Synthesis and reactivity of gold(III) complexes containing cycloaurated iminophosphorane ligands

Article abstract of DOI:10.1016/j.ica.2009.04.024

Transmetallation reactions of ortho-mercurated iminophosphoranes (2-ClHgC₆H₄)Ph₂P{double bond, long}NR with [AuCl₄]^- gives new cycloaurated iminophosphorane complexes of gold(III) (2-Cl₂AuC₆H₄)Ph₂P{double bond, long}NR [R = (R,S)- or (S)-CHMePh, p-C₆H₄F, ^tBu], characterised by NMR and IR spectroscopies, ESI mass spectrometry and an X-ray structure determination on the chiral derivative R = (S)-CHMePh. The chloride ligands of these complexes can be readily replaced by the chelating ligands thiosalicylate and catecholate; the resulting derivatives show markedly higher anti-tumour activity versus P388 murine leukaemia cells compared to the parent chloride complexes. Reaction of (2-Cl₂AuC₆H₄)Ph₂P{double bond, long}NPh with PPh₃ results in displacement of a chloride ligand giving the cationic complex [(2-Cl(PPh₃)AuC₆H₄)Ph₂P{double bond, long}NPh]⁺, indicating that the P{double bond, long}N donor is strongly bonded to the gold centre.

Full text of DOI:10.1016/j.ica.2009.04.024

Inorganica Chimica Acta 362 (2009) 3669–3676
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
Synthesis and reactivity of gold(III) complexes containing cycloaurated
iminophosphorane ligands
*
Kelly J. Kilpin, William Henderson , Brian K. Nicholson
Department of Chemistry, University of Waikato, Private Bag 3105, Hamilton, New Zealand
a r t i c l e i n f o
a b s t r a c t
Article history:
Transmetallation reactions of ortho-mercurated iminophosphoranes (2-ClHgC6H4)Ph2P@NR with
[AuCl4]ꢀ gives new cycloaurated iminophosphorane complexes of gold(III) (2-Cl2AuC6H4)Ph2P@NR
[R = (R,S)- or (S)-CHMePh, p-C6H4F, tBu], characterised by NMR and IR spectroscopies, ESI mass spectrom-
etry and an X-ray structure determination on the chiral derivative R = (S)-CHMePh. The chloride ligands
of these complexes can be readily replaced by the chelating ligands thiosalicylate and catecholate; the
resulting derivatives show markedly higher anti-tumour activity versus P388 murine leukaemia cells
compared to the parent chloride complexes. Reaction of (2-Cl2AuC6H4)Ph2P@NPh with PPh3 results in
displacement of a chloride ligand giving the cationic complex [(2-Cl(PPh3)AuC6H4)Ph2P@NPh]+, indicat-
ing that the P@N donor is strongly bonded to the gold centre.
Received 31 October 2008
Received in revised form 6 April 2009
Accepted 17 April 2009
Available online 3 May 2009
Keywords:
Gold complexes
Iminophosphorane complexes
Ligand substitution
X-ray crystal structures
Ó 2009 Elsevier B.V. All rights reserved.
1. Introduction
in 1977 [9]. Later, examples of the ortho-manganated and –rheni-
ated iminophosphorane Ph3P@NPh were reported [10], and
Gold complexes are well-known to show interesting biological
[1–3] and catalytic [4,5] behaviour. Complexes containing cycloau-
rated ligands are of interest for the stabilisation of the gold(III) cen-
tre that they impart. However, the number of ligands that form
cycloaurated gold(III) complexes with C,N donor atoms are rela-
tively few compared with other platinum group metals. The syn-
theses and reactivity of such Au(III) complexes is the subject of a
recent review [6].
ortho-metallated palladium(II) [8,11–13], iridium, rhodium [14]
and main group complexes [15] are also known. Recently, we have
reported the synthesis of cycloaurated Ph3P@NPh 1 from the ortho-
mercurated compound 2 [16]. Following this, a report on a ‘green-
er’ synthetic route to 1 from the corresponding gold(I) complex 3
was published along with evidence that 1 and related compounds
are effective at catalysing the addition of electron rich arenes and
2-methylfuran to methyl vinyl ketone [17]. In addition there are
a number of ortho-metallated derivatives of related silyl-imino-
phosphoranes [15,18–20]. This paper gives new examples of both
cycloaurated and ortho-mercurated N-alkyl and -aryl iminophos-
phoranes; the reactivity and a preliminary assessment of the bio-
logical activity of these compounds is also reported.
Iminophosphoranes (of the general formula R3P@NR0) are
attractive candidates as ligands not only because the phosphorus
atom introduces a powerful (31P NMR) spectroscopic probe, but
also due to their facile synthesis. Using either the Staudinger reac-
tion between an organic azide and a tertiary phosphine ðR3Pþ
N3R0 ! R3P@NR0Þ, or the Kirsanov reaction between a phosphine
dibromide and a primary amine ðR3PBr2þ H2NR0 ! R3P@NR0Þ, the
organic moieties of the iminophosphorane can easily be modi-
fied to give ligands with different steric and electronic properties
[7].
Transition metal iminophosphorane complexes are well known
in the literature. Bielsa et al. have recently published a paper which
thoroughly summarises the different coordination modes of imino-
phosphorane ligands, commenting however that examples of
ortho-metallated complexes are scarce [8]. The first ortho-metallat-
ed complexes, formed directly by the reaction of the iminophos-
phorane ligands (Ar3P@NR; Ar = Ph, p-CH3C6H4; R = m-CH3C6H4,
p-CH3C6H4, p-CH3OC6H4) with Na2PdCl4, were reported by Alper
2. Results and discussion
Scheme 1 summarises the syntheses of both ortho-mercurated
and cycloaurated iminophosphoranes. The Kirsanov reaction be-
tween primary alkylamines and Ph3PBr2 yields the alkylamino-
phosphonium salts [R3P-NH-R0]Br almost quantitatively.
Subsequent deprotonation of these salts by addition of a base
yields the neutral iminophosphorane ligands. The use of aryl-
amines in the synthesis gives the iminophosphorane ligands
directly [7]. Lee and Singer have reported that the deprotonation
of aminophosphonium salts takes place readily and in high yields
using KOH as the base [21]. However, our experience showed
that yields were often lower than reported and the resulting
iminophosphoranes were unstable towards hydrolysis, especially
when the N-bonded substituent was –CHMePh.
* Corresponding author. Fax: +64 7 838 4219.
E-mail address: w.henderson@waikato.ac.nz (W. Henderson).
0020-1693/$ - see front matter Ó 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2009.04.024

3670
K.J. Kilpin et al. / Inorganica Chimica Acta 362 (2009) 3669–3676
Ph
Ph
Ph
ture 13) appear to be stable towards air and light. The compounds
Ph
Ph
Ph
Ph
Ph
N
N
N
P
P
P
were analysed by mass spectrometry, IR and NMR spectroscopy
and microelemental analysis. In addition, an X-ray structure deter-
mination of the chiral complex 14 was carried out.
