Synthesis, characterization and crystal structure of triphenylphosphine copper(I) methylpyruvate thiosemicarbazones

Article abstract of DOI:10.1016/j.poly.2009.01.013

This paper reports the synthesis of a series of methylpyruvate thiosemicarbazone derivatives containing, on the terminal nitrogen, substituents of different nature and size and namely, ethyl, phenyl and methylphenyl. These ligands were reacted with bis(triphenylphosphine)copper(I) nitrate and acetate to produce the respective complexes: [Cu(PPh₃)₂(Et-Hmpt)]₂(NO₃)₂ (1), [Cu(PPh₃)₂(Ph-Hmpt)]NO₃ (2), [Cu(PPh₃)₂(MePh-Hmpt)]NO₃ (3), [Cu₂(O₂CCH₃)(Et-pt)(PPh₃)₂] · H₂O (4), [Cu(Ph-mpt)(PPh₃)] (5) and [Cu₂(MePh-mpt)₂(PPh₃)₂] (6). All of them were characterized by elemental analysis, IR, ¹H NMR, EPR spectroscopy and, for compounds 1, 2, 4, and 6, by X-ray crystallography. The characterization revealed that the coordinating behaviour of the ligands is influenced by a series of factors, predominant among which is the hard soft nature of the atoms involved in the interactions with the metal centre. The complexes obtained from the nitrate copper(I) salt are formed by cationic molecules with a nitrate as a counterion, while those derived from the acetate salt present deprotonated ligands and a few unexpected features. In particular, one of the compounds (4) is a mixed valence dinuclear complex with an acetate oxygen and the thiosemicarbazone sulfur acting as bridging between the two Cu(I) and Cu(II) ions. Another one (6) presents instead a Cu(I)-Cu(I) sulfur bridged binuclear cluster.

Full text of DOI:10.1016/j.poly.2009.01.013

Polyhedron 28 (2009) 1160–1168
Contents lists available at ScienceDirect
Polyhedron
journal homepage: www.elsevier.com/locate/poly
Synthesis, characterization and crystal structure of triphenylphosphine
copper(I) methylpyruvate thiosemicarbazones
a
Marisa Belicchi-Ferrari a, Franco Bisceglie a, Enrico Buluggiu b, Giorgio Pelosi a, , Pieralberto Tarasconi
*
a Dipartimento di Chimica Generale ed Inorganica, Chimica Analitica, Chimica Fisica, Viale GP Usberti 17/A, Università degli Studi di Parma, 43100 Parma, Italy
b Dipartimento di Sanità Pubblica – Sezione di Fisica, Via Volturno 39, Università degli Studi di Parma, 43100 Parma, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 2 July 2008
Accepted 23 January 2009
Available online 27 February 2009
This paper reports the synthesis of a series of methylpyruvate thiosemicarbazone derivatives containing,
on the terminal nitrogen, substituents of different nature and size and namely, ethyl, phenyl and meth-
ylphenyl. These ligands were reacted with bis(triphenylphosphine)copper(I) nitrate and acetate to
produce the respective complexes: [Cu(PPh3)2(Et-Hmpt)]2(NO3)2 (1), [Cu(PPh3)2(Ph-Hmpt)]NO3 (2),
[Cu(PPh3)2(MePh-Hmpt)]NO3 (3), [Cu2(O2CCH3)(Et-pt)(PPh3)2] ꢀ H2O (4), [Cu(Ph-mpt)(PPh3)] (5) and
[Cu2(MePh-mpt)2(PPh3)2] (6). All of them were characterized by elemental analysis, IR, 1H NMR, EPR
spectroscopy and, for compounds 1, 2, 4, and 6, by X-ray crystallography. The characterization revealed
that the coordinating behaviour of the ligands is influenced by a series of factors, predominant among
which is the hard soft nature of the atoms involved in the interactions with the metal centre. The com-
plexes obtained from the nitrate copper(I) salt are formed by cationic molecules with a nitrate as a coun-
terion, while those derived from the acetate salt present deprotonated ligands and a few unexpected
features. In particular, one of the compounds (4) is a mixed valence dinuclear complex with an acetate
oxygen and the thiosemicarbazone sulfur acting as bridging between the two Cu(I) and Cu(II) ions.
Another one (6) presents instead a Cu(I)–Cu(I) sulfur bridged binuclear cluster.
Keywords:
Copper
Thiosemicarbazone
Mixed-valence complex
Ó 2009 Elsevier Ltd. All rights reserved.
1. Introduction
copper in its lower oxidation state and, second, tertiary phosphines
are reported to enhance the solubility of Cu(I) thioamide com-
Thiosemicarbazones are being extensively studied for their bio-
logical and chemical properties. The former give them interesting
pharmacological properties (antiviral, antibacterial or antitumoral
activities) and the amount of recent literature concerning this sub-
ject is steadily increasing [1–10]. The chemical properties are, on
the other hand, more wanting, even though the nature of thiosemi-
carbazones as polydentate ligands makes them very versatile mol-
ecules. In spite of the wide literature reporting data concerning
copper(II) complexes with thiosemicarbazones, reports containing
structural considerations about copper(I) complexes are more lim-
ited [2–6,10,11–22].
plexes, that otherwise would be insoluble in many organic solvents
[3].
In the literature it is reported that, in their neutral form, biden-
tate thiosemicarbazones bind to a metal in their E configuration,
generally via the S donor atom (Scheme 1, Ia), while if the hydra-
zinic NH hydrogen is deprotonated they generally change into
the Z form and bind to the metal in N,S-chelating mode (Scheme
1, Ib) [3,5,10,11–22]. Recently we have noticed that, more than
the deprotonation, it is the nature of the ligand and the presence
or absence of soft ligand competitors that influence the coordina-
tion mode of these ligands [23].
Our research group has been working on the biological activity
of these compounds and recently we have started a research pro-
gramme aimed at understanding how the oxidation state of copper
influences the biological activity of these compounds. In this
framework, our attention is presently devoted to the synthesis of
copper(I) thiosemicarbazones and to their stabilization in aqueous
solution. We have started our experiments from bis(triphenylphos-
phine)copper(I) salts, namely from the nitrate and the acetate, for
two reasons: first, triphenylphosphine (tpp) is known to stabilize
In fact, acting on the solvent and on the copper salt counterions,
we have managed to impose a chelating behaviour (Z configura-
tion) to the reported ligands even though these were protonated.
