Influence of solvent on aromatic interactions in metal tris-bipyridine complexes

Article abstract of DOI:10.1021/ja973938+

The conformational properties of a series of iron(II) and ruthenium(II) tris-bipyridine complexes have been investigated in a range of solvents. The complexes are equipped with pendant aromatic esters attached by flexible aliphatic linkers, and aromatic i

Full text of DOI:10.1021/ja973938+

3402
J. Am. Chem. Soc. 1998, 120, 3402-3410
Influence of Solvent on Aromatic Interactions in Metal
Tris-Bipyridine Complexes
Gloria A. Breault,† Christopher A. Hunter,*,‡ and Paul C. Mayers‡
Contribution from the Krebs Institute for Biomolecular Science, Department of Chemistry, UniVersity of
Sheffield, Sheffield S3 7HF, U.K., and Zeneca Pharmaceuticals, Mereside, Alderley Park, Macclesfield,
Cheshire SK10 4TG, U.K.
ReceiVed NoVember 17, 1997
Abstract: The conformational properties of a series of iron(II) and ruthenium(II) tris-bipyridine complexes
have been investigated in a range of solvents. The complexes are equipped with pendant aromatic esters
attached by flexible aliphatic linkers, and aromatic interactions between the edge of the bipyridine units and
the face of the aromatic esters cause the complexes to fold up in solution. The extent of folding is assessed
1
using H chemical shifts and found to be strongly solvent-dependent. Strong intramolecular edge-to-face
aromatic interactions leading to stable folded structures are found in both polar solvents (water and alcohols)
and nonpolar solvents (chlorinated hydrocarbons), but solvents of intermediate polarity such as DMSO destabilize
the folded conformation. These results indicate that the aromatic interactions are dominated by a substantial
electrostatic contribution in organic solvents but are sufficiently nonpolar to take advantage of solvophobic
effects in polar solvents. This solvent dependence is likely to be a characteristic feature of any molecular
recognition process which involves a mixture of both polar and nonpolar interactions.
Introduction
studied the influence of solvent on a different host-guest system
where the major interactions controlling complexation are
H-bonds.5 In this case, the complexes are stable in nonpolar
organic solvents but become less stable as the polarity of the
solvent increases, and solvent molecules begin to compete for
H-bond sites on the host and guest.6,7
Desolvation is an important factor in controlling molecular
recognition phenomena in solution: the hydrophobic effect,1
removal of nonpolar residues from aqueous solvent, is one of
the major forces which drives protein folding;2 synthetic
receptors with carefully crafted cavities which are able to
completely exclude solvent, and hence do not need to be
desolvated for complexation to occur, show remarkably high
affinities for complementary guests.3 Diederich et al. carried
out a comprehensive study of the influence of solvent on the
interaction of a nonpolar aromatic guest (pyrene) with a com-
plementary nonpolar aromatic cavity.4 A straightforward cor-
relation was obtained between the stability of the host-guest
complex and the polarity of the solvent. The complex is very
stable in polar solvents which do not compete for the binding
sites and which contribute favorably to the binding free energy
through solvophobic effects. As the solvent polarity decreases,
the solvophobic contribution to binding decreases and competi-
tion of the solvent for binding sites on the host and guest
increases. The net result is a difference of 6 orders of magnitude
in the host-guest association constant.4c Hamilton et al. have
The picture which emerges from these kinds of experiments
is that complexes involving nonpolar interactions are stabilized
by polar solvents and complexes involving polar interactions
are stabilized in nonpolar solvents. This suggests that the way
in which a molecular recognition system responds to changes
in solvent can be used to derive information about the nature
of the interactions which stabilize it. However, most systems
of interest are complex and feature a mixture of different types
of noncovalent interaction. What is the role of desolvation in
these cases? For example, proteins which contain a suitable
mixture of polar and nonpolar interactions are able to stabilize
the same folded state in both polar and nonpolar solvents.8 In
this paper, we describe a system which shows a more complex
solvent dependence and discuss the implications for the nature
of the noncovalent interactions involved.
Experimental approaches to the study of noncovalent interac-
tions in synthetic model compounds have focused on two
scenarios: intermolecular interactions in host-guest systems4,9
and intramolecular interactions in conformationally flexible
‡ University of Sheffield.
† Zeneca Pharmaceuticals.
(1) (a) Hildebrand, J. H. Proc. Natl. Acad. Sci. U.S.A. 1979, 76, 194.
(b) Mirejovsky, D.; Arnett, E. M. J. Am. Chem. Soc. 1983, 105, 1112. (c)
Breslow, R.; Guo, T. Proc. Natl. Acad. Sci. U.S.A. 1990, 87, 167. (d) Muller,
N. Acc. Chem. Res. 1990, 23, 23. (e) Dill, K. A. Science 1990, 250, 297.
(f) Herzfeld, J. Science 1991, 253, 88. (g) Sharp, K. A.; Nicholls, A.; Fine,
R. F.; Honig, B. Science 1991, 252, 106. (h) Nelson, J. C.; Saven, J. G.;
Moore, J. S.; Wolynes, P. G. Science 1997, 277, 1793. (i) Lokey, R. S.;
Iverson, B. L. Nature 1995, 375, 303-305.
(5) Garcia-Tellado, F.; Goswami, S.; Chang, S. K.; Geib, S.; Hamilton,
A. D. J. Am. Chem. Soc. 1990, 112, 7393.
(6) Hydrogen-bonding hosts with improved complexation properties in
more polar solvents have been reported: (a) Furuta, H.; Magda, D.; Sessler,
J. L. J. Am. Chem. Soc. 1991, 113, 978. (b) Kelly-Rowley, A. M.; Cabell,
L. A.; Anslyn, E. V. J. Am. Chem. Soc. 1992, 114, 1900. (c) Fan, E.; Van
Arman, S. A.; Kincaid, S.; Hamilton, A. D. J. Am. Chem. Soc. 1993, 115,
369.
(2) (a) Dill, K. A. Biochemistry 1990, 29, 7133. (b) Lins, L.; Brasseur,
R. Faseb J. 1995, 9, 535.
(3) Chapman, K. T.; Still, W. C. J. Am. Chem. Soc. 1989, 111, 3075-
(7) For discussions of solvent effects upon hydrogen bonding, see:
Adrian, J. C.; Wilcox, C. S. J. Am. Chem. Soc. 1991, 113, 678.
(8) (a) Arakawa, T.; Goddette, D. Arch. Biochem. Biophys. 1985, 240,
21. (b) Sonnichsen, F. D.; Van Eyk, J. E.; Hodges, R. S.; Sykes, B. D.
Biochemistry 1992, 31, 8790. (c) Storrs, R. W.; Truckses, D.; Wemmer, D.
E. Biopolymers 1992, 32, 1695.
3077.
(4) (a) Ferguson, S. B.; Sanford, E. M.; Seward, E.; Diederich, F. J.
Am. Chem. Soc. 1991, 113, 5410-5419. (b) Smithrud, D. B.; Wyman, T.
