A new tris(phosphonomethyl) monoacetic acid cyclam derivative: Synthesis, acid-base and metal complexation studies

Article abstract of DOI:10.1002/ejic.201000919

A new cyclam-based ligand, [4,8,11-tris(phosphonomethyl)-1,4,8,11- tetraazacyclotetradec-1-yl]acetic acid (H₇te3p1a) was synthesised and characterised. The potentiometricand ³¹P NMR spectroscopic titrations of the complexes inan aqueous solution at an ionic strength of 0.10 or0.50 mol dm^-3 in [N(CH₃)₄]NO₃, respectively, indicated that the ligand is highly basic. The stepwise protonations were dominated by extensive proton relocations and were influenced by the intramolecular hydrogen bonds. The thermodynamic stability constants for the complexes formed by H₇te3p1a in the presence of metal ions were determined by potentiometric titrations. The Cu²⁺ complexes exhibited very high stability, those of Zn²⁺, Cd²⁺, Pb ²⁺, La³⁺, Sm³⁺, Gd³⁺, Ho ³⁺ and Lu³⁺ presented high stability while that of Ca ²⁺ had lower stability, as expected. The ligand and its complexes behaviour is similar to that of the tetrakis(phosphonomethyl)cyclam derivative, H₈tetp. The NMR spectroscopic studies of the Zn²⁺ and Cd²⁺ complexes pointed to the involvement of the three phosphonate moieties in the coordination of the latter ion, while several isomer species were generally present in solution for the Zn²⁺ complex. A major isomer was stabilised in solution upon heating. In this isomer only one of the methylphosphonate arms was coordinated to the metal centre and this isomer probably adopts a distorted square pyramidal geometry with the macrocyclic ring in the type I conformation.

Full text of DOI:10.1002/ejic.201000919

FULL PAPER
DOI: 10.1002/ejic.201000919
A New Tris(phosphonomethyl) Monoacetic Acid Cyclam Derivative:
Synthesis, Acid-Base and Metal Complexation Studies
[a]
[a,b]
[c]
[c]
[c]
Luís M. P. Lima, Rita Delgado,*
Jan Plutnar, Petr Hermann,* and Jan Kotek
Keywords: Macrocyclic ligands / Cyclam derivatives / Metal complexes / Thermodynamics / Stability constants /
Coordination
A new cyclam-based ligand, [4,8,11-tris(phosphonomethyl)-
,4,8,11-tetraazacyclotetradec-1-yl]acetic acid (H te3p1a)
Ho3+ and Lu3+ presented high stability while that of Ca2+ had
lower stability, as expected. The ligand and its complexes
behaviour is similar to that of the tetrakis(phosphonomethyl)-
1
7
was synthesised and characterised. The potentiometric
and 31P NMR spectroscopic titrations of the complexes in
an aqueous solution at an ionic strength of 0.10 or
8
cyclam derivative, H tetp. The NMR spectroscopic studies of
the Zn2 and Cd complexes pointed to the involvement of
the three phosphonate moieties in the coordination of the lat-
ter ion, while several isomer species were generally present
in solution for the Zn2 complex. A major isomer was stabi-
+
2+
0
.50 moldm–3 in [N(CH
3 4 3
) ]NO , respectively, indicated that
the ligand is highly basic. The stepwise protonations were
dominated by extensive proton relocations and were influ-
+
enced by the intramolecular hydrogen bonds. The thermo- lised in solution upon heating. In this isomer only one of the
dynamic stability constants for the complexes formed by
te3p1a in the presence of metal ions were determined by
methylphosphonate arms was coordinated to the metal cen-
tre and this isomer probably adopts a distorted square py-
ramidal geometry with the macrocyclic ring in the type I con-
formation.
H
7
2
+
potentiometric titrations. The Cu complexes exhibited very
2
+
2+
2+
3+
3+
3+
high stability, those of Zn , Cd , Pb , La , Sm , Gd
,
Introduction
able chemical properties are high thermodynamic stability
and kinetic inertness, fast complexation and high solubility
in physiological media, as well as a high hydrophilicity and
Tetraaza macrocycles are efficient chelators of metal ions
since they are able to form strong complexes with a variety
of cations ranging from alkaline to post-transition metals.
The addition of further donor atoms to the macrocyclic
backbone in the form of pendant arms increases the coordi-
nation ability of the ligands and enables fine-tuning of the
properties towards specific metal ions. Among these macro-
cyclic chelators, those based on cyclen (1,4,7,10-tetraaza-
cyclododecane) and cyclam (1,4,8,11-tetraazacyclotetra-
[
5,8,9]
a low charge on the complex.
A common strategy to
improve the efficiency of radiopharmaceuticals is to attach
the complex to a targeting biomolecule, such as a mono-
clonal antibody or its fragments, a small peptide or a recep-
tor ligand, in the form of a bifunctional chelator (BFC).
This BFC must contain a reactive group that can be used
for the attachment of the complex to the biomolecule with-
out disturbing the coordination sphere of the metal
[
1–3]
decane) account for the vast majority of the studies.
The main interest in these compounds lies in the possible
medical applications of their metal complexes as contrast
[
5,10–12]
ion.
Ligands based on the cyclam skeleton are interesting
since they can suit the needs of very diverse ions, such as
the transition metals and lanthanides. The tetraacetic
agents in magnetic resonance imaging (MRI) or as radio-
pharmaceuticals for diagnosis and therapy.[
4–8]
In order to
(
H teta) and tetrakis(phosphonomethyl) acid (H tetp) de-
4
8
be used in metal chelates, macrocyclic ligands must form
strong and nonlabile complexes. In this respect, the desir-
rivatives of cyclam have been studied for a long time
Scheme 1). A cyclam derivative containing one methyl-
(
phosphonate and three acetate pendant arms, H te3a1p,
[
a] Instituto de Tecnologia Química e Biológica, Universidade
5
Nova de Lisboa,
has recently been studied and displayed a very good coordi-
nation ability towards the transition metal ions and also a
good coordination ability towards the lanthanide(III)
Av. da República, 2780-157 Oeiras, Portugal
Fax: +351-214411277
E-mail: delgado@itqb.unl.pt
[
13,14]
[
[
b] Instituto Superior Técnico, Universidade Técnica de Lisboa,
Av. Rovisco Pais, 1049-001 Lisboa, Portugal
c] Department of Inorganic Chemistry, Faculty of Science,
Univerzita Karlova (Charles University),
Hlavova 2030, 12840 Prague 2, Czech Republic
E-mail: petrh@natur.cuni.cz
ions.
H te3p1a, as an analogue of H tetp with an acetate arm
We have thus decided to develop a new ligand,
7
8
replacing one of the methylphosphonate arms. This ligand
can be considered to be a bifunctional chelator since a suit-
able chemical modification of the single carboxylic group
can easily lead to bioconjugates. In this work, we report
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejic.201000919.
Eur. J. Inorg. Chem. 2011, 527–538
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
527

L. M. P. Lima, R. Delgado, J. Plutnar, P. Hermann, J. Kotek
FULL PAPER
on the synthesis and detailed acid-base and complexation pended by a Mannich-type reaction with paraformaldehyde
properties of the ligand as studied by potentiometry and and triethyl phosphite under anhydrous conditions, but for-
multinuclear NMR spectroscopy.
mation of a byproduct hampered the yield and purity of
ester 4. Based on the mass spectra of the reaction mixture
+
[
m/z 626.7, formula (C H N O P ) ], the byproduct is
27
57
4
8 2
probably a cyclam formaldehyde aminal with two appended
methylphosphonate ester arms, one of which is bound to a
quaternary nitrogen atom. It is known that formaldehyde
can react with the two secondary amines that are separated
by a propylene chain to form a six-membered aminal[
with a structure that hinders the introduction of a fourth
pendant arm. Efforts to minimise this side-reaction, by
decreasing the reaction temperature and/or the excess
amount of paraformaldehyde added, also resulted in a de-
crease in the yield of the desired ester 4. The optimal condi-
tions were found to be the slow addition of paraformalde-
hyde over the course of several days, which eventually de-
creased the amount of byproduct in the reaction mixture to
17]
Scheme 1. The structure of the cyclam ligands.
10 to 20%. A final hydrolysis of both the carboxylic and the
phosphonic esters by heating 4 under reflux in azeotropic
aqueous hydrochloric acid afforded the target ligand, to-
gether with the corresponding hydrolysed byproduct and a
few other minor impurities. The product was purified by
cationic exchange chromatography as described in the Exp.
Sect.
Results and Discussion
Synthesis of the Ligand
The synthesis of H te3p1a was achieved in five steps
7
starting from cyclam (Scheme 2). The mono N-functionali-
sation of cyclam with a methylcarboxylic ester moiety was
achieved from 1,4,8-tris(trifluoroacetyl)-1,4,8,11-tetraaza- Acid–Base Behaviour
cyclotetradecane (1), which was synthesised by a reported
method for the selective triprotection of the secondary
The acid-base properties of H te3p1a were studied by
7
[
15]
31
amines of cyclam.
Although a number of other macro- both potentiometric and P NMR spectrometric titrations
cyclic polyamine protection methods have been reported in in aqueous solution. The overall protonation constants of
[
16]
the literature, this trifluoroacetyl protection proved to be the ligand and the respective standard deviations, as ob-
particularly effective for our purpose because the relevant tained from fitting the experimental data, are collected in
reaction of cyclam was extremely selective and high yield- the Supporting Information (Table S1), together with the
ing. The subsequent N-alkylation step with tert-butyl bro- literature values for the related ligands, H teta, H te3a1p
4
5
moacetate proceeded under mild conditions in good yield. and H tetp. The calculated stepwise protonation constants
8
The removal of the trifluoroacetyl protection groups was for all of the ligands, as well as the cumulative value of their
easily achieved under very mild conditions in nearly quanti- first five protonation constants, are presented in
[
13,18,19]
tative yield. The methylphosphonic ester groups were ap- Table 1.
Scheme 2. The synthetic pathway for H
7
te3p1a.
528
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 527–538