Ph
AuCl2
HgCl
AuPPh3
1
2
3
2.1. X-ray crystal structure of (2-Cl2AuC6H4)Ph2P@N-(S)-CHMePh 14
In order to fully characterise (2-Cl2AuC6H4)Ph2P@N-(S)-
CHMePh 14 and compare it with the N-aryl complex 1, an X-ray
crystal structure determination was carried out. Fig. 1 shows the
structure of the complex with the atom labelling scheme, while Ta-
ble 1 gives important bond parameters. The crystal is a rare exam-
ple of a P1, Z = 1 structure. The gold atom is essentially square
planar with C(12) showing the greatest deviation from the coordi-
nation plane [defined by C(12), N(1), Au(1), Cl(1) and Cl(2)]. Be-
cause of the greater trans influence of the aryl carbon, the Au–
Cl(1) bond length is greater than the Au–Cl(2) bond length trans
to the nitrogen. The metallacyclic ring made up of Au(1), N(1),
P(1), C(11) and C(12) is significantly puckered, as in the previously
characterised cycloaurated iminophosphorane 1 [16]. Again, the
nitrogen atom shows the greatest deviation [0.288(5) Å above the
plane]. The bond lengths and angles in 14 are generally similar
to those in the N-phenyl complex 1 [16]. However, compared to
1, the Au(1)–N(1) bond of 14 is longer [2.072(10) versus
2.034(4) Å] and the trans chloride Au(1)–Cl(2) of 14 is slightly
longer [2.297(3) versus 2.289(1) Å].
+
R
N
Br
Ph
Ph
P
1. 2 n-BuLi
H
N
HgBr
P
R
2. HgCl
2
-
[AuCl
]
4
9
R = (
10 R = (
R
,
S
)-CHMePh
S)-CHMePh
t
R
4
5
6
R = Bu
Ph
N
R = (
R = (
R
S
,
S
)-CHMePh
)-CHMePh
P
Ph
AuCl
2
KOH
-
[AuCl
]
4
13 R = (
14 R = (
15 R =
16 R = Bu
R,
S
)-CHMePh
)-CHMePh
p-C H F
R
R
S
Ph
N
Ph
Ph
N
P
P
1. PhLi
2. HgCl
Ph
6
4
t
HgCl
2
11 R =
12 R = Bu
p-C H F
6 4
7
8
R = p-C H F
6 4
t
t
R = Bu
2.2. Spectroscopic and mass spectrometric characterisation
Scheme 1. Synthetic routes used for ortho-mercurated and cycloaurated
iminophosphoranes.
Because of the presence of multiple inequivalent phenyl rings
and coupling to phosphorus in the ortho-mercurated and cycloau-
Boubekeur et al. have recently shown that isolation of the imi-
nophosphorane is not necessary – reaction of alkylaminophospho-
nium salts with two equivalents of n-butyllithium leads to
deprotonatation of the phosphonium salt and ortho-lithiation of
the resulting iminophosphorane in one step [22]. We have adapted
this methodology to synthesise ortho-mercurated iminophosphor-
anes in a one-pot reaction. Thus, when the alkylaminophosphoni-
um salts were reacted with two equivalents of n-butyllithium
followed by HgCl2, the ortho-mercurated bromide compounds
were isolated in good yields. In contrast, the aryliminophosphora-
ne ligands are appreciably more stable. They can be isolated in
good yields and reacted with PhLi followed by HgCl2, as previously
described [16,17], to give the ortho-mercurated compounds, also in
good yields.
Vicente et al. [23] have demonstrated that direct mercuration of
Ph3P@NC6H4Me gives the ortho-mercurated compound with the
mercury bound to the p-tolyl ring, the wrong isomer for subse-
quent cycloauration. The use of Ph3P@NBut eliminates the possibil-
ity of metallation on the N-substituent and when reacted directly
with Hg(OAc)2 in THF it appears that mercuration takes place in
the desired position – on the ortho-position of a P-bonded phenyl
ring. However, yields were low compared with the synthetic route
that proceeds via the ortho-lithiated intermediate, and appreciable
decomposition of the ligand to [Ph3P-NH-But]+ and Ph3P@O is ob-
served by 31P{1H} NMR during the reaction.
Fig. 1. ORTEP view of (2-Cl2AuC6H4)Ph2P@N-(S)-CHMePh 14, showing the atom
numbering scheme. Thermal ellipsoids are shown at the 50% probability level.
Table 1
Transmetallation of either the ortho-mercurated chloride or
bromide species with NMe4[AuCl4] produced the cycloaurated
complexes in good yields. In all cases the gold(III) dichloride is
formed and no evidence of mono- or di-brominated compounds
are seen, even when the bromo-mercury iminophosphoranes are
used.
Using the above methodology, the ortho-mercurated com-
pounds 9–12 and the cycloaurated complexes 13–16 were success-
fully synthesised. With the exception of 14 which decomposed
slowly over time, all the compounds (including the racemic mix-
Important structural parameters for (2-Cl2AuC6H4)Ph2P@N-(S)-CHMePh 14, with esds
in parentheses.
Atoms
Lengths (Å)
Atoms
Angles (°)
Au(1)–Cl(1)
Au(1)–Cl(2)
Au(1)–N(1)
Au(1)–C(12)
P(1)–N(1)
2.371(2)
2.297(3)
2.072(10)
2.045(8)
1.615(10)
1.790(6)
1.473(12)
Cl(1)–Au(1)-Cl(2)
N(1)-Au(1)–C(12)
C(12)–Au(1)–Cl(2)
N(1)–Au(1)–Cl(1)
N(1)–P(1)–C(11)
P(1)–N(1)–C(41)
89.70(9)
85.1(4)
92.4(3)
93.0(3)
101.9(4)
131.4(8)
P(1)–C(11)
N(1)–C(41)

K.J. Kilpin et al. / Inorganica Chimica Acta 362 (2009) 3669–3676
3671
rated complexes, both the 1H and 13C{1H} NMR spectra are com-
plex and difficult to fully interpret. 31P{1H} NMR remains the most
useful experiment. The ortho-mercurated complexes have chemi-
cal shifts ranging from 2.2 ppm (12) to 14.8 ppm (9/10), with
3J(199Hg–P) coupling averaging 330 Hz. Upon transmetallation to
the cycloaurated complexes, there is a significant downfield shift
in the signal of approximately 50 ppm, consistent with the forma-
tion and inclusion of the phosphorus in a 5-membered ring [24].
The ESI-MS of the ortho-mercurated complexes 9–12 all show
strong [M+H]+ ions which can easily be identified by the character-
istic isotope pattern of mercury. Complex 12 also indicated the
presence of the diarylmercury complex 2-Hg({C6H4}Ph2P@NtBu)2;
however, there was no evidence of this compound in the 31P{1H}
NMR spectrum or microelemental analysis. This form of ionisation
is consistent with little N–Hg interaction in these species, leaving
the N available for protonation.
This type of adduct is the predominant gold species when monitor-
ing the progression of the transmetallation reaction by ESI-MS.
2.3. Reactivity of cycloaurated iminophosphorane complexes with
dianionic ligands
The chloride ligands on C,N cyclometallated Au(III) compounds
can be readily replaced by dianionic ligands such as catecholate
[26] and thiosalicylate [27–29], forming new five- and six-mem-
bered chelate rings, respectively. Such complexes are of interest
because other analogues containing these anionic ligands have
shown good anti-tumour activity. The iminophosphorane com-
plexes 1 and 16 show analogous behaviour and when reacted with
either catechol or thiosalicylic acid in methanol with excess tri-
methylamine, the cycloaurated complexes 17–20 can easily be iso-
lated in high yields by addition of water to the reaction mixture.
