(Terpyridine)(acetylacetonate)ruthenium(II) complex with a zwitterionic form of phosphoniophenylcyanamide ligand

Article abstract of DOI:10.1016/j.inoche.2003.08.016

The complex [Ru(tpy)(acac)(PPh₄cyd)][PF₆] (where tpy: 2,2′:6′,2″-terpyridine, acac: acetylacetone and PPh ₄cyd: 4-triphenyl-phosphoniophenylcyanamide) has been synthesized. It contains the zwitterionic form of the phosphoniophenylcyanamide ligand. Characterizations were carried out using IR, ES-MS, UV-Vis, electrochemistry and ¹H, ¹³C and ³¹P NMR spectroscopy.

Full text of DOI:10.1016/j.inoche.2003.08.016

Inorganic Chemistry Communications 6 (2003) 1400–1405
www.elsevier.com/locate/inoche
(Terpyridine)(acetylacetonate)ruthenium(II) complex with
a zwitterionic form of phosphoniophenylcyanamide ligand
a
Lucinda Dudd , Matthew Hart , David Ring , Elodie Sondaz ,
a
a
a
a,
b,1
Jacques Bonvoisin *, Yannick Coppel
a
CEMES/CNRS, Molecular Electronics Group, BP 4347, 29 rue Jeanne Marvig, F-31055 Toulouse Cedex 4, France
b
LCC/CNRS, UPR8241, 205 route de Narbonne, F-31077 Toulouse Cedex 4, France
Received 5 June 2003; accepted 21 August 2003
Published online: 28 September 2003
Abstract
The complex [Ru(tpy)(acac)(PPh4cyd)][PF6] (where tpy: 2,20:60,200-terpyridine, acac: acetylacetone and PPh4cyd: 4-triphenyl-
phosphoniophenylcyanamide) has been synthesized. It contains the zwitterionic form of the phosphoniophenylcyanamide ligand.
1
Characterizations were carried out using IR, ES-MS, UV–Vis, electrochemistry and H, 13C and 31P NMR spectroscopy.
Ó 2003 Elsevier B.V. All rights reserved.
Keywords: Ruthenium(II) complex; Cyanamide; NMR; Zwitterion; Phosphonium
1. Introduction
synthesis of a new neutral building block [Ru(tpy)(acac)
Ipcyd)] and an unexpected result, the obtaining of
[Ru(tpy)(acac)(PPh4cyd)]þ which contains a zwitterionic
form of the phosphoniophenylcyanamide ligand. This
last is shown in Scheme 1.
As well as the synthetic work, we also present here the
characterization of the complex through a series of pro-
cedures including, UV–Vis spectroscopy, mass spec-
trometry, electrochemistry and detailed NMR.
Our previous work led to the synthesis of [Ru(tpy)
(bpy)Ipcyd)]þ (where Ipcyd: 4-Iodophenylcyanamide
anion) [1]. The strategy was to find good candidates
giving a strong interaction between distant paramag-
netic sites following the building blocks approach [2].
Unfortunately, the desired paramagnetic form of the
complex (Ru(III) state) could not be reached because of
the proximity in energy of the ruthenium and cyanamide
orbitals making the ruthenium impossible to oxidize. In
the present work, we have replaced the bipyridine ligand
with a better donor group such as acetylacetone which
has shown its capacity to tune the redox potential of
ruthenium complexes [3]. The Iodophenylcyanamide li-
gand was chosen for two reasons: firstly, because of the
NCNÕs efficiency to mediate electronic or magnetic in-
teractions [4] and secondly, because iodide has proven to
be an efficient reaction center for further metal-catalyzed
coupling reactions [5]. Here, we show the successful
2. Experimental
2.1. Materials
All chemicals and solvents were reagent grade or
better. [Ru(tpy)Cl3] [6] and 4-Iodophenylcyanamide
(IpcydH) [1] were prepared according to literature pro-
cedures. Weakly acidic Brockmann I type alumina (Al-
drich) was used. All manipulations were carried out
under an atmosphere of dry Ar.
2.2. Physical measurements
* Corresponding author. Tel.: +33-5-62-25-78-52; fax: +33-5-62-25-
79-99.
E-mail address: jbonvoisin@cemes.fr (J. Bonvoisin).
Electrospray spectra (positive mode) were obtained
with a Perkin–Elmer Sciex (Nermag R10-R10). IR were
1
Tel.: +33-5-61-17-54-28; fax: +33-5-61-30-03.
