Structural and electronic properties of (2,2-trans)-dirhodium(II) tetrakis(N-phenylacetamidate)

Article abstract of DOI:10.1016/S0022-328X(99)00556-2

We have synthesized and characterized the first known 2,2-trans isomer of the N-substituted dirhodium(II) tetrakisacetamidate, Rh₂(RNAc)₄, class of compounds. The bis benzonitrile adduct exhibits a unique orthogonal arrangement of the axial aromatic rings in the solid state. Structural and electronic features suggest the presence of π-backbonding.

Full text of DOI:10.1016/S0022-328X(99)00556-2

www.elsevier.nl/locate/jorganchem
Journal of Organometallic Chemistry 596 (2000) 90–94
Structural and electronic properties of (2,2-trans)-dirhodium(II)
tetrakis(N-phenylacetamidate)
Cassandra T. Eagle a,1, David G. Farrar a,*, Grant N. Holder a,
William T. Pennington b, Rosa D. Bailey b
a Department of Chemistry, Appalachian State Uni6ersity, Boone, NC 28608, USA
b Department of Chemistry, Clemson Uni6ersity, Clemson, SC 29634-1905, USA
Received 29 July 1999
Abstract
We have synthesized and characterized the first known 2,2-trans isomer of the N-substituted dirhodium(II) tetrakisacetamidate,
Rh2(RNAc)4, class of compounds. The bis benzonitrile adduct exhibits a unique orthogonal arrangement of the axial aromatic
rings in the solid state. Structural and electronic features suggest the presence of p-backbonding. © 2000 Elsevier Science S.A. All
rights reserved.
Keywords: Aromatic rings; p-Backbonding; Rhodium acetate
boxamidate) complexes can form four different iso-
meric structures: 2,2-cis (I); 2,2-trans (II); 3,1 (III); and
4,0 (IV).
1. Introduction
Rhodium acetate was first discovered to catalyze
carbenoid transformations more than 25 years ago [1].
Since that discovery, the dirhodium(II) tetrakis(car-
boxylate) core has been extensively modified to develop
selective carbenoid catalysts [2]. Such modifications
have included altering the electronic demand of the
metal, making electron-deficient catalysts by using per-
fluorocarboxylate ligands and electron-rich catalysts
through the use of carboxamidate ligands [3]. Other
modifications involve the production of enantioselective
catalysts using chiral carboxylate and carboxamidate
ligands [4] and axially chiral complexes derived from
achiral orthometallating aryl phosphines [5]. These
modifications have been driven largely by guesswork
and chemical intuition due to the paucity of informa-
tion regarding the structural and electronic composition
of the rhodium carbenoid intermediate.
Bear and co-workers characterized (by cyclic voltam-
metry and UV–vis spectroscopy) two dirhodium ac-
etamidates, Rh2(NHC(O)CF3)4 [6] and Rh2(PhNAc)4
[7], where Ac=C(O)CH3. The trifluoroacetamidate
complex was synthesized as a single isomer and pre-
sumed to be the 2,2-cis isomer. Synthesis of the N-
phenyl acetamidate complex produced two isomers,
assigned as the 2,2-cis and 3,1 isomers. Later synthesis
and structural analysis of (2,2-cis)-Rh2(NHAc)4, (2,2-
cis)-Rh2(NPhAc)4 and (3,1)-Rh2(NPhAc)4 indicated
that the conformation of these acetamidate analogs,
and by extension, previous analogs, were correctly as-
signed [8].
In the course of our efforts to shed light on this
elusive, highly transient intermediate, we have synthe-
sized the heretofore unreported 2,2-trans isomer of a
rhodium carboxamidate. Dirhodium(II) tetrakis(car-
Doyle and co-workers recently isolated and obtained
solid-state structures for their chiral rhodium acetami-
date analogs with the 3,1 (Rh2(4S-MACIM)4) [9] and
* Corresponding author.
1 Also corresponding author.
0022-328X/00/$ - see front matter © 2000 Elsevier Science S.A. All rights reserved.
