Formation of ternary complexes of iron(III) cations in solution and gas phase

Article abstract of DOI:10.1071/CH13179

Formation of labile 1:1:1 ternary mononuclear complexes of iron(iii) cation with η3-terdentate meridional binders was studied using electrospray ionisation mass spectrometry (ESI-MS) titration and UV-Vis titration in solution phase. Low selectivities towards formation of ternary heteroleptic complexes in the solution phase vs. symmetric 2:1 complexes were obtained with combinations of dianionic 2,6-bis[hydroxy(methyl)amino]-1,3,5-triazine (BHT) ligands with monoanionic terdentate ligands such as 2-[(2-pyridinylmethylene)amino]phenol. Moderate selectivities were observed in formation of ternary iron(iii) complexes of iron(iii) between BHT ligands and neutral terdentate ligands such as pyridin-2-ylmethylpyridin-2-ylmethyleneamine. Results obtained by MS titrations were in a reasonable agreement with UV titration data indicating that quantitative ESI MS spectrometry can be applied to these labile iron(iii) complexes. CSIRO 2013.

Full text of DOI:10.1071/CH13179

RESEARCH FRONT
CSIRO PUBLISHING
Aust. J. Chem. 2013, 66, 791–797
Full Paper
http://dx.doi.org/10.1071/CH13179
Formation of Ternary Complexes of Iron(III) Cations
in Solution and Gas Phase
A
A B
,
Xing Xin Liu and Artem Melman
A
Department of Chemistry and Biomolecular Science, Clarkson University, Potsdam,
NY 13676, USA.
B
Corresponding author. Email: amelman@clarkson.edu
Formation of labile 1 : 1 : 1 ternary mononuclear complexes of iron(III) cation with Z3-terdentate meridional binders was
studied using electrospray ionisation mass spectrometry (ESI-MS) titration and UV-Vis titration in solution phase. Low
selectivities towards formation of ternary heteroleptic complexes in the solution phase vs. symmetric 2 : 1 complexes were
obtained with combinations of dianionic 2,6-bis[hydroxy(methyl)amino]-1,3,5-triazine (BHT) ligands with monoanionic
terdentate ligands such as 2-[(2-pyridinylmethylene)amino]phenol. Moderate selectivities were observed in formation of
ternary iron(III) complexes of iron(III) between BHT ligands and neutral terdentate ligands such as pyridin-2-ylmethylpyridin-
2-ylmethyleneamine. Results obtained by MS titrations were in a reasonable agreement with UV titration data indicating
that quantitative ESI MS spectrometry can be applied to these labile iron(III) complexes.
Manuscript received: 15 April 2013.
Manuscript accepted: 24 May 2013.
Published online: 11 June 2013.
Introduction
bonds through redox reactions. Coordination bonding has been
used for self-assembly of metal oxide framework coordination
polymers,[2–4] molecular grids,[5–8] molecular cages and nano-
capsules,[9–11] helicates,[12,13] topologically defined macro-
cycles,[14–17] trigger type molecules,[18] artificial base pairing
in DNA,[19] and other applications.[20–22] The majority of these
assemblies consist of a single binding motif, i.e. it is homoleptic
in nature thus putting a major limitation on their use.[23]
In contrast to the intrinsic complementarity of hydrogen bond-
ing, metal ions exposed to a mixture of different ligands
in solution phase form a mixture of homoleptic and heteroleptic
complexes even for the simplest case of octahedral complexes
with two tridentate ligands. To date, there are few reports on
metal coordination-driven dynamic heteroleptic aggregates that
exist as exclusive species in solution,[23] often referred to as
‘mixed complexes’,[24] ‘mixed ligand complexes’,[25,26] and
‘ternary complexes’.[27,28] Regardless of the terminology used,
selective formation of these species can be achieved through
simple mixing of two different ligands A and B and a metal
Preferential binding of different types of ligands by metal
cations is a basic process that is far from being properly
understood. Generally, it follows the well known tendency of
‘hard’ metal cations such as Mg2þ or Fe3þ to bind anionic
oxygen ligands such as carboxylates or hydroxamates while
‘soft’ metal cations such as Agþ or Hg2þ preferentially bind
‘soft’ nitrogen, sulfur, or phosphine ligands.[1] However,
because most metal cations form four or more coordination
bonds, the highest stability of metal complexes may require a
combination of ‘hard’ and ‘soft’ ligands. This situation is
commonly observed in metalloproteins. For example, the ‘hard’
iron(III) cation in transferrin protein is simultaneously bound by
‘hard’ carboxylate and phenolate ligands but also by a ‘soft’
imidazole ligand. The preferential formation of complexes with
different types of ligands has not been rationalised and no the-
oretical or empirical rules have been suggested. Despite the
progress achieved in understanding factors influencing energies
of covalent bonding using computational approaches, applica-
tion of these methods to coordination bonding in solution phase
is hindered by a strong contribution of solvation energy, which
cannot be computed with sufficient precision. Understanding
factors that influence formation of metal complexes with both
‘hard’ and ‘soft’ ligands, and finding ligands that display com-
plementarity of coordination bonding can be important for
several applications.
