[Ru(II)(hedta)]- complexes of 2,2'-dipyridylamine (dpaH) and a bifunctional tethered analog, N,N,N',N'-tetrakis(2-pyridyl)adipamide (tpada)

Article abstract of DOI:10.1016/S0020-1693(99)00498-3

[Ru(II)(hedta)L]^- complexes (hedta^3- = N-hydroxyethylethylenediamine-N,N,N'-triacetate); L = dpaH (2,2'-dipyridylamine) and tpada (N,N,N',N'-tetrakis(2-pyridyl)adipamide)) have been studied by ¹H NMR and electrochemical methods in aqueous solution. The bidentate rings of dpaH and tpada are differentiated as shown by NMR upon coordination to Ru(II) due to differences in the local environment. The dpa-R headgroup of each ligand binds 'in-plane' with the en backbone of hedta^3- and with one pyridyl ring being nearer the amine of hedta^3- having the pendant glycinato group (matching the known arrangement with bpy (2,2'-bipyridine)). Ru(II/III) E(1/2) values follow the order dpaH (0.32 V) < tpada (0.47 V) < bpy (0.54 V), showing that dpaH is a weaker π-acceptor ligand than bpy, and that the withdrawing carbonyl functionality enhances the π-acceptor capacity for the tpada ligand, approaching the stability imparted by bpy. Only the 1:1 [Ru(II)(hedta)(dpaH)]^- complex forms even in the presence of excess dpaH. [Ru(II)(hedta)(dpaH)] has a pK(a) of the dipyridylamine proton of approximately 5.0 with [Ru(II)(hedta)(dpa^-)] undergoing aquation (k(H₂O)= 1.4 10^-2 s^-1) and OH^--assisted dissociation (k(OH) = 1.33 10⁴ M^-1 s^-1). The {[Ru(II)-(hedta)]₂(tpada)}^2- complex serves as a water-soluble model as to how {[ML']₂(tpada)} complexes might act as an extended bridge between two metal binding sites, potentially those of metallo-derivatized DNA strands, or between one DNA strand and a protein crosslink. In this model M represents an appropriate metal for DNA derivatization such as Ru(II), Pt(II) or Pd(II) and L' represents the attachments to DNA nucleobase sites, aminocarboxylates/peptide coordination for antitumor purposes. (C) 2000 Elsevier Science S.A.

Full text of DOI:10.1016/S0020-1693(99)00498-3

www.elsevier.nl/locate/ica
Inorganica Chimica Acta 303 (2000) 30–39
[RuII(hedta)]− complexes of 2,2%-dipyridylamine (dpaH) and a
bifunctional tethered analog,
N,N,N%,N%-tetrakis(2-pyridyl)adipamide (tpada)
Rex E. Shepherd *, Ya Chen, Richard A. Kortes, Matthew S. Ward
Department of Chemistry, Uni6ersity of Pittsburgh, Pittsburgh, PA 15260, USA
Received 24 June 1999; accepted 10 September 1999
Abstract
[RuII(hedta)L]− complexes (hedta3− =N-hydroxyethylethylenediamine-N,N,N%-triacetate); L=dpaH (2,2%-dipyridylamine) and
tpada (N,N,N%,N%-tetrakis(2-pyridyl)adipamide)) have been studied by 1H NMR and electrochemical methods in aqueous solution.
The bidentate rings of dpaH and tpada are differentiated as shown by NMR upon coordination to RuII due to differences in the
local environment. The dpa–R headgroup of each ligand binds ‘in-plane’ with the en backbone of hedta3− and with one pyridyl
ring being nearer the amine of hedta3− having the pendant glycinato group (matching the known arrangement with bpy
(2,2%-bipyridine)). RuII/III E1/2 values follow the order dpaH (0.32 V)Btpada (0.47 V)Bbpy (0.54 V), showing that dpaH is a
weaker p-acceptor ligand than bpy, and that the withdrawing carbonyl functionality enhances the p-acceptor capacity for the
tpada ligand, approaching the stability imparted by bpy. Only the 1:1 [RuII(hedta)(dpaH)]− complex forms even in the presence
of excess dpaH. [RuII(hedta)(dpaH)] has a pKa of the dipyridylamine proton of approximately 5.0 with [RuIII(hedta)(dpa−)]
undergoing aquation (kH O=1.4×10−2
s s
−1) and OH−-assisted dissociation (kOH=1.33×104 M−1 −1). The {[RuII-
2
(hedta)]2(tpada)}2− complex serves as a water-soluble model as to how {[ML%]2(tpada)} complexes might act as an extended
bridge between two metal binding sites, potentially those of metallo-derivatized DNA strands, or between one DNA strand and
a protein crosslink. In this model M represents an appropriate metal for DNA derivatization such as RuII, PtII or PdII and L%
represents the attachments to DNA nucleobase sites, aminocarboxylates/peptide coordination for antitumor purposes. © 2000
Elsevier Science S.A. All rights reserved.
Keywords: Ruthenium complexes; Dipyridylamine complexes; Polydentate ligand complexes
of dpaH are flexible in their coordination with metal
centers, adopting either nearly coplanar or tilted pyridyl
1. Introduction
ring planes that vary from 2° to 42° of tilt. This can be
2,2%-Dipyridylamine (dpaH) is an aromatic amine in
some ways similar to 2,2%-bipyridine (bpy), but the
induced either electronically or stereochemically. For
example, CuII complexes span the 2–42° range based
central amine unit introduces several differences: (1)
on the various other ligands; structures of [Cu(dpa-
dpaH coordinates with six-membered chelate rings in-
stead of five-membered rings for bpy; (2) the coordi-
H)2]2+, and [Cu(dpaH)2X]+ and [Cu(dpaH)L%] (L%=
acac, etc.) have been widely studied in this regard [2–5].
nated pyridine rings are not necessarily forced to near
[RuII(dpaH)3]2+ and the series [Ru(bpy)3−n(dpaH)n]2+
coplanarity as in bpy; (3) the ligand field strength of
dpaH is greater than ethylenediamine, but closer to that
of a single aromatic amine, e.g. that dpaH is less of a
p-acceptor ligand than bpy [1]. The two pyridine rings
(n=0–2) have been shown by 1H and 13C NMR
methods to have rapidly flapping dpa pyridyl donors
that undergo an inversion of the relative tilt of the two
pyridyl planes of dpa. These motions average the chem-
ical shifts of the ring H-6 and H-6% protons of [Ru-
(dpaH)3]2+ [6a]. The dihedral angle of 35° was found
for the only reported structure of a RuII complex,
[Ru(Me2bpy)2(dpaH)](PF6)2 [6d].
