatment of the known dichloro dimer [(η5-C5H5)-
a Complete listings of bond lengths and angles are provided as
Supporting Information. b Counterion omitted for clarity. c Tabu-
lation taken from one of two independent molecules; values for
the more disordered molecule are available as Supporting Infor-
mation.
reduction in dfπ* back-donation, the chromium-CO
bond distances in complex 4 increase to 1.901(3) Å,
notably longer than in the lower valent complex 1
(1.814(6) and 1.817(5) Å). Accordingly, the C-O bond
lengths of the cationic complex are approximately 3%
shorter than those in complex 1, resulting in an almost
150 cm-1 increase in the average carbonyl stretching
frequency in the infrared spectrum of the cationic
complex. Such changes in bond distances and infrared
absorptions have been previously reported for redox
pairs of chromium dicarbonyl complexes, albeit with
entirely different ligand sets.8b,11
4c
CrCl2]2 with 4 equiv of silver hexafluorophosphate in
acetonitrile affords the tris(acetonitrile) complex 7 in
high yield.
Scheme 1
Cationic η3-allyl complexes 4-6 are indefinitely stable
as solids at room temperature, unique among previously
reported chromium(III) η3-allyl compounds.4,5 In donor
solvents, however, the complexes are susceptible to
rapid disproportionation. In acetonitrile, for example,
complex 4 is completely consumed after 3 h to give a
1:1 mixture of the neutral η3-allyl complex 1 and the
(11) Orpen, A. G.; Connelly, N. G. Organometallics 1990, 9, 1206-
1210. Adams, C. J.; Bartlett, I. M.; Connelly, N. G.; Harding, D. J.;
Hayward, O. D.; Mart´ın, A. J.; Orpen, A. G.; Quayle, M. J.; Rieger, P.
H. J. Chem. Soc., Dalton Trans. 2002, 4281-4288.

4464 Organometallics, Vol. 24, No. 18, 2005
Norman et al.
Our attempts to induce substitution of the carbonyl
ligands in cationic complex 4 without concomitant redox
disproportionation have met no notable success. Upon
treatment with trimethylamine N-oxide, a suspension
of complex 4 in DME provides only an intractable
product mixture, while heating a suspension in toluene
gives a variable amount of the neutral η3-allyl dicarbo-
nyl complex 1 and an uncharacterizable blue material.
Displacement of the carbonyl ligands with strong neu-
tral donors also failed to yield products retaining the
allyl ligand. For example, the reaction of complex 4 with
the stable N-heterocyclic carbene 1,3-bis(2,4,6,-trimethyl-
phenyl)imidazole-2-ylidene (IMes) gave unidentified
metal-containing product(s). A diagmagnetic organic
Figure 2. Solid state molecular structure of complex 8;
0.5 equiv of interstitial THF is omitted. Non-hydrogen
atoms are represented by Gaussian ellipsoids at the 20%
probability level. Hydrogen atoms are shown with arbi-
trarily small thermal parameters. Selected bond lengths
(Å) and angles (deg): Cr-C(1) ) 1.789(2), Cr-C(2) )
1.798(3), Cr-C(3) ) 2.150(2), Cr-C(4) ) 2.140(2), Cr-
1
fraction was also isolated, which by H NMR spectros-
copy proved to be mostly the protonated carbene,
IMesHPF6 (III), a known compound,12 along with a trace
of the tentatively identified allylated analogue, 1,3-bis-
(2,4,6,-trimethylphenyl)-2-(1-propenyl)imidazolium hexa-
fluorophosphate, compound IV (eq 3). The major reac-
C(10) )2.231(3), Cr-C(11)
) 2.228(2), Cr-C(12) )
2.223(2), Cr-C(13) ) 2.217(2), Cr-C(14) ) 2.218(3),
O(1)-C(1) ) 1.182(3), O(2)-C(2) ) 1.180(3), C(3)-C(4) )
1.414(3), C(4)-C(5) ) 1.523(3), C(5)-C(6) ) 1.490(3), C(7)-
C(8) ) 1.340(3), N(1)-C(6) ) 1.343(3), N(2)-C(6) )
1.341(3); C(1)-Cr-C(10) ) 102.89(10), C(1)-Cr-C(2) )
85.89(10), C(1)-Cr-C(3) ) 109.99(9), C(1)-Cr-C(4) )
84.07(9), C(3)-Cr-C(4) ) 38.48(8), C(3)-C(4)-C(5) )
119.10(19), C(4)-C(5)-C(6) ) 110.27(17), N(1)-C(6)-N(2)
) 106.85(17).
