Synthesis, solution thermodynamics, and X-ray study of CuII [12]metallacrown-4 with GABA hydroxamic acid: An unprecedented crystal structure of a [12]MC-4 with a γ-aminohydroxamate

Article abstract of DOI:10.1002/chem.200601035

The solution equilibria of γ-aminobutanehydroxamic acid (GABAha) with H⁺ and Cu²⁺ were investigated by potentiometry, titration calorimetry, spectrophotometry, NMR spectroscopy, and ESI-MS. The thermodynamic parameters of the Cu^II [12]metallacrown-4 obtained for GABAha were compared with those of the corresponding complexes of (S)-α-Alaha and β-Alaha. The stability (-ΔG⁰) sequence was β-Alahaa ? α-Alaha > GABAha, whereas the order of formation enthalpies (-ΔH⁰) was β-Alaha ? GABAha > α-Alaha. These data were interpreted on the basis of the dimensions of the chelate rings and the planarity of the metallamacrocycles. The Cu^II [12]metallacrown-4 ([12]MC-4) complex of GABAha was isolated and its crystal structure, which is the first reported for a [12]MC-4 of a γ-aminohydroxamic acid, fully supports the structural features interpreted from the thermodynamic data.

Full text of DOI:10.1002/chem.200601035

DOI: 10.1002/chem.200601035
Synthesis, Solution Thermodynamics, and X-ray Study of CuII
[12]Metallacrown-4 with GABA Hydroxamic Acid: An Unprecedented
Crystal Structure of a [12]MC-4 with a g-Aminohydroxamate**
Matteo Tegoni,[a] Luca Ferretti,[a] Francesco Sansone,[b] Maurizio Remelli,[c]
Valerio Bertolasi,[d] and Francesco Dallavalle*[a]
Abstract: The solution equilibria of g-
aminobutanehydroxamic acid
(ꢀDG0) sequence was b-Alaha@a-
Alaha>GABAha, whereas the order
of formation enthalpies (ꢀDH0) was b-
the dimensions of the chelate rings and
the planarity of the metallamacrocy-
cles. The CuII [12]metallacrown-4
([12]MC-4) complex of GABAha was
isolated and its crystal structure, which
is the first reported for a [12]MC-4 of a
g-aminohydroxamic acid, fully supports
the structural features interpreted from
the thermodynamic data.
(GABAha) with H+ and Cu2+ were in-
vestigated by potentiometry, titration
calorimetry, spectrophotometry, NMR
spectroscopy, and ESI-MS. The ther-
modynamic parameters of the CuII
Alaha@GABAha>a-Alaha.
These
data were interpreted on the basis of
Keywords: aminohydroxamic acids ·
crown compounds · metallacycles ·
structure elucidation · thermody-
namics
[12]metallacrown-4
obtained
for
GABAha were compared with those of
the corresponding complexes of (S)-a-
Alaha and b-Alaha. The stability
Introduction
molecular squares,[3] metallacalixarenes,[4] metallahelicates,[5]
and metallacrowns[1,6]) resembles that of related organic
compounds (cryptates, cyclophanes, calixarenes, helicates,
and crown ethers, respectively), and is obtained by concep-
tually replacing the carbon atoms of the organic macrocycle
backbone with a metal–heteroatom coordination unit. Thus,
topological analogues of [12]crown-4 ligands ([12]C-4) can
be constructed, for example, by replacing the two methylene
Metallacrowns (MC) are a class of compounds belonging to
the large family of metallamacrocycles.[1] The molecular ar-
chitecture of metallamacrocycles (e.g., metallacryptates,[2]
[a] Dr. M. Tegoni, Dr. L. Ferretti, Prof. F. Dallavalle
Dipartimento di Chimica Generale ed Inorganica
Chimica Analitica, Chimica Fisica
ꢀ ꢀ ꢀ
groups of each macrocycle arm with M N coordination
units (Scheme 1 and ref. [1]). This structure, called [12]met-
allacrown-4 ([12]MC-4), is a 12-membered macrocycle capa-
ble of encapsulating a core metal ion into a four-oxygen-
atom cavity. While [12]MC-4 complexes generally contain
Cu2+ as a core cation,[1,7,8] for [15]MC-5 complexes, the five-
oxygen cavity is of suitable dimensions to encapsulate
Università degli Studi di Parma
Viale G.P. Usberti 17A, 43100 Parma (Italy)
Fax : (+39)0521-905-557
E-mail: francesco.dallavalle@unipr.it
[b] Dr. F. Sansone
Dipartimento di Chimica Organica e Industriale
Università degli Studi di Parma
Viale G.P. Usberti 17A 43100 Parma (Italy)
(III),[9] calcium(II),[10] or uranyl(II) ions.[10]
lanthanideACHTREUNG
[c] Prof. M. Remelli
Dipartimento di Chimica, Università di Ferrara
Via L. Borsari 46, 44100 Ferrara (Italy)
[d] Prof. V. Bertolasi
Centro di Strutturistica Diffrattometrica
Dipartimento di Chimica, Università di Ferrara
Via L. Borsari 46, 44100 Ferrara (Italy)
[**] GABA=g-aminobutanoic acid
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
Scheme 1. Macrocyclic backbone of the [12]crown-4 ether and its inor-
ganic analogue, [12]metallacrown-4.
1300
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 1300 – 1308

FULL PAPER
Several CuII metallacrown complexes have been synthe-
sized during the last decade by using aminohydroxamic
acids (HL), which can act as (O,Oꢀ)–(NH2,Nꢀ) bridging bis-
chelating ligands. In particular, [12]MC-4 complexes
[Cu5L4Hꢀ4]2+ were synthesized by using b-aminohydroxamic
acids (e.g., b-alanine- and b-phenylalaninehydroxamic
However, since 2001 it has been established that also a-
aminohydroxamates can form CuII [12]MC-4 complexes in
solution and
a tensioned-cup structure has been pro-
posed.[11,12] More recently, solid pentacopper complexes of
(S)-a-phenylalanine- and 2-picoline-hydroxamic acids were
isolated and characterized,[13] and metallahelicates contain-
ing 28 copper(II) atoms were synthesized with d- and l-nor-
valinehydroxamic acids and characterized by X-ray diffrac-
tion.[5] Moreover, we have demonstrated that also the (S)-
glutamic-g-monohydroxamic acid, which gives rise to seven-
membered (NH2,Nꢀ) chelate rings, can form a CuII [12]met-
allacrown-4 in solution.[14] These results demonstrate that
the formation of metallacrowns is also possible in solution
by using conceptual building blocks different to those ex-
pected by the “metallacrown structural paradigm”. In addi-
tion, recent papers have reported results of NMR studies on
the conversion of a-aminohydroxamates [12]MC-4 into
[15]MC-5 induced by lanthanide ions,[13,15] from studying in
detail the capability of an appropriate central metal ion to
rearrange the building blocks with an expansion of the met-
allacrown ensemble, as predicted by Pecoraro and co-work-
ers.[9] These data, together with thermodynamic studies,
have proved that a metallacrown is a real self-assembly, in
which the building blocks are only conceptually defined and
not present in solution prior to the metallacrown forma-
tion.[12,14] In other words, under favorable experimental con-
ditions, the metal ions and the ligand molecules can directly
arrange to form the supramolecular structure.