* Corresponding author. Tel.: +1-412-624 8510; fax: +1-412-624
8611.
E-mail address: shepherd@pitt.edu (R.E. Shepherd)
0020-1693/00/$ - see front matter © 2000 Elsevier Science S.A. All rights reserved.
PII: S0020-1693(99)00498-3

R.E. Shepherd et al. / Inorganica Chimica Acta 303 (2000) 30–39
31
tory. Complexes such as 3 with L=NO may have
several possible medical uses in the antisepsis or antitu-
mor arena.
There are fewer reports of dpaH complexes of
Ru(II), Pt(II) or Pd(II), metals of current biomedical
interest. Pt(II) and Pd(II) complexes have been pre-
pared having formulas of [MII(dpaH)(AA)] and
[MII(dpaH)(pyc)]+ (AA=amino acid anion, pyc=
pyridine-2-carboxylate; MII=PdII, PtII). Some of these
have shown antitumor activity toward P388 leukemia
cells that equal or exceed the activity of cisplatin [7,8].
The binding of these PtII and PdII–dpaH complexes to
calf thymus DNA is reported to be non-covalent via the
minor groove [7,8].
Derivatized ligands, with an R group attached to the
central amine of dpaH in place of H, have found uses
as electrochemical mediators in the detection of glucose
using Os(dpa–R) funtionalities as linkages to electrode
surfaces [9a], and as models of hydrolytic enzymes for
ATP and UTP with CuII(dpa–R) tethered to an
adenosine recognition site [9b]. The concept of two dpa
chelating sites linked by a suitable tether has not been
exploited previously in the preparation of bimetallic
complexes.
2. Experimental
2.1. Material
K[Ru(hedta)Cl] was prepared as described previously
[21–23]. 2,2%-Dipyridylamine was obtained from
Aldrich. Ar gas was used as supplied by Valley Welding
Supply. Ar was passed through Cr(II) scrubbing towers
followed by a deionized water rinse tower to remove
traces of O2 from the inert gas stream and to saturate
the inert gas with H2O vapor in order to avoid dehy-
dration of aqueous samples. When samples were pre-
pared in D2O for NMR studies, the tank Ar gas was
used as received to avoid enrichment of the D2O with
unwanted amounts of HOD. All other reagents were
reagent grade.
The previous reports of bpy complexes of Ru(II)-poly-
aminopolycarboxylates, including [Ru(hedta)(bpy)]−,
[10–13] serve as a useful starting point for interpreting
aspects of the coordination of dpaH and dpa-like-lig-
ands toward Ru(II) and heavier metals such as Pt(II) and
II/III
Pd(II). The Ru
E
value and the 1H NMR chemical
1/2
shifts for ligand coordinated to [RuII(hedta)]− can be
used to estimate the p-acceptor power of the ligand. We
have synthesized a ligand relative of dpaH, N,N,N%,N%-
tetrakis(2-pyridyl)adipamide (tpada):
2.2. Sample preparation
The starting K[Ru(hedta)Cl] was weighed out to-
gether with the ligand of interest to achieve 0.80:1.00,
1:00:1:00, 1:00:2.00 or 2:00:1.00 [RuII(hedta)]−: ligand
stoichiometries upon reduction to Ru(II) as specified in
the individual experiments. The samples were prepared
in Ar-purged 10 ml glass round-bottom flasks, sealed
with rubber septa. A total of 3.0–5.0 ml of D2O
together with a few small chips of Zn/Hg and a small
rice-sized stirring bar were included with the
[Ru(hedta)]/ligand samples. Stirring was maintained
magnetically while a stream of Ar purged the 10 ml
flask and an NMR tube receiver, connected in series by
The tpada ligand may be suited for the delivery of
Ru(II), Pt(II) or Pd(II) metal centers as interstrand
cross-linking reagents toward DNA, much in the manner
reported for Farrell’s polyamine groove-spanning
chelates [14–18] and related binuclear Pt(II) complexes
designed by Beck [19] and by Taylor [20].
In this paper we present data concerning the [RuII-
(hedta)(dpaH)]− and {[Ru(hedta)]2(tpada)}2− com-
plexes which allow the p-acceptor power of the ligands
to be established as bpy\1/2(tpada)\dpaH\dpa−.
Further studies with tpada in forming binuclear metal
chelates in the extended arrangement as for
{[Ru(hedta)]2(tpada)}2−, or in the multidentate arrange-
ment toward one metal center as in species 3, are being
explored for a variety of transition metals in our labora-
1
syringe needles and syringe tubing. For H NMR ex-
periments the solvent was adjusted initially with DCl
solution to give pD ꢀ3 in order to assure reduction of
Ru(III) by the Zn/Hg surface. Transfer of the reduced
[RuII(hedta)L]− samples was made under Ar pressure
through the tubing connections. After filling an NMR
tube with the desired solution via the septum top/needle
assembly, the tube was removed and the septum area

32
R.E. Shepherd et al. / Inorganica Chimica Acta 303 (2000) 30–39
was wrapped with parafilm to provide an additional
barrier toward O2.
g of K (0.0546 mol) metal in small pieces was added via
one side arm, keeping exposure to the atmosphere
minimal. The necks of the flask contained a center
reflux condenser, an inlet valve for a flowing N2 stream
and the glass-stopper port for adding reagents. The exit
tube for gas passed out through a mineral oil trap to
provide pressure release. After purging for sufficient
time to remove O2 from the apparatus, the gas flow was
stopped and the exit was connected to a drying tube
filled with dryerite to prevent atmospheric moisture
from reentry. The sample was stirred for a day to allow
time for the K metal to react with the dpa to form the
K+(dpa−) salt. It was observed that the K flakes were
coated with yellow K+(dpa−) at this time. To complete
the formation of K+(dpa), the solution was heated to
40°C while maintaining the stirring. This induced more
rapid conversion to the K+(dpa) product as the K
surface was continuously exposed to dpaH in solution.
