The synthesis and structural characterisation of 2-arsa- and 2-stiba-1,3-dionato complexes of s- and p-block elements

Article abstract of DOI:10.1039/b210043f

The syntheses and structural characterisations of a number of 2-arsa- and 2-stiba-1,3-dionato complexes of main group elements are reported. The synthetic methods that were employed in this study include the reaction of lithium heterodionate complexes, [Li{η²-O,O-OC(R)EC(R)O}], E = As or Sb, R = alkyl or aryl, with main group metal halides or the reactions of arsa- and stiba-enols, O=C(R)E=C(R)OH, with either main group alkyl or amide complexes. The hetero-ligands in the prepared complexes have been shown by X-ray crystallographic studies to coordinate the main group metal centre in several modes, that is, in a delocalised form (η²-O,O coordination) as in [{M[η²-O,O-OC(R)EC(R)O](L)}₂], M = Li or Na, E = As or Sb, R = Ad (adamantyl) or Ph, L = Et₂O or DME; [Na{η²-O,O-OC(Mes*)EC(Mes*)O}(DME)₂], E = As or Sb, Mes* = C₆H₂Bu₃^t-2,4,6, and [Mg{η²-O,O-OC(Ph)AsC(Ph)O}₂(DME)]; in a localised diacylpnictido form (η¹-E coordination) as in P[As{C(O)R}₂]₃, R = But or Ph; or as a combination of the two modes (η-E-: η²-O,O-) as in [LiCl(Et₂O)(DME){η²-O,O-Cl₃InAs[C(O)R] ₂}].

Full text of DOI:10.1039/b210043f

View Article Online / Journal Homepage / Table of Contents for this issue
The synthesis and structural characterisation of 2-arsa- and
2-stiba-1,3-dionato complexes of s- and p-block elements
Stephen Bruce,a David E. Hibbs,a Cameron Jones,*a Jonathan W. Steed,b Ryan C. Thomasa and
Thomas C. Williamsa
a
Department of Chemistry, Cardiff University, P.O. Box 912, Park Place,
Cardiff CF10 3TB, U.K.
Department of Chemistry, King’s College London, Strand, London WC2R 2LS, U.K.
b
Received (in Montpellier, France) 10th October 2002, Accepted 23rd November 2002
First published as an Advance Article on the web 6th February 2003
The syntheses and structural characterisations of a number of 2-arsa- and 2-stiba-1,3-dionato complexes
of main group elements are reported. The synthetic methods that were employed in this study include the
reaction of lithium heterodionate complexes, [Li{Z2-O,O–OC(R)EC(R)O}], E ¼ As or Sb, R ¼ alkyl or aryl,
=
=
with main group metal halides or the reactions of arsa- and stiba-enols, O C(R)E C(R)OH, with either main
group alkyl or amide complexes. The hetero-ligands in the prepared complexes have been shown by X-ray
crystallographic studies to coordinate the main group metal centre in several modes, that is, in a delocalised
form (Z2-O,O coordination) as in [{M[Z2-O,O–OC(R)EC(R)O](L)}2], M ¼ Li or Na, E ¼ As or Sb, R ¼ Ad
(adamantyl) or Ph, L ¼ Et2O or DME; [Na{Z2-O,O–OC(Mes*)EC(Mes*)O}(DME)2], E ¼ As or Sb,
Mes* ¼ C6H2But3-2,4,6, and [Mg{Z2-O,O–OC(Ph)AsC(Ph)O}2(DME)]; in a localised diacylpnictido form
(Z1-E coordination) as in P[As{C(O)R}2]3 , R ¼ But or Ph; or as a combination of the two modes
(Z1-E–: Z2-O,O–) as in [LiCl(Et2O)(DME){Z2-O,O–Cl3InAs[C(O)R]2}].
Introduction
Results and discussion
One of our main research interests lies with developing the
chemistry of ligand systems in which one or more CR frag-
ments (R ¼ H or alkyl) of classical hydrocarbon-based ligand
systems are substituted with the valence isoelectronic heavier
group 15 elements, As or Sb.1 In this context we have recently
been examining the transition metal coordination chemistry of
the 2-arsa and 2-stiba-1,3-dionates, [OC(R)EC(R)O]ꢀ (1, E ¼
As or Sb, R ¼ alkyl or aryl). Although these anions are closely
related to the ubiquitous b-diketonato class of ligands they
have proved to be much more versatile in their coordination
behaviour. For example, whereas b-diketonates generally bind
transition metals in variations of the chelating Z2-O,O mode,
the presence of a low coordinate pnictide centre in 1 allows
them to coordinate metals in a variety of ways, which include
Z2-O,O-; Z1-E; Z1-E:Z1-O:Z2-CO; Z1-E:Z2-O,O- and m-Z1:Z1-
E.2,3 In addition, the weakness of the Sb–C bond of 2-stiba-
1,3-dionates has led to their use as precursors to previously
inaccessible complex types such as the first example of a func-
tionalised distibene, which is stabilised within the coordination
sphere of a transition metal fragment: cis-[Pt(PEt3)2{Z2-Sb,Sb–
Group 1
In a prior study, the synthesis and structural characterisation
of the tert-butyl substituted lithium heterodionates, [{[Li{Z2-
O,O–OC(But)EC(But)O}(DME)0.5]2} ], E ¼ As 2, Sb 3, and
1
[{Li[Z2-O,O–OC(But)AsC(But)O](OEt2)}2] 4, have been
reported.6 It was found that in these syntheses the optimum
yield was achieved when the ratio of acid chloride and lithium
pnictide reactants was 2:3. In addition, it was revealed that
both E(SiMe3)3 and LiCl were generated as by-products and
so the mechanism of formation of 2–4 was proposed to be as
outlined in Scheme 1.
As 2–4 represent the only three structurally characterised
lithium arsa- or stibadionates known to date, it was decided
to prepare a number more to shed light on the degree of gen-
erality of the synthetic technique mentioned above and to
allow structural comparisons to be carried out. To this end
adamantyl acid chloride, AdC(O)Cl, was reacted with [Li{Sb-
(SiMe3)2}] in a 2:3 ratio, which afforded the stibadionato
lithium complex, 5, in high yield after recrystallisation from
diethyl ether (Scheme 1). Altering the reaction stoichiometry
lowered the yield of 5 but under all conditions Sb(SiMe3)3
was shown to be a reaction by-product. This strongly suggests
the mechanism of formation shown in Scheme 1 is in opera-
tion. All the spectroscopic data for 5 are as expected and
are similar to those for the closely related complex, 3. The for-
mer is, however, significantly more thermally stable than the
latter (5 dec. 98 ꢁC, 3 dec. 65 ꢁC), presumably due to the
enhanced kinetic stability afforded by the bulkier adamantyl
substituents.
t
t
4
=
C(Bu )(O)Sb SbC(O)(Bu )}].
Despite the coordinative versatility that heterodionates have
shown towards transition metals, little work has looked at
their interaction with main group elements. Indeed, this can
be confined to the preparation of the lithium salts of these
ligands, [Li{Z2-O,O–OC(R)EC(R)O}],5,6 which are invariably
used as starting materials in the transition metal based studies.
In addition, we have made a preliminary report7 on the synth-
esis and structural characterisation of the first tris(diacylarseni-
do)phosphines, P[As{C(O)R}2]3 , R ¼ But or Ph. We have
sought to extend these results to the formation of a variety
of heterodionate complexes of both s- and p-block elements
and to compare their chemistry with that of their b-diketonate
analogues. The results of this study are reported herein.
