Control of redox potential by deprotonation of coordinated 1H-imidazole in complexes of 2-(1H-imidazol-2-yl)pyridine

Article abstract of DOI:10.1002/hlca.200590033

Complexes [ML₃]²⁺ of the bidentate ligand 2-(1H-imidazol-2-yl)pyridine were prepared with iron(II), cobalt(II), and ruthenium(II). The electronic spectra suggest the ligand to be a weaker σ-donor and π-acceptor than the closely related 2,2′-bipyridine. The complexes are readily deprotonated by addition of base, and the effect of the deprotonation is to lower the M^III/M^II redox potential by roughly 900 mV. This is roughly 75% of the drop observed for related complexes of 2,6-di-1H-imidazol-2-ylpyridine, and suggests the effect to be largely coulombic in origin.

Full text of DOI:10.1002/hlca.200590033

Helvetica Chimica Acta ± Vol. 88 (2005)
487
Control of Redox Potential by Deprotonation of Coordinated 1H-Imidazole in
Complexes of 2-(1H-Imidazol-2-yl)pyridine
by Gilles Stupka, Ludovic Gremaud, and Alan F. Williams*
Department of Inorganic, Analytical and Applied Chemistry, University of Geneva, 30 quai Ernest Ansermet,
CH-1211 Geneva 4
(
phone: ‡ 41223796425; fax: ‡ 41223796830; e-mail: Alan.Williams@chiam.unige.ch)
Dedicated to Professor Andr e Merbach on the occasion of his 65th birthday
Complexes [ML
3
]2 of the bidentate ligand 2-(1H-imidazol-2-yl)pyridine were prepared with iron(II),
‡
cobalt(II), and ruthenium(II). The electronic spectra suggest the ligand to be a weaker s-donor and p-acceptor
than the closely related 2,2'-bipyridine. The complexes are readily deprotonated by addition of base, and the
III
II
effect of the deprotonation is to lower the M /M redox potential by roughly 900 mV. This is roughly 75% of the
drop observed for related complexes of 2,6-di-1H-imidazol-2-ylpyridine, and suggests the effect to be largely
coulombic in origin.
Introduction. ± Although 1H-imidazole is easily the most important N-containing
ligand in biological systems, 1H-imidazole-containing ligands have been much less
studied in abiotic coordination chemistry than the related pyridine ligands. However,
1
H-imidazole is naturally a bifunctional ligand, possessing a pyrrolic N-atom in
addition to the pyridinic N-atom, which acts as a donor to transition metals. This NH
functionality is available for H-bonding, and may also be deprotonated. The H-bonding
property has been intensively studied in connection with metalloporphyrins [1], and H-
bonding of 1H-imidazole coordinated to an iron porphyrin has been shown to shift the
redox potential of the metal center by some 60 mV [2], stabilizing the iron(III)
oxidation state. Coordination of 1H-imidazole significantly lowers the pK value of the
a
pyrrolic H-atom [3 ± 7] by ca. 5 ± 7 log units, allowing deprotonation to occur in weakly
basic solution (pH8 ± 10). The resulting imidazolate can act as a bridging ligand, as is
well known in superoxide dismutase [8] and in many synthetic systems [9 ± 12]. A less
explicitly studied effect of deprotonation is the change in the metal redox potential. We
have shown recently [13][14] that the ligand 2,6-di-1H-imidazol-2-ylpyridine (1) forms
2‡
stable complexes [M(1)2] with a number of transition metals, and that these
complexes are readily deprotonated in basic solution. For M ˆ Co, Fe, and Ru, the
III
II
potential of the M /M couple is lowered by roughly 1300 mV upon full deprotonation
of the complex. We were interested to see whether this large effect is observed with
other 1H-imidazole-containing ligands, and, if so, to what extent the change in redox
potential is dependent on the number of 1H-imidazoles deprotonated. We report here
2‡
our results for the complexes [M(2) ] , which have three potentially acid pyrrolic
3
2‡
protons compared to four in [M(1) ] .
2
The ligand 2-(1H-imidazol-2-yl)pyridine (2) has been known for many years, but its
chemistry has been relatively little studied in comparison with that of the closely
ꢀ
2005 Verlag Helvetica Chimica Acta AG, Zürich

4
88
Helvetica Chimica Acta ± Vol. 88 (2005)
related 2,2'-bipyridine (bipy). Chiswell, Lions, and Morris [15] reported the syntheses
II
of a number of complexes of 2 with a variety of M ions and noted that deprotonation
occurred. Eilbeck and Holmes [16] a few years later reported the stability constants for
II
the formation of complexes of 2 with M ions (M ˆ Fe, Co, Ni, Cu, Zn, Cd, and Hg).
