Heteropolymetallic complexes containing 1,1′-diphenylphosphino-ferrocene

Article abstract of DOI:10.1016/j.ica.2006.10.021

The paper explores the capability of [(dppf)Pt(H-nbu₂-DTO)]Cl (2) (dppf = 1,1′-diphenylphosphinoferrocene; H-nbu₂-DTO = di-nbutyl-dithioxamidate) to act as a starting module for heterometallic linear chains. Actually, the reaction of 2 with [RuCl₂(p-cymene)]₂ affords the heterotrimetallic complex [Cl(p-cymene) Ru(μ-nbu₂-DTO κ-N,N Ru κ-S,S Pt)Pt(dppf κ-P,P Pt) ]₂ (4). However 2, allowed to stand, provides a blue compound of formula [(dppf)Pt(H-nbu₂-DTO)]_nCl_n (3), the most reliable value of n being 6. The oxidation behavior of the new species 2-4 has also been investigated. In particular, the oxidation behavior of cyclic compound 3 is quite unusual, and suggests a large delocalization of the HOMO over the whole multicomponent molecule.

Full text of DOI:10.1016/j.ica.2006.10.021

Inorganica Chimica Acta 360 (2007) 1929–1934
www.elsevier.com/locate/ica
Heteropolymetallic complexes containing
1,10-diphenylphosphino-ferrocene
*
Santo Lanza , Federique Loiseau, Giuseppe Tresoldi, Scolastica Serroni,
*
Sebastiano Campagna
`
Dipartimento di Chimica Inorganica, Chimica Fisica e Chimica Analitica, Universita di Messina, Salita Sperone 31, 98166, Messina, Italy
Received 29 September 2006; accepted 14 October 2006
Available online 27 October 2006
Abstract
The paper explores the capability of [(dppf)Pt(H–nbu2–DTO)]Cl (2) (dppf = 1,10-diphenylphosphinoferrocene; H–nbu2–DTO = di-
nbutyl-dithioxamidate) to act as a starting module for heterometallic linear chains. Actually, the reaction of 2 with [RuCl2(p-cymene)]2
affords the heterotrimetallic complex [Cl(p-cymene) Ru(l-nbu2–DTO j-N,N Ru j-S,S Pt)Pt(dppf j-P,P Pt) ]2 (4). However 2, allowed to
stand, provides a blue compound of formula [(dppf)Pt(H–nbu2–DTO)]nCln (3), the most reliable value of n being 6. The oxidation behav-
ior of the new species 2–4 has also been investigated. In particular, the oxidation behavior of cyclic compound 3 is quite unusual, and
suggests a large delocalization of the HOMO over the whole multicomponent molecule.
ꢀ 2006 Elsevier B.V. All rights reserved.
Keywords: Heteropolymetallic complexes; Diphenylphosphinoferrocene (ddpf); Platinum complexes; Ruthenium complexes; Cyclic voltammetry
1. Introduction
binucleating ligands suitable to link metal ions in successive
steps; reliable approaches have exploited heterotopic
ligands, which can accomplish a sequential synthesis since
the different reactivity of the various chelating systems. In
this context, oxamato ligands have been extensively used
to generate families of polymetallic compounds according
to a sequential strategy [3]; however, the limitation of this
generation of oxamato complexes was the isolation of only
heterobimetallic compounds. Notwithstanding, oxamato
ligands properly designed have overcome this inconve-
nience and heterotrimetallic and heterotetrametallic com-
plexes have been recently obtained [4,5].
Another binucleating ligand, 1,10-phenanthroline-5,6-
dione, has been used as a building block for polymetallic
systems: it possesses a bifunctional character and exhibits
different reactivity of its functionalities. This binucleating
ligand has provided bi and trimetallic derivatives in some
cases structurally characterized [6–9]. In the past few years,
we have found that platinum fragments linked to a depro-
tonated secondary dithioxamide through the sulfur-chelat-
ing system can function as a ligand complex: the N–Hꢀ ꢀ ꢀN
A frontier research in coordination and organometallic
chemistry is the functional assembling by bottom-up
approach [1] of submicrometric aggregates, potentially use-
ful as molecular devices. The electronic, geometric, and
structural properties of the molecular components used
as building blocks determine interaction and forces fea-
tured by the multicomponent final structure. Therefore it
is not surprising if the use of complexes as ligands for the
step-by-step syntheses of polymetallic compounds is
assuming increasing importance in the modern inorganic
chemistry [2]. In particular, the sequential synthesis of lin-
ear metallic chains plays an important role since such a
synthetic procedure allows to insert the individual proper-
ties of various components into a multicomponent molecu-
lar system. Many efforts have been devoted to find
*
Corresponding authors. Tel.: +39 0906765718; fax: +39 090393756 (S.