PII: S0022-328X(99)00556-2

C.T. Eagle et al. / Journal of Organometallic Chemistry 596 (2000) 90–94
91
Table 1
Comparison of bond lengths of several isomeric analogs a
2,2-trans-
Rh2(PhNAc)4
· 2NCPh
2,2-cis-
Rh2(NPhAc)4
· 2DMSO
3,1-
Rh2(NPhAc)4
· 2DMSO
2,2-cis-Rh2(MACIM)4 3,1-Rh2(MACIM)4
4,0-Rh2(MACIM)4
· 2NCCH3
· 2NCCH3
· 2NCCH3
RhꢀRh
RhꢀO
2.4220
2.039
2.448
2.038
2.059
2.397
2.055
2.037
2.4586
2.081
2.009
2.199
2.241
2.4597
2.059
2.023
2.215
2.230
2.4447
2.047
2.039
2.179
2.268
b
RhꢀN(eq) 2.061 b
RhꢀN(ax) 2.2047
2.2478
a
,
All bond lengths are given in A.
b RhꢀO and RhꢀN(eq) bond lengths are given as average lengths.
4,0 (Rh2(4S-MACIM)4) [10] isomeric orientations. Syn-
thesis of each of these complexes, as well as two other
chiral carboxamidates, produced three isomers, the 2,2-
cis, 3,1 and 4,0.
2.2. Structural analysis
The crystal structure of 1 was obtained as the bis
benzonitrile adduct (Fig. 1). The structural features of
the 2,2-trans isomer are similar in many respects to
those of other, previously characterized, isomeric
,
analogs (Table 1). The RhꢀRh bond length (2.422 A) is
2. Discussion
similar to those reported by Bear and Doyle. The
RhꢀN bonds of the equatorial acetamide ligands are
slightly longer than those reported by Bear and Doyle;
while the RhꢀO bonds are slightly shorter.
2.1. Synthesis
We report herein the first isolation and structural
characterization of the 2,2-trans isomer of a rhodium
carboxamidate analog: 2,2-trans-dirhodium(II) tetrak-
is(N-phenylacetamidate) [Rh2(PhNAc)4] (1). This iso-
mer was synthesized, along with the 2,2-cis and 3,1
isomers, by refluxing a mixture of rhodium acetate and
N-phenylacetamide in chlorobenzene, over 7 days, us-
ing a Soxhlet extractor. The thimble of the Soxhlet
extractor was charged with sodium carbonate and sand,
and replaced every 24 h [11]. The isomers were sepa-
rated by flash chromatography on silica gel. This differs
significantly from the method used by Bear and co-
workers, who allowed the rhodium acetate to react in a
melt of the acetamide.
The acetamide bridges are twisted slightly from pla-
narity, with NꢀRhꢀRhꢀO dihedral angles from 9.03 to
11.89. This twisting, which is likely due to crystal
packing forces, is present in Doyle’s chiral rhodium
acetamidate complexes and Bear’s (3,1)-Rh2(NPhAc)4
complex, but is essentially absent in Bear’s less steri-
cally encumbered Rh2(NHAc)4 and (2,2-cis)-
Rh2(NPhAc)4 analogs. The phenyl rings attached to the
acetamide bridges are nearly perpendicular to the plane
of the amide bridge. While this arrangement effectively
eliminates conjugation between the two fragments, it is
necessitated for steric reasons. A co-planar arrangement
would not only block the axial site from coordination,
but more importantly, would result in the ortho protons
Doyle and co-workers, on the other hand, conducted
their syntheses in a manner analogous to that employed
by us. However, their reflux rate was such that the
Soxhlet cycling time was as frequent as every 30 s,
resulting in reaction times of 2–22 h. Doyle and co-
workers report three isomeric products from their reac-
tions, the 2,2-cis, 3,1 and 4,0 isomers, while we observe
only the 2,2-trans, 2,2-cis and 3,1 isomers. In the syn-
thesis of Rh2(4S-MACIM)4, Doyle and co-workers re-
port the 4,0 isomer is initially formed, but disappears
after 18 h of heating at reflux. This suggests that the
previously observed isomers may be kinetic products,
while the 2,2-trans isomer is a thermodynamic product,
evidenced only after extended reaction periods. Addi-
tional studies in this area are on-going.
Fig. 1. ORTEP of Rh2[N(C6H5)C(O)CH3]4·2NCC6H5 (1) showing the
co-planar arrangement of the axial benzonitrile ligands with the
proximal nitrogen atoms of the acetamidate ligands.

92
C.T. Eagle et al. / Journal of Organometallic Chemistry 596 (2000) 90–94
,
of the phenyl rings approaching within 0.7 A of the
CH3 moiety of the acetamide.