Boggess and Martin [17] showed that the pyrrolic protons were indeed acidic in the
2
‡
complexes of formula [M(2)3] (M ˆ Fe, Co, Ni, Cu, and Zn) with pK values in the
a
2‡
range 8 ± 11. The complex [Fe(2)3] has attracted a fair amount of attention, since it
shows spin-crossover behavior in the solid state [18][19] and in solution [20 ± 22]. A
2‡
coordination polymer containing [Fe(2) (4,4'-bipy)] repeat units was recently
2
reported to show spin-crossover properties [23], and 2 was also used as a spectator
ligand in the design of phosphatase model compounds [24]. To our knowledge, the only
crystal structure of a complex of 2 is the coordination polymer [Cu(2)(OH )(dicy-
2
anamide)](NO ) [25]. None of these publications discussed the electrochemistry,
3
although Baker and Goodwin [19] noted that the deprotonation of [Fe(2)3]2 led to the
‡
neutral low-spin iron(III) complex [Fe(2 À H) ]. The electronic spectrum [26] and
3
‡
2
electrochemistry [27] of the complex [Ru(2) ] have been reported, but no reference
3
to possible deprotonation was made. The only case where the effect of deprotonation
2
‡
was discussed was [Ru(bipy) (2)] , where Haga [28] reported that the deprotonation
2
III
II
of ligand lowered the Ru /Ru potential by 380 mV. The current work reports the
2‡
synthesis of the complexes [M(2)3] (M ˆ Fe, Co, Ni, and Ru) and the effect of
deprotonation on the redox potentials. Some preliminary results on manganese
complexes are also presented.
Results. ± Synthesis. The complexes [M(2)3]2 were generally prepared by simple
mixing of alcoholic solutions of ligand with aqueous solutions of metal salt. Partial
evaporation of the solution under vacuum followed by cooling gave the complex salts as
‡
microcrystalline solids, which were characterized by elemental analysis and electro-
II
spray mass spectroscopy. The Ru complex was prepared by refluxing RuCl with a
3
slight excess of 2 in ethylene glycol, followed by precipitation with ammonium
III
III
hexafluorophosphate. The deprotonated complexes of Co and Fe were generated by
2‡
addition of NaOHsolution to solutions of [M( 2)3] whereupon the neutral complex
2
‡
[
M(2 À H) ] precipitated; when applied to [Ru(2) ] , this gave a precipitate of
3
3
2‡
Na[Ru(2 À H) ] showing that oxidation had not occurred. The [Ni(2) ] complex
3
3
showed a change in spectrum upon adding base, but the precipitate did not give a
satisfactory analysis. Attempts to obtain manganese complexes were not successful.
Mixing manganese(II) salts with ligand 2 gave solutions whose UV/VIS spectra were
2‡
consistent with the formation of [Mn(2)3] but addition of base gave dark green
solutions that were unstable and gave precipitates of MnO . However, it was possible to
2

Helvetica Chimica Acta ± Vol. 88 (2005)
489
synthesize a higher-valence compound of manganese, [Mn O (2) ](ClO ) , following a
2
2
4
4 3
method developed for bipyridine complexes [29].
UV/VIS Spectra. UV/VIS Spectra were recorded in MeOH. Ligand 2 shows two
2‡
bands at 272 and 292 nm attributed to p-p* transitions. The complexes [M(2)3] (M ˆ
Co, Ni) show these two bands slightly red-shifted together with the appearance of a new
3
weaker band around 340 nm (cf. Fig. 1). The intensity of this band (e ꢀ 8000 dm ´
À1
À1
mol ´ cm ) and the fact that the position does not change greatly with change in the
metal ion leads us to associate this with a ligand-centered transition, which is a useful
marker of complexation of the ligand. Bands attributed to d-d transitions were
II
II
observed at 974 nm (e 6.7) for Co and 878 (e 5) and 538 nm (e 7) for Ni . These are
red-shifted in comparison to the equivalent bipy complexes [30], indicating that 2
produces weaker ligand-field splitting than bipy, and consistent with the spin-crossover
2‡
6
II
II
behavior of [Fe(2)3] mentioned above [18 ± 22]. The d complexes of Ru and Fe )
show the same ligand-based transitions, but any d-d bands are masked by strong
II
absorption due to a metal-to-ligand charge transfer (MLCT) at 485 nm for Fe and
II
4
35 nm for Ru (cf. Fig. 1). This last value is close to that reported for this complex in
aqueous solution [27]. For both metals, the MLCT band is blue-shifted in comparison
II
to the analogous bipy complexes, where the band is observed at 532 nm (Fe ) [30] and
II
2‡
4
52 nm [27]. The MLCT band for the Fe complex is less intense than for [Fe(bipy) ]
3
as a result of the spin-crossover behavior [20][31].