The solution appeared yellow and a small amount of
solid was present. A total of 4.0 ml (0.0273 mol).
Adipoyl chloride (Aldrich) was transferred to a glass
Hamilton syringe inside a glove bag to prevent H2O
vapor exposure. The syringe was connected to the
reaction flask via the addition portal which was sealed
with a septum. With stirring, the adipoyl chloride was
slowly added to the sample over a period of 2 h. The
syringe was finally rinsed with a small volume of DME
to assure the transfer of all of the adipoyl chloride to
the reaction flask. Stirring was maintained at room
temperature. A yellow–white solid appeared as the
additions continued. The solids were filtered off as
product although significant amount of solid appeared
from the filtrate with evaporation. The solid product
was washed with cold water, ethanol and ether. The
product was not appreciably soluble in diethyl ether.
This is a fortunate result since the unreacted dpaH is
soluble in diethyl ether. Any unreacted dpaH can be
conveniently removed. The product was virtually insol-
uble in H2O unless treated with HCl; the product was
soluble in CHCl3. Small amounts of the product were
dissolved in CDCl3 and in D2O/DCl in order to obtain
1H and 13C NMR data. The dried solid (tpada) was
used for preparation of the [RuII(hedta)]-derivatized
sample assuming the molecular weight of 452.53 im-
2.3. Instrumentation
1H and 13C NMR spectra were obtained on Bruker
AC300 NMR spectrometers following standard proce-
dures reported previously [24–27]. 1H NMR spectra
obtained in D2O were referenced against DSS (0.00
ppm). Those obtained in CHCl3 for ligands were refer-
enced against added TMS (0.00 ppm). 13C NMR spec-
tra were referenced against p-dioxane (69.1 ppm) or
TMS (0.00 ppm).
Electrochemical studies were performed on approxi-
mately 2.61×10−3 M solutions in Ar or N2-purged
0.100 M NaCl plus the desired complex at 22°C. These
samples were prepared from reacting the proper mass
amounts of K[Ru(hedta)Cl] and ligand dissolved in the
electrolyte solvent in contact with Zn/Hg chips in a
glass bubbler apparatus having septum-sealed 14/20
joints. The glass bubbler was connected to a slowly
flowing Ar stream which provided the constant mixing
during reaction of dpaH or tpada with [RuII(hedta)-
(H2O)]−. Samples were transferred by syringe tech-
niques into an N2-purged cell compartment of an IBM
225 electrochemical analyzer for cyclic voltammetry
and differential-pulse polarography detection. The stan-
dard three-electrode arrangement was used having a
glassy carbon working electrode, a platinum wire auxil-
iary electrode, and a saturated sodium chloride calomel
electrode reference. Calibration methods have been re-
ported previously [28,29]. A flowing stream of Ar or N2
was maintained above the electrochemical samples to
prevent air oxidation. Prior work has shown sample
stability up to 24 h by these methods, although the
samples in the present study were analyzed within 1 h
of contact time within the sample cell. pH adjustment
was made by the addition of 1 M HCl or 1 M NaOH
via a syringe into the cell contents. Mixing was pro-
vided by the rice-sized stirring bar rotation and by the
mixing provided by Ar or N2 bubbling the sample via a
small tube that could be raised above the sample sur-
face or lowered to provide purging and mixing.
1
2.4. Ligand synthesis
plied by the formula. The H NMR data showed that
the isolated product was free of unreacted dpaH since
the integration of pyridyl ring protons to aliphatic
protons is correctly 2:1 (16:8). The isolated product
exhibits traces of the DME solvent coprecipitated in the
lattice of the solid; but DME does not interfere in the
coordination of [RuII(hedta)]−.
2.4.1. tpada
1,2-Dimethoxyethane (J.T. Baker) was dried for 6 h
with stirring over pieces of Na metal. The DME solvent
was then vacuum distilled, collecting the vapors in a
dry ice chilled receiver tube. A total of 200 ml of the
dried solvent was placed in a three-necked 500 ml
round-bottom flask with a stirring bar. A total of 9.34
g (0.0546 mol) of 2,2%-dipyridylamine recrystallized
from hot ethanol/acetone and vacuum dried, was added
to the solvent. Magnetic stirring was maintained as 2.13
2.4.2. Computer-assisted structural analysis
Molecular mechanics energy minimization calcula-
tions were performed for {[Ru(hedta)]2(tpada)}2− using
CHEM PRO-3D software on a 7100 MacIntosh Power PC

R.E. Shepherd et al. / Inorganica Chimica Acta 303 (2000) 30–39
33
Table 1
1H NMR shifts for dpaH and tpada a
Ligand/solvent
H-6
H-5
H-4
H-3
–CH2–CH2–
dpaH/D2O
8.17(d)
8.27(d)
8.65(d)
8.44(d)
7.02(t)
7.23(t)
7.77(t)
7.18(t)
7.76(t)
8.03(t)
8.32(t)
7.73(t)
7.28(d)
7.23(d)
7.41(d)
7.48(d)
dpaH2 +/D2O
tpadaH22+/D2O
tpada/CDCl3
1.61(s)
1.67(m)
2.35(s)
2.27(t)
a dpaH22+, tpadaH22+, refer to the dipyridyl protonation of the neutral ligand.
computer. The software allows for print out of a spacial
perspective of all atom coordinates with size of atoms
as perceived from the viewer’s perspective.