tion pathway thus appears to be the deprotonation of
the starting complex (presumably at the allyl ligand),
accompanied by a less favorable nucleophillic addition
to the η3-allyl terminal position. By way of comparison,
the reaction between the neutral η3-allyl complex 1 and
IMes surprisingly gives the diamagnetic zwitterionic
alkylation product 8 in excellent yield (eq 4). The
Scheme 2
structure of this complex was determined by spectro-
scopic analysis and confirmed in the solid state by X-ray
crystallography (Figure 2).13,14 Although the reactions
with IMes do not lead to decarbonylation of either the
neutral or cationic η3-allyl complexes, the products
reveal unprecedented ligand-centered reactivity for the
stabilized carbene, which more typically functions as an
ancillary ligand in organometallic systems.15,16 It is not
obvious why the more electrophilic chromium(III) com-
plex 4 favors deprotonation over nucleophilic addition,
despite the obviously low activation barrier for the
nucleophilic pathway in the neutral congener.
The investigation of phosphine substitution was
equally disappointing. In contrast to the chemistry of
the neutral chromium(II) η3-allyl complex, the reaction
of chromium(III) complex 4 with monodentate or biden-
tate tertiary phosphine proceeds at or below room
temperature, but leads only to the isolation of diamag-
netic cationic cyclopentadienylchromium(II) bis(phos-
phine) complexes in low yield (Scheme 2). Thus, treat-
ment of a suspension of complex 4 in DME with 2 equiv
of trimethylphosphine leads to the formation of a yellow-
green solid, identified spectroscopically as the trans-bis-
(trimethylphosphine)cyclopentadienylchromium(II) di-
carbonyl cation 9. A second product, isolated from the
supernatant, was determined to be the known allyltri-
methylphosphonium hexafluorophosphate by spectro-
scopic comparison to authentic material prepared in-
(12) The major product, IMesHPF6, was identified by comparison
with an authentic sample prepared independently: Arduengo, A. J.,
III; Gamper, S. F.; Tamm, M.; Calabrese, J. C.; Davidson, F.; Craig,
H. A. J. Am. Chem. Soc. 1995, 117, 572-573. The trace product was
tentatively identified based on broad 1H NMR signals between 5.0 and
6.5 ppm, indicative of a terminal olefin.
(13) Crystal data for complex 8 (C31H34CrN2O2‚0.5C4H8O, -80 °C);
monoclinic, I2/a (an alternate setting of C2/c [No. 15]), a ) 42.1994-
(18) Å, b ) 8.4552(4) Å, c ) 37.1949(16) Å, â ) 115.8337(7)°, V )
11945.0(9) Å3, Z ) 16, Fcalcd ) 1.234 g cm-3, µ ) 0.416 mm-1, R1
)
2
0.0482 (Fo g 2σ(Fo2)), wR2 ) 0.1396 (all data).
(14) Although anionic 18e- Cr(0) complexes of the form η5-CpCrL2-
(olefin)- appear to be unprecedented, corresponding neutral (η6-arene)-
CrL2(olefin) analogues have been reported; see: Angelici, R. J.; Busetto,
L. Inorg. Chem. 1968, 7, 1935-1936, and references therein.
(15) Herrmann, W. A. Angew. Chem., Int. Ed. 2002, 41, 1290-1309.
Herrmann, W. A. Adv. Organomet. Chem. 2002, 48, 1-69.
(16) The high nucleophilicity of the N-heterocyclic carbene toward
the neutral allyl complex 1 is both noteworthy and surprising; no
reaction is observed upon addition of excess PMe3, even at 65 °C.