]2 (black line) and [Fe(2)
‡
]
2‡
(grey line) showing the band at
Fig. 1. UV/VIS Spectra (MeOH) of [Ni(2)
3
3
II
3
50 nm indicating complexation of the ligand and the MLCT transition at 485 nm for the Fe complex
Addition of base induces deprotonation. For the NiII complex, this results in a
doubling of the intensity of the ligand-based band at 340 nm but in only a slight red shift
À1
of ca. 100 cm of the d-d transitions, so we may assume that the ligand field is not
II
changed significantly. For Co , we observe a decrease in the intensity of the ligand
bands at 272 and 292 nm with a dramatic increase in the band at 340 nm. The d-d

4
90
Helvetica Chimica Acta ± Vol. 88 (2005)
III
II
transition at 974 nm disappears, consistent with the oxidation to Co . The Fe com-
II
plex shows a behavior similar to that of the Co complex for the ligand-based
transitions. The MLCT band disappears, and a new band, assigned to ligand-to-metal
III
charge transfer (LMCT) for an Fe complex is observed at 739 nm. Baker and
Goodwin reported a similar spectrum [19] for this complex and pointed out the
disagreement with the spectrum reported by Lions et al. [15]. We suspect that this
II
arises from only partial oxidation of Fe in the latter case. As we discuss below, the
potential is lowered less than in the complex of ligand 1, and it is possible that the
II
isolated complex is contaminated with Fe , which has a higher extinction coefficient
III
in the VIS than the Fe complex, and thus confers a red color on the solid. In support
II
of this, the Ru complex (which has a similar redox potential, vide infra) does not
undergo oxidation upon addition of base, but is deprotonated and gives a precipitate
of Na[Ru(2) ]. The spectrum of this deprotonated salt shows the usual increase
3
in intensity at 340 nm, and the MLCT band is split into two bands at 423 and
469 nm.
II
The Mn complex gave a spectrum in which the appearance of a band at 340 nm
indicates that complexation had taken place. Addition of base gave a dark green
solution, suggesting that oxidation had occurred, but the solution was unstable and gave
3‡
an intractable green precipitate. The complex [Mn O (2) ] could be isolated as its
2
2
4
perchlorate salt. The UV/VIS spectrum is very similar to that observed for the
analogous bipy and phen complexes [29], but attempts to deprotonate the imidazole
led to decomposition of the complex.
Isomers. Since 2 does not possess the full symmetry of 2,2'-bipyridine (bipy),
n‡
complexes of the type [M(2)3] exist as two isomers, mer (ˆ(OC-6-21)) and fac
(
ˆ(OC-6-22)). One would not expect any great difference in stablility between mer
and fac isomers given the similar sizes and donor powers of the pyridine and imidazole
moieties, and consequently one would expect a statistical value of 3 :1 for the mer/fac
ratio. This was observed for the similar complex tris(2-(1-methyl-1H-benzimidazol-2-
yl)pyridine)cobalt(II) [32], where the kinetics of isomerization and the pressure
dependence were studied, and the mer/fac isomerism of pyridinyl-1H-benzimidazole
ligands around ruthenium was discussed recently by Piguet and collaborators [33]. In
1
2‡
the present work, the H-NMR spectrum of [Ru(2) ] showed clearly a mixture of
3
isomers. In the mer isomer, the three ligands are all inequivalent, giving a total of 18
different protons in the aromatic region, while, in the fac isomer, the C axis makes all
3
three ligands equivalent and gives six aromatic-proton signals. Fig. 2 shows the
2‡
aromatic region of the spectrum of [Ru(2) ] . The four d around 6.6 ppm are
3
attributed to one of the imidazole protons, and integration indicates that the mer/fac
ratio is close to the statistical value of 3 :1. For most of the other complexes, the
1
H-NMR spectra were broadened by paramagnetic effects, and no information could be
obtained, but, for [Co(2 À H) ], a well-resolved diamagnetic spectrum was observed,
3
II
III
confirming the oxidation of Co to Co . In this case, the spectrum in the region of
6
ppm showed three peaks of equal intensity and one of ca. 20% of this intensity,
implying a mer/fac ratio of ca. 15 :1. Because of the statistical weighting in favor of the
mer form, this corresponds to an energy difference of only a few kJ/mol. Following
Lever [34], we assume that the mer and fac forms have essentially identical electronic
spectra and electrochemistry.

Helvetica Chimica Acta ± Vol. 88 (2005)
491
Fig. 2. 1H-NMR Spectrum of [Ru(2)
]
2‡
: the signals at ca. 6.6 ppm may be used to quantify the mer/fac ratio (see
text)
3
II
II
II
Electrochemistry. ± The cyclic voltammograms of the Co , Fe , and Ru complexes
II
were studied, and the results are summarized in the Table 1. The Ni complex showed
II
II
no observable electrochemical activity. The Fe and Ru complexes gave essentially
reversible waves (cf. Fig. 3), but the Co peak separations were much greater. The
potentials for the protonated complexes lie in the expected range. Thus, the potentials
2‡
for [M(2)3] (M ˆ Co and Fe) are close to those reported for the related 2-(1-methyl-
II
1
H-benzimidazol-2-yl)pyridine complexes [35], and that for Ru is close to the value
/3‡
]2 (left) and [Fe(2 À H)
À/0
Fig. 3. Cyclic voltammograms (MeCN) of [Fe(2)
3
3
]
(right)

4
92
Helvetica Chimica Acta ± Vol. 88 (2005)
]2 and the
‡
Table. Electrochemical Data Obtained by Cyclic Voltammetry (MeCN) for Complexes [M(2)
3
À
Deprotonated Complexes [M(2 À H)
3
] . Potentials are given with respect to the SCE. DE
between anodic and cathodic peaks.
p
is the separation
Solvent
Ligand ˆ 2 Ligand ˆ 2 À H
DE1/2/mV
E
1/2/V
DE
p
/mV
E
1/2/V
p
DE /mV
Co
Fe
Ru
DMF
MeCN
DMF
0.268
0.835
0.836
125
81
91
À 0.616
0.004
À 0.083
325
139
125
À 884
À 831
À 919
previously reported in MeCN [27]. In general, the potentials are lower than for the
equivalent bipy complexes but higher than those for the related complexes of ligand 1
[
13][14]. This is in agreement with the observations of the MLCT bands discussed
II
above. If one expresses the value for Ru relative to the NHE, a value for Leverꢁs
electrochemical parameter [34] E of 0.18 is obtained.