1387-7003/$ - see front matter Ó 2003 Elsevier B.V. All rights reserved.
doi:10.1016/j.inoche.2003.08.016

L. Dudd et al. / Inorganic Chemistry Communications 6 (2003) 1400–1405
1401
pulse used were all gaussians truncated at the 1% level.
Durations of the shaped pulses were kept as short as
possible.
8
7
6
5
2.3. Syntheses of complexes
3
P
4
2.3.1. [Ru(tpy)(acac)Cl ] (1) (blue grey)
2
The synthesis was adapted from a literature proce-
dure [3]. Ethanol (150 ml) was added to Ru(tpy)Cl3
(440 mg, 1.0 mmol). The apparatus was flushed with Ar,
then triethylamine (1.5 ml, 11 mmol) and acetylacetone
(acac) (0.5 ml, 5 mmol) were added using a plastic sy-
ringe. Upon addition, the dark brown solution turned a
deep red. The solution was heated under reflux for 4 h.
The resulting violet solution was evaporated to dryness
to leave a dark green solid. Dichloromethane (100 ml)
was added and the solution was then stirred under Ar
for 10 min, before being filtered through Celite and
washed with dichloromethane until the filtrate was clear.
The solution was concentrated (ꢀ30 ml) and hexane
(50 ml) was added to it to precipitate the final product of
[Ru(tpy)(acac)Cl] which was then filtered and washed
consecutively with hexane, water and diethyl ether be-
fore being air dried (0.289 g, 61.6%). NMR 1H (CD2Cl2
d ¼ 5:35): 8.72 (2H, d, 5.6 Hz), 8.16 (2H, d, 8.1 Hz), 8.09
(2H, d, 7.9 Hz), 7.84 (2H, ddd, 7.5, 6.3 and 1.7 Hz),
7.53–7.54 (2H, dd, 5.6 and 1.4 Hz and 1H, dd, 7.5 and
8.1 Hz), 5.40 (1H, s), 2.49 (3H, s), 1.27 (3H, s). Anal.
Calcd. (%) for C20H18ClN3O2Ru ꢁ 1/2(CH2Cl2): C, 48.1;
H, 3.6; N, 8.5. Found: C, 48.7; H, 3.6; N, 8.5.
1
N
C
N
N
8
Ru
N
O
1
2
4
5
6
7
N
1
4
O
2
3
3
Scheme 1. Schematic drawing of [Ru(tpy)(acac)(PPh4cyd)]þ.
recorded in KBr pellets on a Perkin–Elmer FT-IR 1725.
UV–Vis electronic spectra were obtained with a Shi-
madzu UV-3100. Cyclic voltammograms were obtained
with an Autolab system (PGSTAT100) (0.1 M tetrabu-
tylammonium hexafluorophosphate, TBAH as sup-
porting electrolyte) at 25 °C. A three electrode cell was
used comprising a 1 mm Pt-disk working electrode, a Pt
wire auxiliary electrode, and an aqueous saturated cal-
omel (SCE) reference electrode. NMR spectra were re-
corded on a Bruker AMX400 spectrometer equipped
with a 5 mm triple resonance inverse probe with dedi-
cated 31P channel operating at 400.13 MHz for 1H,
161.97 MHz for 31P and 100.61 MHz for 13C. All
chemical shifts for 1H and 13C are relative to TMS using
1H (residual) or 13C chemical shifts of the solvent as a
secondary standard. H, 13C {1H}, 13C {1H, 31P} and
31P {1H} spectra were recorded at 293 K in CD2Cl2.
Signal assignments were made with the aid of gradi-
ent-enhanced H COSY45. H–13C correlation spectra
using a gradient-enhanced HMQC sequence (delay was
optimized for JCH of 140 Hz) was obtained with 40
scans per increment. A gradient-enhanced HMBC ex-
periment was performed allowing 62.5 ms for long-range
coupling evolution (176 scans were accumulated). All
2D experiments were realized with phosphorus selective
decoupling. Typically, 2048 t2 data points were collected
for 256 t1 increments. The FID along t1 was zero-filled
to 512 points prior to Fourier transformation.
