Stable, cationic alkyl-olefin complexes of ruthenium(II) and rhodium(III): effects of ligand geometry upon olefin insertion/alkyl migration

Article abstract of DOI:10.1021/om000674i

The syntheses of ruthenium(II) and rhodium(III) model complexes of the putative active species in the iron(II)/MMAO polymerization system are reported. The cationic ruthenium methyl ethylene complex 3 is a stable, isolable compound, which does not produce polyethylene under a variety of conditions. Likewise, the analogous rhodium complex 12 shows no appreciable polymerization activity, despite its similarity to existing rhodium complexes which are active catalysts for olefin polymerization. X-ray crystal structures of the cationic ruthenium bis(ethylene) complex 7 and the five-coordinate rhodium bis(triflate) methyl complex 9 are also presented.

Full text of DOI:10.1021/om000674i

Organometallics 2000, 19, 4995-5004
4995
Sta ble, Ca tion ic Alk yl-Olefin Com p lexes of
Ru th en iu m (II) a n d Rh od iu m (III): Effects of Liga n d
Geom etr y u p on Olefin In ser tion /Alk yl Migr a tion
Eric L. Dias, Maurice Brookhart,* and Peter S. White
Department of Chemistry, University of North Carolina at Chapel Hill,
Chapel Hill, North Carolina 27599-3290
Received August 4, 2000
The syntheses of ruthenium(II) and rhodium(III) model complexes of the putative active
species in the iron(II)/MMAO polymerization system are reported. The cationic ruthenium
methyl ethylene complex 3 is a stable, isolable compound, which does not produce
polyethylene under a variety of conditions. Likewise, the analogous rhodium complex 12
shows no appreciable polymerization activity, despite its similarity to existing rhodium
complexes which are active catalysts for olefin polymerization. X-ray crystal structures of
the cationic ruthenium bis(ethylene) complex 7 and the five-coordinate rhodium bis(triflate)
methyl complex 9 are also presented.
In tr od u ction
tive precursors for the polymerization of ethylene to
either oligomers5 or high molecular weight polymer,3,4
depending upon the steric bulk of the ligand employed
(eq 1). Upon treatment with modified methylaluminox-
Following the discovery that transition-metal com-
plexes could catalyze the polymerization of ethylene and
R-olefins, a considerable amount of effort has been
dedicated to the design of new catalyst systems.1 By
varying the catalyst structure, the molecular weight,
composition, and tacticity of the resulting polymers can
be altered, allowing the properties of the material to be
controlled.1c-h While the majority of catalysts are based
on early transition metals, the advent of the nickel and
palladium diimine catalysts,2 which can produce poly-
ethylene with various degrees of branching depending
on the polymerization conditions,2b,h has sparked a great
deal of interest in late-transition-metal catalysts as
well.1a,b
ane (MMAO), exceptionally active catalysts are pro-
duced, consuming ethylene at rates of up to 106
turnovers/h.3-5 The polymers produced in this manner,
however, often exhibit multimodal polydispersities,
which can drastically affect the mechanical properties
of the materials. In addition, the high activities of the
catalysts along with the required excess cocatalyst
(MAO or MMAO) make this an increasingly difficult
system to study.
Recently, it was reported by Bennett at DuPont,3a,b
our group,3c and Gibson’s group4 that the 2,6-diketimi-
noylpyridine complexes of iron(II) dihalides 1 are effec-
(1) For reviews of organometallic catalysis in olefin polymerization,
see: (a) Ittel, S. D.; J ohnson, L. K.; Brookhart, M. Chem. Rev. 2000,
100, 1169-1203. (b) Britovsek, G. J . P.; Gibson, V. C.; Wass, D. F.
Angew. Chem., Int. Ed. 1999, 38, 428-447. (c) Kaminsky, W. J . Chem.
Soc., Dalton Trans. 1998, 1413-1418. (d) Soga, K.; Shiono, T. Prog.
Polym. Sci. 1997, 22, 1503-1546. (e) Bochmann, M. J . Chem. Soc.,
Dalton Trans. 1996, 255-270. (f) Brintzinger, H. H.; Fischer, D.;
Mulhaupt, R.; Rieger, B.; Waymouth, R. M. Angew. Chem., Int. Ed.
Engl. 1995, 34, 1143-1170. (g) Mohring, P. C.; Coville, N. J . J .
Organomet. Chem. 1994, 479, 1-29. (h) Gupta, V. K.; Satish, S.;
Bhardwaj, I. S. Rev. Macromol. Chem. Phys. C 1994, 34, 439-514.
(2) (a) Svejda, S. A.; J ohnson, L. K.; Brookhart, M. J . Am. Chem.
Soc. 1999, 121, 10634-10635. (b) Guan, Z. B.; Cotts, P. M.; McCord,
E. F.; McLain, S. J . Science 1999, 283, 2059-2062. (c) McLain, S. J .;
Feldman, J .; McCord, E. F.; Gardner, K. H.; Teasley, M. F.; et al.
Macromolecules 1998, 31, 6705-6707. (d) Mecking, S.; J ohnson, L. K.;
Wang, L.; Brookhart, M. J . Am. Chem. Soc. 1998, 120, 888-899. (e)
Killian, C. M.; J ohnson, L. K.; Brookhart, M. Organometallics 1997,
16, 2005-2007. (f) Killian, C. M.; Tempel, D. J .; J ohnson, L. K.;
Brookhart, M. J . Am. Chem. Soc. 1996, 118, 11664-11665. (g) J ohnson,
L. K.; Mecking, S.; Brookhart, M. J . Am. Chem. Soc. 1996, 118, 267-
268. (h) J ohnson, L. K.; Killian, C. M.; Brookhart, M. J . Am. Chem.
Soc. 1995, 117, 6414-6415.
By analogy with several well-defined olefin polymer-
ization catalysts,1 the putative active species in this
system is believed to be the cationic iron alkyl complex
2, which may have a molecule of ethylene bound under
the polymerization conditions. Although theoretical
studies using cationic iron(II) alkyl model compounds
have been performed,6 experimental evidence for the
(4) (a) Britovsek, G. J . P.; Bruce, M.; Gibson, V. C.; Kimberley, B.
S.; Maddox, P. J .; Mastroianni, S.; et al. J . Am. Chem. Soc. 1999, 121,
8728-8740. (b) Britovsek, G. J . P.; Gibson, V. C.; Kimberley, B. S.;
Maddox, P. J .; McTavish, S. J .; et al. Chem. Commun. 1998, 849-
850.
(5) Small, B. L.; Brookhart, M. J . Am. Chem. Soc. 1998, 120, 7143-
7144.
(6) (a) Deng, L. Q.; Margl, P.; Ziegler, T. J . Am. Chem. Soc. 1999,
121, 6479-6487. (b) Griffiths, E. A. H.; Britovsek, G. J . P.; Gibson, V.
