Superbasic amidine monodentate ligands in fac -[Re(CO)3(5, 5′-Me2bipy)(Amidine)]BF4 Complexes: Dependence of amidine configuration on the remote nitrogen substituents

Article abstract of DOI:10.1021/ic100714m

Addition of various RNH₂ to fac-[Re(CO)₃(5,5′- Me₂bipy)(CH₃CN)]BF₄ (1) converts the acetonitrile ligand to the amidine ligand (a superbase) in fac-[Re(CO) ₃(5,5′-Me₂bipy)(HNC(CH₃)NHR)]BF ₄ products. Each complex has four conceivable isomers (E, E′, Z, and Z′) because the amidine CN bonds have double-bond character, and the two remote NHR group substituents are different. The reaction of 1 in acetonitrile is complete in 6 to 96 h (25 °C) and forms fac-[Re(CO) ₃(5,5′-Me₂bipy)(HNC(CH₃)NHR)]BF ₄ E′ and Z isomers. Only the E′ isomer formed crystals (R = methyl, isopropyl, isobutyl, tert-butyl, and benzyl). Upon dissolution of such crystals in acetonitrile-d₃, NMR spectra with highly dominant E′ signals gradually changed (～15 min at room temperature) to spectra with signals for an equilibrium mixture of E′ and Z isomers. Such slow E′-to-Z isomer interchange is also indicated by 2D ROESY NMR data used primarily to assign solution structure. Equilibrium ratios (E′:Z) of ～65:35 for R = methyl, isopropyl, and isobutyl and 83:17 for R = tert-butyl demonstrate that increasing the remote NHR group steric bulk above a threshold size favors the E′ isomer. Consistent with this trend, fac-[Re(CO) ₃(5,5′-Me₂bipy)(HNC(CH₃)NH ₂)]BF₄, with a remote NH₂ (low bulk) group, favors the Z isomer. In contrast, although the remote NH(benzyl) group in fac-[Re(CO)₃(5,5′-Me₂bipy)(HNC(CH ₃)NH(CH₂C₆H₅)]BF₄ has only moderate bulk, the E′ isomer has high abundance as a result of favorable 5,5′-Me₂bipy/benzyl stacking, evidence for which is present in both solid and solution states. The fac-[Re(CO)₃(5, 5′-Me₂bipy)(HNC(CH₃)NHR)]BF₄ E isomer can be detected in solvents of low polarity. However, the Z′ isomer was not observed, undoubtedly because unfavorable remote-group clashes with the equatorial ligands destabilize this isomer. Challenge studies with a 5-fold excess of 4-dimethylaminopyridine in acetonitrile-d₃ establish that fac-[Re(CO)₃(5,5′-Me₂bipy)(HNC(CH ₃)NHCH(CH₃)₂)]BF₄ is robust because the isopropylamidine ligand was not displaced, consistent with the superbase character of amidine ligands.

Full text of DOI:10.1021/ic100714m

Inorg. Chem. 2010, 49, 7035–7045 7035
DOI: 10.1021/ic100714m
Superbasic Amidine Monodentate Ligands in fac-[Re(CO)3(5,50-Me2bipy)(Amidine)]BF4
Complexes: Dependence of Amidine Configuration on the Remote
Nitrogen Substituents
Theshini Perera, Frank R. Fronczek, Patricia A. Marzilli, and Luigi G. Marzilli*
Department of Chemistry, Louisiana State University, Baton Rouge, Louisiana 70803
Received April 13, 2010
Addition of various RNH2 to fac-[Re(CO)3(5,50-Me2bipy)(CH3CN)]BF4 (1) converts the acetonitrile ligand to the amidine
ligand (a superbase) in fac-[Re(CO)3(5,50-Me2bipy)(HNC(CH3)NHR)]BF4 products. Each complex has four conceivable
isomers (E, E0, Z, and Z0) because the amidine CN bonds have double-bond character, and the two remote NHR group
substituents are different. The reaction of 1 in acetonitrile is complete in 6 to 96 h (25 °C) and forms fac-[Re(CO)3(5,50-
Me2bipy)(HNC(CH3)NHR)]BF4 E0 and Z isomers. Only the E0 isomer formed crystals (R = methyl, isopropyl, isobutyl,
tert-butyl, and benzyl). Upon dissolution of such crystals in acetonitrile-d3, NMR spectra with highly dominant E0 signals
gradually changed (∼15 min at room temperature) to spectra with signals for an equilibrium mixture of E0 and Z isomers.
Such slow E0-to-Zisomer interchange is also indicated by 2D ROESY NMR data used primarily to assign solution structure.
Equilibrium ratios (E0:Z) of ∼65:35 for R = methyl, isopropyl, and isobutyl and 83:17 for R = tert-butyl demonstrate that
increasing the remote NHR group steric bulk above a threshold size favors the E0 isomer. Consistent with this trend,
fac-[Re(CO)3(5,50-Me2bipy)(HNC(CH3)NH2)]BF4, with a remote NH2 (low bulk) group, favors the Z isomer. In contrast,
although the remote NH(benzyl) group in fac-[Re(CO)3(5,50-Me2bipy)(HNC(CH3)NH(CH2C6H5)]BF4 has only moderate
bulk, the E0 isomer has high abundance as a result of favorable 5,50-Me2bipy/benzyl stacking, evidence for which is present
in both solid and solution states. The fac-[Re(CO)3(5,50-Me2bipy)(HNC(CH3)NHR)]BF4 E isomer can be detected in
solvents of low polarity. However, the Z0 isomer was not observed, undoubtedly because unfavorable remote-group
clashes with the equatorial ligands destabilize this isomer. Challenge studies with a 5-fold excess of 4-dimethylaminopyr-
idine in acetonitrile-d3 establish that fac-[Re(CO)3(5,50-Me2bipy)(HNC(CH3)NHCH(CH3)2)]BF4 is robust because the
isopropylamidine ligand was not displaced, consistent with the superbase character of amidine ligands.
Introduction
the acetonitrile reaction chemistry of ReI complexes has
recently become a subject of scrutiny.7-9
The chemistry of amidine complexes of several metals
(including platinum, iridium, cobalt, and manganese) has been
described in several reviews.1-3 Because the fac-{MI(CO)3}
(M = Tc or Re) core is important in radiopharmaceuticals,4-6
Both of the amidine carbon-to-nitrogen bonds have double-
bond character. The bond between the amidine carbon
and the superbasic coordinated nitrogen10-12 leads to two
configurations dictating the description of amidine stereo-
chemistry as E or Z (Figure 1). When the remote nitrogen has
two different substituents, the two resulting configurations
about the bond to the amidine carbon lead to a total of four
conceivable configurations depicted in Figure 1 (E, E0, Z, and
Z0 labels follow previous usage).13,14 Whereas detection of
isomers involving the two configurations about the bond
*To whom correspondence should be addressed. E-mail: lmarzil@lsu.edu.
(1) Kukushkin, V. Y.; Pombeiro, A. J. L. Chem. Rev. 2002, 102, 1771–
1802.
(2) Michelin, R. A.; Mozzon, M.; Bertani, R. Coord. Chem. Rev. 1996,
147, 299–338.
(3) Baker, J.; Kilner, M. Coord. Chem. Rev. 2000, 133, 219–300.
(4) Alberto, R.; Schibli, R.; Schubiger, A. P.; Abram, U.; Pietzsch, H. J.;
Johannsen, B. J. Am. Chem. Soc. 1999, 121, 6076–6077.
(5) Lipowska, M.; Marzilli, L. G.; Taylor, A. T. J. Nucl. Med. 2009, 50,
454–460.
ꢀ
(10) Raczynska, E. D.; Darowska, M. J. Org. Chem. 2004, 69, 4023–4030.
(11) Decouzon, M.; Gal, J.-F.; Maria, P.-C. Rapid Commun. Mass.
€
(6) Bartholoma, M.; Valliant, J.; Maresca, K. P.; Babich, J.; Zubieta, J.
Spectrom. 1993, 7, 599–602.
€
~~
(12) Kaljurand, I.; Koppel, I. A.; Kutt, A.; Room, E.-I.; Rodima, T.;
Koppel, I.; Mishima, M.; Leito, I. J. Phys. Chem. A 2007, 111, 1245–1250.
(13) Marzano, C.; Sbovata, S. M.; Bettio, F.; Michelin, R. A.; Seraglia,
R.; Kiss, T.; Venzo, A.; Bertani, R. J. Biol. Inorg. Chem. 2007, 12, 477–493.
(14) Belluco, U.; Benetollo, F.; Bertani, R.; Bombieri, G.; Michelin,
R. A.; Mozzon, M.; Tonon, O.; Pombeiro, A. J. L.; da Silva, F. C. G. Inorg.
Chim. Acta 2002, 334, 437–447.
Chem. Commun. (Cambridge, U.K.) 2009, 493–512.
(7) Arroyo, M.; Miguel, D.; Villafane, F.; Nieto, S.; Perez, J.; Riera, L.
Inorg. Chem. 2006, 45, 7018–7026.