Lanza).
E-mail address: Lanza@chem.unime.it (S. Lanza).
0020-1693/$ - see front matter ꢀ 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2006.10.021

1930
S. Lanza et al. / Inorganica Chimica Acta 360 (2007) 1929–1934
hexafluorophosphate was used as the supporting electro-
N
N
S
S
S
S
N
N
S
S
N
N
lyte and its concentration was 0.05 M. Cyclic voltammo-
grams were obtained at scan rates of 20, 50, 200, and
500 mV/s. For reversible processes, half-wave potentials
(versus SCE) were calculated as the average of the catho-
dic and anodic peaks. The criteria for reversibility were
the separation of 60 mV between cathodic and anodic
peaks, the close to unity ratio of the intensities of the
cathodic and anodic currents, and the constancy of the
peak potential on changing scan rate. The number of
exchanged electrons was measured with differential pulse
voltammetry (DPV) experiments performed with a scan
rate of 20 mV/s, a pulse height of 75 mV, and a duration
of 40 ms, and by taking advantage of the presence of fer-
rocene used as the internal reference. Experimental uncer-
tainty on redox potentials is 10 mV.
M
M
MLn
B
C
A
Scheme 1.
frame of the S,S-coordinated ligand is nucleophilic enough
to split chloro-bridged dimers so that bimetallic complexes
can be readily obtained [10]. Following our previous
results, we started to build up linear polymetallic chains
through a modular use of bischelate dithioxamide com-
plexes of type A (Scheme 1), [M(HR2DTO)2] (M = diva-
lent transition metal ions, R = alkyl groups), mono-
chelate [LnM(HR2DTO)] precursors (B), and non-ligating
fragments coming from chloro-bridged dimers (C).
In trimetallic complexes of the type C–A–C (C = Pd-
(g3-allyl); Ru(p-cymene)Cl; Rh(2-phenyl-pyridine N,C);
Rh(cyclo-octadiene); Rh(pentamethyl-cyclopentadienyl)Cl;
A = platinum(II)-bis(di-alkyl-dithiooxamidate)) have already
been published [11].
The present paper starts from the fact that one of the
methods used to prepare heteropolynuclear compounds is
based on the use of ferrocenyl substrates containing hetero-
atoms (mainly nitrogen, sulfur, phosphorus, or oxygen)
and/or unsaturated groups with good donor abilities,
which allow the coordination of one or more metal ions
[12]. Thus we have reacted 1,10-diphenyl-phosphino ferro-
cene with a module like B, i.e., the new platinum complex
(Me2SO)ClPt(di-nbutyldithiooxamidate j-S,S Pt); the
bimetallic complex (1,10-diphenylfosphino-ferrocenyl j-
P,P Pt)Pt(di-nbutyldithiooxamidate j-S,S Pt) has been
obtained, which could be viewed as a starting module for
a monodirectional growth of a polymetallic chain. The
use of a ferrocenyl derivative as a fragment in a polymetal-
lic chain offers the opportunity to test the mutual electronic
influence among metals connected by binucleating ligands.
2.2. Synthesis of [(Me2SO)ClPt(H–nbu2–DTO)]
(H–nbu2–DTO = di-nbutyl-dithioxamidate) (1)
cis-[Pt(Me2SO)2Cl2] (422 mg, 1 mmol) was suspended in
100 ml of chloroform together with five grams of sodium
bicarbonate. The equimolar amount of dibutyldithioxa-
mide (232 mg, 1 mmol) was added in small portions to the
stirred suspension. The solution turned orange-yellow and
was leaved under magnetical agitation for 1 h. At the end
the orange solution was separated from sodium bicarbonate
by filtration. The filtrate was concentrated to a small vol-
ume (25 ml); then petroleum ether (75 ml) was added and
370 mg of 1 precipitated as a yellow-orange powder
1
(yields = 68%). H NMR (CDCl3, 298 K): d = 0.93 [t, 6H,
3
N–(CH2)3–CH3, JH–H = 7.0 Hz], 1.39 ppm (m, 4H, N–
CH2–CH2–CH2–CH3), 1.70 (m, 4H, N–CH2–CH2–CH2–
3
CH3), 3,45 [s, SO(CH3)2, JPt–H = 37.4 Hz], 6.58 [t, 2H,
3
N–CH2–CH2–CH2–CH3, JH–H = 7.0 Hz], 6.63 [t, 2H, N–
3
CH2–CH2–CH2–CH3, JH–H = 7.0 Hz] ppm. 13C{1H}
NMR (CDCl3, 298 K): d = 13.7 [s, 2C, N–(CH2)3–CH3],
20.4 (s, 2C, N–CH2–CH2–CH2–CH3), 30.45 (s, 1C, N–
CH2–CH2–CH2–CH3), 30.80 (s, 1C, N–CH2–CH2–CH2–
2
CH3), 44.40 [s, 2C, SO(CH3)2, JPt–C = 47.2 Hz], 48.4 0 (s,
2. Experimental
1C, N–CH2–CH2–CH2–CH3), 48.70 (s, 1C, N–CH2–CH2–
CH2–CH3), 182.00 (s, CS), 184.30 (s, CS) ppm. Anal. Calc.