1D selective DPFGSE-NOE [7] were obtained with a
mixing time of 800 ms. Gradients were sine shaped in all
experiments with durations of 1 or 1.7 ms. The selective
2.3.2. [Ru(tpy)(acac)Ipcyd] (2) (dark blue)
An ethanol/water mixture (200:40 ml) was added to 1
(393 mg, 0.83 mmol), and the blue solution was then
degassed for 10 min, before AgBF4 (486 mg, 2.5 mmol)
was added, at which the solution turned green. The so-
lution was left at reflux for 4 h. The solution was al-
lowed to cool down before being filtered through Celite
and washed with ethanol. The solution was then con-
centrated before being degassed. IpcydH (1.68 g, 6.88
mmol) was added to the solution, and it was then left
under Ar at 40 °C for 48 h, after which the solution was
then allowed to cool before being evaporated to dryness.
This was then dissolved in dichloromethane and was
purified using column chromatography (acidic alumina,
solvent: dichloromethane, eluent: 1% ethanol:dichlo-
romethane). The column produced nine bands, the first
was a pale yellow, corresponding to the free ligand, the
second was a pale violet/pink, followed by a white band.
After this was a dark blue band, then another white
band followed by a violet band, a dark blue, a pale blue
and a red band. The first dark blue band was collected
and then evaporated to dryness to yield a lacquer of 2
1
1
1
1
1
(120 mg, 25%). NMR H (CD2Cl2 d ¼ 5:35): 8.70 (2H,
d, 5.6 Hz), 8.15 (2H, d, 8.1 Hz), 8.08 (2H, d, 8.1 Hz),
7.90 (2H, ddd, 8.1, 7.5 and 1.5 Hz), 7.60–7.52 (2H, dd,

1402
L. Dudd et al. / Inorganic Chemistry Communications 6 (2003) 1400–1405
7.5 and 5.6 Hz and 1H, t, 8.1 Hz), 7.06 (2H, d, 8.7 Hz),
5.88 (2H, d, 8.7 Hz), 5.37 (1H, s), 2.52 (3H, s), 1.37 (3H,
s). MS (ES, CH3CN): [M + Hþ]þ ¼ 678.1; calcd. 677.99.
[Ru(tpy)(acac)(HCN) + Hþ]þ ¼ 462.0. IR: mNCN ¼ 2176
cmꢂ1. Anal. Calcd. (%) for C27H22IN5O2Ru: C, 47.9; H,
3.3; N, 10.4. Found: C, 47.3; H, 2.5; N, 9.8.
mNCN ¼ 2181:4 cmꢂ1, mPF ¼ 847:0 cmꢂ1. MS (ES) m=z:
6
calcd. for C45H37N5O2PRu (M–PF6)þ 812.17. Found
812.3. Anal. Calcd. (%) for C45H37F6N5O2P2Ru: C,
56.5; H, 3.9,; N, 7.3. Found: C, 55.3; H, 3.7; N, 7.1.
3. Results and discussion
2.3.3. [Ru(tpy)(acac)(PPh4cyd )][PF6] (3) (violet)
Dimethylformamide (5 ml) was added to NiCl2 ꢁ
6H2O (43 mg, 0.18 mmol) and triphenyl phosphine
(191 mg, 0.71 mmol), which caused the solution to turn
blue. The solution was degassed before activated zinc
(80 mg, 1.24 mmol) was added to the solution, which
immediately turned green. The solution was then left for
90 min, during which time the solution changed color
from green, through yellow and orange, to orange/red
before it was heated to 40 °C, and after 30 min of
heating the solution had turned brick red. 2 (120 mg,
0.18 mmol) was then dissolved in dimethylformamide,
degassed, and added to the brick red solution, which
turned dark blue. The reaction was then left overnight.
A saturated solution of KPF6 (15 ml) was added to the
solution to form a precipitate, which was filtered
through a sinter and left to air dry. The precipitate was
redissolved in dichloromethane and purified using col-
umn chromatography (acidic alumina, solvent: dichlo-
romethane, eluent: 1% ethanol:dichloromethane). At
first there were two separate bands on the column, but
the second band merged with the first to yield an impure
product. This was then recrystalized using dichlorome-
thane/ether to leave a powder corresponding to the title
complex 3 (86 mg, 50%), which was filtered through a
sinter and left to air dry. IR (KBr pellet, cmꢂ1):
3.1. Synthesis
The synthesis of [Ru(tpy)(acac)Cl] (1) and [Ru(tpy)
(acac)Ipcyd] (2) were adapted from literature procedure
[1,3]. A general scheme is shown in Scheme 2.
[Ru(tpy)(acac)Cl] (1) was prepared from the appro-
priate Ru(tpy)Cl3 and the appropriate b-diketone.