A New Cyclam-Based Ligand
H
Table 1. The stepwise protonation constants (logK ) of the ligands much higher than those for H te3a1p (21.66) and H teta
5 4
–3
i
as determined by potentiometry; T = 298.2 K; I = 0.10 moldm
in [N(CH ]NO
(
20.68). This is in agreement with the previously mentioned
influence of the phosphonomethyl pendants on the in-
H
8
tetp[b] creased basicity of the macrocyclic amines.
3
)
4
3
.
Equilibrium quotient[a]
HL]/[L][H]
L]/[LH][H]
L]/[LH ][H]
L]/[LH ][H]
L]/[LH ][H]
L]/[LH ][H]
L]/[LH ][H]
L]/[LH ][5H]
L]/[L][H]
H
te3p1a
H
teta[b]
H
te3a1p[c]
7
4
5
[
24.88[d,e,f] 10.59
10.08
11.78[d]
9.88
6.34
3.85
2.63
2.14
–
25.28[d,e]
[H
2
H
4
3
2
8.08
6.60
5.54
2.83[
4.15
3.29
1.84
–
–
–
8.85
7.68
6.23
5.33
2.28
–
3
H
6
5
4
5
e]
[
H
8
7
6
4.33
[H
7
–
[
H
5
45.10
29.95
34.48
48.04
[
[
[
a] The charges of the equilibrium species are omitted for clarity.
[
18,19]
[13]
31
b] Ref.
e] The constant (logβ
[c] Ref. [d] Determined by P NMR spectroscopy.
) corresponds to a double protonation equi-
2
2
librium ([H
ionic strength; the logβ
2
L]/[L][H] ). [f] Converted from measurements at higher
obtained at I = 0.5 moldm–3 was 24.17.
2
The ligand H te3p1a contains a total of eleven basic sites
7
corresponding to the seven ionisable –OH groups of the
pendant arms and the four tertiary amines. The potentio-
metric method used enabled accurate determination of the
protonation constants corresponding to six of those basic Figure 1. The 31P NMR spectroscopic titration of H te3p1a in
7
–
3
3 4 3
sites. Polyaza macrocycles with appended methylphos- water at I ≈ 0.50 moldm in [N(CH ) ]NO . The phosphorus reso-
nance and the nitrogen atom labelling scheme adopted for the li-
gand can be seen in Scheme 3. In Figure S1 of the Supporting
Information this image can be seen in colour.
phonate groups usually present very high values for the first
protonation constant(s), which can hinder the accurate de-
termination of their value(s) by potentiometry, as pre-
[
13,18,20–26]
viously observed.
This trend has been explained
The next three stepwise protonation constants, at inter-
by the effect of spreading the high electron density of the mediate pH values, were expected to correspond to the pro-
fully deprotonated phosphonate group to the nearest nitro- tonation of three aminomethylphosphonate groups. How-
[
3]
gen atom, or by the formation of strong intramolecular ever, the value determined for the first of these constants
hydrogen bonds.[21]
(8.08) was much higher than the other two (6.60 and 5.54),
In the case of H te3p1a, the first two protonations have which could be explained by the presence of intramolecular
7
very high values and only an overall constant that corre- hydrogen bonds.[21] A similar difference can be seen be-
sponds to this double protonation could be determined by tween the fourth and fifth protonation constants for
3
1
[18,19]
the P NMR spectroscopic titration in the alkaline region H tetp
(
where similar interactions are expected. The
pH 9–14); the measurement was performed in an aqueous sixth constant was attributed to the protonation of the acet-
solution and the ionic strength was maintained at ate group. Lastly, only a global constant could be deter-
8
–
3
0
.50 moldm with [N(CH ) ]NO . The spectra generally mined for the two remaining protonations at low pH. It was
3 4 3
presented three resonances that correspond to the non- not possible to obtain a separate constant for H te3p1a,
7
equivalent phosphorus atoms; for a detailed discussion, see which indicated that these last two protonations could be
the description of the entire NMR spectroscopic titration attributed to a concerted process analogous to the proton-
below (Figure 1). Determination of the constant was ation of the first two nitrogen atoms. Similar concerted pro-
achieved using the two resonances that correspond to the tonation phenomena have been previously observed for cy-
[
18,19,21,23–25,27]
phosphorus atoms bound to the amines that are being pro- clam itself and for several of its derivatives.
tonated (b and bЈ). The value thus obtained (24.17) was Further protonations on the ligand could not be deter-
converted to one valid at an ionic strength of 0.10 moldm–3 mined as they occur at an extremely low pH.
(see the details in the Exp. Sect.); this was later used as a
In order to assist with the assignment of the protonation
fixed constant in the refinement of the remaining proton- sites for the various protonated ligand species, the proton-
31
ation constants. It was not possible to determine the pro- ation of H te3p1a was studied with an additional P NMR
7
6
–
tonation constant of (Hte3p1a) due to its low abundance spectroscopic titration in an aqueous solution over a pH
in the equilibrium. This indicated that the values deter- range of –1 to 9. This titration was performed in conditions
mined for the first two stepwise protonation constants similar to those discussed in the previous paragraph, but
should be very close, which can be explained by a proton- without rigorous control of the ionic strength. Combination
ation process that is highly concerted for the first two pro- of the experimental data from the two titrations gave the
tons and must occur on two nitrogen atoms of the ring. The pH dependence of the 31P NMR spectroscopic chemical
value obtained for the logβ of H te3p1a (24.88) is almost shifts throughout the entire pH range (Figure 1 and Figure
2
7
as high as the corresponding value for H tetp (25.28) and S1 of Supporting Information in colour). The interpretation
8
Eur. J. Inorg. Chem. 2011, 527–538
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
529

L. M. P. Lima, R. Delgado, J. Plutnar, P. Hermann, J. Kotek
FULL PAPER
Scheme 3. The proposed protonation sequence of H
7
te3p1a, as well as the phosphorus resonances and the nitrogen atom labelling scheme.
of the 31P NMR spectroscopic chemical shift changes is still group. If this interaction is strong enough to stabilise a par-
mostly empirical as different classes of phosphorus com- ticular ligand conformation in solution, as has been ob-
[
21,23]
pounds show diverse behaviour. However, a significant served in similar compounds,
it will also affect the sig-
amount of data has been published with respect to the ami- nal of the phosphonate group appended to N11. The
[
28–31]
noalkylphosphonate groups in acyclic
cyclic
and macro- smaller upfield shift of δb relative to δbЈ (indicative of a
[
13,20–22,25,26,32–35]
31
compounds. In the P NMR spec- smaller protonation effect due to the hydrogen bonding)
troscopic titration of H te3p1a, one of the resonances (δ ) confirmed the assignment of resonance δb to the phos-
7
a
shows a different pH dependence from those of the other phonate group appended to N11.
two signals (δ and δ ). The H,P-HMQC correlation spec- The next three protonations should occur at the oxygen
trum acquired at pH ca. 8.5 (see Figure S10 in the Support- atoms of the three phosphonate groups to form the
b
bЈ
2–
ing Information) enabled the assignment of the phosphorus (H te3p1a) species and should lead to the dissociation of
5
resonances to the respective phosphonate groups. Thus, the the hydrogen bond previously described. Indeed, all the
δa resonance belongs to the phosphonate group bound to phosphorus resonances changed sharply at the intermediate
the N8 atom, the δ resonance belongs to the N11-bound pH (9 Ͼ pH Ͼ 5). Resonances δ and δ shifted strongly
b
b
bЈ
phosphonate and the δbЈ resonance belongs to the N4- downfield but the effect was too extensive to be explained
bound phosphonate pendant arm. On the basis of this as- only by the protonation of the oxygen atoms of the phos-
signment and the literature data for other macrocyclic ami- phonate groups. Therefore, it must also reflect a simulta-
noalkylphosphonate ligands, the protonation sequence was neous deprotonation of the corresponding amines at N4
proposed and the protonation sites were tentatively as- and N11 and redistribution of the two ring protons to N1
signed (Scheme 3).
and N8. This assumption is supported by the simultaneous
Starting from the fully deprotonated (te3p1a)7– species at upfield shift of δ to a lesser degree, which can be explained
a
high pH, there is a momentary simultaneous upfield shift by the overall effect of the protonation of N8 and of one
of all the resonances (pH Ͼ12). This shows that proton- of the oxygen atoms of the appended phosphonate group.
ation starts on the nitrogen atoms of the ring, as expected, The subsequent protonation should take place at the oxy-
but in a rather delocalised fashion that includes at least the gen atom of the acetate pendant arm bound to N1 to form
–
three amines with the appended phosphonate groups. How- the (H te3p1a) species, and that should lead to another
6
ever, briefly afterwards, δ starts to shift downfield while redistribution of the two ring protons back to N4 and N11.
a
δb and δbЈ continue to shift upfield concurrently Indeed, the simultaneous upfield shift of δ and δ and the
b
bЈ
(
12 Ͼ pH Ͼ 10). These upfield shifts indicated that proton- downfield shift of δ (4 Ͼ pH Ͼ 2) is in agreement with the
a
ation occurs on the nitrogen atoms opposite to each other, deprotonation at N1 and N8 and the concurrent proton-
N4 and N11, due to the electrostatic repulsion effect, to ation at N4 and N11. Protonation should then take place
5
–
form a double protonated species (H te3p1a) . This also at the deprotonated amines N1 and N8 to form the
2
+
indicated that protonation is simultaneous at these nitrogen (H te3p1a) species. This was confirmed by the sharp up-
8
atoms, as anticipated during the determination of the pro- field shift of δ and later, to a lesser degree, the upfield shift
a
tonation constants. At the same time, the smaller and unex- of δ and δ (2 Ͼ pH Ͼ –0.5). However, a slight downfield
b
bЈ
pected downfield shift of δ can only be explained by forma- shift is momentarily observed for one of the resonances
a
tion of a N–H···O hydrogen bond between a protonated (around pH 1). This could be indicative of either the depro-
amine and an oxygen atom of the phosphonate group la- tonation of an ammonium centre or the protonation of a
belled “a” (represented by a dashed line in Scheme 3). An second oxygen atom of a phosphonate group. Since the de-
interaction of this kind is more likely to occur with N11 protonation of an ammonium is unlikely at this stage with-
than with N4 as it is one bond closer to the phosphonate out further proton redistribution, the observation may be
530
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 527–538