(8) Perera, T.; Marzilli, P. A.; Fronczek, F. R.; Marzilli, L. G. Inorg.
Chem. 2010, 49, 2123–2131.
(9) Yam, V. W.; Wong, K. M.; Cheung, K. Chem. Commun. (Cambridge,
U.K.) 1998, 135–136.
r
2010 American Chemical Society
Published on Web 07/01/2010
pubs.acs.org/IC

7036 Inorganic Chemistry, Vol. 49, No. 15, 2010
Perera et al.
Figure 1. Conceivable fac-[Re(CO)3(L)(HNC(CH3)NHR)]þ isomers,
in which L is a bidentate ligand denoted by N-N donor atoms.
between the amidine carbon and the metal-bound nitrogen
is a common occurrence for amidine ligands as well as imino-
ether ligands,15 there have been few reports of isomers
resulting from restricted rotation about the other amidine
carbon-to-nitrogen bond, in part because amidine ligands
often have a symmetrical remote NH2 or NR2 grouping, thus
precluding such isomers. Also, in previously studied ReI
carbonyl compounds, ring formation restricted the amidine
stereochemistry.7,8
Most reports on monodentate amidine ligands of interest
to the current study on complexes with the fac-{ReI(CO)3}
core involve pseudo square planar PtII complexes, about
which many studies have been performed in view of the
demonstrated cytotoxicity of many iminoether and amidine
Pt complexes.13,16-18 Iminoether and amidine ligands are
related. Natile and co-workers have contributed substantially
to this field because the first PtII compound with a trans
configuration shown to have anticancer activity was an
iminoether complex;19 these and other investigators later
extended such studies to the evaluation of ketimine20 and
amidine13,21,22 Pt complexes.
Figure 2. [Re(CO)3(L)(HNC(CH3)NHR)]þ isþomers observed upon
treatment of [Re(CO)3(5,50-Me2bipy)(CH3CN)] with RNH2 in aceto-
nitrile at 25 °C.
thought to be stabilized by strong intramolecular H-bonds
between chloride and the remote NH group.22 The amidine
ligands in cis-[Pt(HNC(CH3)N(CH3)2)2Cl2] lack a remote
NH group, however, and adopt the E configuration.23 Belluco
et al. reported that, on the basis of 1H NMR data, the addition
reactions of primary amines and secondary amines to cis- and
trans-[PtCl2(PhCN)2] afforded a complicated mixture of ami-
dine complexes, with the amidine having a mixture of E, E0, Z,
and Z0 configurations in CD2Cl2.14 In a more recent study,
Marzano et al. conducted additional chemical studies of some
of these amidine complexes because they have promising
antitumor activity.13 These investigators noted that the reac-
tion forming cis-[PtCl2(HNC(Ph)NHCH3)2] afforded a mix-
ture of E, E0, Z, and Z0 isomers, unlike such reactions with
acetonitrile derivatives; heating converted the product to the
Z isomer stabilized by intramolecular H-bonding to chloride
ligands.13 Both steric effects and H-bonding play a role in
dictating stereochemistry in these Pt amidine complexes.22,23
Nucleophilic attack of pyrazole on a coordinated, metal-
activated nitrile resulted in the formation of a pyrazolylami-
dino chelate ring in ReI carbonyl compounds.7 We recently
reported some unusual ReI amidine complexes formed by
attack of primary or secondary amine terminal groups of
polyamines on coordinated acetonitrile, in one case giving a
seven-membered chelate ring.8 The starting complex in that
study8 had three coordinated acetonitrile ligands, and the
attacking amines were complicated.
In reports on Pt complexes, the dependence of the amidine
ligand configuration on the steric bulk and the presence of NH
groups of the remote amidine NR2, NHR, or NH2 group in the
ligand have been assessed.14,22,23 Both amidine ligands in
trans-[Pt(HNC(CH3)NHCH3)2Cl2] have the Zconfiguration,22
(15) Cornacchia, D.; Pellicani, R.; Intini, F. P.; Pacifico, C.; Natile, G.
Inorg. Chem. 2009, 48, 10800–10810.
(16) Natile, G.; Coluccia, M. Coord. Chem. Rev. 2001, 216-217, 383–410.
(17) Intini, F. P.; Pellicani, R. Z.; Boccarelli, A.; Sasanelli, R.; Coluccia,
M. Eur. J. Inorg. Chem. 2008, 4555–4561.
(18) Sbovata, S. M.; Bettio, F.; Mozzon, M.; Bertani, R.; Venzo, A.;
Benetollo, F.; Michelin, R. A.; Gandin, V.; Marzano, C. J. Med. Chem. 2007,
50, 4775–4784.
(19) Coluccia, M.; Nassi, A.; Loseto, F.; Boccarelli, A.; Mariggio, M. A.;
Giordano, D.; Intini, F. P.; Caputo, P.; Natile, G. J. Med. Chem. 1993, 36,
510–512.
(20) Boccarelli, A.; Intini, F. P.; Sasanelli, R.; Sivo, M. F.; Coluccia, M.;
Natile, G. J. Med. Chem. 2006, 49, 829–837.
(21) Natile, G.; Intini, F. P.; Bertani, R.; Michelin, R. A.; Mozzon, M.;
Sbovata, S. M.; Venzo, A.; Seraglia, R. J. Organomet. Chem. 2005, 690,
2121–2127.
(22) Bertani, R.; Catanese, D.; Michelin, R. A.; Mozzon, M.; Bandoli, G.;
Dolmella, A. Inorg. Chem. Commun. 2000, 3, 16–18.
(23) Michelin, R. A.; Bertani, R.; Mozzon, M.; Sassi, A.; Benetollo, F.;
Bombieri, G.; Pombeiro, A. J. L. Inorg. Chem. Commun. 2001, 4, 275–280.
To assess ReI amidine chemistry, we have now investigated
amidine products formed by treating fac-[Re(CO)3(5,50-
Me2bipy)(CH3CN)]BF4 (a complex with only one coordi-
nated acetonitrile) with ammonia and amines (Figure 2). In
the resulting complexes, such asfac-[Re(CO)3(5,50-Me2bipy)-
(HNC(CH3)NHR)]BF4, the monodentate amidine ligand
can conceivably have any of the four possible configurations.
Unlike in the cases of many other Re and Pt complexes, these
configurations will not be confounded by H-bonding inter-
actions or controlled by ring formation. Only steric and solvent

Article
Inorganic Chemistry, Vol. 49, No. 15, 2010 7037
Table 1. Crystal Data and Structural Refinement for [Re(CO)3(5,50-Me2bipy)(HNC(CH3)NHR)]BF4 Complexes
R=isopropyl
R=isobutyl
R=tert-butyl
R=benzyl
R=H
2
3
4
5
7
empirical formula
C
20H24N4O3Re
BF4
C21H26N4O3Re
0.96(BF4) 0.04(Br)
0.33(C2H3N)
C
21H26N4O3Re
0.974(BF4) 0.026(Br)
C
24H24N4O3Re
BF4
C17H18N4O3Re
BF4
3
3
3
3
3
3
3
3
fw
space group
641.44
P21/c
8.4609(5)
14.4443(10)
19.4276(12)
90
102.446(4)
90
2318.5(3)
90
669.73
P1
655.30
P21/c
8.4318(10)
20.314(3)
14.2875(15)
90
102.370(7)
90
2390.4(5)
90
689.48
P21/n
10.4196(10)
22.3643(15)
11.2768(10)
90
104.117(3)
90
2548.4(4)
90
599.36
P21/c
13.040(3)
12.143(3)
13.423(4)
90
103.453(12)
90
2067.1(9)
180
˚
a (A)
8.5566(5)
17.5817(14)
18.8096(3)
106.934(3)
91.331(4)
103.427(5)
2620.4(3)
100
˚
b (A)
˚
c (A)
R (deg)
β (deg)
γ (deg)
3
˚
V (A )
T (K)
Z
F
4
1.838
5.30
66.2
0.024
0.053
8796/310
4
1.698
4.76
64.0
0.034
0.078
18182/663
4
1.821
5.19
65.2
0.032
0.067
8717/321
4
1.797
4.83
72.0
0.030
0.067
11976/344
4
1.926
5.94
50.1
0.048
0.091
3649/274
calc (mg/m3)
abs coeff (mm-1
2θmax (deg)
)
R indicesa
wR2=[I > 2σ(I)]b
data/param
P
P
P
P
a R= ||Fo| - |Fc||/ |Fo|. b wR2 = [ w(Fo2 - Fc2)2/ w(Fo2)2]1/2, in which w = 1/[σ2(Fο2) þ (dP)2 þ (eP)] and P = (Fο þ 2Fc2)/3, d = 0.0218,
2
0.0341, 0.0284, 0.0275, and 0.0331, and e=1.2987, 0.9602, 0, 1.9993, and 2.9329 for complexes 2-5 and 7, respectively.
˚
effects and intrinsic electronic structures of the amidine ligands
will influence geometry and stability. Two-dimensional NMR
spectroscopy, in conjunction with structural characterization
of several complexes by single-crystal X-ray crystallography,
has been utilized to evaluate which monodentate amidine
ligand configurations (Figure 1) are favored in solution.