In this paper we report our studies concerning the use of a
potentially terdentate S,N,O ligand, namely methylpyruvate thio-
semicarbazone [Hmpt], and the influence of the thiosemicarbazide
terminal nitrogen substituents on the coordination behaviour and
the stability of these compounds. To this scope we used a series of
previously synthesized derivatives of methylpyruvate thiosemicar-
bazone containing, on the terminal nitrogen, substituents of differ-
ent nature and size: ethyl, phenyl and methylphenyl (Scheme 2)
[24].
* Corresponding author. Tel.: +39 0521 905 420; fax: +39 0521 905 557.
E-mail address: giorgio.pelosi@unipr.it (G. Pelosi).
0277-5387/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.poly.2009.01.013

M. Belicchi-Ferrari et al. / Polyhedron 28 (2009) 1160–1168
1161
Z
E
S
R2
N
R1
R
R1
NH
S
N NH
O
S
N NH
O
- H+
NH
S
N
R3
N NH
N
M
NH
NH
N
N
R3
R
R2
S
O
M
O
O
O
Ia
Ib
Scheme 1. R1, R2 and R3 can be either H or organic radicals.
Et-Hmpt
Ph-Hmpt
MePh-Hmpt
Scheme 3. Schematic view of the used ligands.
H
N
H
N
2.3. Synthesis of the complexes
O
N
R1
To a hot stirred methanolic 40 mL solution of the proper ligand
was added an amount of solid [Cu(PPh3)2NO3] for 1 (0.502 g of the
ligand, 2.47 mmol, and 1.606 g of metal salt, 2.47 mmol), for 2
(0.610 g of the ligand, 2.43 mmol, and 1.580 g of metal salt,
2.43 mmol), for 3 (0.484 g of the ligand, 1.82 mmol, and 1.187 g
of metal salt, 1.82 mmol) together with additional 40 mL of meth-
anol. Similarly, to a hot stirred methanolic 40 mL solution of the
proper ligand was added an amount of solid [Cu(PPh3)2CH3COO]
for 4 (0.610 g of the ligand, 3.00 mmol, and 3.890 g of metal salt,
6.00 mmol), for 5 (0.480 g of the ligand, 1.91 mmol, and 1.238 g
of metal salt, 1.91 mmol), for 6 (0.505 g of the ligand, 1.91 mmol,
and 1.233 g of metal salt, 1.91 mmol) together with additional
40 mL of methanol. The mixture was allowed to reach the reflux
temperature. When the inorganic salt was completely dissolved,
the resulting solution appeared dark yellow and was left under
magnetic stirring for 2 h. The solution was then cooled down to
room temperature and by slow evaporation of the solvent, crystals
suitable for X-ray diffraction were obtained for complexes 1, 2, 4
and 6.
O
S
Scheme 2. Chemical drawing of Hmpt ligand (R1 = ethyl, phenyl and 3-
methylphenyl).
These ligands were then reacted with bis(triphenylphos-
phine)copper (I) nitrate and acetate and the complexes so ob-
tained, [Cu(PPh3)2(Et-Hmpt)]2(NO3)2 1, [Cu(PPh3)2(Ph-Hmpt)]NO3
2, [Cu(PPh3)2(MePh-Hmpt)]NO3 3, [Cu2(O2CCH3)(Et-pt)(PPh3)2] ꢀ
H2O 4, [Cu(Ph-mpt)(PPh3)] 5 and [Cu2(MePh-mpt)2(PPh3)2] 6, are
reported here. All of them have been characterized by elemental
analysis, IR, 1H NMR spectroscopy and compounds 1, 2, 4, and 6
also by X-ray crystallography. Compound 4 contains two copper
atoms with two totally different geometries suggesting, together
with the charge balance, the presence of two different oxidation
states. For this compound also an EPR spectroscopy analysis was
carried out.
2. Experimental
2.3.1. [Cu(PPh3)2(Et-Hmpt)]2(NO3)2
Yield: 2.01 g, 75% (based on metal). Mp: 166 °C. FT-IR (KBr,
cmꢁ1) 3443, m, 3214, m,
(NH); 3053, m, (CHaromatic); 2980, m,
(CHaliphatic); 1725, s, (CO); 1570, s, (CN); 1435, s, (NCS);
1384, s, (NO); 774, m, (CS). Anal. Calc. for C86H86N8O10P4S2Cu2:
C, 60.52; H, 5.08; N, 6.56; S, 3.76. Found: C, 60.52; H, 5.08; N, 6.56;
S, 3.76%. 1H NMR data (d, ppm; DMSO–d6): 1.14 (t, 3H, CH3CH2–),
2.12 (s, 3H, CH3CN–), 3.60 (m, 2H, CH3CH2–), 3.76 (s, 3H, CH3O–),
7.25 (t, b, 18H, PPh3 Hp + Ho); 7.37 (t, 12H, PPh3 Hm); 8.53 (bs,
1H, NH); 11.61 (s, 1H, C(S)NH).
1
2.1. Materials and techniques
m
m
[Cu(PPh3)2NO3] and [Cu(PPh3)2(CH3COO)] were prepared by
reduction of the corresponding Cu(II) salt using a fourfold excess
of PPh3 in CH3OH at reflux temperature for 1 and 3 h, respectively.
The white powder formed in the reaction flask was filtered to sep-
arate the product from Ph3PO and washed first with EtOH and then
with diethyl ether. Cu(NO3)2 ꢀ 3H2O and Cu(CH3COO)2 ꢀ H2O were
procured from Carlo-Erba, while PPh3 was purchased from Aldrich.