Thermally Stable Cr(III) η3-Allyl Complexes
Organometallics, Vol. 24, No. 18, 2005 4465
Figure 3. Solid state molecular structure of complex 9,
showing only the cationic fragment. Non-hydrogen atoms
are represented by Gaussian ellipsoids at the 20% prob-
ability level. Hydrogen atoms are shown with arbitrarily
small thermal parameters. Selected bond lengths (Å) and
angles (deg): Cr-C(1) ) 1.843(3), Cr-P(1) ) 2.3509(8),
Cr-C(10) ) 2.193(5), Cr-C(11) ) 2.199(4), Cr-C(12) )
Figure 4. Solid state molecular structure of complex 10
showing only the cationic fragment; 1.5 equiv of interstitial
acetone is omitted. Non-hydrogen atoms are represented
by Gaussian ellipsoids at the 20% probability level. Hy-
drogen atoms are shown with arbitrarily small thermal
parameters. Selected bond lengths (Å) and angles (deg):
Cr-C(1) ) 1.854(6), Cr-C(2) ) 1.834(5), Cr-P(1) )
2.3994(13), Cr-P(2) ) 2.3889(13), Cr-C10, 2.171(5), Cr-
C(11) ) 2.186(5), Cr-C(12) ) 2.234(5), Cr-C(13) )
2.242(5), Cr-C(14) ) 2.187(5), O(1)-C(1) ) 1.146(7), O(2)-
C(2) ) 1.162(7), P(1)-C(3) ) 1.866(4), C(3)-C(4) )
1.511(7), P(2)-C(4) ) 1.820(5); C(1)-Cr-C(10) ) 88.1(2),
C(1)-Cr-C(2) ) 78.6(2), P(1)-Cr-C(12) ) 87.29(14), P(2)-
Cr-C(13) ) 82.43(14), P(1)-Cr-P(2) ) 77.84(5), P(1)-
C(3)-C(4) ) 110.9(3), P(2)-C(4)-C(3) ) 105.7(3).
2.1997(5), O-C(1)
)
1.150(4); C(1)-Cr-C(10)
)
118.70(8), C(1)-Cr-C(1′) ) 111.26(17), P(1)-Cr-P(1′) )
129.66(4), P(1)-Cr-C(1) ) 76.11(4), P(1)-Cr-C(10) )
83.45(13).
Scheme 3
Conclusion
A general methodology for the preparation and isola-
tion of cationic cyclopentadienylchromium(III) η3-allyl
dicarbonyl complexes has thus been developed. These
new compounds are remarkably thermally stable, but
have a strong tendency to disproportionate at sufficient
concentration in donor solvents. The thermal stability
is reasonably attributed to the presence of the strongly
π-acidic carbonyl ligands, which maintain the allyl
ligand in the thermodynamically most stable η3-bonding
mode. In more electron-rich chromium(III) η3-allyl
complexes,4,5 the allyl ligand is likely equilibrating
between η3- and η1-hapticity, allowing the less strongly
bound η1-isomer to undergo facile chromium-carbon
bond homolysis and reduction to chromium(II). Com-
parison of the structures of the neutral and cationic
η3-allyl complexes reveals a significant change in the
coordination geometry of the cation, with the carbonyl
ligands positioned significantly closer to the cyclo-
pentadienyl ring than observed in the neutral chro-
mium(II) analogue. Despite the lability of the carbonyl
ligands in chromium(III) complexes 4-6, we have as yet
been unable to modify the ligand set without concomi-
tant loss of the allyl fragment.
dependently.17 Similar reaction conditions lead to the
formation of cis-[1,2-bis(diphenylphosphino)ethane]-
cyclopentadienylchromium(II) dicarbonyl cation 10 from
treatment with 1 equiv of bis(diphenylphophino)ethane,
along with the tentatively assigned doubly allylated
phosphonium salt. Complexes 9 and 10 presumably
arise from the association of one phosphine to the metal
with concomitant isomerization of the allyl ligand from
η3- to η1-hapticity (Scheme 3, path a). Subsequent
homolytic loss of allyl radical and association of a second
phosphine provide the observed product. The allylphos-
phonium salts must then result from an independent
pathway triggered by nucleophilic addition of phosphine
to the allyl ligand (Scheme 3, path b), followed by
dissociation of the alkene from the now Cr(I) center.
