Organogold(III) metallacyclic chemistry. Part 2. Synthesis and characterisation of the first gold(III) ureylene complexes. Crystal structures of [{C6H3(CH2 NMe2)-2-(OMe)-5}Au{RNC(O)N R}] (R=Ph and C(O)Me

Article abstract of DOI:10.1016/S0022-328X(97)00712-2

The reactions of the cyclometallated gold complex [Au{C₆H₃(CH₂NMe₂)-2-(OMe)-5}Cl ₂] 3a with N,N'-diphenylurea or N,N'-diacetylurea using an excess of silver(I) oxide in refluxing dichloromethane gave

Full text of DOI:10.1016/S0022-328X(97)00712-2

Journal of Organometallic Chemistry 557 (1998) 231–241
Organogold(III) metallacyclic chemistry. Part 2. Synthesis and
characterisation of the fi¸rs¹t¹g¹ol¹d¹(I¹II¹) ¹u¹re¹yl¹en¹e¹c¹oºmplexes. Crystal
structures of [{C6H3(CH2NMe2)−2−(OMe)−5}Au{RNC(O)NR}]
»¹¹¹¹¹¹¼
(R=Ph and C(O)Me
Maarten B. Dinger, William Henderson *
Department of Chemistry, The Uni6ersity of Waikato, Pri6ate Bag 3105, Hamilton, New Zealand
Received 3 October 1997
Abstract
¸¹¹¹¹¹¹º
The reactions of the cyclometallated gold complex [Au{C6H3(CH2NMe2)-2-(OMe)-5}Cl2] 3a with N,N’-diphenylurea or
N,N’-diacetylurea using an excess of silver(I) oxide in refluxing dichloromethane gave the aura-ureylene complexes
¸¹¹¹¹¹¹¹¹¹¹¹¹¹º
[{C6H3(CH2NMe2)−2−(OMe)−5}Au{NRC(O)NR}] R=Ph 6a and R=Ac [Ac=C(O)Me] 6b respectively, containing the
»¹¹¹¹¹¼
Au¸–¹N¹–¹C¹(O¹)¹–ºN four-membered ring. This ring system has not been reported for gold previously, but is known for other metals
e.g. Pt, Pd, Rh, Ir, Ru, Os and Re. The complex 6a was also successfully synthesised by reaction of 3a, N,N’-diphenylurea and
sodium hydride in THF at room temperature. The complexes have been fully characterised by 1H- and 13C-NMR and IR
spectroscopies, electrospray mass spectrometry, and single crystal X-ray structures are reported for both complexes. The latter
shows the aura-ureylene rings of both complexes are planar. Reaction of 6a with phenyl isocyanate gave the expected
¸¹¹¹¹¹¹¹¹¹¹¹¹¹¹º
six-membered ring insertion product [{C6H3(CH2NMe2)−2−(OMe)−5}Au{NPhC(O)NPhC(O)NPh}] 11, although dimethyl
»¹¹¹¹¹¹¹¹¹¹¹¹¹¼
acetylenedicarboxylate proved unreactive. Pure samples of 11 readily deinsert phenyl isocyanate in solution, to regenerate 6a. This
is also clearly observed in electrospray mass spectrometry studies. The biological properties of 6b have been assessed and the
compound possesses high cytotoxicity and moderate antitumour activity. © 1998 Elsevier Science S.A. All rights reserved.
Keywords: Gold; Metallacycle; Crystal Structure; Ureylene
phenanthroline) with the dilithio salt of N,N%-dipheny-
lurea [3]. Dinuclear iron complexes [6] and tri- and
1. Introduction
tetra-nuclear palladium complexes [7], containing bridg-
Ureylene complexes of transition metals incorporat-
¸¹¹¹¹¹¹¹º
ing urea dianions have also been prepared.
ing the M–N–C(O)–N ureylene ring system are well
We recently reported that a range of platinum(II)
known. Examples of ruthenium, palladium, platinum,
ureylene complexes 1a–1h (summarised in Scheme 1)
rhodium, iridium, rhenium and osmium complexes
have been synthesised by a variety of methods. These
can also be synthesised, in good yield, by reaction of
platinum(II) chloride complexes with the appropriate
include reactions of low-valent metal complexes with
parent urea in the presence of silver(I) oxide [8]. The
various isocyanates or organic azides [1–4], by the
presence of one sufficiently acidic urea hydrogen atom
oxidative addition of N,N%-di-p-toluene-sulfonylurea
[2,3,5] and by reaction of [Pd(OAc)2(phen)] (phen=o-
appeared to be essential, with no reaction occurring
between [PtCl2(COD)] (COD=cyclo-octa-1,5-diene)
and N,N%-dimethylurea. However, if N-phenyl-N%-
methyl urea or N-acetyl-N%-methyl urea were used, high
* Corresponding author. Fax: +64
w.henderson@waikato.ac.nz
7
8384219; e-mail:
0022-328X/98/$19.00 © 1998 Elsevier Science S.A. All rights reserved.
PII S0022-328X(97)00712-2

232
M. B. Dinger, W. Henderson / Journal of Organometallic Chemistry 557 (1998) 231–241
date appear to mimic many silver(I) oxide reactions
undergone by L2PtCl2 complexes. For example, reac-
tion of 3a with diphenacyl sulfone or dimethyl 1,3-ace-
tonedicarboxylate in refluxing dichloromethane in the
presence of silver(I) oxide afforded in high yields the
complexes 4 and 5 [12], respectively, analogous to the
platinum(II) systems [13,14]. Additionally, recent publi-
cations [15–17] regarding the biological activity of
gold(III) compounds, which often rival those of plat-
inum, have further enhanced gold(III) organometallic
complexes as lucrative systems, demanding closer inves-
tigation.
Gold(I) amide complexes are relatively sparse in the
literature, but compounds derived for example from
barbituric acid [18], succinimide [19], purines [20], and
phthalimide [21] have been reported. Fewer still have
been reported for gold(III) [15,22], despite the ability of
amide ligands to stabilise high oxidation states [e.g.
copper(III) [23,24].
Here we report our studies on the silver(I) oxide and
sodium hydride mediated syntheses of the first gold
ureylene complexes, and their reactivity toward
dimethyl acetylenedicarboxylate (DMAD) and phenyl
isocyanate. The biological activity of one derivative is
also reported.
Scheme 1.
2. Results and discussion
¸¹¹¹¹¹º
yields
of
¸¹¹¹¹¹¹º [Pt{NPhC(O)NMe}(COD)]
1e
or
The silver(I) oxide mediated reactions of
¸¹¹¹¹¹¹¹º
[Pt{NAcC(O)NMe}(COD)] 1f respectively could be
obtained.