[Cu5L5Hꢀ5]]n+ were
acids), whereas [15]MC-5 complexes [MACHTREUNG
synthesized by using a-aminohydroxamates (e.g., a-alanine-
and a-phenylalaninehydroxamic acids) in the presence of
lanthanidesACHTREUNG
(III) or uranyl(II).[1] The structures of [12]metal-
lacrown-4 and [15]metallacrown-5 complexes with b- and a-
aminohydroxamates are presented in Scheme 2. Notably,
CuII metallacrowns are only structural, and not functional,
analogues of the parent organic compounds, because vacant
metallacrowns (with no metal ions in the cavity) have not
yet been isolated.[1]
Scheme 2. Structures of the CuII [12]metallacrown-4 (left) and [15]metal-
lacrown-5 (right) complexes with b- and a-alaninehydroxamate, respec-
tively (Mn+ = Ln3+, Ca2+, UO22+).
The aim of the present work was the evaluation of the
effect of the (Namine,Nhydroximate) chelate ring dimensions on
the thermodynamic parameters (DG0, DH0, DS0) of the
[12]MC-4 CuII complexes of pure a-, b-, and g-aminohy-
droxamic acids. For this purpose, we synthesized the g-ami-
nobutanehydroxamic acid (GABAha, HL) and performed a
complete characterization of its protonation and CuII-com-
plexation equilibria, by means of potentiometry, calorimetry,
spectrophotometry, 1H NMR spectroscopy, and ESI-MS.
The structural hypothesis put forward for the [Cu5L4Hꢀ4]2+
metallacrown found in solution was confirmed by X-ray dif-
fraction analysis on a single crystal.
b- and a-Aminohydroxamic acids satisfy the topological
requirements for the synthesis of planar [12]MC-4 and
[15]MC-5, respectively. The [12]MC-4 complexes are topo-
logically related to squares in which the metals (or the li-
gands) are at the vertices, whereas [15]MC-5 complexes are
related to pentagons. The theoretical internal angles at the
vertices in pentagons are 1088, and in squares, 908. By bi-
secting the chelate rings of b-Alaha, we obtain an angle
close to 908, whereas the same angle for a-Alaha is close to
1088 (Scheme 3 and ref. [1]). Thus, these units can be consid-
ered conceptual building blocks of the metallacrown ensem-
ble. In the case of a derivatives, the wider angle leads to a
15-membered ring to obtain a planar arrangement of the
structure. As a result, a [12]MC-4 contains both five- and
six-membered rings, whereas a [15]MC-5 contains only five-
membered rings. The many publications dealing with this
concept constitute the basis of this “metallacrown structural
paradigm”.[1]
Results and Discussion
Synthesis of g-aminobutanehydroxamic acid (GABAha):
For ligand synthesis, we followed a published procedure that
describes as starting point the preparation of N-Cbz-protect-
ed methyl g-aminobutanoate. This is then reacted with hy-
droxylamine and, finally, the Cbz group is removed by cata-
lytic hydrogenation in the presence of HCl.[16] Several at-
tempts to reproduce this synthesis were not completely suc-
cessful. In fact, in our hands, all of these steps gave yields
much lower than those reported, and difficulties were en-
countered in the purification of both intermediates and the
g-aminobutanehydroxamic acid hydrochloride.
Better results were obtained by exploiting a method
based on the use of trichlorotriazine (TCT) as a convenient
Scheme 3. Structures of the conceptual building blocks of metallacrowns
of b- (left) and a- (right) aminohydroximates.
Chem. Eur. J. 2007, 13, 1300 – 1308
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1301

F. Dallavalle et al.
coupling reagent to obtain hydroxamic acids of N-protected
amino acids, as described by Giacomelli and co-workers.[17]
Under these conditions, the reaction of 1 with O-benzyl-hy-
droxylamine resulted in the isolation of 2 in good yields
(Scheme 4, procedure a). A further improvement in the re-
CF3COOH) in 60–65% overall yield (Scheme 4 procedure
ii). Attempts to remove the protecting groups without acid
or in the presence of HCl led to the cyclization of the com-
pound or to hydrolysis of the hydroxamate function, respec-
tively.
Protonation equilibria of GABAha: The thermodynamic pa-
rameters (DG0, DH0, DS0) obtained for the protonation of
GABAha (HL) and for acetohydroxamic acid (Acha, HL)
are listed in Table 1. Acetohydroxamic acid was chosen as a
reference for a pure (O,Oꢀ) hydroxamate chelator: the pro-
tonation equilibria of this were reinvestigated and gave re-
sults in excellent agreement with recent literature data
(logK=9.30,
DH0 =ꢀ22.2 kJmolꢀ1,
DS0 =
98.3 JKꢀ1 molꢀ1).[19] The two protonation processes of the
GABAha anion (Lꢀ) overlap and cannot be distinguished
by potentiometry.[20]
To elucidate the protonation microequilibria, GABAha
1
was studied by conducting H NMR titrations in D2O solu-
tion. Three sets of chemical-shift values, corresponding to
the signals of the three methylene groups, were recorded at
appropriate pH values and were processed together in the
calculations. The resulting macroconstants (Table 1) are not
significantly different from those determined by potentiome-
try. Figure 1 shows the experimental chemical-shift values of
the three groups as a function of pH, superimposed on the
species-distribution diagram.
Scheme 4. i) Procedure a: NH2OBn hydrochloride, N-methylmorpholine,
DMAP, trichlorotriazine, dry CH2Cl2, 08C then RT, 48 h; Procedure b:
NH2OBn hydrochloride, N-methylmorpholine, ethylchloroformate, dry
CH2Cl2, 08C, 1 h; ii) H2 (1 atm), Pd/C (10%), CF3COOH, CH3OH, RT,
12h.
The chemical-shift values of the g-methylene group, being
the closest one to the amino group, were used to determine
the protonation microconstants. The cgp values (the molar
fraction of the ligand with the amino group in the protonat-
ed form) as a function of pH was computed by using the
equation of Szakµcs and Noszµl [Eq. (1)]:[21]
action of 1 to 2 was achieved by using ethylchloroformiate
as activating agent (Scheme 4, procedure b).[18] Due to the
shorter reaction time and the higher yields (>90%), this
second route proved to be particularly convenient. The sub-
sequent simultaneous removal of benzyl groups was per-
formed by using hydrogen gas (H2) and Pd/C (10%) in the
presence of trifluoroacetic acid to give 4-aminobutanehy-
droxamic acid as the trifluoroacetate salt (3, GABAha·
ꢀ1
dgꢀdgd
dgpꢀdgd
1 þ ðKa1ꢀKaAÞ½HŠ
cgp ¼
¼
ð1Þ
ꢀ1
ꢀ2
1 þ Ka1½HŠ þ Ka1Ka2½HŠ
Table 1. Overall thermodynamic parameters for protonation and Cu2+-complex formation of GABAha and Acha in aqueous solution. T=298.15 K, I=0.1m
(KCl). Standard deviations on the last figure are given in parentheses.