Ru(tpy)Cl3 was placed in ethanol with a 5-fold excess of
b-diketone and an 11-fold excess of triethylamine, and
heated at reflux for 4 h. The triethylamine serves as a
reducing agent toward the Ru(tpy)Cl2 and possibly as a
deprotonating agent for the b-diketone. The solvent was
removed, and the residual solid was reprecipitated from
methylene chloride and hexane. Product purity was
evaluated by cyclic voltammetry and thin layer chro-
matography. The chloride ligand on Ru(tpy)(acac)Cl is
easily displaced in refluxing ethanol/water by adding
AgBF4, permitting the synthesis of [Ru(tpy)(acac)L],
where L is an anionic ligand such as 4-Iodophenylcya-
namide (or Ipcyd). This ligand is of particular interest
since it can serve as precursor to bimetallic complexes [8].
The utility of the [Ru(tpy)(acac)Ipcyd] complex as start-
ing material is illustrated here by its capability to form
[Ru(tpy)(acac)(PPh4cyd)][PF6] (3) in the presence of
stoechiometric amount of Ni(II)Cl2, triphenylphosphine
Abs. EtOH
N
N
N
N
Reflux / Ar
Cl
Cl
Ru Cl
N
Ru
O
N
O
Cl
1. AcacH / NEt3
2. CH2Cl2 / Hexane
1
1. AgBF4
2. IpcydH
Water - ETOH / Ar
PF6
3. CH2Cl2 / Hexane
I
P
N
DMF / Ar
C
N
N
N
40˚C
C
O
N
N
N
Ru
O
N
O
O
1. NiCl2.6H20 +
P(Ph)3 + Zn powder
N
Ru
N
2. KPF6
Scheme 2. General scheme of synthesis.

L. Dudd et al. / Inorganic Chemistry Communications 6 (2003) 1400–1405
1403
PPh3
Ph3P
PPh3
Ph
Ph
Ph
Ph
+
[Ni(PPh3)2]
[Ni(PPh3)2]
I
Ni
Ru
P
I
Ph
NCN
Ph
I
NCN
NCN
Ru
Ru
Scheme 3. Mechanism of formation of the phosphonium salt, Ru being the abbreviation for Ru(tpy)(acac).
and zinc powder in DMF. The mechanism of the reac-
tion seems to involve oxidative addition of the aryl halide
to a nickel(0)–phosphane complex, followed by the re-
ductive elimination of the phosphonium ion and a loss of
the halide ion from the metal center as it is shown on
Scheme 3 [9].
Initially, under these conditions, we expected to get
the homometallic dimer [10] and in this aspect, it did not
lead to the formation of the desired complex. Instead of
that, at the homocoupling step the complex appears to
have reacted with the PPh3 to give a phosphine com-
pound. One possible explanation is that the electron rich
environment around the metal and the cyanamide group
were helping to stabilize the positive charge on the
phosphorus thus stabilizing the phosphine compound 3.
performed at 293 K, a temperature at which it was clear
that rotation around all the phosphorus phenyl bonds
1
was rapid on the NMR time scale. All the H and 13C
signals were unambiguously assigned on the basis of
chemical shifts, spin–spin coupling constants, splitting
patterns and signal intensities, and by using 1H–1H
COSY, 1H–13C HMQC and 1H–13C HMBC experi-
ments. The 1H and 13C chemical shifts as well as proton–
proton, proton–phosphorus and carbon–phosphorus
coupling constants are given in supplementary materials.
The 31P NMR spectrum of the complex displayed a sin-
glet at 24.4 ppm.
For the triphenylphosphoniophenylcyanamide unit,
the 1H NMR spectrum (Fig. 1) shows a typical AA0
MM0NX spin system pattern for the triphenylphosphine
protons (A ¼ HPHOS6, M ¼ HPHOS7, N ¼ HPHOS8, X ¼ P)
3
3
with JAM ¼ 3JA M ¼ 7:2 Hz, JMN ¼ 3JM N ¼ 7:5 Hz,
0
0
0
0
3.2. NMR spectroscopy
4JAN ¼ 4JA N ¼ 1:2 Hz, JAA ¼ JMM ¼ 1:0 Hz and
4
4
0
0
0
In the present study, 1H, 13C and 31P NMR spec-
troscopy were used for the characterization of the
[Ru(tpy)(acac)(PPh4cyd)][PF6] complex. This study was
JAM ¼ JA M ¼ 0:6 Hz obtained from computer simu-
5
5
0
0
lation. The proton spectrum shows also an AA0MM0X
pattern for the phenylcyanamide protons (A ¼ HPHOS2
,
Fig. 1. 1H NMR spectra of [Ru(tpy)(acac)(PPh4cyd)][PF6] complex in CD2Cl2 at 293 K (* are mainly due to traces of unreacted materials such as
[Ru(tpy)(acac)Cl]).