[Au{C6H3(CH2NMe2)-2-(OMe)-5}Cl2 3a with either
N,N%-diphenylurea or N,N%-diacetylurea in refluxing
dichloromethane for 2 h, afford in good yields the
ureylene complexes 6a and 6b, respectively. These reac-
tions proceed markedly faster than for the synthesis of
platinum ureylene complexes, where typically a reaction
time of 24 h was required for completion. The com-
pound 6a forms bright orange needles, whereas 6b is
isolated as cream-coloured blocks. This striking colour
difference can be attributed to the conjugation im-
parted by the phenyl rings of the ureylene ligand
present in 6a, and was also noted for the analogous
platinum ureylene systems [8]. The synthesis of 6a was
also achieved using sodium hydride as the base in THF
at room temperature.
Ureylene complexes have also been inadvertently
synthesised by reaction of [PtCl2(COD)] with N,N%,N%%-
triacetylguanidine in the presence of silver(I) oxide,
during our studies into the synthesis of guanidine dian-
ion complexes [9], which are isoelectronic with ureylene
systems. Apparently the guanidine hydrolysed under
the reaction conditions used, to give the ureylene com-
plex 2, which was isolated as the sole product. The
ureylene complexes were also spectroscopically ob-
served, although in smaller amounts, for the iridium
and osmium systems studied. When analogous ruthe-
nium and rhodium compounds were used, the expected
guanidine complex was the only reaction product. Di-
rect synthesis of the ureylene complexes was readily
achieved in near quantitative yields by reaction of the
appropriate metal precursor, N,N%-diacetylurea and sil-
ver(I) oxide.
The complexes 6a and 6b were unambiguously char-
acterised on the basis of elemental microanalysis, NMR
spectroscopy, electrospray mass spectrometry, and sin-
gle crystal X-ray crystallographic studies.
Recently we have begun to focus our attention on
the metallacyclic chemistry of gold(III), which despite
being isoelectronic with platinum(II) has been largely
neglected in this important area ¸o¹f¹c¹h¹em¹¹is¹trºy. The
cyclometallated gold precursors [Au{C6H3(CH2NMe2)-
2-R-5}Cl2] R=OMe 3a [10] or R=H 3b [11] have
proven to be good analogues of platinum(II), and to
The crystal structures of both complexes were carried
out in order to fully characterise the reaction product
of both N,N%-diphenylurea and N,N%-diacetylurea dian-
ions with 3a, the geometry of the compounds, and for
comparison with the ureylene structures determined for
platinum and palladium. Additionally, complex 6b
showed a very strong [2M+H]+ion in the electrospray

M. B. Dinger, W. Henderson / Journal of Organometallic Chemistry 557 (1998) 231–241
233
mass spectrum (see Section 2.3), and a possible di-
2.1. Discussion of the structures of 6a and 6b
merisation of the complex was considered. It has
been shown in our laboratory that the reaction of
NCCH2C(O)NHC(O)OEt with 3a gives initially the
monomer complex 7, but eventually crystallises as the
poorly soluble dimer 8, thus forming an eight-membered
ring [25]. Diagnostic distinction between monomer or
dimer systems is not readily achievable by spectroscopic
methods. Also, the determination of the structures of
both 6a and 6b allows direct comparison of two elec-
tronically different ureylene substituents.
The structures confirm the expected monomeric
reaction products analogous to ureylene complexes re-
ported previously for platinum [3,8] and palladium(II)
[4]. Selected bond lengths and angles are presented in
Tables 1 and 2, respectively, with ORTEP perspective
views of 6a and 6b shown in Figs. 1 and 2 together with
the atom numbering schemes. The equivalent bond
lengths and angles for 6a and 6b are remarkably similar,
with the majority being identical within two stan-

234
M. B. Dinger, W. Henderson / Journal of Organometallic Chemistry 557 (1998) 231–241
Table 1
dard deviations. This shows that the complexes
formed by the N,N%-diphenylurea and the N,N%-di-
acetylurea dianions, despite being somewhat electroni-
cally different, nonetheless give extremely similar
structures.
1
˚
Selected interatomic bond lengths (A) for 6a and 6b·2H2O with
estimated standard deviations in parentheses
6a
6b·12H2O
Bond
Length
Bond
Length
The geometry about the gold atom of complex 6a
is, as expected, planar with no atom deviating from a
least-squares plane defined by N(1), N(2), Au, N(3)
AuꢀN(1)
AuꢀN(2)
AuꢀN(3)
AuꢀC(31)
Au.C(1)
2.003(6)
2.092(6)
2.088(6)
2.025(7)
2.584(8)
AuꢀN(1)
AuꢀN(2)
AuꢀN(3)
AuꢀC(11)
Au· · ·C(1)
2.014(5)
2.092(6)
2.090(5)
2.021(6)
2.622(7)
˚
and C(31) by more than 0.065(3) A, for N(1). Simi-
larly, for 6b the analogous plane drawn through N(1),
N(2), Au, N(3) and C(11) has no atom deviating by
˚
PhNC(O)NPh Ligand
AcNC(O)NAc Ligand
more than 0.120(3) A, for N(2). The four-membered
C(1)ꢀO(1)
C(1)ꢀN(1)
C(1)ꢀN(2)
N(1)ꢀC(11)
N(2)ꢀC(21)
1.219(8)
C(1)ꢀO(1)
C(1)ꢀN(1)
C(1)ꢀN(2)
N(1)ꢀC(2)
N(2)ꢀC(4)
C(2)ꢀO(2)
C(4)ꢀO(3)
1.222(9)
ureylene rings formed by Au, N(1), N(2) and C(1) are
also planar, with the largest deviation being 0.014(4)
1.387(10)
1.383(9)
1.404(9)
1.421(9)
1.415(8)
1.392(9)
1.390(9)
1.371(8)
1.230(8)
1.218(9)
˚
and 0.031(4) A [for C(1)] for 6a and 6b, respectively.
The carbonyl groups also lie in this plane, as ex-
pected. Thus the gold ureylene structures, as for plat-
inum, closely resemble the isoelectronic cabonato
complexes [26], rather than the puckered ring system
reported previously for the aurathietane-3,3-dioxide
complex 4 [12]. The major distortion from a regular
square-planar arrangement is the N(1)–Au–N(2) bite
angle, which is 64.3(2)° for 6a and 64.1(2)° for 6b, a
geometrical constraint imposed by the ureylene lig-
and. These are comparable to those previously re-
For these reasons the most probable explanation for
the observation is crystal-packing forces. The two car-
bonyl groups C(2)–O(2) and C(4)–O(3) both point
towards the cyclometallated ligand, thus minimising
steric interactions of the bulkier methyl groups, C(3)
and C(5). The torsion angles of the ureylene phenyl
rings of 6a are 39.1(2) and 49.2(2)° for C(26)–C(21)–
N(2)–C(1) and C(16)–C(11)–N(1)–C(1), respectively.