cation Mnþ
:
A þ B þ Mnþ Ð A-M-B þ A-M-A þ B-M-B
Selective mononuclear heteroleptic binding can be obtained
either by enhancing stability of the heteroleptic complex
A-M-B, or by simultaneous destabilisation of both homoleptic
complexes A-M-A and B-M-B. This has been achieved using
several strategies including formation of pentacoordinated
metal complexes with tridentate ligand A and bidentate ligand
For example, metal-ligand interactions are extensively used
for self-assembly owing to high thermodynamic stability of
coordination bonds that can be rapidly formed under very mild
conditions. The inventory of available metal cations allows
a high degree of control over geometry of the bonding, as
well as a possibility for directed disassembly of coordination
B
[29–31]; topologically constrained systems with introduction of
a macrocycle into one of the ligands effectively precludes
formation of its homoleptic complexes.[32] However, the most
Journal compilation Ó CSIRO 2013
www.publish.csiro.au/journals/ajc

792
X. X. Liu and A. Melman
Me
R ϭ
O
N
N
N
N
O
N
NH
N
N
N
N
R
R
Fe
N
HEHT
H2BHT
- HEHT, - Hϩ
O
N
O
- H2BHT
Me
Me
Fe(BHT)(EHT) 5
ϩ
Me
Me
Me
N
N
N
O
HO
N
N
O
N NH
N
N
N
N
N
N
N
R
R
Fe
R
N
N
N
R
Fe
- 3Hϩ
N
N
N
O
O
O
HN
N
Me
Me
Me
Fe(EHT)ϩ2
R
Fe(BHT)(HBHT) 3
4
R
N
N
N
N
Fe3ϩ
N
N
N
N
N
NH
OH
OH
OH
N
Et
2
HEHT
1
H2BHT
Fig. 1. Dynamic equilibria in formation of ternary complex 5 and symmetric complex 3 from H2BHT ligand 1 and
HEHT ligand 2 with iron(III) cations, which provides a 97 : 3 ratio of 5 : 3 under weakly acidic conditions at which
deprotonation of complex 3 is suppressed.
direct approach to selective formation of ternary complexes
involves using two ligands possessing different electric charges.
This effect has been observed in selective formation of
copper(II) complexes involving a-amino acids as ligand A and
2,20-bipyridine as ligand B, and in formation of ternary nickel(II)
complexes involving nitrilotriacetic acid as ligand A and a
hexahistidine tag of recombinant proteins as ligand B. Selective
formation of ternary ligands around iron(III) cation was achieved
with a pair of complementary Z3-terdentate meridional binders
(pincer ligands) H2BHT 1 and HEHT 2 (BHT ¼ 2,6-bis[hydroxy
complexes between H2BHT ligand 1 and other tridentate merid-
ional binders, and we determine the reaction equilibrium using
ESI mass-spectrometry.
Results and Discussion
We chose ligands 6 and 7 (Fig. 2) as a typical monoanionic
counterpart, to be used with the dianionic BHT ligand 1, to form
neutral complexes of type 10 and 11. As a more distant analogy,
we also tested neutral tridentate meridional binders, terpyridines
8 and 9, that were expected to form cationic ternary complexes
of type 12 and 13. Synthesis of all of these ligands was done in
one step using reactions of pyridine-2-aldehyde or imidazole-4-
aldehyde with either 2-aminomethylpyridine (compounds 8 and
9) or 2-aminophenol (compounds 6 and 7).
The ESI-MS (positive mode) spectrum of BHT iron(III)
complex 3 shows a strong peak of [(HBHT)2Fe]þ at m/z 566
(Fig. 3a). Along with it there was a strong peak of a free
[H3BHT]þ cation at m/z 257, yet no traces of [BHTFe]þ
complex (m/z 312) were present. Instead, a 3 : 1 complex
[H2BHT(HBHT)Fe]þ at m/z 822 was observed. The formation
of a 3 : 1 BHT-iron complex has not been previously observed in
solution phase.[34] Its formation can be attributed to dispropor-
tionation of kinetically labile complex 3 under conditions of the
(methyl)amino]-1,3,5-triazine,
EHT ¼ 2-(ethylmethylene)-
both
hydrazino-6-hydroxy(methyl)amino-1,3,5-triazine),
belonging to the hydroxyamino-1,3,5-triazine family
(Fig. 1).[33] Preferential formation of heteroleptic complex 5
was caused by lower stability of homoleptic complex 3 due to
weaker energy of the Fe-OH coordination bond, while homo-
leptic complex 4 was less stable than heteroleptic complex 5
because of intrinsically lower energy of the coordination bond
between a ‘hard metal’ iron(III) cation and ‘soft’ sp2 nitrogen-
containing ligands. As a result, under weakly acidic conditions
the equilibrium is shifted towards formation of complex 5 while
only minor amounts of homoleptic complex 3 are present and no
complex 4 is observed. Depending on substituents at the triazine
ring, the ratio of heteroleptic : homoleptic complexes ranged
from 94 : 6 to 97 : 3. Neutralisation of pH results in complete
reversal of the equilibrium towards deprotonated homoleptic
complex 3.