C.; Gould, I. R. Chem. Commun. 1999, 1333-1334.
(3) (a) Bennett, A. M. A. WO9827124 to DuPont. (b) Bennett, A. M.
A. CHEMTECH 1999, 29, 24. (c) Small, B. L.; Brookhart, M.; Bennett,
A. M. A. J . Am. Chem. Soc. 1998, 120, 4049-4050.
10.1021/om000674i CCC: $19.00 © 2000 American Chemical Society
Publication on Web 10/21/2000

4996 Organometallics, Vol. 19, No. 24, 2000
Dias et al.
stability and reactivity of species such as 2 is still
somewhat lacking. To understand this system in further
detail, we chose to synthesize the analogous ruthenium-
(II) complex, which is expected to exhibit polymerization
activity lower than that of the iron system without the
need for a cocatalyst or activator, making it more
amenable to study.
In this paper, we report the synthesis and isolation
of the stable, cationic ruthenium methyl ethylene com-
plex 3 and the corresponding carbon monoxide adduct
4. Neither of these complexes are effective initiators for
Direct alkylation of 6 with MeLi, MeMgBr, or MgMe2
gave inconsistent results at best, even under strict
temperature control. While reaction of MeLi in THF did
in one case produce what appeared to be an alkylated
complex, this was not reproducible to a satisfactory
degree. As a result, we chose to form a cationic ruthe-
nium complex prior to alkylation.
Reaction of 6 with 1 equiv of NaB(Ar′)4 (Ar′ ) 3,5-
bis(trifluoromethyl)phenyl) under an ethylene atmo-
sphere in dichloromethane produces the bis(ethylene)
complex 7 (eq 3). The reaction is surprisingly clean, and
the polymerization of ethylene and R-olefins or the
copolymerization of olefins and CO. The synthesis of an
analogous isoelectronic, dicationic rhodium methyl eth-
ylene complex, which behaves in a similar fashion, is
also presented. Because the nickel and palladium di-
imine complexes both act as initiators for olefin polym-
erization, reasons for this apparent break in the periodic
trend will be discussed.
after recrystallization from dichloromethane/pentane
(ca. 1/1 v/v) the product is obtained as a stable, dark
red crystalline solid. Both ethylene molecules remain
bound after drying in vacuo overnight.
Resu lts a n d Discu ssion
Syn th esis of th e Ca tion ic Ru th en iu m Meth yl
Eth ylen e Com p lexes 3 a n d 4. The ruthenium complex
3 was synthesized in three steps, starting from [(p-
cymene)RuCl2]2.7 Refluxing a solution of the 2,6-
diketiminoylpyridine ligand 5a 3 and [(p-cymene)RuCl2]2
in ethanol8 under an ethylene atmosphere produces,
upon cooling, the ethylene complex 6 as a purple,
microcrystalline solid in 85% yield (eq 2). It should be
noted, however, that this reaction does not work when
the substituents in the ortho positions are too bulky.
For instance, if the ligand 5b derived from 2,6-diiso-
propylaniline is used instead of 5a , no reaction is
observed. Several attempts were made to synthesize
complexes of 5b from a wide variety of ruthenium
precursors, with little success.
The crystal structure of 7 reveals many details about
the ruthenium complex with this particular ligand that
otherwise would not have been apparent (Figure 1). The
geometry of 7 is best described as a distorted octahe-
dron. Using the nitrogen atoms N(5), N(13), and N(24)
along with the central ruthenium atom to define the
equatorial plane of the complex, the N(5)-Ru-Cl angle
of 82° indicates significant deviation from the expected
90° angle (Table 2). The ethylene molecules are also
bound in a nonstandard coordination geometry. Using
the centroid of C(1) and C(2), the N(5)-Ru-(C-C) angle
is 95.2°, and likewise the N(5)-Ru-(C-C) angle for C(3)
and C(4) is 167°sa deviation of 13° from the equatorial
plane! It is immediately apparent from space-filling
models that the o-methyl substituents on the ligand aryl
groups are at least in part responsible for the atypical
geometry, as the Ru-Cl bond length places the chlorine
atom directly in the most sterically demanding region
(7) Bennett, M. A.; Huang, T.-N.; Matheson, T. W.; Smith, A. K.
Inorg. Synth. 1982, 21, 75.
(8) (a) Cetinkaya, B.; Cetinkaya, E.; Brookhart, M.; White, P. S. J .
Mol. Catal. A 1999, 142, 101-112. (b) Bianchini, C.; Lee, H. M.
Organometallics 2000, 19, 1833-1840.

Alkyl-Olefin Complexes of Ru(II) and Rh(III)
Organometallics, Vol. 19, No. 24, 2000 4997
weak NOE to these methyl groups and a strong NOE
to the other aryl-methyl groups at 2.2 ppm, consistent
with the distorted geometry observed in the crystal
structure.
Alkylation of 7 was effected in dichloromethane using
(tmeda)MgMe2,9 favored due to its solubility in nonco-
ordinating solvents which are incapable of displacing
ethylene from the complex. Isolation of the crude
product, however, indicated that there was an equilib-
rium between the cationic ruthenium methyl ethylene
complex 3 and the ruthenium methyl ethylene chloro
complex (eq 4). To drive the equilibrium completely to
F igu r e 1. X-ray crystal structure of 7. Thermal ellipsoids
are drawn at the 50% probability level.
Ta ble 1. Su m m a r y of Cr ysta llogr a p h ic Da ta
for 7 a n d 9
7
9
formula
mol wt
BClC61F24H47N3Ru C36F6H46N3O6RhS2
1425.35
red
triclinic
P1h
10.0974(6)
16.7117(10)
18.8344(12)
105.224(1)
91.486(1)
97.609(1)
3033.5(3)
2
897.81
color
orange
cryst syst
space group
a (Å)
b (Å)
c (Å)
orthorhombic
Pcab
16.1396(10)
22.7895(15)
25.2719(16)
R (deg)
â (deg)
γ (deg)
V (Å3)
90
90
90
9295.3(10)
Z
8
Dcalcd (Mg m3)
wavelength (Å)
1.560
0.710 73
0.42
1.389
3, addition of 1 equiv of NaB(Ar′)4 was required (eq 5).
After evaporation of the solvent in vacuo, the product 3
was extracted from the magnesium salts at -78 °C, with
a solvent mixture of 1/4 v/v dichloromethane/pentane.
Upon removal of the solvent at room temperature, the
product is obtained as a brittle, dark green foam, in
relatively low yield (25%). Alternatively, the product can
be purified by column chromatography (Florisil) under
an inert atmosphere, generally giving much higher
yields (73%).