for C12H25ClN2OPtS2 (508.00): H, 4.96; C, 28.37; N, 5.51;
Cl, 6.90. Found: H, 4.85; C, 28.41; N, 5.60; Cl, 6.88%.
2.1. General procedures
cis-[PtCl2(Me2SO)2] [13] and di-nbutyl-dithiooxamide
[14] were prepared according to published methods. Other
reagents and solvents were used as received. NMR spectra
were recorded with a Bruker AMX 300 spectrometer. The
following Bruker programs were used: zg, zgpg, Jmod,
COSY, NOESY.TP. Electrochemical measurements were
carried out in argon-purged 1,2-dichloroethane at room
temperature with a PAR 273 multipurpose equipment
interfaced to a PC. The working electrode was a glassy
carbon (8 mm2, Amel) electrode. The counter electrode
was a Pt wire, and the reference electrode was an SCE
separated with a fine glass frit. The concentration of the
complexes was about 5 · 10ꢁ4 M. Tetrabutylammonium
2.3. Synthesis of [(dppf)Pt(H–nbu2–DTO)]Cl
(dppf = diphenyl-phosphino-ferrocene) (2)
[(Me2SO)ClPt(H–nbu2–DTO)] (1) (539 mg, 1 mmol) was
reacted in chloroform (50 ml) with the stoichiometric
amount of dppf (554 mg, 1 mmol). The orange solution
turns pale yellow immediately after the addition of dppf.
On adding petroleum ether (100 ml), [(dppf)Pt(HbuD-
TO)]Cl precipitated as a pale yellow powder (600 mg,
1
59.1% yield). H NMR (CDCl3, 298 K): d = 0.77 [t, 6H,
3
N–(CH2)3–CH3, JH–H = 6.8 Hz], 1.19 (m, 4H, N–CH2–
CH2–CH2–CH3), 1.54 (m, 4H, N–CH2–CH2–CH2–CH3),

S. Lanza et al. / Inorganica Chimica Acta 360 (2007) 1929–1934
1931
3.30 (t, 4H, N–CH2–CH2–CH2–CH3, 3JH–H = 6.8 Hz), 4.18
(s, 4H, 3,30 Cp protons); 4.34 (s, 4H, 2,20 Cp protons); 7.1–
7.4 (m, 20H, phenyl protons) ppm. 13C{1H} NMR (CDCl3,
298 K): d = 14.00 (s, 2C, N–(CH2)2–CH3), 28.80 (s, 2C, N–
CH2–CH2–CH2–CH3), 31.60 (s, 2C, N–CH2–CH2–CH2–
CH3), 53.4 (s, 2C, N–CH2–CH2–CH2–CH3), 73.3 (m, 4C,
3,30 Cp carbons), 75.80 (m, 4C, 2,20 Cp carbons), 127.9
(m, 8C, meta phenyl carbons), 130.2 (m, 4C, C ipso),
131.1 (s, 4C, para phenyl carbons); 134.8 (m, 8C, ortho phe-
The solution turned red and was allowed to stand for
2 h. Then the solvent was removed and the crude product,
dissolved in the minimum amount of chloroform, was
placed at the head of a neutral alumina column equili-
brated with petroleum ether. The red orange eluate was
concentrated to a small volume (10 ml); then petroleum
ether (40 ml) was added and 368 mg of 4 was collected as
a red orange powder (yield = 55%). 1H NMR (CDCl3,
3
298 K): d = 0.76 [t, 6H, N–(CH2)3–CH3, JH–H = 6.7 Hz],
3
nyl carbons); 186 (t, 2C, CS, JP–C = 5.8 Hz) ppm. 31P
1.10 [m, 4H, N–(CH2)2–CH2–CH3], 1.18 [d, 6H, cymene
3
NMR (CDCl3, 298 K): d = 18.5 (s, Pt–P, JP–Pt = 3222 Hz)
–CH(CH3)2, JH–H = 6.7 Hz], 1.55 (m, 4H, N–CH2–CH2–
ppm. Anal. Calc. for C H47ClFeN2P2PtS2 (1016.34): H,
4.66; C, 52.00; N, 2.76; Cl, 3.49. Found: H, 4.55; C,
52.09; N, 2.72; Cl, 3.51%.
