Cobalt and silver complexes of terdentate pyrazine-based amide ligands and assembly of monocobalt building blocks through a silver connector

Article abstract of DOI:10.1002/ejic.200801187

Two terdentate pyrazine-based amide ligands have been prepared from methyl pyrazine-2-carboxylate and 2-(amino-methyl)pyridine (HL^1M) or 2- 2-aminoethyl)pyridine (HL^1E) in order to probe the potential of the "spare" nitrogen atom "out: the back" of the pyrazine ring to coordinate to a different metal ion and thereby act as a linker between complexes. Two inert cobalt(III) complexes, [Co^III(L ^1M)₂](BF₄) 1/4H₂O and Co ^III(L^1E)₂](BF₄)-1/2H₂0, have been prepared as building blocks and the silver(I) coordination of the ligands also probed, forming {[Ag^I(HL^1M)]BF ₄}∞, and [Ag^I₂(HL^1LE) ₂]- (BF₄)₂. The [Co^III(L ^1E)₂](BF₄) building block has been successfully connected to a second such complex by coordination of silver(I) to a "spare" pyrazine nitrogen atom on each complex, resulting in [{Co ^III(L^1E)₂}₂Ag^I](BF ₄)(N0₃)₂, All five complexes have been structurally characterised. Mass spectra and cyclic voltammetry studies on "aged" (kept in solution in air for 2 d) samples clearly showed that the cobalt complex of the methylene-linked ligand was prone to slow ligand oxidation, forming [Co^III(L^1Mox)(L^1M)](BF ₄) and [Co^III(L^1Mox)₂](BF ₄). Fresh samples of [Co^III(L^1M) ₂](BF₄)-1/4H₂0 and o^III(L ^1M)₂](BF₄)-1/2H₂0 undergo a chemically reversible one- electron reduction in dry acetonitrile, at -0.71 and -0.48 V vs. 0.01 M AgN0₃/Ag, respectively, consistent with the methylene-linked ligand being better able to stabilise the higher oxidation state of cobalt than the ethylene-linked ligand.

Full text of DOI:10.1002/ejic.200801187

FULL PAPER
DOI: 10.1002/ejic.200801187
Cobalt and Silver Complexes of Terdentate Pyrazine-Based Amide Ligands and
Assembly of Monocobalt Building Blocks through a Silver Connector
Ryan M. Hellyer,[a] David S. Larsen,[a] and Sally Brooker*[a]
Keywords: Pyrazine / Cobalt / Silver / Mixed-metal complexes / Electrochemistry / Amides
Two terdentate pyrazine-based amide ligands have been
prepared from methyl pyrazine-2-carboxylate and 2-(amino-
complexes have been structurally characterised. Mass spec-
tra and cyclic voltammetry studies on “aged” (kept in solu-
tion in air for 2 d) samples clearly showed that the cobalt
complex of the methylene-linked ligand was prone to slow
ligand oxidation, forming [CoIII(L1Mox)(L1M)](BF4) and [CoIII-
(L1Mox)2](BF4). Fresh samples of [CoIII(L1M)2](BF4)·¼H2O and
[CoIII(L1E)2](BF4)·½H2O undergo a chemically reversible one-
electron reduction in dry acetonitrile, at –0.71 and –0.48 V vs.
methyl)pyridine (HL1M) or 2-(2-aminoethyl)pyridine (HL1E
)
in order to probe the potential of the “spare” nitrogen atom
“out the back” of the pyrazine ring to coordinate to a dif-
ferent metal ion and thereby act as a linker between com-
plexes. Two inert cobalt(III) complexes, [CoIII(L1M)2](BF4)·
¼H2O and [CoIII(L1E)2](BF4)·½H2O, have been prepared as
building blocks and the silver(I) coordination of the ligands
also probed, forming {[AgI(HL1M)]BF4}ϱ and [AgI2(HL1LE)2]-
(BF4)2. The [CoIII(L1E)2](BF4) building block has been suc-
cessfully connected to a second such complex by coordina-
tion of silver(I) to a “spare” pyrazine nitrogen atom on each
complex, resulting in [{CoIII(L1E)2}2AgI](BF4)(NO3)2. All five
0.01
M AgNO3/Ag, respectively, consistent with the methyl-
ene-linked ligand being better able to stabilise the higher
oxidation state of cobalt than the ethylene-linked ligand.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2009)
ion bridging of “spare” donor atoms “out the back” of the
tetrametallic [2ϫ2] grid. In parallel with developing the
“all-pyrazine” ligand analogues of H2L2M and H2L2E for
this purpose, a task that involves significant synthetic chal-
lenges,[7] a test system has been prepared, which is the sub-
ject of this paper.
Introduction
The self-assembly of molecular grids, especially tetra-
metallic [2ϫ2] grids, has been studied extensively over the
past decade.[1] Such complexes have interesting and poten-
tially useful magnetic and electronic properties. For exam-
ple, the Lehn group has described the first tetrametallic
spin-crossover complex[2] and has also reported the highest
number of well-resolved and reversible one-electron re-
ductions for a molecular compound;[3] in both cases the
complexes concerned were tetrametallic [2ϫ2] grids. Such
complexes are of great interest not only for reasons of fun-
damental science but also because of their potential devel-
opment for use in future high-density storage devices.[1–3]
The bis(terdentate) pyrazine-based diamide ligands
H2L2M and H2L2E (Figure 1) have been used to produce
tetrametallic [2ϫ2] grids when complexed with octahedral
metal(II) ions in a 1:1 stoichiometry.[5] In order to access
larger arrays of metal ions, rather than constructing ever
larger ligands (i.e. with more binding pockets per ligand
strand), our approach is to instead connect the [2ϫ2] grids
together in a manner similar to that proposed and demon-
strated by Lehn and co-workers,[6] i.e. by subsequent metal-
Figure 1. Pyrazine diamide ligands H2L2M and H2L2E used, by us
and others, in previous work.[4,5] Dotted ellipses indicate the por-
tion of these ligands that corresponds to HL1M and HL1E (top
and bottom, respectively), the ligands used in this work (see also
Scheme 1).
[a] Department of Chemistry and MacDiarmid Institute for Ad-
vanced Materials and Nanotechnology, University of Otago,
Dunedin, New Zealand
HL1M and HL1E, which are very similar to H2L2M and
H2L2E but with only one alkylpyridine arm, have been pre-
pared (Figure 1, Scheme 1). On binding two of these
terdentate ligands to an octahedral metal ion, one of the
Fax: +64-3-4797906
E-mail: sbrooker@chemistry.otago.ac.nz
Supporting information for this article is available on the
WWW under http://www.eurjic.org or from the author.