C, H, N, S analyses were obtained with a Carlo-Erba 1108 instru-
ment. IR spectra were recorded using KBr pellets on a Nicolet
5PC FT-IR spectrophotometer in the 4000–400 cmꢁ1 range. 1H
NMR spectra were recorded on a Bruker AC300 spectrometer at
300 MHz in d6–DMSO with TMS as the internal reference. EPR
spectra, both at room and liquid nitrogen temperature, were re-
corded in phase-sensitive detection by a X-band (9-GHz) Varian
V-4502 spectrometer with a 100 kHz field modulation.
m
m
m
m
m
m
2.3.2. [Cu(PPh3)2(Ph-Hmpt)]NO3
Yield: 1.69 g, 77% (based on metal). Mp: 168 °C. FT-IR (KBr,
cmꢁ1) 2970, mw,
(CHaliphatic); 1731, s, (CO); 1563, s, (CN);
1435, s, (NCS); 1384, s, (NO); 743, m, (CS). Anal. Calc. for
2
m
m
m
m
m
m
C47H43N4O5P2SCu: C, 62.62; H, 4.81; N, 6.21; S, 3.55. Found: C,
62.54; H, 4.91; N, 6.52; S, 2.83%. 1H NMR data (d, ppm; DMSO–
d6): 2.25 (s, 3H, CH3CN–), 3.75 (s, 3H, CH3O–), 7.25 (t, b, 18H,
PPh3 Hp + Ho); 7.37 (t, 12H, PPh3 Hm); 7.50–7.72 (m, 5H, aromatic
ligand H), 10.05 (bs, 1H, NH); 11.46 (s, 1H, C(S)NH).
2.2. Preparation of the ligands
Methylpyruvate thiosemicarbazones were prepared by conden-
sation of methylpyruvate with N1-substituted thiosemicarbazides
in a 1:1 molar ratio in methanol using the procedure reported in
the literature [24]. In this way methylpyruvate-N1-ethylthiosemic-
arbazone (Et-Hmpt ꢀ H2O, used to prepare complexes 1 and 4),
methylpyruvate-N1-phenylthiosemicarbazone (Ph-Hmpt, used to
2.3.3. [Cu(PPh3)2(MePh-Hmpt)]NO3
3
Yield: 1.20 g, 72% (based on metal). Mp: 183 °C. FT-IR (KBr,
cmꢁ1) 3052, m,
(CO); 1567, s,
m
(CHaromatic); 2970, mw,
m
(CHaliphatic); 1720, s,
(NO); 744, m,
m
m
m
(CN); 1435, s, (NCS); 1384, s, m
m
(CS). Anal. Calc. for C48H45N4O5P2SCu: C, 62.98; H, 4.95; N, 6.12;
S, 3.50. Found: C, 63.16; H, 4.87; N, 6.45; S, 2.85%. 1H NMR data
(d, ppm; DMSO–d6): 2.18 (s, 3H, CH3CN–), 2.31 (s, 3H, CH3Ph),
3.76 (s, 3H, CH3O–), 7.25 (t, b, 18H, PPh3 Hp + Ho); 7.37 (t, 12H,
PPh3 Hm); 7.50–7.70 (m, 5H, aromatic ligand H), 10.11 (bs, 1H,
NH); 11.40 (s, 1H, C(S)NH).
prepare complexes
2
and 5) and methylpyruvate-N1-(3-
methyl)phenylthiosemicarbazone (MePh-Hmpt, used to prepare
complexes 3 and 6) were synthesised. Scheme 3 reports the struc-
ture of the used ligands.

1162
M. Belicchi-Ferrari et al. / Polyhedron 28 (2009) 1160–1168
Table 1b
Selected crystal data.
2.3.4. [Cu2(O2CCH3)(Et-pt)(PPh3)2] ꢀ H2O 4
Yield: 5.38 g, 75% (based on metal). Mp: 178 °C. FT-IR (KBr,
cmꢁ1) 3442, m,
(CN); 1567, m,
m
m
(NH); 1650, s,
(CO) coordinated carboxylates; 695, m,
m
(C@O) pyruvate; 1618, m,
m
Compound
(4)
(6)
m
(CS).
Formula
Molecular mass
Space group
a (Å)
C44H44Cu2N3O5P2S1
915.90
P1
C120H116Cu4N12O8P4S4
2360.53
P1
Anal. Calc. for C44H44N3O5P2SCu2: C, 57.70; H, 4.84; N, 4.59; S,
3.50. Found: C, 57.11; H, 4.76; N, 4.44; S, 3.50%.
ꢀ
ꢀ
11.841(3)
12.276(3)
16.896(4)
72.40(2)
79.65(3)
68.23(2)
2167.8(9)
2
944
1.402
1.15
0.71069
11.561(3)
19.518(3)
26.054(5)
94.61(1)
96.22(1)
90.10(1)
5825(2)
2
2448
1.346
0.91
0.71069
b (Å)
c (Å)
2.3.5. [Cu(Ph-mpt)(PPh3)] 5
Yield: 0.72 g, 65% (based on metal). Mp: 158 °C. FT-IR (KBr,
a
(°)
cmꢁ1) 3400, mb,
m
(NH); 3048, m, (CHaromatic); 1735, m,
m (CO);
b (°)
c
(°)
1564, m,
m
(CN); 746, ms, (CS). Anal. Calc. for C29H27N3O2PSCu:
m
V (Å3)
C, 60.46; H, 4.72; N, 7.29; S, 5.57. Found: C, 59.98; H, 4.96; N,
7.01; S, 5.23%. 1H NMR data (d, ppm; DMSO–d6): 2.20 (s, 3H,
CH3CN–), 3.75 (s, 3H, CH3O–), 7.25 (t, b, 9H, PPh3 Hp + Ho); 7.37
(t, 6H, PPh3 Hm); 7.55–7.72 (m, 5H, aromatic ligand H), 10.08 (bs,
1H, NH).
Z
F(000)
Dcalc (Mg/m3)
l
(mmꢁ1
)
k (Å)
Radiation
Mo K
a
Mo Ka
h range(°)
hkl range
3–23.29
3–23.31
2.3.6. [Cu2(MePh-mpt)2(PPh3)2] 6
ꢁ13, 13;ꢁ13, 13; ꢁ18, 18 ꢁ12, 12; ꢁ21, 21; ꢁ28, 28
Yield: 0.68 g, 60% (based on metal). Mp: 197 °C. FT-IR (KBr,
Crystal size (mm)
No. of measured
reflections
No. of unique reflections
No. of refined parameters
0.1, 0.1, 0.4
17480
0.2, 0.2, 0.5
44871
cmꢁ1) 3400, mb,
m
(NH); 3025, m, (CHaromatic); 1715, m,
m (CO);
1610, m,
m
(CN); 742, ms, (CS). Anal. Calc. for C60H58N6O4P2S2Cu2:
m
6228
514
16694
1225
0.93, ꢁ0.30
C, 61.06; H, 4.95; N, 7.12; S, 5.43. Found: C, 61.18; H, 4.87; N, 7.13;
S, 5.73%. 1H NMR data (d, ppm; DMSO–d6): 2.21 (s, 3H, CH3CN–),
2.35 (s, 3H, CH3Ph), 3.80 (s, 3H, CH3O–), 7.25 (t, b, 9H, PPh3
Hp + Ho); 7.35 (t, 6H, PPh3 Hm); 7.55–7.70 (m, 5H, aromatic ligand
H), 10.17 (bs, 1H, NH).