L
Deprotonation of the complexes results in a dramatic drop in the potential for the
III
II
M /M couple, consistent with the aerial oxidation of the complexes of Fe and Co. The
average decrease is just under 900 mV, to be compared with a drop of 1250 mV
2‡
observed upon deprotonation of [M(1) ] . The change is thus roughly proportional to
2
the number of 1H-imidazole moieties complexed to the metal ion and corresponds to
300 mV per 1H-imidazole unit. This value may be compared to the drop of 60 mV
estimated for H-bonding to 1H-imidazole in porphyrin systems [2] and is close to the
value of 380 mV reported by Haga [28] for the effect of deprotonating
2
‡
[
Ru(bipy) (2)] . The Lever E value calculated for deprotonated 2 from the Ru
2
L
complex is 0.02, corresponding in his classification to a weakly p-acid unsaturated
amine [34].
3‡
The electrochemistry of the mixed-valence manganese complex [Mn O (2) ]
2
2
4
IV
IV
III
showed a signal corresponding to Mn2 ! Mn Mn at ‡ 1.21 V (DE ˆ 131 mV),
p
slightly lower than the values for the phen and bipy complexes, in agreement with the
general tendency observed for 2. An irreversible wave attributed to the reduction to
III
Mn was observed just below 400 mV.
2
Discussion. ± The results presented here confirm and complete the observations
with ligand 1. The observation that the drop in redox potential is roughly constant for
each ligand, and proportional to the number of deprotonated 1H-imidazole moieties
argues that the effect is due to the coulombic effect of building up negative charge on
the ligand and thereby stabilizing a higher oxidation state for the complexed metal by a
Madelung potential. As judged by the insignificant change in the position of the d-d
bands for the Ni, the deprotonation does not greatly change the interaction of the
ligand with the metal, and, if specific orbital effects were involved, we would expect
5
6
more-significant differences in the effect of deprotonation between the d /d iron and
6
7
III
II
Ru couples, and the d /d Co /Co couple.
In both ligands 1 and 2, the 1H-imidazole ligand is bound to an aromatic moiety, and
this may play an important role in stablizing the deprotonated 1H-imidazole. In support
of this, deprotonation of histidine does not appear to take place readily in biological
systems. The 1H-imidazole ligands studied by Matsumoto and co-workers [36][37] also

Helvetica Chimica Acta ± Vol. 88 (2005)
493
show deprotonation, but in this case the 1H-imidazole is bound to an imine
functionality that also offers a means of delocalizing the negative charge generated
by deprotonation of the 1H-imidazole. Preliminary experiments in our laboratory [38]
II
suggest that deprotonation of the 1H-imidazole ligand coordinated to M ions does not
occur when bound to purely aliphatic groups. If this is indeed the case, then this is
unlikely to be an important mechanism for biological systems where the 1H-imidazole
is attached to a saturated chain. Nonetheless, the coupling between protonation state of
ligand 2 and the redox potential of the metal offers a possible switching mechanism
controlled by pH.
We thank the Swiss National Science Foundation for the support of this work.
Experimental Part
General. Solvents and starting materials were purchased from Fluka AG (Buchs, Switzerland) and used
without further purification, unless otherwise stated. MeCN and dimethylformamide (DMF) were distilled from
CaH . Pyridine-2-carboxaldehyde was freshly distilled before use. UV/VIS Spectra: Cary 1E or Perkin-Elmer
2
Lambda-900 UV/VIS/NIR spectrometer; quartz cells of 0.1- and 0.01-cm path lengths; lmax (e) in nm. IR
Spectra: Perkin-Elmer Spectrum-One instrument; KBr discs; NMR Spectra: Varian Gemini-300 instrument; at
1
13
3
00 ( H) and 74.44 MHz ( C) at 208; chemical shifts d in ppm with respect to SiMe
4
, J in Hz; the residual solvent
d(H) was used as a reference, i.e., D
.41), and d(C) 67.4 of dioxane for D
2
O d 4.80, CD
3
CN d 1.95, CD
3
OD d 3.31, CD Cl d 7.26, and ((CD SO d
3
3 2
)
À1
2
2
O solns. nÄ in cm . MS: electron impact (EI) at 70 eV with VG 7000E and
Finnigan 4000 instruments, electrospray ionization (ESI) with Finnigan Mat-SSQ-7000 instrument of the Mass
Spectrometry Laboratory, University of Geneva; in m/z (rel. %). Elemental analyses were performed by Dr. H.