A New Cyclam-Based Ligand
explained by the transient formation of a N–H···O hydro- Table 2. The stepwise stability constants (logKMHiL) for the com-
plexes with selected divalent and lanthanide(III) metal ions as de-
termined by potentiometry; T = 298.2 K; I = 0.10 moldm
gen bond between a deprotonated oxygen atom of one
phosphonate group and the proton that is starting to pro-
tonate the nitrogen atom N8. This led to the tentative as-
–
3
3 4 3
[N(CH ) ]NO .
Ion
Equilibrium quotient[a]
7 4
H te3p1aH teta
5 8
H te3a1p H tetp
signment of this resonance to δ because of the shorter dis-
b
Ca2+
[ML]/[M][L]
7.47
10.96
10.01
–
8.4
[b]
6.69
[c]
19.33[d]
tance between the corresponding phosphonate arm and N8
[b]
[c]
[
MHL]/[ML][H]
MH L]/[MHL][H]
[ML]/[MLOH][H]
MLOH]/[ML(OH)
7.2
–
8.73
–
(see above).
[
2
–
–
–
[
c]
Finally, the last protonations should occur simulta-
–
–
10.24
11.74
[c]
[
2
][H]
–
neously at the deprotonated oxygen atoms of all the phos-
phonate groups to eventually form the (H te3p1a) spe- Cu2+
4+
[ML]/[M][L]
MHL]/[ML][H]
MH L]/[MHL][H]
MH L]/[MH L][H]
[MH L]/[MH L][H]
ML]/[MLOH][H]
24.13
7.61
6.80
5.89
–
21.07 21.58
[c]
[c]
[c]
[c]
25.99
[c]
1
1
[
c]
[c]
[
3.46
6.85
3.36
–
8.09
cies, which was confirmed by the final downfield shift of all
the combined resonances (pH Ͻ –0.5). The assignment of
the protonation sites in the lower pH region was supported
by the molecular structure of the zwitterionic form of
[
c]
[c]
[
2
2.35
6.93
[c]
[
3
2
–
–
–
5.89
4
3
–
5.03[
c]
[
c]
[
10.34
11.28
–
H tetp found in the solid state wherein all the amines are Zn2+
[ML]/[M][L]
MHL]/[ML][H]
MH L]/[MHL][H]
[MH L]/[MH L][H]
18.24
7.74
7.04
5.73
–
17.48 18.16
[c]
[c]
[c]
[c]
18.31
[c]
8
[
c]
[c]
[
4.16
3.37
–
7.15
3.53
–
9.13
protonated and each phosphonate group contains only one
[c]
[c]
[
2
7.19
[
36]
6.66[
c]
proton.
3
2
[MH
4
L]/[MH
3
L][H]
–
–
5.50[
c]
[c]
[c]
[c]
[ML]/[MLOH][H]
10.13
10.80 11.21
11.09
Cd2+
[ML]/[M][L]
MHL]/[ML][H]
MH L]/[MHL][H]
[MH L]/[MH L][H]
17.31
8.75
18.0[b] 17.90
[c]
[c]
[c]
16.7[e]
The Thermodynamic Stability of the Metal Complexes
[b]
[
4.04
7.32
3.59
–
–
–
–
–
[b]
[
2
7.74
2.4
–
–
The complexation properties of H te3p1a with the ions
3
2
5.92
11.43
7
[c]
2
+
2+
2+
2+
2+
3+
3+
3+
3+
[ML]/[MLOH][H]
9.56
Ca , Cu , Zn , Cd , Pb , La , Sm , Gd , Ho and
Lu were studied by potentiometric titrations in an aque- Pb2+
ous solution at 298.2 K and at an ionic strength of
3
+
[ML]/[M][L]
15.37
10.55
7.13
14.3
[b]
b]
13.58
8.08
5.27
[c]
[c]
[c]
15.5[e]
[
[
MHL]/[ML][H]
MH L]/[MHL][H]
[MH L]/[MH L][H]
4.75
–
–
–
–
–
[b]
[
2
4.25
–
3
0
.10 moldm in [N(CH ) ]NO . The equilibria were at-
6.18
–
–
–
3.88[c]
–
3
+
4
3
2+
3
2
2
2+
tained quickly with Ca , Cd and Pb , and relatively
[MH
[ML]/[MLOH][H]
4
L]/[MH
3
L][H]
4.87
[c]
2
+
11.74
9.86
quickly with Cu ; hence, these systems could be studied by
La3+
[ML]/[M][L]
17.0
9.7
8.0
6.9
3.2
–
12.15[f] 12.07[g] 18.02[
[f]
conventional “in-cell” titrations. The equilibria with Zn2+
f]
[f]
[f]
[
MHL]/[ML][H]
MH L]/[MHL][H]
[MH L]/[MH L][H]
–
–
–
–
–
–
–
9.27
8.65
and especially with the lanthanides(III), however, were at-
tained slowly in the region of pH 7–10. For these systems,
it was thus necessary to perform “out-of-cell” titrations in
that pH region and the final equilibrium was attained after
ca. 3 weeks of incubation at 298.2 K. Furthermore, the sol-
ubility of the heavier lanthanide(III) complexes was im-
paired due to the formation of insoluble species in the
acidic pH region, which forced us to perform all titrations
with less than 0.9 equiv. of the metal ion. In the calculation
of the stability constants for most of the systems, data from
[
2
[g]
3
2
4.12
–
8.28[
f]
[
MH
MH
ML]/[MLOH][H]
4
L]/[MH
3
L][H]
6.32
f]
[
[
5
L]/[MH
4
L][H]
–
3.37[
[f]
[f]
[g]
9.5
7.58 10.9
10.64
Sm3+
[ML]/[M][L]
MHL]/[ML][H]
[MH L]/[MHL][H]
MH L]/[MH L][H]
MH L]/[MH L][H]
[MH L]/[MH L][H]
18.4
9.1
8.0
6.6
3.3
–
14.15[f] 14.38
[g]
19.11
9.63
8.58
[f]
[f]
[f]
[f]
[g]
[
–
–
–
–
–
6.68
[g]
2
6.44
4.02
–
[g]
[
[
3
2
7.77[
f]
4
3
6.15
3.35[f]
5
4
–
[f]
[g]
[f]
[ML]/[MLOH][H]
9.3
7.37 10.6
8.78
the “out-of-cell” titrations were fitted together with the data Gd3+
from the fully stabilised parts of the conventional titrations.
The calculated constants presented higher standard devia-
tion values, notoriously in the case of the lanthanides(III),
due to the lower precision of the “out-of-cell” measure-
ments and the smaller number of data points in the decisive Ho3+
13.77[h] 14.62
[g]
[ML]/[M][L]
19.1
9.4
–
–
–
–
–
–
[
g]
[
MHL]/[ML][H]
MH L]/[MHL][H]
MH L]/[MH L][H]
[MH L]/[MH L][H]
–
–
–
–
–
6.55
[g]
[
2
7.9
6.05
3.96
–
[g]
[
3
2
5.8
4
3
3.6
[
g]
[ML]/[MLOH][H]
10.9
10.3
15.78[f] 15.21
[g]
20.03
[f]
[ML]/[M][L]
19.5
9.0
7.7
6.1
2.7
–
[g]
[f]
[
MHL]/[ML][H]
MH L]/[MHL][H]
MH L]/[MH L][H]
[MH L]/[MH L][H]
MH L]/[MH L][H]
ML]/[MLOH][H]
–
–
–
–
–
6.57
5.75
3.98
–
9.52
pH region. The values of the stability constants of H te3p1a
7
[g]
[g]
[f]
[
2
9.25
and their respective standard deviations, as obtained from
fitting the experimental data, are collected in the Support-
ing Information (Table S2), together with the literature val-
ues for the related ligands, H teta, H te3a1p and H tetp.
[f]
f]
[
3
2
7.58
5.60[
4
3
3.10[
f]
[
[
5
4
–
[f]
[g]
[f]
9.9
7.03 10.5
10.09
4
5
8
Lu3+
[ML]/[M][L]
MHL]/[ML][H]
[MH L]/[MHL][H]
19.9
8.7
15.31[i] 15.49
[g]
[g]
[g]
–
–
–
–
–
–
The calculated stepwise constants for all of the ligands are
[
–
–
–
–
–
6.68
5.68
presented in Table 2.[
13,14,18,19,24,37–40]
2
8.5
[g]
[
MH
MH
ML]/[MLOH][H]
3
L]/[MH
2
L][H]
7.1
3.70
–
Under the experimental conditions used, only mononu-
clear complexes were found for all of the metal ions. It is
likely that the presence of excess metal ion could lead to
the formation of dinuclear complexes, especially in the case
[
4
L]/[MH
3
L][H]
3.8
[g]
[
11.0
10.4
[
a] The charges of the equilibrium species are omitted for clarity. [b]
[37]
[13]
Ref. [c] Ref. [d] The constant corresponding to the equilibrium
of the lanthanides(III) and copper(II). The experimental quotient [MHL]/[M][L][H], ref.[24] [e] Ref.[38] [f] Ref.[18,19] [g] Ref.[14]
limitations described above prevented us from using a [h] Ref.[ [i] Ref. with I = 0.20 moldm NaNO
39]
[40]
–3
.
3
Eur. J. Inorg. Chem. 2011, 527–538
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
531