1404
L. Dudd et al. / Inorganic Chemistry Communications 6 (2003) 1400–1405
Table 1
UV–Vis and electrochemical data
Complex
UV–Vis dataa
k/nm (e/103 Mꢂ1 cmꢂ1
Redox potentialsa;b
1=2/mV (DE, mV)
)
E
1
2
3
279
320
402
563
629
251
(83)
244
(88)
381
(68)
(26.6)
277
(51)
(20.3)
301
(38)
(7.8)
390
(5.3)
568
(3.6)
634
(10.2)
382
(37.7)
(5.5)
560
(5.1)
624
277
(62.9)
316
(57.1)
(9.4)
(7.7)
a In CH2Cl2.
b Versus SCE, 0.1 V/s.
M ¼ HPHOS3, X ¼ P) with 3JAM ¼ 3JA M ¼ 8:4 Hz, 4JAA
¼
cationic charge on the phosphonium substituent com-
pared to the unsubstituted ones. In addition, for com-
pound 2, one can notice an irreversible process with an
anodic peak around 1000 mV and a cathodic peak at 650
mV, These two waves are very probably due to irre-
versible oxidation of Iodophenylcyanamide ligands as
demonstrated by our previous studies on [Ru(tpy)(bpy)
Ipcyd] [1]. These irreversible waves are also present for
compound 3 but with the anodic peak shifted to higher
potentials (1400 mV) even though the cathodic peak
stays at the same potential (650 mV).
0
0
0
JMM ¼ 2:2 Hz and 5JAM ¼ JA M ¼ 0:0 Hz.
4
5
0
0
0
In the 13C{1H} NMR spectrum (supplementary ma-
terials), the phenyl carbons are observed in the usual
aromatic region except for the strongly shielded CPHOS1
signal (d 97.0) and the strongly deshielded CPHOS4 signal
(d 162.8). These chemical shifts can be tentatively ex-
plained by p electron density transfer within the delo-
calized p system. The cyanamide carbon signal appears
as a broad singlet (d 123.4, half-height linewidth 8 Hz)
[11].
The terpyridine unit is symmetrical with one set of
signals for the central pyridine and one set of signals for
the two equivalent terminal pyridines.
UV–Vis data are shown in Table 1. Absorption bands
look like the same in shape but differ in intensity. Three
regions can be distinguished. In the UV region, the two
intense bands (around 280 and 320 nm) are attributable
to terpyridine ligand (p ! pꢃ). In the visible region,
the two lower energy absorption bands (between 560
and 634 nm) are more MLCT in character, i.e.,
dpðRuðIIÞÞ ! pꢃðtpyÞ [1]. The band which lies between
382 and 402 nm in the three complexes is more con-
troversial. Its intensity in complex 3 and presence in
complex 1 suggest that it could be an intra-ligand charge
transfer which is enhanced in complex 3 because of the
strongly polarized Nꢂ–Ph–Pþ entity.
The acetylacetone unit is unsymmetrical with two
C@O resonances CACAC1, CACAC3 (dC 187.2, 187.1), two
methyl resonances CACAC4 and CACAC5 (dH=dC 1.38/27.2
and 2.48/28.5) and one methine resonance CACAC2
1
(dH=dC 5.38/99.3) in the H and 13C spectra. The de-
shielded HACAC4 resonance can be explained by a lo-
calization of these protons in the deshielding cone of the
pyridine rings and possibly also of the phenylcyanamide
ring. This hypothesis was confirmed by the observation
of weak NOE correlations of HACAC4 with the terpyri-
dine HTERP1 protons and with the phenylcyanamide
HPHOS2 in 1D NOE experiments (supplementary mate-
rial). The same 1D NOE experiment applied to HACAC5
shows only a NOE correlation with the methine
In conclusion, this work presents the successful cou-
pling of a phosphine onto a new iodo-functionalized
Ru(II) complex leading to a stable cationic ruthenium
complex containing a zwitterionic form of a phos-
phoniophenylcyanamide ligand which represents an
advancement in chemical knowledge. Progress are made
to use this iodo-functionalized Ru(II) complex, [Ru(tpy)
(acac)Ipcyd)] as starting material for building long di-
nuclear complex which can be seen as molecular wires
useful for molecular electronics.