[Ru(hedta)L]− +L X [Ru(hedta)(L)2]−
(2)
and cis-polar, the main species being cis-equatorial
[32–34]. The bis complexes have both L ligands in the
equatorial plane; the same stereochemistry is adopted
for bidentate (LL) donors such as bpy and o-phen. The
[RuII(hedta)]− complexes behave identically with
that of mono and bis L complexes of [Ru(edta)]2−
3. Results and discussion
3.1. Ligands
1
[33,34]. [Ru(hedta)(bpy)]− shows no H NMR differen-
tiation of the H-6 and H-6% protons [11], but
[Ru(hedta)(dpaH)]− exhibits such a differentiation.
However, the dpaH ligand also exhibits differentiated
H-6 and H-6% resonances in [Ru(bpy)2(dpaH)]2+ [6] and
trans-[Ru(dpaH)2(NO)(OH)]2+ [35]. The influence is
attributed to geometric differences in the former [6] and
has been attributed to hindrance of the dpaH flapping
motion in the latter [35].
The tpada ligand was synthesized as described in
Section 2. H NMR data are given in Table 1 for these
1
ligands and for dpa for comparison. Solubility prob-
lems for the tpada ligand required acidification with
CF3COOD to afford solubility as tpadaH22+
.
For this reason, data for the related dpaH2+ [38]
have been given. Protonation of the pyridyl rings of
dpaH produce downfield shifts of −0.10 ppm (H-6),
−0.21 ppm (H-5) and −0.27 (H-6). Comparable ef-
fects can be assumed for tpadaH22+. The shifts of the
tether protons of tpadaH22+ are in similar chemical
environments as for the a and b methylene protons of
n-butanoic acid. The aCH and bCH shifts are 2.31 and
1H NMR samples of [RuII(hedta)]−:dpaH were pre-
pared at 1:1, 1:2 and 2:1 ratios with [RuII(hedta)]−=
2.61×10−2 M. The test of the molar ratios is necessary
in order to show whether two dpaH ligands can dis-
place all three carboxylates of the original
[Ru(hedta)(H2O)]− complex, whether a bridged-binu-
clear species forms in which each pyridyl donor of
dpaH might donate to one [Ru(hedta)]− center, or if
the coordination behavior of dpaH is simply as a
bidentate chelate in the manner of bpy in
[Ru(hedta)(bpy)]−. Substitution reactions forming bis-
substituted [Ru(hedta)L2]− complexes are complete
within 1 h for L=py, pym or pz complexes. However,
both steps of the reaction appear to be nearly as rapid
as the initial substitution process for bpy since only the
bidentate species is detectable within the time to mix a
2
2
1.68 ppm, respectively, for n-butanoic acid which confi-
rms the assignments in Table 1 for tpadaH22+ 13C
.
NMR data for the free tpada ligand was obtained in
CHCl3 referenced against p-dioxane. The 13C shift data
for tpada is given in Table 2, fully confirming the
presence of all chemical functionalities of the tpada
structure. The assignments of C-3 and C-5 carbons are
interchangeable, being virtually identical shifts and
both carbons having one proton which eliminates their
resolution from a proton-coupled spectrum.
3.2. [Ru(hedta)(dpaH)]− complex
Table 2
13C NMR shifts for the tpada ligand in ppm
[Ru(hedta)(H2O)]− reacts with monodentate N-hete-
rocyclic bases in two sequential additions for L=
pyridine, pyrimidine, pyrazine [30,31] (Eqs. (1) and (2)).
There are three possible stereochemical isomers for the
1:1 complexes, [Ru(hedta)L]− that have been identified:
cis-equatorial, trans-equatorial
Ring C
C-2
C-3
C-4
C-5
C-6
154.14
121.61
127.75
121.83
148.59
Tether C
CꢀO
CH2CO CH2CH2
35.66 24.27
172.94
[Ru(hedta)(H2O)]− +L X Ru(hedta)L− +H2O
(1)

34
R.E. Shepherd et al. / Inorganica Chimica Acta 303 (2000) 30–39
ligand must be bidentate toward one Ru(II) center
rather than bridging between two separate Ru(II) sites.
Therefore, the nature of the [Ru(hedta)(dpaH)]− com-
plex appears to be analogous to [Ru(hedta)(bpy)]− in
adopting a bidentate coordination. This is presumably
in the equatorial plane which produces trans-O car-
boxylate coordination, well-known for the [Ru(hedta)-
(L)2]− series. However, the hedta3− chelate rings lead
to differentiation and broadening of the H-6, H-6%
resonances of dpaH.
1
The conclusions drawn on the basis of the dpaH H
NMR splitting pattern are further supported by obser-
vations of the RuII/III wave by differential-pulse polar-
ography (DPP) on 2.61×10−3 M solutions. At 1:1
ratio [Ru(hedta)(dpaH)]− exhibits a single DPP wave at
0.24 V versus NHE for a neutral solution (see Fig. 2).
However, if the sample is made acidic (pH$5.50) the
wave shifts slightly to 0.29 V. Under basic solution
conditions (pH$9.50) the wave shifts markedly to 0.07
V. When the sample is acidic, scanning beyond the
original waves to positive potentials shows no evidence
of ligand loss (which promotes RuIII/IV waves at high
potentials (0.8–1.2 V)). However, the neutral sample
shows a small additional wave at 0.91 V versus NHE of
12.5% amplitude of the RuII/III 0.24 V wave. Under
basic solution, the 0.91 V wave, indicative of a 2e−
oxidation is markedly enhanced, being nearly twice the
amplitude of 0.07 V wave. These observations are
consistent with a pH-dependent redox couple (Eq. (3))
for the wave in the 0.29–0.07 V region:
Fig. 1. 300 MHz 1H NMR spectrum of [RuII(hedta)(dpaH)]− in
D2O. [complex]=2.61×10−2 M, T=25°C, pD=6.0.
sample and to tune the NMR for recording data. That
is, ring closure to form the bidentate complex with bpy
is more rapid than addition of a second monodentate
donor. For the [Ru(hedta)]−:dpaH sample at 1:2 ratio a
reaction time of 5 h, much longer than deemed neces-
sary to reach the equilibrium state in which both dpaH
ligands have added to the Ru(II) center, showed a set of
resonances at exactly the same shift as the 1:1 sample
after 7 h, as well as an equivalent of free ligand.