+554 mV. Diagnostic criteria applied to the oxidation
shows that the peak separation (Ep,a−Ep,c or DEp)
increases from 85 to 196 mV with scan rates increasing
from 50 to 2000 mV s−1. However, the value of E1/2 is
independent of scan rate, as is the current function,
The most intriguing structural feature is the perpen-
dicular disposition of the axial benzonitrile ligands. The
planes of the aromatic rings are essentially perpendicu-
lar to one another. This is in contrast to a series of
phenyl isonitrile adducts of rhodium acetate character-
ized by Eagle et al., in which the aromatic rings are
co-planar [12]. Fenske–Hall calculations on the
rhodium acetate isonitrile complexes indicate that the
observed co-planarity is a consequence of extended
conjugation between the axial ligands, through the
rhodium–rhodium core via p-backbonding [13]. The
orthogonal arrangement of the benzonitrile fragments
is also likely to be a consequence of p-backbonding.
The RhꢀRh core of the Rh2(NPhAc)4 fragment is
isolobal with Rh2(OAc)4, containing an orthogonal pair
of filled d–p* molecular orbitals (MOs). Likewise, the
benzonitrile fragments are isolobal with the
phenylisonitrile fragments, containing a pair of vacant
p* orbitals available to receive electron density from
the rhodium’s filled p* orbital. The Fenske–Hall calcu-
lations show p-backbonding from both of the filled
d–p* MOs to each of the vacant isonitrile p* MOs,
indicating that extended conjugation can occur between
the nitrile fragments even with the orthogonal arrange-
ment of the aromatic rings.
The perpendicular arrangement of the aromatic rings
of 1 renders each ring co-planar with the pair of
equatorial nitrogens on the proximal rhodium atom.
This orientation maximizes possible overlap of the p-
type MOs of the benzonitrile fragment with the more
electron-donating nitrogen atoms (relative to the oxy-
gen atoms) of the acetamide fragments. Thus, the con-
formation of the axial ligands could be electronic in
nature, a consequence of favorable orbital overlap.
Alternatively, the conformation could result from an
electrostatic attraction between the electropositive ben-
zonitrile aryl hydrogens and the electronegative phenyl
amidate p-cloud.
i
p,a/61/2. The ratio of cathodic to anodic peak currents
declines from unity at low scan rates to a value of 0.89
at 2000 mV s−1. These criteria indicate a quasi-re-
versible electron transfer mechanism for this oxidative
process, i.e. one characterized by moderately slow elec-
tron-transfer kinetics.
The second oxidation appears at an E1/2 of +1798
mV (average value) versus Ag ꢀ Ag+. This potential
increases with increasing scan rate, from 1779 (50 mV
s
−1) to 1824 mV (2000 mV s−1). The value of DEp
increases from 91 (50 mV s−1) to 202 mV (2000 mV
s
−1). The current ratio increases with increasing scan
rate; the current function drops slightly with increasing
scan rate. Taken together, these criteria indicate a
chemical reaction following the generation of the +2
ion, or an EC mechanism.
In acetonitrile, coordination of the solvent to the
axial sites on the rhodium ions results in a shift in
potential for the observed oxidations relative to the
same complex observed in 1,2-dichloroethane. The first
oxidation is shifted to an E1/2 value of +310 mV, a
cathodic shift of +244 mV relative to that observed in
the non-coordinating solvent. Again, E1/2 values were
invariant with increasing scan rate. The value of DEp
increased from 77 (50 mV s−1) to 101 mV (2000 mV
s
−1), much smaller values than were observed in 1,2-
dichloroethane. The current ratio declined from unity
at low scan rates to 0.87 at 2000 mV s−1, while the
current function was relatively invariant with scan rate.
The electron-transfer mechanism of this first oxidation
was, of course, unchanged, although it appears that
coordination of the nitrile ligands has increased the rate
of electron transfer for the formation of the +1 ion.
The mechanism of the second oxidation was also not
changed by altering the solvent, though the same poten-
tial shift was observed. The value of +1566 mV (aver-
age value) was 232 mV cathodic of the E1/2 observed in
1,2-dichloroethane. The value of DEp shifted to a maxi-
mum of 125 mV at 2000 mV s−1, only 62% of the value
observed in 1,2-dichloroethane. Other diagnostic crite-
ria were unchanged; the current ratio still increases with
increasing scan rate, while the current function de-
creases by a small amount over the same range. This
indicates a chemical reaction following the generation
of the +2 ion.
Steric explanations for the perpendicular orientation
of the axial ligands are unconvincing. It is unlikely that
both rings would be co-planar with the proximal pair of
nitrogens, rather than have a random disposition, as
seen in several rhodium acetate arylamine complexes
[14], especially in light of the lack of symmetry in 1.