CH2–CH3), 2.41 (s, 3H, cymene –CH3), 2.76 [sl, 1H, cym-
44
3
ene –CH(CH3)2, JH–H = 6.7 Hz], 3.82 (ABX2 double trip-
let,
2H,
N–CH2–CH2–CH2–CH3,
3JH–H = 6.7 Hz,
Jgem = 13.8 Hz), 3.92 (ABX2 double triplet, 2H, N–CH2–
3
2.4. Synthesis of [(dppf)Pt(H–nbu2–DTO)]6Cl6 (3)
CH2–CH2–CH3, JHꢁH = 6.7 Hz; Jgem = 13.8 Hz), 4.25
(br s, 4H, 3,30 Cp protons), 4.45 (br s, 4H, 2,20 Cp protons),
5,55 (dd, unresolved AA0XX0 spin system, 4H, cymene ring
protons), 7.2–7.7 (m, 20H, phenyl protons) ppm. 13C{1H}
NMR: d = 13.70 [s, 2C, N–(CH2)3–CH3], 18.50 (s, 1C,
cymene –CH3), 20.50 [s, 2C, N–(CH2)2–CH2–CH3], 22.2
[s, 2C, cymene –CH(CH3)2], 29.00 (s, 2C, N–CH2–CH2–
CH2–CH3), 31.20 [s, 1C, cymene –CH(CH3)2], 62.10 (s,
2C, N–CH2–CH2–CH2–CH3), 71,40 (m, 4C, 3,30 Cp car-
bons), 74,10 (m, 4C, 2,20 Cp carbons), 84.70, 85.10 (2s,
A concentrated solution of 2 (254 mg, 0.25 mmol, in
10 ml of CHCl3), allowed to stand at room temperature
for two days, slowly turns to green. Then this green solution
is put at the head of an alumina column equilibrated with
petroleum ether. Elution with chloroform/methanol 98:2
mixture separates 2 from 3. This latter remains as a deep
blue compound at the head of the column. When 2 is com-
pletely removed, 3 is eluated with a chloroform/methanol
90:10 mixture. The solvent was removed and the blue resi-
due, dissolved in the minimum amount of CHCl3 (10 ml),
was precipitated as a deep blue powder with petroleum light
(90 ml). By remaking the procedure (four times) 80 mg of
pure 3 were collected (31% yield). 1H NMR (CDCl3,
298 K): d = 0.33 (m, 12H, N–CH2–CH2–CH2–CH3), 0.82
4C, cymene C2,3,5,6), 100.00, 105.00 (2s, 2C, cymene C1,4
)
128.00 (m, 8C, meta phenyl carbons); 132.00 (m, 4C, para
phenyl carbons), 134 (m, 8C, ortho phenyl carbons) ppm.
31P NMR: d = 17.60 (s, JP–Pt0 = 3101 Hz) ppm. Anal.
Calc. for C54H60Cl2FeN2P2PtRuS2 (1286.07): H, 4.70; C,
50.43; N, 2.18; Cl, 5.44. Found: H, 4.65; C, 50.09; N,
2.12; Cl, 5.13%.
3
(t, 36H, N–CH2–CH2–CH2–CH3, JH–H = 6.8 Hz), 0.86
(m, 24H, N–CH2–CH2–CH2–CH3); 1.14 (m, 12H, N–
CH2–CH2–CH2–CH3), 2.21 (ABX2 double triplet, 12H,
3JH–H = 6.6 Hz; Jgem = 12.8 Hz), 3.21 (ABX2 double triplet,
3. Results and discussion
3
12H, JH–H = 6.6 Hz; Jgem = 12.8 Hz), 4.20, 4.23 (2s, 24H,
[Cl(Me2SO)Pt(H–nbu2–DTO)] (1, H–nbu2–DTO = di-
normalbutyl-dithiooxamidate, Scheme 2) was prepared by
reacting equimolar quantities of di-normalbutyl-dithiooxa-
mide and cis-[PtCl2 (Me2SO)2]suspended in chloroform in
the presence of sodium bicarbonate.