NMR spectroscopy, ESI mass spectrometry and microanalysis con-
firm the formulation. Compound 18 appears to crystallise with one
molecule of water as microanalysis of two independent samples
gave a composition which is in agreement with the proposed
formula.
In contrast, the primary route of ionisation for the cycloaurated
complexes appears to be loss of a chloride ligand to give the ions
[M-Cl]+, protonation to give [M+H]+ being precluded by the strong
N–Au bond. As previously observed, addition of a small amount of
highly coordinating pyridine produces the species [M-Cl+py]+ [25].
R
R
O
O
O
Ph
Ph
N
Ph
Ph
N
O
S
P
P
Au
Au
17 R = Ph
18 R = But
19 R = Ph
20 R = But
O
Me
Me
O
N
O
N
O
Au
Au
S
S
R
21
22 R = Me
23 R = H
Ph
PF6
Ph
N
Cl
P
Ph
Au
PPh3
24
Et
Me
Me
Au
Ph
Et
N
S
Me
Me
S
N
Ph
Ph
N
S
S
Me
Me
S
S
Et
Et
PF6
N
P
S
S
Et
N
Au
N
Au
N
Et
25
26
27
Interestingly, the complex 16 also shows the ion [M+Me4N]+ with
traces and the cation carried over from Me4N[AuCl4] to Me4NCl
in the synthesis. The Me4N+ cation would be expected to have very
high ionisation efficiency in ESI-MS when compared to a neutral
gold complex; no evidence of Me4N+ is observed in NMR spectra.
The 31P{1H} NMR spectra of the new cycloaurated derivatives
show only a slight difference in chemical shift from the original
dichloride species. The chemical shift of 1 is 65.5 ppm [16] whereas
17 and 19 have shifts of 60.9 and 54.1 ppm, respectively. Likewise,
the dichloride 16 has a shift of 46.1 ppm, which moves slightly to

3672
K.J. Kilpin et al. / Inorganica Chimica Acta 362 (2009) 3669–3676
55.7 and 43.2 ppm in 18 and 20, respectively. Such data indicate
that the original cycloaurated iminophosphorane ring remains in-
tact upon derivatisation; if the ring was opened an upfield shift
in the 31P{1H} spectrum would be expected [24].
Both the catechol and thiosalicylate derivatives show clean
mass spectra and, with the exception of 19, have strong parent ions
of the type [M+H]+ or [M+Na]+ when the sample is doped with
NaCl. However, for the complex 19, the parent ion is [2-
Au({C6H4}Ph2P@NPh)2]+, even when the sample is doped with
NaCl. The presence of this type of diarylgold species in ESI-MS
spectra of thiosalicylate complexes has been seen previously [29].
To confirm the geometry of the thiosalicylate complexes, an X-
ray crystal structure of 20 was carried out. Fig. 2 shows the molec-
ular structure with the atom labelling scheme, and Table 2 gives
selected bond lengths and angles. As with other gold(III) thiosalicy-
late complexes [27–29] the ligand is coordinated to the gold with
the two highest trans influence donors (aryl C and thiolate S) mutu-
ally cis to each other. The geometry around the gold is square pla-
nar with the largest deviation from the N(1), C(12), Au(1), S(1) and
O(2) plane being for C(12), which sits 0.1620(7) Å below the plane.
The fold angle between the Au coordination plane [defined by
Au(1), S(1) and O(2)] and the plane of the thiosalicylate ligand
(excluding the oxygen atoms) is 43.34(4)°, which falls in the range
of 30.1(8)° (21) and 59.6° (22) previously seen in gold(III) thiosal-
icylate systems [27–29]. The bite angle of the thiosalicylate ligand
is 91.46(4)°, comparable to those seen in (22) and (23) [89.82(8)°
and 90.1(2)°, respectively]. The five-membered ring created by
the cycloaurated iminophosphorane is also puckered with the
greatest deviations for N(1) (0.3226(7) Å above the plane) and
P(1) (0.2788(7) Å below the plane). The puckering of the thiosalicy-
late and the iminophosphorane rings can clearly be seen in Fig. 3.
The fold angle between the Au coordination plane and that of the
thiosalicylate ring is 43.3°, towards the lower end of the range
found for a series of Pt(II) thiosalicylate complexes [30].
Fig. 3. ORTEP diagram of 20, showing the fold angle of the thiosalicylate ligand and
puckering of the iminophosphorane ring. For clarity, only the ipso carbons of the
uncoordinated phenyl rings and the tertiary carbon of the But group are shown.
Ellipsoids are shown at 50% probability.
2.4. Reactivity with phosphines
Previously, reaction of C,N cyclometallated Au(III) dichloride
complexes with PPh3 has given an indication of the strength of
the N–Au bond [6]. If the bond is strong, then a chloride ion will
be displaced to give a cationic complex. Alternatively, if the N–
Au bond is relatively weak, it will be cleaved and coordination of
the phosphine will give a neutral phosphine complex. The behav-
iour is dependent on the ligand. For example, when the damp com-
plex (Scheme 2a) is reacted with one equivalent of PPh3
substitution of the chloride takes place [31]. In contrast, when pap-
AuCl2 (Scheme 2b) is reacted in the same manner the nitrogen is
displaced and the neutral phosphine complex is obtained [32,33].
To assess the strength of the Au–N bond in Au(III) iminophos-
phoranes, 1 was reacted with one molar equivalent of PPh3 to-
gether with NH4PF6 in dichloromethane, producing the
compound 24 where the PPh3 has displaced the chloride ligand.
The products were initially characterised by ESI-MS and 31P{1H}
NMR. The 31P{1H} NMR shows the expected two signals at 60.6
and 41.0 ppm in a 1:1 ratio. The first arises from the iminophos-
phorane, and is only slightly shifted from that of the parent dichlo-
ride (65.5 ppm) indicating that the phosphorus is still present in a
ring. This was unambiguously confirmed by an X-ray crystal struc-
ture determination.
The molecular structure of the cation of 24 is shown in Fig. 4,
and important bond parameters are presented in Table 3. The
(a)
Me
Me
Au
Me
Me
Au
N
N
Cl
Cl
Cl
PPh3
Fig. 2. ORTEP diagram of 20, showing atom numbering scheme with ellipsoids
shown at 50% probability.
PPh3
Table 2
Ph
N
(b)
Ph
Au
Selected bond parameters for 20, with esds in parentheses.
N
Ph3P
Atoms
Lengths (Å)
Atoms
Angles (°)
Cl
Cl
N
N
Cl
Cl
PPh3
N(1)–Au(1)
C(12)–Au(1)
O(2)–Au(1)
S(1)–Au(1)
S(1)–C(3)
2.1196(13)
2.0194(16)
2.0676(13)
2.2760(4)
1.7618(17)
1.226(2)
N(1)–Au(1)–C(12)
O(2)–Au(1)–S(1)
C(3)–S(1)–Au(1)
O(1)–C(1)–O(2)
C(1)–O(2)–Au(1)
O(1)–C(1)–C(2)
85.90(6)
91.46(4)
103.08(6)
121.52(17)
130.92(11)
119.45(15)
Au
O(1)–C(1)
O(2)–C(1)
1.306(2)
Scheme 2. Reaction of (a) dampAuCl2 and (b) papAuCl2 with PPh3.