Electrochemical data show there is no free
[Ru(hedta)(H2O)]− remaining after 7 h for a 1:1 ratio.
[RuII(hedta)(dpa)]− +H3O++e−
X [RuII(hedta(dpaH)− +H2O]
(3)
1
In Fig. 1 of the H NMR spectrum there are broad
singlets for the 1:1 complex at 8.81 and 8.08 ppm for
H-6 and H-6% protons of a coordinated dpaH, 7.68 ppm
for coordinated dpaH’s H-4(t) set, 7.13 ppm for H-3(d)
set and 6.98 ppm for an H-5(t) set. The H-4, H-5 and
H-3 sets integrate to show that these are pairwise
matching but differentiated H-4, H-4%, H-5, H-5%, and
H-3, H-3% sets of resonances. Additionally in the 1:2
sample under acidic conditions which assure solubility
of dpaH, there are equal amplitude resonances to that
of the coordinated dpaH for the free protonated ligand
at 8.24, 7.97 and 7.19 ppm for the H-6, H-6%, H-4, H-4%,
and the overlapped H-3, H-3% and H-5, H-5% protons,
respectively. These results show that dpaH displaces
two coordination sites of [Ru(hedta)(H2O)]−, but will
not displace the remaining two carboxylate donors even
in the presence of excess ligand.
Comparison of the [Ru(hedta)]−:dpaH samples at 1:1
and 2:1 ratios showed identical coordinated dpaH spec-
tra, allowing 8 h to achieve equilibria. Since the H-6
and H-6% resonances remain differentiated, the dpaH
Fig. 2. Differential pulse voltammogram of [RuII(hedta)(dpaH)]−
.
[complex]0=2.61×10−3 M, v=0.10 (NaCl), T=22°C, under N2
purge; dashed line, pH 9.50; dash-dot-dash line, pH 7.00; solid line,
pH 5.50.

R.E. Shepherd et al. / Inorganica Chimica Acta 303 (2000) 30–39
35
near 0.07 (pH 9.5) or 0.22 V (pH 7.0), H2O or
OH− displaces the dpa− ligand within the scanning
time from 0.07 up to 1.00 V. This produces
[RuIII(hedta)(OH)2]2− and dpaH at the electrode sur-
face. The [RuIII(hedta)(OH)2]2− undergoes further oxi-
dation by a 2e− process at 0.91 V [21,19]. The
[RuVO(hedta)]2− species generated at the electrode sur-
face at 0.91 V reacts with another [RuIII(hedta)-
(OH)2]2− species to form the well-known bridged
Ru(IV) species [21,29]. Since the amplitude of the 0.91
V wave increases from pH$7 to 9.5, OH− must be
involved in the displacement of dpa− as summarized in
Scheme 1.
If we assume parallel solvent H2O and OH− path-
ways for the displacement of dpa−, the kinetic split
illustrated from Scheme 1 places limits on kH O50.014
2
s
−1 and kOH− $1.33×104 M−1 s−1. The solvent path-
way carries less than 3% of the events at pH 9.5. Under
acidic conditions the H2O-dependent pathway would
survive if the dpa− ligand remained deprotonated.
However, the scans at pH55.50 showed almost no
wave at 0.91 V. Nor was there a large wave at approx-
imately 1.15 V versus NHE indicative of [RuIII(hedta)-
(H2O)] undergoing oxidation. Hence, the RuIII complex
is stable to aquation as the [RuIII(hedta)(dpaH)] com-
plex, and it exhibits a pH-independent redox couple
with E1/2$0.32 V for solutions of pH54.0 (Eq. (5)):
Fig. 3. pH-dependence of the RuIII/II reduction potentials for
[RuII(hedta)(dpaH)− and [RuII(hedta)(dpa−)]2− species. [com-
plex]0=2.61×10−3 M, T=22°C, under an N2 purge; E1/2 was
independent of pH below 5.0.
A plot of the measured E1/2 versus pH is given in Fig.
3. The theoretical half-cell value under 1.00 M H3O+
standard state conditions would be +0.63 V. However,
dissociation of carboxylates and/or a switch to the
half-reaction involving a neutral dpaH ligand coordi-
nated to both RuIII and the RuII form preclude opera-
tion of this couple in the range of 1.00 M H3O+. The
dependence of the E1/2 to a 59.5 mV/pH change (Fig. 3)
implies that the RuIII complex is deprotonated and that
the equilibrium 4 has a pKa of approximately 5.0.
E $0.32 V
1/2
RuIII(hedta)(dapH)]+e− _` [RuII(hedta)(dpaH)]−
(5)
3.3. {[Ru(hedta)]2(tpada)}2− complex
The binuclear tpada complex was prepared using a
2:1 ratio of [Ru(hedta)(H2O)]− to the ligand tpada as in
Eq. (6). The purpose of these experiments is to show
that the tpada ligand can be a useful carrier of two
metal sites at the termini of the central tether of tpada.