The function of the inorganic salt was to remove HCl
from the tight ion pair {(Me2SO)ClPt(H–nbu2–DTO)+,
(Clꢁ)} once formed, being the ion pair unstable in solution
[15].
3,30 Cp protons), 4.46, 4.55 (2s, 24H, 2,20 Cp protons),
7.21–7.78 (m, 120H, phenyl protons) ppm. 13C{1H} NMR
(CDCl3): d = 4.00 (s, 12C, N–CH2–CH2–CH2–CH3), 20.90
(s, 12C, N–CH2–CH2–C H2–CH3), 28.53 (s, 12C, N–CH2–
CH2–CH2–CH3), 55.84 (s, 12C, N–CH2–CH2–CH2–CH3),
74.00, 74.40 (2m, 24C, 3,30 Cp carbons), 76.00, 76.10 (2m,
24C, 2,20 Cp carbons), 128.13, 128.20 (2m, 48C, meta phenyl
carbons) 129.80 (m, 24C, ipso phenyl carbons), 131.90 (s,
24C, para phenyl carbons), 134.58, 134.65 (2m, 48C, ortho
Compound 1 was reacted with the equimolar amount
of dppf (diphenylphosphinoferrocene) according to
Scheme 3.
3
phenyl carbons); 183.26 (t, CS, JP–C = 5.8 Hz) ppm. 31P
NMR (CDCl3, 298 K): d = 18.9 (s, Pt–P, JP–Pt = 3242 Hz)
ppm. Anal. Calc. for (C44H47N2S2P2ClFePt)6 (6098.04): H,
4.66; C, 52.00; N, 2.76; Cl, 3.49. Found: H, 4.49; C, 51.67;
N, 2.70; S, 6.35; Cl, 3.43%.
n
bu
O
2.5. Synthesis of [(dppf)Pt(l-H–nbu2–DTO j-S,S0 Pt j-
N,N0 Ru)Ru(p-cymene)Cl]Cl (4)
S
S
S
N
Pt
H
Cl
N
nbu
Compound 2 (508 mg, 0.5 mmol) was dissolved in
100 ml of chloroform. To the resulting yellow solution
153 mg (0.25 mmol) of [RuCl2(p-cymene)]2 was added.
1
Scheme 2.

1932
S. Lanza et al. / Inorganica Chimica Acta 360 (2007) 1929–1934
n
bu
n
bu
Ph
P
Ph
P
O
Ph
Ph
Pt
S
S
S
N
S
S
N
-Me SO
2
+
Pt
H
H
Cl
Fe
Fe
N
P
P
Cl
N
Ph Ph
Ph Ph
nbu
nbu
2
1
Scheme 3.
n
bu
n
bu
Ph
P
Ph
P
Ph
Ph
Pt
6
[RuCl (η -
p
-cymene)]
S
S
N
2
2
S
S
N
Cl
Cl
Pt
H
Cl
Fe
Ru
Fe
6
N
N
P
P
η -p-Cy
-HCl
Ph Ph
Ph Ph
nbu
nbu
2
4
Scheme 4.
In Scheme 3, the binuclear Fe–Pt complex (2) is depicted
with the dppf in ‘‘synclinal staggered’’ conformation. Such
a diphosphinoferrocene arrangement in complexes showing
g2-chelating dppf seems to be largely preferred [16]. How-
ever, an eclipsed nature of the C5 rings in a dppf platinum
complex has been shown with a deviation angle of 2.4ꢁ [17].
rings of each phosphorus atom in dppf ligand become
non-equivalent since the rigidity of the cyclic structure;
thus the non-1:2:1 triplets of the phenyl carbons2 [18],
observed in 2, appear splitted in the 13C NMR spectrum
of 3. Innumerable attempts of crystallization of 3 merely
produced amorphous powders; for this reason we find it
hazardous to propose any temptative structure for 3 only
on the basis of NMR evidences, which allow to say only
that 3 is the result of a cyclic self-assembling of platinum
dithioxamidate metal fragments, alternated by dppf reels
as spacers.