B.; Diederich, F. J. Am. Chem. Soc. 1991, 113, 5420-5426. (c) Smithrud,
D. B.; Diederich, F. J. Am. Chem. Soc. 1990, 112, 339-343.
S0002-7863(97)03938-3 CCC: $15.00 © 1998 American Chemical Society
Published on Web 03/31/1998

Influence of SolVent on Aromatic Interactions
Scheme 1. Synthesis of Ligand 6a
J. Am. Chem. Soc., Vol. 120, No. 14, 1998 3403
a Reagents, conditions, and yields: (a) dehydrogenated raney nickel, reflux, 48 h (38%); (b) NBS, cat. AIBN, CCl4, reflux, 4 h (47%); (c)
(EtO2C)2CHLi, DMF, 50 °C, 1 h (65%); (d) wet DMSO, cat. NaCl, reflux, 2.5 h (84%); (e) NaOH, H2O, reflux, 2 h then H3O+ (90%); (f) PhOH,
EDC, cat. DMAP, CH2Cl2, 0 °C to rt, 18 h (74%).
molecules.10 However, the limited solubility of these systems
has meant that, with a few notable exceptions, the experiments
have been carried out in a limited range of solvents. In this
paper, we describe the properties of a new system for studying
aromatic interactions based on metal tris-bipyridine complexes
linked to pendant aromatic esters by flexible side chains. We
demonstrate that these complexes fold up in a process driven
by intramolecular aromatic interactions and that the extent to
which the system is folded can be determined by changes in
1H NMR chemical shift in a variety of solvents. The folding
process is strongly solvent -dependent, and the unusual trends
in stability provide new insight into the nature of the aromatic
interactions involved.
decarbethoxylated using the procedure of Krapcho et al.13 to
give 414 (84%). Subsequent base hydrolysis gave the diacid 5
(90%) which was coupled with phenol using 1-(3-(dimethyl-
amino)propyl)-3-ethylcarbodiimide hydrochloride (EDC) to give
phenyl diester 6 (74%) (Scheme 1).
Iron(II) complexes of 4 and 6 were prepared quantitatively
by treatment of the ligand with a solution of 0.33 equiv of the
required iron salt: Fe(ClO4)2‚6H2O in 5% methanol/CH2Cl2
gave the perchlorate salts, and FeCl2‚4H2O in methanol gave
the chloride salts (Scheme 2).
The corresponding ruthenium(II) complexes were prepared
by treatment of 4 with RuCl3 in refluxing ethylene glycol,
followed by base hydrolysis and ion exchange to give RuII53
(PF6)2 in 77% yield. EDC coupling with phenol then gave the
hexaphenyl ester RuII63 (PF6)2 in 70% yield, and that with
n-pentanol gave the hexapentyl ester RuII73 (PF6)2 (47%)
(Scheme 3). Unsymmetrical ruthenium(II) complexes were
prepared by treatment of 1 with RuCl3‚3H2O and LiCl in
refluxing DMF to give the cis-dichloride RuII12‚Cl2 (83%), and
subsequent reaction with a slight excess of 4 in aqueous ethanol
followed by ion exchange gave RuII4‚12 (PF6)2 (72%).15 Base
hydrolysis followed by an EDC coupling with phenol gave the
diphenyl ester RuII6‚12 (PF6)2 (72%) (Scheme 4).
Synthesis
5,5′-Dimethyl-2,2′-bipyridine (1) was prepared11 and con-
verted to dibromide 212 following literature procedures.
A
modified malonate-type reaction with diethylmalonate lithium
salt in DMF gave tetraester 3 in 65% yield which was
(9) Adams, H.; Harris, K. D. M.; Hembury, G. A.; Hunter, C. A.;
Livingstone, D.; McCabe, J. F. J. Chem. Soc., Chem. Commun. 1996, 2531-
2532. (b) Adams, H.; Carver, F. J.; Hunter, C. A.; Morales, J. C.; Seward,
E. M. Angew. Chem., Int. Ed. Engl. 1996, 35, 1542-1544. (c) Ferguson,
S. B.; Diederich, F. Angew. Chem., Int. Ed. Engl. 1986, 25, 1127-1129.
(d) Kato, Y.; Conn, M. M.; Rebek, J. J. Am. Chem. Soc. 1994, 116, 3279-
3284. (e) Aoyama, Y.; Asakawa, M.; Yamagishi, A.; Toi, H.; Ogoshi, H.
J. Am. Chem. Soc. 1990, 112, 3145-3151. (f) Aoyama, Y.; Asakawa, M.;
Matsui, Y.; Ogoshi, H. J. Am. Chem. Soc. 1991, 113, 6233-6240. (g) Rebek,
J.; Nemeth, D. J. Am. Chem. Soc. 1986, 108, 5637-5638. (h) Jeong, K. S.;
Rebek, J. J. Am. Chem. Soc. 1988, 110, 3327-3328. (i) Muehldorf, A. V.;
Vanengen, D.; Warner, J. C.; Hamilton, A. D. J. Am. Chem. Soc. 1988,
110, 6561-6562. (j) Zimmerman, S. C.; Vanzyl, C. M.; Hamilton, G. S. J.
Am. Chem. Soc. 1989, 111, 1373-1381. (k) Zimmerman, S. C.; Kwan, W.
S. Angew. Chem., Int. Ed. Engl. 1995, 34, 2404-2406. (l) Zimmerman, S.
C. Top. Curr. Chem. 1993, 165, 71-102. (m) Schneider, H. J. Angew.
Chem., Int. Ed. Engl. 1991, 30, 1417-1436.
(10) (a) Eliel, E. L.; Allinger, N. L.; Angyal, S. J.; Morrison, G. A.
Conformational Analysis; Interscience: New York, 1965. (b) Gallo, E. A.;
Gellman, S. H. J. Am. Chem. Soc. 1993, 115, 9774-9788. (c) Newcomb,
L. F.; Haque, T. S.; Gellman, S. H. J. Am. Chem. Soc. 1995, 117, 6509-
6519. (d) Newcomb, L. F.; Gellman, S. H. J. Am. Chem. Soc. 1994, 116,
4993-4994. (e) Paliwal, S.; Geib, S.; Wilcox, C. S. J. Am. Chem. Soc.
1994, 116, 4497-4498. (f) Cozzi, F.; Cinquini, M.; Annunziata, R.; Dwyer,
T.; Siegel, J. S. J. Am. Chem. Soc. 1992, 114, 5729-5733. (g) Cozzi, F.;
Cinquini, M.; Annunziata, R.; Siegel, J. S. J. Am. Chem. Soc. 1993, 115,
5330-5331. (h) Cozzi, F.; Ponzini, F.; Annunziata, R.; Cinquini, M.; Siegel,
J. S. Angew. Chem., Int. Ed. Engl. 1995, 34, 1019-1020. (i) Cozzi, F.;
Siegel, J. S. Pure Appl. Chem. 1995, 67, 683-689.
Results and Discussion
Evidence for Intramolecular Aromatic Interactions. Dur-
ing our work with a range of tris-bipyridine metal complexes
such as FeII63, FeII43, RuII63, and RuII73, we noticed that the
1H NMR chemical shifts of protons H3 and H4 were very
sensitive to the nature of the ester substituent on the end of the
1
chain. For example, in chloroform the H NMR signal due to
H3 is shifted 0.54 ppm upfield in FeII63 (ClO4)2 relative to the
signal due to the same proton in FeII43 (ClO4)2 (Figure 1). If
1
we compare the H NMR spectra of the uncomplexed free
(11) (a) Sasse, W. H. F.; Whittle, C. P. J. Chem. Soc. 1961, 1347-
1350. (b) Badger, G. M.; Sasse, W. H. F. J. Chem. Soc. 1956, 616-620.
(12) Ebmeyer, F.; Vo¨gtle, F. Chem. Ber. 1989, 122, 1725-1727.
(13) Krapcho, A. P.; Lovey, A. J. Tetrahedron Lett. 1973, 957-960.