0.710 73
µ (mm-1
)
0.53
cryst dimens (mm)
temp (°C)
0.30 × 0.30 × 0.10
0.35 × 0.35 × 0.15
-100
-100
ω
mode
scan width (deg)
octants
ω
0.3
0.3
(h,k,l
(h,k,l
50.0
56 256
8233
6975
2θmax (deg)
no. of rflns
no. of unique rflns
no. of data with
I > 2.5σ
50.0
31 116
10 727
8797
R(F)
Rw(F)
0.050 (0.059)
0.063 (0.065)
0.039 (0.051)
0.041 (0.046)
2.960
largest peak in final 1.350
diff map (e Å-3
GOF
)
2.90
5.37
Ta ble 2. Selected Dista n ces a n d An gles for 7
Distances (Å)
Ru(1)-C(2)
Ru(1)-C(3)
Ru(1)-N(5)
2.229(4)
2.265(4)
1.992(3)
Ru(1)-N(24)
Ru(1)-Cl(1)
2.140(3)
2.4060(11)
Angles (deg)
N(5)-Ru(1)-Cl(1) 81.60(10) N(5)-Ru(1)-c[C(1)C(2)]
95
N(5)-Ru(1)-N(24) 77.40(13) N(5)-Ru(1)-c[C(3)C(4)] 167
of the ligand. In fact, the distance between the chlorine
atom and the methyl groups is shorter than the sum of
their respective van der Waals radii, explaining the
difficulty encountered when using ligand 5b, which
contains much bulkier isopropyl substituents.
Because the product tends to form an intractable oily
residue when exposed to solvents other than dichlo-
romethane, attempts to crystallize 3 have been unsuc-
cessful to date. The complex was therefore characterized
by NMR spectroscopy (Figure 3). As expected, the
ruthenium-methyl (1) resonance appears upfield at
-1.01 ppm in the 1H NMR spectrum. Two distinct
1
A 2D H NOESY measurement, which was acquired
to provide a complete assignment of the peaks in the
1D 1H NMR spectrum, provides evidence that this
geometry is also present in solution (Figure 2), ruling
out distortions due to crystal-packing forces. While the
ethylene molecule bound in the axial position (3.3 ppm)
displays a very strong NOE to the aryl-methyl groups
at 1.7 ppm as expected, the ethylene molecule bound in
the equatorial position (4.4 ppm) displays only a very
(9) Graser, T.; Kopf, J .; Thoennes, D.; Weiss, E. J . Organomet. Chem.
1980, 191, 1-6.

4998 Organometallics, Vol. 19, No. 24, 2000
Dias et al.
F igu r e 2. 1H NOESY of 7 acquired at room temperature in CD2Cl2. Significant cross-peaks: (A) ethylene (4)-ethylene
(5); (B) methyl (1)-ethylene (4); (C) methyl (2)-ethylene (5). The larger intensity of B compared to that of C provides an
assignment of the respective ethylene and methyl resonances. The very weak cross-peak from methyl (1)-ethylene (5) is
not shown for clarity.
resonances for the aryl-methyl groups (2, 3) are ob-
served at 1.38 and 1.95 ppm, and a single, bound
ethylene molecule (5) appears at 3.43 ppm. A 2D 1H
NOESY was also collected to establish that the complex
actually possesses the depicted square-pyramidal ge-
ometry (Figure 4). It is readily apparent that there is a
strong NOE between the ruthenium-methyl group (1)
and the bound ethylene (5), indicating that they are in
a cis arrangement. The ruthenium-methyl group (1)
also displays a strong NOE to only one set of aryl-
methyl resonances at 1.95 ppm (3), indicating that it is
in an axial position. Finally, the bound ethylene (5)
displays two cross-peaks of equal intensity to both aryl-
methyl resonances (2, 3), providing conclusive evidence
that it is bound in the equatorial position as illustrated.
(Figure 4).
The CO complex 4 is easily obtained by addition of 1
equiv of CO via syringe to a solution of 3 under an inert
atmosphere (eq 6). After 30 min, the solution changes
color from dark green to orange, and upon evaporation
of the solvent a brittle, dark orange foam is obtained
(similar to 3).
Complex 4 was easily verified to be a carbonyl
complex by both the CO stretch in the IR spectrum at
ν ) 2003 cm-1 (KBr) and the peak at 183.68 ppm in
the 13C NMR spectrum (CD2Cl2). As with complex 3, 4
was also characterized by 2D H NMR spectroscopy to
unambiguously establish its stereochemistry (Figure 5).
1
Sch em e 1
Because complex 3 is five-coordinate, it might be
expected that steric congestion around the Ru-methyl
group could be relieved if the aryl groups were canted,
i.e. not perpendicular to the plane of the ligand back-
bone, as depicted in Scheme 1. Rotation about the
N-C(ipso) bonds in the complex has the net result of
placing the aryl-methyl substituents (3) closer to, and
the aryl-methyl groups (2) further from, the ketimine
methyl groups (4) of the ligand backbone. Evidence for
this behavior is provided by the magnitudes of the
respective cross-peakssthe NOE between the backbone
methyl groups (4) and the aryl-methyl groups (3) is
noticably stronger than that for the methyl groups (2)

Alkyl-Olefin Complexes of Ru(II) and Rh(III)
Organometallics, Vol. 19, No. 24, 2000 4999
F igu r e 3. 1H NMR spectrum of 3 in CD2Cl2. Peak assignments 2 and 3 are based on the 2D NOESY spectrum (Figure 4).
C-C bond formation. Even after the mixture is stirred
for several hours under 400 psi of ethylene at room
temperature, there is no evidence of either polymer or
R-olefins. Extended heating at 80 °C in chlorobenzene,
on the other hand, results in the very slow production
of 1-butene. Because 3 is not stable at this temperature,
the identity of the active species is not clear, given that
butene production continues after all of the complex has
apparently decomposed to unidentified products. Pho-
tolysis of a dichloromethane solution of 3 under an
ethylene atmosphere only results in formation of the
precursor complexes 6 and 7. The use of other solvents
is severely limited by the tendency of 3 to produce an
intractable oil when exposed to nonpolar, nonchlori-
nated solvents. When using toluene or hexanes, how-
ever, there is still no polymer obtained.
The NOE between the methyl group and the bound
ethylene is very strong, confirming their cis arrange-
ment around the metal center. In addition, the NOE’s
between the ethylene and the aryl methyl groups do not
have the same intensity, as was observed for 7. J udging
by the relative intensities of these cross-peaks, however,
it can be concluded that while this complex is also a
distorted octahedron, the deviations from ideal geometry
are not as severe as with 7.
Attem p ted P olym er iza tion s Usin g 3. Because of
the remarkable stability of 3 toward methyl migration
to ethylene at room temperature, it was not expected
that this complex would be a particularly active polym-
erization catalyst. When a dichloromethane solution of
3 is heated to 70 °C (in a closed system) under an
ethylene atmosphere, however, there is still no apparent
Complex 3 was screened for activity in the copolym-
erization of ethylene/CO as well. Upon exposure of a
dichloromethane solution of 3 to 1 atm of ethylene/CO,
the color immediately changes from dark green to
orange, and no polymer is obtained after extended
periods of time. Presumably, the orange product is 4,
resulting from coordination of carbon monoxide to 3.