For 6b the torsion angles are 11.4(2) and 22.3(2)° for
C(5)–C(4)–N(2)–C(1) and C(3)–C(2)–N(1)–C(1).
ported for the platinum and palladium(II) complexes
¸¹¹¹¹¹¹º
1d [8] and [Pd{NPhC(O)NPh}(phen)] 9 [4], which
have bite angles of 64.7(3) and 65.6(1)°, respectively
for the ureylene ligands.
Due to the non-symmetrical nature of the five-
membered cyclometallated ring, the bond lengths
from the gold atom to the ureylene nitrogen atoms,
N(1) and N(2), are not equal. Since the ortho-carbon
on the cyclometallated ring has a higher trans influ-
ence [27] than co-ordinated nitrogen, the Au–N(2)
Table 2
˚
distances are significantly longer [2.092(6) A for both
Selected bond angles (°) for 6a and 6b·12H2O with estimated standard
˚
6a and 6b] than Au–N(1) [2.003(6) A for 6a and
deviations in parentheses
˚
2.014(6) A for 6b]. Comparing these lengths to the
6a
Bonds
6b·12H2O
Bonds
closest related platinum(II) ureylene complex that has
Angle
Angle
had its structure determined 1d, which has Pt–N dis-
˚
˚
tances of 2.048(8) A for Pt–NAd and 2.021(8) A for
N(1)ꢀAuꢀN(2)
N(2)ꢀAuꢀN(3)
N(1)ꢀAuꢀC(31)
N(3)ꢀAuꢀC(31)
64.3(2)
108.6(2)
106.3(3)
80.9(3)
N(1)ꢀAuꢀN(2)
N(2)ꢀAuꢀN(3)
N(1)ꢀAuꢀC(11)
N(3)ꢀAuꢀC(11)
64.1(2)
108.9(2)
107.0(2)
80.7(2)
Pt–NPh, it is clear the average Au–NPh lengths for
˚
6a and 6b [ca. 2.05(1) A] are very much consistent
with the platinum ureylene system.
It is evident from Figs. 1 and 2 that the OMe
groups for 6a and 6b point in opposite directions,
with the Me group orientated towards the ureylene
ligand for 6a and away for 6b. The water of crystalli-
sation [O(5)] in 6b [which is hydrogen-bonded to O(3)
PhNC(O)NPh ligand
AuꢀN(1)ꢀC(1)
AuꢀN(2)ꢀC(1)
AuꢀN(1)ꢀC(11)
AuꢀN(2)ꢀC(21)
N(1)ꢀC(1)ꢀN(2)
N(1)ꢀC(1)ꢀO(1)
N(2)ꢀC(1)ꢀO(1)
C(1)ꢀN(1)ꢀC(11)
C(1)ꢀN(2)ꢀC(21)
AcNC(O)NAc ligand
AuꢀN(1)ꢀC(1)
AuꢀN(2)ꢀC(1)
AuꢀN(1)ꢀC(2)
AuꢀN(2)ꢀC(4)
N(1)ꢀC(1)ꢀN(2)
N(1)ꢀC(1)ꢀO(1)
N(2)ꢀC(1)ꢀO(1)
C(1)ꢀN(1)ꢀC(2)
C(1)ꢀN(2)ꢀC(4)
N(1)ꢀC(2)ꢀO(2)
N(2)ꢀC(4)ꢀO(3)
97.7(5)
93.9(5)
98.2(4)
95.6(4)
137.3(5)
143.1(5)
103.9(6)
127.1(7)
128.9(8)
124.6(6)
122.5(6)
133.8(4)
133.4(5)
102.0(6)
128.8(6)
129.1(7)
125.2(6)
127.1(6)
119.7(6)
121.4(7)
˚
(O· · ·O distance 2.813(9) A)] is 3.627(9) and 4.424(9)
˚
A from O(2) and O(4) (in adjacent molecules), respec-
tively, so the orientation of the OMe group is proba-
bly not directed by hydrogen-bonding interactions.
Additionally, the phenyl group in 6a is bulkier than
the C(O)Me in 6b, so steric effects are also unlikely.

M. B. Dinger, W. Henderson / Journal of Organometallic Chemistry 557 (1998) 231–241
235
¸¹¹¹¹¹¹¹¹¹¹¹¹¹º
Fig. 1. Molecular structure of [{C6H3(CH2NMe2)−2−(OMe)−5}Au»{N¹P¹h¹C¹(O¹)¼NPh}] 6a, showing the atom numbering scheme.
2.2. Spectroscopic characterisation
edly due to the deshielding of this moiety by the
electronegative metal centre [the electronegativity of
gold(III) is quite high [28]. The same effect is noted, but
to a lesser extent, for the remaining N–C groups, and
those trans to the higher trans-influence C, are
deshielded to a greater extent than those trans to N, as
expected. This information can be used to unambigu-
ously assign signals in these positions.
A combination of 2D experiments (HSQC, HMBC,
COSY and NOESY) were used to unambiguously as-
1
sign the H- and 13C-NMR spectra of 6a and 6b. Due
to the differing trans-influence groups on the or-
thometallated ligand, the ureylene ligand, despite being
symmetrical itself, gives two sets of resonances for its
substituents. These could be readily assigned by 1D
NOE difference or 2D NOESY experiments, which
show clear through space correlation to either the para-
methoxy group, or the NMe groups. Table 3 shows a
comparison of the relevant 13C chemical shifts of start-
ing materials and products.
The IR spectrum (in the 1580–1700 cm−1 region) of
the diphenyl ureylene complex 6a, bears a remarkable
similarity to the platinum ureylene complexes 1a–1h, in
showing two bands at 1650 and 1592 cm−1. This
compares with values recorded for 1a [8] of 1648 and
1594 cm−1. The spectrum recorded for 6b is compli-
cated by the presence of two additional carbonyl moi-
eties and is consequently of little diagnostic value.
A large downfield shift of the central carbonyl reso-
nance is observed upon co-ordination. This is undoubt-

236
M. B. Dinger, W. Henderson / Journal of Organometallic Chemistry 557 (1998) 231–241
¸¹¹¹¹¹¹¹¹¹¹¹¹¹º
Fig. 2. Molecular structure of [C6H3(CH2NMe2)−2−(OMe)−5}Au{NAcC(O)NAc}]·12H2O 6b·21H2O, showing the atom numbering scheme.