GABAha (HL)
Acha (HL)
Potentiometry/calorimetry
NMR
logb
Potentiometry/calorimetry
Species
logb
ꢀDG0 [kJmolꢀ1
]
ꢀDH0 [kJmolꢀ1
]
DS0 [JKꢀ1 molꢀ1
]
logb
ꢀDG0 [kJmolꢀ1
]
ꢀDH0 [kJmolꢀ1
]
DS0 [JKꢀ1 molꢀ1
]
HL
H2L+
s[a]
10.22(1) 58.3
18.92(1) 108.0
44.2(5)
67.6(5)
–
47(2)
136(2)
–
–
–
10.19(1)
18.95(2)
0.01
–
75
9.35(1)
–
1.13
–
53.4
22.4(2)
–
104.0(6)
–
–
–
–
–
–
–
–
1.78
–
174
–
–
–
–
U[b]
9.4”10ꢀ4
280
6.1”10ꢀ4
248
n[a]
156
[CuL]+
–
–
–
–
–
–
–
–
–
–
–
–
–
–
7.85(2)
14.01(5) 80.0
–
–
–
3.06
–
44.8
13.5(3)
27.0(3)
–
–
105(1)
178(1)
[CuL2]
[CuLH]2+
17.25(1) 98.4
33.28(3) 189.9
36.72(6) 209.6
2.89
–
60.2(3)
108.0(4)
93(2)
–
128(1)
275(1)
391(6)
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
[CuL2H2]2+
2+
[Cu5L4Hꢀ4
s[a]
]
–
–
–
–
–
U[b]
2.0”10ꢀ4
306
7.4”10ꢀ4
144
n[a]
275
–
49
[a] s=[ꢀwi(E0i ꢀEic)2/(nꢀm)]1/2 =sample standard deviation; wi =1/si2, in which si is the expected error on each experimental electromotive force value or chemi-
cal-shift value (Ei0); n=number of observations, m=number of parameters refined. [b] U=ꢀwi(Q0i ꢀQic)2, in which wi is the statistical weight (unitary in the pres-
ent work), and Q0i and Qci are the experimental and calculated heats, respectively.
1302
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 1300 – 1308

CuII [12]Metallacrown-4
FULL PAPER
KaH ½HLHŠ cH
ð5Þ
KT ¼
¼
¼
cA
KaA
½HLAŠ
in which cA and cH are the molar fractions of the HLA and
HLH species, respectively (Scheme 5). The KT constant indi-
cates that the hydroxamic group is the most acidic (cH =
0.877(3)), so that the 87.7% of the HL species is in the zwit-
terionic HLH form. It has been reported that the ammonium
group is the most acidic in a-Alaha, whereas the most acidic
group in b-Alaha is the hydroxamic function.[20] In
GABAha, as the length of the spacer between the two
groups increases, the pKa values become closer to those of
acetohydroxamic acid and an alkylamine (see below).
Figure 1. 1H NMR chemical-shift values of the methylene groups for the
Taking into account the protonation microequilibria, the
protonation enthalpies for the amino and hydroxamate
groups can be calculated by using the Equations (6) and (7):
titration of GABAha in D2O ([L]tot =1.66”10ꢀ2 m).
in which Ka1 and Ka2 are the dissociation macroconstants of
the ligand obtained by NMR titrations, [H] is the free-
proton concentration ([H]=10ꢀpH), dg is the observed chem-
ical shift of the methylene group in the g position at the con-
sidered pH value, and dgp and dgd are the corresponding
chemical-shift values for the g-methylene group in the H2L+
and Lꢀ species, respectively, as determined by using the
DH10 ¼ cADHH0 þ cHDHA0
DH02 ¼ DH0A þ DH0H
ð6Þ
ð7Þ
in which DH01 and DH02 are the experimental global protona-
tion enthalpies for Lꢀ and HL (Table 1), and DHA0 and DH0A
are the protonation enthalpies for the amino and hydrox-
HypNMR2004 program (dgp =3.031(6); ddg =2.579(8)).[22] KaH
,
AHCTREaUNG mate functions, respectively. The same equations can be
KaA, KaAH, KaHA, and KT are the dissociation microconstants,
as described in Scheme 5. The observed and expected chem-
used for the DS0A and DSH0 calculations. Both enthalpy and
entropy values for the protonation of the amino group
(DH0A =ꢀ47.6(7) kJmolꢀ1, DSA0 =40(3) JKꢀ1 molꢀ1) are in
good agreement with those reported for the amino group of
4-aminobutanoic
acid
(DH0 =ꢀ51.0 kJmolꢀ1,
DS0 =
30 JKꢀ1 molꢀ1).[23] Those of the hydroxamate group (DH0H =
ꢀ20.0(9) kJmolꢀ1, DS0H =95(3) JKꢀ1 molꢀ1) are close to the
values
of
Acha
(DH0 =ꢀ22.4 kJmolꢀ1
,
DS0 =
104.0 JKꢀ1 molꢀ1).
Cu2+/GABAha mononuclear complexes: Table 1 shows the
speciation model proposed to interpret the potentiometric
data concerning the Cu2+/GABAha system. This model con-
sists of three complexes: [CuLH]2+
, , and
[CuL2H2]2+
Scheme 5. Microscopic dissociation equilibria of GABAha. The pKa
values determined by H NMR titration are reported, with standard devi-
ations in parentheses.
1
[Cu5L4Hꢀ4]2+. Our proposed hypothesis for the coordination
of GABAha to Cu2+ involves an (O,Oꢀ) chelation in both
[CuLH]2+ and [CuL2H2]2+, which was confirmed by calori-
metric and spectroscopic data, as discussed below. A repre-
sentative distribution diagram is shown in Figure 2and the
calculated visible spectra of the complex species are report-
ed in Figure 3.
1
ical-shift values for the H NMR titration of GABAha, to-
gether with the calculated values of dp and dd for the three
methylene groups, are reported in the Supporting Informa-
tion. Nonlinear least-squares regression on the observed and
calculated cgp gave a pKaA =9.67(1). The other four micro-
constants were calculated by using Equations (2)–(5):
ESI-MS measurements confirmed the presence of
[Cu5L4Hꢀ4]2+ and [CuL2H2]2+ together with peaks attributa-
ble to the [CuL]+ and [CuL2H]+ ions (see Table 2). The
latter two species are most likely derived from the depro-
pKaH ¼ ꢀlog ðKa1ꢀKaA
Þ
ð2Þ
ð3Þ
AHCTREUNG
tonation of [CuLH]2+ and [CuL2H2]2+, respectively. If the
two [CuL]+ and [CuL2H]+ complexes are added to the spe-
ciation model previously employed to process the potentio-
metric data, only the former is successfully retained in the
refinement, with a negligible improvement of the curve fit-
ting. In the corresponding distribution diagram (not shown),
[CuL]+ is present in the pH range 4.8–6.5, (max. 30% if Cu/
ꢀ
ꢁ
Ka1Ka2
pKaAH ¼ ꢀlog
KaA
ꢀ
Ka1Ka2
ꢁ
pKaHA ¼ ꢀlog
KaH
ð4Þ
Chem. Eur. J. 2007, 13, 1300 – 1308
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1303

F. Dallavalle et al.
Calorimetric results reported in Table 1 show that the for-
mation of mononuclear complexes is both enthalpically and
entropically favored. The enthalpic effect for the reaction
Cu2+ +LH!
[CuLH]2+ can be estimated by subtracting the
ACHTREUNG
neat contribution due to the amino-group protonation
(ꢀ47.6 kJmolꢀ1, see above) from the DH0 value of
[CuLH]2+ (ꢀ60.2kJmol ꢀ1, Table 1). The calculated value
(ꢀ12.6 kJmolꢀ1) is in very good agreement with the enthal-
py of formation of [CuL]+ of Acha (ꢀ13.5 kJmolꢀ1,
Table 1), thus, confirming the (O,Oꢀ) chelation of GABAha
on Cu2+. The bis complexes, [CuL2H2]2+ of GABAha and
[CuL2] of Acha, give overall enthalpy of formation values
(ꢀ108.0 and ꢀ27.0 kJmolꢀ1, respectively) that are almost
double those for the corresponding 1:1 species (see Table 1),
supporting the (O,Oꢀ) chelation hypothesis for both ligands.