Syn th esis of Isoelectr on ic Rh od iu m (III) Com -
p lexes. Because of the surprising inactivity of 3 toward
olefin polymerization, we also decided to explore its
isolectronic rhodium(III) analogue. Because the rhod-
ium(III) complex would be dicationic, it was expected
to be more electrophilic and therefore more reactive
than 3. Given that Wang and Flood have demonstrated

5000 Organometallics, Vol. 19, No. 24, 2000
Dias et al.
F igu r e 4. 1H NOESY of 3 acquired at room temperature in CD2Cl2. Significant cross-peaks: (A) methyl (1)-methyl (3);
(B) methyl (1)-ethylene (5); (C) ethylene (5)-methyl (2) and ethylene (5)-methyl (3); (D) methyl (4)-methyl (2) and
methyl (4)-methyl (3). The difference in the relative intensities of cross-peaks D provide evidence for twisting of the aryl
groups around the N-C(ipso) bonds (Scheme 1). The cross-peaks C (not enlarged) are of approximately equal intensity.
Ta ble 3. Selected Dista n ces a n d An gles for 9
Distances (Å)
Rh(1)-C(7)
Rh(1)-N(31)
Rh(1)-N(5)
2.026(3)
1.911(3)
2.091(3)
Rh(1)-O(11)
Rh(1)-O(21)
2.0977(22)
2.3578(21)
Angles (deg)
C(7)-Rh(1)-N(31) 91.37(12) C(4)-N(5)-C(61)-C(62) -71.9(5)
N(5)-Rh(1)-N(31) 79.92(11) C(1)-N(2)-C(41)-C(46) 76.0(5)
that the complexes (Cn)Rh(Me)(X)2 (Cn ) trimethyltri-
azacyclononane, X ) OTf, BF4) are effective initiators
for the polymerization of ethylene,10 the analogous
rhodium complexes employing the 2,6-diketiminoylpy-
ridine ligands 5 were attractive targets as polymeriza-
tion initiators. The bis(triflate) complex 9 was readily
11
synthesized in two steps from [Rh(C2H4)2Cl]2 via the
compound [(5b)RhCl] (8), as shown in eq 7.12
The crystal structure of 9 (Figure 6) reveals a few
details worth noting. One significant feature is the
difference between the Rh-O bond lengths for the two
triflate groups. While the triflate bound cis to the methyl
group has a normal Rh-O bond length of 2.098 Å (Table
3), the triflate that is bound trans to the methyl group
has a very long Rh-O bond (2.358 Å). Because of the
electronic trans influence of the methyl group in con-
junction with the steric crowding imposed by the iso-
propyl groups, this triflate might better be described as
a contact ion, with the rhodium metal center being five-
coordinate. In accordance with this description, the aryl
rings on the imines are twisted dramatically to accom-
modate the rhodium-bound methyl group, making the
opposite site much less accessiblesthe same torsional
configuration that is proposed for 3 on the basis of
spectroscopic data.
(10) (a) Wang, L.; Lu, R. S.; Bau, R.; Flood, T. C. J . Am. Chem. Soc.
1993, 115, 6999-7000. (b) Wang, L.; Flood, T. C. J . Am. Chem. Soc.
1992, 114, 3169-3170.
(11) Cramer, R. Inorg. Synth. 1974, 15, 14-16.
(12) Haarmann, H. F.; Ernsting, J . M.; Kranenburg, M.; Kooijman,
H.; Veldman, N.; et al. Organometallics 1997, 16, 887-900.

Alkyl-Olefin Complexes of Ru(II) and Rh(III)
Organometallics, Vol. 19, No. 24, 2000 5001
F igu r e 5. 1H NOESY of 4 acquired at room temperature in CD2Cl2. Significant cross-peaks: (A) methyl (1)-methyl (2),
(B) methyl (1)-ethylene (5); (C) ethylene (5)-methyl (2) and ethylene (5)-methyl (3); (D) methyl (4)-methyl (2) and
methyl (4)-methyl (3). The difference in intensities of cross-peaks C indicate deviation from ideal octahedral geometry, as
observed in 7. The relatively equal intensities of cross-peaks D suggest that the aryl rings remain approximately
perpendicular to the ligand backbone.
exchanging the remaining triflate, a separate route was
adopted for the synthesis of a dicationic, rhodium(III)
methyl ethylene complex that would be isoelectronic and
isostructural with the ruthenium complex 3.
The first attempted route to a dicationic rhodium
methyl ethylene complex is shown in eq 8. The cationic
F igu r e 6. X-ray crystal structure of 9. Thermal ellipsoids
are drawn at the 50% probability level.
In contrast to what one would expect, reaction of 9
with Na(BAr′)4 exchanges the triflate that is cis to the
methyl group, as confirmed by a crystal structure of the
corresponding aquo complex 13 (see the Supporting
Information). Because of the problems encountered in

5002 Organometallics, Vol. 19, No. 24, 2000
Dias et al.
ethylene complex 10 is readily synthesized by the
reaction of 8 with NaB(Ar′)4 under an ethylene atmo-
sphere. Alkylation of 10 with methyl iodide could only
be effected upon heating a dichloromethane solution to
45 °C for 4 h, during which time the color changes from
orange to an intensely dark reddish pink. Upon removal
of the solvent, the only product isolated is the five-
coordinate complex 11, in which the iodide has appar-
ently displaced ethylene from the metal center. The
crystal structure of 11 (Supporting Information) indi-
cates that the methyl and iodide substituents are in a
cis geometry, as expected, with the iodide trans to the
pyridine moiety. In addition, the aryl rings are slightly
canted to relieve steric congestion at the methyl-bound
position, as is observed for both 3 and 9.
al.,10 these systems once again do not exhibit any
activity for the polymerization of ethylene. Because the
ligand 5b and triazacyclonane are expected to be fairly
similar electronically, the difference must be due to the
geometry imposed by the ligandssi.e., fac-triazacy-
clononane versus mer-5b. The meridional geometry of
these ligands necessarily differentiates the remaining
binding sites around the metal center, creating distinct
axial (cis to pyridine) and equatorial (trans to pyridine)
positionssa feature that is lacking in both the diimine
(Ni,Pd)2 and triazacyclononane (Rh)10 systems. By anal-
ogy, this may also be responsible for the inactivity
observed with the ruthenium complex 3. Experimen-
tally, it has been observed that both the rhodium and
ruthenium complexes are exceptionally stable with the
alkyl and ethylene substituents in the axial and equato-
rial sites, respectively. Because migration of the methyl
group to the ethylene is expected to form a species in
which the newly formed metal-alkyl bond is in the
equatorial position, the stability of such a species must
now be considered as well.