Interestingly, when an attempt was made to prepare the phe-
nyl substituted analogues of 2 and 4,that is 6 and 7, by the 2:3
reaction of PhC(O)Cl with [Li{As(SiMe3)2}], only low yields
were obtained and much of the arsenic starting material
466
New J. Chem., 2003, 27, 466–474
DOI: 10.1039/b210043f
This journal is # The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2003

View Article Online
solution, which involves a rapid breaking and reforming of
bridging Li–O bonds as has been suggested for 2 and 4.
Attempts to investigate this process with variable temperature
NMR spectroscopic experiments were thwarted by the low
solubility of the complexes in non-coordinating solvents at
temperatures below 0 ꢁC.
The X-ray crystal structures of 5–7 were obtained and
are depicted in Figs. 1–3 (see Table 1). All are dimeric and
contain essentially planar central Li2O2 four-membered rings.
The lithium centres in each compound are chelated by the
heterodionate ligand and have an additional intermolecular
coordination from the bridging oxygen centre of the other
heterodionate ligand in the dimer. Compounds 5 and 7 have
further lithium coordination from diethyl ether molecules
whilst DME molecules chelate the 5-coordinate lithium centres
in 6. This arrangement is identical to that in the phosphorus
analogue of 68 but differs from the situation in 2 where the
DME molecules bridge dimeric units into an infinite polymeric
chain. The differences here most probably arise from the
greater steric crowding about the lithium centres in 2, which
disfavour DME chelation.
In each of 5–7 the 6-membered LiOCECO chelate ring is
non-planar with the E and Li centres sitting slightly out of
the least squares plane defined by both C and O centres.
Despite this, there appears to be a significant degree of deloca-
lisation over the heterodionate ligands as the E–C and C–O
bond lengths lie between those normally seen for double and
single bonded interactions.9 It must be said, however, that
the E–C bond lengths are closer to single than double bonds,
which probably reflects a relative accumulation of negative
charge at the hetero-centres.
Scheme 1 Reagents and conditions: (i) DME, ꢀLiCl; (ii) Li{E-
(SiMe3)2}, ꢀE(SiMe3)3 ; (iii) RC(O)Cl, ꢀLiCl; (iv) Li{E(SiMe3)2},
ꢀE(SiMe3)3 .
remained unreacted. It was eventually found that the optimum
reaction stoichiometry was 2:1 and that no As(SiMe3)3 was
generated. Therefore, a different reaction mechanism to that
which yielded 2–5 must be in operation. This is proposed to
be as outlined in Scheme 2. Instead of salt elimination occur-
ring there are two sequential SiMe3Cl eliminations and an
intramolecular rearrangement that gives a high yield of the
products, 6 or 7, depending on which solvent is employed to
recrystallise the crude product. Evidence for this mechanism
came from a GC/MS analysis of the reaction volatiles, which
shows both SiMe3Cl and (Me3Si)2O as major reaction by-pro-
ducts. The latter is a known hydrolysis product of SiMe3Cl. It
is not known why two very different mechanisms are in opera-
tion to form such products but the reason probably lies with
the smaller steric bulk of the Ph group relative to the But sub-
stituent, the electronic differences between these groups, or a
combination of both. It was attempted to form the stibadio-
nate analogues of 6 and 7 using similar reaction conditions
but the products proved to be unstable at ambient temperature
and decomposed, affording elemental antimony and many
other products.
There are no known examples of 2-bisma-1,3-dionato com-
plexes, which is not surprising given the expected decrease in
their thermal stability relative to their already thermally frail
antimony counterparts. Despite this, an attempt was made to
prepare a sterically hindered lithium bismadionato complex,
Again, the spectroscopic data for 6 and 7 are similar to those
for the analogous tert-butyl substituted complexes, 2 and 4.6
As for these complexes the NMR data suggest a more symme-
trical structure in solution than was observed in the solid state
(vide infra) because the acyl carbon centres are chemically
equivalent. This is probably due to a fluxional process in
˚
Fig. 1 Molecular structure of compound 5. Selected bond lengths (A)
and angles (ꢁ): Sb(1)–C(12) 2.119(4), Sb(1)–C(1) 2.187(3), O(1)–C(1)
1.236(4), O(1)–Li(1) 1.916(7), O(2)–C(12) 1.266(4), O(2)–Li(1)0
1.914(7), O(2)–Li(1) 1.926(7), O(3)–Li(1) 1.950(7), C(12)–Sb(1)–C(1)
95.36(13), C(1)–O(1)–Li(1) 130.8(3), C(12)–O(2)–Li(1)0 146.6(3),
C(12)–O(2)–Li(1) 125.2(3), Li(1)0–O(2)–Li(1) 88.1(3), O(1)–C(1)–
Sb(1) 124.4(3), O(2)–C(12)–Sb(1) 124.9(3), O(2)0–Li(1)–O(1) 125.8(4),
O(2)0–Li(1)–O(2) 91.9(3), O(1)–Li(1)–O(2) 98.9(3), O(2)0–Li(1)–O(3)
106.6(3), O(1)–Li(1)–O(3) 118.5(3), O(2)–Li(1)–O(3) 110.3(3). Symme-
Scheme 2 Reagents and conditions: (i) DME, ꢀSiMe3Cl; (ii)
PhC(O)Cl, ꢀSiMe3Cl.
0
try operation : ꢀx + 2, ꢀy, ꢀz + 1.
New J. Chem., 2003, 27, 466–474
467

View Article Online
[Li{Z2-O,O–OC(Mes*)BiC(Mes*)O}], Mes* ¼ C6H2But3-2,4,6,
by the reaction of 2 equiv. of Mes*C(O)Cl with 3 equiv. of
[Li{Bi(SiMe3)2}] in an analogous fashion to the preparation
of 5. This reaction, however, unexpectedly led to moderate
yields of the known dibismuthane, (Me3Si)2BiBi(SiMe3)2 ,10
and the a-diketone, {Mes*C(O)}2 . Presumably, these arose
from the initial formation of Mes*C(O)Bi(SiMe3)2 , which
underwent a thermally or photolytically induced homolytic
Bi–C bond cleavage, whilst a subsequent coupling of the
generated fragments gave the observed products. As a result of
this outcome no further attempts were made to prepare a bis-
madionate complex.
It was deemed of sufficient interest to attempt the prepara-
tion of heterodionate complexes of the heavier alkali metal,
sodium. The rationale for this was that sodium heterodionates
were considered as having potential as transfer reagents in the
synthesis of lanthanide-heterodionate complexes from lantha-
nide halides where sodium halide elimination would be the
reaction driving force. The analogous lithium complexes are
not suitable in this respect. They are, however, useful in the
formation of arsa- and stiba-enol compounds, 8, when treated
with protonating agents. These hetero-enols have previously
been shown to exist predominantly in the enol, and not the
keto tautomeric form, in both solution and the solid state.5,6
Therefore, the treatment of several of these with the sodium
amide, [Na{N(SiMe3)2}], was investigated and in all cases
the expected sodium heterodionate complexes, 9–11, were
formed in moderate to high yields (Scheme 3).