»¹¹¹¹¹¹¼
The water of crystallisation has been omitted.
2.3. Electrospray mass spectrometry
exhibit appreciable stability even at high cone voltages
(ca. 100 V), as evidenced by the persistence of signifi-
cant parent ions. This contrasts with the platinum
ureylene complexes 1a–1h, which show only fragmenta-
tion ions at high voltages. This is consistent with the
notion of the extra stability imparted by two ring
systems reported for complexes 4 and 5, and has also
been observed previously for platinum(II) complexes
with two ring systems [30].
Electrospray mass spectrometry (ESMS) continues to
provide excellent utility for the analysis of inorganic
and organometallic compounds [29]. As expected for
complexes containing oxygen and nitrogen atoms [in
OMe, CꢁO, and NC(O)N groups] capable of associa-
tion with protons, the complexes 6a and 6b give strong
parent ions [M+H]+, with no observable fragmenta-
tion at low cone voltages (B20 V). The complexes
Fragmentation for 6a and 6b (cone voltage \80 V),
appears to proceed via different mechanisms. For 6a,
loss of both the gold and ureylene moieties is observed
giving rise to a strong peak at m/z 164, assigned as
Table 3
Comparison of 13C chemical shifts of 6a and 6b compared to the
respective urea starting materials
[C6H3(CH2NMe2)-2-(OMe)-5]+. In contrast, 6b shows
¸¹¹¹¹¹¹º
l
13C (ppm)
loss of AcNCO to give [Au{C6H3(CH2NMe2)-2-(OMe)-
5}NHAc]+, analogous to the fragmentation of 1a [8],
which showed loss of phenyl isocyanate to give an ion
assigned as [Pt{NHPh}(COD)]+.
Group
6a
N,N%-dipheny- 6b
N,N%-diacety-
lureaa
lureab
Also noteworthy is the propensity of 6b to give very
strong reproducible [2M+H]+ions, initially tentatively
interpreted as solid state dimer formation. The reason
for this is not fully clear, and was not observed for 6a,
but is most likely due to the presence of three carbonyl
groups in 6b.
Ring CꢁO
169.4 152.6
144.2 139.7
163.1 151.0
176.5 171.8
172.8 171.8
N–C trans C
N–C trans N 141.7 139.7
Au–C 145.0
—
143.0
—
a Recorded in (CD3)2SO.
b Recorded in CDCl3.

M. B. Dinger, W. Henderson / Journal of Organometallic Chemistry 557 (1998) 231–241
237
¸¹¹¹¹¹¹¹¹¹¹¹¹¹º
Fig. 3. Positive-ion electrospray mass spectra of [{C6H3(CH2NMe2)−2−(OMe)−5}Au{NPhC(O)NPhC(O)NPh}] 11 (=M), showing peak
»¹¹¹¹¹¹¹¹¹¹¹¼
assignments. (a) Cone voltage 20 V; (b) Cone voltage 50 V.
The isolated crystalline compound is stable, but when
dissolved in dichloromethane or chloroform, readily
deinserts phenyl isocyanate to regenerate the starting
material 6a. For this reason NMR spectra of 11 were
obtained in the presence of excess phenyl isocyanate.
Since complex 11 contains four inequivalent aromatic
rings, the proton NMR contains many overlapping
signals. For this reason the HOHAHA (Homonuclear
Hartman–Hahn transfer) experiment was utilised. This
unambiguously verified the expected presence of three
phenyl rings, and one trisubstituted aromatic ring. Cou-
pled with the NOESY experiment, this allowed com-
plete assignment of the 1H-NMR spectrum of 11
(detailed in Section 3).
2.4. Reaction of 6a with phenyl isocyanate and
dimethyl acetylenedicarboxylate
The insertion of phenyl isocyanate into the four-
¸¹¹¹¹¹¹¹º
membered Pd–N–C(O)–N ring of the diphenylurey-
¸¹¹¹¹¹¹º
lene palladium complex [Pd{NPhC(O)NPh}(phen)] 9
has been reported to give in high yield the triphenylbi-
¸¹¹¹¹¹¹¹¹¹¹¹º
ureto derivative [Pd{NPC(O)NPhC(O)NPh}(phen) 10
[4].Similarly, some platinum ureylene complexes also
appear to have a high reactivity toward unsaturated
molecules, such as phenyl isocyanate, dimethyl
acetylenedicarboxylate (DMAD), and carbon disulfide,
which rapidly insert into the four membered ureylene
metallacycles to give the expected six-membered ring
insertion products [31].
We have investigated the insertion chemistry of 6a
with both DMAD and phenyl isocyanate. When
DMAD is added to a stirred dichloromethane solution
of 6a, no reaction occurred, even when refluxed for 5 h.
Lack of reactivity towards DMAD insertion has also
been noted for 1a [31]. However when phenyl isocyanate
was added to a solution of 6a, a rapid colour change
from bright orange to bright yellow was observed. The
The complex 11 displays very broad NMe2 and
4
NCH2 1H-NMR signals, and lack of JH,H coupling on
any of the three biureto phenyl rings, despite well
4
resolved JH,H coupling observed on the orthometal-
lated benzylamine ring. Additionally, the 13C-NMR
spectrum shows four resolved signals corresponding to
only two inequivalent CꢁO groups. We believe this is
the result of room temperature fluxionality, and is
probably due the slow inversion of a non-planar six-
¸¹¹¹¹¹¹¹¹¹¹¹¹¹º
product was characterised as the expected insertion
membered Au–N–C(O)–N–C(O)–N ring system
present in 11. Evidence in support of this fluxionality
was determined by the acquisition of a high tempera-
¸¹¹¹¹¹¹¹¹¹¹¹¹¹¹º
complex
[{C6H3(CH2NMe2)−2−(OMe)−5}A-
u{NPhC(O)NPhC(O)NPh}] 11 on the basis of elemen-
t»al¹¹m¹ic¹ro¹a¹n¹al¹ys¹is¹, ¹d¹et¼ailed NMR studies, and ESMS.
ture (55°C) H-NMR spectrum, which showed marked
1

238
M. B. Dinger, W. Henderson / Journal of Organometallic Chemistry 557 (1998) 231–241
sharpening of the NMe2 and NCH2 resonances. Unfor-
tunately the carbonyl signals could not be resolved in
the high temperature 13C-NMR spectrum in the time-
scale of the experiment, but could be expected to coa-
lesce into two signals. The broadening and additional
resonances are unlikely to be caused by rapid insertion/
deinsertion of phenyl isocyanate, since no signal corre-
sponding to the deinserted product 6a is observed in the
ESMS spectrum at low cone voltages. Indeed ESMS
analysis of 11 shows solely strong parent ions ([M+
H]+and [2M+H]+) at cone voltage=20 V (Fig. 3(a)),
with the loss of phenyl isocyanate becoming the domi-
nant ion only with higher voltages (Fig. 3(b)). The
facile loss of phenyl isocyanate has been previously
used to prepare the ureylene complexes seem to proceed
much more rapidly for the gold(III) system, reactivity
toward insertion of phenyl isocyanate and DMAD is
comparable to the platinum(II) ureylene complexes.