The dicationic methyl ethylene complex 12, containing
two BF4 anions, was finally obtained by the reaction of
11 with excess AgBF4 under an ethylene atmosphere
(eq 9). Complex 12 was not isolated as a pure solid,
In a separate paper, ab initio calculations of both the
ruthenium and iron systems which support theseem-
pirical conclusions will be presented.13 The electronic
nature of these ligands, in conjunction with the geom-
etry of the intermediates, has unforeseen consequences
upon the alkyl migration step. In particular, migration
of an alkyl group from a position cis to the pyridine
moiety to one trans to it is extremely unfavorable
energetically in the case of rutheniumsa situation
which may be exacerbated by the unnaturally short
Ru-N(pyridine) bond. For iron, the accessibility of
higher multiplicity spin states may serve to alleviate
this problem. Further details regarding this unforeseen
difference in reactivities will be forthcoming.
Finally, these results may have implications regard-
ing the nature of the active species for the (pybox)Ru-
(C2H4)Cl2/MAO polymerization system reported re-
cently.14 Given the quite small amounts of high-
molecular-weight polyethylene produced in this system,
the question of whether a ruthenium pybox alkyl
complex (similar to 3) is actually responsible for this
activity naturally arises.15
although it could be observed spectroscopically (Sup-
porting Information).
Attem pted P olym er ization s u sin g Rh odiu m Com -
p lexes 9 a n d 12. As with the ruthenium complexes,
the triflate complex 9 does not produce polyethylene,
or even butene, under conditions similar to those used
for 7. In the case of 12, which was generated in situ
under an ethylene atmosphere, no further reaction takes
place, even after extended heating at 40 °C.
Exp er im en ta l Section
Con clu sion s
All solvents used were dried according to the Grubbs/Dow
procedure,16 and all manipulations were done either in a
drybox or under argon or ethylene atmosphere utilizing
standard Schlenk and cannula techniques, unless otherwise
stated. NMR samples were prepared under an inert atmo-
sphere, utilizing CD2Cl2 that was dried over CaH2 and vacuum-
transferred before use. NMR spectra were recorded on a
Bruker 400 MHz spectromoter at room temperature, unless
otherwise noted. [Ru(p-cymene)Cl2]2,7 [Rh(C2H4)2Cl]2,11 the
Although iron and ruthenium are both group VIII
transition metals, it is immediately obvious that there
are significant differences between them. The iron
dichloride complexes of the ligands 5a and 5b are five-
coordinate, possessing a square-pyramidal geometry in
which the halides are in a cis configuration.3,4 In the
case of ruthenium, however, these complexes are always
octahedral, with the halides in a trans geometry.8
Electronically, the iron complexes are commonly high-
spin,3,4 while their ruthenium counterparts are inevi-
tably low-spin. These are mainly empirical observations
regarding these two systems, but they may have far-
reaching effects regarding reactivity, as evidenced by
the lack of polymerization activity for the ruthenium
complex 3.
(13) Prosenc, M. H.; Brookhart, M. Manuscript in preparation.
(14) Nomura, K.; Warit, S.; Imanishi, Y. Macromolecules 1999, 32,
4732-4734.
(15) In ref 1a, p 1191, it was mistakenly indicated that “attempts
to reproduce this work were unsuccessful”. We do not doubt the
reproducibility of these results and regret this misstatement. We meant
to convey that by analogy with previous transition-metal halide/MAO
polymerization systems, the active species is expected to be a cationic
Ru pybox alkyl ethylene complex; however, given the inactivity of 3,
it is likely that there is a trace amount of a different active species
that is responsible for the observed polymerization activity.
(16) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;
Timmers, F. J . Organometallics 1996, 15, 1518-1520.
The analogous rhodium complexes 9 and 12 provided
even more of a surprise. Despite their overall similarity
to the triazacylononane complexes studied by Flood et

Alkyl-Olefin Complexes of Ru(II) and Rh(III)
Organometallics, Vol. 19, No. 24, 2000 5003
ligand 5a ,3a and (tmeda)MgMe2 were synthesized according
weighed into a vial and dissolved in 1 mL of CH2Cl2, which
was placed within a larger vial containing 10 mL of pentane.
The assembly was sealed, and the crystals were grown over
several days at room temperature.
9
to published procedures. NaB(Ar′)4 was purchased from Boul-
der Scientific and used without further purification. Elemental
analyses were done by Atlantic Microlabs.
Syn th esis of [(5a )Ru (C2H4)(CH3)][B(Ar ′)4] (3). Inside the
drybox, 500 mg of 7 (0.35 mmol, 1 equiv), 50 mg of (tmeda)-
MgMe2 (0.29 mmol), and 320 mg of NaB(Ar′)4 (0.36 mmol, 1.03
equiv) were weighed into three separate Schlenk flasks. The
flasks were capped with rubber septa and placed under an
ethylene atmosphere on a Schlenk line. CH2Cl2 (10 mL) was
added to both 7 and the (tmeda)MgMe2 via syringe, and the
solution of 7 was added to the (tmeda)MgMe2 solution via
cannula. The reaction mixture turns dark green upon addition
and was stirred for 30 min at room temperature. CH2Cl2 (5
mL) was added to the Na(BAr′)4 flask, and the reaction
mixture was added to the resulting slurry via cannula. After
the mixture was stirred for 2 h at room temperature, the solids
were removed by cannula filtration and washed with CH2Cl2
and the combined filtrate was dried in vacuo to give an oily,
dark green-brown residue. The crude product can be purified
by two methods.
Meth od 1. The crude product was dissolved in 4 mL of CH2-
Cl2, and the solution was cooled to -78 °C in a dry ice/acetone
bath. A 16 mL amount of pentane was added via syringe, and
the mixture was stirred at -78 °C for 20 min to precipitate a
dark brown solid. The dark green supernatant is collected by
cannula filtration at -78 °C and the solvent removed in vacuo.
The product obtained in this manner is a dark green, brittle
foam, obtained in 27% yield.
Crystallographic data were collected at -100 °C on a Bruker
SMART 1K diffractometer using Mo KR radiation. Initial cell
dimensions and orientation matrices were determined and the
frames used to evaluate the quality of the crystals. A full
sphere of intensity data were collected with the CCD set to
the appropriate maximum 2θ (50-60°). The Bruker routine
SAINT was used to integrate the frames and produce a set of
intensities, which were corrected for absorption effects using
SADABS. All subsequent computations were performed using
the NRCVAX17a suite of programs using scattering factors
taken from ref 17b. The structures were solved using direct
methods to locate the non-hydrogen atoms. Experimental data
are summarized in the Supporting Information.