[RuIII(hedta)(dpaH)] X [RuIII(hedta(dpa)]− +H3O+
(4)
Data from the [Ru(dpaH)3]2+ and [Co(dpaH)3]3+
complexes establish pKa values of approximately 11.0
and 5.0 for dpaH ligands coordinated to Ru(II) and
Co(III), respectively [6b,c]. Since the influence of
Ru(III) is usually about 1 pKa unit more acidic than the
analogous Co(III) or Rh(III) metal clusters on the pKa
of a coordinated ligand such a H2O or imidazole when
the remaining ligand are neutral amines [36], and since
an edta4− anionic ligand environment raises the pKa of
a second ligand by about 1.5 log units [34b], one
anticipates the pKa of [RuIII(hedta)(dpaH)] to be ap-
proximately 5.5 in good agreement with the electro-
chemically-based estimate.
The pKa of the Ru(III) complexes cannot be obtained
directly because of the aquation of the [RuIII(hedta)-
(dpa)]− complex in basic solution. This is observed in
the DPP scans for solutions of pH 7 and above. When
[RuIII(hedta)(dpa)]− is present as shown by the wave
(6)

36
R.E. Shepherd et al. / Inorganica Chimica Acta 303 (2000) 30–39
the H-3 population is near the amide carbonyl. Further
away the H-6 and H-6% protons are strongly differenti-
ated by the placement of the dpa–R headgroup in the
equatorial plane. Since 50% of each of these rings is
aligned with the carbonyl, each H-6 or H-6% set is
further electronically perturbed, such that two types of
H-6 and two types of H-6% are created upon coordina-
tion. We assign the two rotamers A and B as in Table
3 for the coordinated tpada complex. The tether chain
protons exhibit small downfield shifts due to inductive
effects caused by metallation of the dpa–R headgroups
of approximately −0.05 ppm for the central methyl-
enes, and −0.35 ppm for the methylenes nearer the
carbonyl and metal center.
The electrochemical data for {[Ru(hedta)]2-
(tpada)}2− presented in Fig. 5 are consistent with
aspects already discussed for the [Ru(hedta)(dpaH)]−
complex. The binuclear tethered complex exhibits only
one DPP wave for the RuII/III couple in either acidic or
neutral pH media at 0.47 V versus NHE. In basic
solution (pH$10.0) the wave appears at 0.43 V, and
again tpada dissociation from the RuIII(hedta) unit is
observed by a new wave at 0.91 V versus NHE that is
absent in neutral or acidic solution. These results
are anticipated based on the outcome of the
[Ru(hedta)(dpaH)]− study. There should be no influ-
ence due to a titratable N–H proton with the binuclear
tpada complex. The small change to 0.43 V (basic
solution) versus NHE compared to 0.47 V (neutral
solution) may involve some replacement of an axial
Scheme 1.
The 1H NMR features of the resultant binuclear
complex are shown in Fig. 4. At first there appears to
be two sets of differentiated H-6 and H-6% protons.
However, the splittings would require J5–6 or J5%–6%
couplings of 32 and 27 Hz, too large for normal pyridyl
rings. Values nearer 5.0 Hz are observed for
[PtII(dpaH)(glycine)]+ or [PtII(dpaH)(2-pyridine-car-
boxylate]+ as models of coordinated dpaH chro-
mophores [7,8]. A structural model of the headgroups
of tpada coordinated to [RuII(hedta)] proved to be
informative. Additional differentiated H-3 and H-3%
resonances appear at 7.84 and 7.72 ppm indicative of
the fact that the amide carbonyl functional group is not
rapidly rotating. This places the carbonyl in close con-
tact with the H-3 or H-3% protons. Planarity of the
amide carbonyl functionality forces a differentiation of
the H-3 and H-3% protons in order to preserve delocal-
ization between the dpa–R pyridyl rings. Thus, 50% of
Fig. 4. 300 MHz 1H NMR spectrum of {[RuII(hedta)]2(tpada)]}2− in D2O. [complex]0=1.31×10−2 M, T=25°C, pD=6.0.

R.E. Shepherd et al. / Inorganica Chimica Acta 303 (2000) 30–39
37
Table 3
1H NMR shifts for {[Ru(hedta)]2(tpada)} (shifts in ppm) for tether protons
H-6
H-6%
H-5
H-4
H-3
H-3%
Rotamer A
Rotamer B
9.08
8.97
8.51
8.40
7.44
7.44
7.94
7.94
7.84
7.84
7.71
CH2–CꢀO
CH2CH2
Rotamers A, B
2.61
1.72
carboxylate by OH− or a medium effect. The shift rela-
tive to neutral solution (0.04 V) is much less than for the
dpaH complex (0.17 V). Once oxidized to RuIII, both the
dpa− ligand or the dpa–R units of tpada dissociate in
basic solutions, as detected by the appearance of the
RuIII/IV wave at 0.91 V. The product is 38.3% converted
to [RuIII(hedta)(OH)2]2− in 7.5 s. Therefore, the loss of
tpada is about six times slower than dpa− at pH 9.5. The
presence of this feature in the DPP scan for the tpada
complex also rules out that the 0.91 V oxidation wave
has anything to do with association of the depronated
amide’s lone-pair for RuIII in the [RuIII(hedta)(dpa)]−
complex, but rather is a feature of ligand dissociation
from the RuIII species in each case.
4. Conclusions
2,2%-Dipyridylamine (dpaH) has been widely studied
in its CuII complexes, but far less so with softer metal
centers. Herein, coordination of dpaH to [RuII(hedta)]−
has been utilized to establish that dpaH is much less p-
accepting than bpy. DpaH stabilizes RuII versus RuIII by
0.22 V less than bpy. Also, dpaH and dpa− dissociate
from [RuIII(hedta)], whereas bpy does not readily do so.
Beginning studies of the dpaH-related ligand, tpada,
in which two dpa headgroups are linked by a six-
atom tether linkage are reported. The tpada ligand is
shown to form binuclear metal species of the type
{[(LM)]2(tpada)}. In the present study, the hedta3− lig-
and assists the solubilization of RuII attached to tpada.