2.3. Cyclic 6oltammetry
The cyclic voltammetric behavior of 1 in 1,2-
dichloroethane and acetonitrile is reminiscent of that
reported for Bear’s 2,2-cis-Rh2(PhNAc)4 isomer [7]. In
the non-coordinating 1,2-dichloroethane, two oxida-
tions are observed within the potential window af-
forded by the solvent/supporting electrolyte. The first
oxidation appears at a half-wave potential (E1/2) of
2.4. Visible spectroscopy
The complex exhibits a broad visible absorption
which varies depending on the solvent used. In 1,2-
dichloroethane, a featureless absorption at umax=497.8

C.T. Eagle et al. / Journal of Organometallic Chemistry 596 (2000) 90–94
93
nm is observed. In acetonitrile, this absorption is cen-
tered at umax=510.4 nm, a shift of 12.6 nm. This red
shift is consistent with our cyclic voltammetric observa-
tions, indicating reduced energy of the HOMO for the
complex in the p-donor solvent.
phenylacetamide was removed by sublimation, leaving
a purple residue. This residue was purified by flash
column chromatography (silica, hexane–ethyl acetate).
Three colored bands eluted in the order: green (1), blue
(2), and green (3). No detectable traces of a fourth band
were evident. The eluting solvent was removed from
each fraction under reduced pressure. Each fraction
was then dried in a vacuum oven at 80°C for 8 h. Total
yield of product from Rh2(O2CCH3)4 was 69%; of that
18% was fraction 1, 71% was fraction 2, and 11% was
3. Experimental
NMR spectra were obtained with a Varian Gemini
2000 spectrometer (300 MHz) using CDCl3 as solvent.
Spectra were referenced to solvent. Elemental analysis
was performed by Galbraith Laboratories. Visible spec-
tra were obtained using a Shimadzu UV-2401PC UV–
vis spectrophotometer. Cyclic voltammetry was
performed with a BioAnalytical Systems, Inc. model
100 B Workstation.
1
fraction 3. H-NMR: l 1.67 (s, 12H), 7.00 (d, 8H), 7.14
(t, 4H), 7.29 (t, 4H). 13C-NMR: l 21.41, 124.58, 126.01,
129.44, 149.23, 179.21. Elemental Analysis for
C32H32N4O4Rh2: Calc. C, 51.77; H, 4.34; N, 7.55. Anal.
C, 51.38; H, 3.99; N, 7.65%.
Rh2(N{C6H5}COCH3)4 · 2NCC6H5 was obtained as
red crystals by adding benzonitrile, without mixing, to
a solution of 1 in acetone and allowing crystal growth
over several days.
X-ray crystallography data: molecular formula,
C46H42N6O4Rh2, Mr=948.68, T=294(1) K, mono-
clinic C2/c (No. 15), a=30.444(6), b=10.657(2), c=
3
,
,
26.138(5) A, i=90.39(3) °, V=8480(4) A ; Z=8;
D
calc. =1.49 Mg m−3; F(000)=3856. Data were col-
4. Supplementary material
lected in a red platelet crystal of size 0.048×0.12×0.14
mm3 using a Rigaku AFC7R (18 kW) diffractometer
with graphite-monochromated Mo–Ka radiation (u=
X-ray crystallographic material has been deposited
with the Cambridge Crystallographic Centre, deposi-
tion number CCDC 134587. Copies of this information
may be obtained free of charge from The Director,
CCDC, 12 Union Road, Cambridge, CB2 1EZ, UK
(fax: +44-1223-336033; e-mail: deposit@ccdc.cam.
ac.uk or www: http://www.ccdc.cam.ac.uk).
,
0.71073 A), in the q range 2.5–25°. A total of 8799
reflections were measured, 7526 unique (Rint=0.025),
and 4693 observed (I\2|(I)); empirical absorption
correction, v=0.83 mm−1, transmission factors=
0.93–1.00. The structure was solved by direct methods
and refined by full-matrix least-squares on F. Final
residual values were R=0.045, Rw=0.043 for observed
data and R=0.085, Rw=0.050 for all data.
Acknowledgements
Cyclic voltammetric measurements were made under
a blanket of dry nitrogen gas using 0.1 M tetra-N-
butylammonium hexafluorophosphate as supporting
electrolyte. This had been purified by recrystallizing
three times from ethanol and was dried prior to use.
Electrochemical solvents were 1,2-dichloroethane or
acetonitrile. The latter was distilled over CaH2 under
inert atmosphere immediately prior to use. The working
electrode was a Pt disc and the auxiliary electrode
consisted of a Pt coil. All electrochemical measurements
were referenced against the Ag ꢀ Ag+ couple. Potentials
were not corrected for liquid junction, but were instead
calibrated daily against an internal standard
(ferrocene).