Unfortunately, no other tractable metal-containing
products could be isolated from these reactions, leaving
the fate of the reduced chromium species as yet unde-
termined. Although NMR spectroscopy was sufficient
to assign the stereochemistry about the metal center,
the structural determination of both complexes was
complemented by X-ray crystallography (Figures 3,
4).18,19 cis- and trans-Bis(phosphine) complexes similar
to 9 and 10 have been previously reported.8,20-22
(19) Crystal data for complex 10 (C33H29CrF6O2P3‚1.5C3H6O,
-80 °C); orthorhombic, P212121 (No. 19), a ) 12.0893(10) Å, b )
16.7108(14) Å, c ) 19.4470(16) Å, V ) 3928.7(6) Å3, Z ) 4, Fcalcd
)
)
1.359 g cm-3, µ ) 0.477 mm-1, R1 ) 0.0676 (Fo g 2σ(Fo2)), wR2
2
0.1911 (all data).
(20) Schubert, U.; Ackermann, K.; Janta, R.; Voran, S.; Malisch, W.
Chem. Ber. 1982, 115, 2003-2008.
(21) (a) Cooley, N. A.; MacConnachie, P. T. F.; Baird, M. C.
Polyhedron 1988, 7, 1965-1972. (b) Watkins, W. C.; Hensel, K.; Fortier,
S.; Macartney, D. H.; Baird, M. C. Organometallics 1992, 11, 2418-
2424.
(22) Salsini, L.; Pasquali, M.; Zandomeneghi, M.; Festa, C.; Leoni,
P.; Braga, D.; Sabatino, P. J. Chem. Soc., Dalton Trans. 1990, 2007-
2012.
(17) Cardaci, G. J. Chem. Soc., Dalton Trans. 1984, 815-818.
(18) Crystal data for complex 9 (C13H23CrF6O2P3, -80 °C); orthor-
hombic, Pmmn (No. 59), a ) 10.7360(7) Å, b ) 11.0577(7) Å, c )
8.0243(5) Å, V ) 952.61(10) Å3, Z ) 2, Fcalcd ) 1.639 g cm-3, µ ) 0.913
mm-1, R1 ) 0.0333 (Fo g 2σ(Fo2)), wR2 ) 0.0869 (all data).
2

4466 Organometallics, Vol. 24, No. 18, 2005
Norman et al.
lished the presence of 1,5-hexadiene; details are provided as
Supporting Information. The burgundy residue was triturated
with 3 × 5 mL of diethyl ether and the combined ether extracts
were concentrated under vacuum to give the neutral chromium
allyl complex 1 (6 mg, 47%), identified spectroscopically by
comparison to authentic material.6 The remaining purple
residue was recrystallized from acetonitrile/diethyl ether at
-35 °C to give [(η5-C5H5)Cr(NCCH3)3](PF6)2, 7 (14 mg, 43%),
as determined by comparison of the infrared spectrum and
combustion analysis to authentic material from independent
synthesis, given below.
Experimental Section
General Procedures. All manipulations of air-sensitive
compounds were performed under a nitrogen atmosphere using
standard Schlenk techniques or in a nitrogen-filled drybox.
Ethereal and hydrocarbon solvents were purified by distillation
from sodium or potassium benzophenone ketyl. Acetonitrile
was purified by distillation from calcium hydride and degassed.
Acetone was dried from boric anhydride and degassed. Infrared
(IR) spectra were obtained in solution (KBr solution cells) or
in the solid state (as Nujol mulls on KBr disks) and are
reported in wavenumbers (cm-1) calibrated to the 1601 cm-1
absorption of polystyrene. Chemical shifts are reported in the
δ scale, referenced to residual protiated solvent. Combustion
analyses were performed by the University of Alberta Micro-
analysis Laboratory. Some of the chromium compounds failed
to afford consistent elemental analysis, even when using highly
purified crystalline samples suitable for X-ray crystallography.