Further studies of gold(III) metallacyclic chemistry will
be described in subsequent papers.
3. Experimental
Melting points were measured in air on a Reichert
hotstage apparatus and are uncorrected. Infrared spec-
tra were recorded as KBr discs on a BioRad FTS-40
spectrophotometer. Electrospray mass spectra were
recorded in positive-ion mode on a VG Platform II
instrument, using a 1:1 mixture of MeCN/H2O as the
mobile phase. Elemental analyses were performed by
the Campbell Microanalytical Laboratory, University
of Otago.
observed
in
the
ESMS
analysis
of
¸¹¹¹¹¹¹¹¹¹¹¹¹¹¹º
[Pt{NPhC(O)NMeC(O)NMeC(O)NPh}(COD)] 12, the
reaction product of 1e and phenyl isocyanate [31].
2.5. Biological acti6ity
Proton and all inverse 2D-NMR experiments
(HSQC, Heteronuclear Single Quantum Coherence;
HMBC, Heteronuclear Multiple Bond Correlation;
NOESY; and HOHAHA, Homonuclear Hartmann–
Hahn transfer) were recorded on a Bruker DRX 400
spectrometer at 400.13 and 100.61 MHz for the proton
and carbon channels respectively, with SiMe4 (l 0.0) as
the external standard. The 13C-{1H}-NMR spectra were
recorded on either the instrument above, or on a
Bruker AC300 spectrometer at 75.47 MHz. All NMR
analyses were carried out in CDCl3, with the exception
of the 13C-NMR spectrum of N,N%-diphenylurea which
was acquired in D6-DMSO.
All syntheses were carried out under a dry, oxygen-
free, nitrogen atmosphere, using solvents which were
dri¸ed¹¹an¹d¹¹fr¹es¹hºly distilled prior to use. The compound
[Au{C6H3(CH2NMe2)-2-(OMe)-5}Cl2] 3a [10] was pre-
pared as reported, by transmetallation of the orthomer-
curated complex [Hg{C6H3(CH2NMe2)-2-(OMe)-5}Cl]
[32] with Me4NAuCl4. Dimethyl acetylenedicarboxylate
(Aldrich) and phenyl isocyanate (BDH) were obtained
from commercial sources, and used as received. The
sodium hydride (Koch-Light) used was oil-free and
freshly powdered. N,N%-diphenylurea was prepared by
the condensation of phenyl isocyanate and aniline in
diethyl ether, m.p. 239–240° (lit. [33] 238°). N,N%-di-
acetylurea was synthesised by reacting excess acetyl
chloride with urea in refluxing benzene, m.p. 155–157°
(lit. [33] 154–155°).
Assays for antimicrobial, antiviral and antitumour
activities were determined for 6a. The results are sum-
marised in Table 4. The compound showed high cyto-
toxicity towards the bacteria and fungi tested, and
completely killed the BSC-1 cell line used for antiviral
testing. Antitumour activity was moderate, requiring
7377 ng ml−1 to inhibit 50% of cell growth in P-388
tumour cells.
2.6. Conclusions
The gold(III) ureylene complexes reported here are
analogous to those previously prepared for platinu-
m(II), and the cyclometalled gold(III) compound 3a
continues to provide an excellent platinum(II) ana-
logue. Although the silver(I) oxide mediated reactions
Table 4
(a) Antitumour (P-388) and antiviral/cytotoxicity assay results
IC50a
7377
HSV1b
—
PV1c
—
Cytd
4+e
(b)Antimicrobial/antifungalf activitiesg of 6b (2 mg loaded on disc)
Ec
9
Bs
14
Pa
9
Ca
8
Tm
10
Cr
6
a
The concentration of sample in ng/ml required to reduce the cell
growth of the P-388 leukemia cell line (ATCC CCL 46) by 50%.
b Herpes simplex type I virus (strain F, ATCC VR 733) grown on the
BSC cell line (ATCC CCL 26)
3.1. Preparation of 6a
c Polio type I virus (Pfiser vaccine strain) grown on the BSC cell line.
d Cytotoxicity to BSC cells.
e 4+denotes antiviral or cytotoxic zone over whole well (100% zone).
To
a
Schlenk
flask
(25
containing
ml) was
degassed
added
f
Ec, Escherichia coli; Bs, Bacillus subtilis; Pa, Pseudomonas aerugi-
dic¸h¹lo¹ro¹m¹e¹th¹a¹nºe
nosa; Ca, Candida albicans; Tm, Trichophyton mentagrophytes; Cr,
Cladosporium resinae.
[Au{C6H3(CH2NMe2)-2-(OMe)-5}Cl2] 3a (0.050 g,
0.116 mmol), N,N%-diphenylurea (0.025 g, 0.118 mmol)
and silver(I) oxide (0.092 g, excess). The mixture was
g Inhibition zone as excess radius (mm) from a 6 mm (diameter) disc
containing 2 mg of 6b.

M. B. Dinger, W. Henderson / Journal of Organometallic Chemistry 557 (1998) 231–241
239
refluxed under nitrogen for 2 h. The silver salts were
filtered off and the solvent removed under reduced
pressure giving a bright orange oil. This was recrys-
0.116 mmol), N,N%-diacetylurea (0.017 g, 0.118 mmol)
and silver(I) oxide (0.088 g, excess) were refluxed under
nitrogen for 2 h. The silver salts were filtered off and
the solvent removed under reduced pressure giving a
pale yellow oil. This was recrystallised by slow evapora-
tion of an ether solution at −20°C, to give cream
crystals of 6b (0.044 g, 76%).
tallised by vapour diffusion of ether into
a
dichloromethane solution, to give bright orange needles
of 6a (0.034 g, 51%).
Alternatively, complex 3a (0.050 g, 0.116 mmol),
N,N%-diphenylurea (0.025 g, 0.118 mmol) and sodium
hydride (0.5 g, excess) were added to a Schlenk flask
containing freshly distilled degassed THF (25 ml), and
stirred under nitrogen for 1 h. The resulting bright
orange solution was filtered, and the THF removed
under reduced pressure. The residue was dissolved in
dichloromethane and filtered again. Crystallisation by
slow addition of diethyl ether gave 6a (0.040 g, 60%).