electrospray ionisation. With excess H2BHT ligand
1
(13 equiv.) peaks at m/z 257 and 822 increased in intensity,
and new peaks at m/z 513 [H(HBHT)2]þ, 535 [(HBHT)2Na]þ,
and 1077 [H(HBHT)4Fe]þ were observed.
It can be anticipated that preferential formation of 1 : 1 : 1
ternary complexes can be a general property for ligands posses-
sing a different number of anionic groups. In that case preferen-
tial formation of either a symmetric homoleptic complex such as
3 or ternary heteroleptic complex such as 5 can also be
dynamically controlled by the acidity of the media. Extending
these studies here we report the formation of ternary iron(III)
Addition of tridentate ligand 6 to a solution containing 15 : 1
ratio of H2BHT ligand 1 and iron(III) acetate resulted in forma-
tion of a peak corresponding to ternary heteroleptic metal
complex 10 (m/z 508) along with a peak of protonated ligand
6 with m/z 199 (Fig. 3b). While all peaks of BHT-iron complex 3
and BHT ligand 1, including [H2BHT(HBHT)Fe]þ (m/z 822)
were observed, there were no traces for a 2 : 1 complex of ligand

Ternary Complexes of Iron(III)
793
H
N
Monoacidic
Neutral
Me
Me
N
N
N
N
N
N
N
O
O
N
N
R
N
R
N
Fe
Fe
N
N
N
N
HN
N
O
O
O
N O
OH
N
OH
Me
Me
Me
Me
10
11
H
N
6
7
Neutral
Neutral
N
N
N
N
N
N
N
O
O
N
N
ϩ
ϩ
N
N
HN
N
R
N
R
N
Fe
Fe
N
N
N
N
N
N O
O
N
N
8
Me
Me
9
12
13
Fig. 2. Structures of ligands 6–9 and expected ternary complexes 10–13 produced from these ligands with H2BHT ligand 1.
(HBHT)2Feϩ
257.0
566.0
(a)
(b)
H5BHTϩ2
513.1
100
0
100
0
H4BHT2Naϩ
535.1
H3BHT2FeNaϩ
257.8
566.0
567.0
822.1
514.1
588.0
Complex 10
507.9 H5BHTϩ2
513.0
(HBHT)2Feϩ
565.9
257.0
H4BHT2Naϩ
535.0
100
0
100
0
H3BHT2FeNaϩ
587.9
514.0
505.9
257.8
477.0
566.9
536.0 556.9
455.0
(HBHT)2Feϩ
565.9
H5BHTϩ2
513.0
257.0
H4BHT2Naϩ
535.0
(c)
(d)
100
0
100
0
H3BHT2FeNaϩ
587.9
Complex 11
566.1
514.0
536.0
557.0
257.8
496.9
466.0
Complex 12
507.0
(HBHT)2Feϩ
566.0
484.1
257.0
100
0
100
0
H4BHT2Naϩ
535.1
508.0
H3BHT2FeNaϩ
H5BHTϩ2
491.0
615.0
257.8
567.0 588.0
557.1
513.1
472.0
(HBHT)2Feϩ
566.0
257.0
(e)
100
0
100
0
H4BHT2Naϩ
535.1
H3BHT2FeNaϩ
587.9
Complex 13
566.1
H5BHTϩ2
513.2
565.9
496.0
479.0
480
567.1
580
258.1
500
462.1
460
500
520
540
560
600
1000
1500
m/z
m/z
Fig. 3. ESI-MS spectra (left) and expanded spectra for m/z 400–600 (right) of (a) complex 3 in the presence of 6.5 equiv. of ligand 1; (b) solution containing a
15 : 1 : 4 mixture of 1 : iron(III) : 6; (c) solution containing a 15 : 1 : 4 mixture of 1 : iron(III) : 7; (d) solution containing a 15 : 1 : 4 mixture of 1 : iron(III) : 8;
(e) solution containing a 15 : 1 : 4 mixture of 1 : iron(III) : 9.