L. M. P. Lima, R. Delgado, J. Plutnar, P. Hermann, J. Kotek
FULL PAPER
higher metal-to-ligand ratio to check the presence of such mation constants for the MHL species of these complexes
species. The value of their constants, however, should be are slightly higher than the first protonation constant for
quite low and should be similar to those determined for the free ligand phosphonate groups (8.08). This indicates
[41]
the transition metal complexes of H teta. In addition, for that a deprotonation of an amine group occurred, while the
4
most of the applications of this type of ligand, the metal values of the formation constants for the more protonated
complex has to be prepared with an excess of the ligand in species should correspond to deprotonations of the phos-
order to guarantee that full chelation is achieved. Besides phonate groups.
the fully deprotonated complex (ML), several protonated
The stability constants for the H te3p1a lanthanide(III)
7
species (MHL to MH L) exist at an intermediate pH and a complexes increase along the series, which has already been
4
hydroxo complex (MLOH), probably corresponding to the observed for other tetraazamacrocyclic ligands with acetate
[
19]
deprotonation of a coordinated water molecule, could be and methylphosphonate pendant arms.
found at alkaline pH. lower than for those for the H tetp complexes but are much
The values are
8
Overall, and with the exception of calcium(II) and cad- higher than those for the H teta complexes, which con-
4
mium(II), the stability constant values (for the formation of firmed the importance of the overall basicity of the ligand.
the ML species) showed that H te3p1a forms more stable The formation constants for the MHL and MH L species
7
2
complexes than H teta and complexes that are almost as are generally higher than those for the protonation of the
4
stable as those of H tetp. This could have been anticipated phosphonate pendants of the free ligand. Thus, we assumed
8
from the chemical structure of H te3p1a, which is more that the formation of these complexes occurred through the
7
similar to that of H tetp than to that of H teta, and is in generally accepted mechanism for the complexation of lan-
8
4
agreement with the assumption that the stability of this type thanide(III) ions by tetraazamacrocyclic ligands with pen-
of complex depends mainly on the basicity of the nitrogen dant arms containing oxygen donor atoms. In this mecha-
atoms of the macrocycle.[
3]
nism, the lanthanide(III) ions are initially captured by the
Interestingly, the stability of this calcium(II) complex is pendant groups and only bind to the amines in the ring
[
42]
lower than that of the H teta complex as the complex spe- upon their deprotonation.
Given the high coordination
4
cies started to form in alkaline pH and its abundance was demands of the lanthanide(III) ions, it is expected that all
only significant after the complete deprotonation of the li- of the pendant arms, as well as the macrocyclic amines of
gand. The formation constant values for the calcium(II) the ligand, will be involved in the coordination of such ions.
protonated species are lower than the first two protonation
The distribution diagrams of the representative species
constants for the amine groups of H te3p1a (above 12) but are presented in the Supporting Information (Figures S2 to
7
are higher than the highest protonation constant for the S9). The distribution diagrams show that, at physiological
phosphonate groups of the ligand (8.08). These values point pH (7.4), the copper(II), zinc(II) and cadmium(II) com-
to the deprotonation of the nitrogen atoms of the proton- plexes exist as mixtures of deprotonated, mono- and dipro-
ated complex species, which suggests that the metal ion is tonated species, while the lead(II) and lanthanide(III) com-
initially coordinated by the pendant arms only.
plexes exist mainly as mono- and diprotonated species. This
The stability constants for the copper(II) and zinc(II) corresponds to species bearing highly negative charges that
complexes are very close to those for H tetp, especially for are not very appropriate for biological use.
8
zinc(II), which has a very similar stability with all the li-
Assessment of the efficiency of the ligands in binding
gands discussed. The formation constants for the MHL metal ions by a simple comparison of the stability constant
species of these complexes are slightly lower than the first values could be misleading as they depend to a large degree
protonation constant ascribed to the phosphonate group of on the basicity of the ligand, as seen above. For a better
the free ligand (8.08), which indicates that deprotonation of comparison of the complexation behaviour of the various
a noncoordinated phosphonate group occurred. The forma- ligands, the pM values (–log[M]) for the H te3p1a com-
7
tion constants for the MH L and the MH L species, how- plexes were calculated. These take into account the proton-
2
3
ever, are slightly higher than those ascribed for the proton- ation constants and were calculated at pH 7.4 on the basis
ations of the other phosphonate groups of the free ligand of the global set of stability constants describing each sys-
(6.60 and 5.54, respectively). These unexpected protonation tem. These values, which express the amount of free metal
constants for the free ligand are lower, however, because ion in solution at this pH, are presented in Table 3, together
[
13,14]
of the existence of a hydrogen bond, as suggested above. with the literature values for the other ligands.
Therefore, the higher constants found for the protonated Several conclusions can be drawn from the calculated
complexes could be due to deprotonation of the noncoordi- pM values. For the calcium(II) complexes, most of the
nated phosphonate groups, which suggests that these com- metal ion is free in solution, which was expected since the
plexes are formed initially by coordination of the ring complex species are only formed in significant quantities
amines, and only one or more of the pendant arms are in- at a pH higher than 7.4. For the copper(II), zinc(II) and
volved in the deprotonation.
cadmium(II) complexes, H te3p1a seems to be less efficient
7
For the cadmium(II) and lead(II) complexes, the stability than the other ligands in most cases, which is probably be-
constants are close to those of the H tetp complexes, being cause there are more donor groups than the expected coor-
8
higher and lower, respectively, and they fit the trend of the dination number of these metal complexes. However, the
stability variation with the four ligands discussed. The for- copper(II) complex of H te3p1a has a pCu value much
7
532
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 527–538

A New Cyclam-Based Ligand
Table 3. The calculated pM[a] values for the metal complexes at pH
7.4.
Ion
Ca2
Cu
H
7
te3p1a
H
4
teta
H
5
te3a1p
[b]
8
H tetp
+
[b]
[b]
[b]
[b]
[b]
[d]
[d]
–[c]
5.00
13.71
7.96
8.28
7.82
9.24
9.97
10.79
10.69
11.60
5.00
14.20
10.61
11.15
7.45
6.52
8.71
5.01
14.79
11.46
11.26
7.45
5.67
7.56
7.78
8.37
8.67
2+
[b]
[b]
[b]
[b]
[d]
[d]
[d]
[d]
[d]
[b]
[b]
14.45
7.87
2
+
Zn
Cd
Pb
La
2
+
[c]
–
2+
[c]
–
3+
[d]
[d]
9.69
10.65
–
3
+
+
+
Sm
Gd
Ho
3
3
[c]
–
[d]
[d]
10.67
–
11.99
–
3
+
[c]
Lu
[
a] The values were calculated for 100% excess of the ligand and
were based on the stability constants collected in Table S2 at a
concentration of c
meaningful due to the incomplete set of constants available. [d]
= 1.00ϫ10–5 moldm . [b] Ref.
–3
[13]
[c] Not Figure 2. The 31P{ H} NMR spectrum of the Cd complex of
1
2+
M
H te3p1a at pH 9.9, with excess ligand present (the ligand peaks
are labelled with *).
7
[
14]
Ref.
closer to that of the other complexes, which is probably due
to the plasticity of this metal ion that allows it to adopt a
more distorted geometry. The pM values for the lead(II)
and lanthanide(III) complexes, however, were closer to that
The spectra of the zinc(II) complex proved to be more
complicated due to the possible presence of more than one
isomer in solution. Throughout most of the alkaline pH
range, the P{ H} spectra generally showed several broad
and unresolved resonance signals together with some
sharper ones. In order to elucidate the structure of the
31
1
of H tetp and higher than those for the other ligands, as
8
shown by their stability constants. These conclusions indi-
cate that the stability constant values alone must be ana-
lysed with care, and that any detailed comparison between
ligands of different basicity must be made with regard to
the pH range of interest. Ultimately, the pM values demon-
major species/isomer of the zinc(II) complex with H te3p1a
7
in solution, a series of NMR spectroscopic measurements
were performed on a sample prepared by prolonged heating
of the metal ion/ligand mixture in order to ensure that the
final complex isomer was present. The sample was prepared
at a pH of ca. 9.5 since at this pH the major species was at
maximum abundance (after heating the sample). The 1D
strated that H te3p1a is a very efficient ligand for the com-
7
plexation of all the metal ions studied, at physiological pH,
except for calcium(II).
1
13
1
31
H, C{ H} and P NMR spectra, together with the 2D
H,H-COSY, H,H-NOESY, H,C-HMBC and H,P-HMQC
Spectroscopic Studies of the Zinc(II) and Cadmium(II)
Complexes
correlations, were acquired and the 31P T /T relaxation
1
2
times were determined.
1
In an attempt to understand the coordination modes of
The 31P{ H} NMR spectra exhibit three major signals,
the metal complexes of H te3p1a, NMR spectroscopic which have relative intensities of 1:1:1 and chemical shifts
7
2
+
1
13
31
2+ 31
studies of the Zn ( H, C and P) and the Cd ( P) of 16.32, 16.86 and 17.09 ppm (Figure 3), together with
complexes were carried out in an alkaline aqueous solution some other minor signals, whose total intensity is ca. 15%
where the complexes are present mostly as the ML species. and that probably correspond to other isomers present in
The spectrum of the cadmium(II) complex that was ac- the solution. This pattern did not change with additional
quired at pH 9.9 (Figure 2) shows three well-resolved reso- heating of the solution. The relaxation time (Table 4) mea-
nances assigned to three nonequivalent phosphorus nuclei surements indicated that the signals at δP = 16.32 and
(at δ = 13.3, 13.8 and 14.7 ppm). All of the resonances are 16.86 ppm should correspond to phosphonate groups that
shifted relative to those of the free ligand. Additionally, all have a similar chemical environment whereas the third sig-
2
three resonances exhibit satellite signals ( JP,Cd values are nal exhibits significantly different values. This indicated that
2
0.2 Hz, 15.3 Hz and 13.8 Hz for the 14.7 ppm, 13.8 ppm one of the phosphonate groups is coordinated to the central
and 13.3 ppm resonances, respectively) that arise from the ion and the remaining two groups are not, or vice versa.
1
11/113
13
1
coupling to the
Cd isotopes. Therefore, only one con-
In the C{ H} NMR spectra, the carboxyl group signal
formational isomer of the complex was present in solution appears at δ = 178.7 ppm compared to 179.8 ppm for the
and it involved coordination of the metal ion to all three of free ligand at pH 8.5, which indicates a noncoordinated
the phosphonate groups. The coordination number of the acetate pendant arm. This is in agreement with the assump-
metal ion should be seven if all of the nitrogen atoms and tion that the carboxylate group is not coordinated in the
the three phosphonate pendant arms are coordinated. The presence of the more basic and fully deprotonated phos-
most probable conformation of the cyclam skeleton for this phonate pendant arms.
coordination mode is the type I conformation;[ the same
43]
The H NMR spectra shows a set of barely resolved sig-
1
one was also suggested as the most probable conformation nals, except for the protons of the –CH – groups of the
2
for the zinc(II) complex (see below).
pendant arms that are discussed in the following text. The
Eur. J. Inorg. Chem. 2011, 527–538
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
533