(14) (a) Hamilton, A. D.; Rubin, H.-D.; Bocarsly, A. B. J. Am. Chem.
Soc. 1984, 106, 7255-7257. (b) Hamilton, A. D. Private communication.
(15) (a) Lay, P. A.; Sargeson, A. M.; Taube, H. Inorg. Synth. 1986, 24,
291-299. (b) Sprintschnik, G.; Sprintschnik, H. W.; Kirsch, P. P.; Whitton,
D. G. J. Am. Chem. Soc. 1977, 99, 4947-4954.

3404 J. Am. Chem. Soc., Vol. 120, No. 14, 1998
Scheme 2. Synthesis of Iron(II) Complexesa
Breault et al.
a Reagents, conditions, and yields: (a) 0.33 equiv of Fe(ClO4)2‚6H2O, 5% methanol in CH2Cl2, 30 min (quantitative); (b) 0.33 equiv of FeCl2‚4H2O,
methanol, 30 min (quantitative).
a
Scheme 3. Synthesis of Ruthenium(II) Complexes RuII63 (PF6)2 and RuII73 (PF6)2
a Reagents, conditions, and yields: (a) RuCl3, HOCH2CH2OH, 200 °C, 0.5 h then NH4PF6/H2O; (b) NaOH, H2O, 100 °C, 1 h then HCl/NH4PF6
n
(77% from 4); (c) PhOH, EDC, cat. DMAP, CH2Cl2, 0 °C to rt, 18 h (70%); (d) C5H11OH, EDC, cat. DMAP, CH2Cl2, 0 °C to rt, 18 h (47%).

Influence of SolVent on Aromatic Interactions
J. Am. Chem. Soc., Vol. 120, No. 14, 1998 3405
Scheme 4. Synthesis of RuII12‚6 (PF6)2
a
Figure 1. Chemical shift changes observed for FeII63 (ClO4)2 and
free ligand 6 in chloroform. Shifts for the bipyridine protons of FeII63
(ClO4)2 are relative to those for FeII43 (ClO4)2 and the phenolic protons
are relative to those of 6. Shifts for 6 are relative to those for ethyl
ester 4.
Figure 2. Variation of chemical shift of bipyridine H3 with concentra-
tion for RuII63 (PF6)2 and RuII73 (PF6)2 in acetone-d6. ∆δ values are
quoted relative to those obtained at the highest recorded concentration
(57 mM).
acetone (Figure 2). For this system, the difference between the
chemical shift of the signal due to H3 in RuII63 (PF6)2 and the
corresponding proton in the alkyl ester reference compound
RuII73 (PF6)2 is -0.37 ppm. Dilution of the complexes resulted
in changes in chemical shift of less than 0.05 ppm, and both
complexes exhibited almost identical changes over the same
concentration range. Thus the small concentration-dependent
changes in chemical shift which are observed are not related to
the large difference observed between the aromatic and alkyl
esters. This indicates that even if ion pairing does take place
to a significant extent in this system, it has a negligible effect
on the observed differences in 1H NMR chemical shift in which
we are interested.
A more likely explanation of these observations is that the
large shifts observed for the aromatic esters are caused by
intramolecular aromatic interactions in the complex. These
interactions lead to close proximity of the phenyl esters and
the bipyridine groups in the metal complex which in turn causes
ring current-induced changes in chemical shift. The bipyridine
a Reagents, conditions, and yields: (a) RuCl3‚3H2O, LiCl, DMF,
reflux, 7 h (83%); (b) 1.1 equiv of 4, 50% aq ethanol, reflux, 3 h
then NH4PF6 (72%); (c) NaOH, water, reflux, 2.5 h then H3O+/
NH4PF6 (96%); (d) PhOH, EDC, cat. DMAP, CH2Cl2, 0 °C to rt, 18 h
(75%).
ligands 6 and 4, there is a 0.09 ppm downfield shift of H3, so
the very large difference observed in the metal complexes is
clearly not due to through-bond effects (Figure 1).
One possibility is that the large shifts are caused by ion-
1
pairing interactions with the counterions. However, H NMR
dilution studies revealed very small concentration-dependent
changes in chemical shift. Typical results are shown for the
signals due to proton H3 in RuII63 (PF6)2 and RuII73(PF6)2 in

3406 J. Am. Chem. Soc., Vol. 120, No. 14, 1998
Breault et al.
Figure 3. Conformational equilibria for tris-bipyridine complexes of type MII63 X2.
Figure 4. Chemical shift changes observed for RuII12‚6 (PF6)2 relative
Figure 5. NOEs observed in a 400 MHz ROESY spectrum of FeII63
(ClO4)2 in CDCl3.
to RuII12‚4 (PF6)2 in acetone-d6.
protons H3 and H4 experience large upfield shifts in FeII63 (PF6)2
relative to FeII43 (PF6)2, whereas the phenyl ester protons are
not significantly shifted relative to those of the free ligand 6
which implies that the bipyridine protons lie over the face of
the phenyl ester rings (Figure 1). Figure 3 illustrates confor-
mational equilibria which could explain the NMR data. In
conformation A, the ester group is directed away from the
complex and there are no aromatic interactions. Conformations
B and C show two different types of geometry in which there
is an interaction between the bipyridine protons and the face of
a phenyl ester ring. Large differences in chemical shift are not
observed between the uncomplexed free ligands 4 and 6 which
implies that bending back of the phenyl ester onto its own
bipyridine unit as in conformation C is not very probable.
However, it is possible that complexation of the bipyridine by
the metal cation significantly polarizes protons H3 and H4 which
increases the electrostatic interaction of these protons with the
π-electrons on the face of the phenyl ester ring. Therefore, to
distinguish conformations B and C, we prepared the unsym-
metrical complex RuII6‚12 (PF6)2.
the uncomplexed free ligands 6 and 4. The analogous H3 and
H4 protons on the dimethylbipyridine units (1) however show
significant upfield shifts (the values in Figure 4 are averaged,
since the two pyridine rings of each bipyridine unit are
nonequivalent but behave similarly). Correcting for the influ-
ence of metal coordination on the chemical shifts of the
bipyridine protons, it is clear that there are large upfield shifts
of the bipyridine protons on 1 and no shifts on 6 which proves
that the aromatic interaction in this system is due to type B
conformations rather than type C conformations (Figure 3). The
magnitudes of the shifts observed in RuII6‚12 (PF6)2 (-0.15 and
-0.09 ppm) are approximately one-half those observed in RuII63
(PF6)2 (-0.37 and -0.12 ppm), because only one phenyl ester
ring can interact with each set of dimethylbipyridine protons
whereas in complex RuII63 (PF6)2 in conformation B two phenyl
ester groups interact with each set of bipyridine protons.
Further evidence for the folded conformation of the aromatic
esters was obtained from two-dimensional ROESY experiments.
Cross-peaks connecting the signals due to the bipyridine protons
H3 and H4 and the signal due to the o-phenyl ester protons (CP1
and CP2 in Figure 5) indicate that these two parts of the
molecule are close in space (Figure 5). These observations are
clearly consistent with the folded conformation B shown in
Figure 3.