1162
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 1162–1171

Co and Ag Complexes of Terdentate Pyrazine-Based Amide Ligands
pyrazine nitrogen atoms out the back of each ligand will
be spare. Subsequent coordination of these spare nitrogen
donors in the “building block” complex to another metal
ion will allow us to test whether these complexes can be
linked together. Whilst cobalt(III) is not our usual first row
metal ion of choice as it is diamagnetic, in this case it was
chosen for the building block complexes as the +3 charge
should challenge the ability of the spare pyrazine nitrogen
atom with regard to binding to the linker metal ion (the
[2ϫ2] grid complexes of H2L2M and H2L2E prepared to date
feature +2 oxidation states; so the all-pyrazine analogues of
these complexes should be able to more readily bind to the
linker-metal ion). Silver(I) was chosen as the linker-metal
ion as it is labile and frequently forms approximately linear
bonds. For completeness, the silver(I) coordination chemis-
try of these two ligands was also probed. In this paper, we
report the results of these investigations, including the step-
wise synthesis of the two ligands, the synthesis and struc-
tures of five cobalt and silver complexes, and a careful and
illuminating electrochemical study of the cobalt(III) com-
plexes.
Figure 2. SciFinder Scholar search fragment (X = C or N, Y =
one- or two-carbon atom linker); “hits” included complexes of the
ligands, HL3, H2L4, H2L5, H2L6, HL7, HL7ox and HL8 (shown
here), as well as complexes of the ligands shown in Figure 1.
of ligands such as those shown in Figure 2, is most relevant
to this work.[11–14] No silver complexes were found.
The HL7 and HL8 ligands (Figure 2) are the pyridine-
based analogues of our pyrazine-based ligands HL1M and
HL1E (Figure 1, Scheme 1). Nonoyama and Yamasaki re-
ported the synthesis, from cobalt(II) reagents in air, of
[CoIII(L7)2]+, [CoIII(L8)2]+ and CoII(HL8)2X2. They found
that a base, NaOH, had to be used to deprotonate HL8,
whereas HL7 spontaneously deprotonated in the presence
of the metal ion.[11] Mascharak and co-workers, in the
course of studies aimed at modelling the activation of bleo-
mycins (glycopeptide antibiotics that cleave DNA), sub-
Scheme 1. Synthesis of the ligands used in this work. (i) SOCl2/
MeOH, (ii) K2CO3/CH2Cl2, (iii) 2 equiv. 2-(2-aminoethyl)pyridine,
(iv) 1 equiv. 2-(2-aminoethyl)pyridine, (v) 1 equiv. 2-(aminomethyl)-
pyridine.
A SciFinder Scholar[8] substructure search (search frag- sequently found that, whereas the ligand (L7)– is stable in
ment shown in Figure 2 with X = N) revealed that, with the air, when ligated to FeIII or CoIII centers it is oxidised by
exception of the handful of complexes of H2L2M and H2L2E air to (L7ox)– (Figure 2).[14] They proposed that the likely
that we and others have reported,[4,5] only three papers re- driving force for the formation of the oxidised ligand is that
port metal complexes featuring a pyrazine-amide-linker- conversion of the methylene group to a carbonyl group is
pyridine motif. All three are communications, published in energetically favourable due to the resulting conjugation
2007 or 2008, and appeared after our own research program with the aromatic ring. Interestingly, they were unable to
was well underway. Firstly, Hong and co-workers published oxidise the ligand by adding ferrocinium tetrafluoroborate
two communications in 2007 describing the preparation and to the complexation reaction. However, simply stirring the
use of [FeIII(L3)(CN)3]– as building block (Figure 2) for reaction solution in air gave [CoIII(L7ox)2]ClO4. The deter-
FeIII–CN–MnIII ferrimagnetic chains, one of which exhibits mination of the single-crystal structure, but no electrochem-
a metamagnetic transition.[9] Then, in 2008, Hubin and co- ical study, was reported.[14] It should be noted that the
workers[10] communicated the structure of a trimetallic imide, HL7ox, has also been prepared by other routes, for
mixed-valent complex of HL1M (Figure 1), [{CoIII(L1M)2}2- example by copper(II)-catalysed hydrolysis of 1,3,5-tris(2-
{CoII(H2O)2Cl2}]Cl2. They did not isolate the monomeric pyridyl)triazine followed by EDTA extraction (to remove
species [CoIII(L1M)2]X.
the copper ions)[15] and by deliberate organic synthesis from
In contrast, the analogous pyridine-amide-linker-pyridine pyridine-2-carboxylic acid, by reaction with PCl5 followed
motif has been far more extensively studied. A SciFinder by liquid NH3.[16] Simultaneous with the work of Masch-
Scholar[8] substructure search for the same fragment but arak and co-workers, Kajiwara and co-workers reported the
with X = C (Figure 2) revealed over 100 papers concerning preparation of [CoIII(L7ox)2](ClO4) from preformed HL7ox
such complexes. Of these, the handful of cobalt complexes, and that it undergoes two reversible reductions in MeCN,
Eur. J. Inorg. Chem. 2009, 1162–1171
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1163

R. M. Hellyer, D. S. Larsen, S. Brooker
FULL PAPER
at –0.28 and –1.66 V vs. SSCE (ca. –0.22 and –1.60 V vs. synthesis was repeated and the solid material isolated by
0.01  AgNO3/Ag[17]), processes that they assumed to be overnight crystallisation within 12 h. Electrochemical ex-
metal-centred.[13]
periments on [CoIII(L1M)2](BF4) showed that this complex
was stable in the solid state in air for at least two weeks.
Results and Discussion
Ligand Synthesis
N-(2-Pyridylmethyl)pyrazine-2-carboxamide (HL1M) and
N-[2-(2-pyridyl)ethyl]pyrazine-2-carboxamide (HL1E) were
prepared in two or three steps from commercially available
pyrazine-2-carboxylic acid (1) as shown in Scheme 1. Con-
version of 1 to the methyl ester 2·HCl was achieved by
esterification with methanol and thionyl chloride, and the
free base 2 obtained from that by treatment with potassium
carbonate.
The addition of either 2 equiv. of 2-(2-aminoethyl)pyr-
idine to 2·HCl, or of 1 equiv. of 2-(2-aminoethyl)pyridine
to 2 in methanol resulted in the formation of the desired
new ligand HL1E. Silica gel column chromatography al-
lowed the purification of the ligand, affording HL1E in
quantitative or 39% yield, respectively. Surprisingly, the
synthesis of HL1E from the free base 2 resulted in a lower
yield: not all of the ester reacted but no amine was left. A
repeat of this experiment with a slight excess of amine
should improve the yield considerably. However, as the use
of 2·HCl worked well this was not pursued.
Figure 3. In the case of the methylene-linked ligand, ligand oxi-
dation appears to lead to the formation of [CoIII(L1M)-
(L1Mox)](BF4) (right) and [CoIII(L1Mox)2](BF4) (left).
Single-crystal X-ray analyses (Figures 4 and 5) confirmed
that in both cases the cobalt(II) ion had been air-oxidised
to cobalt(III). Two mono-deprotonated (L1E)– or (L1M)– li-
gands and one BF4 anion are present per cobalt centre, im-
plying a +III oxidation state. Consistent with this, the Co–
N bond lengths are appropriate for low-spin cobalt(III)
{1.91–1.99 Å for [CoIII(L1E)2](BF4)·2CH3CN and 1.873–
1.946 Å for [CoIII(L1M)2](BF4)·¼H2O}.
The methylene analogue HL1M was prepared from 2
and 1 equiv. of 2-(2-aminoethyl)pyridine. After column
chromatography, the product was further purified by vap-
our diffusion of diethyl ether into a dichloromethane solu-
tion to give colourless rod-like crystals of HL1M in 75%
yield.