Note: We omit the fac-designation when discussing specific
complexes. Also, to simplify the text, we shall use the terms,
E, E0, Z, or Z0 isomer, to designate the isomers of the entire
complex with the amidine ligand in the respective configura-
tions.
with graphite-monochromated Mo KR (λ=0.71073 A) radia-
tion. Data reduction included absorption corrections by the
multiscan method, with HKL SCALEPACK.27 All X-ray struc-
tures were determined by direct methods and difference Fourier
techniques and refined by full-matrix least-squares by using
SHELXL97.28 All non-hydrogen atoms were refined anisotropi-
cally. All H atoms were visible in differencemaps, but were placed
in idealized positions, except for N-H hydrogen atoms, which
were refined individually where possible. A torsional parameter
was refined for each methyl group. For compound 3, the con-
tribution to the structure factors from disordered solvent was
removed by using SQUEEZE,29 amounting to 4/3 molecules of
acetonitrile per unit cell. For compounds 3 and 4, BF4- sites were
shared by a small amount of bromide, and the occupancies of
BF4- and Br- were constrained to sum to unity. The occupancies
were fixed in final refinements for 3. The structure of compound 6
was determined from a crystal having more substantial substitu-
Experimental Section
Starting Materials. Re(CO)5Br was synthesized as described
in the literature.24 Re2(CO)10, 5,50-dimethyl-2,20-bipyridine (5,50-
Me2bipy), isopropylamine, isobutylamine, tert-butylamine,
benzylamine, anhydrous methylamine and ammonia in steel
cylinders, and AgBF4 were obtained from Aldrich. [Re(CO)3-
(CH3CN)3]BF4 was synthesized by a slight modification of a
known procedure.25 The synthesis of [Re(CO)3(5,50-Me2bipy)-
(CH3CN)]BF4 (1) from [Re(CO)3(CH3CN)3]BF4 is described
elsewhere.26 Elemental analyses were performed by Atlantic
Microlabs, Atlanta, GA.
NMR Measurements. 1H NMR spectra were recorded on a 400
MHz Bruker spectrometer. Peak positions are relative to TMS or
solvent residual peak, with TMS as reference. All NMR data were
processed with TopSpin and Mestre-C software. For specific
assignments of signals listed in the synthetic section below, please
see tables in the text and the Supporting Information.
-
tional disorder for the anion, apparently about 52% BF4 and
48% ReO4-. In compound 3, the isobutyl group of one of the two
independent cations is disordered into two orientations with
0.725(7)/0.275(7) occupancy. Crystal data and details of refine-
ments are listed in Table 1, except for compound 6, for which the
cation is illustrated in the Supporting Information.
Synthesis of [Re(CO)3(5,50-Me2bipy)(HNC(CH3)NHCH-
(CH3)2)]BF4 (2). Isopropylamine (52 μL, 0.60 mmol) was added
to an acetonitrile solution (6 mL) of [Re(CO)3(5,50-Me2bipy)-
(CH3CN)]BF4 (1, 0.04 g, 0.06 mmol), and the reaction mixture
was stirred at room temperature for 24 h. The volume was
reduced to ∼1 mL by rotary evaporation, and addition of
diethyl ether (∼10 mL) produced a yellow crystalline material,
which was collected on a filter and washed with diethyl ether and
dried; yield, 40 mg (63%). An NMR spectrum recorded im-
mediately upon dissolution of this material showed mostly one
X-ray Data Collection and Structure Determination. Intensity
data were collected at low temperature on a Nonius Kappa
CCD diffractometer fitted with an Oxford Cryostream cooler
1
set of signals. H NMR signals (ppm) in acetonitrile-d3: 8.84
(s, 2H, H6/60), 8.28 (d, 2H, H3/30), 8.04 (d, 2H, H4/40), 6.10
(b, 1H, NH), 4.51(1H, NH), 3.14 (m, 1H, CH), 2.47 (s, 6H, 5/50-
CH3), 2.07 (s, 3H, CCH3), 0.74 (d, 6H, 2CH3). However, an
NMR spectrum recorded after 15 min shows an equilibrium
mixture of E0 and Z isomer signals, as observed in the reactions
monitored by NMR spectroscopy and described below.
(24) Schmidt, S. P.; Trogler, W. C.; Basolo, F. Inorg. Synth. 1990, 28, 160–
165.
(25) Edwards, D. A.; Marshalsea, J. J. Organomet. Chem. 1977, 131,
73–91.
(26) Perera, T.; Fronczek, F. R.; Marzilli, P. A.; Marzilli, L. G., manu-
script in preparation.
(27) Otwinowski, Z.; Minor, W. Macromolecular Crystallography, Part
A, Methods in Enzymology; New York Academic Press: New York, 1997; Vol.
276, pp 307-326.
(28) Sheldrick, G. M. Acta Crystallogr., Sect. A 2008, A64, 112–122.
(29) Spek, A. L. J. Appl. Crystallogr. 2003, 36, 7–13.

7038 Inorganic Chemistry, Vol. 49, No. 15, 2010
Perera et al.
Amine reactions of 1 (10 mM in acetonitrile-d3, 600 μL) were
monitored by NMR spectroscopy; we refer to this as the 10 mM
solution. The first spectrum recorded at 5 min showed only
reactants. On addition of a 10% excess of isopropylamine (5 μL)
to the 10 mM solution, NMR signals indicative of a mixture of E0
and Z isomers of 2 were observed within 30 min (at 6 h, E0:Z =
64:36), and these signals continued to grow (while maintaining
the same ratio of isomers) until no starting complex remained the
next day. 1H NMR signals (ppm) in acetonitrile-d3 (see Figure 2
for atom numbering): 8.84 (s, H6/60), 8.74 (s, H6/60), 8.28 (over-
lapping d, H3/30), 8.04 (d, H4/40), 6.10 (b, NH), 5.57 (b, NH),
5.33(NH), 4.51(NH), 3.69 (m, CH), 3.14 (m, CH), 2.47 (s, 5/50-
CH3), 2.07 (s, CCH3), 1.89 (s, CCH3), 1.21 (d, 2CH3), 0.74 (d,
2CH3). Slow evaporation of this acetonitrile solution yielded
volume approximate because the volatile methylamine was added
from an inverted container) produced a yellow precipitate; yield,
32 mg (52%). 1H NMR signals (ppm) in acetonitrile-d3: 8.83
(s, 2H, H6/60), 8.28 (d, 2H, H3/30), 8.05 (d, 2H, H4/40), 6.80 (b, 1H,
NH), 4.50 (1H, NH), 2.25 (d, 3H, NCH3), 2.47 (s, 6H, 5/50-CH3),
2.07 (s, 3H, CCH3). Anal. Calcd for C18H20BF4N4O3Re: C, 35.25;
H, 3.29; N, 9.13. Found: C, 35.36; H, 3.26; N, 9.02.
On addition of a 10% excess of methylamine (∼5 μL) to the
10 mM solution, NMR signals indicative of a mixture of E0 and
Z isomers of 6 were observed within 20 min (E0:Z = 66:34) and
continued to grow, maintaining the same ratio of isomers until
1
no starting complex signal remained (∼6 h). H NMR signals
(ppm) in acetonitrile-d3: 8.83 (s, H6/60), 8.73 (s, H6/60), 8.28
(overlapping d, H3/30), 8.05 (d, H4/40), 6.80 (b, NH), 5.90 (b,
NH), 5.29 (NH), 4.50 (NH), 2.87 (d, NCH3), 2.25 (d, NCH3),
2.47 (s, 5/50-CH3), 2.07 (s, CCH3), 1.86 (s, CCH3). Upon slow
evaporation, the resulting solution yielded X-ray quality crys-
tals of the E0 isomer.
1
X-ray quality crystals of the E0 isomer. H NMR spectrum in
acetonitrile-d3: identical to that of the bulk precipitate.
[Re(CO)3(5,50-Me2bipy)(HNC(CH3)NHCH2CH(CH3)2)]BF4
(3). The method described above but with isobutylamine (60 μL,
0.60 mmol) produced a yellow crystalline material; yield, 35 mg
Synthesis of [Re(CO)3(5,50-Me2bipy)(HNC(CH3)NH2)]BF4
(7). The method described above but with ammonia bubbling
through for ∼5 min (medium flow rate) produced a yellow
crystalline precipitate; yield, 25 mg (41%). 1H NMR spectrum in
acetonitrile-d3: identical to that given below. Ammonia gas was
bubbled through a 10 mM solution of [Re(CO)3(5,50-Me2bipy)-
(CH3CN)]BF4 in acetonitrile-d3 (600 μL), and the solution was
monitored by NMR spectroscopy. NMR signals of a mixture of
E0 and Z isomers of 7 were observed. 1H NMR signals (ppm) in
acetonitrile-d3: 8.79 (s, H6/60), 8.75 (s, H6/60), 8.29 (overlapping
d, H3/30), 8.04 (d, H4/40), 6.30 (b, NH), 5.93 (b, NH), 5.45 (NH),
5.33 (NH), 2.48 (s, 5/50-CH3), 2.12 (s, CCH3), 1.83 (s, CCH3).