6 with iron(III) cation (expected m/z 450). Analogous changes in
MS were observed using ligand 7, although both the peak
corresponding to the ternary complex 11 (m/z 497) and the peak
of protonated ligand 7 (m/z 188) were observed with very low
intensity (Fig. 3c). Peaks of ternary complexes 12 (m/z 507) and
13 (m/z 496) were also observed in solutions containing 4 equiv.
of ligand 8 or 9, 15 equiv. of BHT ligand 1, and 1 equiv. of
iron(III) acetate (Fig. 3d, e). In all of these cases no peaks with
m/z corresponding to 2 : 1 symmetric complexes of ligands 7–9
with either iron(III) or iron(II) cations were detected.
A study of the equilibrium between symmetric complex 3
and ternary complex 10 was done by titration of a solution of
ligand 1 (3.75 mM) and iron(III) acetate (0.25 mM) in a 1 %
solution of acetic acid in ethanol with ligand 6 (Fig. 4). Excess of
ligand 1 was used to shift the equilibrium towards formation of
the less stable symmetric complex 3 thus increasing the

794
X. X. Liu and A. Melman
(a)
(b)
(c)
Me
N
N
N
O
N
N
R ϭ
O
N
1.2
1.0
0.8
0.6
1.0
0.9
N
R
Fe
N
O
O
6
Me
H2BHT
- 6, - Hϩ
- H2BHT
Fe(BHT)(6) 10
ϩ
Me
Me
- 3Hϩ
N
N
N
N
OHO
O
N
0.8
0.7
0.6
N
N
N
R
R
Fe
N
N
Fe
N
N
0.4
0.2
O
N
O
N
O
Me
Me
14
Fe(BHT)(HBHT) 3
Not observed
Fe(6)ϩ2
R
0
N
N
Fe3ϩ
ϩ
ϩ
0
5
10
450
550
650
N
N
N
N
N
OH
Wavelength [nm]
Concentration of 6 [mM]
OH
OH
6 (0 to 4 mM)
0.5 mM
1 (7.5 mM)
Fig. 4. (a) Equilibria between ternary complex 10 and symmetric complex 3. No traces of symmetric complex 14 were observed. (b) Visible spectra upon
titration of BHT ligand 1 (3.75 mM) and iron(III) acetate (0.25 mM) with 0–22 mM of ligand 6. (c) Changes in absorbance at 521 nm with addition of ligand 6.
precision of determining the equilibrium constant. Addition of
ligand 6 (11 mM solution) resulted in changes in the absorbance
spectra in the visible range; the maximum shifted from 550 nm
(e ¼ 2716 Mꢀ1 cmꢀ1) for symmetric complex 3 to 521 nm, along
with increased absorbance (e ¼ 4042 Mꢀ1 cmꢀ1). Additionally,
a very strong absorbance band below 450 nm appeared which
obscured the second isosbestic point at lower wavelength. These
changes proceed progressively with the addition of ligand 6,
slowing down after addition of ,6 mM of ligand 6 but not
reaching saturation. These changes are consistent with forma-
tion of ternary complex 10 which does not further convert to a
symmetric 2 : 1 complex of ligand 6 with iron(III) cation under
these conditions.
These data allowed estimation of an equilibrium constant
between symmetric complex 3 and ternary complex 10. It can
be assumed that 50 % change in absorbance at 506 nm, upon
addition of 4 mM of ligand 6, indicates a 50 % drop in concen-
tration of complex 3. Under these conditions 50 % of iron(III)
cations are bound in complex 10, which allows us to estimate the
equilibrium constant between complexes 3 and 10. Substituting
BHT ligand with A and ligand 6 with B, equilibrium constants
were calculated at 480 nm according to Eqn 1:
of ligand 1 (7.5 mM) and iron acetate (0.25 mM) with ligand 8
(Fig. 5b) caused a gradual shift of the maximum from 574 nm
(e ¼ 2620 Mꢀ1 cmꢀ1) to 615 nm (e ¼ 1700 Mꢀ1 cmꢀ1) with sat-
uration practically achieved after addition of 8 mM of ligand 8.
Fifty percent change in absorbance was achieved at concentra-
tion of ligand 8 ¼ 1.0 mM, which corresponds to an equilibrium
constant of 10. Titration under the same conditions with ligand 9
proceeded analogously (Fig. 5c), with the 50 % change in
absorbance achieved only after addition of 3.0 equiv. of ligand
9, corresponding to an equilibrium constant of only 2.6.