˚
Fig. 2 Molecular structure of compound 6. Selected bond lengths (A)
and angles (ꢁ): As(1)–C(8) 1.915(2), As(1)–C(7) 1.942(2), O(1)–C(7)
1.240(2), O(1)–Li(1) 1.942(3), O(2)–C(8) 1.259(2), O(2)–Li(1)
2.012(3), O(2)–Li(1)0 2.038(4), O(3)–Li(1) 2.153(4), O(4)–Li(1)
2.092(4), C(8)–As(1)–C(7) 99.19(7), C(7)–O(1)–Li(1) 137.3(2), C(8)–
O(2)–Li(1) 133.7(2), C(8)–O(2)–Li(1)0 120.3(2), Li(1)–O(2)–Li(1)0
88.0(2), O(1)–C(7)–As(1) 127.50(13), O(2)–C(8)–As(1) 127.64(13),
O(1)–Li(1)–O(2) 91.49(14), O(1)–Li(1)–O(2)0 111.3(2), O(2)–Li(1)–
1
In the case of 9 the H NMR spectrum was very similar to
that of its lithium analogue, 6, and was indicative of one mole-
cule of DME per sodium centre. As a result it was proposed
that this compound had a related dimeric structure. When
the very bulky supermesityl ligand, Mes*, was employed the
1H NMR spectra of both the arsa- and stibadionate complexes
were very similar and pointed towards two chelating DME
molecules per sodium centre. The most likely structural
arrangement that would allow this was thought to be mono-
meric with octahedrally coordinated sodium centres. This
was confirmed in the solid state by X-ray crystallography.
The molecular structures of 9–11 are depicted in Figs. 4–6
(see Table 1). Compound 9 is isostructural and isomorphous
to 6 and as in that compound the As–C and C–O bond lengths
lie between those expected for double and single interactions.
The only significant difference between the two structures are
the metal-oxygen distances, which are, not surprisingly, mark-
edly longer in the sodium complex. The other sodium hetero-
dionate complexes, 10 and 11, are isostructural but not
isomorphous due to the inclusion of half a molecule of crystal-
lisation of DME for each molecule of 10 in its crystal structure.
In addition, in 11 there are two crystallographically indepen-
dent half molecules in the asymmetric unit, which sit on 2-fold
rotation axes. The full molecules that are generated by these
symmetry elements have very similar geometries and so only
one is shown in Fig. 6. As in the previously mentioned dimeric
heterodionato lithium and sodium complexes the bond lengths
within the heterodionate fragments are consistent with signifi-
cant delocalisation. One difference that was observed between
the monomeric and dimeric complexes is that in the dimeric
complexes the 6-membered LiOCECO rings are significantly
distorted from planarity, whereas in 10 and 11 the NaOCECO
rings are essentially planar. In addition, 6-coordinate metal
centres are not observed in the dimers whilst in 10 and 11
the sodium centres possess distorted octahedral geometries
arising from two chelating DME molecules and one heterodio-
nate ligand.
0
O(2)0 92.0(3). Symmetry operation : ꢀx + 1, ꢀy, ꢀz.
˚
Fig. 3 Molecular structure of compound 7. Selected bond lengths (A)
and angles (ꢁ): As(1)–C(8) 1.934(8), As(1)–C(1) 1.946(10), As(2)–C(22)
1.869(10), As(2)–C(15) 1.955(11), O(1)–C(1) 1.222(13), O(1)–Li(1)
1.83(2), O(2)–C(8) 1.230(12), O(2)–Li(1) 1.92(2), O(2)–Li(2)
1.903(18), O(3)–C(15) 1.276(13), O(3)–Li(2) 1.919(18), O(4)–C(22)
1.306(12), O(4)–Li(2) 1.918(19), O(4)–Li(1) 2.054(19), Li(1)–O(1S)
1.96(2), Li(2)–O(2S) 1.948(19), C(8)–As(1)–C(1) 98.1(4), C(22)–
As(2)–C(15) 102.4(4), C(1)–O(1)–Li(1) 130.2(9), C(8)–O(2)–Li(1)
128.2(8), C(8)–O(2)–Li(2) 124.6(8), Li(1)–O(2)–Li(2) 92.3(8), C(15)–
O(3)–Li(2) 133.7(10), C(22)–O(4)–Li(2) 133.2(8), C(22)–O(4)–Li(1)
122.2(8), Li(2)–O(4)–Li(1) 87.9(7), O(1)–C(1)–As(1) 128.3(7), O(3)–
C(15)–As(2) 124.7(8), O(4)–C(22)–As(2) 126.8(8).
Group 2
Prior to this work there were no reported examples of group 2
heterodionate complexes. This contrasts with the chemistry of
468
New J. Chem., 2003, 27, 466–474

View Article Online
Table 1 Crystal data for compounds 5, 6, 7, 9, 10ꢂDME0.5 , 11, 12, 13, and 15ꢂEt2O
5
6
7
9
10ꢂDME0.5
11
12
32H30
13
15ꢂEt2O
Chemical
formula
C26H40O3-
SbLi
C18H20-
AsLiO4
382.20
Monoclinic
P21/n
C18H20
-
C18H20
-
C48H83
-
C46H78
-
C
-
C18H38
-
C46H40
As3O7P
960.51
Triclinic
-
AsLiO3
366.20
AsNaO4
398.25
AsNaO7
870.05
NaO6Sb
871.82
As2MgO6
AsCl3InLiO5
637.51
FW
Crystal system
Space group
529.29
Monoclinic
P21/c
684.71
Monoclinic
Cc
Triclinic
P1
Monoclinic
P21/n
Monoclinic
P21/c
Monoclinic
C2
Orthorhombic
Pbca
¯
P1
˚
a/A
11.7360(5)
17.3904(8)
12.5818(6)
90
12.5633(3)
7.5209(6)
18.9855(8)
90
8.5317(3)
9.7835(5)
11.7665(6)
112.465(3)
102.759(3)
96.788(3)
862.55(7)
2
12.5803(10)
7.6097(7)
19.678(2)
90
14.440(3)
19.134(4)
19.562(4)
90
27.506(3)
10.440(2)
19.595(3)
90
8.4986(6)
22.694(2)
16.2467(11)
90
17.155(3)
18.028(4)
18.480(4)
90
13.075(3)
14.159(3)
15.026(3)
97.50(3)
111.62(3)
117.04(3)
2149.0(7)
2
˚
b/A
˚
c/A
a/ꢁ
b/ꢁ
g/ꢁ
98.619(3)
90
105.605(2)
90
103.446(10)
90
109.36(3)
90
121.486(12)
90
105.137(4)
90
90
90
3
˚
U/A
2538.9(2)
4
100(2)
1727.8(2)
4
100(2)
1832.2(3)
4
150(2)
5099.1(18)
4
4798.5(13)
4
3024(7)
4
5715(2)
8
Z
T/K
100(2)
1.98
100(2)
0.72
150(2)
0.63
100(2)
1.50
150(2)
2.279
150(2)
2.406
m(Mo–Ka)/
mmꢀ1
0.55
1.98
1.89
Reflections
collected
Unique
20 437
5803
34 967
4451
10 780
6662
4196
3715
58 037
11 668
4690
4585
34 110
4962
91 098
6540
35 408
9793
reflections
Rint
0.0573
0.0504
0.0977
0.0730
0.0340
0.1102
0.0390
0.0365
0.0932
0.0300
0.0436
0.1199
0.0657
0.0471
0.1045
0.0734
0.0765
0.1943
0.0600
0.0681
0.1630
0.1246
0.0665
0.1521
0.0441
0.0311
0.0699
R1 [I > 2s(I)]
wR2 (all data)
˚
group 2 b-diketonate complexes, which has been well devel-
oped, primarily due to the wide utilisation these complexes
have found as precursors to thin films in MOCVD technol-
ogy.11 In this study many attempts were made to prepare both
arsa- and stibadionato complexes of group 2 metals via the
reaction of hetero-enols, 8, with metal alkyls and amides. In
all cases involving the stiba-enols, reactions occurred but
decomposition of the products rapidly took place above 0 ꢁC
to give elemental antimony and a mixture of other products.