When the experiment was repeated in CDCl3, a mixture of iso-
mers formed. 1H NMR signals (ppm) in CDCl3: 8.75 (s, H6/60),
8.60 (s, H6/60), 8.27 (overlapping d, H3/30), 7.92 (d, H4/40), 6.15
(b, NH), 5.86 (b, NH), 5.58 (NH), 2.50 (s, 5/50-CH3), 2.21 (s,
CCH3), 2.17 (s, CCH3). Slow evaporation of this chloroform
solution yielded X-ray quality crystals.
1
(54%). H NMR signals (ppm) in acetonitrile-d3: 8.84 (s, 2H,
H6/60), 8.29 (d, 2H, H3/30), 8.04 (d, 2H, H4/40), 6.29 (b, 1H,
NH), 4.52 (1H, NH), 2.49 (m, 2H, CH2), 2.47 (s, 6H, 5/50-CH3),
2.08 (s, 3H, CCH3), 1.82 (m, 1H, CH), 0.56 (d, 6H, CH3).
On addition of a 10% excess of isobutylamine (6 μL) to the
10mMsolution, NMRsignals ofa mixtureofE0 and Z isomersof
3 were observed within 10 min, and the reaction was complete the
next day. 1H NMR signals (ppm) in acetonitrile-d3: 8.84 (s, H6/
60), 8.73 (s, H6/60), 8.29 (overlapping d, H3/30), 8.05 (d, H4/40),
6.29 (b, NH), 5.85 (b, NH), 5.34 (NH), 4.52 (NH), 3.06 (m, CH2),
2.49 (m, CH2), 2.47 (s, 5/50-CH3), 2.08 (s, CCH3), 1.87 (s, CCH3),
1.82 (m, CH), 1.22 (m, CH), 0.95 (d, CH3), 0.56 (d, CH3).
X-ray quality crystals of the E0 isomer of 3 were produced
upon slow evaporation of the solution of the crystalline material
(10 mg) in a 1:5 (v/v) mixture of acetonitrile/diethyl ether.
[Re(CO)3(5,50-Me2bipy)(HNC(CH3)NHC(CH3)3)]BF4 (4). The
method described above but with tert-butylamine (65 μL, 0.60
mmol) produced a yellow crystalline precipitate (yield, 32 mg,
49%), but the reaction time was longer (4 days). 1H NMR signals
(ppm) in acetonitrile-d3: 8.85 (s, 2H, H6/60), 8.30 (d, 2H, H3/30),
8.06 (d, 2H, H4/40), 6.11 (b, 1H, NH), 4.30 (1H, NH), 2.47 (s, 6H,
5/50-CH3), 2.01 (s, 3H, CCH3), 0.80 (s, 9H, CH3).
Challenge Reactions. A 5 mM solution of [Re(CO)3(5,50-
Me2bipy)(HNC(CH3)NHCH(CH3)2)]BF4 (2) in acetonitrile-d3
(600 μL) was treated with a 5-fold excess of 4-dimethylamino-
pyridine (2.0 mg, 25 mM), and the solution was monitored by 1H
NMR spectroscopy. A similar experiment was conducted in
CDCl3.
On addition of a 10% excess of tert-butyl amine (6.5 μL) to the
10 mMsolution, NMR signals of a mixtureof E0 and Z isomersof
4 were observed only after 1 h (reaction time, ∼4 days). 1H NMR
signals (ppm) in acetonitrile-d3: 8.85 (s, H6/60), 8.73 (s, H6/60),
8.30 (overlapping d, H3/30), 8.06 (d, H4/40), 6.11 (b, NH), 5.97
(b, NH), 5.38 (NH), 4.30 (NH), 2.47 (s, 5/50-CH3), 2.01 (s, CCH3),
1.95 (s), 1.38 (s, CH3), 0.80 (s, CH3). The resulting solution
yielded X-ray quality crystals upon slow evaporation.
[Re(CO)3(5,50-Me2bipy)(HNC(CH3)NHCH2C6H5)]BF4 (5).
The method described above but with benzylamine (66 μL,
0.60 mmol) produced a yellow crystalline precipitate; yield, 38
mg (55%). 1H NMR signals (ppm) in acetonitrile-d3: 8.64 (s, 2H,
H6/60), 7.96 (d, 2H, H3/30), 8.05 (d, 2H, H4/40), 7.15 (t, 1H), 7.06
(t, 2H), 6.91 (b, 1H, NH), 6.09 (d, 2H), 4.38 (1H, NH), 3.94
(d, 2H, CH2), 2.47 (s, 6H, 5/50-CH3), 2.18 (s, 3H, CCH3).
On addition of a 10% excess of benzylamine (7 μL) to the
10 mM solution, NMR signals of a mixture of E0 and Z isomers
of 5 were observed within 15 min, and the reaction was complete
the next day. 1H NMR signals (ppm) in acetonitrile-d3: 8.76 (s,
H6/60), 8.64 (s, H6/60), 8.27 (d, H4/40), 8.05 (overlapping d, H3/
30 Z and H4/40 E0), 7.96 (d, H4/40), 7.38 (t, benzyl), 7.31 (t,
benzyl), 7.22 (d, benzyl), 7.15 (t, benzyl), 7.06 (t, benzyl), 6.91 (b,
NH), 6.32 (b, NH), 6.09 (d, benzyl), 5.54 (NH), 4.45 (d, CH2),
4.30 (NH), 3.94 (d, CH2), 2.44 (s, 5/50-CH3), 2.18 (s, CCH3), 1.84
(s, CH3). The resulting solution yielded X-ray quality crystals
upon slow evaporation.
Results and Discussion
Synthesis. Syntheses of [Re(CO)3(5,50-Me2bipy)(HNC-
(CH3)NHR)]BF4 complexes were carried out in aceto-
nitrile (R=isopropyl (2), isobutyl (3), tert-butyl (4), benzyl
(5), and methyl (6)) at room temperature (Figure 2). For
R=H (7), acetonitrile or chloroform was used. Reactions
were monitored at intervals of 10 min, 1 h, and 1 to 4 days
(sometimes also 5 min, 30 min, or 6 h) by NMR spectros-
copy. Times required for completion of reaction varied
(∼6 h for 6; ∼1 day for 2, 3, 5, 6, and 7; and ∼4 days for 4).
For compounds 2 to 6, the ratio of E0 to Z isomers re-
mained the same throughout the course of the reaction.
For 7, which has a symmetrical remote nitrogen group,
the isomer designation is restricted to E and Z; experi-
mentally, a trace amount of the E isomer was observed in
addition to the major isomer, Z.
Structural Results. Complexesstructurally characterized
here, having the general formula, [Re(CO)3(5,50-Me2bipy)-
(HNC(CH3)NHR)]BF4 (R = alkyl, benzyl or H, Figures 3
and 4, and Supporting Information, Figure S1), exhibit a
distorted octahedral structure, with the three carbonyl
ligands occupying one face. The remaining three coordina-
tion sites are occupied by the two nitrogen atoms of the
Synthesis of [Re(CO)3(5,50-Me2bipy)(HNC(CH3)NHCH3)]BF4
(6). The method described above but with methylamine (∼30 μL,

Article
Inorganic Chemistry, Vol. 49, No. 15, 2010 7039
Figure 3. ORTEP plots of the cations in [Re(CO)3(5,50-Me2bipy)(E0-HNC(CH3)NHCH(CH3)2)]BF4 (2), [Re(CO)3(5,50-Me2bipy)(E0-HNC(CH3)-
NHCH2CH(CH3)2)]BF4 (3), [Re(CO)3(5,50-Me2bipy)(E0-HNC(CH3)NHC(CH3)3)]BF4 (4), and [Re(CO)3(5,50-Me2bipy)(E0-HNC(CH3)NHCH2C6H5)]BF4
(5). Thermal ellipsoids are drawn with 50% probability.
5,50-Me2bipy ligand and a nitrogen atom of the neutral
monodentate amidine ligand. Crystal data and details of
the structural refinement for these complexes are summar-
ized in Table 1. The atom numbering systems in the
ORTEP figures are used to describe the solid-state data.
The Re-C bond distances (not shown) involving the CO
group trans to the amidine group are not significantly
different from those of the other Re-C bonds.
The Re-N bond lengths of the amidine complexes
(Table 2) are comparable to those found for typical Re sp2
30
˚
nitrogen bond lengths, which range from 2.14 to 2.18 A.
Likewise, in all cases the bond distances from the amidine
carbon (C16) to the two nitrogen atoms, N3 and N4 (see
Figure 2), are closer to the average sp2 C double-bond length
3
˚
to N (CdN, ∼1.28 A) than to the average sp C single-bond
31
0
˚
length to N (C-N, 1.47 A). For example, in [Re(CO)3(5,5 -
Me2bipy)(HNC(CH3)NHCH(CH3)2)]BF4 (2) (Figure 3),
These [Re(CO)3(5,50-Me2bipy)(HNC(CH3)NHR)]BF4
complexes (R = isopropyl (2), isobutyl (3), tert-butyl (4),
benzyl (5), and methyl (6)) have an amidine ligand in the
E0 configuration in the solid state (Figure 3 and Support-
ing Information, Figure S1), even though NMR data
show that both E0 and Z isomers exist in acetonitrile
(discussed below). The molecular structure of [Re(CO)3-
(5,50-Me2bipy)(HNC(CH3)NH2)]BF4 (7) reveals that this
non sterically bulky amidine ligand has the Z configura-
tion (Figure 4), in contrast to the structures found here for
all other crystals.