The differences in 1H NMR chemical shift between the
bipyridine protons of RuII6‚12 (PF6)2 and the corresponding alkyl
ester reference compound RuII4‚12 (PF6)2 in acetone are shown
in Figure 4. H3 and H4 of the bipyridine unit containing the
pendant phenyl esters (6) are shifted slightly downfield: the
shifts are in fact very similar to the differences observed for
We have also investigated the conformational properties of

Influence of SolVent on Aromatic Interactions
J. Am. Chem. Soc., Vol. 120, No. 14, 1998 3407
Figure 7. Variation of chemical shift with temperature for bipyridine
H3 of FeII63 (ClO4)2 relative to FeII43 (ClO4)2 in C2D2Cl4 (g295 K)
and CDCl3 (e295 K). The discontinuity at 295 K is due to the change
in solvent at this temperature.
Table 2. Changes in Chemical Shift of Iron(II) Complexes FeII63
a
(ClO4)2 and FeII63 Cl2 Relative to FeII43 (ClO4)2 and FeII43 Cl2
∆δ (ppm)
solvent
Z (kcal mol-1
)
H3
H4
H6
counterion
Figure 6. Typical low-energy conformation of RuII12‚6 obtained from
a Monte Carlo conformational search using the Macromodel imple-
mentation of the MM2 force field with chloroform solvation.
-
CDCl3
63.2
64.2
65.7
68.5
71.1
71.3
76.3
83.6
94.6
-0.54 -0.26 +0.02
ClO4
ClO4
ClO4
ClO4
ClO4
ClO4
-
CD2Cl2
acetone-d6
DMF-d7
DMSO-d6
CD3CN
iPrOD-d7
CD3OD
D2O
-0.48 -0.20
-0.34 -0.15
-0.25 -0.07
b
b
b
-
-
-
Table 1. Temperature-Dependent Changes in Chemical Shift of
FeII63 (ClO4)2 Relative to FeII43 (ClO4)2
-0.12 -0.08 +0.03
-0.32 -0.12 +0.01
-
-0.51 -0.18
b
Cl-
Cl-
Cl-
high temperature (Cl2DCCDCl2)
low temperature (CDCl3)
-0.56 -0.23 -0.06
-0.81 -0.31 -0.09
T (K)
∆δ H3 (ppm)
T (K)
∆δ H3 (ppm)
295
313
330
346
366
-0.500
-0.488
-0.473
-0.462
-0.445
295
268
258
243
228
-0.537
-0.568
-0.574
-0.593
-0.614
a All spectra were recorded at a concentration of 8 mM with the
exception of the D2O sample where a lower concentration (ca.. 2 mM)
was used due to limited solubility. b Values could not be obtained due
to overlapping signals in FeII63.
a All spectra were recorded at a concentration of 8 mM.
the aromatic and alkyl esters are significantly reduced, while
at low temperatures the differences increase. This is consistent
with the model in Figure 3 where there is an equilibrium
between a folded conformation (B) stabilized by attractive
aromatic interactions and an unfolded disordered conformation
(A). High temperatures shift the equilibrium toward the
disordered conformation (A), and low temperatures favor the
more ordered state (conformation B). It seems that the two
extreme conformations illustrated in Figure 3 are never fully
populated within the temperature range studied.
this system using molecular mechanics calculations. The X-ray
crystal structure of ruthenium(II) tris-bipyridine was retrieved
from the Cambridge Crystallographic Database and used as a
starting point for molecular modeling studies.16 One phenyl
ester side chain was attached, and a conformational search was
carried out using the Macromodel17 implementation of the MM2
force field with chloroform solvation and constraining the
ruthenium(II) tris-bipyridine core to the X-ray structure geom-
etry. In other words, we probed the conformational properties
of the flexible side chain only. Not surprisingly, a large number
of different low-energy conformations were obtained. However,
a number of these structures corresponded to conformation B,
and none of the low-energy structures were in conformation C,
which is consistent with the experimental observations. A repre-
sentative type B conformation is shown in Figure 6. We do
not suggest that this is the optimal or most populated conforma-
tion; it simply illustrates one of the conformations which is
populated to a significant extent and which is responsible for
the large ring current shifts observed in this system.
These experiments demonstrate that there is a strong intramo-
lecular aromatic interaction in this system. The differences
1
between the chemical shifts of the H NMR signals due to the
bipyridine protons H3 and H4 in the aromatic esters and the
corresponding alkyl ester control compounds provide a simple
measure of the position of the equilibrium between conformation
A and conformation B (Figure 3) and hence a direct measure
of the strength of the intramolecular aromatic interaction. The
solubility of metal complexes of this type can easily be con-
trolled by choice of counterions, so this represents an ideal
system for investigating the influence of solvent on the mag-
nitude of aromatic interactions.
Solvent Dependence of Aromatic Interactions. The per-
chlorate and chloride salts of the iron(II) complexes and the
hexafluorophosphate salts of the ruthenium(II) complexes
allowed us to record 1H NMR spectra in a wide range of
solvents. The results are summarized in Tables 2 and 3 and
Figure 8. The differences in chemical shift which we use to
quantify the aromatic interactions show an interesting variation
with solvent polarity (quantified by the parameter Z).18 Large
upfield shifts are observed in polar solvents, e.g., water, and as
The temperature dependence of the NMR shifts was inves-
tigated for FeII63 (ClO4)2 and FeII43 (ClO4)2 in tetrachloroethane-
1,1,2,2-d2 (for high temperatures) and CDCl3 (for low temper-
atures). The results are shown in Table 1 and Figure 7. At
higher temperatures, the differences in chemical shift between
(16) Fletcher, D. A.; McMeeking, R. F.; Parkin, D. J. Chem. Inf. Comput.
Sci. 1996, 36, 746-749.
(17) Mohamadi, F.; Richards, N. G. J.; Guida, W. C.; Liskamp, R.;
Lipton, M.; Caufield, C.; Chang, G.; Hendrickson, T.; Still, W. C. J. Comput.
Chem. 1990, 11, 440-467.

3408 J. Am. Chem. Soc., Vol. 120, No. 14, 1998
Breault et al.
Table 3. Changes in Chemical Shift of Ruthenium(II) Complex
tramolecular interactions between the pendant aromatic ester
groups and the aromatic core of the complex. These interactions
cause the complexes to fold up in solution, and the extent of
folding is strongly solvent-dependent. However, the relationship
between solvent polarity and folding (or the strength of the
aromatic interactions) is not simple. The aromatic interactions
are strongest in water, where there is a substantial hydrophobic
contribution. As the solvent polarity decreases, the interactions
decrease until we reach DMSO. Then the interactions start to
become more favorable again, and by the time we reach
chloroform, they are almost as strong as they are in water. This
indicates that there is a substantial electrostatic contribution to
the aromatic interactions which dominates in very nonpolar
solvents. The bipyridine rings are strongly polarized by the
metal centers in these systems, and so the electrostatic compo-
nent of this interaction is likely to be much larger than in
interactions between simple aromatic rings. However, the
results clearly demonstrate that electrostatics are an important
factor even in interactions between relatively nonpolar groups.
We believe that the biphasic behavior of this system is likely
to be representative of the solvent dependence of the majority
of molecular recognition events which involve combinations
of polar and nonpolar (or solvophobic) interactions. It is no
coincidence that the solvent which lies at the bottom of the curve
in Figure 8 is DMSO: it is one of the best general solvents
known because it competes effectively for both polar and
nonpolar interaction sites.
RuII63 (PF6)2 Relative to RuII73 (PF6)2
a
∆δ (ppm)
solvent
Z (kcal mol-1
)
H3
H4
H6
CDCl3
CD2Cl2
acetone-d6
DMF-d7
DMSO-d6
CD3CN
63.2
64.2
65.7
68.5
71.1
71.3
-0.61
-0.45
-0.37
-0.21
-0.20
-0.30
-0.23
-0.21
-0.12
-0.05
-0.04
-0.11
-0.01
-0.02
-0.02
+0.06
-0.06
+0.02
a All spectra were recorded at a concentration of 8 mM.