M.p. 185–190°C. Found: C, 48.0; H, 4.4; N, 7.2%;
C23H24N3O2Au requires: C, 48.3; H, 4.2; N, 7.4%. IR:
w(CO region) 1650 (vs), 1592 (m) cm−1. ESMS: (Cone
voltage=5 V) m/z 588 ([MNH4]+, 30%), 572 ([MH]+,
100%). (Cone voltage=50 V) m/z 588 ([MNH4]+,
25%), 572 ([MH]+, 100%), 164 ([C6H3(CH2NMe2)-2-
(OMe)-5]+, 17%). (Cone voltage=100 V) m/z 572
([MH]+, 35%), 164 ([C6H3(CH2NMe2)-2-(OMe)-5]+,
100%). 1H-NMR: (400.13 MHz) l 7.38 (2H, dd,
M.p. 168–171°C. Found: C, 35.9; H, 4.0; N, 8.3%;
C15H20N3O4Au requires: C, 35.8; H, 4.0; N, 8.4%. IR:
w(CO region) 1721 (vs), 1665 (m), 1641 (s), 1629 (s),
1596 (m) cm−1. ESMS: (Cone voltage=20 V) m/z
1007 ([2MH]+, 33%), 504 ([MH]+, 100%). (Cone
voltage =50 V) m/z 1007 ([2MH]+, 38%), 504 ([MH]+,
100%), 419 ([MH–AcNCO]+, 92%). (Cone voltage=
80 V) m/z 1007 ([2MH]+, 8%), 504 ([MH]+, 70%), 419
1
([MH–AcNCO]+, 100%). H-NMR: (400.13 MHz) l
7.38 (1H, d, 4J6%,4%=2.49 Hz, H-6%), 6.96 (1H, d, 3J3%,4%
=
=
),
3
4
8.27 Hz, H-3%), 6.74 (1H, dd, J4%,3%=8.29 Hz, J4%,6%
2.51 Hz, H-4%), 4.17 (2H, s, CH2
6
), 3.75 (3H, s, OCH3
3.45 (6H, s, NCH6 3), 2.57 (3H, s, NC(O)CH3 trans N),
6
2.49 (3H, s, NC(O)CH3 trans C). 13C-NMR: (100.61
MHz) l 176.5 (s, NC(O)CH3 trans C), 172.8 (s,
NC(O)CH3 trans N), 163.1 (s, CꢁO), 157.2 (s, C-5%),
6
143.0 (s, C-1%), 138.1 (s, C-2%), 122.9 (d, C-3%), 122.1 (d,
C-6%), 114.7 (d, C-4%), 75.6 (t, CH2), 55.6 (q, OCH3),
53.8 (q, NCH3), 28.8 (q, NC(O)CH3 trans C), 28.1 (q,
NC(O)CH3 trans N).
4
3J2%%,3%%=8.09 Hz, J2%%,4%%=1.37 Hz, H-2%%,6%%), 7.33–7.25
3
(6H, m, H-3%%,5%%,2%%%,6%%%,3%%%,5%%%), 7.13 (1H, tt, J4%%,3%%/5%%
=
=
4
3
7.23 Hz, J4%%,2%%/6%%=0.60 Hz, H-4%%), 7.03 (1H, d, J3%,4%
8.16 Hz, H-3%), 7.02 (1H, t, J4%%%,3%%%/5%%%=8.49 Hz, H-4%%%),
6.68 (1H, dd, J4%,3%=8.26 Hz, J4%,6%=2.50 Hz, H-4%),
3
3
4
3.3. Preparation of 11
4
6.14 (1H, d, J6%,4%=2.48 Hz, H-6%), 4.11 (2H, s, CH2
6 ),
3.34 (3H, s, OCH3), 2.95 (6H, s, NCH3
6
6
). 13C-NMR:
To a stirred solution of 6a (0.025 g, 0.044 mmol) in
dichloromethane (5 ml) was added two drops of phenyl
isocyanate. A marked colour change from bright or-
ange to bright yellow occurred almost immediately.
Evaporation of the solvent, and recrystallisation of the
residue from dichloromethane and diethyl ether gave 11
as yellow rosettes (0.020 g, 67%). NMR studies revealed
that gradual deinsertion of phenyl isocyanate to regen-
erate 6a occurred over 24 h, if an excess of the iso-
cyanate was not present.
(75.47 MHz) l 169.4 (s, CꢁO), 157.8 (s, C-5%), 145.0 (s,
C-1%), 144.2 (s, C-1%%%), 141.7 (s, C-1%%), 136.5 (s, C-2%),
129.0 (d, C-3%%,5%%), 128.7 (d, C-3%%%,5%%%), 128.5 (d, C-
2%%,6%%), 126.7 (d, C-2%%%,6%%%), 124.9 (d, C-4%), 123.5 (d,
C-4%%%), 123.5 (d, C-3%), 115.2 (d, C-4%), 114.8 (d, C-6%),
73.9 (t, C6 H2), 55.2 (q, OCH3), 52.0 (q, NC6 H3).
M.p. 175–176°C. Found: C, 50.5; H, 4.1; N, 7.6%;
C30H29N4O3Au·12CH2Cl2 requires: C, 50.0; H, 4.1; N,
7.6%. IR: w(CO region) 1648 (vs), 1627 (s), 1592 (m)
cm−1
. ESMS: (Cone voltage=20 V) m/z 1382
([2MH]+, 22%), 691 ([MH]+, 100%). (Cone voltage
=50 V) m/z 1382 ([2MH]+, 8%), 692 ([MH]+, 69%),
572 ([MH–PhNCO]+, 100%). (Cone voltage=80 V)
m/z 1382 ([2MH]+, 5%), 691 ([MH]+, 72%), 572
([MH–PhNCO]+, 100%). H-NMR: (400.13 MHz) l
1
3
8.01 (2H, d, J2%%%,3%%%=8.29 Hz, H-2%%%,6%%%), 7.86 (2H, d,
3J2%%,3%%=8.29 Hz, H-2%%,6%%), 7.44 (2H, d, J2%%%%,3%%%%=8.15
3
3
Hz, H-2%%%%,6%%%%), 7.39 (2H, t, J3%%,2%%/4%%=7.80 Hz, H-3%%),
3
7.38 (2H, t, J3%%%%,2%%%%/4%%%%=6.53 Hz, H-3%%%%/5%%%%), 7.34 (2H,
3.2. Preparation of 6b
3
3
t, J3%%%,2%%%/4%%%=8.08 Hz, H-3%%%/5%%%), 7.31 (1H, t, J4%%%%,3%%%%/
3
In
a similar fashion to the synthesis of 6a,
5%%%%=8.15 Hz, H-4%%%%), 7.11 (1H, t, J4%%,3%%/5%%=7.82 Hz,
H-4%%), 7.11 (1H, t, J4%%%,3%%%/5%%%=8.88 Hz, H-4%%%), 6.99
¸¹¹¹¹¹¹¹º
3
[Au{C6H3(CH2NMe2)-2-(OMe)-5}Cl2] 3a (0.050 g,

240
M. B. Dinger, W. Henderson / Journal of Organometallic Chemistry 557 (1998) 231–241
(1H, t, 3J3%,4%=8.33 Hz, H-3%), 6.69 (1H, dd, 3J4%,3%=8.30
Hz, 4J4%,6%=2.42 Hz, H-4%), 6.20 (1H, d, 4J6%,4%=2.40 Hz,
H-6%), 5.25 (1H, s, CH2Cl2 of crystallisation), 4.22 (2H,
final cycle of the full-matrix least-squares refinement
based on F2, all non-hydrogen atoms were assigned
anisotropic temperature factors, and all hydrogen atom
positions determined by calculation. The refinement
converged with R1=0.0446 for 3233 data with
I]2|(I), 0.0882 for all data; wR2=0.0937
s, br, CH6 2), 3.33 (3H, s, OCH6 3), 2.79 (6H, s, br, NCH6 3).