The same suggestion arises from the inspection of the
spectrophotometric parameters obtained for the protonated
complexes (lmax =778 and 664 nm for [CuLH]2+ and
[CuL2H2]2+, respectively, see Figure 3), which are in good
agreement with the literature data for ligands behaving as
simple hydroxamates (Acha, [CuL2]: lmax =653 nm;[24] (S)-N-
Figure 2. Representative distribution diagram of the system Cu2+
GABAha (Cu/L=1:2, [Cu2+ tot =3.0”10ꢀ3 m).
/
]
(3-hydroxycarbamoyl-propionyl)-proline,
[CuL]:
lmax =
760 nm; [CuL2]2ꢀ, lmax =656 nm[18]).
Cu2+/GABAha [12]MC-4 complex: Both potentiometric re-
sults (Table 1 and Figure 2) and ESI-MS spectra (Table 2)
agree with the formation of high amounts of the [12]metal-
lacrown-4 [Cu5L4Hꢀ4]2+ at pH>6.5. The ESI-MS spectra of
a Cu2+/GABAha solution at pH 8.10 were recorded at dif-
ferent cone voltages and are reported in the Supporting In-
formation. The most abundant ions are either [Cu5L4Hꢀ4]2+
or [[Cu5L4Hꢀ4]CF3COO]+ (m/z=391 or 894, respectively).
As the cone voltage increases from 30 to 200 V, a decrease
in the intensity of the doubly charged ion in favor of an in-
Figure 3. Calculated visible spectra for the CuII complexes of GABAha.
Table 2. Detected ESI-MS ions for Cu2+ complexes with GABAha. The
m/z values were calculated on the basis of the most abundant peak of the
multiplet.
creasing
intensity
of
the
single-charged
ions
[[Cu5L4Hꢀ4]CF3COO]+ and [[Cu5L4Hꢀ4]Cl]+ was observed.
No other significantly intense peaks were observed. Notably,
at a cone voltage as high as 200 V, the peak corresponding
to the metallacrown complex is still present, whereas signifi-
cant peaks attributable to fragments of the supramolecular
ensemble are absent.
Species
[[CuL]]+
m/z
Species
m/z
180
294
298
[[Cu5L4Hꢀ4]]2+
391
818
894
[[CuL2H2]CF3COO]+
[[CuL2H]]+
[[Cu5L4Hꢀ4]Cl]+
[[Cu5L4Hꢀ4]CF3COO]+
L=1:4). The [CuL]+ stoichiometry requires the deprotona-
tion of the NH3 group at pHꢁ5, and this can only happen
It is interesting to compare the thermodynamic parame-
ters referring to the GABAha metallacrown formation
(logb=36.72; DH0 =ꢀ93 kJmolꢀ1, DS0 =391 JKꢀ1 molꢀ1,
Table 1) with those of the corresponding compounds formed
by a-Alaha (logb=40.16;[12] DH0 =ꢀ85 kJmolꢀ1, DS0 =
484 JKꢀ1 molꢀ1),[25] and b-Alaha (logb=49.39;[12] DH0 =
ꢀ166 kJmolꢀ1, DS0 =388 JKꢀ1 molꢀ1)[25]. Logb values indi-
cate that the stability order for the [12]MC-4 is b-Alaha@a-
Alaha>GABAha, which is reflected in the starting forma-
tion pH of 2.5, 4.5, 6.5, respectively (ref. [12] and Figure 2).
Given the constant (O,Oꢀ) chelation enthalpy to Cu2+ for
all metallacrowns, the differences in ꢀDH0 values mainly
represent the strength of the (Namine,Nhydroximate) chelation
rings. In the case of a-Alaha, this is affected by the nonpla-
nar, tensioned structure of the metallamacrocycle. This de-
termines the ꢀDH0 sequence b-Alaha@GABAha>a-
+
if the proton is displaced by Cu2+. However, a tridentate be-
havior of the ligand with an additional coordination of NH2
is prevented by the shortness of the alkyl chain. On the
other hand, an alternative bidentate chelation mode with
formation of a (NH2,Nꢀ) seven-membered chelate ring is in-
consistent with the behavior of b-alaninehydroxamate, for
which a six-membered (NH2,Nꢀ) chelation was not ob-
served.[7,12] Moreover, if the spectrophotometric data are
processed with the inclusion of the [CuL]+ species, meaning-
less results (negative molar-absorbance values for
[CuL2H2]2+) are produced. In conclusion, only the specia-
tion model reported in Table 1 appears to be reliable. This
means that [CuL]+ and [CuL2H]+ ions observed in the ESI-
MS spectra are formed during the ionization process.
1304
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 1300 – 1308

CuII [12]Metallacrown-4
FULL PAPER
Table 3. Selected bond lengths [Š] and angles [8].
Alaha, which is different from the stability order. Absorp-
tion maxima of b-Alaha, GABAha, and a-Alaha metalla-
crowns (lmax =611,[12] 640, 648 nm,[12] respectively) corre-
spond to n˜max(d–d) =1.637, 1.563, and 1.543”104 cmꢀ1, and
represent a decreasing field strength to which less-negative
DH0 values (ꢀ166, ꢀ93, ꢀ85 kJmolꢀ1) correspond. This be-
havior is in agreement with that reported for CuII complexes
of tetraamines, both linear and cyclic, for which a linear re-
lationship between n˜max(d–d) and DH0 was found.[26]
ꢀ
ꢀ
Cu1 O1
1.932(2)
Cu1 O3
1.913(2)
ꢀ
ꢀ
Cu2 Cl1
2.774(1)
1.958(2)
Cu3 O1w
2.349(3)
1.928(2)
1.950(3)
1.961(3)
ꢀ
ꢀ
Cu2 O1
Cu3 O3
ꢀ
ꢀ
C)u2 O21.958C(u23
O4
ꢀ
ꢀ
Cu2 N3’
1.996(3)
2.021(3)
Cu3 N1
ꢀ
ꢀ
Cu2 N4’
.005(3)Cu3 N22
O1-Cu1-O1’
O1-Cu1-O3’
180.0
88.7(2)
O1-Cu1-O3
O3-Cu1-O3’
91.3(2)
180.0
X-ray data for [12]MC-4 of b-Alaha[7,25] and GABAha
(see below) demonstrated the planarity of both complexes;
therefore, the principal differences in their structures are
the dimensions of the peripheral (Namine,Nhydroximate) chelation
rings. Thus, the lower logb value determined for GABAha
[12]MC-4 reflects the weakness of its seven-membered rings
with respect to the six-membered rings of b-Alaha. By con-
trast, even if a tensioned-cup configuration was proposed
for the [12]MC-4 of a-Alaha,[12] accounting for its lowest for-
mation enthalpy, the formation of stronger five-membered
rings with a more favorable chelate effect accounts for the
higher stability of this metallacrown relative to that of
GABAha. Thus, the stability of the [12]MC-4 of a-Alaha is
determined by mainly entropic factors.