[(η3-Allyl)(η5-cyclopentadienyl)dicarbonylchromium]-
PF6‚(DME)0.5, 4. In a Schlenk flask under an inert atmo-
sphere, (η5-C5H5)Cr(CO)2(C3H5), 16 (405 mg, 1.89 mmol), was
dissolved in 20 mL of DME and cooled to 0 °C. A solution of
NOPF6 (331 mg, 1.89 mmol) in 15 mL of DME at 0 °C was
added via cannula. Immediately a color change from orange
to green was observed with effervescence and formation of a
green precipitate. After stirring for 20 min, the suspension was
warmed to room temperature and 20 mL of diethyl ether
added. The precipitate was collected on a frit, washed with
2 × 10 mL of diethyl ether, and dried to give allylchromium
[(η5-Cyclopentadienyl)tris(acetonitrile)chromium]-
(PF6)2, 7, from [(η5-C5H5)CrCl2]2. In a Schlenk flask under
an inert atmosphere [(η5-C5H5)CrCl2]2 (50 mg, 0.133 mmol)
4c
was dissolved in 20 mL of acetonitrile. A solution of silver
hexafluorophosphate (134 mg, 0.532 mmol) in 10 mL of
acetonitrile was added. A white precipitate formed immedi-
ately, and the solution changed color from green to dark
purple. The mixture was stirred for 2 h and filtered through
a frit layered with Celite, and the filtrate was mixed with
15 mL of diethyl ether. After 2 days at -35 °C, deep purple
crystals were deposited. The supernatant was removed and
the crystals were washed with 2 × 10 mL of diethyl ether and
dried to give the tris(acetonitrile) complex 7 (102 mg, 72%).
IR (νCN, cm-1, Nujol): 2325, 2296. Anal. Calcd for C11H14CrN3-
P2F12: C, 24.92; H, 2.66; N, 7.93. Found: C, 25.42; H, 2.50; N,
7.71.23
(η5-C5H5)Cr(CO)2(η2-2-allyl-1,3-dimesitylimidazolyl), 8.
To
a 1 (17.6 mg,
solution of (η5-C5H5)Cr(CO)2(C3H5),
complex 4 as a green DME solvate (625 mg, 82%). IR (νCO
,
0.082 mmol), in benzene was added 1,3-bis(2,4,6-trimethylphe-
nyl)imidazolin-2-ylidene (IMes) (25 mg, 0.082 mmol) dissolved
in a minimum of benzene. After standing for 3 h at room
temperature, light red crystals had deposited. The supernatant
was decanted and the crystals were washed with 2 × 5 mL of
diethyl ether. Drying under vacuum gave the zwitterionic
adduct 8 (41 mg, 89%). IR (νCO, cm-1, acetone): 1845, 1757.
1H NMR (500 MHz, acetone-d6): δ 7.84 (s, 2H, NCHCHN);
7.22 (br s, 2H, Hmeta); 7.16 (br s, 2H, Hmeta); 3.86 (s, 5H, C5H5);
3.63 (dd, J ) 15.0, 2.5 Hz, 1H, CH2); 2.39 (s, 6H, CH3-para);
2.26 (s, 6H, CH3-ortho); 2.14 (s, 6H, CH3-ortho); 1.69 (m, 1H,
cm-1, Nujol): 2070, 2032. Anal. Calcd for C10H10CrO2PF6‚
0.5DME: C, 35.66; H, 3.74. Found: C, 34.87; H, 3.80. Crystals
suitable for analysis by X-ray diffraction were grown from an
equimolar solution of complex 1 and NOPF6 in dilute DME
(∼46 mM), upon standing undisturbed at RT for 16 h.
[(η3-Crotyl)(η5-cyclopentadienyl)dicarbonylchromium]-
PF6, 5. In a Schlenk flask under an inert atmosphere,
(η5-C5H5)Cr(CO)2(C4H7), 26 (46 mg, 0.202 mmol), was dissolved
in 5 mL of DME and cooled to 0 °C. A solution of NOPF6
(35 mg, 0.202 mmol) in 10 mL of DME at 0 °C was added via
cannula. Immediately a color change from yellow to light green
was observed with concomitant effervescence and formation
of a light green precipitate. After stirring for 20 min, the
suspension was warmed to room temperature and 10 mL of
diethyl ether added. The precipitate was collected on a frit,
washed with 2 × 5 mL of diethyl ether, and dried to give
crotylchromium complex 5 as a pure yellow-green powder
(53 mg, 70%). IR (νCO, cm-1, Nujol): 2065, 2040, 2031, 2022.