Finding additional pairs of tridentate ligands capable of
selective formation of ternary complexes necessitates devel-
opment of newer methods for determination of equilibrium
constants. Detection of the formation of kinetically labile
ternary complexes such as 10–13, and determination of the
corresponding equilibrium constants between heteroleptic and
homoleptic metal complexes, can constitute a substantial
challenge. Data on the stoichiometry and the stability constants
of kinetically labile metal complexes formed in solution are
generally obtained from potentiometric, spectrophotometric,
or calorimetric measurements. However, despite their reliabil-
ity and high precision in the case of complicated systems, these
methods are considerably less valuable because of the intrinsi-
cally poor ability to identify these species. NMR spectroscopy,
the most commonly used method to study dynamic equilibria of
complexes, is frequently unsuitable for these purposes as most
transition metal cations are paramagnetic. Mass spectral data of
metal complexes[35,36] provide a very useful characterisation
technique in these cases. It has been successfully used for
detection of ternary copper complexes of amino acids with
neutral bidentate ligands.[37–41]
½FeABꢁ½Aꢁ
K ¼
ð1Þ
½FeA2ꢁ½Bꢁ
where [FeAB] ¼ [FeA2] ¼ 0.1125 mM, [A] ¼ 7.38 mM, and
[B] ¼ 3.98 mM so that the equilibrium constant is less than
1.85. Analogous results were observed in titration of complex
3 under the same conditions with monoacidic ligand 7 (Fig. 5a).
The observed lack of selectivity towards formation of ternary
complexes is in contrast to the high selectivity in the BHT/EHT
ligands with equilibrium constants in the range of 30–35.[34] The
drop in selectivity for ligands 6 and 7 is likely caused by lower
acidity of OH groups (pKa in the range of 9–10) in comparison to
pKa 8 of the hydroaminotriazine group.[34]
ESI-MS spectra were obtained for ternary complexes 10/11
and 12/13; these showed substantial differences in response
factors between otherwise very similar compounds. To quantify
the relative concentration of species in the solution, particularly
of symmetric complex 3 and ternary complex 10, a quantitative
ESI MS analysis was performed. Mixtures containing 0.5 mM of
iron(III) acetate (total amount 5 nmol), 7.5 mM of BHT ligand 1
(total amount 75 nmol), and 0–4 mM of ligand 6 (total amount 0
In contrast to the absence of selectivity in formation of
ternary complexes 10 and 11, substantially higher selectivity
was achieved with neutral ligands 8 and 9. Titration of a solution

Ternary Complexes of Iron(III)
795
1.2
1.2
1.0
0.8
0.6
0.4
0.2
0
1.2
1.0
0.8
0.6
1.1
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.4
0.2
0
400
500
600
700
400
500
600
700
400
500
600
700
Wavelength [nm]
Wavelength [nm]
Wavelength [nm]
1.4
0.70
0.69
0.68
0.67
0.66
0.65
0.64
0.63
0.62
1.3
1.2
1.1
1.0
0.9
0.8
0.7
1.3
1.1
0.9
0.7
0
2
4
6
8
0
2
4
6
8
10
0
5
10
Concentration of 7 [mM]
Concentration of 8 [mM]
Concentration of 9 [mM]
Fig. 5. Changes in visible spectra (above) and absorbance (below) upon titration of (a) BHT ligand 1 (3.75 mM) and iron(III) acetate (0.25 mM) with ligand 7;
(b) BHT ligand 1 (7.5 mM) and iron(III) acetate (0.5 mM) with ligand 8; (c) BHT ligand 1 (7.5 mM) and iron(III) acetate (0.5 mM) with ligand 9.
to 4 nmol) were directly injected into an ESI probe through 5 mL
loop followed by elution with methanol at 5 mL minꢀ1. Because
no traces of 2 : 1 symmetric iron(III) complex 14 were discovered
in the solution, it can be assumed that the equilibrium takes place
only between iron complexes 10 and 3. In that case, the
concentration of ternary complex 10 can also be detected
through the decrease of the intensity of the peak of complex 3
at m/z 566. The original peak intensities showed considerable
variation in response factors in experiments with different
ligands 6–9, and sometimes even with different concentrations
of ligands. In order to decrease these variations, instead of using
the absolute intensity of counts, we measured relative intensities
using the [H2BHT]þ peak of ligand 1 with m/z 257 as an internal
standard. In all experiments ligand 1 is present in large excess, so
its concentration does not appreciably change in the equilibrium
between complex 3 and a ternary complex.
As can be seen in Fig. 6a, addition of ligand 6 predictably
increased the amount of complex 10 (m/z 508, blue) while
decreasing symmetric complex 3 (m/z 566, red). Increasing
the concentration of ligand 6 to 3 mM caused only a 35 %
reduction in the intensity of the peak for complex 3 which
corresponds to an equilibrium constant of 1.3.
Analogous experiments were conducted with ligands 7-9.
Addition of ligand 7 was found to continuously increase the
peak of corresponding ternary complex 11 with m/z 497
(Fig. 6b). However, the increase was very small and the
decrease of peak of symmetric complex 3 at m/z 566 was
not significant, indicating that the equilibrium constant is less
than 1. In contrast, addition of ligands 8 or 9 caused a
substantial drop in intensity of signal for complex 3. For
example, addition of 1.2 mM of ligand 8 caused a 73 % drop
in intensity of peak of complex 3 which corresponds to
equilibrium constant 6.8 (Fig. 6c). For ligand 9, addition of
1 mM of the ligand caused a decrease of 54 % relative
intensity of peak for complex 3 (Fig. 6d) which corresponds
to an equilibrium constant of 3.0.