Cl1-Cu2-O1
Cl1-Cu2-O2
Cl1-Cu2-N3’
Cl1-Cu2-N4’
O1-Cu2-O2
O1-Cu2-N3’
O1-Cu2-N4’
O2-Cu2-N3’
O2-Cu2-N4’
N3’-Cu2-N4’
103.0(1)
91.3(1)
96.1(1)
88.2(1)
81.2(1)
88.9(1)
165.6(1)
164.8(1)
88.6(1)
99.0(1)
O1w-Cu3-O3
O1w-Cu3-O4
O1w-Cu3-N1
O1w-Cu3-N2
O3-Cu3-O4
O3-Cu3-N1
O3-Cu3-N2173.1(1)
O4-Cu3-N1
O4-Cu3-N293.8(1)
N1-Cu3-N298.7(1)
96.3(1)
94.7(1)
92.3(1)
88.8(1)
81.2(1)
85.8(1)
165.8(1)
encapsulated in the central cavity and is coordinated in a
square-planar mode to four hydroximate oxygen atoms. The
Cu2and Cu3 atoms display a square-pyramidal geometry in
which the fifth ligand consists of a Clꢀ anion and a water
O1w molecule, respectively. The Cu–O and Cu–N distances
are in agreement with those of other similar CuII metalla-
Crystal structure of [[Cu5L4Hꢀ4]Cl
ORTEP view of the CuII metallacrown [[Cu5L4Hꢀ4]Cl2-
(H2O)2] with GABAha is shown in Figure 4. Selected intera-
2ACHTREU(GN H2O)2]·9H2O: An
ACHTREUNG
tomic distances and angles are given in Table 3. The com-
macrocycles.[7,8,27–30]
ACHTREUNG
ꢀ
ꢀ
The Cu2 Cl1 and Cu3 O1w apical-bond lengths, 2.774(1)
and 2.349(3) Š, respectively, are significantly greater than
ꢀ
ꢀ
the average Cu Cl and Cu Ow bond lengths, 2.323 and
2.189 Š, respectively, in pentacoordinated CuII complexes.[31]
ꢀ
ꢀ
These data are consistent with the Cu Cl and Cu Ow
apical-bond elongation, due to the Jahn–Teller effect, ob-
served in CuII complexes.
In the crystal packing, the complexes are arranged in a
herring-bone mode (Figure 5) interacting with each other by
means of O1w···Cl hydrogen bonds (O1w···Cl1=
3.155(3) Š). The other water molecules surrounding the
metallamacrocycle are linked by hydrogen bonds with N, O,
Figure 4. An ORTEP view of the [[Cu5L4Hꢀ4]Cl2(H2O)2] complex of
GABAha showing the thermal ellipsoids at 40% probability level.
plex lies on a crystallographic center of symmetry situated
on the central Cu1 atom and assumes a “flattened chair”
conformation. The structure consists of 12chelate rings,
from which eight are five-membered, and the (NH2,Nꢀ) che-
lation leads to the formation of four puckered seven-mem-
bered rings at the corners of the metallacrown complex. The
binucleating anionic ligands connect the four peripheral CuII
cations through the hydroximate oxygen atoms and the
amino and hydroximate nitrogen atoms. The Cu1 cation is
Figure 5. Stick representation of the metallacrown units in the crystal,
projected down the crystallographic a axis. The neutral complexes are ar-
ranged in a herringbone mode and interact with each other by means of
O1w···Cl1 hydrogen bonds.
Chem. Eur. J. 2007, 13, 1300 – 1308
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1305

F. Dallavalle et al.
CH2COOH), 1.70–1.57 ppm (m, 2H; CH2CH2CH2); 13C NMR (75 MHz,
[D]CHCl3, 300 K): d=178.3 (s; COOH), 156.6 (s; CONH), 136.2(s; Ar),
128.4, 128.0 (d; Ar), 66.6 (t; PhCH2), 40.1 (t; NHCH2), 31.0 (t; CH2CO),
and Cl atoms, stabilizing the crystal packing. The remaining
uncoordinated water molecules are in disordered positions.
ꢀ
24.7 ppm (t; CH2CH2CH2); IR (KBr pellet): n˜ =3333 (N H), 1689 (C=
O(OH) and C=O
(urethane), overlapped), 1549 cmꢀ1 (dN H); MS (ESI):
ꢀ
m/z (%): 260 (85) [M+Na]+, 282 (100) [MꢀH+Na]+; elemental analysis
calcd (%) for C12H15NO4 (237.26): C 60.75, H 6.37, N 5.90; found: C
60.87, H 6.46, N 5.81.
Conclusion
Comparison of the thermodynamic parameters (DG0, DH0,
and DS0) of GABAha and b-Alaha [12]MC-4 complexes
showed that the overall stability of the metallamacrocycle of
the latter ligand is greatest, which is due to a more favorable
enthalpic effect, as the entropic contributions are similar.
This confirms the observations of Pecoraro and co-workers
that b-aminohydroamates are appropriate for the synthesis
of the most stable [12]metallacrown-4 compounds. However,
we have demonstrated that g- and a-aminohydroxamic acids
are also capable of forming CuII metallacrowns and we have
measured their thermodynamic properties in solution. Met-
allacrowns of g-aminohydroxamates are the less stable, al-
though their formation is slightly more exothermic than that
of a-aminohydroxamates. However, the stability of [12]MC-
4 with a-aminohydroxamates is higher than that of g-amino-
hydroxamates on the basis of a more favorable formation
entropy, determined in particular by a stronger chelate
effect for the a derivatives.
Synthesis of benzyl N-carbobenzyloxy-g-aminobutanehydroxamate (2):
Procedure a: Acid 1 (1.50 g, 6.33 mmol), N-methylmorpholine (2.15 mL,
1.96 mmol), dimethylaminopyridine (DMAP, catalytic), and O-benzylhy-
droxylamine hydrochloride (1.3 g, 8.23 mmol) were dissolved in dry
CH2Cl2 (20 mL) and the solution was cooled at 08C. 2,4,6-Trichloro-
[1,3,5]-triazine (cyanuric chloride, TCT, 0.580 g, 3.16 mmol) was added to
the mixture and the reaction was kept at RT for 48 h. The color of the so-
lution changed from bright- to pale-yellow. The reaction mixture was fil-
tered on Celite and washed with HCl (20 mL, 1m, 3”), 4% NaHCO3
aqueous solution (20 mL, 2”), and brine (20 mL). The organic phase was
dried with Na2SO4, was evaporated to dryness in vacuo, and the solid ob-
tained was triturated with ethyl ether to give the product as a white
powder in 70% yield (1.51 g). Procedure b: Ethylchloroformiate
(0.800 mL, 8.40 mmol) and N-methylmorpholine (0.922 mL, 8.39 mmol)
were added to a solution of acid 1 (2.00 g, 8.39 mmol) in dry CH2Cl2
(60 mL) at 08C and under a N2 stream. In a separate flask, N-methylmor-
pholine (1.016 mL, 9.24 mmol) was added to a suspension of O-benzylhy-
droxylamine hydrochloride (1.48 g, 9.24 mmol) in dry CH2Cl2 (20 mL) at
08C and under a N2 stream. The suspension obtained was added to the
reaction mixture and the flask was kept at 08C for 1 h. The solvent was
removed in vacuo and the crude material was dissolved in ethyl acetate
(200 mL). The organic phase was washed with HCl (100 mL, 1m, 1”),
4% NaHCO3 water solution (100 mL, 1”), and brine (100 mL, 2”), and
the solvent was removed in vacuo. The crude white solid obtained was
suspended in cold diethyl ether (10 mL) and the product was collected by
filtration in 92% yield. TLC (silica): CH2Cl2/methanol 9:1, Rf =0.75; m.p.
The crystal structure of the first CuII [12]MC-4 of a g-ami-
nohydroxamate is reported; it fully supports the structural
hypothesis proposed to interpret the thermodynamic data in
aqueous solution.