Anal. Calcd for C11H12CrO2PF6: C, 35.40; H, 3.24. Found: C,
35.34; H, 2.92.
Hcentral); 1.44 (dd, J ) 15.0, 11.5 Hz, 1H, CH2); 1.32 (dd, J )
8.0, 2.0 Hz, 1H, Hsyn); -0.19 (dd, J ) 9.5, 2.0 Hz, 1H, Hanti).
13C NMR (125 MHz, C6D6): δ 258.3 (CO); 258.1 (CO); 153.2
(NCN); 141.9 (Cpara); 135.9 (Cortho); 135.5 (Cortho); 132.0 (Cipso);
130.8 (Cmeta); 130.5 (Cmeta); 123.7 (NCCN); 85.9 (C5H5); 35.9
(Cmethylene); 34.6 (Ccentral); 29.5 (Cterminal); 21.1 (CH3-para); 18.1
(CH3-ortho); 17.8 (CH3-ortho). The given spectroscopic assign-
ments were deduced from homo- and heterocorrelated NMR
spectroscopy; details are provided as Supporting Information.
Anal. Calcd for C31H34CrN2O2: C, 71.79; H, 5.40; N, 6.61.
Found: C, 71.21; H, 5.22; N, 6.83. Crystals suitable for X-ray
diffraction analysis were grown from an equimolar mixture
of complex 1 and IMes in THF upon standing at RT for 12 h,
providing complex 8 as the 0.5 THF solvate.
[(η3-Cyclohexenyl)(η5-cyclopentadienyl)dicarbonyl-
chromium]PF6, 6. In a Schlenk flask under an inert atmo-
sphere, (η5-C5H5)Cr(CO)2(C6H9), 36 (46 mg, 0.180 mmol), was
dissolved in 5 mL of DME and cooled to 0 °C. A solution of
NOPF6 (35 mg, 0.197 mmol) in 10 mL of DME at 0 °C was
added via cannula. Immediately a color change from yellow
to orange was observed with concomitant effervescence and
formation of an orange-red precipitate. After stirring for
20 min, the suspension was warmed to room temperature and
10 mL of diethyl ether added. The precipitate was collected
on a frit, washed with 2 × 5 mL of diethyl ether, and dried to
give cyclohexenylchromium complex 6 as an orange-red powder
(69 mg, 96%). IR (νCO, cm-1, Nujol): 2039, 2002. Anal. Calcd
for C13H14CrO2PF6: C, 39.11; H, 3.53. Found: C, 38.60; H, 3.42.
[(η5-Cyclopentadienyl)tris(acetonitrile)chromium]-
(PF6)2, 7, from [(η3-allyl)(η5-cyclopentadienyl)dicarbonyl-
[(η5-Cyclopentadienyl)bis(trimethylphosphine)-
dicarbonylchromium]PF6, 9. To a suspension of (η5-C5H5)-
Cr(CO)2(C3H5)]PF6‚(DME)0.5, 4 (111 mg, 0.275 mmol), in 20 mL
of DME was added PMe3 (577 µL, 1.0 M in THF, 0.577 mmol).
The color of the mixture immediately turned from dark green
to light green, and a light green precipitate formed. After
30 min the solid was collected on a frit, washed with 2 ×
10 mL of diethyl ether, and dried to give complex 9 as a yellow-
green powder (48 mg, 33%). IR (νCO, cm-1, Nujol): 1959, 1885.
1H NMR (400 MHz, acetone-d6): δ 5.25 (t, J ) 2.0 Hz, 5H,
C5H5); 1.84 (2nd order m, 18 H, PMe3). 31P NMR (162 MHz,
acetone-d6): 52.94 (s, PMe3); -138.03 (sept, J ) 708.1 Hz, PF6).