Eder, University of Geneva.
Electrochemistry. Cyclic voltammograms were recorded by using a BAS-CV-50W potentiostat connected to
a personal computer. A three-electrode system consisting of a stationary Pt-disk working electrode, a Pt counter
electrode, and a nonaqueous Ag/AgCl reference electrode was used. Bu
electrolyte. Before each measurement, the system was standardized with the ferrocene/ferrocenium couple
1/2 ˆ ‡0.50 vs. SCE [39]) in DMF, and with [Ru(bipy) ](ClO [40] in MeCN. Potentials are the mean of
anodic and cathodic peaks and are given with respect to SCE.
-(1H-Imidazol-2-yl)pyridine (2) [15][20]. A soln. of freshly distilled pyridine-2-carboxaldehyde (22.65 g,
.21 mol) was added to a soln. of glyoxal (ˆethanedial; 29 ml, 40% aq. soln.) in EtOHat 0 8. Ice-cold
concentrated aq. ammonia (85 ml) was rapidly added, and the mixture was stirred in an ice bath for 1 h and
allowed to warm to r.t.. The mixture was extracted with Et SO ), and evaporated.
4 6
NPF (0.1m) served as an inert
(
E
3
4 2
)
2
0
2
O (10 Â 100 ml), dried (Na
2
4
The residue was treated with charcoal in refluxing EtOHand recrystallized from MeOH: 2 (9 g, 30%). White
solid. M.p. 1358. UV/VIS (MeOH): 272 (11900), 292 (15400). IR (KBr): ca. 3350w, 1614s, 1594s, 1568w, 1479s,
1
4
458s, 1414m, 1381m, 1307m, 1280m, 1171w, 1135m, 1108m, 992s, 955s, 788s, 737m, 708s, 620m, 503w, 466w,
1
3
3
03m. H-NMR (300 MHz, (D
6
)DMSO): 8.58 (d, J ˆ 4.2, 1 H); 8.03 (d, J ˆ 8.1, 1 H); 7.86 (td, J ˆ 7.8, 1.5, 1 H);
‡
.
7.34 (ddd, J ˆ 7.5, 4.8, 1.5, 1 H); 7.13 (s, 2 H). EI-MS: 145 (100, M ).
3
1
Tris[2-(1H-imidazol-2-yl-kN )pyridine-kN]cobalt(2‡) Diperchlorate ), ([Co(2)
Co(ClO ´ 6 H O (0.336 g, 0.92 mmol) in H O (5 ml) was added to a soln. of 2 (0.40 g, 2.75 mmol) in EtOH
5 ml). The resulting orange soln. was concentrated and, upon cooling, an orange powder precipitated which
was filtered and washed with Et O: [Co(2) ](ClO (0.600 g, 94%). UV/VIS (MeOH): 268 (26300), 293
32400), 974 (6.7). IR (KBr): 3244m, 3158m, 2934m, 1613s, 1571m, 1472s, 1295m, 1103 (br.), 965m, 929m, 794s,
3 4 2
](ClO ) ). A soln. of
4
)
2
2
2
(
2
3
4 2
)
(
2
‡
2‡
7
2
51m, 704s, 623s, 483w, 415w, 383w, 264w. ESI-MS (MeOH): 174.9 ([Co(2) ] , 100), 246.9 ([Co(2) ] , 15).
3
Anal. calc. for [Co(2)
3
](ClO
4
)
2
´ H
2
O: C 40.52, H3.25, N 17.72; found: C 40.82, H3.51, N 17.89.
1
Tris[2-(1H-imidazol-2-yl-kN)pyridine-kN]iron(2‡) Diperchlorate ) ([Fe(2)
3 4 2
](ClO ) ). A soln. of Fe-
(
(
ClO
4 2 2 2
) ´ 6 H O (0.417 g, 1.15 mmol) in H O (5 ml) was added to a soln. of 2 (0.50 g, 3.45 mmol) in MeOH
5 ml). The resulting red soln. was concentrated, and upon cooling, a purple powder precipitated which was
1)
Caution: Dry perchlorates may explode and should be handled in small quantities and with the necessary
precautions [41].

4
94
Helvetica Chimica Acta ± Vol. 88 (2005)
filtered and washed with Et
2
O: [Fe(2)
3 4 2
](ClO ) (0.706 g, 89%). UV/VIS (MeOH): 265 (21200), 295 (32100),
4
7
85 (2350). IR (KBr): 3250 (br.), 3158w, 1614s, 1567m, 1471s, 1450m, 1294m, 1103 (br.), 964m, 930m, 788s, 753s,
2
‡
2‡
3
03s, 623s, 504w, 455w, 411w, 384w. ESI-MS (MeOH): 173.3 ([Fe(2)
2
]
, 55, 245.9 ([Fe(2)
]
, 30). Anal. calc. for
). A soln. of
[
Fe(2)
3
](ClO
4
)
2
´ H
2
O): C 40.70, H3.27, N 17.79; found: C 40.74, H3.53, N 17.86.