Syn th esis of [(5a )Ru (C2H4)Cl2] (6). A 500 mg amount of
[Ru(p-cymene)Cl2]2 (0.816 mmol, 1 equiv) and 618 mg of 5a
(1.67 mmol, 2 equiv) were added to a Schlenk flask connected
to a Schlenk line under an ethylene atmosphere. The flask
was evacuated and refilled with ethylene, and 25 mL of
degassed, absolute ethanol was added. The flask was equipped
with a reflux condenser vented to an oil bubbler, purged for
15 min, and then heated to gentle reflux with stirring for 12
h. During this time, the product precipitated as a purple,
microcrystalline solid. The reaction mixture was cooled to room
temperature and the supernatant removed by cannula filtra-
tion. After it was dried overnight in vacuo, the product was
extracted from insoluble Ru0 impurities using CH2Cl2. Con-
Meth od 2. The crude product was dissolved in 5 mL of CH2-
Cl2 and loaded via cannula onto a Florisil column (5 cm) under
argon. The product was eluted with approximately 50 mL of
CH2Cl2, and the dark green band was collected in a Schlenk
flask. The solvent was removed in vacuo, and the product was
isolated as a dark green, brittle foam in 73% yield (350 mg).
The product should be stored in a drybox at low temperatures
1
centration of the filtrate yielded 770 mg of 6 (83% yield). H
NMR (CD2Cl2): δ 8.163 (ABB′ d, 2H, pyridine Hmeta), 8.02
(ABB′ t, 1H, pyridine Hpara), 7.05-7.11 (m, 6H, 2,6-(CH3)2-
C6H3), 4.45 (s, 4H, C2H4), 2.67 (s, 6H, NdCCH3), 2.15 (s, 12H,
2,6-(CH3)2-C6H3). 13C{1H} NMR: δ 174.9, 157.2, 148.58, 132.1,
131.1, 128.8, 126.4, 125.3, 82.7, 21.0, 19.0.
Syn th esis of [(5a )Ru (C2H4)2Cl][B(Ar ′)4] (7). Inside a
drybox, 500 mg of 6 (0.878 mmol, 1 equiv) and NaB(Ar′)4 (0.878
mmol, 1 equiv) were weighed into a Schlenk flask, which was
capped with a rubber septum and placed under an ethylene
atmosphere on a Schlenk line. CH2Cl2 (20 mL) was added via
syringe, and the reaction mixture was stirred at room tem-
perature for 2.5 h. The solution was transferred to a separate
Schlenk flask via cannula filtration, and the remaining NaCl
was washed with 2 × 5 mL of CH2Cl2. The combined fractions
were dried in vacuo to give a dark brown solid.
1
to prevent decomposition. H NMR (CD2Cl2): δ 8.10 (ABB′ d,
2H, pyridine Hmeta), 8.01 (ABB′ t, 1H, pyridine Hpara), 7.71 (br
m, 8H, B(Ar′)4 Hortho), 7.55 (br s, 4H, B(Ar′)4 Hpara), 7.14-7.27
(m, 6H, 2,6-(CH3)2-C6H3), 3.43 (s, 4H, C2H4), 2.50 (s, 6H, Nd
CCH3), 1.95 (s, 6H, 2,6-(CH3)2-C6H3), 1.38 (s, 6H, 2,6-(CH3)2-
C6H3), -1.01 (s, 3H, Ru-CH3). 13C{1H} NMR: δ 174.6, 162.0
(q, 1J BC ) 50 Hz, B(Ar′)4), 155.8, 148.7, 135.1 (s, B(Ar′)4), 133.2,
2
129.7, 129.1 (qm, J CF ) 32 Hz, B(Ar′)4), 129.1, 128.2, 127.7,
1
127.6, 124.9 (q, J CF ) 272 Hz, B(Ar′)4), 124.8, 117.8 (m,
B(Ar′)4), 78.8, 18.9, 18.6, 18.4, -4.8. Anal. Calcd: C, 52.34; H,
3.37; N, 3.05. Found: C, 52.10; H, 3.33; N, 3.17.
Under argon, the solid was dissolved in 10-15 mL of
CH2Cl2, and an equal volume of pentane was added via
syringe. When the mixture was stirred, the product precipi-
tated as a dark red crystalline solid. The flask was cooled to 0
°C in an ice bath to obtain more crystals, and the product was
isolated by cannula filtration. Inside the drybox, 1.075 g of 7
was isolated (86% yield). The product is stable under an inert
atmosphere at room temperature or below but tends to slowly
decompose at higher temperatures. 1H NMR (CD2Cl2): δ 8.45
(ABB′ t, 1H, pyridine Hpara), 8.33 (ABB′ d, 2H pyridine Hmeta),
7.72 (br m, 8H, B(Ar′)4 Hortho), 7.55 (br s, 4H, B(Ar′)4 Hpara),
7.07-7.17 (m, 6H, 2,6-(CH3)2-C6H3), 4.38 (s, 4H, C2H4), 3.35
(s, 4H, C2H4), 2.51 (s, 6H, NdCCH3), 2.24 (s, 6H, 2,6-(CH3)2-
C6H3), 1.82 (s, 6H, 2,6-(CH3)2-C6H3). 13C{1H} NMR: δ 180.3,
Syn th esis of [(5a )Ru (C2H4)(CH3)(CO)][B(Ar ′)4] (4). In-
side the drybox, 50 mg of 3 (0.36 mmol, 1 equiv) was weighed
into a Schlenk flask, which was capped with a rubber septum
and placed on a Schlenk line under argon. CH2Cl2 (5 mL) was
added via syringe, and the stopcock to the argon inlet was
closed. CO (1 mL, 1.1 equiv) was added via a gastight syringe,
and the reaction mixture was stirred for 30 min at room
temperature, during which time the solution changed color
from dark green to orange. The solvent was removed in vacuo,
1
leaving the product 4 as a brittle foam. H NMR (CD2Cl2): δ
8.23 (m, 3H, pyridine Hmeta, Hpara), 7.72 (br m, 8H, B(Ar′)4
Hortho), 7.55 (br s, 4H, B(Ar′)4 Hpara), 7.18-7.22 (m, 6H, 2,6-
(CH3)2-C6H3), 3.35 (s, 4H, C2H4), 2.49 (s, 6H, NdCCH3), 2.07
(s, 6H, 2,6-(CH3)2-C6H3), 2.02 (s, 6H, 2,6-(CH3)2-C6H3), -0.57
(s, 3H, Ru-CH3). 13C{1H} NMR: δ 183.7 (CO), 175.5, 162.0
(q, 1J BC ) 50 Hz, B(Ar′)4), 154.1, 148.0, 136.9, 135.1 (s, B(Ar′)4),
1
162.1 (q, J BC ) 50 Hz, B(Ar′)4), 160.0, 146.6, 140.2, 135.2 (s,
B(Ar′)4), 131.0, 130.5, 130.0, 129.2 (qm, 2J CF ) 32 Hz, B(Ar′)4),
1
128.9, 128.0, 126.8, 125.0 (q, J CF ) 272 Hz, B(Ar′)4), 117.9
2
130.1, 129.7, 129.3, 129.1 (qm, J CF ) 32 Hz, B(Ar′)4), 128.1,
(m, B(Ar′)4), 86.2, 74.7, 20.5, 20.0, 19.0. Anal. Calcd: C, 51.40;
H, 3.32; N, 2.95. Found: C, 51.36; H, 3.29; N, 2.94.