Max. and min. resolutions 0.75, ꢁ0.38
F map (e Åꢁ3
||Fo| ꢁ |Fc||/
D
)
R =
R
R
|Fo|
0.0467
0.1429
0.0650
0.1431
Rw2
Weighting scheme
1/[
r
2(F2o) + (0.0963P)2]
1/[r
2(F2o) + (0.0383P)2]
P = (max(F2o,0) + 2F2c )/3.
2.4. X-ray crystallography
Relevant data concerning data collection and details of structure
refinement are summarized in Tables 1a and 1b. Intensity data for 1
first and the SADABS [25] procedure for 4 and 6. The crystals of com-
pounds 4 and 6 gave very poor diffraction and data could be col-
lected only up to 23.30° of theta. The structures were solved
using direct methods (SIR-97 [26]). Refinements were carried out
by full matrix least-squares cycles SHELXL97 [27] for all compounds.
Most hydrogen atoms were calculated with standard geometry and
refined in riding position. The program PARST [28] was employed for
the geometrical description and PLUTON [29] and MERCURY [30] for the
drawings.
were collected on a CAD4 diffractometer with Cu K
those of 2 on a Philips PW1100 and those of 4 and 6 on a SMART
1000 Bruker AXS diffractometers with Mo K radiation. Data from
a radiation,
a
crystals of compound 1 were affected by twinning and were not
corrected for absorption, while for compounds2, 4 and 6 absorption
corrections were carried out using the psi-scan procedure for the
Table 1a
Selected crystal data.
3. Results and discussion
Compound
(1)
(2)
3.1. Synthesis and spectroscopy
Formula
Molecular mass
Space group
a (Å)
b (Å)
c (Å)
C86H86Cu2N8O10P4S2
C47H43Cu1N4O5P2S1
901.39
P21/n
14.205(4)
32.117(8)
9.952(2)
90
90.40(2)
90
4540(2)
4
1872
1.319
0.647
0.71069
Mo Ka
3–30.05
ꢁ19, 19; ꢁ45, 45; 0, 14
0.3, 0.4, 0.5
26205
1706.71
Pca21
19.057(4)
18.376(3)
24.365(8)
90
90
90
8532(4)
4
3552
Ligands were synthesized according to the procedure reported
in [24] and complexes by reacting these ligands with bis(triphen-
ylphosphine)copper(I) nitrate and with bis(triphenylphosphine)
copper(I) acetate. Both ligands and complexes were produced in
good yields. In complexes 1, 2 and 3, obtained from the nitrate salt,
the ligand is protonated while in complexes 4, 5 and 6, that are ob-
tained from the acetate salt, it is deprotonated. The syntheses were
carried out in methanol in order to deter possible problems arising
from transesterification or hydrolysis processes that could modify
the ligands. Compound 4 was isolated with difficulty and only after
many attempts. The best conditions were found using a 2:1 ratio of
ligand with respect to the metal content. Moreover, in comparison
with the syntheses of the other complexes, compound 4 was ex-
posed to atmospheric oxygen and moisture more than the other
ones being recalcitrant to crystallize. The higher acetate content
stabilizing the anionic form of the ligand, together with the in-
creased acidity of the hydrazine nitrogen due to copper(I) oxidised
to copper(II), promoted the hydrolysis of the methyl group from
the ester improving the electrophilicity of the carboxyl carbon
and speeding up the process of demethylation.
a
(°)
b (°)
c
(°)
V (Å3)
Z
F(000)
Dcalc (Mg/m3)
1.329
2.28
1.54184
l
(mmꢁ1
)
k (Å)
Radiation
h range (°)
hkl range
Crystal size (mm)
No. of measured reflections
No. of unique reflections
No. of refined parameters
Cu Ka
3–67.02
0, 21; 0, 20; ꢁ27, 27
0.4, 0.5, 0.5
12779
12772
867
13240
560
0.62, ꢁ0.77
Max. and min. resolutions
map (e Åꢁ3
||Fo| ꢁ |Fc||/
D
F
0.55, ꢁ0.43
)
R =
R
R
|Fo|
0.0678
0.1811
0.0473
0.1532
Rw2
Weighting scheme
1/[r
2(F2o) + (0.125P)2]
1/[r
2(F2o) + (0.0741P)2]
The 1H NMR spectrum of complex 1 exhibits the NHC@S signal at
about 11.40 ppm, showing that the ligand is present in its neutral
P = (max(F2o,0) + 2F2c )/3.

M. Belicchi-Ferrari et al. / Polyhedron 28 (2009) 1160–1168
1163
form and the structure of the complex in the solid state persists in
DMSO solution; the deshielded NHCH2CH3 signal at 8.53 ppm upon
complexation can be related to the change from an E to a Z confor-
mation of the ligand and the coordination to the metal of the S and
N3 atoms of the ligand [24]. Even for complexes 2 and 3 the signal
attributed to NHC@S and the deshielded resonance of the NHR nu-
cleus reveals that the ligands are in their neutral form and in a Z
conformation showing that the three complexes behave in a similar
way. The 1H NMR spectra of complexes 4 and 5 show the lack of the
signal of the NHC@S group confirming that in solution the ligand is
deprotonated. The resonances at ca 10.10 ppm, downfield with re-
spect to the resonances of the same NH group in the free ligand,
show that also these compounds are in a Z conformation. Due to
its paramagnetism, it was not possible to record the 1H NMR spectra
of complex 4. As regards as IR spectra, the absorptions in the region
3510–3150 cmꢁ1 usually attributed to
m(OH) and m(NH) stretching
frequencies are assigned with difficulty, because of numerous
hydrogen bonding interactions involving amino and imino groups
(NHꢀꢀꢀS, NHꢀꢀꢀONO2ꢁ and NHꢀꢀꢀOH2). No relevant shifts are observed
by the comparison of the vibrational stretching C@O bonds of com-
plexes with the absorption of the free ligands [24], showing that the
metal centre does not bind via pyruvate oxygen. An exception is re-
ported for complex 4, where the C@O absorption at 1650 cmꢁ1 (ca
70 cmꢁ1 at lower frequencies with respect the free ligand) shows
that the C@O group is involved in coordination. For all complexes
m
(C@N) and m(C@S) are significantly shifted at lower frequencies
showing that every ligand bind the metal ion via N,S. For complexes
1, 2 and 3 the nitrate group vibration can be easily find at
1384 cmꢁ1 with a strong absorption, while for complex 4, the typ-
ical stretching bands (symmetrical and asymmetrical) for CH3COOꢁ
are identified. In agreement with diffraction and/or NMR data, no
counterion can be observed in the IR spectra of compounds 5 and 6.