13C-NMR: (75.47 MHz) l 158.0 (s, C-5%), 156.5 (s,
CꢁO), 156.2 (s, CꢁO), 153.43 (s, CꢁO), 153.36 (s, CꢁO),
145.2 (s, C-1%%), 144.0 (s, C-1%%%), 144.0 (s, C-1%), 139.9 (s,
C-1%%%%), 135.9 (s, C-2%), 130.5 (d, C-2%%%%,6%%%%), 129.1 (d,
C-3%%,5%% or C-3%%%%,5%%%%), 128.5 (d, C-3%%%,5%%%), 128.3 (d,
C-3%%,5%% or C-3%%%%,5%%%%), 127.6 (d, C-4%%%%), 126.9 (d, C-
2%%%,6%%%), 125.8 (d, C-2%%,6%%), 124.6 (d, C-4%% or C-4%%%),
124.3 (d, C-4%% or C-4%%%), 123.4 (d, C-3%), 116.4 (d, C-4%),
{w=1/[|2(F2o)+(0.0339P)2+0.0000P]
where
P=
(F2o+2F2c)/3}, and GoF=1.010. No parameter shifted
in the final cycle. The final difference map showed no
peaks or troughs of electron density greater than +1.33
and −2.02 eA−3 respectively, both adjacent to the gold
˚
atom.
114.7 (d, C-6%), 75.1 (t, CH2), 55.4 (q, OC
6 H3), 52.4 (q,
NCH3).
6
3.5. X-ray crystal structure of 6b ·12H2O
Unit cell dimensions and intensity data were obtained
on a Siemens CCD detector mounted on a P4 diffrac-
tometer at the University of Canterbury. A crystal of
dimensions 0.60×0.35×0.12 mm was used for the
study and a total of 5504 reflections (of which 4068 were
unique) in the range 1.89BqB30.48° were collected at
148(2) K, with monochromatic Mo–Kh X-rays (u=
˚
0.71073 A). The data collection nominally covered over
a hemisphere of reciprocal space, by a combination of
two sets of exposures. In the first each exposure covered
0.3° for a total of 52° in ꢀ. The second run covered 360°
in ƒ (the mounting axis) also using 0.3° increments
between frames. The crystal to detector distance was 4.0
cm. The data set was corrected empirically for absorp-
tion using SADABS34 (Tmax., min. =1.00, 0.57).
Crystal data: C15H2ON3O4Au·12H2O, Mr=513.32,
triclinic, space group P1 (no. 2), a=8.890(1), b=
3.4. X-ray crystal structure of 6a
˚
8.983(1), c=11.201(1) A, h=74.78(1), i=89.77(1),
3
˚
Unit cell dimensions and intensity data were obtained
on a Siemens CCD SMART diffractometer at the
University of Auckland. A crystal of dimensions 0.23×
0.09×0.03 mm was selected for the study and a total of
4909 reflections (of which 4751 were unique) in the
range 1.51BqB28.14° were collected at 130(2) K, with
k=76.32(1)°, U=837.1(1) A , Dc=2.037 g cm−3, Z=
2, F(000)=496, v(Mo–Kh)=8.814 mm−1
.
Solution and refinement:The structure was solved by
the direct methods (TREF) option of SHELXS-96 [35],
and the majority of the heavy atom positions deter-
mined. All further non-hydrogen atoms were located
routinely (SHELXL-96 [36]). In the final cycle of the
full-matrix least-squares refinement based on F2, all
non-hydrogen atoms were assigned anisotropic temper-
ature factors, and all hydrogen atom positions deter-
mined by calculation. The refinement converged with
R1=0.0396 for 3592 data with I]2|(I), 0.0427 for all
˚
monochromatic Mo–Kh X-rays (u=0.71073 A). The
data collection nominally covered over a hemisphere of
reciprocal space, by a combination of three sets of
exposures; each set had a different ƒ angle for the
crystal and each exposure covered 0.3° in ꢀ. The crystal
to detector distance was 5.0 cm. The data set was
corrected empirically for absorption using SADABS
[34] (Tmax., min. =0.86, 0.52).
data;
wR2=0.0971
{w=1/[|2(F2o)+(0.0526P)2+
0.0000P] where P=(F2o+2F2c)/3}, and GoF=1.006.
Crystal data: C23H24N3O2Au, Mr=571.42, mono-
No parameter shifted in the final cycle. The final differ-
clinic, space group P21/c, a=13.5099(5), b=17.683(1),
ence map showed significant peaks and troughs of
3
−3
˚
˚
˚
c=9.028(1) A, i=92.37(1)°, U=2154.7(1) A , Dc=
electron density (maximum +3.24 and −3.94 eA
,
1.761 g cm−3, Z=4, F(000)=1112, v(Mo–Kh)=
respectively), adjacent to the gold atom, which can
attributed to poor absorption correction beyond our
control.
After elucidation of the structure, an extra peak of
electron density was present in the penultimate differ-
ence map that was clearly not part of the main molecule.
6.851 mm−1
.