3
Tris[2-(1H-imidazol-2-yl-kN )pyridine-kN]nickel(2‡) Bis(tetrafluoroborate) ([Ni(2)
3
](BF
O (5 ml) was added to a soln. of 2 (0.80 g, 5.5 mmol) in EtOH
5 ml). The resulting violet soln. was concentrated, and, upon cooling, a pink-violet powder precipitated, which
was filtered and washed with Et O: [Ni(2) ](BF (0.855 g, 70%). UV/VIS (MeOH): 264 (26400), 300 (38000),
38 (7), 878 (5). IR (KBr): 3302 (br.), 3159w, 2939w, 1615s, 1572m, 1493m, 1472s, 1295m, 1071 (br.), 964m,
4 2
)
4 2 2 2
Ni(BF ) ´ 6 H O (0.622 g, 1.83 mmol) in H
(
2
3
4 2
)
5
9
2
‡
30m, 794s, 753s, 704s, 640m, 521m, 422w, 386w, 280w. ESI-MS (MeOH): 174.2 ([Ni(2)
2
]
, 100), 246.7
O: C 42.03, H3.38, N 18.38; found: C 42.13, H3.70, N 18.45.
](PF ).
(0.150 g, 0.72 mmol) was refluxed in ethylene glycol (30 ml) for
O (30 ml) added, and the product precipitated with a conc. aq. NH PF soln.
The red-orange complex was filtered and washed with H O: [Ru(2) ](PF (0.416 g, 70%). UV/VIS (MeOH):
94 (43460), 435 (13800). IR (KBr): 3386m, 1614m, 1495s, 1469s, 1163m, 1115m, 1102m, 840 (br.), 782m, 751m,
2
‡
(
[Ni(2)
3
]
, 25). Anal. calc. for [Ni(2)
3
](BF
4
)
2
´ H
2
3
Tris[2-(1H-imidazol-2-yl-kN )pyridine-kN]ruthenium(2‡) Bis(hexafluorophosphate). ([Ru(2)
3
6 2
)
A mixture of 2 (0.344 g, 2.37 mmol) and RuCl
h. The soln. was cooled to r.t., H
3
1
2
4
6
2
3
6 2
)
2
7
1
3
02m, 558s, 397w, 348w, 301w. H-NMR (300 MHz, (D
6
)DMSO): 8.17 (d, J ˆ 7.8, 4 H); 7.95 (m, 4 H); 7.76
3
3
3
3
3
(
1
(
d, J ˆ 5.4, 2 H); 7.66 (d, J ˆ 5.7, 2 H); 7.59 (d, J ˆ 0.9, 2 H); 7.57 (d, J ˆ 0.9, 2 H); 7.30 (m, 4 H); 6.60 (d, J ˆ
3
3
3
.2, 1 H); 6.57 (d, J ˆ 1.2, 1 H); 6.53 (d, J ˆ 1.2, 1 H); 6.51 (d, J ˆ 1.2, 1 H). ESI-MS (MeOH): 268.4
2
‡
[Ru(2)
Tris[2-(pyridin-2-yl-kN)-1H-imidazolato-kN ]cobalt ([Co(2 À H)
.21 mmol) in a minimum of MeOHwas added 0.84 m aq. NaOH(1 ml, 4 equiv.): The resulting red-brown soln.
O: [Co(2 À H) ] (0.082 g,
0%). UV/VIS (MeOH): 264 (21100), 291 (17000), 346 (19600). IR (KBr): 1612s, 1558m, 1519s, 1462m, 1447m,
427m, 1344m, 1215m, 1144m, 1103m, 984m, 758s, 522m, 467m, 416w, 401w, 301w, 277w. H-NMR (300 MHz,
D )DMSO): 8.05 ± 7.95 (m); 7.90 (d); 7.83 (d); 7.20 ± 7.33 (m); 7.18 (s); 7.15 (s); 7.07 (s); 7.03 (s); 6.78 (d); 6.66
6
d); 6.19 (s); 5.97 (s); 5.87 (s); 5.71 (s). EI-MS : 491 ([Co(2 À H)
3 3 6 2
] , 100). Anal. calc. for [Ru(2) ](PF ) : C 34.87, H2.56, N 15.25; found: C 34.88, H3.23, N 15.20.
1
3
3 4 2
]). To a soln. of [Co(2) ](ClO ) (0.15 g,
0
was cooled to 08 and the precipitated red powder was filtered and washed with Et
8
1
(
(
2
3
1
3
], 65), 347 ([Co(2 À H)
2
], 55), 203 ([Co(2 À
H)], 9), 145 (L, 100). Anal. calc. for [Co(2 À H)
3
] ´ 3/2 H O: C 55.60, H4.08, N 24.31; found: C 55.75, H4.46, N
2
2
4.40.
1
Tris[2-(pyridin-2-yl-kN)-1H-imidazolato-kN ]iron ([Fe(2 À H)
3
]). To a soln. of [Fe(2)
N (3 ml, 4 equiv.). The resulting red-violet soln.