13C NMR (100 MHz, acetone-d6): δ 245.8 (t, J ) 51.1 Hz, CO);
chromium]PF6‚(DME)0.5
. Allylchromium(III) complex 4
(22 mg, 0.061 mmol) was dissolved in 5 mL of acetonitrile.
After stirring for 3 h, the green solution had turned burgundy
and the solvent was removed to a trap under vacuum. The
volatile fraction was subjected to GC analysis, which estab-
(23) The composition and bond connectivity of tris(acetonitrile)
complex 7 was confirmed by a preliminary X-ray crystal structure
analysis, which was abandoned prior to final refinement.

Thermally Stable Cr(III) η3-Allyl Complexes
Organometallics, Vol. 24, No. 18, 2005 4467
91.0 (s, C5H5); 19.6 (dd, J ) 15.6 2.0 Hz, PMe3). Crystals
suitable for X-ray diffraction were grown from acetone at
-35 °C. Anal. Calcd for C13H23CrO2P3F6: C, 29.9; H, 4.44.
Found: C, 30.84; H, 4.29.
(s, Cpara); 131.5 (t, J ) 3.9 Hz, Cmeta); 130.4 (t, J ) 5.1 Cortho);
129.6 (t, J ) 4.8 Hz, Cortho); 93.2 (s, C5H5); 28.7 (dd, J ) 39.8
Hz, CH2CH2). Anal. Calcd for C33H29CrO2P3F6: C, 55.32; H,
4.08. Found: C, 55.14; H, 4.11. Crystals suitable for X-ray
diffraction were grown from a 1:1 mixture of acetone and
diethyl ether at -35 °C.
{(η5-Cyclopentadienyl)[1,2-bis(diphenylphosphino)-
ethane)]dicarbonylchromium}PF6, 10. To a suspension of
(η5-C5H5)Cr(CO)2(C3H5)]PF6‚(DME)0.5, 4 (125 mg, 0.309 mmol),
in 20 mL of DME was added dppe (117 mg, 0.294 mmol). The
color of the mixture immediately turned from dark green to
light green, and after 30 min the solvent was removed under
vacuum. The residue was triturated with 2 × 10 mL of pentane
and the remaining material dissolved in 10 mL of acetone.
Diethyl ether (5 mL) was added and the solution was stored
at -35 °C for 12 h to yield a yellow precipitate. The solid was
collected on a frit, washed with 2 × 10 mL of diethyl ether,
and dried to give complex 10 as a yellow powder (45 mg, 21%).
IR (νCO, cm-1, Nujol): 1971, 1917. 1H NMR (500 MHz, acetone-
d6): δ 7.83 (m, 5H, Ph); 7.62 (m, 6H, Ph); 7.33 (m, 5H, Ph);
7.23 (m, 4H, Ph); 4.86 (t, J ) 1.5 Hz, 5H, C5H5); 3.43 (2nd
order m, 2H, CH2); 3.12 (2nd order m, 2H, CH2). 31P NMR
(162 MHz, acetone-d6): 106.78 (s, dppe); -137.93 (sept, J )
708.1 Hz, PF6). 13C NMR (100 MHz, acetone-d6): δ 249.0
(t, J ) 28.7 Hz, CO); 137.6 (t, J ) 24.6 Hz, Cipso); 133.6 (t, J )
3.9 Hz, Cmeta); 133.4 (t, J ) 22.2 Hz, Cipso); 132.2 (s, Cpara); 132.1
Acknowledgment. The authors thank Mr. Wayne
Moffat for assistance with infrared and gas chromato-
graphic analyses. We also acknowledge Profs. Josef
Takats and Martin Cowie for helpful discussions on the
solid state structural comparison of complexes 1 and 4.
Financial support from the Natural Sciences and En-
gineering Research Council (NSERC) of Canada and the
University of Alberta is gratefully acknowledged.
Supporting Information Available: Parameters and
measurements from the GC analysis; details of the two-
dimensional NMR spectroscopy and assignments for complex
8; details of the X-ray structure determinations for complexes
1, 4, 8, 9, and 10. This material is available free of charge via
the Internet at http://pubs.acs.org.
OM050415Z