The authors gratefully acknowledge financial support
from Research Corporation, a Supplemental Award of
the Camille and Henry Dreyfus Foundation Grant
Program in Chemistry for Liberal Arts Colleges and
Appalachian State University Research Council. The
authors also thank Johnson Matthey Inc. for the loan
of rhodium salts.
References
[1] R. Paulissen, H. Reimlinger, E. Hayez, A.J. Hubert, Ph. Teyssie,
Tetrahedron Lett. (1973) 2233.
[2] M.P. Doyle, A. McKervey, T. Ye, Modern Catalytic Methods
for Organic Synthesis with Diazo Compounds: From Cyclo-
propanes to Ylides, Wiley, New York, 1998.
[3] For examples see: A. Padwa, D.J. Austin, A.T. Price, M.A.
Semones, M.P. Doyle, P. Marina, W.R. Winchester, A. Tran, J.
Am. Chem. Soc. 115 (1993) 8669 and references therein.
[4] For examples see: M.P. Doyle, Aldrichimica Acta 29 (1996) 3
and references therein.
[5] D.F. Taber, S.C. Malcolm, K. Bieger, P. Lahuerta, M. Sanau,
S.-E. Stiriba, J. Perez-Prieto, M.A. Monge, J. Am. Chem. Soc.
121 (1999) 860.
Rh2(N{C6H5}COCH3)4
was
prepared
from
Rh2(O2CCH3)4 [15] (1.00 g) and phenylacetamide (15.00
g), which were added to 150 ml of chlorobenzene, dried
according to standard methods. The mixture was
heated to reflux, and the reflux apparatus fitted with a
Soxhlet extractor. The thimble was charged with sand
and sodium carbonate (dried at 100°C for 2 days). The
thimble was replaced every 24 h. After 7 days, the
solvent was removed by vacuum distillation and the

94
C.T. Eagle et al. / Journal of Organometallic Chemistry 596 (2000) 90–94
[10] M.P. Doyle, C.E. Raab, G.H.P. Roos, V. Lynch, S.H. Simonsen,
[6] (a) A.M. Dennis, D. Lancon, K.M. Kadish, J.L. Bear, J. Chem.
Soc. Chem. Commun. (1982) 399. (b) K.M. Kadish, D. Lancon,
A.M. Dennis, J.L. Bear, Inorg. Chem. 21 (1982) 2987.
[7] (a) J. Duncan, T. Malinski, T.P. Zhu, Z.S. Hu, K.M. Kadish,
J.L. Bear, J. Am. Chem. Soc. 104 (1982) 5507. (b) J.L. Bear, T.P.
Zhu, T. Malinski, A.M. Dennis, K.M. Kadish, Inorg. Chem. 23
(1984) 674.
Inorg. Chim. Acta 266 (1997) 13.
[11] M.P. Doyle, V. Bagheri, T.J. Wandles, N.K. Harn, D.A.
Brinker, C.T. Eagle, K.-L. Loh, J. Am. Chem. Soc. 112 (1990)
1906.
[12] C.T. Eagle, D.G. Farrar, C.U. Pfaff, J.A. Davies, C. Kluwe, L.
Miller, Organometallics 17 (1998) 4523.
[8] (a) M.Q. Ahsan, I. Bernal, J.L. Bear, Inorg. Chem. 25 (1986)
260. (b) M.Y. Chavan, X.Q. Lin, I. Bernal, J.L. Bear, K.M.
Kadish, Inorg. Chem. 25 (1986) 1281. (c) R.S. Lifsey, X.Q. Lin,
M.Y. Chavan, M.Q. Ahsan, K.M. Kadish, J.L. Bear, Inorg.
Chem. 26 (1987) 830.
[13] A.L. Sargent, M.E. Rollog, C.T. Eagle, Theor. Chem. Acc. 97
(1997) 283.
[14] (a) F.A. Cotton, T.R. Felthouse, Inorg. Chem. 20 (1981) 600. (b)
Y.B. Koh, G.G. Christoph, Inorg. Chem. 17 (1978) 2590.
[15] Rh2(O2CCH3)4 was prepared according to the method of G.A.
Rampel, P. Legzdins, H. Smith, G. Wilkinson, Inorg. Synth. 13
(1972) 90.
[9] M.P. Doyle, Q.-L. Zhou, C.E. Raab, G.H.P. Roos, S.H. Simon-
sen, V. Lynch, Inorg. Chem. 35 (1996) 6064.
.