K.J. Kilpin et al. / Inorganica Chimica Acta 362 (2009) 3669–3676
3673
Table 4
Anti-tumour (P388 murine leukaemia) activities for selected gold(III) iminophos-
phoranes and related systems.
a
Complex
IC50
(ng mLꢀ1
)
(lM)
1
7546
20,021
<487
655
<487
658
10.7
33.4
<0.74
1.03
<0.69
0.97
16
17
18
19
20
a
Concentration of sample required to reduce the cell growth of the P388 murine
leukaemia cell line by 50%.
NMR spectrum is observed. However, isolation and microanalysis
of this complex as a white solid produced microelemental data
completely inconsistent with the expected formulation.
Reaction of dampAuCl2 with one equivalent of NaS2CNMe2 pro-
duced the cationic complex 25. However, if two equivalents of
dithiocarbamate were used a neutral complex was obtained where
one dithiocarbamate ligand was coordinated in a bidentate fashion,
the other as a monodentate ligand, as shown by an X-ray crystal
structure analysis, 26. Rapid interchange between the ligands oc-
curs on the NMR timescale [39]. Analogous results are seen with
oxazoline gold(III) complexes [40].
When 1 is reacted with a molar equivalent of NaS2CNEt2 and
NH4PF6 in dichloromethane the complex 27 is obtained as a yellow
solid. The analogous reaction with two equivalents of NaS2CNEt2
also gives a yellow complex which is identical to 27 (on the basis
of 1H and 13P{1H} NMR experiments). No evidence of the complex
analogous to 26 is observed. These observations indicate that the
iminophosphorane complexes contain a robust cyclometallated
ring.
Fig. 4. ORTEP diagram of [(2-Cl(PPh3)AuC6H4)Ph2P@NPh]+. Ellipsoids are shown at
30% probability.
Table 3
Selected bond parameters for [(2-Cl(PPh3)AuC6H4)Ph2P@NPh]PF6 ꢁ 0.5Et2O, with esds
in parentheses.
Atoms
Lengths (Å)
Atoms
Angles (°)
Au(1)–Cl(1)
Au(1)–P(2)
Au(1)–C(12)
Au(1)–N(1)
P(1)–N(1)
2.3463(11)
2.3030(12)
2.057(5)
2.079(5)
1.620(6)
1.419(7)
1.401(9)
N(1)–Au(1)–C(12)
C(12)–Au(1)–P(2)
P(2)–Au(1)–Cl(1)
Cl(1)–Au(1)–N(1)
Au(1)–N(1)–P(1)
P(1)–N(1)–C(41)
N(1)–P(1)–C(11)
84.0(2)
92.88(17)
91.11(4)
92.26(13)
108.3(3)
124.9(4)
100.6(3)
N(1)–C(41)
C(11)–C(12)
geometry around the gold atom is essentially square planar and
N(1) shows the greatest deviation [0.113(3) Å] from the coordina-
tion plane. The X-ray crystal structure shows that the phosphine is
coordinated trans to the nitrogen, thus placing the two softest li-
gands (the phosphine and aryl groups) mutually cis to each other,
giving the most thermodynamically stable configuration (anti-
symbiosis) [34]. This is also observed in other phosphine substi-
tuted Au(III) cationic complexes, and the Au–P bond lengths are
comparable with those previously seen [35–38]. Upon substitution
of the chloride with the bulky phosphine group the bond angle be-
tween the two monodentate ligands has increased [from 90.70(4)°
in 1 to 92.88(17)° in 24] which in turn has decreased the bite angle
of the iminophosphorane ligand [from 84.86(17)° to 84.0(2)°].
However, the ligand remains significantly puckered, with N(1)
and P(1) showing the greatest deviation from the mean plane of
the ring [defined by N(1), P(1), Au(1), C(11) and C(12)].
2.6. Biological activity
A number of gold(III) complexes have previously been found to
show anti-tumour activity [6,41,42]. The biological activity of 1 has
been reported, and showed low anti-cancer activity against the
P388 murine leukaemia cell line [16]. In order to determine if this
is characteristic of cycloaurated iminophosphoranes, the activity of
the alkyl compound 16 was evaluated. The bicyclic catecholate (17
and 18) and thiosalicylate (19 and 20) derivatives were also evalu-
ated, since related derivatives have been previously been found to
exhibit high anti-tumour activity [29].
Results of the biological assays are presented in Table 4. In con-
trast to the dichloride complexes 1 and 16 which show low anti-tu-
mour activity, the catecholate and thiosalicylate complexes show a
tenfold increase in the level of activity. In particular the complexes
17 and 19, originating from complex 1, show a high level of activ-
ity. These observations confirm that the presence of catecholate or
thiosalicylate ligands tends to lead to complexes with good activ-
ity, and warrants further detailed studies on these systems.
2.5. Attempted displacement of the N-donor ligand
In order to evaluate our hypothesis that displacement of the N-
donor would result in a significant upfield shift in the 31P spectra
from a monodentate iminophosphorane ligand we attempted to
synthesise complexes of this type. Previously this has been
achieved with gold(III) complexes by replacement of the relatively
weakly bound chloride ligands with stronger cyanide [39] or
dithiocarbamate [40,41] groups.
Cyanide complexes of the damp system have been spectroscop-
ically characterised. Reaction of dampAuCl2 (see Scheme 2) with
two equivalents of KCN gave the neutral complex dampAu(CN)2,
whereas excess cyanide gave the anionic [Au(damp)(CN)3]ꢀ com-
plex as the potassium salt [28]. When 1 was reacted with excess
KCN the corresponding anion could be seen in the ESI mass spec-
trum, and an unexpected new peak at 69.5 ppm in the 31P{1H}
3. Conclusions
A range of new ortho-mercurated and cycloaurated iminophos-
phorane complexes and their derivatives have been synthesised
and fully characterised by NMR spectroscopy and ESI mass spec-
trometry. X-ray crystal structures of three of these compounds
have been determined. The Au–N coordinate bond is relatively
strong because when 1 is reacted with PPh3 the Au–N bond re-
mains intact and a chloride ligand is substituted by PPh3. Attempts
at cleaving this bond by reacting 1 with ligands such as CNꢀ and
dithiocarbamates were unsuccessful. Both chloride ligands can be

3674
K.J. Kilpin et al. / Inorganica Chimica Acta 362 (2009) 3669–3676
replaced by chelating dianionic ligands to give bi-metallacyclic
compounds with enhanced anti-tumour activity, possibly due to
the reduced reactivity of the complexes.
diately give a dark orange solution. After stirring for 1.5 h a white
precipitate had formed. HgCl2 (0.730 g, 2.69 mmol) was added and
the mixture stirred for an additional hour. Workup as for 9 gave
crystals of (2-ClHgC6H4)Ph2P@NC6H4F (0.927 g, 57%). Found: C,
47.5; H, 2.9; N, 2.4; C24H18NFPClHg requires C, 47.5; H, 3.0; N,
2.3%. NMR: 1H: d (ppm) 6.69–6.75, 6.80–6.85, 7.27–7.34, 7.40–
7.70 (all m, Ar–H); 31P{1H}: d (ppm) 9.0 (br), 3J(199Hg–P) =
320 Hz. ESI-MS: m/z: 607 (100%, [M+H]+), 942 (35%, [2-
4. Experimental
4.1. General procedures
The known aminophosphonium salts 4 and 5 and the imino-
phosphoranes 7 and 8 were prepared by literature procedures; 6
was prepared by a completely analogous method to 5 [7,21]. Com-
plex 1 was prepared by the literature procedure [16]. n-Butyllith-
ium (either 1.6 M in hexanes or 2.0 M in cyclohexane, Aldrich),
PhBr (AJAX Chemicals), HgCl2, catechol, 30% aqueous Me3N (BDH)
sodium diethyldithiocarbamate trihydrate (BDH), Hg(OAc)2 (Riedel
de Haën) and thiosalicylic acid (Sigma) were used as received. Me4-
N[AuCl4] was prepared by reaction of HAuCl4 ꢁ 4H2O with Me4NCl.