Figure 8. Differences in chemical shift between H3, H4, and H6 of
FeII63 (ClO4)2 and RuII63 (PF6)2 and their respective reference complexes
as a function of solvent polarity.
Experimental Section
the solvent polarity decreases, the magnitude of the shift
decreases. This is consistent with a solvophobic description of
aromatic interactions: the interactions are strongest in polar
solvents which are unable to properly solvate the nonpolar
surfaces of the aromatic rings.4 However, as the solvent polarity
decreases further, the strength of the aromatic interaction goes
through a minimum at DMSO and then starts to increase again.
By the time we reach chloroform, the magnitude of the
interaction is comparable to that in water. This suggests that
in nonpolar solvents, electrostatic interactions become dominant
and lead to large attractive interactions between the aromatic
rings.19
Further attempts to investigate the relative strengths of the
interactions were made using 2D ROESY experiments on FeII63
(ClO4)2 in CDCl3 and DMSO-d6 and FeII63 Cl2 in CD3OD. A
qualitative estimate of the average distance between the inter-
acting aromatic rings was obtained by integrating the corre-
sponding NOE cross-peaks (CP1 and CP2 in Figure 5) and
normalizing the intensities relative to the H3-H6 cross-peak
(CP3 in Figure 5): the H3-H6 distance is constrained by the
covalent and coordination bonding in the complex and so is
solvent-independent. The normalized cross-peaks labeled CP1
and CP2 in Figure 5 are significantly more intense in the
CD3OD and CDCl3 spectra compared with the DMSO-d6
spectrum. Thus the aromatic rings spend less time in close
proximity in DMSO than in the other two solvents in accord
with the chemical shift experiments.
1H and 13C NMR spectra were recorded on a Bruker AC 250 or
AM 250 MHz spectrometer operated in Fourier transform mode.
Chemical shifts were referenced to TMS as an internal standard. All
solvent-dependent measurements were made on the AM 250 instrument.
2D ROESY spectra were recorded on a Bruker AMX 400 MHz
spectrometer. FAB+ mass spectra were recorded on a Kratos MS80
using a matrix of m-nitrobenzyl alcohol (NOBA) and positive electro-
spray spectra on a Kratos MS25. UV/visible spectra were recorded
on a Phillips PU 8720 scanning spectrophotometer using quartz cuvettes
in either dry CH2Cl2 or spectroscopic grade methanol. IR spectra were
recorded on a Perkin-Elmer Paragon 1000 spectrometer, and samples
were prepared as a Nujol mull. Melting points were recorded on a
Reichter Kofler hot stage melting point apparatus and are uncorrected.
Microanalyses were performed using a Perkin-Elmer 2400 CHN
elemental analyzer working at 975 °C. Where specified, CH2Cl2 was
dried by distillation from CaH2. Dry ruthenium trichloride was prepared
by heating the trihydrate at 140 °C for 48 h and was assumed to have
composition RuCl3.
Molecular modeling was carried out using the Macromodel v4.5
implementation of the MM2 force field17 with chloroform solvation.
The core of the complex (metal and bipyridine rings) was constrained
to the X-ray crystal structure geometry found for Ru(bipy)3 in the
Cambridge Structure Database, and Monte Carlo searching was
employed to locate local energy minimum conformations for the flexible
sidearm.
5,5′-Bis(2,2-dicarbethoxyethyl)-2,2′-bipyridine (3). Lithium hy-
dride (0.159 g, 20 mmol) was added in one portion to a solution of
diethyl malonate (5.86 g, 36.6 mmol) in dimethylformamide (DMF)
(10 mL) and stirred for 1 h. Solid 2 (1.00 g, 2.92 mmol) was added
in one portion and stirred for 30 min at 20 °C followed by 1 h at 50
°C. Water (10 mL) was added to the mixture which was then extracted
with diethyl ether (3 × 50 mL). The combined extracts were washed
with water (3 × 50 mL), dried (Na2SO4), filtered, and evaporated to
give a yellow oil from which white crystals formed overnight. The
crystals were removed by filtration, washed well with petroleum ether
(bp 40-60 °C), and dried in vacuo to yield the title compound (0.948
g, 65%): Rf ) 0.4 (Et2O); mp 123-124 °C; 1H NMR (250 MHz,
CDCl3) δ 1.22 (12H, t, J ) 7 Hz, -CH3), 3.26 (4H, d, J ) 8 Hz, ArCH2-
), 3.66 (2H, t, J ) 8 Hz, ArCH2CH-), 4.16 (8H, m, -OCH2-), 7.67 (2H,
Conclusions
1
The H NMR chemical shifts of the metal tris-bipyridine
complexes reported above indicate that there are strong in-
(18) Z values were obtained from Gordon, A. J.; Ford, R. A. The
Chemist’s Companion; Wiley: New York, 1972; pp 22-23 and references
therein. Isotope effects for deuterated solvents were ignored. For a discussion
of the derivation of Z values, see: Kosower, E. M. J. Am. Chem. Soc. 1958,
80, 3253-3270.
(19) (a) Hunter, C. A.; Sanders, J. K. M. J. Am. Chem. Soc. 1990, 112,
5525-5534. (b) Hunter, C. A. Chem. Soc. ReV. 1994, 23, 101-109.

Influence of SolVent on Aromatic Interactions
J. Am. Chem. Soc., Vol. 120, No. 14, 1998 3409
dd, J ) 8, 2 Hz, 4-pyridine H), 8.27 (2H, d, J ) 8 Hz, 3-pyridine H),
8.52 (2H, d, J ) 2 Hz, 6-pyridine H); 13C NMR (62.9 MHz, CDCl3)
δ 14.0, 31.7, 53.3, 61.7, 120.6, 133.5, 137.3, 149.6, 154.6, 168.4; FAB+
MS m/z ) 501 (100, MH+), C26H32N2O8 requires 500.53; υ ) 1720
cm-1 (CdO). Calculated for C26H32N2O8: C, 62.39; H, 6.44; N, 5.60.
Found: C, 62.41; H, 6.20; N, 5.52.
138.3, 140.0, 150.4, 154.2, 156.9, 170.5; FAB+ MS m/z ) 1513 (15,
[M - ClO4]+), 1414 (70, [M - 2ClO4]+), 960 (100, [M - bipy]2+).