L. M. P. Lima, R. Delgado, J. Plutnar, P. Hermann, J. Kotek
FULL PAPER
very rare[2] and conclude that type I represents the structure
of the zinc(II) complex in solution.
Figure 3. The 31P{ H} NMR spectrum of the H
1
te3p1a zinc(II)
7
7
Figure 4. The isomers of the H te3a1p complexes.
complex at pH ≈ 9.5.
Table 4. The NMR relaxation times of the phosphorus atoms in
the H
te3p1a Zn2+ complex at pH ≈ 9.5.
Several coordination modes are still possible, however,
because any of the four pendant arms might be coordi-
nated. The species with three or four arms coordinated may
be excluded since coordination numbers higher than six for
zinc(II) complexes are very rare and thus only the isomers
where one or two pendant arms are involved in the coordi-
nation are possible. As stated above, coordination of the
acetate arm (bound to N1, see numbering in Scheme 3) was
7
Relaxation time NMR chemical shift of the phosphorus atom (ppm)
1
6.32
16.86
17.09
T
T
1
2
[s]
[ms]
0.66Ϯ0.01
23Ϯ2
0.63Ϯ0.04
19Ϯ4
0.77Ϯ0.01
48Ϯ2
presence of an AB spin system for the –CH COOH pen- excluded on the basis of its 13C NMR chemical shift; thus,
2
dant arm (3.24 and 3.42 ppm, JA,B = 16.5 Hz; confirmed only the phosphonate groups can be considered for coordi-
by the H,C-HMBC correlation spectra) and an ABX spin nation. The difference in the 31P relaxation times of the
system for the –CH PO H pendant arms (2.64 and phosphonate arms neighbouring the acetate group (δ =
2
3
2
2
.75 ppm, JA,B ≈ JA,X ≈ JB,X ≈ 14 Hz; 2.60 and 2.74 ppm, 16.86 and 17.09 ppm) indicates that one of them is coordi-
JA,B ≈ JA,X ≈ JB,X ≈ 14 Hz; 2.56 and 2.78 ppm, JA,B ≈ JA,X
≈
nated whereas the other one is not (the values of the 31
P
JB,X ≈ 14 Hz, determined from the H,P-HMQC correlation relaxation times for the free ligand are: T11 = 1.64Ϯ0.03,
spectrum, Figure S13 in the Supporting Information) indi- T12 = 1.33Ϯ0.03, T13 = 1.33Ϯ0.01 s, and T21 = 39.1Ϯ0.8,
cated the coordination of all of the nitrogen atoms bearing T22 = 45.5Ϯ1.0, T23 = 50.8Ϯ0.8 ms). We assumed that, for
the pendant arms. The proton signals for the acetate pen- steric reasons, simultaneous coordination of the neighbour-
dant arm show two sets of cross-peaks in the H,H-NOESY ing phosphonates (either bound to N4/N8 or N8/N11) was
spectrum. The signal at δ = 3.04 ppm shows interaction highly improbable for the type I isomer. As the simulta-
with signals at 2.87, 2.78, 2.75 and 2.68 ppm, and the signal neous coordination of the distant arms (bound to N4 and
at 3.24 interacts with the signals at δ = 3.09, 2.78 and N11) is excluded by the difference in their relaxation times,
2.75 ppm (Figure S12). The H,H-COSY spectrum indicates we concluded that only one of the pendant arms is coordi-
that the signals at δ = 2.87 and 2.68 ppm should be assigned nated to the central zinc(II) atom. The modelling results
to the protons of the propylenediamine fragment of the cy- indicated that the N4-bound phosphonate had to be coordi-
clam ring close to the acetate group and the signal at δ = nated as only then would the interactions that resulted in
3.09 ppm belongs to one of the protons of the ethylenedi- the observed NOE H,H cross-peaks be possible. If the N8-
amine fragment close to the acetate pendant arm. The sig- bound pendant arm was coordinated, the respective pro-
nals at δ = 2.75 and 2.78 ppm are assigned to the protons tons would be too far away from the acetate protons and
2
–
of the –CH PO
side arms: the former one to the phos- would be shielded by the oxygen atom of the phosphonate
2
3
phonate group at δ = 16.86 ppm and the latter one to that group.
P
at δ = 17.09 ppm. Contacts of the acetate arm with two
From the findings described above, it is evident that, un-
P
methylphosphonate groups were observed in the H,H- der the conditions of the measurements, the macrocyclic
NOESY spectra, which suggested that only type I or II framework is folded in the type I conformation and only
[
43]
structures can represent the complex in solution since in one of the pendant arms, the N4-bound methylphos-
the remaining types, either contacts to only one pendant phonate one, is coordinated. Thus, it can be suggested that
[43]
arm (type III and IV) or no contact to any pendant arm the coordination number of the central metal ion is five and
[
43]
(
type V) would be observed. The five possible spatial ar- the sixth possible coordination site remains unoccupied.
rangements of the pendant arms together with the chelate The simulated structure is depicted in the Supporting Infor-
ring conformations involving the macrocycle are shown in mation (Figure S14). This is the most common conforma-
Figure 4. Based on the literature data, we can exclude the tion for the complexes of N,NЈ,NЈЈ,NЈЈЈ-tetrasubstituted cy-
[
2,44]
arrangement represented by the structure of type II as it is clam derivatives,
and their divalent metal complexes
534
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 527–538