Being the above cyclization a slow process, it is possible
to exploit 2 as a ligand complex, and link a further metal
fragment according to the known procedures [10,11].
Thus, by the reaction of 2 with an half molar quantity of
[RuCl2(g6-p-cymene)]2, the following process took place
(Scheme 4).
1
The H and 13C NMR of 2 show a single butylic pattern.
Because of the rapid conformational flipping of chelated
dppf, both proton and carbon spectra of dppf in 2 show
chemical equivalence between 2,3 and 20,30 CH in each
cyclopentadienyl fragment. Phosphorus resonances appear
as a singlet flanked by 195Pt satellites.
Concentrated chloroform solutions of 2, allowed to
stand, slowly turn to green. By chromatography of the
green mixture on an alumina column, the pale yellow 2
is separated from a deep blue product 3. Despite the dra-
matic change in color, both 2 and 3 have the same compo-
sition; i.e., in 3 there is a dppf unit per Pt(HbuDTO)
fragment. Compound 3 is a cyclic compound of formula
[2]n where the most probable value n is 6.1 The cyclic nat-
ure of 3 is mainly suggested by the presence of only one P–
Pt signal and only one butylic pattern, more easily detect-
able in 13C NMR spectra . Furthermore, in the 13C NMR
spectrum of 3 four multiplets are detectable for 2 and 3
carbons of ferrocene cp rings. This means that 2 and 3 car-
bons of one ring are unequivalent with respect to the sim-
ilar carbons of the other ring in the same ferrocenyl reel
probably because dppf spacers are blocked in an eclipsed
synclinal conformation of Cp rings. Moreover, the phenyl
The tetrahedral arrangement of ligands around ruthe-
nium atom in 4 causes the plane passing through arene cen-
troid and chlorine to be a symmetry plane, while that one
containing nitrogen atoms, which also contains the plati-
num square plane, is not a symmetry plane. Consequently,
1
in the H NMR spectrum of 4 there is only one butylic
spectral pattern, in which methylene hydrogens, being
chemically non-equivalent, exhibit complex ABX2
resonancies.
We wish to point here that the process in Scheme 4 rep-
resents a route for a systematic access to trinuclear hetero-
trimetallic complexes.
1
In principle, tetrameric and octameric conformations cannot be
2
excluded, but an hexamer is by far to be preferred for geometrical
reasons. Other evidences (electrochemistry, see later in the main text)
strongly support the proposed nuclearity.
Aromatic and Cp carbons appear in the spectrum as quintets, non-
1:3:1 triplets, doublet of doublets, because of virtual coupling [18]. We
have reported these signals as multiplets.

S. Lanza et al. / Inorganica Chimica Acta 360 (2007) 1929–1934
1933
CV experiments, we assigned the first process of 3 to a one-
electron oxidation and the second one to two closely-
spaced one-electron processes dealing with two weakly
interacting sites.
+
6
P
To rationalize this behavior it is useful to consider that
bridging Pt(II) centers may allow a good electronic interac-
tion between redox-active groups: this is the case for the
‘rod-like’ organometallic complex trans-[Pt(ethynylferro-
cene)2(PPh3)2], for which the two ferrocene-centered oxida-
tions are splitted by 260 mV [20]. Although the Pt(II)
center in 3 is quite different from that in trans-[Pt(ethynyl-
ferrocene)2(PPh3)2], and the geometrical arrangements of
the species are also different, it is reasonable to assume that
two nearby dppf subunits in 3 are significantly coupled via
the Pt(II) centers. Since all the dppf subunits in 3 are iden-
tical to one another (as evidences by NMR spectra), the
coupling can easily be extended to all the dppf subunits
which are present in the cyclic multinuclear array. The con-
sequence is that the HOMOs of the six redox centers can
give rise to delocalized orbitals which are stabilized com-
pared to the HOMO of the ‘isolated’ subunit, and to delo-
calized ones which result to be highly destabilized. It is one
of these latter delocalized orbitals which is held responsible
for the oxidation process at +0.38 V. A platinum-centered
assignment for the oxidation process can be ruled out since
the precursor [Pt(dithioxamide)(dmso)Cl] (1) complex does
not undergo any oxidation process at potential less positive
than +1.70 mV and the trinuclear species [Ru(Cp)Cl(l-
dithioxamide)Pt(dppf)]+ (4) exhibits the first oxidation pro-
cess at +1.10 V.