FeII63 Cl2: mp ca. 180 °C; ES+ MS m/z ) 706 (100, [M - 2Cl]2+);
1H NMR (250 MHz, CD3OD) δ 2.63 (24H, m, -CH2CH2-), 6.87 (12H,
d, J ) 7.5 Hz, o-phenol H), 7.24 (6H, d, J ) 1.5 Hz, 6-pyridine H),
7.31 (6H, m, p-phenol H), 7.42 (12H, m, m-phenol H), 7.81 (6H, dd,
J ) 8, 1.5 Hz, 4-pyridine H), 8.04 (6H, d, J ) 8 Hz, 3-pyridine H);
5,5′-Bis(2-carbethoxyethyl)-2,2′-bipyridine (4). 3 (7.50 g), water
(2.025 g), and sodium chloride (2.10 g) were refluxed in dimethyl
sulfoxide (DMSO) (100 mL) for 2.5 h with the exclusion of light. After
the mixture cooled, diethyl ether (400 mL) was added and the ether
layer washed with water (5 × 150 mL) before being dried (Na2SO4)
and filtered through 5 cm of silica gel. The solvent was then removed
in vacuo to yield the title compound as a white crystalline solid (4.50
g, 84%): Rf ) 0.50 (10% methanol/CH2Cl2); mp 87-89 °C; 1H NMR
(250 MHz, CDCl3) δ 1.22 (6H, t, J ) 7 Hz, -CH3), 2.55 (4H, t, J ) 8
Hz, -CH2-), 2.98 (4H, t, J ) 8 Hz, -CH2-), 4.12 (4H, q, J ) 7 Hz,
-OCH2-), 7.65 (2H, dd, J ) 8, 2 Hz, 4-pyridine H), 8.26 (2H, d, J )
8 Hz, 3-pyridine H), 8.52 (2H, d, J ) 2 Hz, 6-pyridine H); 13C NMR
(62.9 MHz, CDCl3) δ 14.2, 28.0, 35.4, 60.6, 120.6, 135.9, 136.8, 149.2,
154.4, 172.4; FAB+ MS m/z ) 357 (100, MH+), C20H24N2O4 requires
356.42; υ ) 1722 cm-1 (CdO). Calculated for C20H24N2O4: C, 67.40;
H, 6.79; N, 7.86. Found: C, 67.33; H, 6.77; N, 7.85.
λmax (ꢀ) 255.6 (39 500), 265.7 (35 100), 304.3 (76 400), 394.0 (3100),
490.0 (7000), 518.3 (8100). Calculated for C84H72N6O12FeCl2‚2H2O:
C, 66.36; H, 5.04; N, 5.53; Cl, 4.66. Found: C, 66.35; H, 5.03; N,
5.46; Cl, 4.96.
Tris(5,5′-bis(2-carbethoxyethyl)-2,2′-bipyridine)iron(II) Perchlo-
rate and Chloride Salts (FeII43 (ClO4)2 and FeII43 Cl2). FeII43
1
(ClO4)2: mp 112-115 °C; H NMR (250 MHz, MeCN-d3) δ 1.08
(18H, t, J ) 7 Hz, -CH3), 2.45 and 2.73 (24H, br, -CH2CH2-), 3.90
(12H, m, -OCH2-), 7.15 (6H, d, J ) 1.5 Hz, 6-pyridine H), 7.93 (6H,
dd, J ) 8, 1.5 Hz, 4-pyridine H), 8.37 (6H, d, J ) 8 Hz, 3-pyridine
H); FAB+ MS m/z ) 1223 (3, [M - ClO4]+), 1125 (20, [M - 2ClO4]+),
562 (100, [M - 2ClO4]2+).
Calculated for C60H72N6O20FeCl2: C, 54.43; H, 5.48; N, 6.35; Cl,
5.36. Found: C, 54.32; H, 5.52; N, 6.51; Cl, 5.65.
FeII43 Cl2: mp 105-106 °C; ES+ MS m/z ) 562 (100, [M - 2Cl]2+);
1H NMR (250 MHz, CD3OD) δ 1.11 (18H, t, J ) 7 Hz, -CH3), 2.58
and 2.78 (2 × 12H, 2t, J ) 6.5 Hz, -CH2CH2-), 3.94 (12H, 2q, J ) 7
Hz, -OCH2-), 7.30 (6H, d, J ) 1.5 Hz, 6-pyridine H), 8.05 (6H, dd, J
5,5′-Bis(2-carboxyethyl)-2,2′-bipyridine (5). Sodium hydroxide
(0.85 g, 21.25 mmol) and 4 (1.50 g, 4.208 mmol) were refluxed for 2
h in water (10 mL). After cooling and neutralization with concen-
trated HCl (36%, 2.15 g, 21.25 mmol), the product was removed by
filtration, washed well with water and diethyl ether, and then dried in
vacuo at 60 °C overnight to give the title compound as a white
solid (1.13 g, 90%): Rf ) 0.0 (15% methanol/CH2Cl2); mp 246-249
) 8, 1.5 Hz, 4-pyridine H), 8.60 (6H, d, J ) 8 Hz, 3-pyridine H); 13
C
NMR (62.9 MHz, CD3OD) δ 14.5, 28.6, 34.8, 61.6, 124.6, 139.9, 142.2,
155.2, 158.9, 173.5; λmax (ꢀ) 254.8 (30 700), 267.1 (25 700), 304.6
(62 800), 354.5 (3000), 490.0 (4200), 520.0 (5200).
1
°C; H NMR (250 MHz, DMSO-d6) δ 2.62 and 2.89 (2 × 4H, 2t,
Tris(5,5′-bis(2-carboxyethyl)-2,2′-bipyridine)ruthenium(II) Hexa-
fluorophosphate (RuII53 (PF6)2). Dried ruthenium trichloride (0.194
g, 0.935 mmol) and 4 (1.000 g, 2.806 mmol) were heated at reflux in
ethylene glycol (6 mL) for 45 min. After the addition of NH4PF6 (0.610
g, 3.74 mmol) in water (6 mL), the solution was refrigerated overnight
and the crystals that separated were removed by filtration and washed
well with water before being suspended in water (10 mL). Sodium
hydroxide (0.822 g, 20.55 mmol) was added and the mixture heated at
reflux for 45 min until the solid dissolved to give a clear red solution.
After the mixture cooled, 36% HCl (2.2 mL, 25 mmol) and NH4PF6
(1.5 g, 9.20 mmol) were added and the solution returned briefly to
reflux. The crystals that separated upon cooling were removed by
filtration, washed well with water, and dried in vacuo at 60 °C to give
the title compound as a bright-orange solid (0.931 g, 77%): mp 122-
J ) 8 Hz, -CH2CH2-), 7.79 (2H, dd, J ) 8, 2 Hz, 4-pyridine H),
8.27 (2H, d, J ) 8 Hz, 3-pyridine H), 8.54 (2H, d, J ) 2 Hz, 6-pyri-
dine H), 12.25 (2H, br, -CO2H); 13C NMR (62.9 MHz, DMSO-d6) δ
27.8, 35.1, 120.3, 137.1, 137.5, 149.6, 153.7, 174.0; FAB+ MS m/z )
301 (100, MH+), C16H16N2O4 requires 300.30. Calculated for
C16H16N2O4‚0.25H2O: C, 63.04; H, 5.46; N, 9.19. Found: C, 62.91;
H, 5.18; N, 9.17.
5,5′-Bis(2-carbophenoxyethyl)-2,2′-bipyridine (6). 5 (0.195 g,
0.649 mmol) and phenol (0.306 g, 3.246 mmol) in dry CH2Cl2 (10
mL) were cooled in ice before the addition of 1-(3-(dimethylamino)-
propyl)-3-ethylcarbodiimide hydrochloride (EDC) (0.435 g, 2.27 mmol)
and 4-(dimethylamino)pyridine (DMAP) (8 mg, 0.0649 mmol). After
the mixture stirred for 18 h at room temperature, the organic layer was
washed with 1 M HCl (5 mL), 1 M NaOH (10 mL), and water (2 × 10
mL). The CH2Cl2 solution was dried (Na2SO4), filtered, and evaporated.