The reactions involving arsa-enols were cleaner but in most
cases inseparable mixtures of products resulted. The one
exception to this was the moderately yielding formation of
the bis(arsadionato)magnesium complex, 12, from the reaction
depicted in Scheme 3. The compound is moderately thermally
stable (dec. 137 ꢁC) and its NMR spectroscopic data are com-
patible with a monomeric octahedral complex incorporating
one chelating DME molecule. The infrared spectrum of 12 dis-
plays strong absorptions (n 1524, 1548 cmꢀ1) in the expected
region for a dionato ligand chelating a metal centre.
complexes, that is 2.006
A
ave. in [Mg{OC(But)C(H)-
C(But)O}2(TMEDA)],12 but longer than the interactions of the
˚
DME ligand with the magnesium centre (Mg–O 2.130 A ave).
As in the previously mentioned sodium and lithium arsadio-
nate complexes the ligands are delocalised, as evidenced by
the As–C (1.922 A ave.) and C–O (1.253 A ave.) bond lengths,
which are intermediate between double and single bonded
interactions.9 In addition, both arsadionate ligands are essen-
tially planar and co-planar with the phenyl substituents. Pre-
sumably the latter observation arises from the conjagative
stability that this arrangement affords the complex.
˚
˚
An X-ray crystal structure analysis of 12 confirmed this and
showed the magnesium centre to have a distorted octahedral
geometry (Fig. 7, Table 1). The distances from the metal centre
˚
to the arsadionate oxygen centres are all similar (2.016 A ave.)
and close to those normally seen for magnesium b-diketonate
˚
Fig. 4 Molecular structure of compound 9. Selected bond lengths (A)
and angles (ꢁ): As(1)–C(8) 1.916(3), As(1)–C(1) 1.939(3), O(1)–C(1)
1.242(4), O(2)–C(8) 1.248(4), O(1)–Na(1)0 2.201(3), O(2)–Na(1)0
2.261(3), Na(1)–O(4) 2.295(3), Na(1)–O(2) 2.342(4), Na(1)–O(3)
2.554(17), C(8)–As(1)–C(1) 101.24(14), O(1)–C(1)–As(1) 127.6(3),
O(2)–C(8)–As(1) 127.1(3), C(8)–O(2)–Na(1) 120.4(3), C(8)–O(2)–
Na(1)0 120.4(3), C(1)–O(1)–Na(1)0 140.3(3), O(1)0–Na(1)–O(2)
110.08(14), O(2)0–Na(1)–O(2) 89.17(12). Symmetry operation 0: ꢀx,
ꢀy + 1, ꢀz.
Scheme 3 Reagents and conditions: (i) DME, Na{N(SiMe3)2},
ꢀHN(SiMe3)2 ; (ii) DME, MgBu2 , ꢀBuH.
New J. Chem., 2003, 27, 466–474
469

View Article Online
Fig. 5 Molecular structure of compound 10. Selected bond lengths
ꢁ
˚
(A) and angles ( ): As(1)–C(1) 1.911(2), As(1)–C(20) 1.943(2), Na(1)–
O(2) 2.2351(17), Na(1)–O(1) 2.3036(17), Na(1)–O(5) 2.3777(19),
Na(1)–O(3) 2.3967(18), Na(1)–O(6) 2.4177(18), Na(1)–O(4)
2.4830(19), O(1)–C(1) 1.243(3), O(2)–C(20) 1.236(3), C(1)–As(1)–
C(20) 101.60(9), O(2)–Na(1)–O(1) 83.45(6), O(3)–Na(1)–O(4)
66.25(6), O(5)–Na(1)–O(6) 68.28(6), C(1)–O(1)–Na(1) 135.61(14),
C(20)–O(2)–Na(1) 136.27(13), O(1)–C(1)–As(1) 128.20(16), O(2)–
C(20)–As(1) 129.06(15).
Fig. 7 Moleculꢁar structure of compound 12. Selected bond lengths
˚
(A) and angles ( ): As(1)–C(8) 1.916(10), As(1)–C(1) 1.938(9), As(2)–
C(15) 1.908(10), As(2)–C(22) 1.928(9), Mg(1)–O(3) 2.000(6), Mg(1)–
O(2) 2.007(6), Mg(1)–O(4) 2.028(7), Mg(1)–O(1) 2.029(7), Mg(1)–
O(6) 2.129(6), Mg(1)–O(5) 2.131(6), O(1)–C(1) 1.253(9), O(2)–C(8)
1.251(10), O(4)–C(22) 1.261(9), O(3)–C(15) 1.250(9), C(8)–As(1)–C(1)
99.1(4), O(3)–Mg(1)–O(4) 88.5(3), O(2)–Mg(1)–O(1) 88.4(3), O(6)–
Mg(1)–O(5) 74.5(3), C(1)–O(1)–Mg(1) 132.3(6), O(1)–C(1)–As(1)
127.6(8), C(15)–As(2)–C(22) 99.2(4), C(8)–O(2)–Mg(1)–137.7(7),
C(15)–O(3)–Mg(1) 137.5(7), C(22)–O(4)–Mg(1) 132.0(6), O(2)–C(8)–
As(1) 126.8(7), O(3)–C(15)–As(2) 127.2(8), O(4)–C(22)–As(2) 127.6(8).
Group 13
There have been no reported examples of arsa- or stibadionato
group 13 complexes. In order to initiate this field the reactions
of 2 and 3 with ‘‘GaI’’13 were examined. The intention here
was to utilise the bulky tert-butyl substituents on the dionato
ligands to stabilise gallium carbene analogues, [:Ga{Z2-O,O–
OC(But)EC(But)O}], E ¼ As, Sb, related examples of which,
derived from diazabutadiene14 and b-diketiminato15 ligands,
have recently been reported. Unfortunately, these reactions
were not clean and led to intractable mixtures of products. It
is noteworthy that a similar result occurred when 2 was reacted
with GaCl3 in 1:1 or 3:1 stoichiometries.
More success was had when 2 was treated with 1 equiv. of
InCl, though the outcome was not expected. When the reac-
tion was carried out in a Et2O–DME mixture, the unusual
complex, 13, was formed in good yield (Scheme 4). This is best
considered as containing an anionic arsenic centre that forms a
donor-acceptor As–In bond but also possesses an uncoordi-
nated lone pair. The arsadionate ligand in the complex is act-
ing as a localised diacylarsenide, which additionally chelates a
cationic Li(OEt2)(DME) fragment through dative interactions
with its oxygen centres. This unusual complex can be thought
of as an intermediate in a salt elimination reaction that would
give [InCl2{Z2-O,O–OC(But)AsC(But)O}]. However, heating
solutions of 13 to 50 ꢁC did not lead to this complex but an
inseparable mixture of many products. In addition, attempts
to prepare 13 by the direct treatment of InCl3 with 2 in a
DME–Et2O solvent mixture led to no reaction, presumably
because of the low solubility of InCl3 in this solvent system.