O: [Fe(2 À H) ] (0.092 g,
0%). UV/VIS (MeOH): 289 (23900), 331 (21800), 739 (730). IR (KBr): 1611s, 1558m, 1519s, 1449s, 1393m,
283m, 1203m, 1160m, 1101m, 984m, 781s, 756s, 710s, 505w, 455m, 438m, 411m, 385w, 301w. Anal. calc. for
3 4 2
](ClO ) (0.15 g,
0
.21 mmol) in a minimum of MeOHwas added 0.29 m aq. Et
3
was cooled to 08 and the precipitated red-violet powder filtered and washed with Et
9
1
2
3
[
Fe(2 À H)
3
] ´ 2 H
2
O: C 54.97, H4.22, N 24.04; found: C 55.30, H4.54, N 24.22.
1
Sodium Tris[2-(pyridin-2-yl-kN)-1H-imidazolato-kN ]ruthenium(1À) (Na[Ru(2 À H)
3
]). To a soln. of
[
3 6 2
Ru(2) ](PF ) (0.1 g, 0.12 mmol) in a minimum of MeOHwas added 0.48 m aq. NaOH(1 ml, 4 equiv.). The
resulting brown soln. was filtered and cooled to 08, and the precipitated red-brown powder filtered and washed
with Et ] (0.046 g, 69%). UV/VIS (MeOH): 298 (33200), 423 (8150), 469 (6480). IR (KBr):
385 (br.), 1603s, 1550m, 1501s, 1456m, 1328m, 1145m, 1099m, 972m, 937m, 779s, 751m, 710m, 642m, 508m,
2
O: Na[Ru(2 À H)
3
3
4
(
6
1
1
3
3
50w, 399m, 340w, 300w. H-NMR (300 MHz, CD
3
OD): 7.82 (m, 4 H); 7.73 (d, J ˆ 5.4); 7.66 (d, J ˆ 5.7); 7.62
3
d, J ˆ 5.4); 7.52 (m); 7.09 (s, 1 H); 7.08 (s, 1 H); 7.07 (s, 1 H); 7.05 (s, 1 H); 6.8 (m, 4 H); 6.43 (s, 1 H );
À
.37(s, 1 H); 6.36 (s, 1 H). ESI-MS (MeOH): 534.1 ([Ru(2 À H)
3
] , 30). Anal. calc. for [Ru(2 À H)
3 2
] ´ Na ´ H O ´
/4 MeOH: C 49.92, H 4.15, N 21.77; found: C 49.99, H 3.63, N 21.64.
Tetrakis[2-(1H-imidazol-2-yl-kN )pyridine-kN]di-m-oxo-dimanganese(3‡) Triperchlorate ) ([MnIIIMnI-
3
1
VO
2
4 4 3 2 2 2
(2) ](ClO ) ). To a soln. of [Mn(OAc) ´ 4 H O (0.169 g, 0.69 mmol) in H O (2.5 ml) was added 2 (0.3 g,
2
.07 mmol) in EtOH(2 ml). Acetate buffer (4 ml) was added. The yellow soln. was cooled to 0 8 in an ice bath
and KMnO (0.047 g, 0.3 mmol) in H O (2 ml) was added dropwise. The resulting green soln. was stirred for
5 min at 08 and filtered. Conc. NaClO soln. was added to precipitate the product as a green powder, which was
filtered and washed with Et O and Et
756w, 1615m, 1571m, 1474s, 1294m, 1103 (br.), 971m, 928w, 790s, 747m, 703s, 655m, 624s, 411w, 366w, 279w.
4
2
1
4
III
IV
2
2 2 4 4 3
O: [Mn Mn O (2) ](ClO ) (0.12 g, 24%). IR (KBr): 3567m, 3094w,
2
3
‡
3‡
2 2 4 4 2 2 4
ESI-MS (MeCN): 295.6 ([Mn O (2) (MeCN) ] , 70), 281.9 ([Mn O (2) (MeCN) ] , 20%). Anal. calc. for
3
[
2 2 4 4 3 2
Mn O ](2) (ClO ) ´ 2 H O: C 36.36, H3.03, N 15.90; found: C 36.28%, H3.23, N 15.84.

Helvetica Chimica Acta ± Vol. 88 (2005)
495
REFERENCES
[
[
[
[
[
[
[
[
1] R. Quinn, J. Mercer-Smith, J. N. Burtstyn, J. S. Valentine, J. Am. Chem. Soc. 1984, 106, 4136.
2] P. OꢁBrien, D. A. Sweigart, Inorg. Chem. 1985, 24, 1405.
3] T. R. Harkins, H. Freiser, J. Am. Chem. Soc. 1954, 78, 1143.
4] G. C. Hayward, H. A. O. Hill, J. M. Pratt, R. J. P. Williams, J. Chem. Soc. A 1971, 196.
5] J. M. Harrowfield, V. Norris, A. M. Sargeson, J. Am. Chem. Soc. 1976, 98, 7282.
6] M. F. Hoq, R. E. Shepherd, Inorg. Chem. 1984, 23, 1851.
7] C. R. Johnson, W. W. Henderson, R. E. Shepherd, Inorg. Chem. 1984, 23, 2754.