An electron paramagnetic resonance study was also performed
for compound 4 in order to provide information about the elec-
tronic structure of the metal and its immediate surrounding. The
sample was in polycrystalline form because we experienced some
difficulty to find single crystals of dimensions large enough to give
an acceptable signal to noise ratio. The resulting derivative absorp-
tions, with superimposed a DPPH (diphenyl-b-picrylhydrazyl)
marker (g = 2.0036), are shown in Fig. 1a and b. The two spectra
are quite similar, apart from the observation at the liquid nitrogen
temperature of a more intense signal with some emphasization of
a number of not completely resolved weak peaks of a possible
superhyperfine nature, see below. The increase of the peak-to-peak
intensity in decreasing temperature is usual in EPR, reminiscent of
the fact that a simple isolated paramagnetic centre follows the 1/T
Curie’s law.
Fig. 1. Derivative X-band EPR spectra of a polycrystalline sample of 4 at room
temperature (a) and at T = 77 K (b). The observed intensities of the signals a and b
were in the approximate ratio 1:4.
Assuming complete axial symmetry, we estimate:
g? ꢄ 2:065 ꢂ 0:007 and jA j 6 30 ꢃ 10ꢁ4 cmꢁ1
ð3Þ
?
Although in general the measured g values may not coincide
with the local single-ion values if exchange is present, here we
have no doubt about this because the 1 point symmetry makes
the two copper(II) per unit cell magnetically undistinguishable.
The g values are strongly indicative of a magnetically isolated
Cu2+ ion with unpaired spin in a substantially dx2
ground state
ꢁy2
orbital. In this case, in fact, it should roughly be [31]:
g== ¼ g0 ꢁ 8
2k=D1
g? ¼ g0 ꢁ 2
2k=D2
a
;
ð4Þ
a
;
Both resonance signals appear typical for a Cu2+ ion in octahe-
dral field with tetragonal distortion and show a well resolved four
lines hyperfine pattern in the low field range. They may be de-
scribed by a spin Hamiltonian of the form:
where g0 is the free electron g value 2.0023, k is the spin-orbit
coupling constant for the free copper ion with a vale close to
ꢁ828 cmꢁ1, and D1 and D2 stand for the optical energy splittings
due to the crystal-field between the ground state and the dxy and
dxz,yz orbitals, respectively.
$
$
~
~
~
~
H ¼
l
BS ꢀ g ꢀH þ S ꢀ A ꢀI; with S ¼ 1=2; ICu ¼ 3=2:
ð1Þ
The reducing a2 factor measures the fraction of time spent by
the unpaired electron at the copper ion and is directly related to
the covalence of the bonds [32–34]. A light anisotropy of this factor
arising from the excited orbitals dxy and dxz,yz in (Eq. (4)) has been
neglected for simplicity.
Here the first term represents the Zeeman interaction of the
copper unpaired spin S = 1/2 with the applied static magnetic field
H (lB = Bohr magneton) and the second one describes the hyper-
fine interaction of the unpaired electron with the copper nucleus
(I = 3/2)$.
$
The value jA==j appears somewhat low as compared with usual
axial values for copper(II) complexes, frequently found in the range
150–200 ꢃ 10ꢁ4 cmꢁ1. The reduced value has possibly to be related
to the strong spin delocalization due to the Cu(II)–S very covalent
bond. In fact, this shouldresult in two distinct effects. A low presence
of the unpaired spin at the copper nucleus with reduction of both the
dipolar and Fermi-contact terms of the hyperfine interaction, and an
The g and A tensors are approximately axial with maximum
values:
g== ¼ 2:274 ꢂ 0:002 and jA==j ¼ ð104 ꢂ 2Þ ꢃ 10ꢁ4 cmꢁ1
ð2Þ
However, the presence of a rhombic contribution is also evi-
dent, more pronounced in the high temperature spectrum, but
the overlap of various non-resolved hyperfine peaks prevents a
precise determination of the in-plane constants.
induced in-plane rhombic distortion with mixing between the dx2
ꢁy2

1164
M. Belicchi-Ferrari et al. / Polyhedron 28 (2009) 1160–1168
Table 2
and d3z2
[34].
orbitals which should also reduce the axial dipolar term
ꢁr2
Selected parameters of the hydrogen bonds involving the nitrate ions in compound 1.
As a further experimental fact, we observe that the powder
spectrum at liquid nitrogen temperature shows the presence of
several groups of three equiintense weak peaks, see Fig. 1b. The
measured splittings correspond to energies in the range 13–
16 ꢃ 10ꢁ4 cmꢁ1. Both the number of peaks and the energy splitting
involved seem to suggest the presence of a superhyperfine interac-
tion between the magnetic copper and one nitrogen nucleus
(IN = 1), in line with the structure of Fig. 5.