˚
the bond distances are 1.308(3) A (N3-C16) and 1.339(3)
˚
˚
A (N4-C16). The N3-C16 (1.266(12) A) and N4-C16
(1.359(12) A) bond distances of 7 (Figure 4) are generally
˚
similar to the relevant distances (Table 2) of [Re(CO)3(5,50-
Me2bipy)(HNC(CH3)NHR)]BF4 complexes 2 to 5.
(30) He, H.; Lipowska, M.; Xu, X.; Taylor, A. T.; Carlone, M.; Marzilli,
L. G. Inorg. Chem. 2005, 44, 5437–5446.
(31) Allen, F. H.; Kennard, O.; Watson, D. G.; Brammer, L.; Orpen,
A. G.; Taylor, R. J. Chem. Soc. Perkin Trans. 2 1987, S1–S19.

7040 Inorganic Chemistry, Vol. 49, No. 15, 2010
Perera et al.
The relevant angles of the amidine moiety are all ∼120°
(Table 2), and these facts all indicate electron delocaliza-
tion along the N-C-N bonds. Furthermore, in all cases
as exemplified by 2, the N4-C16 bond is slightly but
significantly longer than the N3-C16 bond (Table 2),
suggesting slightly less double-bond character. These
slight differences are reflected in the ease of rotation
about the N-C bonds.
The orientation of the amidine ligand of 4 is different
from that of compounds 2, 3, and 7. (Supporting Infor-
mation, Figure S2 shows different orientations of the
amidine ligands when the Me2bipy plane is oriented in a
similar manner). These differences in orientation are
present in the solid state; there are no indications to imply
their presence in solution.
The molecular structure of the benzylamidine complex,
[Re(CO)3(5,50-Me2bipy)(HNC(CH3)NHCH2C6H5)]BF4
(5, Figure 3), shows that the phenyl ring of the amidine
moiety is stacked above the bipyridine ligand. The clos-
est C-to-C non-bonded distances are 3.584, 3.586, and
˚
3.953 A. Stacking is depicted in side and top-down views
of the stacked rings in the Supporting Information,
Figure S3. Effects of this stacking on NMR shifts of the
bipyridine signals are discussed later.
NMR Spectroscopy. General Considerations. All com-
plexes reported in this study were characterized by NMR
spectroscopy in acetonitrile-d3, and selected complexes were
studied in other solvents. COSY and ROESY experiments
aided in the assignment of the signals of complexes 2, 3, and
4 in acetonitrile-d3 and of 2 in CDCl3 and CD2Cl2. (For the
NMR discussion, the atom numbering in Figure 2 is used.)
NMR spectra recorded immediately after dissolving crystals
of [Re(CO)3(5,50-Me2bipy)(HNC(CH3)NHR)]BF4 (R =
isopropyl (2), isobutyl (3), tert-butyl (4), benzyl (5) and
methyl (6)) complexes in several solvents showed two to
three sets of signals, which changed in intensity until
equilibrium was reached. In general, one set of signals
grew slowly, and all other sets (one or two sets, depending
on solvent) decreased with time.
Because rotation about the bond between the amidine
carbon and the Re-bound N is expected to be slow com-
pared to the rotation about the bond between the amidine
carbon and the remote N, this behavior indicates that the
first signals observed are from the E0 or E isomer (the solid
has the E0 isomer). This reasoning allowed us to assign
unambiguously the initial set(s) of peaks to an isomer
with the amidine ligand in the E0 or E configuration and
the second set to the isomer with the amidine ligand in the
Z0 or Z configuration. However, 2D NMR data in com-
bination with studies in several solvents and in mixtures
of solvents allowed us to conclude that in every case the
E0 and Z isomers of [Re(CO)3(5,50-Me2bipy)(HNC(CH3)-
NHR)]BF4 complexes coexisted in polar solvents such as
acetonitrile-d3. Only in solvents of low polarity do we
observe signals for the E isomer. The Z0 isomer was not
Figure 4. ORTEP plot of the cation in [Re(CO)3(5,50-Me2bipy)(Z-HNC-
(CH3)NH2)]BF4 (7). Thermal ellipsoids are drawn with 50% probability.
0
˚
Table 2. Selected Bond Distances (A) and Angles (deg) for [Re(CO)3(5,5 -Me2bipy)(HNC(CH3)NHR)]BF4 Complexes
R=isopropyl
R=isobutyl
R= tert-butyl
R=benzyl
R=H
2
3
4
5
7
Bond Distances
Re-N1
Re-N2
Re-N3
N3-C16
N4-C16
2.1829(19)
2.1819(18)
2.1810(18)
1.308(3)
2.167(3)
2.178(3)
2.174(3)
1.307(4)
1.331(4)
2.183(2)
2.178(3)
2.158(2)
1.302(4)
1.340(4)
2.1746(19)
2.1874(18)
2.181(2)
1.298(3)
1.341(3)
2.169(6)
2.180(6)
2.156(7)
1.266(12)
1.359(12)
1.339(3)
Bond Angles
N1-Re-N2
75.14(7)
85.31(7)
81.33(7)
109.5(18)
135.43(16)
114.9(18)
122.9(2)
121.45(19)
115.64(18)
125.25(18)
74.89(10)
83.92(10)
80.76(10)
111(3)
135.4(2)
113(3)
123.1(3)
121.4(3)
115.4(3)
124.9(3)
74.88(9)
79.73(9)
79.89(9)
111(2)
138.7(2)
110(2)
123.6(3)
120.2(3)
116.2(3)
126.8(2)
75.24(7)
85.78(7)
77.56(7)
115(2)
135.59(16)
110(2)
123.4(2)
120.8(2)
115.8(2)
123.9(2)
75.3(3)
80.4(3)
84.6(3)
112.5
135.1(7)
112.5
N1-Re-N3
N2-Re-N3
Re-N3-H3N
Re-N3-C16
C16-N3-H3N
N3-C16-N4
N3-C16-C17
N4-C16-C17
C16-N4-C18
123.2(10)
121.6(10)
115.1(11)
a
a Not applicable.

Article
Inorganic Chemistry, Vol. 49, No. 15, 2010 7041
found in any solvent. As an illustration of our solution
studies and 2D NMR analysis of amidine complexes in
polar solvents, we first discuss 2 (Figure 3) in acetonitrile-
d3; we then extend the discussion to other compounds in
this solvent and eventually to other solvents.
to solvent. For example in acetonitrile-d3, CDCl3, CD2Cl2,
and DMSO-d6, the N4H and CH coupling values to
the CH2 signal of [Re(CO)3(5,50-Me2bipy)(HNC(CH3)-
NHCH2CH(CH3)2)]BF4 (3) fall within the range of 6.0-
6.6 Hz for all isomers (E0, E, or Z). The J values of the
isobutyl methyl signal coupling to the CH signal are
∼6.6-6.7 Hz in all solvents. The 5,50-Me2bipy H3/30-to-
H4/40 coupling constants have nearly invariant values of
8.3-8.4 Hz in allsolvents for allisomers of allcompounds.
NMR Studies of [Re(CO)3(5,50-Me2bipy)(HNC(CH3)-
NHCH(CH3)2)]BF4 (2) in Acetonitrile-d3. Upon dissolu-
tion of 2 (Figure 3) in acetonitrile-d3, the NMR spectrum
showed two sets of signals (Figure 5). The less intense
set of signals grew in intensity until equilibrium was
reached (∼15 min, Figure 5). Integration as well as the slow
isomerization of the complex on dissolution allowed us
to identify all signals attributable to each of the two sets.
Two singlets for the CCH3 groups derived from aceto-
nitrile occur at 1.89 (Z0 or Z) and 2.07 (E0 or E) ppm. The
four broad peaks between 4.5 and 6.1 ppm (Figure 6) were
identified as NH signals by addition of K2CO3/D2O. A
CH multiplet at 3.15 ppm (integrating to one proton) for
the E0 or E isomer has a COSY cross-peak with the CH3
doublet (0.74 ppm) from the isopropyl group (Figure 7).
This CH multiplet correlated in turn with an NH signal at
6.10 ppm, assigning this signal to N4H (Table 3). (See
Figure 2 for designations of the N3H and N4H protons.)
Likewise, a similar correlation was observed for peaks of
the Z0 or Z isomer of 2; the CH3 doublet (ppm) correlated
with the CH multiplet (3.69 ppm), which correlated with
the N4H signal at 5.57 ppm (Figure 7).