A New Cyclam-Based Ligand
presenting another ligand conformation, for example trans- perature of 298.2 K, unless otherwise noted. The chemical shifts
(
δ) are given in ppm and the coupling constants (J) in Hz. For the
III, are directly accessible only under special condi-
tions.
1
13
[
44–46]
H and C NMR spectra, tert-butyl alcohol was used as the in-
3
1
3 4
ternal reference, while for the P NMR spectra, H PO was used
as the external reference (inner-capillary method). The resonance
assignments are based on peak integration, peak multiplicity and
Conclusions
2D correlation experiments. The electrospray ionisation mass spec-
tra (ESI-MS) and the elemental analyses were carried out by the
ITQB Analytical Services.
A new macrocyclic ligand that combines methylphos-
phonate and acetate pendant arms, H te3p1a, was synthe-
7
sised from cyclam by a convenient triprotection methodol- 11-[(tert-Butyloxycarbonyl)methyl]-1,4,8-tris(trifluoroacetyl)-1,4,8,11-
tetraazacyclotetradecane (2): Compound 1 (2.44 g, 5 mmol) was
dissolved in acetonitrile (40 mL) and potassium carbonate (1.38 g,
ogy. The compound presented a very high overall basicity,
which was between that of H teta and H tetp, and was
closer to the latter ligand, as expected. The protonation of
4
8
1
0 mmol) was added. Tert-butyl bromoacetate (0.89 mL, 6 mmol)
was then added in one portion and the mixture was heated to ca.
33 K under a nitrogen atmosphere for 4 d. The mixture was evap-
H te3p1a proceeds in a very dynamic sequence with trans-
7
3
location of the protons between several basic sites and with
the protonated species stabilised by intramolecular hydro-
gen bonds, as demonstrated by 3 P NMR spectroscopy. The
orated to dryness and then dissolved in chloroform (100 mL), the
solids were filtered off and washed with chloroform (2ϫ50 mL).
The combined organic solutions were dried (anhydrous MgSO ),
4
1
ligand forms complexes of high thermodynamic stability filtered, evaporated to dryness and dried under vacuum to yield 2
with the lanthanide(III) and zinc(II) ions, but the formation in the form of a yellowish oil (2.6 g, 86%) that was judged pure by
f 3
of these complexes is rather slow. A fast complexation was TLC (R = 0.6 in CHCl /EtOAc, 1:1). ESI-MS: m/z = 546.6 [M –
+
+
+
tBu + 2 H] , 602.6 [M + H] , 624.9 [M + Na] .
observed for the copper(II) and calcium(II) complexes. The
copper(II) complex shows very high thermodynamic sta-
bility while that of calcium(II) exhibits a low stability. A
1
(
-[(tert-Butoxycarbonyl)methyl]-1,4,8,11-tetraazacyclotetradecane
3): Compound 2 (2.47 g, 4.1 mmol) was suspended in a solution
greater selectivity for copper(II) than for zinc(II) and the of potassium carbonate (400 mL, 10% w/v solution in methanol/
lanthanides(III) was evident from the large difference in the water, 5:2). The mixture was vigorously stirred in a closed flask at
values of their stability constants. The NMR spectra for the room temperature for 3 d, during which time the solids completely
dissolved. The solution was evaporated to dryness and then dis-
cadmium(II) and zinc(II) complexes, which present similar
solved in dichloromethane (100 mL), the solids were filtered off
and washed with dichloromethane (2ϫ50 mL). The combined or-
thermodynamic stability, indicated that while all three of
the phosphonate arms are coordinated to cadmium(II),
ganic solutions were dried (anhydrous MgSO
to dryness and dried under vacuum to yield 3 in the form of a
yellow oil (1.2 g, 93%) that was judged pure by TLC (R = 0.1 in
EtOH/25 % aq. NH , 10:1). ESI-MS: m/z = 258.8 [M – tBu +
4
), filtered, evaporated
only one methylphosphonate group is coordinated to
zinc(II) in the major isomer that was observed at a slightly
alkaline pH, which thus supports the pentacoordination of
the zinc(II) centre. It was shown that the acetate pendant
arm was not coordinated to zinc(II) in the complex and
should not be bound to the metal ion in the other divalent
metal ion complexes either, except in the case of the lead(II)
complex. Overall, the coordination ability of the new ligand
is quite good compared to related ligands, and the advan-
f
3
+
+
2
H] , 314.8 [M + H] .
1
1-[(tert-Butyloxycarbonyl)methyl]-1,4,8-tris[(diethoxyphosphoryl)-
methyl]-1,4,8,11-tetraazacyclotetradecane (4): Compound 3 (1.10 g,
.5 mmol) was suspended in triethyl phosphite (10.8 mL, 63 mmol)
3
and a small volume (ca. 2 mL) of dry chloroform was added to the
mixture to achieve complete dissolution. The solution was heated
tage of H te3p1a over H tetp is that it forms less charged
7
8
to ca. 333 K in a flask equipped with an anhydrous CaCl drying
2
complexes and has better solubility in the case of the lan- tube, paraformaldehyde (0.63 g, 21 mmol) was then added in small
thanide(III) complexes.
portions over 48 h, and the mixture was heated for an additional
8 h. The resulting solution was filtered, evaporated to dryness and
4
co-evaporated with toluene (4ϫ50 mL) to remove the volatiles. The
oil was dissolved in chloroform (50 mL) and the organic phase was
extracted with water (2ϫ25 mL) and then evaporated to dryness.
The product was dissolved in minimal water/ethanol (ca. 5 mL, 1:1)
and loaded onto a weak cationic exchange column (Amberlite CG-
50, 20ϫ2.5 cm, H+ form), which had been conditioned with the
water/ethanol (1:1) solvent system. The column was first eluted
with the previous solvent (300 mL) and then with a 25% acetic acid
solution in water/ethanol (1:1) (300 mL). After evaporation, the
neutral fractions contained only a small amount of the non-am-
monic byproducts, while the combined and vacuum dried acidic
fractions yielded 2.5 g of a yellow oil containing 4, which was con-
taminated with less than 20% of an aminal byproduct with only
two (diethoxyphosphoryl)methyl pendant arms. This product was
used directly in the next step without further purification. ESI-MS:
Experimental Section
General: 1,4,8,11-Tetraazacyclotetradecane (cyclam)[
47,48]
15]
and
(1)
1
,4,8-tris(trifluoroacetyl)-1,4,8,11-tetraazacyclotetradecane[
were synthesised by published methods. Paraformaldehyde was fil-
tered from aged solutions of formaldehyde and was dried in a desic-
cator with concentrated sulfuric acid. All the other chemicals were
used as obtained from the commercial sources. The organic sol-
vents were dried by standard methods.[ Thin layer chromatog-
raphy (TLC) was performed on silica gel 60 F254 aluminium sheets
and detection was by ninhydrin spray, Dragendorff’s reagent or by
49]
1
13
1
31
dipping in a 5% CuSO
4
aq. solution. The H, C{ H}, P and
31
1
P{ H} NMR spectra were recorded with Varian VNMRS300
1
UNITY
(
299.940 MHz for H), Varian INOVA
400 (399.951 MHz for
1
H), Bruker Avance III 400 (400.135 MHz for 1H) or Bruker
+
+ 31
m/z = 626.7 (C27
H
57
N
4
O
8
P
2
)
(byproduct), 764.5 [M + H] .
P
Avance III 600 equipped with an Inverse TCI 5 mm H/C/N/D
CryoProbe (600.174 MHz for H) spectrometers, at a probe tem-
NMR (162 MHz, CDCl
26.67 (s, 1P) ppm.
3
): δ = 24.56 (s, byproduct), 26.54 (s, 2P),
1
Eur. J. Inorg. Chem. 2011, 527–538
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
535