Use of these results for determination of equilibrium con-
stants of metal complexes is challenging. Despite the similarity
in structure of ligands 6 and 7, and almost equal equilibrium
constants in formation of ternary complexes 10 and 11, results of
their mass-spectrometric studies were completely different.
These results can be attributed to high kinetic lability of these
complexes and disruption of equilibrium during the ionisation
process which involves protonation of these complexes. More
encouraging results were obtained with quantitative mass spec-
trometry of cationic complexes 12 and 13, where there was
semiquantitative agreement with the results obtained by UV
titration. Better agreement in this case can be explained both by
lower lability of the complexes 12 and 13.
Conclusions
We studied the formation of ternary iron(III) complexes with
BHT and other tridentate meridional ligands by UV-Vis titra-
tion and quantitative mass-spectrometry. A substantial pref-
erence was found in formation of ternary complexes with
neutral but not monoacidic tridentate ligands. Equilibrium
constants obtained by quantitative mass-spectrometry were in
semiquantitative agreement with ones obtained by UV-Vis
titration. These results indicate potential applicability of ESI
MS for the quantitative study of equilibria in formation of
labile metal complexes and the need for additional studies in
the field.

796
X. X. Liu and A. Melman
(a)
(b)
0.020
0.015
0.010
0.005
0
0.025
0.020
0.015
0.010
0.005
0
0
1
2
3
0
1
2
3
Concentration of 6 [mM]
Concentration of 7 [mM]
(c)
(d)
0.25
0.20
0.15
0.10
0.05
0.07
0.06
0.05
0.04
0.03
0.02
0.01
0
0
0
1
2
3
0
1
2
3
Concentration of 8 [mM]
Concentration of 9 [mM]
Fig. 6. Changes in peak height relative to [H3BHT]þ (m/z 257) of complex 3 (m/z 566, red squares) and ternary complexes (a) 10, (b) 11, (c) 12,
and (d) 13 (m/z 508, blue diamonds) upon addition of 0–3 mM of ligands 7–9, respectively, to solutions containing 7.5 mM of BHT ligand 1 and
0.5 mM of iron(III) acetate.
Experimental
and evaporation to dryness, the product was obtained as a solid
1
(32.2 mg, 86 %). H NMR (400 MHz, CDCl3): d 4.89 (s, 2H),
General Information
7.20 (t, J ¼ 7.2 Hz, 1H), 7.32 (d, J ¼ 7.6 Hz, 1H), 7.43 (s, 1H),
Thin layer chromatography was performed on aluminium
sheets precoated with silica gel (Merck, Kieselgel 60 F-254).
Flash chromatographic separations were performed on silica
gel (Merck, Kieselgel 60, 230–400 Mesh ASTM). Unless
mentioned specifically, 1H and 13C NMR spectra were
recorded on a Bruker AMX-400 spectrometer at room tem-
perature in deuterated chloroform (CDCl3) using the residual
solvent peaks for calibration. Unless otherwise stated, all
reagents used are commercially available. Glassware was
oven-dried before use and solvents were purified by conven-
tional methods. Titration experiments were done on Varian
Cary 100 and Lambda 35 Perkin-Elmer UV/Vis spectro-
meters. Ligands 1,[34] 8,[42] and 6[43] were synthesised
according to published procedures.
7.63–7.69 (m, 2H), 8.38 (s, 1H), 8.58 (d, J ¼ 4.4 Hz, 1H).
2-[(E)-(1H-Imidazol-4-ylmethylidene)amino]phenol (7)
To a solution of 4-formylimidazole (19.2 mg, 0.2 mmol) in 1 mL
of dry CH2Cl2 was added dropwise a solution of 2-aminophenol
(21.8 mg, 0.2 mmol) in 1 mL of dry CH2Cl2 with stirring at room
temperature. Anhydrous Na2SO4 (200 mg) was then added, and
the mixture was stirred at room temperature for 2 h. After fil-
tration and evaporation to dryness, the product was obtained as a
solid (30.7 mg, 82 %). 1H NMR (400 MHz, CDCl3): d 6.89
(t, J ¼ 7.6 Hz, 1H), 6.98 (d, J ¼ 7.6 Hz, 1H), 7.16 (t, J ¼ 7.6 Hz,
1H), 7.23–7.26 (m, 1H), 7.59 (s, 1H), 7.79 (s, 1H), 8.60 (s, 1H).