HACAC2
.
3.3. Redox and spectroscopic properties
Cyclic voltammetry (CV) data for the complexes are
given in Table 1. The E1=2 potentials were determined
from the average of the anodic and cathodic peak po-
tentials. All the waves are reversible with peak to peak
separation of about 70–80 mV. Compounds 1 and 2 have
almost the same E1=2 potentials (251 vs 244 mV) because
of the anionic character of the ligand (Cl or Ipcyd) which
stabilizes the Ru(III) state. However, for compound 3,
the wave is shifted to more positive potentials (381 mV),
ca. 150 mV more than 1 and 2. This would agree with the
increasing acceptor character of the ligand due to the
Supplementary materials
Table with NMR spectral data for [Ru(tpy)(acac)
(PPh4cyd)][PF6] complex in CD2Cl2; Figure A: 13C
NMR spectrum of [Ru(tpy)(acac)(PPh4cyd)][PF6] com-
plex in CD2Cl2 at 293 K; Figure B: 1D DPFGSE-NOE

L. Dudd et al. / Inorganic Chemistry Communications 6 (2003) 1400–1405
1405
[4] R.J. Crutchley, Adv. Inorg. Chem. 41 (1994) 273;
with selection of HPHOS2 proton in CD2Cl2 at 293 K
(mixing time 600 ms); Figure C: idem with HACAC4
Figure D: idem with HACAC5; Figure E: idem with
HTERP1
M.A.S. Aquino, F.L. Lee, E.J. Gabe, C. Bensimon, J.E. Greedan,
R.J. Crutchley, J. Am. Chem. Soc. 114 (1992) 5130;
A.R. Rezvani, C.E.B. Evans, R.J. Crutchley, Inorg. Chem. 34
(1995) 4600;
;
.
P.J. Mosher, G.P.A. Yap, R.J. Crutchley, Inorg. Chem. 40 (2001)
1189.
Acknowledgements
[5] K. Sonogashira, Y. Tohda, N. Hagihara, Tetrahedron Lett. 50
(1975) 4467;
S. Takahashi, Y. Kuroyama, N. Sonogashira, N. Hagihara,
Synthesis (1980) 627.
We thank the European Union for an ERASMUS
scholarship (L.D., M.H. and D.R.) and the ‘‘Service
dÕanalyse’’ of LCC (Laboratoire de Chimie de Coordi-
[6] B.P. Sullivan, J.M. Calvert, T.J. Meyer, Inorg. Chem. 34 (1980)
1404.
ꢀ
nation) and the ÔService commun de spectrometrie de
[7] K. Scott, J. Keeler, Q.N. Van, A.J. Shaka, J. Magn. Reson. 125
(1997) 302.
ꢀ
masseÕ of UPS (Universite Paul Sabatier).
[8] E. Sondaz, J. Jaud, J.-P. Launay, J. Bonvoisin, Eur. J. Inorg.
Chem. 8 (2002) 1924.
References
[9] G. Torre, A. Gouloumis, P. Vazquez, T. Torres, Angew. Chem.
Int. Ed. 40 (2001) 2895;
B.E. Segelstein, T.W. Butler, B.L. Chenard, J. Org. Chem. 60
(1995) 12;
[1] E. Sondaz, A. Gourdon, J.-P. Launay, J. Bonvoisin. Inorg. Chim.
Acta 316 (2001) 79.
F.E. Goodson, T.I. Wallow, B.M. Novak, J. Am. Chem. Soc. 119
(1997) 12441.
[2] S. Fraysse, C. Coudret, J.-P. Launay, Tetrahedron Lett. 39 (1998)
7873;
[10] P.M. Griffiths, F. Loiseau, F. Puntoriero, S. Serroni, S. Campa-
gna, Chem. Commun. (2000) 2297.
P.J. Connors, D. Tzalis, A.L. Dunnick, Y. Tor. Inorg. Chem. 37
(1998) 1121.
€
[11] S. Ruile, O. Kohle, H. Pettersson, M. Gratzel, New J. Chem.
(1998) 25.
[3] S.J. Slattery, W.D. Bare, D.L. Jameson, K.A. Goldsby, J. Chem.
Soc. Dalton Trans. (1999) 1347.