L. M. P. Lima, R. Delgado, J. Plutnar, P. Hermann, J. Kotek
FULL PAPER
+
. The term pH is defined as –log[H ]. EЈo and
+
[
4,8,11-Tris(phosphonomethyl)-1,4,8,11-tetraazacyclotetradec-1-
yl]acetic Acid (H te3p1a): The reaction mixture containing 4
2.40 g, ca. 2.5 mmol) was dissolved in aqueous HCl (20%, 50 mL)
Qϫlog[H ] + E
j
+
7
Q were determined by titration of a solution with a known H
(
concentration at the same ionic strength in the acidic pH region.
and the solution was heated under reflux for 48 h. After cooling,
the solution was filtered, evaporated to dryness and co-evaporated
with water (3ϫ50 mL) to remove the excess acid. The resulting oil
was dissolved in minimal water (ca. 5 mL), loaded onto a strong
cationic exchange column (Dowex 50WX4, 15ϫ3 cm, H+ form)
and eluted with water (300 mL) followed by a 5% aq. ammonia
solution (300 mL). The water fractions contained only inorganic
impurities, while the combined alkaline fractions contained the
macrocyclic compounds and were evaporated to dryness and co-
evaporated with water (3ϫ100 mL) to remove the excess ammonia.
The resulting oil was dissolved in water (ca. 5 mL), loaded onto a
weak cationic exchange column (Amberlite CG-50, 20ϫ2.5 cm, H+
form) and eluted with water (500 mL) in fractions of ca. 50 mL.
The fractions containing the desired compound were combined and
the solution was treated with charcoal (1 g), filtered, concentrated
to ca. 50 mL and added dropwise, with vigorous stirring, to anhy-
drous ethanol (250 mL). The white solid was filtered off, washed
The liquid-junction potential, E
j
, was found to be negligible under
+
–
the experimental conditions used. The value of K
w
= [H ]ϫ[OH ]
mol dm by titrating a solution
–
13.80
2
–6
was found to be equal to 10
with a known H+ concentration at the same ionic strength in the
o
alkaline pH region, considering EЈ and Q valid for the entire pH
range. The measurements during the conventional titrations were
carried out with ca. 0.04 mmol of the ligand in a total volume of
ca. 30 mL, both in the absence of the metal ions and in the presence
of each metal ion at ca. 0.9 equiv. relative to the ligand. For all
metal ions, except calcium(II), a back titration was performed at
the end of each direct titration in order to check if equilibrium
had been attained throughout the pH range. Each titration curve
typically consisted of 60–70 points in the 2.5 to 11.5 pH region
and a minimum of two replicate titrations were performed for each
system. In addition, the “out-of-cell” titrations were performed for
the zinc(II) and lanthanide(III) ion systems and consisted of 10 to
12 points in the 7–10 pH region. Each sample was prepared in
with ethanol (2ϫ50 mL) and dried under vacuum to yield 1.2 g of separate vials under the same conditions as for the conventional
H
7
te3p1a·2H
2
O (59% based on 3). C15
H
37
N
4
O
12
P
3
(576.41): calcd.
titrations but with a final volume 10 times smaller. After addition
of the base, the vials were tightly closed and kept at 298.2 K for
20 d, at which time they had attained equilibrium and the electro-
motive force was then measured with a Metrohm 6.0210.100 com-
bined pH electrode, which had been previously calibrated by an
C 31.26, H 6.82, N 9.72; found C 31.00, H 6.79, N 9.83. ESI-MS:
–
1
m/z = 539.2 [M – H] . H NMR (400 MHz, KOH in D
.4): δ = 1.95 (m, 2 H, CH -CH -CH ), 2.01 (m, 2 H, CH
CH HP = 11.3 Hz, 2 H, N-CH
), 2.89 (d, 2 -P), 2.96 (d, 2JHP
1.2 Hz, 4 H, N-CH -P), 3.01 (br. t, 2 H, N-CH -CH
t, 2 H, N-CH -CH ), 3.15 (m, 4 H, N-CH -CH ), 3.21 (m, 4 H, N-
), 3.30 (br. t, 2 H, N-CH -CH ), 3.33 (br. t, 2 H, N-CH
), 3.40 (s, 4 H, N-CH
-COOH) ppm. 13C NMR (101 MHz,
O, pD = 6.4): δ = 23.0 (s, 1C, CH -CH -CH ), 23.3 (s,
-CH -CH ), 50.0, 50.8, 50.9, 51.1, 51.4, 51.5, 51.8, 51.9,
2.0, 52.2, 52.3, 52.6, 53.5, 53.6 (N-CH -CH and N-CH -P), 57.7
-COOH), 177.0 (s, 1C, COOH) ppm. 3 P NMR
162 MHz, KOH in D O, pD = 6.4): δ = 11.52 (br. s), 11.76 (br.
2
O, pD =
6
H
2
2
2
2
-CH -
2
2
J
2
=
1
2
2
2
), 3.08 (br. “in-cell” titration.
2
2
2
2
NMR Spectroscopic Measurements: In order to determine the two
CH
CH
KOH in D
2
-CH
2
2
2
2
-
31
1
highest protonation constants of the ligand, the P{ H} NMR
spectra of H te3p1a in water were recorded at 298.2 K at 16 pH
2
2
7
2
2
2
2
points over the region of pH 9 to 14. A ligand solution was pre-
1C, CH
2
2
2
–3
pared at an ionic strength of 0.010 mol dm and the titrant, a
5
2
2
2
1
2
5% (w/w) aqueous solution of [N(CH
Aldrich and was standardised by titration with a 1.0 mol dm
HNO solution. The titration was carried out in a closed titration
3 4
) ]OH, was obtained from
(s, 1C, N-CH
2
–3
(
2
P
3
s), 13.95 (br. s) ppm. For the complete NMR assignment of all the
nuclei at pH 8.5, see the Supporting Information (Table S3).
cell and the titrant was added with a Crison microBU 2031 auto-
matic burette. The pH was measured with an Orion 420A measur-
ing instrument that was fitted with a Metrohm 6.0210.100 com-
bined electrode, which had been calibrated with commercial buffer
solutions with the standard pH (at 298.2 K) of 7.96 and 11.88.
Atmospheric CO was excluded from the cell during the titration
2
by passing purified nitrogen across the top of the experimental
solution. The measurement was carried out with 0.03 mmol of li-
Thermodynamic Stability Measurements
Preparation of the Reagents and Solutions: The purified water was
obtained from a Millipore Milli-Q demineralisation system. A
stock solution of
H
7
te3p1a was prepared at ca.
]NO was prepared by neutralisation
]OH solution with HNO . The metal ion
solutions were prepared in water at 0.025 to 0.050 moldm from
analytical grade nitrate salts of the metals [kept in excess HNO in
the case of the lanthanide(III) ions], and were standardised by ti-
.5ϫ10–3 mol·dm . [N(CH
–3
)
1
3
4
3
of a commercial [N(CH
3
)
4
3
gand in a total volume of ca. 5 mL and the ionic strength of the
–
3
–3
3 4 3
titration solution was kept at 0.50 moldm with [N(CH ) ]NO .
3
Following each addition of titrant, the pH was measured after
equilibration and a sample of the solution was placed in a 5 mm
NMR tube, which had been adapted with an internal capillary tube
2 2
H
edta.[ Carbonate-free solutions of the titrant
50]
tration with Na
N(CH ]OH were obtained at ca. 0.10 moldm by treating
freshly prepared silver oxide with a solution of [N(CH ]I under a
nitrogen atmosphere, were standardised by application of Gran’s
–3
[
3 4
)
2 3 4
containing D O and H PO for locking and referencing purposes.
)
4
31
1
3
After recording the P{ H} NMR spectra, the sample volume was
returned to the titration cell.
[
51]
method and were discarded as soon as the carbonate concentra-
tion reached ca. 1% of the total amount of the base. For the back
Calculation of Equilibrium Constants: The calculation of the overall
titrations, a 0.100 moldm–3 standard solution of HNO
pared from a commercial ampoule.
was pre-
equilibrium constants
β
M H L (where
β
m h l
M H L
=
[M
) was achieved
by fitting the potentiometric data from the titrations with the
m h l
H L ]/
m
h
l
3
m
h
l
[M] ϫ[H] ϫ[L] and βMH–1L = βML(OH) ϫK
w
HYPERQUAD program.[ The spectroscopic data from the
NMR titration in water were used to calculate the overall constant
for the two highest protonations of H te3p1a at ionic strength of
52]
31
P
Potentiometric Measurements: The potentiometric setup used for
conventional titrations has been previously described. The mea-
surements were carried out at 298.2 Ϯ 0.1 K in solutions with an
[13]
7
–
3
–3
[53]
ionic strength that was maintained at 0.10Ϯ0.01 moldm with
N(CH ]NO3. Atmospheric CO was excluded from the cell dur-
ing the titrations by passing purified N across the top of the exper-
imental solutions. The [H ] of the solutions was determined by
0.50 moldm by fitting the data with the HYPNMR program.
The obtained value was then extrapolated to an ionic strength of
[
3
)
4
2
–
3
[54]
2
0.10 moldm by using the Davies equation,
and this adjusted
+
value was used as a constant in the calculation of the remaining
protonation constants for the ligand. The species distribution dia-
o
measurement of the electromotive force of the cell, E = EЈ
+
536
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 527–538