Donorꢀꢀꢀacceptor (Å)
Angle (°)
Donorꢀꢀꢀacceptor (Å)
Angle (°)
N1ꢀꢀꢀO4
3.01(2)
N1ꢀꢀꢀO6
2.87 (1)
N2ꢀꢀꢀO4
3.18(2)
N2ꢀꢀꢀO6
3.19(1)
N1–H1ꢀꢀꢀO4
149.5(6)
N1AꢀꢀꢀO4
N1A–H1AꢀꢀꢀO4
131.6(6)
3.18(1)
N1–H1ꢀꢀꢀO6
N1AꢀꢀꢀO6
3.17(2)
N1A–H1AꢀꢀꢀO6
134.0(6)
139.3(6)
N2–H2ꢀꢀꢀO4
135.9(5)
N2AꢀꢀꢀO4
N2A–H2AꢀꢀꢀO4
137.4(5)
3.13(1)
N2–H2ꢀꢀꢀO6
N2AꢀꢀꢀO6
2.85(1)
N2A–H2AꢀꢀꢀO6
131.6(4)
142.6(5)
In conclusion, all the results of this EPR study on a polycrystal-
line sample of compound 4 appear to strictly support the proposed
crystallographic view based on two copper centres in different oxi-
dation states.
carboxylate, but in the other the molecule are significantly dis-
torted: the angle between the average plane of the terminal carbox-
ylate and that of the thiosemicarbazide fragment is 23.0(4)°.
A peculiarity of this structure is the way the two nitrate ions
link the two cationic moieties. A system of loose bifurcated hydro-
gen bonds binds each thiosemicarbazone amino and hydrazine NH
groups to two different oxygen atoms from the nitrate ions (see
Fig. 2). These interactions are quite loose and this is reflected in
the long bond distances (see Table 2) observed in the structure
and also in the high thermal parameters presented by the nitrate
ions.
3.2. Crystal structures of the complexes
A
PLUTON drawing representing complex [Cu(PPh3)2(Et-
Hmpt)]2(NO3)2 1 is reported in Fig. 2. The structure can be de-
scribed as formed by two cationic complex molecules with two
bridging nitrates that act as counterions. The copper(I) ion coordi-
nates two triphenylphosphines and an N1-ethyl methylpyruvate
thiosemicarbazone molecule in its neutral form. The thiosemicar-
bazones bind to the metal through the sulfur and the iminic nitro-
gen and form a five-term chelation ring. The coordination bonds
have distances of 2.165(6) and 2.148(6) Å for the Cu–N bonds
and 2.365(2) and 2.368(2) Å for the Cu–S bonds. The Cu–P bonds
range instead from 2.254(2) to 2.281(2) Å. The angles N–Cu–S of
82.3(2)° and 82.9(2)° are fairly small, while the P–Cu–P angles
are wide (129.60(8)° and 130.54(8)°). The copper coordination is
slightly distorted from the ideal tetrahedron but less than expected
in the presence of a potentially terdentate planar ligand. The non-
methylated oxygens of the ligand and the copper atom are 2.802(6)
and 2.799(6) Å apart and do not seem to influence the copper coor-
dination geometry. Noteworthy in both molecules that the cen-
troids between the coordinated N and S atoms lie on the plane
Another feature that characterizes the packing of these mole-
cules is
a
sixfold phenyl embrace (6PE) [35–38] (P1ꢀꢀꢀP1I
I = (x,1 + y,z) distance 7.354 Å) (see Fig. 3) between molecules re-
lated by a translation along y. More precisely the phenyls of the tri-
phenylphosphines involved in this interaction present what the
aforementioned paper calls an ‘‘offset sextuple phenyl embrace”
(OSPE) conformation. To complete the description of the packing
it is noteworthy to observe that the hydrogen bonds with the ni-
trate ions together with these OSPE create a ribbon of molecules
that develops along the y-axis.
The structure of compound 2, [Cu(PPh3)2(Ph-Hmpt)]NO3, is re-
ported in Fig. 4. The asymmetric unit is formed by a complex cat-
ion, [Cu(PPh3)2(Ph-Hmpt)]+, and a disordered nitrate anion bound
through a loose charge-assisted hydrogen bond. In the cation com-
plex, the copper(I) ion coordinates two triphenylphosphines and
an N1-phenyl methylpyruvate thiosemicarbazone molecule in its
neutral form. Also in this case the thiosemicarbazone binds to
the metal through the sulfur and the iminic nitrogen and forms a
five-term chelation ring. The angle N–Cu–S is 81.70(7)° while the
defined by P–Cu–P, as in
geometry.
a typical tetrahedral coordination
The thiosemicarbazone moieties, expected to be planar because
of the extended conjugated system of double bonds, in one of the
cations are in fact fairly planar with a distortion of only 9.8(4)° be-
tween the average thiosemicarbazide plane and that of the terminal
Fig. 2. PLUTON rendering of complex [Cu(PPh3)2(Et-Hmpt)]2(NO3)2 1.

M. Belicchi-Ferrari et al. / Polyhedron 28 (2009) 1160–1168
1165
Fig. 3. MERCURY drawing showing compound 1 with two adjacent molecules linked through the six phenyl embrace.
Fig. 5. PLUTON rendering of compound
4 complex fragment [Cu2(O2CCH3)(Et-
pt)(PPh3)2].
Fig. 4. PLUTON rendering of complex [Cu(PPh3)2(Ph-Hmpt)]NO3
2 showing the
disordered nitrate hydrogen bond interaction (on the right-hand side).
copper(II) and the ligand has hydrolysed and lost the ester methyl
group and has become a carboxylic acid. The asymmetric unit, as a
consequence, contains two metal centres in two different oxidation
states. The two copper atoms are bridged by an acetate ion and by
a shared sulfur from the thiosemicarbazone. The copper(II) atom
coordinates in a square planar fashion the thiosemicarbazone li-
gand. In this case the ligand behaves as O,N,S terdentate using also
its carboxylic oxygen. The fourth position site is occupied by an
acetate oxygen atom. The coordination distances are: Cu(II)–N
1.912(4), Cu(II)–S 2.267(1), Cu(II)–O 1.949(3) and Cu(II)–O(ac)
1.895(3) Å. Those of the tetrahedral copper(I) fragment are:
Cu(I)–O(ac) 2.248(4), Cu(I)–S1 2.351(1) Å, and the two with the
phosphorus atoms of the tpp ligands are Cu–P1 2.238(1) and Cu–
P2 2.277(1) Å.
P–Cu–P is slightly narrower than the previous ones, being
124.28(3)°. Compared to compound 1 the copper tetrahedral coor-
dination is therefore slightly more regular. The coordination bonds
have values of 2.156(2) Å for Cu–N, 2.400(1) Å for Cu–S and the
Cu–P bonds are of 2.282(1) and 2.277(1) Å.