The coupling constants (J) are uninformative and do
not change appreciably from isomer to isomer or solvent
Figure 5. 1H NMR spectra as a function of time after crystals of the
E0 isomerof[Re(CO)3(5,50-Me2bipy)(HNC(CH3)NHCH(CH3)2)]BF4 (2)
were dissolved in acetonitrile-d3 at 25 °C.
A strong NOE cross-peak seen in the ROESY spectrum
(Supporting Information, Figure S4) between the N3H
signal (5.33 ppm) and the CCH3 singlet (1.89 ppm) for the
Z0 or Z isomer, together with a strong NOE peak between
this CCH3 singlet and the CH multiplet (3.69 ppm),
establishes that this is the Z isomer. The absence of a
N4H-CCH3 NOE cross-peak excludes the presence of
the Z0 isomer. There is no NOE peak between the N3H
signal (4.51 ppm) and the CCH3 signal (2.07 ppm) attrib-
utable to the E0 or E isomer, a result consistent with the
assignment of these signals to the E0 isomer. The presence
1
Figure 6. H NMR spectrum of [Re(CO)3(5,50-Me2bipy)(HNC(CH3)-
NHCH(CH3)2)]BF4 (2) in acetonitrile-d3 at 25 °C.
1
Figure 7. H-1H COSY NMR spectrum of [Re(CO)3(5,50-Me2bipy)(HNC(CH3)NHCH(CH3)2)]BF4 (2) in acetonitrile-d3 at 25 °C.

7042 Inorganic Chemistry, Vol. 49, No. 15, 2010
Perera et al.
1
Table 3. Selected H NMR Shifts (ppm) for Re(CO)3[(5,50-Me2bipy)(HNC-
Table 4. Distribution (%) of E0 and Z Isomers of [Re(CO)3(5,50-Me2bipy)(HNC-
(CH3)NHR)]BF4 in Acetonitrile-d3 at 25 °C
(CH3)NHR)]BF4 in Acetonitrile-d3 at 25 °C
CH3
(formerly
R
E0
Z
a
methyl (6)
isopropyl (2)
isobutyl (3)
tert-butyl (4)
benzyl (5)
H (7)
66
64
65
82
83
12
34
36
35
18
17
88
R
isomer N3H
N4H
CH3CN) N4CHn H6/60
methyl (6)
E0
Z
E0
Z
E0
Z
E0
Z
E0
Z
E0
Z
4.50
5.29
4.51
5.33
4.52
5.34
4.30
5.38
4.30
5.54
f
6.80
5.90
6.10
5.57
6.29
5.85
6.11
5.97
6.91
6.32
f
2.07
1.86
2.07
1.89
2.08
1.87
2.01
1.95
2.18
1.84
2.12
1.83
2.25(d)b 8.83
2.87(d)b 8.73
3.14(m)c 8.84
3.69(m)c 8.74
2.49(m)d 8.84
3.06(m)d 8.73
isopropyl (2)
isobutyl (3)
tert-butyl (4)
benzyl (5)
H (7)
(HNC(CH3)NHCH2C6H5)]BF4 (5, Figure 3), the Z isomer
H6/60, H4/40 and H3/30 signals (8.76, 8.05, and 8.27 ppm,
respectively) have the typical values; however, the signals
for the E0 isomer all appear upfield (8.64, 7.96, and 8.05
ppm, respectively) to typical values of the E0 isomer of com-
pounds 2, 3, 4, and 6 (see Experimental Section). (For 5, the
upfield-shifted H3/30 doublet of the E0 isomer overlaps with
the H4/40 doublet of the Z isomer.)
e
e
8.85
8.73
3.94(d)d 8.64
4.45(d)d 8.76
8.79
5.45
6.30,
5.93(b)
8.75
a The N4CHn signals vary in multiplicity according to the R group.
b n = 3. c n = 1. d n = 2. e n = 0. f Not observed.
The upfield shifts of the 5,50-Me2bipy signals of the
E0 isomer of 5 can be reasonably explained only by the
stacking of the phenyl moiety over the 5,50-Me2bipy
ligand in solution. This explanation was apparent when
the complex was first prepared, and it derived support
when the solid-state structure showed a stacking interac-
tion (Supporting Information, Figure S3). As we noted
above and for reasons expanded upon below, the major
form that can be present upon dissolution has to be either
the E0 or the E isomer. The E isomer (Figure 1) cannot
stack and can be excluded because the upfield shifts
cannot be explained without phenyl/5,50-Me2bipy stack-
ing. Free rotation around the Re-N3 bond, which must
occur in solution, explains both the presence of one signal
for each type of 5,50-Me2bipy proton and the upfield shifts
of the H6/60 and H4/40 signals.
Influence of Isomerization Rate on NMR Spectra of [Re-
(CO)3(5,50-Me2bipy)(HNC(CH3)NHCH2CH(CH3)2)]BF4
(3). 1H NMR spectra of 3 (5 mM) recorded from -15 to
55 °C in acetonitrile-d3, showed no change in the equilib-
rium ratios of the H6/60 and the NH NMR signals and no
major shift in these signals. Thus, the E0-to-Z interchange
is too slow on the NMR time scale to influence spectra.
The presence of only small EXSY cross-peaks at 65 °C in
a ROESY spectrum in acetonitrile-d3 (not shown) fur-
ther confirms that the interchange between the E0 and
Z isomers of 3 is slow on the NMR time scale. In addition,
at -15 °C no additional signals, which would indicate the
presence of the E or Z0 isomers, were observed.
Dependence of the E0:Z Equilibrium Ratio on the Nature
of R in Acetonitrile. NMR data for a series of amidine
complexes in acetonitrile-d3 at room temperature allowed
us to determine the E0:Z equilibrium ratio (Table 4). The
similarity of these E0:Z ratios for 2 (R=isopropyl), 3 (R=
isobutyl), and 6 (R = methyl) indicates that a ratio of
∼65:35 may be the “normal” ratio of E0 to Z isomers in
acetonitrile. Put differently, the ratio is dictated by the
relative steric bulk of the N4H and the CCH3, which project
toward the equatorial plane, and by the natural relative
energy of the molecular orbitals. The higher equilibrium
ratio(E0:Z= 82:18) for 4(R=tert-butyl) thanthe baseline
value establishes that as the bulk of the amidine ligand
R substituent rises above a threshold size, such as for R=
tert-butyl, the high bulk favors the E0 isomer. Likewise,
although the benzyl group in 5 has only moderate bulk,
of a very strong NOE peak between the N3H signal
(4.51 ppm) and the CH multiplet (3.15 ppm) confirms
beyond doubt that this isomer indeed has the E0 config-
uration because the distance between the N3H and CH
protons is short for the E isomer (2.18 A) but is too long
˚
(3.51 to 4.39 A, Supporting Information) to give a strong
NOE peak for all of the other isomers. Thus, we conclude
that the two sets of signals observed for 2 in acetonitrile-d3
arise from the E0 and Z isomers. The alternative conclu-
sion that the signals arise from a mixture of rapidly
interconverting (E0/E and Z0/Z) isomers is not supported
by the solvent and temperature dependence studies de-
scribed below.
The 5,50-Me2bipy signals of 2were completely assigned by
using NOE and COSY cross-peaks between the 5/50 CH3
signals and the H6/60 and H4/40 signals for 2 (Supporting
Information, Figure S5). The occurrence of only one set of
5,50-Me2bipy signals for each isomer indicates rapid rota-
tion about the Re-N3 bond.
0
˚
NMR Studies of Other [Re(CO)3(5,50-Me2bipy)(amidine)]-
BF4 Complexes in Acetonitrile-d3. The same methods de-
scribed above revealed that two isomers formed upon dis-
solution of crystals for compounds 3 to 6. For all except 5
(see below), the clear chemical shift patterns for the H6/60
signals in acetonitrile-d3 (Table 3) indicate that these are E0
and Zisomers. For 3and 4, these conclusions were supported
by 2D NMR spectra. For both isomers of [Re(CO)3(5,50-
Me2bipy)(HNC(CH3)NHR)]BF4 (R=isopropyl (2), isobu-
tyl (3), tert-butyl (4), and methyl (6)) complexes, the N3H
signals (N bound to Re) are more upfield (4.30-5.34 ppm)
than the N4H signals (5.57-6.80 ppm) in acetonitrile-d3. The
signals of the methyl group derived from acetonitrile range
from 2.01 to 2.08 ppm for the E0 isomers and from 1.86 to
1.95 ppm for the Zisomers. Thus, for the typical R group, the
isomers are easily identified.
Upfield 5,50-Me2bipy Signals of [Re(CO)3(5,50-Me2bipy)-
(HNC(CH3)NHCH2C6H5)]BF4 (5) Arising from Stacking.