1
105.1–106.08C; H NMR (300 MHz, [D6]DMSO, 300 K): d=10.97 (s, 1H;
C(O)NHOBn), 7.32(m, 11H; Ar and CbzN H), 5.00 (s, 2H;
PhCH2OCO), 4.77 (s, 2H; PhCH2ONH), 3.01–2.93 (m, 2H;
OC(O)NHCH2), 1.96 (t, 2H; J=7.8 Hz, CH2CH2C(O)NH), 1.70–
1.58 ppm (m, 2H; CH2CH2CH2); 13C NMR (75 MHz, [D]CHCl3): d=
170.5 (s; CONHOBn), 157.1 (s; OC(O)NH), 136.5, 135.3 (s; Ar), 129.1,
128.5, 128.1, 128.06 (d; Ar), 78.0 (OCH2Ph), 66.8 (CH2OCO), 39.8 (t;
NHCH2), 30.2(t; CH2CO), 26.1 ppm (t; CH2CH2CH2); IR (KBr pellet):
Experimental Section
Materials and instrumentation: Acetohydroxamic acid was purchased
from Sigma (St. Louis, MO) and its purity was checked by potentiometry
and NMR spectroscopy. Dichloromethane was dried over molecular
sieves (4 Š). If not otherwise specified, all solvents were reagent grade
and were used without further purification. 1H NMR spectra were col-
lected by using a Bruker Avance 300 MHz spectrometer (partially deuter-
ated solvents were used as internal standards). FTIR spectra (4000–
400 cmꢀ1) were recorded by using a Nicolet Nexus FT spectrometer.
Electrospray mass spectra were recorded by using a single quadrupole
ZMD Mass Spectrometer (Micromass, Manchester, UK) fitted with a
pneumatically assisted electrospray probe. Data were processed by using
the spectrometer software (MassLynx NT Version 3.4). Elemental analy-
ses (C, H, N, S) were performed by using a Carlo Erba EA 1108 automat-
ed analyzer. Visible absorption spectra were recorded by using a Perkin–
Elmer Lambda 25 spectrophotometer. Enthalpy values were determined
by titration calorimetry using a Tronac model 450 isoperibol calorimeter
equipped with a 25-mL reaction vessel.
ꢀ
ꢀ
n˜ =3312(N H), 3222 (N H), 1685 (C=O (urethane)), 1653 (C=O (hy-
droxamate)), 1553 cmꢀ1 (dN H); MS (ESI): m/z (%): 365 (100)
ꢀ
[M+Na]+, 707 (10) [2M+Na]+; elemental analysis calcd (%) for
C19H22N2O4 (342.40): C 66.65, H 6.47, N 8.18; found: C 66.34, H 6.82, N
8.31.
Synthesis of g-aminobutanehydroxamic acid trifluoroacetate salt (3): Tri-
fluoroacetic acid (TFA, 0.400 mL, 0.006 mol) was added to a solution of
2 (1.70 g, 0.006 mol) in methanol (30 mL) in which Pd/C (10%, 300 mg)
was previously suspended under nitrogen. Hydrogenation was performed
with a Parr apparatus by using p(H2)=1 atm. After 2h, the catalyst was
filtered off, the solvent was evaporated in vacuo, and a slightly rubbery
product was isolated in 90% yield, corresponding to the trifluoroacetate
salt of 3. M.p. 85.2–86.98C; 1H NMR (300 MHz, [D6]DMSO, 300 K): d=
8.33 (brs, 1H), 2.76 (t, J=7.5 Hz, 2H; NHCH2), 2.04 (t, J=7.5 Hz, 2H;
CH2CO), 1.73 ppm (q, J=7.5 Hz, 2H; CH2CH2CH2); IR (KBr pellet):
Synthesis of N-carbobenzyloxy-g-aminobutanoic acid: Benzylchloro-
formiate (16.7 mL, 50% mol/mol solution in toluene, 0.05 mol) was
added to an aqueous solution (20 mL) of g-aminobutanoic acid (GABA,
4 g, 0.04 mol) and NaOH (3.1 g, 0.08 mol) cooled at ꢀ58C. The reaction
mixture was stirred at this temperature for 1 h and at RT for 3 h, then di-
ethyl ether (20 mL) was added and the organic phase was separated to
remove the excess benzylchloroformiate and toluene. The aqueous phase
was then acidified with HCl (30 mL, 1m) and extracted with diethyl ether
(20 mL, 3”). The combined organic phases were evaporated at reduced
pressure and the product was obtained without further purification as a
white solid in 80% yield. M.p. 63.5–66.38C; 1H NMR (300 MHz,
[D6]DMSO, 300 K): d=12.05 (s, 1H; OH), 7.32 (m, 6H; Ar and NH),
5.00 (s, 2H; PhCH2), 3.07–2.92 (m, 2H; NHCH2), 2.21 (t, J=7.2Hz, 2H;
n˜ =3175 (N H), 1642cm ꢀ1 (C=O); MS (ESI): m/z (%): calcd for
ꢀ
+
+
+
ꢀ
C4H10N2O2: 118 [M] ; found: 119 (100) [M+H] , 102(10) [ M+H OH] ,
+
ꢀ
86 (5) [M+H NHOH] ; elemental analysis calcd (%) for C6H11O4F3N2
(232.16): C 31.04, H 4.78, N 12.07; found C 31.22, H 4.75, N 11.99.
Synthesis of [[Cu5L4Hꢀ4]Cl
CuCl2·2H2O (0.63 mmol) was added with stirring to
GABAha (0.50 mmol) in water (10 mL). The solution was then neutral-
ized by adding KOH solution. Green crystals of the compound
2A
2A
=
GABAha): Solid
a
solution of
a
[[Cu5L4Hꢀ4]Cl HCTRE(UNG H2O)2]·9H2O suitable for X-ray analysis were obtained
by slow evaporation of the solvent. IR (KBr pellet): n˜ =1642cm ꢀ1 (C=O,
shoulder at 1638 cmꢀ1), 1557 cmꢀ1 (N H); MS (ESI): m/z (%): 818 (55)
ꢀ
1306
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 1300 – 1308

CuII [12]Metallacrown-4
FULL PAPER
[Mꢀ11H2OꢀCl]ꢀ; elemental analysis calcd (%) for C16H54Cl2Cu5N8O19
sorptivity (e) of the various species, by using the logb values obtained by
potentiometry as fixed parameters. The 1H NMR data were treated by
using the HypNMR2004 program.[22] Nonlinear least-squares regression
calculations were performed by using SPSS 14.[38] DH0 values were com-
puted from the calorimetric data by means of the least-squares program
DOEC,[39] which minimizes the function: U=ꢀwi(Q0i ꢀQic)2, in which wi
are statistical weights (unitary in the present work), and Q0i and Qic are
the experimental and calculated heats, respectively. A literature DHw0
value of 56.4 kJmolꢀ1 was used in the calculations.[40] Throughout, the
precision of each thermodynamic parameter was reported as the standard
deviation given by the corresponding least-squares program. This is
shown in parentheses as uncertainty on the last significant figure
(Table 1).
(1051.27): C 18.28, H 10.66, N 28.92; found C 18.52, H 10.45, N 28.59.