Mononuclear “Building Block” Cobalt(III) Complexes
Both cobalt complexes were prepared in air in acetoni-
trile by treating 2 equiv. of the ligand with 2 equiv. of trieth-
ylamine before adding 1 equiv. of cobalt(II) tetrafluorobo-
rate hexahydrate. Slow diffusion of diethyl ether vapour into
the acetonitrile solutions gave the products as deep-red sin-
gle crystals. Decanting of the solvent and drying of the resi-
due in vacuo gave [CoIII(L1E)2](BF4)·½H2O and [CoIII
-
(L1M)2](BF4)·¼H2O in 83% and 32% yields, respectively.
Consistent with air oxidation to diamagnetic CoIII, sharp
1H and 13C NMR spectra were obtained and assigned for
each complex.
Figure 4. Perspective view of the cation of [CoIII(L1M)2](BF4)·
¼H2O. Hydrogen atoms and anions omitted for clarity.
Mass spectrometric data obtained on an “aged” sample
Each of the terdentate ligands is bound in a mer configu-
of [CoIII(L1M)2](BF4)·¼H2O, that had been stirred in solu- ration through a pyridine, a deprotonated amide and a pyr-
tion for 3 d in air before being crystallised, revealed signals azine nitrogen atom, forming an N6 octahedral geometry
consistent with the presence of both [CoIII(L1M)(L1Mox)]+ around the cobalt(III) ion. The two outward pointing
and [CoIII(L1Mox)2]+ species (Figure 3). There is precedent “spare” pyrazine nitrogen donors per complex are approxi-
for oxidation of the methylene linker, Mascharak and co- mately orthogonal to one another.
workers having reported that this occurred for the cobalt
Within these complexes, each of the methylene-linked li-
complex of the analogous pyridyl ligand (L7)–, forming gands was found to be very flat [pyridine and pyrazine rings
(L7ox)– (Figure 2).[14] In order to isolate a pure sample of almost coplanar N(1)/N(3) 4.6(2)° and N(21)/N(23)
the non-oxidised ligand product, [CoIII(L1M)2](BF4), the 0.6(1)°], whereas each of the ethylene-linked ligands was
1164
www.eurjic.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 1162–1171

Co and Ag Complexes of Terdentate Pyrazine-Based Amide Ligands
gand and a tetrafluoroborate counterion. The silver(I)
atoms are bound to a pyrazine nitrogen atom of one ligand
and a pyridine nitrogen atom (and weakly to an amide oxy-
gen atom) of another ligand. In the case of the ethylene-
linked ligand a dimeric structure results (second half of the
dimer generated by a centre of inversion), whereas for the
methylene-linked ligand a polymer results (Figures 6 and
7). The polymer consists of two intertwined helical strands
running parallel to one another and interacting through
Ag···Ag [3.3237(5) Å] and π–π interactions [centroid···
centroid 3.704 Å; mean plane···centroid and offset angles:
3.441(1) Å and 21.7°; and 3.553(1) Å and 16.4°; mean
planes intersect 11.3(2)°].
Considering only the nitrogen donor atoms, the silver(I)
atoms in {[AgI(HL1M)](BF4)2}ϱ have an almost ideal linear
geometry [N4–Ag1–N3B 176.4(1)°], whereas in [AgI2-
(HL1LE)2](BF4)2 there is a significant distortion away from
linear [N4–Ag1–N3A 158.6(1)°]. Consistent with this, the
weak Ag–O bonds are shorter in the dimer than in the poly-
mer [2.542(2) vs. 2.615(2) Å] and the Ag–N bonds slightly
longer [2.179(2) and 2.167(2) Å vs. 2.153(3) and 2.158(3) Å].
Figure 5. Perspective view of the cation of [CoIII(L1E)2](BF4)·
2CH3CN. Hydrogen atoms and anions omitted for clarity.
found to be far more twisted [N(1)/N(3) 42.0(1)° and N(21)/
N(23) 40.5(1)°]. The twisting of the latter ligands is facili-
tated by the conformationally more flexible ethylene linker;
the methylene linker provides less flexibility.
Silver(I)-Connected Cobalt(III) Building Blocks
Initially, the reaction of the inert cobalt(III) building
block, [CoIII(L1E)2](BF4)·½H2O with silver(I) nitrate in ace-
tonitrile was carried out in a 1:1 ratio in an attempt to form
Dimeric and Polymeric Silver(I) Complexes
The 1:1 reaction of silver(I) tetrafluoroborate with the a polymeric array of silver(I)-connected (through the
appropriate ligand in acetonitrile in the dark followed by “spare” pyrazine nitrogen donors) cobalt(III) complexes.
the slow diffusion of diethyl ether vapour into the acetoni- Diethyl ether vapour diffusion into the resulting brown-red
trile solutions gave [AgI2(HL1E)2](BF4)2 and {[AgI(HL1M)]- solution gave brown-red block-like single crystals. Elemen-
(BF4)}ϱ as colourless block-shaped crystals in 68% and tal analysis, and the subsequent X-ray crystal structure de-
50% yields, respectively. When triethylamine was used in termination (Figure 8) were consistent with this being a 2:1
attempts to deprotonate the amide group in the ligand, oily, CoIII/AgI complex, {[CoIII(L1E)2]2[AgI(H2O)]}(BF4)(NO3)2
unidentified products were obtained.
(no H2O was present in the dried microanalysis sample),
Single-crystal X-ray analyses (Figures 6 and 7) of these despite the 1:1 reaction stoichiometry. The yield, on the ba-
two silver complexes confirmed that the amide nitrogen sis of the limiting reagent (the cobalt building block com-
atoms of the HL1E and HL1M ligands are not deprotonated. plex) was 94%. When the correct, 2:1 Co/Ag, reaction stoi-
Rather, they remain neutral, and a tetrafluoroborate anion chiometry was employed the product obtained contained a
balances the charge of the silver(I) ion. In both cases, the 2:1 BF4/NO3 ratio, consistent with the ratio present in the
asymmetric unit comprises one silver ion, one neutral li- reaction, not a 1:2 ratio as obtained from the 1:1 reaction.
Figure 6. Perspective view of the cation of [AgI2(HL1E)2](BF4)2. Hydrogen atoms, other than those on the amide nitrogen atoms, and
anions omitted for clarity. A: –x, –y, –z.
Eur. J. Inorg. Chem. 2009, 1162–1171
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1165

R. M. Hellyer, D. S. Larsen, S. Brooker
FULL PAPER
Figure 7. Perspective view of the cation of {[AgI(HL1M)](BF4)}ϱ. Hydrogen atoms, other than those on the amide nitrogen atoms, and
anions omitted for clarity. Two intertwined helical strands run left-right. A: 1 – x, –y, 2 – z; B: 1 – x, –0.5 + y, 1.5 – x.