Dry de-oxygenated diethyl ether and THF were obtained from a
Hg({C6H4}Ph2P@NC6H4F)2+H]+). IR: (P@N) = 1309 cmꢀ1 (vs).
t
4.5. Preparation of (2-ClHgC6H4)Ph2P@NtBu 12
As for 11 above, Ph3P@NtBu 8 (1.199 g, 3.60 mmol) was added
to a solution of PhBr (0.38 mL, 3.60 mmol) and nBuLi (1.6 M,
2.25 mL, 3.60 mmol) in diethyl ether (20 mL). Addition of HgCl2
(0.977 g, 3.60 mmol) followed by workup gave (2-ClHgC6H4)Ph2
P@NtBu (0.909 g, 44%). Found: C, 46.0; H, 4.1; N, 2.5; C22H23NPClHg
requires C, 46.4; H, 4.1; N, 2.5%. NMR: 1H: d (ppm) 1.29 (s, CH3),
7.13–7.19, 7.31–7.64, 7.74–7.81 (all m, Ar–H); 31P{1H}: d (ppm)
2.2, 3J(199Hg–P) = 341 Hz. ESI-MS: m/z: 569 (100%, [M+H]+). IR:
TM
Pure Solv solvent purification system, and metallation reactions
were carried out under a nitrogen atmosphere using standard
Schlenk techniques, with light also being excluded in the case of
cycloauration reactions.
t
(P@N) = 1223 cmꢀ1 (vs).
Positive ion ESI mass spectra were acquired on a VG Platform II
instrument using methanol as the solvent after the compound had
first been dissolved in a small amount of dichloromethane. IR spec-
tra were acquired as KBr disks on a Digilab Scimitar FT-IR spec-
trometer. 31P{1H} and 1H NMR spectra (solvent CDCl3) were
recorded on a Bruker DRX 300 FT NMR spectrometer at 121.5
and 300.1 MHz, respectively. Microelemental analysis was carried
out at the Campbell Microanalytical Laboratory at the University
of Otago. Anti-tumour assays were carried out using previously de-
scribed methods [43]; samples were tested in a 2:1 dichlorometh-
ane:methanol solvent system.
4.5.1. Alternative synthesis
Ph3P@NtBu
(0.300 g, 0.90 mmol) and Hg(OAc)2 (0.288 g,
8
0.90 mmol) were refluxed under nitrogen in dry degassed THF
(20 mL) for 16 h. The solution was allowed to cool, LiCl (0.076 g,
1.79 mmol) was added and stirred for an additional 4 h. The solu-
tion was then filtered and evaporated to dryness. The filtrate was
redissolved in dichloromethane (2 ꢂ 10 mL) and diethyl ether
(20 mL) was added. Storage at ꢀ20 °C gave crystals of (2-
ClHgC6H4)Ph2P@NtBu (0.128 g, 25%), identified by 1H and 31P{1H}
NMR spectroscopy.
4.6. Preparation of (2-Cl2AuC6H4)Ph2P@N-(R,S)-CHMePh 13
4.2. Preparation of (2-BrHgC6H4)Ph2P@N-(R,S)-CHMePh 9
The complex (2-BrHgC6H4)Ph2P@N-(R,S)-CHMePh 9 (0.100 g,
0.15 mmol), Me4N[AuCl4] (0.062 g, 0.15 mmol) and Me4NCl
(0.016 g, 0.15 mmol) were added to degassed acetonitrile (15 mL)
and stirred in a foil-covered flask for 48 h. The solvent was re-
moved and the yellow solid extracted into dichloromethane
(2 ꢂ 10 mL), filtered and diethyl ether was added to the filtrate.
Storage at ꢀ20 °C gave yellow crystals of (2-Cl2AuC6H4)Ph2P@N-
(R,S)-CHMePh) (0.062 g, 64%). Found: C, 47.2; H, 3.5; N, 2.1;
C26H23NPCl2Au requires C, 48.2; H, 3.6; N, 2.2%. NMR: 1H: d
(ppm) 1.62 (d, CH3), 5.85 (m, CH), 6.92–7.01, 7.05–7.08, 7.18–
7.23, 7.32–7.53, 7.58–7.67, 8.17–8.21 (all m, Ar–H); 31P{1H}: d
(ppm) 59.0. ESI-MS: pyridine doped m/z: 654 (100%, [M-
Cl+MeCN]+), 692 (85%, [M-Cl+py]+), 613 (53%, [M-Cl]+). IR:
[Ph3P-NH-(R,S)-CHMePh]+Brꢀ 5 (1.000 g, 2.16 mmol) was sus-
pended in diethyl ether (30 mL) and cooled to ꢀ84 °C in an ethyl
acetate slush bath. nBuLi (1.6 M, 2.7 mL, 4.32 mmol) was added
dropwise with stirring. The solution was allowed to warm to room
temperature to give a clear orange solution that was stirred for an
additional 30 min. HgCl2 (0.586 g, 2.16 mmol) was added in one
portion and after stirring for 1 h the solvent was removed and
the solid extracted into dichloromethane (2 ꢂ 10 mL), filtered
and diethyl ether (20 mL) was added to the filtrate. Storage at
ꢀ20 °C gave white crystals of (2-BrHgC6H4)Ph2P@N-(R,S)-CHMePh
(0.606 g, 42%). Found: C, 47.7; H, 3.6; N, 2.2; C26H23NPBrHg re-
quires C, 47.3; H, 3.5; N, 2.1%. NMR: 1H: d (ppm) 1.64 (d, CH3),
4.68 (m, CH), 7.00–7.69 (m, Ar–H); 31P{1H}:
3J(199Hg–P) = 317 Hz. ESI-MS: m/z: 662 (100%, [M+H]+). IR:
(P@N) = 1163 cmꢀ1 (vs).
d
(ppm) 14.8,
t
(P@N) = 1138 cmꢀ1 (s).
t
4.7. Preparation of (2-Cl2AuC6H4)Ph2P@N-(S)-CHMePh 14
4.3. Preparation of (2-BrHgC6H4)Ph2P@N-(S)-CHMePh 10
Analogous to 13, (2-BrHgC6H4)Ph2P@N-(S)-CHMePh 10
(1.000 g, 1.51 mmol), Me4N[AuCl4] (0.623 g, 1.51 mmol) and
Me4NCl (0.166 g, 1.51 mmol) were reacted in acetonitrile (30 mL)
for 48 h. Workup gave (2-Cl2AuC6H4)Ph2P@N-(S)-CHMePh) as pale
yellow crystals (0.139 g, 14%). The product was identified by 1H
and 31P{1H} NMR spectroscopy.
As for the preparation of 9 above, [Ph3P-NH-(S)-CHMePh]+Brꢀ
6
(2.000 g, 4.34 mmol) was suspended in diethyl ether (30 mL) and
nBuLi (1.6 M, 5.4 mL, 8.68 mmol) added. After addition of HgCl2
(1.178 g, 4.34 mmol), work-up as above gave 1.780 g, (62%) of (2-
BrHgC6H4)Ph2P@N-(S)-CHMePh. The product was identified by 1H
and 31P{1H} NMR spectroscopy.