A New Cyclam-Based Ligand
grams were plotted from the calculated constants with the HYSS
program. The difference, in logunits, between the values for the
protonated (or hydrolysed) and the non-protonated constants pro-
[6] X. Liang, P. J. Sadler, Chem. Soc. Rev. 2004, 33, 246–266.
[7] T. J. Wadas, E. H. Wong, G. R. Weisman, C. J. Anderson, Curr.
Pharm. Des. 2007, 13, 3–16.
[
55]
[
[
[
8] P. Hermann, J. Kotek, V. Kubí cˇ ek, I. Luke sˇ , Dalton Trans.
008, 3027–3047.
9] R. Delgado, V. Félix, L. M. P. Lima, D. W. Price, Dalton Trans.
007, 2734–2745.
10] W. A. Volkert, T. J. Hoffman, Chem. Rev. 1999, 9, 2269–2292.
vided the stepwise (logK) protonation constants (where KMmHhLl
=
2
[M
m
H
h
L
l
]/[M ]ϫ[H]). The errors quoted are the standard
m h–1 l
H L
deviations of the overall stability constants calculated by the fitting
program from all the experimental data for each system.
2
31
1
Spectroscopic Measurements:
A
P{ H} NMR titration of
[11] M. Woods, Z. Kovacs, A. D. Sherry, J. Supramol. Chem. 2002,
H
7
te3p1a in the region of pH –1 to 9 was performed in addition
2, 1–15.
to that described in the previous thermodynamic stability measure-
ments section (between pH 9–14). It was performed in a similar
way except that [N(CH ) ]OH and then conc. HNO aq. solutions
3 4 3
[12] S. Liu, Adv. Drug Deliv. Rev. 2008, 60, 1347–1370.
[
13] L. M. P. Lima, R. Delgado, M. G. B. Drew, P. Brandão, V.
Félix, Dalton Trans. 2008, 6593–6608.
14] L. M. P. Lima, R. Delgado, F. Marques, L. Gano, I. Santos,
Eur. J. Med. Chem. 2010, 45, 5621–5627.
[
were used as the titrants, without rigorous control over the volume
of the titrant that was added and the ionic strength. A solution of
2
+
[15] W. Yang, C. M. Giandomenico, M. Sartori, D. A. Moore, Tet-
rahedron Lett. 2003, 44, 2481–2483.
[
the H
7
te3p1a Cd
complex was prepared in water at ca.
–3
–3
4
ϫ10 moldm of the ligand in a 1:2 metal to ligand ratio for
16] P. Hermann, J. Kotek, Ten-membered Rings or Larger with One
or More Nitrogen Atoms, in: Comprehensive Heterocyclic Chem-
istry III (Eds.: A. R. Katritzky, C. A. Ramsden, E. F. V.
Scriven, R. J. K. Taylor), Elsevier, Oxford, 2008, vol. 14, pp.
613–666.
31
1
the P{ H} NMR studies. The pH was adjusted to 9.9 by the ad-
dition of dilute [N(CH ]OH and the solution was heated over-
night at 353 K. An internal capillary tube containing D O and
PO was used for locking and referencing purposes during the
acquisition of the spectra. The solutions of the H
complex for the multinuclear NMR studies were prepared in D
te3p1a (30 mg) was dissolved in D O (0.5 mL) and the pD was
adjusted to ca. 8 by means of a 2 moldm solution of KOH in
O. Zinc(II) chloride tetrahydrate (15 mg, 1.3 equiv.) was added
to the solution and the pD was adjusted to ca. 9.5 (KOH in D O).
3 4
)
2
H
3
4
2
+
7
te3p1a Zn
O.
[17] R. W. Alder, E. Heilbronner, E. Honegger, A. B. McEwen,
R. E. Moss, E. Olefirowicz, P. A. Petillo, R. B. Sessions, G. R.
Weisman, J. M. White, Z.-Z. Yang, J. Am. Chem. Soc. 1993,
115, 6580–6591.
2
H
7
2
–3
[18] R. Delgado, J. Costa, K. P. Guerra, L. M. P. Lima, Pure Appl.
D
2
Chem. 2005, 77, 569–579.
2
®
[19] F. Marques, L. Gano, M. P. Campello, S. Lacerda, I. Santos,
L. M. P. Lima, J. Costa, P. Antunes, R. Delgado, J. Inorg. Bio-
chem. 2006, 100, 270–280.
The solution was transferred to a Teflon -cap-sealed vial and
heated at 353.2 K for 3 d (until no further change in the 3 P NMR
pattern was observed).
1
[
20] L. Burai, J. Ren, Z. Kovacs, E. Brücher, A. D. Sherry, Inorg.
Chem. 1998, 37, 69–75.
21] J. Kotek, P. Vojtí sˇ ek, I. Císa rˇ ová, P. Hermann, P. Jure cˇ ka, J.
Rohovec, I. Luke sˇ , Collect. Czech. Chem. Commun. 2000, 65,
1289–1316.
22] P. Táborský, P. Lubal, J. Havel, J. Kotek, P. Hermann, I. Luke sˇ ,
Collect. Czech. Chem. Commun. 2005, 70, 1909–1942.
23] J. Havlí cˇ ková, H. Medová, T. Vitha, J. Kotek, I. Císa rˇ ová, P.
Hermann, Dalton Trans. 2008, 5378–5386.
Supporting Information (see footnote on the first page of this arti-
cle): Table of the overall protonation and stability constants for the
[
ligand H
7
te3p1a and its metal complexes, a colour version of the
31
P NMR spectroscopic titration of H te3p1a, the species distribu-
7
[
[
[
[
[
tion diagrams for the free ligand and its complexes, the H,P-
HMQC NMR spectrum as well as the signal assignments for the
1
H and C NMR resonances of the free ligand, the 1H, H,H-
13
7
NOESY and H,P-HMQC NMR spectra of the H te3p1a zinc(II)
24] R. Delgado, L. C. Siegfried, T. A. Kaden, Helv. Chim. Acta
complex and a representation of the suggested solution structure
1990, 73, 140–148.
for the [Zn(te3p1a)]5– complex.
25] S. Füzerová, J. Kotek, I. Císa rˇ ová, P. Hermann, K. Binnemans,
I. Luke sˇ , Dalton Trans. 2005, 2908–2915.
26] M. P. C. Campello, S. Lacerda, I. C. Santos, G. A. Pereira,
C. F. G. C. Geraldes, J. Kotek, P. Hermann, J. Van eˇ k, P. Lubal,
V. Kubí cˇ ek, É. Tóth, I. Santos, Chem. Eur. J. 2010, 16, 8446–
8465.
Acknowledgments
The authors acknowledge the financial support from Fundação
para a Ciência e a Tecnologia (FCT), with co-sponsorship from
the Fundo Europeu de Desenvolvimento Regional (FEDER), from
project PTDC/QUI/67175/2006, and a Ph. D. fellowship for
L. M. P. L. (SFRH/BD/18522/2004). We also thank the Grant Ag-
ency of the Czech Republic (203/09/1056) and the Ministry of Edu-
cation of the Czech Republic (grant number MSM0021620857).
The Analytical Services of ITQB-UNL are acknowledged for pro-
viding analytical data. The NMR spectrometers at ITQB are part
of the National NMR Network and were acquired with funds from
FCT and FEDER. This work was done in the framework of the
COST D38 and BM607 projects.
[27] R. D. Hancock, R. J. Motekaitis, J. Mashishi, I. Cukrowski,
J. H. Reibenspies, A. E. Martell, J. Chem. Soc. Perkin Trans. 2
1996, 1925–1929.
[
[
[
[
[
[
[
28] T. G. Appleton, J. R. Hall, A. D. Harris, H. A. Kimlin, I. J.
McMahon, Aust. J. Chem. 1984, 37, 1833–1840.
29] K. Popov, E. Niskanen, H. Rönkkömäki, L. H. J. Lajunen,
New J. Chem. 1999, 23, 1209–1213.
30] K. Popov, H. Rönkkömäki, L. H. J. Lajunen, Pure Appl. Chem.
2001, 73, 1641–1677.
31] A. Popov, H. Rönkkömäki, K. Popov, L. H. J. Lajunen, A.
Vendilo, Inorg. Chim. Acta 2003, 353, 1–7.
32] C. F. G. C. Geraldes, A. D. Sherry, W. P. Cacheris, Inorg. Chem.
1989, 28, 3336–3341.
33] K. P. Guerra, R. Delgado, L. M. P. Lima, M. G. B. Drew, V.
[
[
1] K. P. Wainwright, Coord. Chem. Rev. 1997, 166, 35–90.
2] M. Meyer, V. Dahaoui-Gindrey, C. Lecomte, R. Guilard, Co-
ord. Chem. Rev. 1998, 178–180, 1313–1405.
Félix, Dalton Trans. 2004, 1812–1822.
34] M. Försterová, I. Svobodová, P. Lubal, P. Táborský, J. Kotek,
P. Hermann, I. Luke sˇ , Dalton Trans. 2007, 535–549.
[3] I. Luke sˇ , J. Kotek, P. Vojtí sˇ ek, P. Hermann, Coord. Chem. Rev.
2
001, 216–217, 287–312.
[35] F. K. Kálmán, Z. Baranyai, I. Tóth, I. Bányai, R. Kíraly, E.
Brücher, S. Aime, X. Sun, A. D. Sherry, Z. Kovács, Inorg.
Chem. 2008, 47, 3851–3862.
[
[
4] C. J. Anderson, M. J. Welch, Chem. Rev. 1999, 99, 2219–2234.
5] S. Liu, D. S. Edwards, Bioconjugate Chem. 2001, 12, 7–34.
Eur. J. Inorg. Chem. 2011, 527–538
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
537

L. M. P. Lima, R. Delgado, J. Plutnar, P. Hermann, J. Kotek
FULL PAPER
[
36] T. Vitha, J. Kotek, J. Rudovský, V. Kubí cˇ ek, I. Císa rˇ ová, P.
[46] C. Bucher, E. Duval, E. Espinosa, J.-M. Barbe, J.-N. Verpeaux,
C. Amatore, R. Guilard, Eur. J. Inorg. Chem. 2001, 1077–1079.
[47] E. K. Barefield, F. Wagner, K. D. Hodges, Inorg. Chem. 1976,
15, 1370–1377.
[48] I. Meunier, A. K. Mishra, B. Hanquet, P. Cocolios, R. Guilard,
Can. J. Chem. 1995, 73, 685–695.
[49] D. D. Perrin, W. L. F. Armarego, Purification of Laboratory
Chemicals, 3rd ed., Pergamon, Oxford, 1988.
[50] G. Schwarzenbach, H. Flaschka, Complexometric Titrations,
Methuen & Co, London, 1969.
[51] F. J. Rossotti, H. J. Rossotti, J. Chem. Educ. 1965, 42, 375–378.
[52] P. Gans, A. Sabatini, A. Vacca, Talanta 1996, 43, 1739–1753.
[53] C. Frassineti, S. Ghelli, P. Gans, A. Sabatini, M. S. Moruzzi,
A. Vacca, Anal. Biochem. 1995, 231, 374–382.
[54] L. D. Pettit, I. Sukhno, V. Buzko, Aqueous Solutions program
suite, Academic Software, IUPAC, 2004.
Hermann, I. Luke sˇ , Collect. Czech. Chem. Commun. 2006, 71,
337–367.
[
[
[
[
[
[
[
37] G. Anderegg, F. Arnaud-Neu, R. Delgado, J. Felcman, K. Po-
pov, Pure Appl. Chem. 2005, 77, 1445–1495.
38] M. I. Kabachnik, Y. M. Polikarpov, Zh. Obsch. Khim. (USSR)
1
988, 58, 1937–1962.
39] E. T. Clarke, A. E. Martell, Inorg. Chim. Acta 1991, 190, 37–
6.
4
40] M. Kodama, T. Koike, A. B. Mahatma, E. Kimura, Inorg.
Chem. 1991, 30, 1270–1273.
41] S. Chaves, R. Delgado, J. J. R. Fraústo da Silva, Talanta 1992,
39, 249–254.
42] J. Moreau, E. Guillon, J.-C. Pierrard, J. Rimbault, M. Port, M.
Aplincourt, Chem. Eur. J. 2004, 10, 5218–5232.
43] B. Bosnich, C. K. Poon, M. L. Tobe, Inorg. Chem. 1965, 4,
1102–1108.
[55] L. Alderighi, P. Gans, A. Ienco, D. Peters, A. Sabatini, A.
Vacca, Coord. Chem. Rev. 1999, 184, 311–318.
[
[
44] E. K. Barefield, Coord. Chem. Rev. 2010, 254, 1607–1627.
45] A. Riesen, M. Zehnder, T. A. Kaden, Acta Crystallogr., Sect.
C 1991, 47, 531–533.
Received: August 27, 2010
Published Online: December 28, 2010
538
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 527–538