No single crystals were obtained from the 2:1 reaction. A their “spare” pyrazine donor atoms [Figure 8; Ag1–N24
product with a 1:1 Co/Ag ratio was not obtained, even 2.209(3) Å, N24–Ag1–N24A 155.4(1)°] forming the 2:1 di-
when an excess (10ϫ) of silver(I) was used.
mer {[CoIII(L1E)2]2[AgI(H2O)]}(BF4)(NO3)2. The silver(I)
atom is also weakly bound to the oxygen atom of a water
molecule of crystallisation [Ag1···O50 2.414(4) Å] causing
the N–Ag–N angle [155.4(1)°] to be distorted somewhat
away from linear. The silver(I) atom also has a very weak
interaction with the nitrate anion [Ag–O 2.780(3) Å]. As ex-
pected, the Co–N distances are within experimental error
of those observed for the mono-cobalt(III) complex
[CoIII(L1E)2](BF4)·2CH3CN, despite the intra-ligand pyr-
azine ring to pyridine ring angles [36.0(1)° and 48.2(1)°] be-
ing more widely spread than for [CoIII(L1E)2](BF4)·
2CH3CN [40.5(1)° and 42.0(1)°].
Electrochemical Studies
Electrochemical analysis of [CoIII(L1E)2](BF4)·½H2O
over a wide range (–2.0 to +2.0 V) showed the presence of
a single chemically reversible reductive process (Figure 9,
top). Close inspection of this process (Figure S3) showed it
to have an E1/2 of –0.48 V (∆E = 75, 80, 90 and 95 mV, for
scan rates of 50, 100, 200 and 400 mVs–1, respectively). A
controlled potential coulometry experiment at –0.59 V re-
sulted in a change in colour from dark-red to pale-tan and
the transfer of 0.96 electron-equiv., confirming that this is
a one-electron reduction process, presumably primarily
metal-centred (CoIII Ǟ CoII). Afterwards, the cyclic vol-
tammogram was practically unchanged, consistent with
Figure 8. Perspective view of the cation of {[CoIII(L1E)2]2[AgI-
(H2O)]}(BF4)(NO3)2. Hydrogen atoms and anions omitted for clar-
ity. A: 1 – x, y, 0.5 – z.
The single-crystal X-ray structure confirmed that, as de- chemical reversibility.
sired, the silver(I) ion (located on a C2 axis) has bridged two
In contrast, an initial electrochemical analysis of the
crystallographically identical cobalt(III) sub-units by one of methylene-linked analogue, carried out on a solution-aged
1166
www.eurjic.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 1162–1171

Co and Ag Complexes of Terdentate Pyrazine-Based Amide Ligands
This, along with the ∆E values and similarities to the
[CoIII(L1E)2](BF4)·½H2O complex (including the same col-
our change from dark-red to tan during the controlled po-
tential electrolysis) indicates that this too is a one-electron
process.
Tests of the hypothesis that the presence of three electro-
chemical processes in the experiment carried out on the
“aged” sample was due to the presence of oxidised versions
of the monomeric complex (Figure 3) were carried out.
Firstly, air was bubbled through an acetonitrile solution of
fresh complex for 2 h. The cyclic voltammogram was un-
changed, indicating that such oxidation may occur quite
slowly. Therefore, in a second test, a “fresh” sample was
stirred in acetonitrile solution in air for 2 d. The resulting
cyclic voltammogram showed a large peak at –0.41 V along
with shoulders at –0.71 and –0.54 V. These peaks occur in
similar positions to those for the “aged” sample, so it ap-
pears that the proposed oxidation does require considerable
time; this is also consistent with our ability to grow crystals
of unoxidised ligand complex overnight (Figure 4).
As anticipated, the reduction potentials confirm that the
methylene-linked ligand (–0.71 V) is better able to stabilise
cobalt(III) over cobalt(II) than the ethylene-linked ligand
can (–0.48 V). Interestingly, both of these amide ligands are
better at this than the rather rigid imide ligand, (L7ox)– (see
Introduction, E1/2 approx. –0.22 V).[13]
Conclusions
Two carefully designed pyrazine-amide-linker-pyridine li-
gands, HL1M and HL1E, that contain a spare nitrogen atom
out the back, have been prepared. The silver(I) complexes
of these ligands were found to be dimeric and polymeric,
[AgI2(HL1E)2](BF4)2 and {[AgI(HL1M)](BF4)}ϱ, respec-
tively. One-off attempts to deprotonate and coordinate the
monoanionic ligands to silver(I) were unsuccessful.
Figure 9. Cyclic voltammograms of [CoIII(L1E)2](BF4)·½H2O (top),
fresh [CoIII(L1M)2](BF4)·¼H2O (middle), and aged [CoIII
-
(L1M)2](BF4) (bottom) that is believed to contain [CoIII(L1M)-
(L1Mox)](BF4) and [CoIII(L1Mox)2](BF4). 200 mVs–1, vs. 0.01 
AgNO3/Ag.
sample, showed the presence of three reductive processes
Two cobalt(III) building blocks, [CoIII(L1M)2](BF4)·
between –2.0 and +2.0 V. Close inspection of these pro- ¼H2O and [CoIII(L1E)2](BF4)·½H2O, were prepared and
cesses (Figure 9, bottom) over a smaller potential range fully characterised. In solution, the methylene-linked ligand
(–0.9 to –0.1 V) showed they had E1/2 values of –0.71, –0.54 in the former complex appears to be sensitive to oxidation,
and –0.41 V at scan rates of 50, 100, 200 and 400 mVs–1 generating complexes of the more conjugated ligand,
(∆E values at 50 mVs–1 were 60, 60 and 50 mV respectively, [CoIII(L1M)(L1Mox)](BF4) and [CoIII(L1Mox)2](BF4). Cyclic
and at 400 mVs–1 were 70, 70 and 60 mV). These three pro- voltammetry was found to be a sensitive tool for monitor-
cesses are believed to be due to the presence of three dif- ing this proposed ligand oxidation. The electrochemical
ferent compounds in solution, [CoIII(L1M)2](BF4), [CoIII
(L1M)(L1Mox)](BF4), and [CoIII(L1Mox)2](BF4), which arise the ethylene-linked ligand is (a) resistant to such ligand oxi-
due to partial ligand oxidation (Figure 3). dation and (b) more readily reduced than the complex of
-
studies also confirmed the expectation that the complex of
Only one chemically reversible process is observed (Fig- the methylene-linked ligand.
ure 9, middle) when a fresh sample of [CoIII(L1M)2](BF4) is
Silver(I) was successfully employed to connect two
studied between –2 and +2 V. Close inspection of this pro- [CoIII(L1E)2](BF4)·½H2O building blocks, by coordination
cess (Figure S4) over a smaller potential range (–1.4 to 0 V) to a spare nitrogen atom from each building block, despite
showed that it has an E1/2 of –0.71 V (∆E = 65, 70, 75, the cobalt(III) withdrawing electron density from the pyr-
80 mV) at scan rates of 50, 100, 200 and 400 mVs–1. A azine ring. With this test result in hand, our focus is now
small bump on the re-oxidation side of the wave indicated firmly on targeting the self-assembly of extended two-di-
the presence of a trace of ligand-oxidised product in the mensional paramagnetic polymetallic systems. These are
mixture. Controlled potential coulometry conducted at likely to be easier to link as (a) the metal ions will be in +2,
–0.81 V indicated that 0.84 electron-equiv. were transferred. not +3, oxidation states and (b) there are eight spare pyr-
Eur. J. Inorg. Chem. 2009, 1162–1171
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1167

R. M. Hellyer, D. S. Larsen, S. Brooker
FULL PAPER
idine (208 µL, 1.74 mmol) in methanol (5 mL) was refluxed for 8 h
and stirred at room temperature for 14 h, then concentrated to dry-
ness in vacuo to give HL1E as an off-white crystalline solid. The
product was purified by column chromatography (silica gel, 5%
MeOH/CH2Cl2, Rf = 0.25). The resulting yellow oil rapidly crystal-
lised on cooling from 30 °C (rotary evaporator temperature) to
20 °C to give a pale lemon-coloured solid in quantitative yield.