Fig. 6 Molecular structure of compound 11. Selected bond lengths
(A) and angles (ꢁ): Sb(1)–C(1) 2.156(11), Na(1)–O(1) 2.282(12),
˚
Na(1)–O(2) 2.381(12), Na(1)–O(3) 2.474(16), O(1)–C(1) 1.222(16),
C(1)–Sb(1)–C(1)0 96.0(6), O(1)–Na(1)–O(1)0 83.0(5), O(2)–Na(1)0–
O(3)0 75.2(5), O(1)–C(1)–Sb(1) 127.5(8). Symmetry operation
:
Scheme 4 Reagents and conditions: (i) InCl, DME–Et2O, ꢀLiCl,
ꢀIn(s) ; (ii) PCl3 , DME, ꢀLiCl; (iii) PhSeCl, DME, ꢀLiCl.
ꢀx + 1, y, ꢀz + 1.
470
New J. Chem., 2003, 27, 466–474

View Article Online
With regards to a mechanism of formation for 13 it seems
likely that in the presence of the arsenic lone pairs of 2, InCl
disproportionates into InCl3 and indium metal with a conco-
mitant formation of the observed complex. Indeed, it is well
known that when treated with group 15 donors InCl can
disproportionate16 and in the current reaction significant
amounts of indium metal deposition were observed.
nates, 2 or 3, with either SnCl2 or PbCl2 which in all cases
led to no reaction occurring, presumably as a result of the
low solubility of the metal salts. To overcome this problem 2
equiv. of the arsa-enol, 8 R ¼ Ph, E ¼ As, were reacted with
[M{N(SiMe3)2}2], M ¼ Sn or Pb (cf. the formation of 9 and
10), but only intractable mixtures of products were formed.
Similarly, the reactions of 2 or 3 with Me2SnCl2 , Me2SiCl2
and Ph3SnCl all afforded oily mixtures of many products that
could not be separated by normal techniques.
The spectroscopic data for 13 are fully consistent with its
proposed formulation. In particular its infrared spectrum
shows acyl C–O stretching bands in the region normally
observed for localised diacylarsenide ligands but at higher fre-
quencies than in delocalised arsadionates. The molecular struc-
ture of 13 is depicted in Fig. 8 (Table 1). It is monomeric and
its lithium centre possesses a distorted trigonal bipyramidal
coordination environment with O(2) and O(4) in apical posi-
tions and all Li–O interactions in the normal range. An exam-
Group 15
In a preliminary communication7 we have recently reported
the synthesis of the first diacylarsenido group 15 compounds,
P[As{C(O)R}2]3 , R ¼ But 14, Ph 15, which were formed in
the 3:1 reactions of the lithium arsadionates, 2 and 6, with
PCl3 (Scheme 4). Interestingly, when the analogous reactions
with either AsCl3 or SbCl3 were carried out, elemental arsenic
or antimony deposited and the known tetraacyldiarsanes,
[As{C(O)R}2]2 ,2 were formed in high yields via oxidative cou-
pling reactions. Reactions of the less stable stibadionate, 3,
with ECl3 , E ¼ P, As or Sb, led to intractable mixtures of pro-
ducts. We have subsequently also found that the reactions of 1
or 2 equiv. of 2 or 6 with PCl3 do not lead to the formation of
˚
˚
ination of the As–C (2.026 A ave.) and acyl C–O (1.204 A)
bond lengths shows them to be normal for single and double
interactions, and to be respectively longer and shorter than
the related bonds in delocalised arsadionato complexes, such
as 6. The geometry about the arsenic centre is heavily distorted
pyramidal with both the In–As–C (92.66ꢁ ave.) and C–As–C
[98.9(3)ꢁ] angles being acute. The acuteness of these angles
implies that a high degree of p-character exists in the bonds
to As(1) and therefore that its lone pair has predominantly
s-character. The indium centre in 13 has a slightly distorted tet-
rahedral environment with the In–Cl bond lengths in the nor-
the mono- or bis(arsino)phosphines, ClnP[As{C(O)R}2]3 ꢀ n
,
n ¼ 1 or 2, but instead the tris(arsino)phosphines, 14 or 15,
are formed leaving unreacted PCl3 .
˚
mal range and an As–In bond length [2.6094(9) A] somewhat
shorter than the mean for all crystallographically determined
The spectroscopic data for both compounds have previously
been discussed and will not be further analysed here, though
full synthetic details for the two compounds are included in
the experimental section. In addition, details of the X-ray crys-
tal structure of 15 are reported here for the first time, Fig. 9
9
In–As bonds (2.725 A). The reason for this lies with the fact
˚
that compound 13 contains the only example of an In–As bond
in which the arsenic centre is 3-coordinate and not bridging
two indium centres.
Group 14
There are no known examples of arsa- or stibadionato-group
14 complexes. In this study a variety of reactions were carried
out to address this paucity but all met with failure. These
included the treatment of 2 equiv. of the lithium heterodio-
Fig. 9 Molecular structure of compound 15. Selected bond lengths
ꢁ
˚
Fig. 8 Molecular structure of compound 13. Selected bond lengths
(A) and angles ( ): As(1)–C(2) 2.023(2), As(1)–C(1) 2.027(3), As(1)–
P(1) 2.3010(9), As(2)–C(15) 2.027(2), As(2)–C(16) 2.037(2), As(2)–
P(1) 2.3057(15), As(3)–C(29) 2.031(2), As(3)–C(30) 2.037(2), As(3)–
P(1) 2.3063(9), O(1)–C(1) 1.211(3), O(2)–C(2) 1.210(3), O(3)–C(15)
1.209(3), O(4)–C(16) 1.207(3), O(5)–C(29) 1.210(3), O(6)–C(30)
1.209(3), C(2)–As(1)–C(1) 90.94(9), C(15)–As(2)–C(16) 93.39(9),
C(29)–As(3)–C(30) 89.18(9), As(1)–P(1)–As(2) 105.79(5), As(1)–P(1)–
As(3) 108.63(3), As(2)–P(1)–As(3) 105.72(4), O(1)–C(1)–As(1)
ꢁ
˚
(A) and angles ( ): In(1)–Cl(3) 2.3680(18), In(1)–Cl(1) 2.3744(19),
In(1)–Cl(2) 2.3748(18), In(1)–As(1) 2.6094(9), As(1)–C(2) 2.024(6),
As(1)–C(1) 2.028(6), O(1)–C(1) 1.209(8), O(1)–Li(1) 1.976(13), O(2)–
C(2) 1.200(7), O(2)–Li(1) 2.015(12), O(3)–Li(1) 2.135(15), O(4)–Li(1)
2.097(14), O(5)–Li(1) 1.990(14), C(2)–As(1)–C(1) 98.9(3), C(2)–
As(1)–In(1) 93.60(17), C(1)–As(1)–In(1) 91.72(18), Cl(3)–In(1)–Cl(1)
106.60(7), Cl(3)–In(1)–Cl(2) 106.84(7), Cl(1)–In(1)–Cl(2) 106.83(8),
Cl(3)–In(1)–As(1) 111.51(6), Cl(1)–In(1)–As(1) 112.55(5), Cl(2)–
In(1)–As(1) 112.14(5), C(1)–O(1)–Li(1) 141.2(5), C(2)–O(2)–Li(1)
140.7(2), O(1)–C(1)–As(1) 123.5(5), O(2)–C(2)–As(1) 123.5(5).