The thiosemicarbazone ligand is strongly distorted from the ex-
pected planarity and the angle between the thiosemicarbazide
average plane and that of the carboxylate fragment is of
24.10(9)°; also the average plane of the phenyl ring is tilted of
36.3(1)° with respect to the thiosemicarbazide group. Also here,
one carboxylate oxygen is oriented in the direction of the copper
atom but presents a long distance Cu–O of 2.780(2) Å.
The packing is formed by ribbons of molecules held together by
hydrogen bonds involving the water molecule [N1–HꢀꢀꢀO1 W
2.936(7) Å 158.8(4)° and O1 WꢀꢀꢀO1Wi 2.982(7) Å i = ꢁx,ꢁy,ꢁz + 2]
and by a weak interaction between the copper(II) atom with the
The crystal structure of compound
4 [Cu2(O2CCH3)(Et-
pt)(PPh3)2] ꢀ H2O is reported in Fig. 5. The solution of this structure
shed light on the anomalies that were observed in the elemental
analysis and the IR spectra. This compound was particularly recal-
citrant to crystallise and, probably due to the long exposure to oxy-
gen and to air moisture, copper(I) has been partially oxidised to
hydrazine nitrogen atom of
a
nearby molecule (Cu2ꢀꢀꢀN2i
3.406(5) Å i = ꢁx + 1,ꢁy,ꢁz + 2) as can be seen in Fig. 6. These rib-
bons are then joined together by a 6PE (sixfold phenyl embrace)

1166
M. Belicchi-Ferrari et al. / Polyhedron 28 (2009) 1160–1168
Fig. 6. MERCURY drawing of the packing of compound 4 showing the ribbon of complex molecules held together through hydrogen bonds formed by water molecules and by a
long interaction between copper(II) and the hydrazine nitrogen of an adjacent molecule.
Fig. 7. PLUTON rendering of the two independent molecules of complex [Cu2(MePh-mpt)2(PPh3)2] 6.
interaction with a distance PꢀꢀꢀP of 7.234 Å that involves the
Table 3
triphenylphosphines.
Selected bond distances for compound 6.
The crystal structure of compound 6 [Cu2(MePh-mpt)2(PPh3)2]
Bond
Distance (Å)
Bond
Distance (Å)
is reported in Fig. 7. The asymmetric unit contains two indepen-
dent enantiomeric complex molecules. Both are planar 2Cu–2S
clusters made up of two copper ions with interatomic distances
of 2.845(1) and 2.844(1) Å, bridged by two sulfur atoms belonging
to two thiosemicarbazone ligands. These ligands are deprotonated
and located on the same face of the cluster plane, bind through the
sulfur and the iminic nitrogen (see Table 3) and are perpendicular
to each other and oriented head to tail (see Fig. 7). The triphenyl-
phosphines are on the opposite side and two phenyls interact edge
Cu1–Cu2
Cu1–S1
Cu1–S2
Cu1–P1
Cu1–N3
Cu2–S1
Cu2–S2
Cu2–P2
Cu2–N6
2.845(1)
2.347(3)
2.379(3)
2.213(3)
2.102(7)
2.385(3)
2.343(3)
2.217(3)
2.124(7)
Cu3–Cu4
Cu3–S3
Cu3–S4
Cu3–P3
Cu3–N9
Cu4–S3
Cu4–S4
Cu4–P4
Cu4–N12
2.845(1)
2.343(3)
2.386(3)
2.207(3)
2.084(7)
2.395(3)
2.343(3)
2.213(3)
2.115(8)

M. Belicchi-Ferrari et al. / Polyhedron 28 (2009) 1160–1168
1167
Fig. 8. MERCURY drawing of the packing of compound 6 showing the phenyl–phenyl edge to face interactions extending in the xy plane.
trophilicity of the carboxyl carbon and speeding up the process of
demethylation. This phenomenon of demethylation had been pre-
viously observed in a work where an analogous ligand was coordi-
nated by zinc(II) [40]. The ligand is therefore the same but the
effects are dramatically different in the coordination of copper(I)
versus copper(II). The fairly rigid ligand is markedly deformed
upon coordination on copper(II) in 4 (see Fig. 9) to allow the car-
boxylic oxygen to bind to the metal ion, while it remains almost
unaltered upon coordination of copper(I) in 1.
Acknowledgment
Fig. 9. PLUTON rendering of the methylpyruvate thiosemicarbazone group deforma-
tion observed passing from a Cu(I) in 1 to a Cu(II) coordination in 4.
Thanks are due to Dr. Mario Ghillani for his valuable contribu-
tion in the recording of EPR spectra.
to face with those of an adjacent molecule forming parallel quadru-
ple phenyl embrace (PQPE) [35–38] (see Fig. 8).
Appendix A. Supplementary data
CCDC 690657, 690658, 690659 and 690660 contains the sup-
plementary crystallographic data for compounds 1, 2, 4 and 6.
These data can be obtained free of charge via http://www.ccdc.ca-
m.ac.uk/conts/retrieving.html, or from the Cambridge Crystallo-
graphic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK;
fax: (+44) 1223-336-033; or e-mail: deposit@ccdc.cam.ac.uk. Sup-
plementary data associated with this article can be found, in the
online version, at doi:10.1016/j.poly.2009.01.013.
4. Conclusions
Copper(I) ions, with their d10 electronic configuration, do not
have an energetically preferred geometry and their coordination
is therefore mainly governed by steric repulsions. Aside from the
two donor species P and S, the remaining space around the copper
ion, in bistriphenylphosphine complexes present in the Cambridge
Structural Database [39], is usually occupied by a fourth ligand,
usually a halide. In previous experiments we have verified that
the absence of other soft species can influence the coordination
and in particular that the imino nitrogen can become a suitable do-
nor for coordination on a soft metal ion in the absence of softer
competitors. What is apparent in the structures reported in this pa-
per is that the control on the coordinating behaviour of a ligand is
tuned by a series of factors among which the predominant is the
hard soft nature of the atoms involved in the interactions with
the metal centre. In our case it is striking the comparison between
compounds 1 and 4. Compound 4 was recalcitrant to crystallisation
and the crystals were obtained after many attempts leaving the
compound exposed to atmospheric oxygen and moisture. The cop-
per(I) oxidised to copper(II) increases both the acidity of the hydra-
zine nitrogen stabilizing the anionic form and promotes the
hydrolysis of the methyl group from the esther improving the elec-
References
[1] T.S. Lobana, Rekha, A.P.S. Pannu, G. Hundal, R.J. Butcher, A. Castineiras,
Polyhedron 26 (2007) 2621.