1
127.4, 127.3, 124.9 (q, J CF ) 272 Hz, B(Ar′)4), 117.8 (m,
B(Ar′)4), 75.7, 20.0 (Ru-CH3), 19.2, 18.8, 18.4. IR (KBr): ν(CO)
2003 cm-1. Anal. Calcd: C, 52.15; H, 3.30; N, 2.99. Found: C,
52.05; H, 3.33; N, 3.02.
Crystals of 7 suitable for X-ray diffraction were grown in a
diffusion chamber inside the drybox. Complex 7 (25 mg) was
Syn th esis of Liga n d 5b. A 2 g (12.26 mmol, 1 equiv)
amount of 2,6-diacetylpyridine was weighed into a 100 mL
round-bottom flask and dissolved in 50 mL of reagent grade
2-propanol. A 6 mL (5.64 g, 28.62 mmol, 2.3 equiv) portion of
(17) (a) Gabe, E. J .; Le Page, Y.; Charland, J .-P.; Lee, F. L.; White,
P. S. J . Appl. Crystallogr. 1989, 22, 384-387. (b) International Tables
for X-ray Crystallography; Kynoch Press: Birmingham, England, 1974;
Vol. IV.

5004 Organometallics, Vol. 19, No. 24, 2000
Dias et al.
90% 2,6-diisopropylaniline was added by pipet, followed by a
catalytic amount (10-20 mg) of camphorsulfonic acid. The
flask was equipped with a reflux condenser, and the reaction
mixture was heated to a vigorous reflux for 30 h, during which
time the product precipitated as a light yellow powder. While
it was still warm, the mixture was filtered and the product
was collected on a Bu¨chner funnel. (Note: It is important to
filter the reaction mixture while it is still warm. When the
supernatant cools to room temperature or below, the mono-
ketimine begins to precipitate out of solution. In fact, this
product often precipitates as a pure solid out of the filtrate
during vacuum filtration.) After drying in vacuo, 5.07 g of 5b
was obtained (86% yield) and used without further purifica-
tion.
Syn th esis of [(5b)Rh Cl] (8). Inside the drybox, 1 g (2.57
mmol, 1 equiv) of [Rh(C2H4)2Cl]2 and 2.48 g (5.14 mmol, 2
equiv) of 5b were weighed into a Schlenk flask, which was
capped with a rubber septum and placed on a Schlenk line
under argon. Toluene (50 mL) was added via syringe, and the
reaction mixture was stirred for 24 h at room temperature,
during which time the product precipitated as a dark green,
microcrystalline solid. The product was isolated by cannula
filtration and dried in vacuo. A total of 3.02 g of compound 8
with CH2Cl2. After the solvent was removed in vacuo, 530 mg
of 10 was obtained as a brown, microcrystalline solid (89%
1
yield). H NMR (CD2Cl2): δ 8.19 (AA′B t, 1H, pyridine Hpara),
7.85 (AA′B d, 2H, pyridine Hmeta), 7.71 (br m, 8H, B(Ar′)4 Hortho),
7.55 (br s, 4H, B(Ar′)4 Hpara), 7.22-7.41 (m, 6H, 2,6-(C3H7)2-
2
3
C6H3), 3.66 (d, J RhH ) 1 Hz, 4H, Rh-C2H4), 2.97 (sept, J HH
) 7 Hz, 4H, ArCH(CH3)2), 2.03 (s, 6H, NdCCH3), 1.27 (d, 3J HH
) 7 Hz, 12H, ArCH(CH3)2), 1.09 (3J HH ) 7 Hz, 12H, ArCH-
(CH3)2). 13C{1H} NMR: δ 174.8 (d, J RhC ) 2 Hz), 162.0 (q,
2
1J BC ) 50 Hz, B(Ar′)4), 155.6 (d, J RhC ) 4 Hz), 141.8, 140.2,
2
2
139.1, 135.1 (s, B(Ar′)4), 129.1 (qm, J CF ) 32 Hz, B(Ar′)4),
1
128.9, 125.7, 125.0, 124.9 (q, J CF ) 272 Hz, B(Ar′)4), 117.8
(m, B(Ar′)4), 82.8 (d, 2J RhC ) 10 Hz, Rh-C2H4), 28.7, 24.5, 23.3,
19.4 (d, 3J RhC ) 2 Hz). Anal. Calcd: C, 54.52; H, 4.03; N, 2.85.
Found: C, 54.44; H, 4.02; N, 2.82.
Syn th esis of [(5b)Rh (CH3)(I)][B(Ar ′)4] (11). Inside the
drybox, 250 mg (0.169 mmol, 1 equiv) of 10 was weighed into
a Schlenk flask, which was capped with a rubber septum and
placed on a Schlenk line under argon. CH2Cl2 (15 mL) was
added to dissolve the compound, followed by 25 µL (57 mg,
0.402 mmol, 2.4 equiv) of methyl iodide. The septum was
replaced with a glass stopper, and the reaction mixture was
heated at 45 °C in a closed system for 4 h, during which time
the color changed from orange to dark red-pink. The reaction
mixture was cooled to room temperature and evaporated to
dryness in vacuo. A total of 265 mg of product 11 was isolated
as a brittle foam and used without further purification (98%
yield). Crystals for X-ray diffraction were grown in a manner
1
3
was isolated (95% yield). H NMR (CD2Cl2): δ 8.51 (t, J HH
)
3
8 Hz, 1H, pyridine Hpara), 7.70 (d, J HH ) 8 Hz, 2H, pyridine
3
Hmeta), 7.18-7.31 (m, 6H, 2,6-(C3H7)2-C6H3), 3.02 (sept, J HH
) 7 Hz, 4H, ArCH(CH3)2), 1.64 (s, 6H, NdCCH3), 1.12 (d, 3J HH
3
) 7 Hz, 12H, ArCH(CH3)2), 1.03 (d, J HH ) 7 Hz, 12H, ArCH-
1
(CH3)2). 13C{1H} NMR: δ 168.1 (d, J RhC ) 3 Hz), 156.5 (d,
2
similar to that for 7. H NMR (CD2Cl2): δ 8.43 (AA′B t, 1H,
2J RhC ) 3 Hz), 146.2, 140.8, 126.7, 125.3, 124.0, 123.5, 28.5,
pyridine Hpara), 8.12 (AA′B d, 2H, pyridine Hmeta), 7.73 (br m,
8H, B(Ar′)4 Hortho), 7.56 (br s, 4H, B(Ar′)4 Hpara), 7.24-7.42 (m,
3
23.7, 23.6, 17.9 (d, J RhC ) 2 Hz). Anal. Calcd: C, 63.92; H,
3
6H, 2,6-(C3H7)2-C6H3), 2.43 (s, 6H, NdCCH3), 2.37 (sept, J HH
6.99; N, 6.78. Found: C, 63.74; H, 6.97; N, 6.71.