4.8. Preparation of (2-Cl2AuC6H4)Ph2P@NC6H4F 15
4.4. Preparation of (2-ClHgC6H4)Ph2P@NC6H4F 11
As for 13, (2-ClHgC6H4)Ph2P@NC6H4F 11 (0.174 g, 0.29 mmol),
Me4N[AuCl4] (0.119 g, 0.29 mmol) and Me4NCl (0.032 g,
0.29 mmol) in acetonitrile (20 mL) for 24 h. Workup gave (2-
Cl2AuC6H4)Ph2P@NC6H4F as bright yellow crystals (0.074 g, 40%).
Found: C, 45.2; H, 3.1; N, 2.2; C24H18NFPCl2Au requires C, 45.2;
A solution of nBuLi (2 mL, 2.69 mmol) was added to a solution of
PhBr (0.28 mL, 2.69 mmol) in diethyl ether (20 mL) and stirred for
15 min. Ph3P@NC6H4F 7 (1.000 g, 2.69 mmol) was added to imme-

K.J. Kilpin et al. / Inorganica Chimica Acta 362 (2009) 3669–3676
3675
H, 2.8; N, 2.2%. NMR: 1H: d (ppm) 6.73–6.80, 6.92–6.97, 7.10–7.17,
7.33–7.38, 7.49–7.62, 7.71–7.78, 8.41–8.46 (all m, Ar–H); 31P{1H}:
d (ppm) 66.9. ESI-MS: m/z: 603 (100%, [M-Cl]+), 1241 (80%, [2 M-
Cl]+); pyridine doped m/z: 682 (100%, [M-Cl+py]+). IR:
and washed with water (3 ꢂ 10 mL) and diethyl ether (1 ꢂ
10 mL) and dried under vacuum to give 19 (0.099 g, 88%). Found:
C, 53.1; H, 3.4; N, 2.0; C3 H23NO2PSAu requires C, 53.1; H, 3.3; N,
1
2.0%. NMR: 1H: d (ppm) 6.88–6.92, 7.01–7.21, 7.28–7.35, 7.46–
7.57, 7.64–7.75, 8.00–8.04, 8.15–8.20 (all m, Ar–H); 31P{1H}: d
(ppm) 54.1. ESI-MS: m/z: 902 (100%, [2-Au({C6H4}Ph2P@NPh)2]+),
703 (30%, [M+H]+), 1404 (25%, [2M+H]+); NaCl doped m/z: 901
(100%, [2-Au({C6H4}Ph2P@NPh)2]+), 1426 (85%, [2M+Na]+), 725
(40%, [M+Na]+), 703 (10%, [M+H]+), 1404 (7%, [2 M+H]+). IR:
t
(P@N) = 1219 cmꢀ1 (s).
4.9. Preparation of (2-Cl2AuC6H4)Ph2P@NtBu 16
As for 13, (2-ClHgC6H4)Ph2P@NtBu 12 (0.400 g, 0.70 mmol), Me4-
N[AuCl4] (0.291 g, 0.70 mmol) and Me4NCl (0.077 g, 0.70 mmol)
were reacted in acetonitrile (20 mL) for 24 h. Workup gave yellow
crystals of (2-Cl2AuC6H4)Ph2P@NtBu (0.312 g, 74%). Found: C,
44.2; H, 3.9; N, 2.4; C22H23NPCl2Au requires C, 44.0; H, 3.9; N,
2.3%. NMR: 1H: d (ppm) 1.34 (s, CH3), 7.31–7.40, 7.55–7.62, 7.67–
7.72, 7.85–7.91, 8.04–8.07 (all m, Ar–H); 31P{1H}: d (ppm) 46.1.
ESI-MS: m/z: 565 (100%, [M-Cl]+), 673 (35%, [M+Me4N]+), 1165
(30%, [2 M-Cl]+); pyridine doped m/z: 644 (100%, [M-Cl+py]+), 565
t
(P@N) = 1316 cmꢀ1 (m),
t
(C@O) = 1626 cmꢀ1 (vs).
4.13. Synthesis of 20
As for 19, (2-Cl2AuC6H4)Ph2P@NtBu 16 (0.101 g, 0.17 mmol) and
thiosalicylic acid (0.026 g, 0.17 mmol) were reacted in methanol
(15 mL). Addition of Me3N (1 mL, excess) and water (20 mL) fol-
lowed by workup gave 0.085 g (73%) of 20 as a yellow solid. Found:
C 51.0, H 4.0, N 2.0; C29H27NO2PSAu requires C, 51.1; H, 4.0; N,
2.1%. NMR: 1H: d 1.40 (s, CH3), 7.01–7.05, 7.15–7.25, 7.32–7.38,
7.43–7.47, 7.53–7.59, 7.63–7.69, 7.86–7.99, 8.14–8.19 (all m, Ar–
H); 31P{1H}: d 43.2 ppm. ESI-MS: m/z: 683 (100%, [M+H]+), 1212
(52%, unidentified), 1364 (30%, [2M+H]+); NaCl doped m/z: 705
(60%, [M-Cl]+). IR:
t
(P@N) = 1183 cmꢀ1 (m).
4.10. Synthesis of 17
The complex (2-Cl2AuC6H4)Ph2P@NPh 1 (0.100 g, 0.16 mmol)
and catechol (0.018 g, 0.16 mmol) were refluxed in methanol
(20 mL). With stirring, aqueous Me3N (1 mL, excess) was added
resulting in the yellow solution immediately turning dark orange.
After refluxing for a further 20 min the solution was allowed to
cool, water (20 mL) was added and stirring continued overnight.
The precipitate that had formed was filtered, washed with water
(2 ꢂ 10 mL) and diethyl ether (10 mL) and dried under vacuum
to give 17 (0.085 g, 81%) as a rose-coloured solid. Found: C, 54.8;
H, 3.6; N, 2.2; C30H23NO2PAu requires C, 54.8; H, 3.5; N, 2.1%.
NMR: 1H: d (ppm) 6.42–6.47, 6.73–6.76, 7.01–7.15, 7.29–7.36,
7.49–7.57, 7.66–7.79, 8.19–8.22 (all m, Ar–H); 31P{1H}: d (ppm)
60.9. ESI-MS: m/z: 658 (100%, [M+H]+), 1315 (60%, [2M+H]+); NaCl
doped m/z: 680 (100%, [M+Na]+), 1337 (34%, [2M+Na]+), 658 (28%,
(100%, [M+Na]+), 1386 (87%, [2M+Na]+). IR:
(m),
(C@O) = 1626 cmꢀ1 (vs).
t
(P@N) = 1310 cmꢀ1
t
4.14. Synthesis of [(2-Cl(PPh3)AuC6H4)Ph2P@NPh]PF6 24
The complex (2-Cl2AuC6H4)Ph2P@NPh 1 (0.104 g, 0.17 mmol),
PPh3 (0.045 g, 0.17 mmol) and NH4PF6 (0.111 g, 0.68 mmol) were
stirred in dichloromethane (10 mL) for 4 h. The solution was fil-
tered and the dark orange filtrate reduced in volume (ꢃ2 mL).