The strong differences in the electronic properties of 2
and 3 indicated by the redox data are also evidenced by
the absorption spectra. The lowest-energy absorption fea-
ture of 2 in dichloromethane solution is a structured band
with maximum at about 420 nm, whereas the lowest-energy
absorption feature of 3 is also a structured band, but the
maximum is displaced at 680 nm. Energy separation
between the maxima within each structured absorption
bands (i.e., the vibrational progression) is about
1500 cmꢁ1 in both 2 and 3. The similarities in the shapes
of the bands in 2 and 3 suggest a similar origin. So, the
red-shift of the low-energy band of 3 with respect to 2
can be directly related to the difference in the first oxidation
potentials of the two compounds3 [21].
P
Fe
S
Pt
S
6
Fig. 1. Proposed formula for cyclic oligomer 3.
4. Electrochemistry
The cyclic voltammogram (CV) of complex 2 in 1,2-
dichloroethane exhibits a reversible one-electron oxidation
process at +1.13 V versus SCE. Such a process is assigned
to the ferrocene/ferrocinium couple; it is strongly shifted to
more positive potentials compared to the oxidation of fer-
rocene (+0.48 V under these experimental conditions), as
expected because of the presence of the diphenylphosphino
substituents. However, it is also much more positive than
the oxidation process found for dppf under the same exper-
imental conditions (+0.67 V), and close to those reported
for (dppf)PtCl2 and (dppf)Pt(i-mnt) (i-mnt = 2,2-dicyano-
1,1-ethylenedithiolate), which undergo reversible dppf-
based oxidation at +1.27 V and +1.22 V versus Ag/Ag+
in dichloromethane [19]. The significant positive shift of
the oxidation in 2 compared to that of dppf is attributed
to the electron withdrawing ability of the dithioxamide
ligand, which drains electronic charge from the dppf
ligands via the Pt(II) center.
The cyclic multinuclear system 3 undergoes a reversible
oxidation process at +0.38 V followed by a quasi-reversible
process at +1.17 V (Fig. 2).
By comparison between the intensities of the peaks of 3
with those of 2 and that of ferrocene, used as standard, in
10.00
0.00
3
Absorption bands similar (although at slightly higher energy) to that
-10.00
-20.00
-30.00
exhibited by 1 have been reported for homoleptic Pt(II) species with
dithioxamide ligands, and have been attributed to charge-transfer (CT)
bands involving an acceptor orbital mainly centered on the peripheral
moiety of the dithioxamide ligands [20]. Visible ferrocenylborane-to-
polypyridine CT bands have also been recently reported [20]. So we
tentatively attribute the lowest energy absorption band of 1 and 2 to a
dppf-to-dithioxamide charge transfer (ligand-to-ligand charge transfer,
LLCT) transition, where probably the donor orbital receives partial
contribution from the platinum orbitals. The red-shift of the band in 2 is
justified by the fact that the donor orbital (in this case, the delocalized
orbitals involving the dppf centers) is much easier to be oxidized.
1.80
1.40
1.00
E (V)
0.60
0.20
-0.20
Fig. 2. Cyclic voltammogram (CV) of complex 3 in 1,2-dichloroethane.
Scan rate 200 mV/s.

1934
S. Lanza et al. / Inorganica Chimica Acta 360 (2007) 1929–1934
[10] (a) S. Lanza, G. Bruno, F. Nicolo, R. Scopelliti, Tetrahedron:
Finally, we wish to point out that a further and definite
Asymmetry 7 (1996) 3347;
(b) S. Lanza, G. Bruno, F. Nicolo, A. Rotondo, R. Scopelliti, E.
Rotondo, Organometallics 19 (2000) 2462.
strong indication concerning the nuclearity of 3 comes
from electrochemical data: comparison between cyclic vol-
tammograms of 2, 3, and the trinuclear species
[Ru(Cp)Cl(l-dithioxamide)Pt(dppf)]+ (4) allows to obtain
integer numbers of exchanged electrons for the redox pro-
cesses of 3 only when a molecular weight corresponding to
the structure in Fig. 1 is used. The details of this method-
ology have been reported by Olbrich et al. [22].
`
[11] (a) S. Lanza, G. Bruno, F. Nicolo, G. Callipari, G. Tresoldi, Inorg.
Chem. 42 (2003) 4545;
(b) S. Lanza, G. Callipari, F. Loiseau, S. Serroni, G. Tresoldi, Inorg.
Chem. 44 (15) (2005) 6717.
[12] Y.N. Kukushikin, Y.E. Viaz’menskii, L.I. Zorina, Russ. J. Inorg.