The chemical and electrochemical behavior of the RuII-
tpada derivatized site is similar to [RuII(hedta)(dpaH)]−
with tpada acting as an even better p-acceptor ligand
than dpaH, approaching the p-acceptor power of bpy.
An energy-minimized computer-generated structure for
{[RuII(hedta)]2(tpada)} is shown in Fig. 6.
It is useful to point out that the RuII/III wave for
{[Ru(hedta)]2(tpada)}2− is more positive of the
[Ru(hedta)(dpaH)]− complex by 0.15 V using the pH-
independent couple for dpaH in the comparison. This in-
dicates that the tpada headgroup is a better p-acceptor/
poorer p-donor than dpaH alone. This is anticipated
on the basis of the withdrawing effect of the carbonyl
group of tpada. Since the E1/2 value for [Ru(hedta)-
(bpy)]− is 0.54 V, the effect of the carbonyl is not large
enough to make tpada as good of a p-acceptor as bpy,
but it is still better than dpaH in its p-acceptor capacity
as detected by the extent that the ligand stabilizes RuII
and makes the E1/2 value of the RuII/III couple more
positive:
The perspective of the RuII site at the left is along an
axis that allows the viewing of the axial carboxylato
RuII(hedta)L
H2O ꢁ dpaH Btpada Bbpy
E1/2 (V) 0.00 0.32 0.47 0.56
As anticipated for such a long insulating tether chain,
there is virtually no coupling except through space be-
tween the [RuII(hedta)]− headgroups. Had there been
such a coupling along the chain, there should be a split-
ting of the RuII/III wave such that the corrected com-
proportionation constant may be calculated by
1/4(e−DE/0.05916). No evidence for a wave separation is
detected in Fig. 4, indicative of insulated metal binding
sites.
Fig. 5. Differential pulse voltammogram of the binuclear complex
{[RuII(hedta)]2(tpada)]2−
.
[complex]=2.61×10−3 M, T=22°C,
v=0.10 (NaCl), under N2 purge. Dashed curve, pH 9.50; solid lines,
pH 7.0 and 3.0 (virtually the same).

38
R.E. Shepherd et al. / Inorganica Chimica Acta 303 (2000) 30–39
Fig. 6. Computer-generated, energy-minimized structure of {[RuII(hedta)]2(tpada)}2− by MM2 calculation.
donors and the plane containing the ‘en-backbone’
donors of the hedta3− portion. Toward the forefront is
the attachment of the ‘dpa-like’ headgroups of the
tpada ligand which has rings that are greatly non-copla-
nar. At the opposite end of the tether, the second RuII
center repeats the motif. But the second ‘dpa-like’
headgroup is rotated to minimize contacts along the
tether and for long-range avoidance of the other RuII
sites’s ‘dpa’ connection. All calculated bond distances
properties as a chelate for softer metal centers for use
as bimetallic delivery agent. The ligand, itself, and its
complexes have solubility limitations which are being
addressed by adding hydroxyl attachments along the
tether chain.
References
[1] (a) C.D. Burbridge, D.M.L. Goodgame, J. Chem. Soc. (A)
(1967) 694. (b) W.P. Jensen, R.A. Jacobson Inorg. Chim. Acta
49 (1981) 199. (c) H.C. Freeman, M.R. Snow, Acta Crystallogr.
18 (1965) 845.
[2] J.E. Johnson, T.A. Beinke, R.A. Jacobson, J. Chem. Soc. (A)
(1971) 1371.
were found to be normal with RuII–N distances near
3−
,
2.0–2.1 A. The hedta
portion of each metal site is
relatively unhindered with respect to the ‘dpa head-
group’ coordination.
The tpada ligand appears to be useful in forming
tethered binuclear metal centers that could be used to
span the major groove of DNA. On-going studies are
aimed at other RuIIL headgroups that have labile coor-
dination sites for making DNA attachments, as well as
tpada chelation. The present {[Ru(hedta)]2(tpada)}2−
complex illustrates how a central Ru2II(tpada) core
might be transported into cells by iminocarboxylate
ligands or small peptides to increase the solubility.
These carrier ligands would need to be replaced by
purine bases of the DNA strand, or stronger donors of
a protein in order for the system to make useful DNA–
DNA or DNA–DNA–protein cross-links. Once the
DNA–DNA or DNA–DNA–protein cross-links are
formed, further repair processes in the major groove are
blocked. Such linkages have shown to be cytotoxic
when the trapped protein is a repair enzyme for re-
moval of metallo-DNA-lesions [15].
[3] J. Sletten, K. Svardal, A. Sorensen, Acta Chem. Scand. 47 (1993)
1091.
[4] (a) C. Tsiamis, A.G. Hatzidimitriou, L.C. Tsvellas Inorg. Chem.
37 (1998) 2903. (b) F. Ugozzoli, A.M.M. Landfrei, N. Marsich,
A. Camus, Inorg. Chim. Acta 256 (1997) 1.
[5] (a) R.A. Jacobson, W.P. Jensen, Inorg. Chim. Acta 114 (1986)
L9. (b) R.A. Jacobson, W.P. Jensen Inorg. Chim. Acta 49 (1981)
1981. (c) G. Giorgi, R. Cini, Inorg. Chim. Acta 151 (1988) 153.
[6] (a) N. Nagao, M. Muhaida, S. Tachinyashika, K. Mizumachi,
Bull. Chem. Soc. Jpn 67 (1994) 1802. (b) D.P. Segers, M.K.
DeArmond, J. Phys. Chem. 86 (1982) 3768. (c) W.L. Johnson,
J.F. Geldard, Inorg. Chem. 18 (1979) 664. (d) P.A. Anderson,
G.B. Deacon, K.H. Haarmann, F.R. Keene, T.J. Meyer, D.A.
Reitsma, B.W. Skelton, G.F. Strosse, N.C. Thomas, J.A. Tread-
way, A.H. White, Inorg. Chem. 34 (1995) 6145.