Diethyl ether was added until the solution went cloudy. Storage
at ꢀ20 °C gave yellow crystals of [(2-Cl(PPh3)AuC6H4)Ph2
P@NPh]PF6 (0.088 g, 52%). Found: C, 50.2; H, 3.5; N, 1.5;
C42H34NF6P3ClAu requires C, 50.9; H, 3.5; N, 1.4%. NMR: 1H: d
(ppm) 6.90–7.03, 7.09–7.14, 7.30–7.34, 7.51–7.57, 7.61–7.71,
7.78–7.87 (all m, Ar–H); 31P{1H}: d (ppm) 60.6 (P@N), 41.0
(PPh3), 1:1 ratio. ESI-MS: m/z: 847 (100%, [(2-Cl(PPh3)AuC6H4)
[M+H]+), 1315 (12%, [2M+H]+). IR: (P@N) = 1254 cmꢀ1 (m).
t
4.11. Synthesis of 18
Ph2P@NPh]+). IR: (P@N) = 1253 cmꢀ1 (w).
t
The complex (2-Cl2AuC6H4)Ph2P@NtBu (0.100 g, 0.17 mmol)
and catechol (0.019 g, 0.17 mmol) were brought to reflux in meth-
anol (15 mL). With stirring, aqueous Me3N (1 mL, excess) was
added resulting in the yellow solution turning dark orange. After
refluxing for a further 15 min the now purple solution was allowed
to cool, water (20 mL) was added, and stirring continued overnight.
The precipitate that had formed was filtered, washed with water
(3 ꢂ 10 mL) and diethyl ether (20 mL) and dried under vacuum
to give 18 (0.067 g, 62%) as a rose-coloured solid. A sample for
microanalysis was recrystallised from dichloromethane/diethyl
ether. Found: C, 51.3; H, 4.2; N, 2.2; C28H27NO2PAu requires C,
52.8; H, 4.3; N, 2.2; C28H27AuNO2P ꢁ H2O requires C, 51.3; H, 4.5;
N, 2.1%. NMR: 1H: d (ppm) 1.52 (s, CH3), 6.45–6.50, 6.53–6.58,
6.67–6.79, 7.13–7.20, 7.34–7.39, 7.56–7.62, 7.66–7.71, 7.91–7.98,
8.11–8.14 (all m, Ar–H) 31P{1H}: d (ppm) 55.7. ESI-MS: m/z: 638
(100%, [M+H]+), 1275 (30%, [2 M+H]+); NaCl doped m/z: 660
(100%, [M+Na]+), 1297 (60%, [2 M+Na]+), 638 (40% [M+H]+). IR:
4.15. Synthesis of 27
The complex (2-Cl2AuC6H4)Ph2P@NPh 1 (0.100 g, 0.16 mmol)
and NaS2CNEt2ꢁ3H2O (0.036 g, 0.16 mmol) were stirred in dichloro-
methane (10 mL) for 2.5 h. NH4PF6 (0.065 g, 0.40 mmol) was added
and the solution stirred for an additional 18 h. The insoluble mate-
rial was removed by filtration and the yellow solution reduced in
volume (ꢃ2 mL). Diethyl ether was added until the solution went
cloudy. Storage at ꢀ20 °C gave yellow microcrystals (0.099 g,
73%) of 27. Found: C, 41.8; H, 3.6; N, 3.2; C29H29N2F6P2S2Au re-
quires C, 41.3; H, 3.5; N, 3.3%. NMR: 1H: d (ppm) 1.30 (t, CH3),
1.43 (t, CH3), 3.71 (q, CH2), 3.82 (q, CH2), 6.83–6.86, 7.06–7.12,
7.17–7.23, 7.28–7.35, 7.39–7.43, 7.47–7.54, 7.57–7.80 (all m, Ar–
H); 31P{1H}: d (ppm) 65.9. ESI-MS: m/z: 697 (100%, [M]+). IR:
t
(P@N) = 1232 cmꢀ1 (w).
t
(P@N) = 1254 cmꢀ1 (m).
4.16. Crystal structure determinations of 14, 20 and 24
4.12. Synthesis of 19
Single crystals of X-ray quality were grown by vapour diffusion
of either diethyl ether (14 and 24) or pentane (20) into a dichloro-
methane solution of the compound at room temperature. Both 14
and 24 crystallised as yellow blocks, 20 as orange plates.
To a stirred methanolic solution (15 mL) of (2-Cl2AuC6H4)
Ph2P@NPh 1 (0.100 g, 0.16 mmol) and thiosalicylic acid (0.025 g,
0.16 mmol), Me3N (1 mL, excess) was added, resulting in the pale
yellow solution immediately becoming darker. Stirring was contin-
ued in the dark for a further 90 min before water (20 mL) was
added. The fine yellow precipitate that had formed was filtered
4.16.1. Data collection
Unit cell dimensions and intensity data (Table 5) were collected
at the University of Canterbury on a Bruker Nonius Apex II CCD

3676
K.J. Kilpin et al. / Inorganica Chimica Acta 362 (2009) 3669–3676
Table 5
free of charge from The Cambridge Crystallographic Data Centre
Crystallographic details for 14, 20 and 24.
via www.ccdc.cam.ac.uk/data_request/cif. Supplementary data
associated with this article can be found, in the online version, at
doi:10.1016/j.ica.2009.04.024.
14
20
24ꢁ0.5Et2O
Formula
Molecular weight
T/K
Crystal system
Space group
a (Å)
C26H23NPCl2Au C29H27NO2PSAu C44H39NO0.5F6PClAu
648.29
93
Triclinic
P1
9.0297(4)
9.2152(4)
9.3699(4)
61.462(2)
63.334(2)
61.134(2)
575.22(4)
1
681.51
93
Monoclinic
P21/n
9.0966(2)
17.4390(4)
15.6240(4)
90
93.998(1)
90
1029.09
93
Monoclinic
C2/c
50.109(2)
9.8843(5)
18.3601(8)
90
111.043(2)
90
References
[1] C.F. Shaw III, Chem. Rev. 99 (1999) 2589.
[2] H.G. Raubenheimer, S. Cronje, Chem. Soc. Rev. 37 (2008) 1998.
[3] P.C.A. Bruijnincx, P.J. Sadler, Curr. Opinion Chem. Biol. 12 (2008) 197.
[4] A.S.K. Hashmi, Angew. Chem., Int. Ed. Eng. 44 (2005) 6990.
[5] A.S.K. Hashmi, R. Salathé, T.M. Frost, L. Schwarz, J.-H. Choi, Appl. Catal. A: Gen.
291 (2005) 238.
[6] W. Henderson, Adv. Organomet. Chem. 54 (2006) 207.
[7] A.W. Johnson, Ylides and Imines of Phosphorus, Wiley, New York, 1993.
[8] R. Bielsa, R. Navarro, T. Soler, E.P. Urriolabeitia, Dalton Trans. (2008)
1203.
[9] H. Alper, J. Organomet. Chem. 127 (1977) 385.
[10] M.A. Leeson, B.K. Nicholson, M.R. Olsen, J. Organomet. Chem. 579 (1999)
243.
[11] D. Aguilar, M.A. Araguees, R. Bielsa, E. Serrano, T. Soler, R. Navarro, E.P.
Urriolabeitia, J. Organomet. Chem. 693 (2008) 417.
[12] R. Bielsa, R. Navarro, E.P. Urriolabeitia, A. Lledós, Inorg. Chem. 46 (2007)
10133.
b (Å)
c (Å)
a
(°)
b (°)
c
(°)
V (Å3)
2472.49(1)
4
1.831
8487.2(7)
8
1.611
Z
Dcalc (g cmꢀ3
Tmax.,min.