121.22(18), O(2)–C(2)–As(1)
120.33(17), O(4)–C(16)–As(2)
122.00(16), O(6)–C(30)–As(3) 119.39(17).
123.12(17),
121.03(17),
O(3)–C(15)–As(2)
O(5)–C(29)–As(3)
New J. Chem., 2003, 27, 466–474
471

View Article Online
(see Table 1). The compound is isostructural and isomorphous
with its tris(diacylphosphino)phosphine analogue, P[P{C(O)-
Ph}2]3 ,7 and as such contains a molecule of diethyl ether of
crystallisation in its asymmetric unit. The structure is also clo-
sely related to that of 14 in that it has a central pyramidal
phosphorus atom [ave. As–P–As 106.7ꢁ, cf. 108.39(11)ꢁ in
the residual 1H or 13C resonances of the solvent used (1H
and 13C NMR), external 85% H3PO4 , 0.0 ppm (31P NMR),
or 1 M LiNO3 , 0.0 ppm (7Li NMR). A 13C NMR spectrum
of 5 could not be obtained due to its very low solubility in nor-
mal deuterated solvents at ambient temperature. Mass spectra
were recorded using a VG Fisons Platform II instrument under
APCI conditions. Melting points were determined in sealed
glass capillaries under argon, and are uncorrected. Microana-
lyses were obtained from the Warwick Microanalytical Ser-
vice. Where reproducible microanalyses were not obtained
the sample was either extremely air sensitive and/or variable
amounts of solvent of crystallisation were present. However,
the NMR spectra of these samples suggested their purity was
greater than 95%. The starting materials 2,6 8,5,6 Li{As-
(SiMe3)2}17 and Li{Sb(SiMe3)2}18 were prepared by literature
procedures. All other reagents were used as received.
˚
14] with As–P bonds (2.304 A ave.) almost equivalent to those
˚
in 14 [2.3054(19) A] but slightly shorter than the mean for all
9
˚
crystallographically determined As–P interactions (2.33 A).
The three diacylarsenido ligands are not planar and contain
localised As–C single and C–O double bonds as was the case
in 13. Another similarity with that compound is the acuteness
of the angles about each arsenic centre [As(1) 96.0ꢁ ave., As(2)
96.3ꢁ ave., As(3) 95.1ꢁ ave.], which again are compatible with a
significant degree of p-character in the bonds involving arsenic
centres.
Syntheses
Group 16
[{Li[g2-O,O–OC(Ad)SbC(Ad)O](OEt2)}2], 5. A solution of
adamantoyl chloride (0.66 g, 3.3 mmol) in DME (20 ml) was
added over 20 min to a solution of [Li{Sb(SiMe3)2}(DME)]
(1.22 g, 3.3 mmol) in DME (20 ml) at ꢀ50 ꢁC. The resulting
brown solution was warmed to ambient temperature and stir-
red overnight in the absence of light. Volatiles were removed
in vacuo and the residue washed with hexane (40 ml) to leave
an orange powder, which was recrystallised from diethyl ether
(10 ml) at ꢀ30 ꢁC to yield 5 as orange blocks (0.56 g, 73%).
There have been no reported complexes of arsa- or stib-
adionates with group 16 elements. In this study a number of
reactions of 2 and 3 with group 16 halides, such as Se2Cl2 ,
SOCl2 , PhSeCl and PhTeCl, were carried out but all led to
the formation of largely inseparable mixtures of decomposi-
tion products. It is worth mentioning that in the 1:1 reaction
t
=
of 2 with PhSeCl one such product, [({PhSeC(Bu ) O}Li-
{Z2-O,O–OC(But)AsC(But)O})2] 16, was isolated in very low
yield ( < 10%) and its preparation could not be reproduced
(Scheme 4). The compound was identified by a poor quality
X-ray crystal structure that is not included here but unambigu-
ously confirmed the molecular connectivity of the compound.
The mechanism of formation of 16 is unknown but presum-
ably involves the selenodiacylarsenide, [PhSeAs{C(O)But}2],
as an unstable intermediate that undergoes As–C bond clea-
vage and rearrangement to generate the selenaester, PhSeC(O)-
But, which displaces DME of solvation from the lithium
arsadionate starting material, 2, to form the observed complex.
Arsenic–carbon bond cleavages in diacylarsenido complexes
have been observed before4 and most likely lead to the decom-
position observed in the other group 16 based reactions men-
tioned above.
1
M.p. 94–96 ꢁC (dec.); NMR: H (400 MHz, C6D6) d 1.12 (t,
3
12H, OCH2CH3 , JHH 8 Hz), 1.66–2.16 (m, 60H, Ad), 3.26
3
(q, 8H, OCH2 , JHH 8 Hz); IR (Nujol, u/cmꢀ1) 1578 s, 1545
s; MS APCI: m/z 135 (Ad+, 100), 147 (AdC+, 5), 164
(AdCO+, 32%).
[{Li[g2-O,O–OC(Ph)AsC(Ph)O](DME)}2], 6. A solution of
benzoyl chloride (0.88 ml, 7.5 mmol) in DME (10 ml) was
added over 20 min to a solution of [Li{As(SiMe3)2}(DME)]
(1.19 g, 3.8 mmol) in DME (20 ml) at ꢀ50 ꢁC. The resulting
orange solution was warmed to ambient temperature and stir-
red overnight in the absence of light. Volatiles were removed
in vacuo and the residue washed with hexane (40 ml) to leave
a red powder, which was recrystallised from DME (10 ml) at
ꢀ30 ꢁC to yield 6 as red rods (1.17 g, 82%). M.p. 106–108 ꢁC
(dec.); NMR: 1H (400 MHz, C6D6) d 2.91 (s, 8H, OCH2),
2.93 (s, 12H, OCH3), 7.42–8.17 (m, 20H, ArH); 13C (101.6
MHz, C6D6) d 58.7 (OCH3), 70.6 (OCH2), 125.8, 128.5,
131.7 (Ar–CH), 146.6 (ipso-C), 241.1 (AsC); 7Li (97.2 MHz,
C6D6) d ꢀ0.42; IR (Nujol, u/cmꢀ1) 1581 s, 1547 s; MS APCI:
m/z 105 (PhCO+, 100%).
Conclusions
In summary, the preparation and structural characterisation of
a variety of group 1, 2, 13 and 15 complexes incorporating 2-
arsa-or 2-stiba-1,3-dionato ligands in either their delocalised
form (Z2-O,O coordination), their localised diacylpnictido
form (Z1-E coordination), or a combination of the two modes
(Z1-E–:Z2-O,O–) has been reported. The study has highlighted
the coordinative versatility of heterodionate ligands and has
shown that they can behave similarly to b-diketonates in their
coordination chemistry but also that significant differences are
possible. The utility of the prepared sodium heterodionates
as transfer reagents in lanthanide complex formation is curr-
ently being explored and will be discussed in a forthcoming
publication.