[7] A.K. Paul, H. Mansuri-Torshizi, T.S. Srivastava, S.J. Chavan,
M. Chitnis, J. Inorg. Biochem. 50 (1993) 9.
[8] A.K. Paul, T.S. Srivastava, S.J. Chitnis, S. Desai, K.K. Rao, J.
Inorg. Biochem. 61 (1996) 179.
[9] (a) S.M. Zakeeruddin, M. Gratzel, D.M. Frazer, Biosens.
Bioelectr. 11 (1996) 305. (b) M. Sawa, K. Ogino, W. Tagaki,
Bull. Chem. Soc. Jpn. 69 (1996) 765.
[10] (a) S. Zhang, R.E. Shepherd, Transition Met. Chem. 17 (1992)
97. (b) Y. Chen, S. Zhang, R.E. Shepherd, Transition Met.
Chem. 22 (1997) 338.
Parallel studies with PdII have recently generated
{[PdCl2]2(tpada)} and [Pd(tpada)Cl]+, 2:1 and 1:1 com-
plexes [37]. NMR evidence has shown that the 2:1
species exists in the extended chain arrangement as for
{[RuII(hedta)]2(tpada)}2−. The 1:1 complex is too
strained to allow for all four pyridyl groups to be
bound as in structure 3. Rather three pyridyl donors
are coordinated to PdII with one pyridyl arm remaining
pendant [37]. The tpada ligand satisfies many desirable
[11] S. Zhang, Y. Chen, R.E. Shepherd, Inorg. Chim. Acta 230
(1995) 77.
[12] Y. Chen, R.E. Shepherd, J. Inorg. Biochem. 68 (1997) 183.
[13] S. Zhang, R.E. Shepherd, Transition Met. Chem. 17 (1992) 199.
[14] N.P. Farrell, in: S.B. Howell (Ed.), Platinum and Other Metal
Compounds in Cancer Chemotherapy, Plenum, New York,
1991, pp. 81–91.

R.E. Shepherd et al. / Inorganica Chimica Acta 303 (2000) 30–39
39
[15] (a) B. VanHoughten, S. Illeyne, Y. Qu, N. Farrell, Biochemistry
32 (1993) 11794. (b) Y. Qu, N. Farrell, Inorg. Chem. 34 (1995)
3573.
[16] (a) N.P. Farrell, Y. Qu, B. VanHoughten, Biochemistry 29
(1990) 9522. (b) T.D. Roberts, B. VanHoughten, Y. Qu, N.P.
Farrell, Nucleic Acids Res. 17 (1989) 9719.
[17] N.P. Farrell, Y. Qu, M.P. Hacker, J. Med. Chem. 33 (1990)
2179.
[18] Y. Qu, N.P. Farrell, J. Inorg. Biochem. 40 (1990) 255.
[19] E. Schuhman, J. Altman, W. Karaghiosoff, W. Beck, Inorg.
Chem. 34 (1995) 2316.
[20] R. Alul, M.B. Cleaver, J.-S. Taylor, Inorg. Chem. 31 (1992)
3636.
[21] S. Zhang, R.E. Shepherd, Inorg. Chem. 27 (1988) 4712.
[22] S. Zhang, R.E. Shepherd, Inorg. Chim. Acta 193 (1992) 217.
[23] K. Shimizu, Bull. Chem. Soc. Jpn. 50 (1977) 2921.
[24] S. Zhang, L.A. Holl, R.E. Shepherd, Inorg. Chem. 29 (1990)
1012.
[25] M.G. Elliott, S. Zhang, R.E. Shepherd, Inorg. Chem. 28 (1989)
3036.
[26] R.E. Shepherd, S. Zhang, F.-T. Lin, R.A. Kortes, Inorg. Chem.
31 (1992) 1457.
[28] R.E. Shepherd, S. Zhang, P. Dowd, G. Choi, B. Wilk, S.-C.
Choi, Inorg. Chim Acta 174 (1990) 249.
[29] S. Zhang, L.A. Holl, R.E. Shepherd, Transition Met. Chem. 17
(1992) 390.
[30] (a) Y. Chen, R.E. Shepherd, Inorg. Chim. Acta 268 (1998) 287.
(b) Y. Chen, R.E. Shepherd, Inorg. Chim. Acta 279 (1998) 85.
[31] R.E. Shepherd, Y. Chen, C.R. Johnson, Inorg. Chim. Acta 267
(1998) 11.
[32] (a) R.E. Shepherd, S. Zhang, Inorg. Chim. Acta 191 (1992) 271.
(b) Y. Chen, F.-T. Lin, R.E. Shepherd, Inorg. Chem. 38 (1999)
973. (c) Y. Chen, R.E. Shepherd, Inorg. Chem. 37 (1998) 1249.
[33] (a) A.A. Diamantis, J.V. Dubrawski, Inorg. Chem. 20 (1981)
1142. (b) A.A. Diamantis, J.V. Dubrawski Inorg. Chem. 22
(1983) 1934.
[34] (a) T. Matsubara, C. Creutz, J. Am. Chem. Soc. 100 (1978) 6255.
(b) T. Matsubara, C. Creutz, Inorg. Chem. 18 (1979) 1956.
[35] N. Nagao, M. Mukaida, S. Tachiyashiki, K. Mizumachi, Bull.
Chem. Soc. Jpn. 67 (1994) 1802.
[36] M.F. Hoq, R.E. Shepherd, Inorg. Chem. 23 (1984) 1851.
[37] T.W. Stringfield, R.E. Shepherd, work in progress.
[38] (a) H. Bock, H. Schodel, T.T.H. Van, R. Dienelt, M. Gluth, J.
Prak. Chem. Zeit. 340 (1998) 722. (b) F.A. Cotton, L.M.
Daniels, G.T. Jordan, C.A. Murillo, Polyhedron 17 (1997) 581.
[27] R.E. Shepherd, S. Zhang, C. Chen, Inorg. Chim. Acta 253
(1996) 65.
.