General Procedure for UV Titration
(E)-1-(1H-Imidazol-4-yl)-N-(pyridin-2-ylmethyl)
methanimine (9)
To a 1 % solution of acetic acid in absolute ethanol (3 mL)
containing BHT ligand 1 (7.5 mM) and iron acetate (0.5 mM) in
a quartz UV cell were added portions of ligands 6–9 (50 mM in
ethanol containing 1 % acetic acid) by aliquots of 6 mL, initially,
and then 12 and 30 mL until a total of 90 mL were added. After
every addition the reaction mixture was stirred and UV-Vis
spectra (350–700 nm) were recorded.
To a solution of 4-formylimidazole (19.2 mg, 0.2 mmol) in 1 mL
of dry CH2Cl2 was added dropwise a solution of 2-picolylamine
(21.6 mg, 0.2 mmol) in 1 mL of dry CH2Cl2 with stirring at room
temperature. Anhydrous Na2SO4 (200 mg) was then added, and
the mixture was stirred for a further 2 h. After filtration

Ternary Complexes of Iron(III)
797
General Procedure for Quantitative ESI-MS Spectrometry
[17] C. D. Pentecost, K. S. Chichak, A. J. Peters, G. W. V. Cave,
S. J. Cantrill, J. F. Stoddart, Angew. Chem. Int. Ed. 2007, 46, 218.
doi:10.1002/ANIE.200603521
[18] L. Zelikovich, J. Libman, A. Shanzer, Nature 1995, 374, 790.
doi:10.1038/374790A0
[19] H. Weizman, Y. Tor, J. Am. Chem. Soc. 2001, 123, 3375. doi:10.1021/
JA005785N
[20] A. J. Blake, N. R. Champness, P. Hubberstey, W. S. Li,
M. A. Withersby, M. Schroder, Coord. Chem. Rev. 1999, 183, 117.
doi:10.1016/S0010-8545(98)00173-8
[21] D. L. Long, E. Burkholder, L. Cronin, Chem. Soc. Rev. 2007, 36, 105.
doi:10.1039/B502666K
[22] B. Linton, A. D. Hamilton, Chem. Rev. 1997, 97, 1669. doi:10.1021/
CR960375W
[23] S. De, K. Mahata, M. Schmittel, Chem. Soc. Rev. 2010, 39, 1555.
doi:10.1039/B922293F
[24] C. De Mello Donega, S. A. Junior, G. F. De Sa, Chem. Commun. 1996,
ESI-MS spectra were measured by a Waters Micromass Micro
QTOF spectrometer using direct injection. To a solution of
acetic acid in absolute methanol (1 % v/v, 3 mL) containing
BHT ligand 1 (7.5 mM) and iron acetate (0.5 mM) was added a
ligand 6–9 (50 mM in methanol containing 1 % acetic acid) by
6 mL portions, initially, and then 12 and 30 mL portions until a
total of 120 mL were added. 5 mL samples of the resultant
solution were injected into the ESI mass spectrometer using a
5 mL minꢀ1 stream of methanol under ESþ mode. The ESI was
operating under the following conditions: capillary voltage
1600 V; sample cone 40 V; extraction cone 4.0 V; desolvation
temperature 20.08C; source temperature 20.08C, with instru-
ment parameters: ion energy 1.0 V; collision energy 6.0 V;
aperture 7.0 V; MCP detector 2100.0 V; acceleration 200.0 V.
1199. doi:10.1039/CC9960001199
[25] P. Laine, A. Gourdon, J. P. Launay, Inorg. Chem. 1995, 34, 5156.
doi:10.1021/IC00125A012
[26] S. A. A. Sajadi, B. Song, H. Sigel, Inorg. Chim. Acta 1998, 283, 193.
doi:10.1016/S0020-1693(98)00097-8
Acknowledgement
This work was supported by National Science Foundation Grant CHE
1150768.
[27] B. E. Fischer, H. Sigel, J. Am. Chem. Soc. 1980, 102, 2998.
doi:10.1021/JA00529A021
References
[28] H. Sigel, B. E. Fischer, B. Prijs, J. Am. Chem. Soc. 1977, 99, 4489.
doi:10.1021/JA00455A047
[1] R. G. Pearson, J. Am. Chem. Soc. 1963, 85, 3533. doi:10.1021/
JA00905A001
[29] T. Yutaka, I. Mori, M. Kurihara, J. Mizutani, N. Tamai, T. Kawai,
M. Irie, H. Nishihara, Inorg. Chem. 2002, 41, 7143. doi:10.1021/
IC0260116
[2] M. Eddaoudi, D. B. Moler, H. L. Li, B. L. Chen, T. M. Reineke,
M. O’Keeffe, O. M. Yaghi, Acc. Chem. Res. 2001, 34, 319.
doi:10.1021/AR000034B
[30] D. Sielemann, A. Winter, U. Florke, N. Risch, Org. Biomol. Chem.
[3] S. Kitagawa, R. Kitaura, S. Noro, Angew. Chem. Int. Ed. 2004, 43,
2004, 2, 863. doi:10.1039/B316633C
2334. doi:10.1002/ANIE.200300610
[31] C. K. Hui, B. W. K. Chu, N. Y. Zhu, V. W. W. Yam, Inorg. Chem.