Method B: A solution of methyl pyrazine-2-carboxylate (2) (1.00 g,
7.25 mmol) and 2-(2-aminoethyl)pyridine (510 µL, 7.25 mmol) in
methanol (40 mL) was refluxed for 4 d, then concentrated to dry-
ness in vacuo to give HL1E as an impure tan solid. The product
was purified by column chromatography (silica gel, 5% MeOH/
CH2Cl2, Rf = 0.25) to afford HL1E as a white solid in 645 mg
azine nitrogen atoms per [2ϫ2] grid complex, and they will
be better aligned with regard to packing of the resulting
super complex in space, so many complementary links
should be possible. Only a handful of such systems have
been studied, all by Lehn and co-workers, which is an indi-
cation of the challenges inherent in such research. However,
the results of this pilot study pave the way for our next
generation of experiments, on the basis of previous work
from our research group, but which nevertheless involve
overcoming significant synthetic challenges.[7]
1
(2.83 mmol, 39%) yield. H NMR (300 MHz, CDCl3): δ = 9.35 (d,
Experimental Section
J = 1.5 Hz, 1 H, 5-H), 8.67 (d, J = 2.4 Hz, 1 H, 3-H), 8.54 (d, J =
7.5 Hz, 1 H, 16-H), 8.47 (dd, J2,5 = 1.5, J = 2.4 Hz, 1 H, 2-H), 8.45
(br. s, 1 H, NH), 7.58 (dt, J = 7.5, J = 1.5 Hz, 1 H, 15-H), 7.15
(m, 2 H, 13-H and 14-H), 3.88 (q, J = 6.6 Hz, 2 H, 10-H), 3.09 (t,
J = 6.3 Hz, 2 H, 11-H) ppm. 13C NMR (300 MHz, CDCl3): δ =
163.0 (C-7), 159.0 (C-12), 149.3 (C-16), 147.0 (C-3), 144.7 (C-6),
144.3 (C-5), 142.6 (C-2), 136.7 (C-15), 123.4 (C-13), 121.6 (C-14),
38.7 (C-10) and 37.1 (C-11) ppm. C12H12N4O (228.25): calcd. C
63.15, H 5.30, N 24.55; found C 63.23, H 5.52, N 24.54.
General: Acetonitrile used was HPLC grade. Pyrazine-2-carboxylic
acid, cobalt(II) tetrafluoroborate hexahydrate, silver(I) tetrafluo-
roborate and silver(I) nitrate were used as supplied. The 10% (v/v)
triethylamine stock solution was prepared from 9 mL of aceto-
nitrile (HPLC grade, dried with calcium hydride) and 1 mL of tri-
ethylamine (distilled from calcium hydride). 1H and 13C NMR
spectra were recorded at 25 °C with a Varian 300 or 500 MHz In-
ova spectrometer. NMR assignments were determined by a combi-
nation of results from HSQC, gHMBC, gCOSY, 1H and 13C NMR
spectra. ESI mass spectra were recorded in acetonitrile with a
Bruker MicrOTOFQ spectrometer. All electrochemical analyses
were conducted under argon with a 1 mmolL–1 solution of complex
in dry acetonitrile with 0.1 molL–1 Bu4N(PF6) (recrystallised from
water and dried in vacuo) electrolyte. All potentials are quoted vs.
0.01 molL–1 AgNO3/Ag. All controlled potential coulometry ex-
periments were carried out until the current reached 1% of the
initial value. The Fc/Fc+ couple occurs at 0.07 V (75 mV) in this
system. Single crystal X-ray data were collected with a Bruker
SMART CCD area detector diffractometer (λ = 0.71073 Å). The
structures were solved by direct methods using SHELXS-97[18] and
refined against F2 using full-matrix least-squares techniques with
SHELXL-97.[18] A summary of the crystallographic data and re-
finement parameters is given in Table 1. CCDC-702959, -702960,
-702961, -702962, -702963 contain the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
HL1M: A solution of methyl pyrazine-2-carboxylate (2) (2.00 g,
14.5 mmol) and 2-(aminomethyl)pyridine (1.50 mL, 14.5 mmol) in
methanol (80 mL) was refluxed for 5 d, before concentrating to
dryness in vacuo, to yield HL1M as a tan solid. The product was
purified by column chromatography (silica gel, 5% MeOH/CH2Cl2,
Rf = 0.25) to afford HL1M as an off-white solid. Diffusion of diethyl
ether into a solution of the product in dichloromethane yielded
2.25 g (10.5 mmol, 73%) of white rod-like crystals of HL1M 1H
.
NMR (500 MHz, CDCl3): δ = 9.42 (s, 1 H, 5-H), 8.86 (br. s, 1 H,
9-H), 8.75 (d, J = 2.5 Hz, 1 H, 3-H), 8.610 (td, J = 5, J = 1.2 Hz,
1 H, 15-H), 8.57 (d, J = 2.5 Hz, 2 H, 2-H), 7.70 (dt, J = 8.0, J =
1.2 Hz, 1 H, 13-H), 7.36 (d, J = 7.5 Hz, 1 H, 12-H), 7.24 (t, J =
5 Hz, 1 H, 14-H) and 4.81 (d, J = 7.5 Hz, 1 H, 10-H) ppm. 13C
NMR (500 MHz, CDCl3): δ = 166.2 (C-7), 156.1 (C-6), 149.1 (C-
15), 147.3 (C-3), 144.6 (C-11), 144.5 (C-5), 142.8 (C-2), 137.2 (C-
13), 122.7 (C-14), 122.3 (C-12) and 44.4 (C-10) ppm. C11H10N4O
(214.23): calcd. C 61.67, H 4.71, N 26.15; found C 61.71, H 4.58,
N 26.37.
[CoIII(L1E)2](BF4)·½H2O: Triethylamine (49 µL, 351 µmol) was
added to a solution of HL1E (80.0 mg, 351 µmol) in acetonitrile
(7 mL) and stirred for 30 min, before cobalt(II) tetrafluoroborate
hexahydrate (60.0 mg, 176 µmol) was added. The resulting orange-
red solution was stirred overnight. The solution was then reduced
in volume in vacuo to ca. 5 mL. Slow diffusion of diethyl ether
vapour into this solution resulted in deep-red, very large (up to
3 mm across), plate-shaped single crystals of [CoIII(L1E)2](BF4)·
2CH3CN. The colourless solution was decanted, and the crystals
dried in vacuo to give 88.8 mg (148 µmol, 83%) of [CoIII(L1E)2]-
Synthesis
2·HCl: Thionyl chloride (880 µL, 12.1 mmol) was added dropwise
to a suspension of pyrazine-2-carboxylic acid (2.00 g, 16.1 mmol) in
methanol (200 mL). The white solid rapidly dissolved and evolved a
gas, leaving a clear colourless solution. After stirring for 2.5 d, the
solution was concentrated to dryness in vacuo to give 2·HCl as a
white solid in quantitative yield. C6H6N2O2·HCl (174.59): calcd. C
41.28, H 4.04, N 16.05; found C 41.19, H 3.87, N 15.76.