The crude product was purified by flash chromatography on silica gel
(2 cm × 15 cm) eluting with 1.5% methanol in CH2Cl2 to give the
title compound as a white solid (0.215 g, 73%): Rf ) 0.3 (5% methanol
1
125 °C; H NMR (250 MHz, acetone-d6) δ 2.61 and 2.81 (2 × 12H,
2m, -CH2CH2-), 7.78 (6H, d, J ) 1.5 Hz, 6-pyridine H), 8.09 (6H, dd,
J ) 8, 1.5 Hz, 4-pyridine H), 8.63 (6H, d, J ) 8 Hz, 3-pyridine H);
13C NMR (62.9 MHz, acetone-d6) δ 28.2, 34.3, 124.6, 138.6, 142.0,
151.9, 156.2, 174.2; FAB+ MS m/z ) 1147 (15, [M - PF6]+), 1001
(100, [M - 2PF6]+); υ ) 1706 cm-1 (CdO); λmax (ꢀ) 256.8 (28 000),
294.3 (78 000), 453.2 (13 000). Calculated for C48H48N6O12RuP2F12‚
3H2O: C, 42.83; H, 4.04; N, 6.24. Found: C, 42.82; H, 3.73; N, 6.40.
1
in CH2Cl2); mp 153.5-155 °C; H NMR (250 MHz, CDCl3) δ 2.94
(4H, t, J ) 8 Hz, -CH2-), 3.14 (4H, t, J ) 8 Hz, -CH2-), 7.03 (4H, d,
J ) 8 Hz, o-phenol H), 7.22 (2H, m, p-phenol H), 7.36 (4H, m,
m-phenol H), 7.74 (2H, dd, J ) 8, 2 Hz, 4-pyridine H), 8.34 (2H, d, J
) 8 Hz, 3-pyridine H), 8.61 (2H, d, J ) 2 Hz, 6-pyridine H); 13C
NMR (62.9 MHz, CDCl3) δ 27.9, 35.5, 120.8, 121.5, 126.0, 129, 135.6,
137.0, 149.3, 150.5, 154.5, 170.9; FAB+ MS m/z ) 453, C28H24N2O4
requires 452.29. Calculated for C28H24N2O4: C, 74.32; H, 5.35; N,
6.19. Found: C, 73.7; H, 5.3; N, 6.5.
Tris(5,5′-bis(2-carbophenoxyethyl)-2,2′-bipyridine)ruthenium-
(II) Hexafluorophosphate (RuII63 (PF6)2). Phenol (87.4 mg, 0.929
mmol) and RuII53 (PF6)2 (0.100 g, 0.077 mmol) were cooled to 0 °C in
dry CH2Cl2 (15 mL) before the addition of DMAP (9.5 mg, 0.077
mmol) and EDC (89.0 mg, 0.464 mmol). After stirring for 48 h, the
CH2Cl2 solution was washed with 1 M HCl (2 × 15 mL), 1 M NaOH
(2 × 15 mL), and water (15 mL). The solution was then stirred
vigorously with a saturated aqueous solution of NH4PF6 (1 mL) for 5
h before the CH2Cl2 layer was separated, washed with water (15 mL),
dried (Na2SO4), filtered, and evaporated. Flash chromatography on
silica gel (20 cm × 1.5 cm) eluting with 2.5% methanol in CH2Cl2
gave the title compound as an orange solid (94 mg, 70%): Rf ) 0.60
(5% methanol in CH2Cl2); mp 175-176 °C; 1H NMR (250 MHz,
CDCl3) δ 2.69 (24H, br, ArCH2CH2CO2Ph), 6.88 (12H, d, J ) 8 Hz,
o-phenol H), 7.24 (6H, m, p-phenol H), 7.35 (12H, m, m-phenol H),
7.53 (6H, d, J ) 1.5 Hz, 6-pyridine H), 7.55 (6H, dd, J ) 8, 1.5 Hz,
4-pyridine H), 7.73 (6H, d, J ) 8 Hz, 3-pyridine H); 13C NMR (62.9
MHz, 5% CD3OD/CDCl3) δ 27.0, 33.6, 121.4, 123.4, 126.0, 129.4,
137.4, 140.4, 150.4, 151.2, 154.7, 170.6; FAB+ MS m/z ) 1603 (100,
General Procedure for the Preparation of Iron(II) Complexes.
To a solution of the iron(II) salt with the required counterion
(Fe(ClO4)2‚6H2O or FeCl2‚4H2O) (0.033 mmol) in the relevant solvent
(5% methanol in CH2Cl2 for ClO4 salts, methanol for Cl- salts) was
-
added ligand 4 or ligand 6 (1 mmol). After stirring for 30 min, the
solvent was removed in vacuo to give a quantitative yield of the product
as a deep-red solid.
Tris(5,5′-bis(2-carbophenoxyethyl)-2,2′-bipyridine)iron(II) Per-
chlorate and Chloride Salts (FeII63 (ClO4)2 and FeII63 Cl2). FeII63
(ClO4)2: 1H NMR (250 MHz, MeCN-d3) δ 2.67 (24H, m, -CH2CH2-),
6.88 (12H, d, J ) 8 Hz, o-phenol H), 7.16 (6H, d, J ) 1.5 Hz,
6-pyridine H), 7.25-7.43 (m, 18H, m- and p-phenol H), 7.81 (6H, dd,
J ) 8, 1.5 Hz, 4-pyridine H), 8.05 (6H, d, J ) 8 Hz, 3-pyridine H);
13C NMR (250 MHz, CDCl3) δ 27.3, 33.7, 121.5, 123.1, 125.9, 129.4,

3410 J. Am. Chem. Soc., Vol. 120, No. 14, 1998
Breault et al.
[M - PF6]+), 1459 (45, [M - 2PF6]+). Calculated for C84H72N6O12-
RuP2F12: C, 57.70; H, 4.15; N, 4.81. Found: C, 57.94; H, 4.42; N,
5.24.
Bis(5,5′-dimethyl-2,2′-bipyridine)(5,5′-bis(2-carboxyethyl)-2,2′-bi-
pyridine)ruthenium(II) Hexafluorophosphate (RuII12‚5 (PF6)2).
RuII12‚4 (PF6)2 (0.112 g, 0.100 mmol) and sodium hydroxide (0.206 g,
5.15 mmol) in water (5 mL) were heated at reflux for 2.5 h. The
solution was cooled before the addition of 36% HCl (0.608 g, 6 mmol)
and NH4PF6 (1.00 g) and then briefly returned to reflux. After cooling,
the suspension was extracted with CH2Cl2 (4 × 15 mL), and the
combined CH2Cl2 extracts were washed with water, dried (Na2SO4),
filtered, and evaporated to give the title compound as a bright-orange
solid (0.102 g, 96%): Rf ) 0.0 (10% methanol/CH2Cl2); mp 188-192
Tris(5,5′-bis(2-carbopentoxyethyl)-2,2′-bipyridine)ruthenium-
(II) Hexafluorophosphate (RuII73 (PF6)2). RuII73 (PF6)2 was prepared
using an identical procedure to that for RuII63 (PF6)2 but substituting
n-pentanol (0.450 g, 5.04 mmol) in place of phenol. The product was
purified by flash chromatography on silica gel (20 cm × 1.5 cm) eluting
with 2% methanol in CH2Cl2 to give the title compound as a
hygroscopic orange solid (62 mg, 47%): Rf ) 0.3 (5% methanol in
CH2Cl2); 1H NMR (250 MHz, CDCl3) δ 0.86 (30H, m, -CH2CH3), 1.43
(12H, m, -OCH2CH2CH2-), 1.61 (12H, m, OCH2CH2-), 2.54 and 2.83
(2 × 12H, 2m, ArCH2CH2CO2R), 3.93 (12H, m, -OCH2-), 7.54 (6H,
d, J ) 1.5 Hz, 6-pyridine H), 7.81 (6H, dd, J ) 8, 1.5 Hz, 4-pyridine
H), 8.34 (6H, d, J ) 8 Hz, 3-pyridine H); FAB+ MS m/z ) 1566 (100,
[M - PF6]+), 1423 (75, [M - 2PF6]+), [M - PF6]+ requires 1567.6711,
found m/z ) 1567.6863; λmax (ꢀ) 265.4 (35 000), 295.5 (93 000), 453.9
(16 000).