8] D. Bordo, A. Pesce, M. Bolognesi, M. E. Stroppolo, M. Falconi, A. Desideri, in ꢂ Handbook of
Metalloenzymesꢁ, Eds. A. Messerschmidt, R. Huber, T. Poulos, and K. Wieghardt, Wiley, Chichester ± New
York ± Weinheim ± Brisbane ± Singapore ± Toronto, 2001.
[9] P. V. Bernhardt, G. A. Lawrance, N. J. Curtis, Polyhedron 1987, 6, 1347.
[
[
[
10] C. Piguet, B. Bocquet, E. Müller, A. F. Williams, Helv. Chim. Acta 1989, 72, 323.
11] R. Quinn, M. Nappa, J. S. Valentine, J. Am. Chem. Soc. 1982, 104, 2588.
12] N. Matsumoto, Y. Motoda, T. Matsuo, T. Nakashima, N. Re, F. Dahan, J.-P. Tuchagues, Inorg. Chem. 1999,
38, 1165.
[
[
[
[
[
[
[
[
[
[
[
13] R. F. Carina, L. Verzegnassi, G. Bernardinelli, A. F. Williams, Chem. Commun. 1998, 2681.
14] G. Stupka, G. Bernardinelli, A. F. Williams, Dalton Trans. 2004, 407.
15] B. Chiswell, F. Lions, B. S. Morris, Inorg. Chem. 1964, 3, 110.
16] W. J. Eilbeck, F. Holmes, J. Chem. Soc. A 1967, 1777.
17] R. K. Boggess, R. B. Martin, Inorg. Chem. 1974, 13, 1525.
18] R. J. Dosser, W. J. Eilbeck, A. E. Underhill, P. J. Edwards, C. E. Johnson, J. Chem. Soc. A 1969, 810.
19] A. T. Baker, H. A. Goodwin, Aust. J. Chem. 1977, 30, 771.
20] K. A. Reeder, E. V. Dose, L. J. Wilson, Inorg. Chem. 1978, 17, 1071.
21] E. V. Dose, M. A. Hoselton, N. Sutin, M. F. Tweedle, L. J. Wilson, J. Am. Chem. Soc. 1978, 100, 1141.
22] J. K. Beattie, K. J. McMahon, Aust. J. Chem. 1988, 41, 1315.
23] G. S. Matouzenko, G. Molnar, N. Br e fuel, M. Perrin, A. Bousseksou, S. A. Borshch, Chem. Mater. 2003, 15,
550.
[
[
[
[
[
[
[
[
[
24] F. Verge, C. Lebrun, M. Fontecave, S. M e nage, Inorg. Chem. 2003, 42, 499.
25] J. Carranza, C. Brennan, J. Sletten, F. Lloret, M. Julve, J. Chem. Soc., Dalton Trans. 2002, 3164.
26] J. G. D. M. Atton, R. D. Gillard, Transition Met. Chem. 1995, 6, 351.
27] G. Orellana, M. L. Quiroga, A. M. Braun, Helv. Chim. Acta 1987, 70, 2073.
28] M. Haga, Inorg. Chim. Acta 1983, 75, 29.
29] S. R. Cooper, M. Calvin, J. Am. Chem. Soc. 1977, 99, 6623.
30] R. A. Palmer, T. S. Piper, Inorg. Chem. 1966, 5, 864.
31] K. A. Reeder, E. V. Dose, L. J. Wilson, Inorg. Chem. 1978, 17, 1071.
32] L. J. Charbonni eÁ re, A. F. Williams, U. Frey, A. E. Merbach, P. Kamalaprija, O. Schaad, J. Am. Chem. Soc.
1997, 119, 2488.
[
[
[
33] S. Torelli, S. Delahaye, A. Hauser, G. Bernardinelli, C. Piguet, Chem.±Eur. J. 2004, 10, 3503.
34] A. B. P. Lever, Inorg. Chem. 1990, 29, 1271.
35] L. J. Charbonni e re, A. F. Williams, C. Piguet, G. Bernardinelli, E. Rivara-Minten, Chem.±Eur. J. 1998, 4,
4
85.
36] I. Katsuki, Y. Motoda, Y. Sunatsuki, N. Matsumoto, T. Nakashima, M. Kojima, J. Am. Chem. Soc. 2002, 124,
29.
37] Y. Sunatsuki, Y. Ikuta, N. Matsumoto, H. Ohta, M. Kojima, S. Iijima, S. Hayami, Y. Maeda, S. Kaizaki, F.
Dahan, J.-P. Tuchagues, Angew. Chem., Int. Ed. 2003, 42, 1614.
[
[
6
[
[
[
38] M. Bergamo, N. Banerji, G. Bernardinelli, A. F. Williams, unpublished results.
39] J. C. Chambron, C. Coudret, J. P. Sauvage, New J. Chem. 1992, 16, 361.
40] A. J. Bard, L. R. Faulkner, ꢂElectrochemical Methods: Fundamentals and Applicationsꢁ, Wiley, New York,
1980.
[41] W. C. Wolsey, J. Chem. Educ. 1978, 55, A355.
Received November 12, 2004