Potentiometric measurements: Stock solutions of GABAha (ca. 0.03m)
were prepared by weight and their titre was checked by potentiometric
titration with standard KOH solutions. Acetohydroxamic acid was dried
in vacuo and its solutions prepared by weight. KOH and HCl aqueous
solutions (ca. 0.2m) were prepared by diluting concentrated Merck Titri-
sol ampoules and were standardized by using the usual procedure of this
laboratory.[32] All solutions were prepared by using freshly boiled, doubly
distilled water. The titrations were carried out in aqueous solution at T=
298.15Æ0.1 K and I=0.1m (KCl) under a N2 stream, using 25-cm3 sam-
ples. The potentiometric apparatus for the automatic data acquisition was
described previously.[32]
A Hamilton combined glass electrode (P/N
238000) was employed and was calibrated in terms of [H+] concentration
by titrating HCl solutions with KOH (pH=ꢀlog[H+]), as reported previ-
ously.[32] The pKw value was 13.77(1). The protonation constants of
GABAha were determined by alkalimetric titration of four samples (4.7–
7.1”10ꢀ3 m). For the CuII complexation equilibria, eight titrations were
carried out by using ligand/copper molar ratios of 0.84:1 to 4:1 and [Cu2+
Crystal structure analysis: The crystal data of compound
[[Cu5L4Hꢀ4]Cl2(H2O)2]·9H2O of GABAha were collected at RT by using
a
Nonius Kappa CCD diffractometer with graphite-monochromated
MoKa radiation. The data sets were integrated by using the Denzo-SMN
package[41] and were corrected for Lorentz, polarization, and absorption
effects (SORTAV)[42]. The structures were solved by direct methods
(SIR97)[43] and were refined by using full-matrix least-squares with all
non-hydrogen atoms anisotropically and hydrogen atoms included on cal-
culated positions, riding on their carrier atoms. Some hydrogen atoms of
the water molecules were found in the difference Fourier maps and were
taken in fixed position during the refinements. The remaining unidenti-
fied hydrogen atoms were not included in the refinement process. The
water molecule O6w was disordered and was refined over two positions
with occupancies 0.5. All calculations were performed by using
SHELXL-97[44] and PARST[45] implemented in the WINGX[46] system of
programs. Crystal data are reported in Table 4. CCDC 608266 contains
the supplementary crystallographic data for this paper. These data can be
obtained free of charge from the Cambridge Crystallographic Data
Centre via www.ccdc.cam.ac.uk/data_request/cif.
]
ranging between 1.5 and 4.0”10ꢀ3 m. The pH range explored was 2.8–
tot
11.1. The protonation constants of acetohydroxamic acid were deter-
mined by alkalimetric titration of six samples (5.0”10ꢀ3 m). For the CuII-
complexation equilibria, four titrations were performed using ligand/
copper molar ratios of 1:6 and 1:7, with [Cu2+ tot =1.0”10ꢀ3 m. The pH
]
range explored was 3.5–6.5; at higher pH values, a precipitate was ob-
served.
Calorimetric measurements: Enthalpy values for both protonation and
Cu2+ complexation of the ligands were determined by titration calorime-
try at T=298.15Æ0.02K and I=0.1m (KCl). The measurements were re-
corded by titrating in duplicate 25-mL aliquots of solutions, of composi-
tion similar to those employed in potentiometry, with standard HNO3.
The reaction heats were corrected for nonchemical contributions[33] and
for the dilution heats computed from literature data.[34]
Spectrophotometric and ESI-MS measurements: CuII-complexation equi-
libria of GABAha were studied by conducting visible spectrophotometric
titrations by recording 17 spectra in the range 400–800 nm (Cu/L=1:2,
Acknowledgements
[Cu2+ tot =3.0”10ꢀ3 m, pH 2.6–10.1). Visible absorption spectra were re-
]
corded by using matched quartz cells of 5-cm pathlength and a 0.1m KCl
solution as a reference. The solutions examined were passed from the po-
tentiometric vessel to the thermostatted cuvette by means of a peristaltic
pump. Electrospray-assisted mass spectrometric studies were performed
on Cu2+/GABAha solutions at appropriate pH values between 3.5 and
Financial support by the University of Parma (FIL 2005) and the Univer-
sity of Ferrara (ex-60%) is gratefully acknowledged.
Table 4. Summary of crystal data, intensity measurements, and structure
refinement.
9.0 (Cu/L=1:2.5, [Cu2+ tot =2.3”10ꢀ3 m) without the addition of KCl, to
]
formula
MW
C16H36Cl2Cu5N8O10·9H2O
1051.27
0.55”0.36”0.10
monoclinic
C2/c
28.5625(7)
12.3687(3)
10.7403 (3)
96.1564(9)
3772.5(2)
4
avoid the formation of KCl clusters. Samples were analysed by direct in-
fusion at 10 mLminꢀ1. In the positive-ion mode, the conditions were: ES
capillary, 3.0 kV; cone, 25–200 V; extractor, 2 V; RF lens, 0.15 V; source
block temperature, 808C; desolvation temperature, 1308C; cone and des-
olvation gas (N2), 1.6 and 8 Lminꢀ1, respectively. Scanning was per-
formed from m/z=100–1800. In the negative-ion mode, the conditions
were: ES capillary, 3.08 kV; cone, 22 V; extractor, 4 V; RF lens, 0.26 V;
source block temperature, 808C; desolvation temperature, 1308C; cone
and desolvation gas (N2), 1.6 and 8 Lminꢀ1, respectively. Scanning was
performed from m/z=100–1800.
crystal size[mm3]
crystal system
space group
a[Š]
b[Š]
c[Š]
b[8]
V[Š3]
Z
1calcd [gcmꢀ3
]
1.851
1
1H NMR titration: H NMR titration of GABAha was performed by col-
q range for data collection[8]
limiting indices
3.2ꢂqꢂ28.0
ꢀ37ꢂhꢂ36
ꢀ16ꢂkꢂ16
ꢀ11ꢂlꢂ14
295
lecting 25 spectra of GABAha in D2O, at appropriate pH, at 298.15K. A
0.75-mL sample of 3 in D2O (1.66”10ꢀ2 m) was prepared by weight and
the pH* was adjusted within the range 7.5–12.1 by addition of small
amounts of concentrated NaOD (approximately 0.9m). No KCl was
added to the solution. A Mettler–Toledo U402M3-S7/200 glass microelec-
trode was calibrated by potentiometric titration in aqueous solution, as
described above. The pH-meter reading in D2O solutions (pH*) was con-
verted into pH as proposed in the literature.[35] NMR spectra were cali-
brated by using sodium [2,2,3,3-D4]-3-(trimethylsilyl)propionate (TSP).
T[K]
m[cmꢀ1
]
29.95
Tmin/Tmax
D1min; D1max [eŠ
0.310/0.744
ꢀ1.22; 1.33
4516
ꢀ3
]
independent reflections measured
observed reflections [Iꢂ2s(I)]
refinement against
final R index (observed reflns)
wR (all reflections)
3605
Calculations: Protonation and Cu2+ complexation constants for
GABAha and Acha were calculated from potentiometric data by using
the HYPERQUAD2003 program.[36] For each system, the data of differ-
jF2 j
0.0467
0.1426
ent titrations were treated together. Visible spectrophotometric data
S
1.034
[37]
were treated with the SPECFIT32program
to calculate the molar ab-
Chem. Eur. J. 2007, 13, 1300 – 1308
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1307

F. Dallavalle et al.
[24] E. Farkas, E. A. Enyedy, G. Micera, E. Garribba, Polyhedron 2000,
19, 1727–1736.
[25] F. Dallavalle, M. Tegoni, L. Ferretti, F. Sansone, M. Remelli, V. Ber-
tolasi, unpublished results.