2: Methyl pyrazine-2-carboxylate hydrochloride (2·HCl) (10.0 g,
57.3 mmol) was suspended in dichloromethane (200 mL). Potas-
sium carbonate (10 g) was added and the mixture stirred for 1 h.
The solution was decanted, filtered (glass sinter), and the solvents
were evaporated to dryness in vacuo to yield 7.02 g (50.9 mmol,
1
(BF4)·½H2O. H NMR (500 MHz, CD3CN): δ = 9.08 (s, 1 H, 5-
H), 8.53 (d, J = 3 Hz, 1 H, 3-H), 8.26 (d, J = 6 Hz, 1 H, 16-H),
8.01 (d, J = 3 Hz, 1 H, 2-H), 7.818 (t, J = 8 Hz, 1 H, 14-H), 7.36
(d, J = 7.5 Hz, 1 H, 13-H), 7.24 (t, J = 7 Hz, 1 H, 15-H), 4.67 (td,
J = 4, J = 14 Hz, 1 H, 10-H), 3.65 (dt, J = 14, J = 2 Hz, 1 H, 11-
H), 3.21 (m, 1 H, 10-H), 2.37 (dt, J = 18, J = 4 Hz, 1 H, 11-H)
ppm. 13C NMR (500 MHz, CD3CN): δ = 171.2 (C-7), 165.6 (C-6),
154.7 (C-16), 150.6 (C-12), 150.2 (C-3), 147.6 (C-5), 146.5 (C-2),
142.0 (C-14), 129.1 (C-13), 125.4 (C-15), 41.6 (C-10), 37.8 (C-11)
ppm. MS (ESI): m/z = 513 [CoIII(L1E)2]+. C24H22BCoF4N8O2·
½H2O (609.24): calcd. C 47.32, H 3.81, N 18.39; found C 47.40, H
3.79, N 18.53.
1
89%) of 2 as a white solid. H NMR (300 MHz, CDCl3): δ = 9.33
(d, J = 1.5 Hz, 1 H, 5-H), 8.78 (d, J = 2.7 Hz, 1 H, 2-H), 8.73 (dd,
J3,5 = 1.5, J3,2 = 2.4 Hz, 1 H, 3-H), 4.05 (s, 3 H, 9-H) ppm.
C6H6N2O2 (138.13): calcd. C 52.17, H 4.38, N 20.28; found C
52.42, H 4.44, N 20.36.
HL1E. Method A: A solution of methyl pyrazine-2-carboxylate hy-
drochloride (120 mg, 870 µmol) and excess 2-(2-aminoethyl)pyr-
1168
www.eurjic.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 1162–1171

Co and Ag Complexes of Terdentate Pyrazine-Based Amide Ligands
Table 1. Details of the crystal-structure determination for the complexes [CoIII(L1E)2](BF4)·2CH3CN, [CoIII(L1M)2](BF4)·¼H2O,
[AgI(HL1LE)]2(BF4)2, {[AgI(HL1M)]BF4}ϱ and [{CoIII(L1E)2}2AgI(OH2)](BF4)(NO3)2.
[CoIII(L1E)2](BF4)·2CH3CN
[CoIII(L1M)2]·(BF4)·¼H2O
[AgI2(HL1E)2]·(BF4)2
Empirical formula
Mr
C28H28BCoF4N10O2
682.34
C22H18.50BCoF4N8O2.25
576.69
C24H24Ag2B2F8N8O2
845.87
Crystal system
Space group
triclinic
P1
monoclinic
P21/c
triclinic
P1
¯
¯
a [Å]
b [Å]
c [Å]
α [°]
8.8824(5)
10.1800(6)
15.9646(10)
93.602(3)
91.676(3)
94.560(3)
1435.30(15)
2
9.9828(13)
8.4576(8)
27.226(3)
90
92.193(4)
90
2297.0(5)
4
1.668
8.0720(5)
8.9336(5)
10.3831(6)
102.802(3)
103.376(2)
90.646(3)
708.78(7)
1
β [°]
γ [°]
V [Å3]
Z
ρcalcd. [g/cm3]
1.579
0.672
1.982
1.475
µ [mm]–1
0.821
F(000)
700
1170
416
Crystal size [mm]
θ range for data collection [°]
Reflections collected
Independent reflections
R(int)
0.45ϫ0.30ϫ0.14
2.01–26.53
30408
5933
0.0280
0.45ϫ0.20ϫ0.15
1.50–26.52
29029
4762
0.0533
0.54ϫ0.14ϫ0.04
2.07–26.00
14517
2720
0.0338
Max./min. transmission
Data/restraints/parameters
Goof (F2)
R1 [IϾ2σ(I)]
wR2 (all data)
1.000/0.911
5933/16/439
1.042
0.0280
0.0701
0.887/0.709
4762/9/363
1.033
0.0476
0.1238
0.943/0.503
2720/1/212
1.058
0.0210
0.0526
{[AgI(HL1M)]·BF4}ϱ
[{CoIII(L1E)2}2AgI(OH2)](BF4)(NO3)2
Empirical formula
Mr
Crystal system
Space group
C11H10AgBF4N4O
408.91
monoclinic
P21/c
C48H46AgBCo2F4N18O11
1363.57
monoclinic
P2/c
a [Å]
b [Å]
c [Å]
α [°]
10.0079(6)
8.4784(5)
15.6539(10)
90
19.3281(14)
12.7845(9)
10.3147(7)
90
β [°]
99.002(3)
90
1311.89(14)
4
2.070
1.590
90.949(4)
90
2548.4(3)
2
1.777
1.121
γ [°]
V [Å3]
Z
ρcalcd. [g/cm3]
µ [mm]–1
F(000)
800
1380
Crystal size [mm]
θ range for data collection [°]
Reflections collected
Independent reflections
R(int)
0.30ϫ0.10ϫ0.05
2.74–26.40
18644
2686
0.0521
0.50ϫ0.10ϫ0.05
2.11–26.02
64154
5023
0.0611
Max./min. transmission
0.924/0.647
1.098
1.000/0.837
1.099
Goof (F2)
R1 [IϾ2σ(I)]
wR2 (all data)
0.0347
0.0877
0.0342
0.0698
[CoIII(L1M)2](BF4)·¼H2O: HL1M (40.0 mg, 187 µmol) and triethyl-
amine [187 µmol, 261 µL of 10% (v/v) stock solution in acetoni-
trile] were stirred in acetonitrile (10 mL) for 30 min. Cobalt(II) tet-
rafluoroborate hexahydrate (31.8 mg, 93.5 µmol) in acetonitrile
(1 mL) was added with stirring, resulting in a dark red solution,
which was stirred overnight. Diffusion of diethyl ether vapour into
the solution over 7 d gave a red solid in a colourless solution. The
colourless solution was decanted and discarded. The red solid was
redissolved in acetonitrile. Diffusion of diethyl ether vapour into
the solution gave red rod-like crystals surrounded by red powder
in a red solution. The solvent was decanted and the solids dried in
vacuo to give 34.8 mg (60.4 µmol, 32%) of [CoIII(L1M)2](BF4)·
¼H2O. 1H NMR (500 MHz, CD3NO2): δ = 9.05 (d, J = 1.0 Hz,
1 H, 5-H), 8.58 (d, J = 3.0 Hz, 1 H, 3-H), 8.25 (d, J = 2.5, J =
1.2 Hz, 1 H, 2-H), 8.18 (dd, J = 7.5, J = 1.2 Hz, 1 H, 15-H), 7.87
(dt, J = 8, J = 1.2 Hz, 1 H, 13-H), 7.67 (dd, J = 8, J = 1.5 Hz, 1
H, 12-H), 7.29 (dt, J = 8, J = 1.5 Hz, 1 H, 14-H), 5.93 (s, 1 H, 10-
H) and 5.90 (s, 1 H, 10-H) ppm. 13C NMR (125 MHz, CD3NO2):
δ = 52.2 (C-10), 125.3 (C-12), 127.0 (C-14), 141.8 (C-13), 146.9 (C-
5), 147.0 (C-2), 150.9 (C-3), 151.7 (C-15), 152.3 (C-6), 168.9 (C-7),
Eur. J. Inorg. Chem. 2009, 1162–1171
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1169

R. M. Hellyer, D. S. Larsen, S. Brooker
FULL PAPER
169.0 (C-11) ppm. MS (ESI): m/z = 485 [CoIII(L1M)2]+, 499 [CoIII
(OL1M)2]+, 513 [CoIII (OL1M)(OL1M)]+. C22H18BCoF4N8O2·
¼H2O (576.68): calcd. C 45.82, H 3.23, N 19.43; found C 45.51, H
3.20, N 19.43.