[{Li[g2-O,O–OC(Ph)AsC(Ph)O](OEt2)}2], 7. A solution of
benzoyl chloride (0.88 ml, 7.5 mmol) in DME (10 ml) was
added over 20 min to a solution of [Li{As(SiMe3)2}(DME)]
(1.19 g, 3.8 mmol) in DME (20 ml) at ꢀ50 ꢁC. The resulting
orange solution was warmed to ambient temperature and stir-
red overnight in the absence of light. Volatiles were removed
in vacuo and the residue washed with hexane (40 ml) to leave
a red powder, which was recrystallised from diethyl ether (15
ml) at ꢀ30 ꢁC to yield 7 as red rods (0.95 g, 70%). M.p. 124–
126 ꢁC (dec.); NMR: 1H (400 MHz, C6D6) d 1.13 (t, 12H,
3
3
Experimental details
OCH2CH3 , JHH 8 Hz), 3.27 (q, 8H, OCH2 , JHH 8 Hz),
7.02–8.22 (m, 20H, ArH); 13C (101.6 MHz, C6D6) d 14.2
(OCH2CH3), 64.5 (OCH2), 125.0, 128.4, 131.2 (Ar–CH),
146.2 (ipso-C), 241.3 (AsC); 7Li (97.2 MHz, C6D6) d ꢀ0.22;
IR (Nujol, u/cmꢀ1) 1595 s, 1579 s; MS APCI: m/z 105
(PhCO+, 100), 77 (Ph+, 53%).
General remarks
All manipulations were carried out using standard Schlenk and
glove box techniques under an atmosphere of high purity
argon or dinitrogen. The solvents diethyl ether, DME and hex-
ane were distilled over either potassium or Na/K alloy then
freeze/thaw degassed prior to use.1H, 31P and 7Li NMR spec-
tra were recorded on either Bruker DPX400 or Jeol Eclipse 300
spectrometers in deuterated solvents and were referenced to
[{Na[g2-O,O–OC(Ph)AsC(Ph)O](DME)}2], 9. A solution of
Na{N(SiMe3)2} (0.35 g, 1.9 mmol) in DME (20 ml) was added
over 20 min to a solution of the diacylarsane 8, R ¼ Ph,
472
New J. Chem., 2003, 27, 466–474

View Article Online
E ¼ As, (0.59 g, 2.0 mmol) in DME (20 ml) at ꢀ50 ꢁC. The
resulting orange solution was warmed to ambient temperature
and stirred overnight in the absence of light. Volatiles were
removed in vacuo and the residue washed with hexane (40
ml) to leave a red powder, which was recrystallised from
DME (5 ml) at ꢀ30 ꢁC to yield 9 as red rods (0.51 g, 67%).
M.p. 118–120 ꢁC (dec.); NMR: 1H (400 MHz, C6D6) d 2.73
(s, 12H, OCH3), 2.99 (s, 8H, OCH2), 7.55–8.33 (m, 20H,
ArH); 13C (101.6 MHz, C6D6) d 59.2 (OCH3), 71.2 (OCH2),
126.2, 127.4, 131.8 (Ar–CH), 148.0 (ipso-C), 238.1 (AsC); IR
(Nujol, u/cmꢀ1) 1642 s, 1598 s; MS APCI: m/z 286.8
[PhC(O)As+, 100], 105 (PhCO+, 65%).
(30 ml), extracted with diethyl ether (50 ml) and filtered; the
filtrate was placed at ꢀ30 ꢁC overnight, yielding 13 as colour-
less crystals (yield 0.18 g, 56% based on proposed InCl dispro-
portionation). M.p. 80 ꢁC (decomp.); NMR: 1H (400 MHz,
C6D6) d 1.10 (s, 18H. But), 1.13 [br, 6H, CH3 (Et2O)] 2.95
[s, 6H, CH3 (DME)], 3.18 [s, 4H, CH2 (DME)], 3.28 [br, 4H,
CH2 (Et2O)]); 13C (101.6 MHz, C6D6) d 15.0 [CH3 (Et2O)],
25.7 [C(CH)3], 52.3 (CMe3), 59.1 [CH3 (DME)], 65.8 [CH2
(Et2O)], 69.8 [CH2 (DME)], 244.9 (AsC); IR (Nujol, u/cmꢀ1
)
AsCO 1693 s, 1675 s; MS APCI: m/z 563 (M+ ꢀ Et2O, 10%).
P[As{C(O)But}2]3 , 14. To a solution of 2 (0.840 g, 2.82
mmol) in DME (50 ml) was added PCl3 (0.082 ml, 0.94 mmol)
at ꢀ78 ꢁC whilst stirring. The resulting yellow solution was
stirred at room temperature for 24 h after which time it had
turned orange. Volatiles were removed in vacuo leaving a yel-
low-orange residue, which was extracted with hexane (3 ꢃ 30
ml). The extract was filtered and place at ꢀ30 ꢁC overnight,
yielding yellow crystals of 14 (0.150 g, 21%). M.p. 184–
186 ꢁC (decomp). NMR: 1H (400 MHz, C6D6) d 1.29 (s,
54H, But); 13C (101.6 MHz, C6D6) d 25.4 [C(CH3], 49.4
[C(CH3)], 225.2 (CO); 31P{1H} (121.7 MHz, C6D6) d ꢀ78.9
(s); IR (Nujol, u/cmꢀ1) 1708 s, 1660 s, 1475 m, 1363 m; MS
APCI: m/z 767.0 (M+, 100); anal. found: C 46.67%, H
7.11%, calcd. for C30H54O6PAs3 : C 47.0%, H 7.05%.
[Na{g2-O,O–OC(Mes*)AsC(Mes*)O}(DME)2], 10. A solu-
tion of Na{N(SiMe3)2} (0.055 g, 0.3 mmol) in DME (20 ml)
was added over 20 min to a solution of the diacylarsane 8,
R ¼ Mes*, E ¼ As, (0.20 g, 0.3 mmol) in DME (20 ml) at
25 ꢁC. The resulting solution was heated at reflux for 3 h after
which time the yellow solution was concentrated in vacuo to 5
ml and placed at ꢀ30 ꢁC overnight to yield 10 as yellow cubic
1
crystals (0.18 g, 71%). M.p. 240–250 ꢁC (dec.); NMR: H (400
MHz, C6D6) d 1.42 (s, 18H, p-But), 1.93 (s, 36H, o-But), 3.18
(s, 8H, OCH2), 3.21 (s, 12H, OCH2CH3), 7.63 (s, 4H, ArH);
13C (101.6 MHz, C6D6) d 30.2, 33.4 (CCH3), 33.4, 37.2
(CCH3), 57.4 (OCH2CH3), 70.0 (OCH2), 121.6 (Ar–CH),
143.7, 146.4, 147.3, (quat. Ar–C), 202.1 (AsC); IR (Nujol, u/
cmꢀ1) 1599 s, 1568 s; MS APCI: m/z 286.8 [Mes*C(O)+,
100%].