2002, 41, 6178. doi:10.1021/IC025685D
[32] R. Kramer, J. M. Lehn, A. Marquis-Rigault, Proc. Natl. Acad. Sci.
[4] B. H. Ye, M. L. Tong, X. M. Chen, Coord. Chem. Rev. 2005, 249, 545.
doi:10.1016/J.CCR.2004.07.006
[5] P. Broekmann, K. H. Dotz, C. A. Schalley, Templates in Chemistry III
1993, 90, 5394. doi:10.1073/PNAS.90.12.5394
2009 (Springer: Berlin).
[33] G. Melman, P. Vimal, A. Melman, Inorg. Chem. 2009, 48, 8662.
doi:10.1021/IC901151N
[34] I. Ekeltchik, J. Gun, O. Lev, R. Shelkov, A. Melman, Dalton Trans.
[6] U. Schlickum, R. Decker, F. Klappenberger, G. Zoppellaro,
S. Klyatskaya, M. Ruben, I. Silanes, A. Arnau, K. Kern, H. Brune,
J. V. Barth, Nano Lett. 2007, 7, 3813. doi:10.1021/NL072466M
[7] T. Kudernac, S. B. Lei, J. Elemans, S. De Feyter, Chem. Soc. Rev.
2009, 38, 402. doi:10.1039/B708902N
[8] N. A. A. Zwaneveld, R. Pawlak, M. Abel, D. Catalin, D. Gigmes,
D. Bertin, L. Porte, J. Am. Chem. Soc. 2008, 130, 6678. doi:10.1021/
JA800906F
2006, 1285. doi:10.1039/B513719E
[35] M. J. Keith-Roach, Anal. Chim. Acta 2010, 678, 140. doi:10.1016/
J.ACA.2010.08.023
[36] V. B. Di Marco, G. G. Bombi, Mass Spectrom. Rev. 2006, 25, 347.
doi:10.1002/MAS.20070
[37] G. R. Blacken, M. H. Gelb, F. Turecek, Anal. Chem. 2006, 78, 6065.
doi:10.1021/AC060509Y
[38] C. L. Gatlin, F. Turecek, J. Mass Spectrom. 1995, 30, 1636.
doi:10.1002/JMS.1190301118
[39] F. Turecek, Mass Spectrom. Rev. 2007, 26, 563. doi:10.1002/MAS.
20137
[40] F. Turecek, A. I. S. Holm, S. Panja, S. B. Nielsen, P. J. Hvelplund,
J. Mass Spectrom. 2009, 44, 1518. doi:10.1002/JMS.1642
[41] F. Turecek, J. W. Jones, A. I. S. Holm, S. Panja, S. B. Nielsen,
P. Hvelplund, J. Mass Spectrom. 2009, 44, 707. doi:10.1002/
JMS.1546
[9] M. Fujita, Chem. Soc. Rev. 1998, 27, 417. doi:10.1039/A827417Z
[10] C. Y. Su, Y. P. Cai, C. L. Chen, M. D. Smith, W. Kaim, H. C. zur Loye,
J. Am. Chem. Soc. 2003, 125, 8595. doi:10.1021/JA034267K
[11] W. Y. Sun, M. Yoshizawa, T. Kusukawa, M. Fujita, Curr. Opin. Chem.
Biol. 2002, 6, 757. doi:10.1016/S1367-5931(02)00358-7
[12] C. Piguet, G. Bernardinelli, G. Hopfgartner, Chem. Rev. 1997, 97,
2005. doi:10.1021/CR960053S
[13] C. Piguet, M. Borkovec, J. Hamacek, K. Zeckert, Coord. Chem. Rev.
2005, 249, 705. doi:10.1016/J.CCR.2004.08.023
[14] C. J. Kuehl, S. D. Huang, P. J. Stang, J. Am. Chem. Soc. 2001, 123,
9634. doi:10.1021/JA0114355
[42] S. Padilla, R. Tejero, J. Adrio, J. C. Carretero, Org. Lett. 2010, 12,
[15] S. Leininger, B. Olenyuk, P. J. Stang, Chem. Rev. 2000, 100, 853.
doi:10.1021/CR9601324
5608. doi:10.1021/OL102605Q
[43] L. M. Liable-Sands, C. Incarvito, A. L. Rheingold, C. J. Qin,
A. L. Gavrilova, B. Bosnich, Inorg. Chem. 2001, 40, 2147.
doi:10.1021/IC0012625
[16] K. S. Chichak, S. J. Cantrill, A. R. Pease, S. H. Chiu, G. W. V. Cave,
J. L. Atwood, J. F. Stoddart, Science 2004, 304, 1308. doi:10.1126/
SCIENCE.1096914