[26] L. Fabbrizzi, P. Paoletti, A. B. P. Lever, Inorg. Chem. 1976, 15, 1502–
1506.
[27] M. Orama, H. Saarinen, J. Korvenranta, Acta Chem. Scand. 1994,
48, 127–133.
[28] J. J. Bodwin, V. L. Pecoraro, Inorg. Chem. 2000, 39, 3434–3435.
[29] Y. Song, J.-C. Liu, Y.-J. Liu, D.-R. Zhu, J.-Z. Zhuang, X.-Z. You,
Inorg. Chim. Acta 2000, 305, 135–142.
[30] D. Gaynor, Z. A. Starikova, S. Ostrovsky, W. Haase, K. B. Nolan,
Chem. Commun. 2002, 506–507.
[31] A. G. Orpen, L. Brammer, F. H. Allen, O. Kennard, D. G. Watson,
R. Taylor, J. Chem. Soc. Dalton Trans. 1989, S1–S83.
[32] F. Dallavalle, G. Folesani, R. Marchelli, G. Galaverna, Helv. Chim.
Acta 1994, 77, 1623–1630.
[1] a) V. L. Pecoraro, A. J. Stemmler, B. R. Gibney, J. J. Bodwin, H.
Wang, J. W. Kampf, A. Barwinski, Prog. Inorg. Chem. 1997, 45, 83–
177; b) J. J. Bodwin, A. D. Cutland, R. G. Malkani, V. L. Pecoraro,
Coord. Chem. Rev. 2001, 216–217, 489–516.
[2] M. S. Lah, B. R. Gibney, D. L. Tierney, J. E. Penner-Hahn, V. L. Pe-
coraro, J. Am. Chem. Soc. 1993, 115, 5857–5858.
[3] a) M. Fujita, J. Yazaki, K. Ogura, J. Am. Chem. Soc. 1990, 112,
5645–5647; b) D. H. Cao, K. Chen, J. Fan, J. Manna, B. Olenyuk,
J. A. Whiteford, P. J. Stang, Pure Appl. Chem. 1997, 69, 1979–1986.
[4] H. Rauter, E. G. Hillgeris, A. Erxleben, B. Lippert, J. Am. Chem.
Soc. 1994, 116, 616–624.
[5] J. A. Johnson, J. W. Kampf, V. L. Pecoraro, Angew. Chem. 2003, 115,
564–567; Angew. Chem. Int. Ed. 2003, 42, 546–549.
[6] M. S. Lah, V. L. Pecoraro, J. Am. Chem. Soc. 1989, 111, 7258–7259.
[7] B. Kurzak, E. Farkas, T. Gløwiak, H. Kozłowski, J. Chem. Soc.
Dalton Trans. 1991, 163–167.
[8] B. R. Gibney, D. P. Kessissoglou, J. W. Kampf, V. L. Pecoraro, Inorg.
Chem. 1994, 33, 4840–4849.
[9] A. J. Stemmler, T. W. Kampf, M. L. Kirk, B. H. Atasi, V. L. Pecor-
aro, Inorg. Chem. 1999, 38, 2807–2817.
[10] A. J. Stemmler, J. W. Kampf, V. L. Pecoraro, Angew. Chem. 1996,
108, 3011–3013; Angew. Chem. Int. Ed. Engl. 1996, 35, 2841–2843.
[11] F. Dallavalle, M. Tegoni, Polyhedron 2001, 20, 2697–2704.
[12] M. Careri, F. Dallavalle, M. Tegoni, I. Zagnoni, J. Inorg. Biochem.
2003, 93, 174–180.
[13] S. H. Seda, J. Janczak, J. Lisowski, Inorg. Chim. Acta 2006, 359,
1055–1063.
[14] M. Tegoni, F. Dallavalle, B. Belosi, M. Remelli, Dalton Trans. 2004,
1329–1333.
[33] L. D. Hansen, T. E. Jensen, S. Mayne, D. J. Eatough, R. M. Izatt, J. J.
Christensen, J. Chem. Thermodyn. 1975, 7, 919–926.
[34] V. B. Parker, Thermal Properties of Aqueous Uni-Univalent Electro-
lytes, Natl. Stand. Ref. Data Ser. (US Natl. Bur. Stand.) 2, US Gov-
ernment Printing Office, 1965.
[35] A. Krezel, W. Bal, J. Inorg. Biochem. 2004, 98, 161–166.
[36] P. Gans, A. Sabatini, A. Vacca, Talanta 1996, 43, 1739–1753.
[37] R. A. Binstead, B. Jung, A. D. Zuberbꢂhler, SPECFIT Global Anal-
ysis System, Version 3.0, Spectrum Software Associates, Marlbor-
ough (MA, USA), 2004.
[38] SPSS 14, SPSS Inc., Chicago (IL, USA), 2006.
[39] C. Rigano, E. Rizzarelli, S. Sammartano, Thermochim. Acta 1979,
33, 211–216.
[15] A. Pacco, T. N. Parac-Vogt, E. Van Besien, K. Pierloot, C. Gçrller-
Walrand, K. Binnemans, Eur. J. Inorg. Chem. 2005, 3303–3310.
[16] H. Hjeds, T. Honorꢁ, Acta Chem. Scand. B 1978, 32, 187–192.
[17] G. Giacomelli, A. Porcheddu, M. Salaris, Org. Lett. 2003, 5, 2715–
2717.
[40] A. Sabatini, A. Vacca, L. Bologni, J. Chem. Soc. Dalton Trans. 1981,
1246–1250.
[41] Z. Otwinowski, W. Minor in Methods in Enzymology, Vol. 276, Part
A (Eds.: C. W. Carter, R. M. Sweet), Academic Press, London, 1997,
pp. 307.
[18] M. Tegoni, F. Dallavalle, M. A. Santos, J. Inorg. Biochem. 2004, 98,
209–218.
[19] M. L. Senent, A. Nino, C. Munoz Caro, S. Ibeas, B. Garcia, J. M.
Leal, F. Secco, M. Venturini, J. Org. Chem. 2003, 68, 6535–6542.
[20] E. Farkas, T. Kiss, B. Kurzak, J. Chem. Soc. Perkin Trans. 2 1990,
1255–1257.
[21] Z. Szakµcs, B. Noszµl, J. Math. Chem. 1999, 26, 139–155.
[22] C. Frassineti, S. Ghelli, P. Gans, A. Sabatini, M. S. Moruzzi, A.
Vacca, Anal. Biochem. 1995, 231, 374–382.
[42] R. H. Blessing, Acta Crystallogr. Sect. A 1995, 51, 33–38.
[43] A. Altomare, M. C. Burla, M. Camalli, G. L. Cascarano, C. Giaco-
vazzo, A. Guagliardi, A. G. Moliterni, G. Polidori, R. Spagna, J.
Appl. Crystallogr. 1999, 32, 115–119.
[44] G. M. Sheldrick, SHELX-97, Program for Crystal Structure Refine-
ment, University of Gçttingen, Gçttingen (Germany), 1997.
[45] M. Nardelli, J. Appl. Crystallogr. 1995, 28, 659.
[46] L. J. Farrugia, J. Appl. Crystallogr. 1999, 32, 837–838.
[23] A. E. Martell, R. M. Smith, NIST Standard Reference Database 46,
Version 7.0, Gaithersburg (MD, USA), 2003.
Received: July 18, 2006
Published online: October 30, 2006
1308
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 1300 – 1308