) 7 Hz, 4H, ArCH(CH3)2), 2.16 (d, 2J RhH ) 3 Hz, 3H, Rh-CH3),
Syn th esis of [(5b)Rh (CH3)(OTf)2] (9). Inside the drybox,
250 mg (0.403 mmol, 1 equiv) of 8 was weighed into a Schlenk
flask, which was capped with a rubber septum and placed on
a Schlenk line under argon. CH2Cl2 (10 mL) was added via
syringe, followed by 250 µL (572 mg, 4.03 mmol, 10 equiv) of
methyl iodide. The reaction mixture was stirred for 30 min at
room temperature, during which time the solution turned from
green to dark orange. Under a positive flow of argon, 210 mg
(0.806 mmol, 2 equiv) of silver triflate was added as a solid,
and the slurry was stirred for 3 h at room temperature, during
which time the solution became light orange. The filtrate was
collected by cannula filtration, and the remaining solid was
extracted with approximately 50 mL of CH2Cl2. The combined
fractions were evaporated to dryness in vacuo, and the solid
obtained was washed with toluene, followed by pentane. After
drying, 330 mg of 9 was isolated as a light orange powder (91%
yield) and used without further purification. If necessary, 9
can be recrystallized from dichloromethane/pentane. Crystals
for X-ray diffraction were grown in a manner similar to that
3
3
1.27 (d, J HH ) 7 Hz, 6H, ArCH(CH3)2), 1.24 (d, J HH ) 7 Hz,
6H, ArCH(CH3)2), 1.19 (d, 3J HH ) 7 Hz, 6H, ArCH(CH3)2), 1.02
(d, J HH ) 7 Hz, 6H, ArCH(CH3)2). 13C{1H} NMR: δ 182.3,
3
1
162.0 (q, J BC ) 50 Hz, B(Ar′)4), 156.7, 143.3, 140.1, 139.2,
139.0, 135.1 (s, B(Ar′)4), 129.8, 129.6, 129.1 (qm, 2J CF ) 32 Hz,
1
B(Ar′)4), 124.9 (q, J CF ) 272 Hz, B(Ar′)4), 124.9, 124.7, 117.8
(m, B(Ar′)4), 32.1, 29.2, 24.9, 24.7, 23.7, 22.8, 20.3, 8.9 (d, 1J RhC
) 25 Hz, Rh-CH3). Anal. Calcd: C, 49.86; H, 3.68; N, 2.64.
Found: C, 49.87; H, 3.68; N, 2.55.
Gen er a tion of [(5b)Rh (CH3)(C2H4)](BF 4)2 (12). In a J .
Young NMR tube, 10 mg of 11 was dissolved in 0.5 mL of CD2-
Cl2 and an excess of AgBF4 was added (5-7 mg). The tube
was immediately frozen in liquid nitrogen and evacuated on
a Schlenk line. After it was thawed, the reaction mixture was
placed under an ethylene atmosphere. While three distinct
species (two major, one minor) are observed initially, only one
remains after heating for 1 h at 40 °C, presumed to be 12 due
to the disappearance of the B(Ar′)4 signals. 1H NMR (CD2Cl2):
δ 8.68 (AA′B t, 1H, pyridine Hpara), 8.43 (AA′B d, 2H, pyridine
1
for 7. H NMR (CD2Cl2): δ 8.38 (AA′B t, 1H, pyridine Hpara),
H
H
meta), 7.47 (AA′B t, 2H, aryl Hpara), 7.39-7.34 (m, 4H, aryl
8.09 (AA′B d, 2H, pyridine Hmeta), 7.18-7.41 (m, 6H, 2,6-
meta), 4.29 (d, 4H, 2J RhH ) 1 Hz), 3.21 (sept, 2H, 3J HH ) 7 Hz,
(C3H7)2-C6H3), 3.62 (sept, 3J HH ) 7 Hz, 2H, ArCH(CH3)2), 2.54
3
3
ArCH(CH3)2), 2.67 (s, 6H, NdCCH3), 2.66 (sept, J HH ) 7 Hz,
(s, 6H, NdCCH3), 2.52 (sept, J HH ) 7 Hz, 2H, ArCH(CH3)2),
2
2.16 (d, 2J RhH ) 2 Hz, 3H, Rh-CH3), 1.33 (d, 3J HH ) 7 Hz, 6H,
2H, ArCH(CH3)2), 1.80 (d, J RhH ) 2 Hz, 3H, Rh-CH3), 1.28-
3
3
1.19 (m, 18H, ArCH(CH3)2), 1.03 (d, J HH ) 7 Hz, 6H, ArCH-
ArCH(CH3)2), 1.27 (d, J HH ) 7 Hz, 6H, ArCH(CH3)2), 1.21 (d,
3J HH ) 7 Hz, 6H, ArCH(CH3)2), 1.00 (d, 3J HH ) 7 Hz, 6H, ArCH-
(CH3)2). 19F NMR: δ -80.4 (s), -81.6 (s). 13C{1H} NMR (triflate
carbons not located): δ 181.0, 159.6, 143.4, 140.8, 140.5, 139.9,
129.3, 128.9, 125.8, 125.3, 29.1, 28.9, 26.3, 25.3, 25.0, 21.9, 20.5,
7.4 (d, 1J RhC ) 25 Hz, Rh-CH3). Anal. Calcd: C, 48.16; H, 5.16;
N, 4.68. Found: C, 48.34; H, 5.19; N, 4.62.
(CH3)2).
Ack n ow led gm en t. We wish to thank DuPont for
funding and Prof. J . L. Templeton for use of laboratory
space.
Syn th esis of [(5b)Rh (C2H4)][B(Ar ′)4] (10). Inside the
drybox, 250 mg (0.403 mmol, 1 equiv) of 8 and 357 mg (0.403
mmol, 1 equiv) of NaB(Ar′)4 were weighed into a Schlenk flask,
which was capped with a rubber septum and placed on a
Schlenk line under an ethylene atmosphere. CH2Cl2 (15 mL)
was added via syringe, and the reaction mixture was stirred
for 4 h at room temperature, during which time the color
changed from green to dark orange. The filtrate was collected
by cannula filtration, and the remaining NaCl was washed
Su p p or tin g In for m a tion Ava ila ble: Figures giving the
1H NMR spectrum of compound 12 and X-ray crystal struc-
tures of compounds 11 and 13 and tables giving atomic
coordinates and isotropic displacement coefficients, anisotropic
thermal parameters, bond distances, angles, and packing
diagrams for all of the crystal structures. This material is
available free of charge via the Internet at http://pubs.acs.org.
OM000674I