For complexes 2, 3, 4, and 6, with non-anisotropic R
groups, the average shifts in acetonitrile-d3 are 8.84
(H6/60), 8.04 (H4/40), and 8.27 (H3/30) ppm for E0 isomer
signals and 8.73 (H6/60), 8.04(H4/40), and 8.29(H3/30) ppm
for Z isomer signals. (The H4/40 and H3/30 signals of the
E0 isomer overlap with the H4/40 and H3/30 signals of the
Z isomer, in some cases.) For [Re(CO)3(5,50-Me2bipy)-

Article
Inorganic Chemistry, Vol. 49, No. 15, 2010 7043
In CD2Cl2, as found in the less polar CDCl3 solvent, 2
exhibits three sets of H NMR signals. Two sets of these
1
signalshaveessentially equalintensityand areconnectedby
E-E0 EXSY peaks (ROESY spectrum at 0 °C, not shown)
as follows (ppm): CH multiplets, 3.51 and 2.95; N3H, 5.11
and 4.06; N4H, 5.96 and 5.85; and CH3 doublets, 1.03 and
0.80. The third set of signals is established as belonging
to the Z isomer by NOE peaks relating the CCH3 signal
(1.97 ppm) to the N3H signal (5.16 ppm) and to the CH
multiplet of the isopropyl group (3.67 ppm). Equilibrium
for 2 in CD2Cl2 (E0:E:Z= 19:19:62) is reached in 3 h.
Mixed-Solvent Studies Exploring the E0 Isomer Signals
of 2. As mentioned above, in the polar solvent aceto-
nitrile-d3, the signals for the E0 isomer could alternatively
represent a rapidly interconverting mixture of E and E0
isomers. In solvents such as CD2Cl2, the separate signals
for these isomers can be observed. Thus, in solvents of low
polarity, it is clear that both isomers are present and that
interchange is slow on the NMR time scale. We designed
an experiment to assess whether in acetonitrile-d3 a sig-
nificant amount of an E isomer in rapid exchange with the
major E0 isomer might be present but undetectable in
acetonitrile-d3.
We initiated our study with a freshly prepared CD2Cl2
solution because equilibrium was not reached for ∼3 h
after crystals of 2 were dissolved in CD2Cl2. The E iso-
mer was initially abundant (E0:E:Z = 37:37:28; Figure 8,
bottom). Addition of 1% and then 2% of acetonitrile-d3
into the CD2Cl2 solution of 2 (these first two additions
were recorded prior to equilibrium to observe abundant
E0 and E isomers) showed an increase in the relative
abundance of the E0 isomer, as reflected in the intensities
of the H6/60 signals (Figure 8). At 21% acetonitrile-d3, the
E0 isomer H6/60 signal was observed but not the E isomer
signal (Figure 8). Thus, the one set of signals observed in
100% acetonitrile-d3 is that of the E0 isomer, with a
negligible (if any) contribution of an E isomer.
1
Figure 8. H NMR spectra illustrating the distribution of E0, E, and Z
isomers of [Re(CO)3(5,50-Me2bipy)(HNC(CH3)NHCH(CH3)2)]BF4 (2)
in CD2Cl2 (with percent acetonitrile-d3 noted on trace) and acetonitrile-d3
at 25 °C. Note: the bottom two spectra were recorded before equilibrium
was reached.
phenyl/5,50-Me2bipy stacking favors the E0 isomer in aceto-
nitrile (Table 4).
Isomers of 2 Present in Less Polar Solvents, CD2Cl2 and
CDCl3. When crystals of [Re(CO)3(5,50-Me2bipy)(HNC-
(CH3)NHCH(CH3)2)]BF4 (2) were dissolved in CD2Cl2
and CDCl3, signals for three isomers were observed
(Figure 8 and Supporting Information, Figure S6). Two
sets of signals maintained the same ratio and decreased
with time. The third set grew in intensity and has all the
characteristics expected for the Z isomer.
NMR Studies of Other [Re(CO)3(5,50-Me2bipy)(amidine)]-
1
BF4 Complexes in CD2Cl2 and CDCl3. H NMR data for
complexes 2, 3, and 4 in CD2Cl2 and CDCl3 (Supporting
Information, Table S1) indicate that complexes 3 and 4
have E0, E (in most cases), and Z isomers, as found above
for 2. In CDCl3 and CD2Cl2 (as in acetonitrile-d3), the
signals of the methyl group derived from acetonitrile have a
more downfield shift (∼2.2 ppm) characteristic for the
E and E0 isomers and a more upfield shift (∼2.0 ppm)
characteristic for the Z isomer. An N3H signal having a
shift upfield of 4.60 ppm (Table 3 and Supporting Informa-
tion, Table S1) in acetonitrile-d3, CD2Cl2, or CDCl3 was
another indication allowing the assignment of signals to
the E0 isomer. As found in the case of every solvent for
compounds with non-anisotropic R groups, the H6/60
signal of the E0 isomer in CD2Cl2 or CDCl3 is more down-
field than that of the E isomer, which in turn is more
downfield than this signal for the Z isomer (Supporting
Information, Table S1). This relationship holds true for all
solvents, including DMSO-d6 (see below).
In CDCl3, of the two sets of signals that decreased with
time, one set had medium intensity (designated as m),
and the other had weak intensity (designated as w). In a
ROESY spectrum of 2 (Supporting Information, Figure
S6) in CDCl3 at 25 °C, EXSY peaks between the respective
signals of the m and w sets are present as follows (ppm):
CH multiplets, 3.50 (m) and 3.13 (w); N3H, 5.22 (m) and
4.30 (w); N4H, 6.07(m) and 5.77(w); and CH3 doublets,
1.07(m) and 0.80 (w). The m signals were assigned to the
E isomer by an N4H-N3H NOE peak. Also, an NOE
cross-peak between the CCH3 (2.23 ppm) signal and the
CH multiplet (3.50 ppm) establishes without doubt that
the m set belongs to the E isomer. The w signals belong to
the E0 isomer, as confirmed above by EXSY peaks. The
most intense set of signals was assigned to the Z isomer by
NOE cross-peaks relating the CCH3 signal (2.03 ppm) to
the N3H signal (5.75 ppm) and to the CH multiplet of the
isopropyl group (3.67 ppm). A CCH3-N3H NOE peak
together with a CCH3-CH NOE peak is uniquely ex-
pected for the Z isomer (Figure 1). Thus, the three sets of
1H NMR signals observed for 2 in CDCl3 belong to the
E0, E, and Z isomers present in the equilibrium ratio of
5:32:63, respectively.
These shift analogies were supported by the spectral
changes upon dissolution of crystals, which also aided in
the assignments. Dissolution of crystals of 3 (R = iso-
butyl) in CDCl3 gave NMR features (Supporting Infor-
mation, Table S2) similar to those observed for 2. A 1H
NMR spectrum recorded in 5 min contained mostly one

7044 Inorganic Chemistry, Vol. 49, No. 15, 2010
Perera et al.
large set of peaks of the E isomer, identified by NMR
shifts that closely resemble those of the E isomer of 2.
Dissolution of crystals of 3 in CD2Cl2 also gave NMR
features similar to those observed for 2 (Supporting
Information, Tables S1 and S2), for which three distinct
sets of signals were observed, with features expected for
the three isomers.
1
A H NMR spectrum of 4 (R = tert-butyl) initially
recorded within 5 min of dissolution in CDCl3 showed
signals for the three isomers, with the E and E0 isomers in
abundance. With time, peaks assignable to the Z isomer
grew. However, the spectrum obtained upon dissolution
of 4 in CD2Cl2 has only two sets of signals. The peaks
clearly assignable to the Z isomer were sharp, and these
grew with time. Although the signals attributable to the
E0 isomer were broad at 25 °C, all signals in the regions
characteristic of the E and E0 isomers, including the
informative H6/60 signal, were sharp at 0 °C; these results
are consistent with the absence of the E isomer in this
solvent (Supporting Information, Tables S1 and S2).
Dependence of the E0:E and the E0:Z Ratios on Solvent
and Remote N Group Steric Bulk. Some E0:E:Z equili-
brium ratios have been mentioned above for different
solvents. In this section we discuss the data further
(Supporting Information, Table S2). Because the situa-
tion is complicated, we begin with a brief summary. The
E0 isomer is favored by high steric bulk and polar solvents.
The E isomer is insensitive to steric bulk but favored by
low solvent polarity. The Z isomer is favored by low steric
bulk and low solvent polarity.
For 2, E0:E:Z equilibrium ratios in CDCl3 (5:32:63)
versus CD2Cl2 (19:19:62) show that while the percentage
of the E0 isomer has increased, the abundance of the
Z isomer has remained more or less constant. In other
words, E0 has increased and E has decreased as the
dielectric constant of the solvent increased from 4.8
(CDCl3) to 9.1 (CD2Cl2). This same relationship is valid
for 3, and as mentioned above, signals for the E isomer
of 4 are no longer observed even in the relatively low pola-
rity solvent, CD2Cl2 (Supporting Information, Table S2).
For all [Re(CO)3(5,50-Me2bipy)(HNC(CH3)NHR)]BF4
compounds, the E0 isomer is least abundant in CDCl3
(5-15%). As shown in Supporting Information, Table
S2, the abundance of the E0 isomer (64-82%) was greater
in the higher dielectric constant (37.5) solvent, aceto-
nitrile-d3, and even higher in DMSO-d6 (dielectric con-
stant = 47.2). Although we do not discuss the NMR
studies in DMSO-d6 in detail, the isomer and signal assign-
ments followed the methods discussed for other solvents.