decanted, and the crystals were dried in vacuo to give
16.4 mg (12.0 µmol, 73%) of [{CoIII(L1E)2}2AgI](BF4)2(NO3).
C48H44AgB2Co2F8N16NO4O3 (1370.33): calcd. C 42.07, H 3.24, N
17.38; found C 41.94, H 3.69, N 17.28.
Supporting Information (see also the footnote on the first page of
this article): Further crystallographic details and diagrams, as well
as additional cyclic voltammetry plots (including scan-rate and full-
range studies).
[AgI2(HL1E)2](BF4)2:
Silver(I)
tetrafluoroborate
(69.2 mg,
355 µmol) in acetonitrile (2 mL) was added to HL1E (81.0 mg,
355 µmol) in acetonitrile (7 mL) resulting in a lemon-coloured solu-
tion, which was stirred at room temperature for 4 h. Diffusion of
diethyl ether vapour into the solution in the dark over 10 d gave
103 mg (79.1 µmol, 67%) of [AgI2(HL1E)2](BF4)2 as microanalyti-
cally pure, clear, colourless, rod-shaped crystals. 1H NMR
(300 MHz, CD3CN): δ = 10.17 (s, 1 H, 5-H), 8.95 (d, J = 2.5 Hz,
1 H, 3-H), 8.93 (d, J = 2.5 Hz, 1 H, 2-H), 8.76 (d, J = 7.5 Hz, 1
H, 16-H), 8.07 (t, J = 7.5 Hz, 1 H, 14-H), 7.69 (d, J = 7.5 Hz, 1
H, 13-H), 7.56 (t, J = 7.5 Hz, 1 H, 15-H), 3.88 (m, 2 H, 11-H),
3.40 (m, 1 H, 10-H) ppm. 13C NMR (125 MHz, CD3CN): δ = 165.5
(C-7), 160.4 (C-12), 153.5 (C-16), 149.8 (C-2), 147.4 (C-6), 146.8
(C-5), 146.6 (C-3), 141.6 (C-14), 126.9 (C-13), 124.8 (C-15), 41.8
(C-11), 41.5 (C-10) ppm. C12H11AgBF4N4O (421.92): calcd. C
34.16, H 2.63, N 13.28; found C 34.23, H 2.93, N 13.14.
Acknowledgments
This work was supported by grants from the Marsden Fund (Royal
Society of New Zealand) and the University of Otago. We thank
Assoc. Prof. A. J. McQuillan for helpful discussions.
[1] a) M. Ruben, J. Rojo, F. J. Romero-Salguero, L. H. Uppadine,
J.-M. Lehn, Angew. Chem. Int. Ed. 2004, 43, 3644; b) L. N.
Dawe, T. S. M. Abedin, L. K. Thompson, Dalton Trans. 2008,
1661.
{[AgI(HL1M)]BF4}ϱ: Silver(I) tetrafluoroborate (36.4 mg, 187 µmol)
in acetonitrile (1 mL) was added to HL1M (40.0 mg, 187 µmol) in
acetonitrile (5 mL) resulting in a colourless solution, which was
stirred in the dark for 30 min. Diffusion of diethyl ether vapour
into the solution in the dark over 3 d gave {[AgI(HL1M)](BF4)}ϱ
as microanalytically pure, clear, colourless, rod-shaped crystals in
[2] E. Breuning, M. Ruben, J.-M. Lehn, F. Renz, Y. Garcia, V.
Ksenofontov, P. Gütlich, E. Wegelius, K. Rissanen, Angew.
Chem. Int. Ed. 2000, 39, 2504.
[3] M. Ruben, E. Breuning, J. P. Gisselbrecht, J.-M. Lehn, Angew.
Chem. Int. Ed. 2000, 39, 4139.
[4] a) E. B. Fleischer, M. B. Lawson, Inorg. Chem. 1972, 11, 2772;
b) E. B. Fleischer, D. Jeter, R. Florian, Inorg. Chem. 1974, 13,
1042; c) I. Mallik, S. Mallik, Synlett 1996, 734; d) D. S. Cati,
H. Stoeckli-Evans, Acta Crystallogr., Sect. E 2004, 60, o210; e)
D. S. Cati, H. Stoeckli-Evans, Acta Crystallogr., Sect. E 2004,
60, m174; f) D. S. Cati, H. Stoeckli-Evans, Acta Crystallogr.,
Sect. E 2004, 60, m177.
[5] a) J. Hausmann, G. B. Jameson, S. Brooker, Chem. Commun.
2003, 2992; b) D. S. Cati, J. Ribas, J. Ribas-Arino, H. Stoeckli-
Evans, Inorg. Chem. 2004, 43, 1021; c) J. Hausmann, S.
Brooker, Chem. Commun. 2004, 1530; d) J. Klingele (nee Haus-
mann), J. F. Boas, J. R. Pilbrow, B. Moubaraki, K. S. Murray,
K. J. Berry, K. A. Hunter, G. B. Jameson, P. D. W. Boyd, S.
Brooker, Dalton Trans. 2007, 633.