1
°C; H NMR (250 MHz, acetone-d6) δ 2.20 (2 × 6H, 2s, -CH3), 2.57
and 2.83 (2 × 4H, 2m, ArCH2CH2COOH), 7.67, 7.80, and 7.87 (3 ×
2H, 3d, J ) 1.5 Hz, 6-pyridine H), 7.98 and 8.09 (4H and 2H, 2m,
4-pyridine H), 8.62 (6H, m, 3-pyridine H); 13C NMR (62.9 MHz,
acetone-d6) δ 18.5, 28.1, 34.2, 124.3, 124.5, 138.7, 139.1, 139.2, 139.3,
142.1, 152.0, 152.1, 152.3, 155.8, 156.1, 173.3; FAB+ MS m/z ) 915
(85, [M - PF6]+), 769 (100, [M - 2PF6]+), 385 (30, [M - 2PF6]2+).
Calculated for C40H40N6O4RuP2F12: C, 45.33; H, 3.80; N, 7.93;
Found: C, 44.8; H, 4.1; N, 7.9.
cis-Dichlorobis(5,5′-dimethyl-2,2′-bipyridine)ruthenium(II) (RuII12‚
Cl2). 1 (1.417 g, 7.69 mmol), RuCl3‚3H2O (1.000 g, 3.82 mmol), and
lithium chloride (1.100 g, 25.9 mmol) were heated at reflux in HPLC
grade DMF (8 mL) for 7 h. The resulting solution was poured into
acetone (25 mL), and after refrigeration overnight the precipitate was
removed by filtration and washed with water (3 × 25 mL) and diethyl
ether (3 × 25 mL) before being dried in vacuo to yield the title
compound as a black solid (1.72 g, 83%) which was used directly in
the next step.
Bis(5,5′-dimethyl-2,2′-bipyridine)(5,5′-bis(2-carbophenoxyethyl)-
2,2′-bipyridine)ruthenium(II) hexafluorophosphate (RuII12‚6 (PF6)2).
To a solution of RuII12‚5 (PF6)2 (50.9 mg, 0.048 mmol), phenol (45.0
mg, 0.478 mmol), and DMAP (0.5 mg) in CH2Cl2 (2 mL) at 0 °C was
added EDC (36.8 mg, 0.192 mmol) in one portion. The mixture was
stirred a room temperature for 18 h before dilution with CH2Cl2 (10
mL) and washing with 1 M HCl (10 mL), 1 M NaOH (10 mL), and
water (10 mL). After drying (Na2SO4), the solvent was removed in
vacuo and the crude product purified by flash chromatography on silica
gel (1 cm × 15 cm) eluting with 2% methanol in CH2Cl2 to give the
title compound as an orange solid (44.5 mg, 75%): mp ca. 150 °C;
1H NMR (250 MHz, acetone-d6) δ 2.09 and 2.13 (2 × 6H, 2s, -CH3),
2.92 (8H, br, -CH2CH2-), 6.95 (4H, d, J ) 8 Hz, o-phenol H), 7.27
(2H, t, J ) 8 Hz, p-phenol H), 7.41 (4H, t, J ) 8 Hz, m-phenol H),
7.70, 7.75, and 7.93 (3 × 2H, 3s, J ) 1.5 Hz, 6-pyridine H), 7.88 and
7.95 (2 × 2H, 2dd, J ) 8, 1.5 Hz, 4- and 4′-dimethylbipy H), 8.15
(2H, dd, J ) 8, 1.5 Hz, 4-pyridine H), 8.43 and 8.51 (2 × 2H, 2d, J )
8 Hz, 3- and 3′-dimethylbipy H), 8.69 (2H, d, J ) 8 Hz, 3-bipyridine
H); 13C NMR (62.9 MHz, acetone-d6) δ 18.4, 28.0, 34.5, 122.4, 124.2,
124.3, 124.6, 126.7, 130.3, 138.7, 139.2, 139.3, 139.0, 139.1, 151.7,
151.9, 152.6, 155.8, 156.3, 171.3; FAB+ MS m/z ) 1068 (98, [M -
PF6]+), 922 (100, M - 2PF6]+); λmax (ꢀ) 265.4 (38 000), 295.5 (99 000),
447.2 (16 000). Calculated for C54H48N6O4RuP2F12‚3H2O: C, 50.27;
H, 4.22; N, 6.51; Found: C, 49.94; H, 4.20; N, 6.56.
Bis(5,5′-dimethyl-2,2′-bipyridine)(5,5′-bis(2-carbethoxyethyl)-2,2′-
bipyridine)ruthenium(II) Hexafluorophosphate (RuII12‚4 (PF6)2). A
solution of crude RuII12‚Cl2 (0.500 g, 0.868 mmol) and 4 (0.340 g,
0.955 mmol) in 50% ethanol/water (20 mL) was heated at reflux for 3
h. After the mixture cooled, a solution of NH4PF6 (1.4 g) in water (20
mL) was added with the formation of an orange precipitate which was
extracted into CH2Cl2 (50 mL). The CH2Cl2 layer was separated and
washed with 1 M HCl (2 × 25 mL), 1 M NaOH (2 × 25 mL), and
water (25 mL) before being dried (Na2SO4), filtered, and evaporated.
The product was purified by flash chromatography on silica gel (2 cm
× 35 cm) eluting with 2% methanol/CH2Cl2, collecting the fast-running
band to give the title compound as an orange solid (0.702 g, 72%): Rf
1
) 0.70 (10% methanol in CH2Cl2); mp ca. 120 °C; H NMR (250
MHz, acetone-d6) δ 1.10 (6H, t, J ) 8 Hz, -OCH2CH3), 2.21 (2 × 6H,
2s, ArCH3), 2.57 and 2.82 (2 × 4H, 2m, ArCH2CH2CO2Et), 3.95 (4H,
m, OCH2CH3), 7.61, 7.80, and 7.85 (3 × 2H, 3d, J ) 1.5 Hz, 6-pyridine
H), 7.95-8.09 (6H, m, 4-pyridine H), 8.62 (6H, m, 3-pyridine H); 13
C
NMR (62.9 MHz, acetone-d6) δ 14.4, 18.4, 28.1, 34.5, 60.9, 124.2,
124.3, 124.4, 138.6, 139.3, 139.0, 152.0, 152.1, 152. 5, 155.7, 155.8,
156.1, 172.4; FAB+ MS m/z ) 971 (100, [M - PF6]+), 826 (75, [M -
2PF6]+), 413 (86, [M - 2PF6]2+); λmax (ꢀ) 265.0 (63 000), 294.1
(140 000), 446.4 (25 000). Calculated for C44H48N6O4RuP2F12: C,
47.34; H, 4.34; N, 7.53; Found: C, 47.6; H, 4.3; N, 7.8.
Acknowledgment. We thank the Lister Institute (C.A.H.),
the EPSRC, and Zeneca Pharmaceuticals (P.C.M.) for financial
support and Dr. J. A. Thomas for useful discussions.
JA973938+