[2] T.S. Lobana, Rekha, R.J. Butcher, Transition Met. Chem. 29 (2004) 291.
[3] T.S. Lobana, Rekha, R.J. Butcher, A. Castineiras, E. Bermejo, P.V. Bharatam,
Inorg. Chem. 45 (2006) 1535.
[4] D.X. West, S.B. Padhye, P.B. Sonawane, Struct. Bond. (Berlin) 76 (1991) 1.
[5] D.R. Smith, Coord. Chem. Rev. 164 (1997) 575.
[6] A.R. Cowley, J.R. Dilworth, P.S. Donnelly, E. Labisbal, A. Sousa, J. Am. Chem. Soc.
124 (2002) 5270.
[7] D.B. Lovejoy, D.R. Richardson, Blood 100 (2002) 666.
[8] L.-J. Ming, Med. Res. Rev. 23 (2003) 697.
[9] M. Belicchi-Ferrari, G. Gasparri-Fava, C. Pelizzi, P. Tarasconi, Inorg. Chim. Acta
97 (1985) 99.
[10] M. Belicchi-Ferrari, G. Gasparri-Fava, M. Lanfranchi, C. Pelizzi, P. Tarasconi,
Inorg. Chim. Acta 181 (1991) 253.
[11] J.S. Casas, M.S. Garcia-Tasende, J. Sordo, Coord. Chem. Rev. 209 (2000) 197.

1168
M. Belicchi-Ferrari et al. / Polyhedron 28 (2009) 1160–1168
[12] M.J.M. Campbell, Coord. Chem. Rev. 15 (1975) 279.
[26] A. Altomare, M.C. Burla, M. Cavalli, G. Cascarano, C. Giacovazzo, A. Gagliardi,
A.G.G. Moliterni, G. Polidori, R. Spagna, J. Appl. Crystallogr. 32 (1999) 115.
[13] S. Padhye, G.B. Kauffman, Coord. Chem. Rev. 63 (1985) 127.
[14] D.X. West, A.E. Liberta, S.B. Padhye, R.C. Chikate, P.B. Sonawane, A.S. Kumbhar,
R.G. Yerande, Coord. Chem. Rev. 123 (1993) 49.
[27] G. Sheldrick, SHELXL97,
A Program for Structure Refinement, University of
Goettingen, Germany, 1997.
[15] R.L. De Lima, L.R.D. Teixeira, T.M.G. Carneiro, H. Beraldo, J. Braz. Chem. Soc. 10
(1999) 184.
[28] M. Nardelli, J. Appl. Crystallogr. 28 (1995) 659.
[29] (a) A.L. Spek, Acta Crystallogr. A46 (1990) C34;
[16] N.T. Akinchan, R. Akinchan, P.M. Drozdewski, Y.H. Yang, T.L. Klein, D.X. West,
Synth. React. Inorg. Met.-Org. Chem. 26 (1996) 1735.
[17] H.G. Zheng, D.X. Zeng, X.Q. Xin, W.T. Wong, Polyhedron 16 (1997) 3499.
[18] S. Lhuachan, S. Siripaisarnpipat, N. Chaichit, Eur. J. Inorg. Chem. (2003) 263.
[19] L.J. Ashfield, A.R. Cowley, J.R. Dilworth, P.S. Donnelly, Inorg. Chem. 43 (2004)
4121.
(b) A.L. Spek, PLATON, A Multipurpose Crystallographic Tool, Utrecht University,
Utrecht, The Netherlands, 1998.
[30] C.F. Macrae, P.R. Edgington, P. McCabe, E. Pidcock, G.P. Shields, R. Taylor, M.
Towler, J. van de Streek, J. Appl. Crystallogr. 39 (2006) 453.
[31] A. Abragam, B. Bleaney, Electron Paramagnetic Resonance of Transition Ions,
Clarendon Press, Oxford, 1970.
[20] M. Belicchi-Ferrari, A. Bonardi, G. Gasparri-Fava, C. Pelizzi, P. Tarasconi, Inorg.
Chim. Acta 223 (1994) 77.
[21] T.S. Lobana, B.S. Sidhu, A. Castineiras, E. Bermejo, T. Nishioka, J. Coord. Chem.
58 (2005) 803.
[22] T.S. Lobana, S. Khanna, R.J. Butcher, A.D. Hunter, M. Zeller, Polyhedron 25
(2006) 2755.
[23] M. Belicchi-Ferrari, F. Bisceglie, C. Cavalieri, G. Pelosi, P. Tarasconi, Polyhedron
26 (2007) 3774.
[32] D. Kivelson, R. Neiman, J. Chem. Phys. 35 (1961) 149.
[33] A.H. Maki, B.R. McGarvey, J. Chem. Phys. 29 (1958) 31.
[34] E. Buluggiu, A. Vera, A.A.G. Tomlinson, J. Chem. Phys. 36 (1972) 5602.
[35] I.G. Dance, M.L. Scudder, J. Chem. Soc., Chem. Commun. (1995) 1039.
[36] I.G. Dance, M.L. Scudder, Chem. Eur. J. 2 (1996) 481.
[37] I. Dance, M. Scudder, J. Chem. Soc. Dalton Trans. 10 (2000) 1579.
[38] I. Dance, M. Scudder, J. Chem. Soc., Dalton Trans. 10 (2000) 1587.
[39] F.H. Allen, Acta Crystallogr. B58 (2002) 380.
[24] M. Belicchi-Ferrari, F. Bisceglie, G. Pelosi, P. Tarasconi, R. Albertini, S. Pinelli, J.
Inorg. Biochem. 87 (2001) 137.
[40] M. Belicchi Ferrari, G. Gasparri Fava, G. Pelosi, P. Tarasconi, Polyhedron 19
(2000) 1895.
[25] G.M. Sheldrick, Acta Crystallogr. A64 (2008) 112.