Chem. (1968) 835.
[13] R.N. Hurd, G. De La Mater, G.C. McElheny, R.J. Tuner, V.H.
Wallingford, J. Org. Chem. 26 (1961) 3980.
[14] (a) I.R. Whittall, A.M. McDonagh, M.G. Humphrey, M. Samoe, in:
F.G.A. Stone, R. West (Eds.), Advances in Organometallic Chem-
istry, vol. 42, Academic Press, San Diego, CA, 1998, p. 291;
(b) O. Crespo, M.C. Gimeno, P.G. Jones, A. Laguna, C. Sarroca,
Chem. Commun. (1998) 1481;
References
[1] V. Balzani, A. Credi, M. Venturi, Molecular Devices and Machines,
Wiley-VCH, 2003.
[2] O. Kahn, Adv. Inorg. Chem. 43 (1995) 179.
[3] (a) Y. Pei, Y. Journaux, O. Kahn, Inorg. Chem. 27 (1988) 399;
(b) Y. Pei, Y. Journaux, O. Kahn, Inorg. Chem. 28 (1989) 100;
(c) M. Mitsumi, H. Okawa, H. Sakiyama, M. Ohba, N. Matsumoto,
T. Kurisaki, H. Wakita, J. Chem. Soc., Dalton Trans. (1993) 2991;
(d) Y. Pei, Y. Journaux, O. Kahn, A. Dei, D. Gatteschi, J. Chem.
Soc., Chem. Commun. (1986) 1300;
(c) K. Severin, R. Bergs, W. Beck, Angew. Chem., Int. Ed. Engl. 37
(1998) 1635;
(d) M.A. Zhuravel, D.S. Gluck, L.M. Liable-Sands, A.L. Rheingold,
Organometallics 17 (1998) 574.
[15] G. Rosace, G. Bruno, L.M. Scolaro, F. Nicolo’, S. Sergi, S. Lanza,
Inorg. Chim. Acta 208 (1) (1993) 59.
[16] G. Bandoli, A. Dolmella, Coord. Chem. Rev. 209 (2000) 161.
[17] W.Y. Wong, G.L. Guo-Liang Lu, K.H. Ka-Ho Choi, J. Organomet.
Chem. 659 (2002) 107.
[18] A.W. Verstuif, J.H. Nelson, L.W. Cary, Inorg. Chem. 15 (1976) 732.
[19] D.-Y. Noh, E.-M. Seo, H.-J. Lee, H.-Y. Jang, M.-G. Choi, Y.H.
Kim, J. Hong, Polyhedron 20 (2001) 1939.
[20] D. Osella, L. Milone, C. Nervi, M. Ravera, J. Organomet. Chem. 488
(1995) 1.
`
(e) J. Ribas, C. Diaz, R. Costa, Y. Journaux, C. Mathoniere, O.
Kahn, A. Gleizes, Inorg. Chem. 29 (1990) 2042.
[4] V.C. Verani, T. Veyhermuller, R. Rentschler, E. Bill, P. Chauduri,
Chem. Commun. 22 (1998) 2475.
[5] K.E. Berg, Y. Pellegrin, G. Blondin, X. Ottenwaelder, Y. Journeaux,
M.M. Canovas, T. Mallah, S. Parson, A. Aukauloo, Eur. J. Inorg.
Chem. (2002) 323.
[6] G.A. Fox, S. Bhattacharya, C.G. Pierpont, Inorg. Chem. 30 (1991)
1689.
[7] W. Paw, R. Eisenberg, Inorg. Chem. 36 (1997) 2287.
[8] H. Liu, M. Du, X. Leng, X.H. Bu, R.H.J. Zhang, Chin. J. Struct.
Chem. 19 (2000) 427.
[21] (a) G. Rosace, G. Giuffrida, G. Guglielmo, S. Campagna, S. Lanza,
Inorg. Chem. 35 (1996) 6816;
(b) L. Ding, K. Ma, G. Durner, M. Bolte, F. Fabrizi de Biani, P.
¨
Zanello, M. Wagner, J. Chem. Soc., Dalton Trans. (2002) 1566.
[22] F. Olbrich, J. Kopf, E. Weiss, Angew. Chem., Int. Ed. Engl. 32 (1993)
1077.
[9] Y.Y. Qiu, Y. Nakano, G.R. Frauenhoff, D.R. Whithcomb, F.
Takusawa, D.H. Busch, Inorg. Chem. 30 (1991) 1503.