For example, an NMR spectrum of 2 recorded in DMSO-
d6 showed mostly one set of signals (E0) initially, and a
second set of peaks grew with time, reaching equilibrium
within 30 min (E0:Z = 78:22). The N4H signals of both the
E0 and Z isomersappear more downfield in DMSO-d6 than
in acetonitrile, consistent with the good H-bonding and
solvation properties of DMSO.
Figure 9. Scheme relating the identified isomers of [Re(CO)3(L)(HNC-
(CH3)NHR)]þ cations. The figure illustrates the likely pathway between
the Z and the E0 isomers, passing through the E isomer. Faster inter-
conversion occurs between E and E0 isomers, as rotation about the bond
between the amidineC and the remoteN isexpectedtobefaster compared
to rotation about the bond between the amidine C and the N bound to Re.
Z0 Isomer. While we have observed evidence for the
presence of E0, E, and Z isomers for complexes 2 and 3
in CDCl3 and CD2Cl2, the Z0 isomer appears not to
be present in several solvents ranging from CDCl3 to
DMSO-d6. The Z0 isomer is probably destabilized by
interligand steric clashes involving the basal plane defined
by the 5,50-Me2bipy N atoms and the trans carbonyl C
atoms (Supporting Information, Figure S7). Such clashes
are likely to be severe in octahedral complexes but less
important in square-planar complexes, and the Z0 isomer
has been reported for PtII complexes.14 Complexes with
the fac-{ReI(CO)3} core are generally sterically un-
demanding compared to other octahedral complexes be-
cause ReI-N bonds are longer than typical M-N bonds
for metal ions in an octahedral environment, and the CO
ligands are relatively non sterically bulky. Nevertheless,
even in this favorable case, steric clashes appear to pre-
clude formation of significant amounts of the Z0 isomer.
The unstable nature of the Z0 isomer suggests that the
pathway for the interconversion of the E0 to Z isomers
passes through the E isomer and not through the Z0
isomer. Furthermore, the findings for 2 in CDCl3,
namely, that the E0/E ratio remained constant while the
amount of Z increased with time and that only E0-to-E
EXSY cross-peaks were present in the ROESY spectrum,
suggest that the E0-to-E interconversion is facile. Thus,
the slow steps in the E0-to-Z interconversion are the E-to-
Z steps. A likely reaction pathway scheme is illustrated in
Figure 9.
NMR Studies of [Re(CO)3(5,50-Me2bipy)(HNC(CH3)-
NH2)]BF4 (7). NMR spectra of 7 in acetonitrile-d3 at
25 °C recorded at 15 min and 1 day were very similar,
containing two very broad NH signals (6.30 and 5.93 ppm)
and one fairly sharp NH signal at 5.45 ppm (each integrat-
ing to one proton) for the major Z isomer (Table 3). Sharp
H6/60 signals were observed, with the smaller signal down-
field, suggesting that ∼12% E0 isomerwasalsopresent. The
signal at 5.45 ppm did not shift with temperature and can be
assigned to N3H of the Z isomer. At 5 °C, the two broad
Compared to all solvents used for 2, 3, and 4, the
abundance of the Z isomer was lowest (13-22%) in
DMSO-d6 (Supporting Information, Table S2). The Z
isomer was favored most in CDCl3 (57-64% abun-
dance). Thus, a low dielectric solvent favors the Z isomer
for all [Re(CO)3(5,50-Me2bipy)(HNC(CH3)NHR)]BF4
compounds studied.

Article
Inorganic Chemistry, Vol. 49, No. 15, 2010 7045
NH signals of equal intensity became sharp (6.60 and
5.81 ppm). At 35 °C, the broad signals merged to give one
peak (6.09 ppm), the total intensity remaining the same.
These results indicate that the signals are from the -NH2
group of [Re(CO)3(5,50-Me2bipy)(HNC(CH3)NH2)]BF4
and that elevated temperature increases the rate of rota-
tion around the C-NH2 bond. The NH peaks of the
minor E0 isomer could not be identified. The H6/60 and
amidine CH3 signals were used to obtain the ratio of the
E0 and Z isomers (Table 4).
Structural analysis of fac-[Re(CO)3(5,50-Me2bipy)-
(HNC(CH3)OCH3)]BF4, an iminoether analogue of the
amidine complexes in this study, has revealed that the
iminoether ligand has the Z configuration in the solid
state.26 The oxygen of the iminoether ligand is less sterically
bulky than the NHR group of [Re(CO)3(5,50-Me2bipy)-
(HNC(CH3)NHR)]BF4 complexes. Combining these find-
ings for [Re(CO)3(5,50-Me2bipy)(ligand)]BF4 complexes in
which the ligand is an amidine or an iminoether, we suggest
that the Z configuration is favored electronically, but the
E0 configuration is favored by steric effects.
Robustness of the Isopropylamidine Ligation. When
[Re(CO)3(5,50-Me2bipy)(HNC(CH3)NHCH(CH3)2)]BF4
(2) in acetonitrile-d3 or in CDCl3 was treated with a 5-fold
excess of the basic 4-dimethylaminopyridine ligand, no
major changes in spectral features of the amidine com-
plex were observed, even up to 2 months, indicating that
the isopropylamidine ligand is not readily replaced. The
NMR signals of [Re(CO)3(5,50-Me2bipy)(4-dimethylamino-
pyridine)]BF4, synthesized as a control, did not change
with time in acetonitrile-d3 or in CDCl3.
possible to influence the stereochemistry. Thus, the config-
uration is influenced primarily by electronic and steric effects.
The E0 isomer crystallized for [Re(CO)3(5,50-Me2bipy)-
(HNC(CH3)NHR)]BF4 (R=methyl, isopropyl, isobutyl,
tert-butyl, and benzyl), whereas the Z isomer crystallized
for [Re(CO)3(5,50-Me2bipy)(HNC(CH3)NH2)]BF4. Increased
steric bulk of the R group favors the E0 configuration of the
amidine ligand because the alkyl group projects away from
the basal plane. The Z configuration is favored electronically,
as evidenced by the structure of [Re(CO)3(5,50-Me2bipy)-
(HNC(CH3)NH2)]BF4 and also by the structure of several
[Re(CO)3(L)(iminoether)]BF4 complexes analyzed in unpub-
lished work.26
The exchange reaction between the E0 and Z isomers
of [Re(CO)3(5,50-Me2bipy)(HNC(CH3)NHR)]BF4 (R = iso-
propyl, isobutyl and tert-butyl) complexes is too slow to be
observed on the NMR time scale. The isomerization rate is
slow because there is double-bond character in the bond
between the amidine C and the N bound to Re. However,
the isomerization rate between the E0 and E isomers is fast
because there is less double-bond character in the bond
between the amidine C and the remote N. Exchange between
the E0 and Z isomers is likely to follow a pathway through
the E isomer because the Z0 isomer (not observed) is likely to
be unstable as a result of strong steric clashes between the R
group and the basal ligands in this isomer.
Acknowledgment. Purchase of the diffractometer was
made possible by Grant LEQSF(1999-2000)-ENH-TR-
13, administered by the Louisiana Board of Regents. This
work was supported in part by the National Institutes of
Health (R37 DK038842).
Conclusions
Supporting Information Available: Crystallographic data for
[Re(CO)3(5,50-Me2bipy)(HNC(CH3)NHCH(CH3)2)]BF4 (2),
[Re(CO)3(5,50-Me2bipy)(HNC(CH3)NHCH2CH(CH3)2)]BF4
(3), [Re(CO)3(5,50-Me2bipy)(HNC(CH3)NHC(CH3)3)]BF4 (4),
[Re(CO)3(5,50-Me2bipy)(HNC(CH3)NHCH2C6H5)]BF4 (5), and
[Re(CO)3(5,50-Me2bipy)(HNC(CH3)NH2)]BF4 (7) in CIF for-
mat; ORTEP plot of the cation in [Re(CO)3(5,50-Me2bipy)(HNC-
(CH3)NHCH3)]BF4 (6); ROESY spectra of 2 in acetonitrile-d3
and CDCl3; figure depicting stacking in 5; and tables of selected
1H NMR shifts for 2, 3, and 4 in CDCl3 and CD2Cl2 and of the
distribution of E, E0, and Z isomers of 2, 3, and 4 in several sol-
vents compared to the distribution in acetonitrile-d3. This material
is available free of charge via the Internet at http://pubs.acs.org.
The [Re(CO)3(5,50-Me2bipy)(CH3CN)]BF4 complex (1)
readily forms [Re(CO)3(5,50-Me2bipy)(amidine)]BF4 com-
plexes upon treatment with amines. The facility of this
reaction for a low-oxidation-state, third-row transition metal
complex very likely reflects the fact that the CO ligands
reduce the electron density on the metal via π back-bonding.
In these complexes, the amidine ligand is attached robustly,
but it does not exhibit a trans influence.
Because the amidine grouping is not in a ring and the
ligand is monodentate, the configuration is not constrained.
Furthermore, there is no interligand hydrogen bonding