[6] M. Ruben, U. Ziener, J.-M. Lehn, V. Ksenofontov, P. Guetlich,
G. B. M. Vaughan, Chem. Eur. J. 2005, 11, 94.
[7] M. G. Cowan, D. S. Larsen, S. Brooker, unpublished results.
[8] Scifinder Scholar 2007, Chemical Abstracts Service, American
Chemical Society, Columbus, Ohio, 2007.
[9] a) J. I. Kim, H. S. Yoo, E. K. Koh, H. C. Kim, C. S. Hong,
Inorg. Chem. 2007, 46, 8481; b) J. I. Kim, H. S. Yoo, E. K. Koh,
C. S. Hong, Inorg. Chem. 2007, 46, 10461.
[10] D. L. Cockriel, J. M. McClain, K. C. Patel, R. Ullom, T. R.
Hasley, S. J. Archibald, T. J. Hubin, Inorg. Chem. Commun.
2008, 11, 1.
1
38.2 mg (93.4 µmol, 50%) yield. H NMR (300 MHz, CD3CN): δ
= 9.26 (d, J2,5 = 1.5 Hz, 1 H, 5-H), 8.78 (d, J = 2.7 Hz, 1 H, 3-H),
8.75 (br. s, 1 H, 9-H), 8.62 (dd, J = 2.4, J2,5 = 1.5 Hz, 1 H, 2-H),
8.53 (m, 1 H, 12-H), 7.81 (dt, J = 7.8, J = 2.1 Hz, 1 H, 13-H), 7.44
(d, J = 7.8 Hz, 1 H, 15-H), 7.33 (m, 1 H, 14-H), 4.73 (d, J = 6.0 Hz,
2 H, 10-H) ppm. 13C NMR (125 MHz, CD3CN): δ = 45.8 (C-10),
118.3 (C-7), 123.3 (C-15), 123.9 (C-14), 138.8 (C-13), 144.3 (C-2),
144.9 (C-5), 145.6 (C-6), 148.8 (C-3), 150.8 (C-12), 158.5 (C-11),
164.5 (C-7) ppm. C11H10AgBF4N4O (408.90): calcd. C 32.31, H
2.47, N 13.70; found C 32.49, H 2.53, N 13.84.
[{CoIII(L1E)2}2AgI](BF4)(NO3)2:
Silver(I)
nitrate
(8.1 mg,
47.7 µmol) was added to a solution of [CoIII(L1E)2](BF4)·½H2O
(29.2 mg, 47.9 µmol) in acetonitrile (4 mL). The resulting dark-red
solution was stirred overnight. Slow diffusion of diethyl ether vap-
our into this solution resulted in deep-red plate-shaped single crys-
tals of [{CoIII(L1E)2}2AgI(OH2)](BF4)(NO3)2. The colourless solu-
tion was decanted and the crystals dried in vacuo to give 30.2 mg
(22.5 µmol, 94% based on the limiting reagent, the cobalt complex)
of [{CoIII(L1E)2}2AgI](BF4)(NO3)2. 1H NMR (300 MHz, CD3CN):
3
δ = 9.07 (d, J = 3.3 Hz, 1 H, 5-H), 8.53 (t, J = 3.3 Hz, 1 H, 2-H),
8.25 (m, 1 H, 16-H), 8.00 (t, J = 3.3 Hz, 1 H, 3-H), 7.81 (m, 1 H,
14-H), 7.34 (dd, J = 7.8, J = 1.8 Hz, 1 H, 13-H), 7.24 (m, 1 H, 15-
H), 4.65 (m, 1 H, 10-H), 3.65 (t, J = 14, J = 3.6 Hz, 1 H, 10-H),
3.19 (dd, J = 14.0, J = 2.4 Hz, 1 H, 11-H), 2.35 (tt, J = 14.0, J =
3.6 Hz, 1 H, 11-H) ppm. 13C NMR (125 MHz, CD3NO2): δ = 41.6
(C-10), 37.8 (C-11), 125.4 (C-15), 129.1 (C-13), 142.0 (C-14), 146.6
(C-6), 147.6 (C-5), 150.2 (C-3), 150.6 (C-2), 154.6 (C-16), 165.6 (C-
12), 171.4 (C-7) ppm. C48H44AgBCo2F4N18O10 (1345.53): calcd. C
42.85, H 3.30, N 18.74; found C 43.14, H 3.32, N 18.46.
[11] M. Nonoyama, K. Yamasaki, Inorg. Chim. Acta 1971, 5, 124.
[12] a) F. A. Chavez, C. V. Nguyen, M. M. Olmstead, P. K. Masch-
arak, Inorg. Chem. 1996, 35, 6282; b) T. Hirao, T. Moriuchi, T.
Ishikawa, K. Nishimura, S. Mikami, Y. Ohshiro, I. Ikeda, J.
Mol. Catal. A 1996, 113, 117; c) F. A. Chavez, J. M. Rowland,
M. M. Olmstead, P. K. Mascharak, J. Am. Chem. Soc. 1998,
120, 9015; d) P. T. Kaye, T. Nyokong, G. M. Watkins, K. W.
Wellington, ARKIVOC 2002, 9, 9; e) A. N. Dwyer, M. C. Gros-
sel, P. N. Horton, Supramol. Chem. 2004, 16, 405; f) N. W. Al-
cock, G. Clarkson, P. B. Glover, G. A. Lawrance, P. Moore, M.
Napitupulu, Dalton Trans. 2005, 518; g) M. Napitupulu, G. A.
Lawrance, G. J. Clarkson, P. Moore, Aust. J. Chem. 2006, 59,
796–804.
[{CoIII(L1E)2}2AgI](BF4)2(NO3): Silver(I) nitrate (16.4 µmol, from
0.5 mL of 32.8 mmolL–1 stock solution in acetonitrile) was added
to a solution of [CoIII(L1E)2](BF4)·½H2O (20.0 mg, 32.8 µmol) in
acetonitrile (5 mL). The resulting dark-red solution was stirred
overnight. Slow diffusion of diethyl ether vapour into this
solution resulted in a deep-red solid. The colourless solution was
[13] T. Kajiwara, R. Sensui, T. Noguchi, A. Kamiyama, T. Ito, In-
org. Chim. Acta 2002, 337, 299.
[14] J. M. Rowland, M. M. Olmstead, P. K. Mascharak, Inorg.
Chem. 2002, 41, 2754.
1170
www.eurjic.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 1162–1171

Co and Ag Complexes of Terdentate Pyrazine-Based Amide Ligands
[15] A. Kamiyama, T. Noguchi, T. Kajiwara, T. Ito, Inorg. Chem.
2002, 41, 507.
[16] H. Chowdhury, S. H. Rahaman, R. Ghosh, S. K. Sarkar, H.-
K. Fun, B. K. Ghosh, J. Mol. Struct. 2007, 826, 170.
[17] A. J. Bard, L. R. Faulkner, Electrochemical methods: fundamen-
tals and applications, 2nd ed., John Wiley & Sons, Inc., New
York, 2001.
[18] G. M. Sheldrick, Acta Crystallogr., Sect. A 2008, 64, 112.
Received: December 7, 2008
Published Online: February 11, 2009
Eur. J. Inorg. Chem. 2009, 1162–1171
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1171