Solution and refinement: The structure was solved by
the Patterson methods option of SHELXS-96 [35], and
the gold position determined. All further non-hydrogen
atoms were located routinely (SHELXL-96 [36]). In the

M. B. Dinger, W. Henderson / Journal of Organometallic Chemistry 557 (1998) 231–241
241
[12] M.B. Dinger, W. Henderson, J. Organomet. Chem. 547 (1997)
243.
This was successfully modelled as a lone oxygen atom
with half occupancy, almost certainly therefore a water
of crystallisation. The atom, O(5), is 2.813(8) A from
O(3), thus within hydrogen bonding distance. The asso-
ciated hydrogen atoms could not be located in the final
electron difference map.
[13] (a) W. Henderson, R.D.W. Kemmitt, L.J.S. Prouse, D.R. Rus-
sell, J. Chem. Soc. Dalton Trans. (1989) 259. (b) Ibid, (1990) 781.
[14] D.A. Clarke, R.D.W. Kemmitt, M.A. Mazid, P. McKenna, D.R.
Russell, M.D. Schilling, L.J.S. Sherry, J. Chem. Soc. Dalton.
Trans. (1984) 1993.
[15] S.L. Best, T.K. Chattopadhyay, M.I. Djuran, R.A. Palmer, P.J.
Sadler, I. So´va´go´, K. Varnagy, J. Chem. Soc. Dalton Trans.
(1997) 3587.
[16] R.V. Parish, J. Mack, L. Hargreaves, J.P. Wright, R.G. Buckley,
A.M. Elsome, S.P. Fricker, B.R.C. Theobald, J. Chem. Soc.
Dalton Trans. (1996) 69.
[17] R.V. Parish, B.P. Howe, J.P. Wright, J. Mack, R.G. Pritchard,
Inorg. Chem. 35 (1996) 1659.
[18] F. Bonati, A. Burini, B.R. Pietroni, J. Organomet. Chem. 317
(1986) 121.
[19] D.M.L. Goodgame, C.A. O’Mahoney, S.D. Plank, D.J.
Williams, Polyhedron 12 (1993) 2705.
[20] E. Colacio, A. Romerosa, J. Ruiz, P. Roma´n, J.M. Gutie´rrez-
Zorrilla, M. Mart´ınez-Ripoll, J. Chem. Soc. Dalton Trans. (1989)
2323.
[21] P. Lange, A. Schier, J. Riede, H. Schmidbaur, Z. Naturforsch
49b (1994) 642.
˚
Acknowledgements
We thank the University of Waikato for financial
support of this work and the New Zealand Lottery
Grants Board for a grant-in-aid. We also thank Allen
Oliver and Assoc. Prof. C.E.F. Rickard (University of
Auckland) and Professor W.T. Robinson (University of
Canterbury) for collection of the X-ray data sets and
Professor B.K. Nicholson for assistance in the crystal-
lography. G. Ellis (University of Canterbury) is ac-
knowledged for the biological assays, and Dr M.R.
Prinsep for helpful discussion. Professor A.L. Wilkins is
thanked for NMR assistance. M.B.D. thanks the Uni-
versity of Waikato and the William Georgetti Trust for
scholarships.
[22] T. Cheung, T. Lai, C. Che, Polyhedron 13 (1994) 2073.
[23] J. Hanss, H.J. Kru¨ger, Angew. Chem. Int. Ed. Engl. 32 (1996)
2827.
[24] R. Ruiz, C. Surville-Barland, A. Aukauloo, E. Anxolabehere-
Mallart, Y. Journaux, J. Cano, M.C. Mun˜oz, J. Chem. Soc.
Dalton Trans. (1997) 745.
References
[25] W. Henderson, A.G. Oliver, B.K. Nicholson, unpublished results.
[26] M.R. Gregg, J. Powell, J.F. Sawyer, Acta Crystallogr. Sect. C 44
(1988) 43.
[27] T.G. Appleton, H.C. Clark, L.E. Manzer, Coord. Chem. Rev. 10
(1973) 335.
[28] K.M. Mackay, R.A. Mackay, Introduction to Modern Inorganic
Chemistry, 4th ed., Blackie, Glasgow and London, 1989.
[29] R. Colton, A. D’Agostino, J.C. Traeger, Mass Spectrom. Rev. 14
(1995) 79.
[30] L.J. McCaffrey,W. Henderson, B.K. Nicholson, J.E. Mackay,
M.B. Dinger, J. Chem. Soc. Dalton Trans. (1997) 2577.
[31] M.B. Dinger, W. Henderson, J. Chem. Soc. Dalton Trans.,
accepted for publication.
[32] P.A. Bonnardel, R.V. Parish, J. Organomet. Chem. 551 (1996)
221.
[33] R.C. Weast, CRC Handbook of Chemistry and Physics, 59th ed.,
CRC, West Palm Beach, Florida, 1978.
[34] R.H. Blessing, Acta Cryst. A51 (1995) 33.
[35] G.M. Sheldrick, SHELXS-96, Program for Solving X-ray Crystal
Structures, University of Go¨ttingen, Germany, 1996.
[36] G.M. Sheldrick, SHELXL-96, Program for Refining X-ray Crys-
tal Structures, University of Go¨ttingen, Germany, 1996.
[1] P. Braunstein, D. Nobel, Chem. Rev. 89 (1989) 1927.
[2] S. Cenini, G. La Monica, Inorg. Chim. Acta 18 (1976) 279.
[3] W. Beck, W. Rieber, S. Cenini, F. Porta, G. La Monica, J. Chem.
Soc. Dalton Trans. (1974) 298.
[4] F. Paul, J. Fischer, P. Oschenbein, J.A. Osborn, Angew Chem.
Int. Ed. Engl. 32 (1993) 1638.
[5] S. Cenini, M. Pizzotti, F. Porta, G. La Monica, J. Organomet.
Chem. 88 (1975) 237.
[6] (a) P.C. Ellgen, J.N. Gerlach, Inorg. Chem. 13 (1974) 1944. (b)
R.J. Doedens, Inorg. Chem. 7 (1968) 2323.
[7] S. Okeya, S. Koshino, M. Namie, I. Nagasawa, Y. Kushi, J.
Chem. Soc. Chem. Commun. (1995) 2123.
[8] M.B. Dinger, W. Henderson, B.K. Nicholson, A.L. Wilkins, J.
Organomet. Chem. 526 (1996) 303.
[9] M.B. Dinger, W. Henderson, B.K. Nicholson, J. Organomet.
Chem. 556 (1998) 77.
[10] P.A. Bonnardel, R.V. Parish, R.G. Pritchard, J. Chem. Soc.
Dalton Trans. (1996) 3185.
[11] J. Vicente, M.T. Chicote, M.D. Bermu´dez, J. Organomet. Chem.
268 (1984) 191.
.