)
1.871
0.5534, 0.0508 0.5267, 0.2164
7298 (0.0358)
5778
0.530, 0.327
Total reflections (Rint
Number of unique
reflections
Number of observed
reflections
)
44,421 (0.0302) 64,794 (0.0376)
10,185
13,986
5273
8940
11,379
R [I > 2
wR2 (all data)
Goodness of Fit
Flack ꢂ parameter
r
(I)]
0.0319
0.077
1.062
0.0198
0.0454
1.016
0.0536
0.1385
1.142
[13] R. Bielsa, A. Larrea, R. Navarro, T. Soler, E.P. Urriolabeitia, Eur. J. Inorg. Chem. 9
(2005) 1724.
[14] K.T.K. Chan, L.P. Spencer, J.D. Masuda, J.S.J. McCahill, P. Wei, D.W. Stephan,
Organometallics 23 (2004) 381.
[15] P. Wei, K.T.K. Chan, D.W. Stephan, Dalton Trans. (2003) 3804.
[16] S.D.J. Brown, W. Henderson, K.J. Kilpin, B.K. Nicholson, Inorg. Chim. Acta 360
(2007) 1310.
[17] D. Aguilar, M. Contel, R. Navarro, E.P. Urriolabeitia, Organometallics 26 (2007)
4604.
[18] S. Wingerter, M. Pfeiffer, T. Stey, M. Bolboacá, W. Kiefer, V. Chandrasekhar, D.
Stalke, Organometallics 20 (2001) 2730.
[19] S. Wingerter, H. Gornitzka, G. Bertrand, D. Stalke, Eur. J. Inorg. Chem. (1999)
173.
[20] S. Wingerter, H. Gornitzka, R. Bertermann, S.K. Pandley, J. Rocha, D. Stalke,
Organometallics 19 (2000) 3890.
[21] K.-W. Lee, L.A. Singer, J. Org. Chem. 39 (1974) 3780.
[22] L. Boubekeur, L. Ricard, N. Mézailles, M. Demange, A. Auffrant, P. Le Floch,
Organometallics 25 (2006) 3091.
[23] J. Vicente, J.A. Abad, R. Clemente, J. López-Serrano, M.C. Ramírez de Arellano,
P.G. Jones, D. Bautista, Organometallics 22 (2003) 4248.
[24] P.E. Garrou, Chem. Rev. 81 (1981) 229.
[25] W. Henderson, C. Evans, Inorg. Chim. Acta 294 (1999) 183.
[26] C.H.A. Goss, W. Henderson, A.L. Wilkins, C. Evans, J. Organomet. Chem. 679
(2003) 194.
[27] W. Henderson, B.K. Nicholson, S.J. Faville, D. Fan, J.D. Ranford, J. Organomet.
Chem. 631 (2001) 41.
[28] D. Fan, C.-T. Yang, J.D. Ranford, J.J. Vittal, P.F. Lee, Dalton Trans. (2003)
3376.
[29] M.B. Dinger, W. Henderson, J. Organomet. Chem. 560 (1998) 233.
[30] W. Henderson, L.J. McCaffrey, B.K. Nicholson, J. Chem. Soc., Dalton Trans.
(2000) 2753.
[31] J. Vicente, M.T. Chicote, M.D. Bermúdez, J. Organomet. Chem. 268 (1984)
191.
[32] J. Vicente, M.T. Chicote, Inorg. Chim. Acta 54 (1981) L259.
[33] J. Vicente, M.T. Chicote, M.D. Bermúdez, Inorg. Chim. Acta 63 (1982) 35.
[34] R.G. Pearson, Inorg. Chem. 12 (1973) 712.
0.005(7)
Diffractometer. Absorption was corrected for by semi-empirical
methods (SADABS [44]).
4.16.2. Solution and refinement
Structures were solved by either the direct methods (24) or
Patterson (14 and 20) options of Shelxs-97 [45]. In all cases the
gold atom was initially located, and all other atoms by a series of
difference maps. Full-matrix least-squares refinement (Shelxl-97
[46]) was based upon F02, with all non-hydrogen atoms refined
anisotropically (with the exception of the cases mentioned below)
and hydrogen atoms in calculated positions.
Complex 24 crystallised with the phosphorus of one PF6ꢀ falling
on an inversion centre. The other half anion was in a hole which
contained both a PF6 and a diethyl ether molecule (50:50 occu-
ꢀ
ꢀ
pancy). Both the PF6 and the diethyl ether showed considerable
disorder and as a result were not refined anisotropically and the
hydrogen atoms were excluded from the diethyl ether. Bond
lengths in these species were constrained.
Complex 14 showed significant ripples around the gold after the
final refinement. As a result, the anisotropic displacement param-
eters for C(12) and C(13) were unrealistic and thus were refined
isotropically.
[35] Y. Fuchita, H. Ieda, A. Kayama, J. Kinoshita-Nagaoka, H. Kawano, S. Kameda, M.
Mikuriya, J. Chem. Soc., Dalton Trans. (1998) 4095.
[36] Y. Fuchita, H. Ieda, Y. Tsunemune, J. Kinoshita-Nagaoka, H. Kawano, J. Chem.
Soc., Dalton Trans. (1998) 791.
[37] Y. Fuchita, H. Ieda, S. Wada, S. Kameda, M. Mikuriya, J. Chem. Soc., Dalton
Trans. (1999) 4431.
[38] Y. Fuchita, H. Ieda, M. Yasutake, J. Chem. Soc., Dalton Trans. (2000) 271.
[39] R.V. Parish, B.P. Howe, J.P. Wright, J. Mack, R.G. Pritchard, R.G. Buckley, A.M.
Elsome, S.P. Fricker, Inorg. Chem. 35 (1996) 1659.
[40] P.A. Bonnardel, R.V. Parish, R.G. Pritchard, J. Chem. Soc., Dalton Trans. (1996)
3185.
[41] E.R.T. Tiekink, Crit. Rev. Oncol./Hematol. 42 (2002) 225.
[42] R.V. Parish, Metal-Based Drugs 6 (1999) 271.
[43] W. Henderson, B.K. Nicholson, A.L. Wilkins, J. Organomet. Chem. 690 (2005)
4971.
Acknowledgements
We thank the University of Waikato for financial support of this
work, the Tertiary Education Commission for a Top Achievers Doc-
toral Scholarship (KJK) and the New Zealand Federation of Gradu-
ate Women for a Merit Award (KJK). We also thank Dr. Jan Wikaira
(University of Canterbury) for collection of X-ray intensity data and
Gill Ellis (University of Canterbury) for biological assays. Ms Fran-
ces Gourdie is thanked for the synthesis of Ph3P@NC6H4F and Mr.
Bevan Jarman for helpful discussions.
[44] R.H. Blessing, Acta Cryst. A51 (1995) 33.
[45] G.M. Sheldrick, Shelxs-97 – a Program for the Solution of Crystal Structures,
University of Göttingen, Germany, 1997.
Appendix A. Supplementary data
[46] G.M. Sheldrick, Shelxl-97 – a Program for the Refinement of Crystal Structures,
University of Göttingen, Germany, 1997.
CCDC 699890, 699891 and 699892 contain the supplementary
crystallographic data for this paper. These data can be obtained