P[As{C(O)Ph}2]3 , 15. To a solution of 6 (0.226 g, 0.59
mmol) in DME (50 ml) was added PCl3 (0.017 ml, 0.197 mmol)
at ꢀ78 ꢁC whilst stirring. The resulting yellow solution was
stirred at room temperature for 24 h after which it had turned
orange. Volatiles were removed in vacuo leaving a yellow-
orange residue, which was washed with hexane and extracted
with diethyl ether (3 ꢃ 20 ml). The extract was filtered and
placed at ꢀ30 ꢁC overnight to yield 15 as yellow crystals
[Na{g2-O,O–OC(Mes*)SbC(Mes*)O}(DME)2], 11. A solu-
tion of Na{N(SiMe3)2} (0.10 g, 0.56 mmol) in DME (20 ml)
was added over 20 min to a solution of the diacylstibane 8,
R ¼ Mes*, E ¼ Sb, (0.37 g, 0.56 mmol) in DME (20 ml) at
25 ꢁC. The resulting solution was warmed to room temperature
over 4 h over which time it became a bright pink colour. The
volatiles were remove in vacuo and the resulting pink powder
was extracted with DME (10 ml). This solution was concen-
trated in vacuo to 5 ml and placed at ꢀ30 ꢁC overnight to yield
11 as purple cubic crystals (0.33 g, 68%). M.p. 86–88 ꢁC (dec.);
NMR: 1H (400 MHz, C6D6) d 1.20 (s, 18H, p-But), 1.72 (s,
36H, o-But), 3.04 (s, 8H, OCH2), 3.09 (s, 12H, OCH2CH3),
7.41 (s, 4H, ArH); 13C (101.6 MHz, C6D6) d 30.2, 31.6
(CCH3), 34.0, 37.2 (CCH3), 57.5 (OCH2CH3), 70.0 (OCH2),
121.9 (Ar-CH), 141.5, 146.2, 147.4, (quat. Ar–C), 210.3
(AsC); IR (Nujol, u/cmꢀ1) 1598 s, 1572 s; MS APCI: m/z
286.8 [Mes*C(O)+, 100%]; anal. found C 62.87, H 8.88%, calcd
for C46H78NaO6Sb: C 63.37, H 9.02%.
1
(0.03 g, 18%). M.p. 130–132 ꢁC (dec.). NMR: H (400 MHz,
C6D6) d 6.65–6.75 (m, 12H, m-ArH), 6.85 (t, 6H, p-ArH,
3
3JHH ¼ 8 Hz), 7.85 (d, 12H, o-ArH, JHH ¼ 8 Hz); 13C
(101.6 MHz, C6D6) d 128.9 (m-Ar), 129.3 (o-Ar), 133.7 (p-
Ar), 140.8 (ipso-Ar), 209.9 (AsC); 31P{1H} (121.7 MHz,
C6D6) d ꢀ73.4; IR (Nujol, u/cmꢀ1) 1654.6 s, 1629.0 s, 1460
m, 1444 m; MS APCI: m/z 317.1 [PAs(C(O)Ph)2+, 100],
285.9 [As(C(O)Ph)2 , 45%]; Anal. found: C 55.45%, H 3.41%,
calcd. for C42H30O6PAs3 : C 56.88%, H 3.39%.
t
2
t
t
=
[({PhSeC(Bu ) O}Li{g -O,O–OC(Bu )AsC(Bu )O})2], 16. A
solution of 2 (0.516 g, 1.35 mmol) in DME (30 ml) was added
to a solution of PhSeCl (0.259 g, 1.35 mmol) in DME (20 ml)
at ꢀ78 ꢁC. The resulting yellow solution was allowed to warm
to room temperature and stirred for 24 h. Volatiles were
removed in vacuo leaving an orange oil, which was extracted
with hexane (30 ml), filtered and the filtrate placed at ꢀ30 ꢁC
overnight yielding 16 as yellow-orange crystals (0.05 g, 8%).
[Mg{g2-O,O–OC(Ph)AsC(Ph)O}2(DME)], 12. A solution of
MgBu2 (0.25 mmol) in DME (20 ml) was added over 20 min to
a solution of the diacylarsane 8, R ¼ Ph, E ¼ As, (0.15 g, 0.5
mmol) in DME (20 ml) at ꢀ50 ꢁC. The resulting orange solu-
tion was warmed to ambient temperature and stirred overnight
in the absence of light. Volatiles were removed in vacuo and the
residue washed with hexane (40 ml) to leave a red powder,
which was recrystallised from DME (5 ml) at ꢀ30 ꢁC to yield
12 as red rods (0.11 g, 67%). M.p. 133–137 ꢁC (dec.); NMR:
1H (400 MHz, C6D6) d 2.90 (s, 6H, OCH3CH3), 3.04 (s, 4H,
OCH2), 7.21–8.33 (m, 20H, ArH); 13C (101.6 MHz, C6D6) d
58.0 (OCH3), 69.5 (OCH2), 124.5, 130.9, 132.6 (Ar–CH),
141.0 (ipso-C), 201.1 (AsC); IR (Nujol, u/cmꢀ1) 1548 s, 1524
s; MS APCI: m/z 77 (Ph+, 100), 105 (PhCO+, 50%).
1
M.p. 83–85 ꢁC (dec.). NMR: H (400 MHz, C6D6) d 1.04 (s,
18H, But), 1.30 (s, 36H, But), 7.04–7.49 (m, 8H, ArH); 13C
(100.6 MHz, C6D6)
d 26.7, 28.3 [C(CH3)3], 48.9, 49.6
(CMe3), 126.5 (m-Ar), 128.7 (o-Ar), 129.2 (p-Ar), 136.4 (ipso-
Ar), 208.1 (CO), 264.9 (AsCO); IR (Nujol, u/cmꢀ1) 1701 s,
1635 s, 1568 s, 1460 m, 1378 m; MS APCI: m/z 227
[PhSeC(H)But+, 20%].
Structure determinations
Crystals of 5, 6, 7, 9, 10, 11, 12, 13 and 15 suitable for X-ray
structure determination were mounted in silicone oil. Crystal-
lographic measurements were made using either Nonius
Kappa CCD or Enraf-Nonius CAD4 diffractometers. The
structures were solved by direct methods and refined on F2
by full-matrix least-squares methods (SHELXL9719 or
SHELXL9320) using all unique data. Crystal data, details of
data collections and refinements are given in Table 1. The
[LiCl(Et2O)(DME){g2-O,O–Cl3InAs[C(O)R]2}], 13. A solu-
tion of 2 (0.452 g, 1.52 mmol) in DME (30 ml) was added
dropwise to a suspension of InCl (0.229 g, 1.52 mmol) in
DME–Et2O (50:50, 25 ml) at ꢀ78 ꢁC. The resulting pale yellow
solution was allowed to warm to room temperature and stirred
for 24 h. Volatiles were removed in vacuo leaving indium
deposits and a white powder, which was washed with hexane
New J. Chem., 2003, 27, 466–474
473

View Article Online
5
6
C. Jones, J. W. Steed and R. C. Thomas, J. Chem. Soc., Dalton
Trans., 1999, 1541.
J. Durkin, D. E. Hibbs, P. B. Hitchcock, M. B. Hursthouse, C.
Jones, J. Jones, K. M. A. Malik, J. F. Nixon and G. Parry,
J. Chem. Soc., Dalton Trans., 1996, 3277.
C. Jones, P. C. Junk and T. C. Williams, J. Chem. Soc., Dalton
Trans., 2002, 2417.
G. Becker, M. Birkhahn, W. Massa and W. Uhl, Angew. Chem.,
Int. Ed. Engl., 1980, 19, 741.
relatively high R factor for the structural determination of 11
results from the poor quality of the crystal used in the experi-
ment. The assignment of the chiral space group C2 to the crys-
tal structure of this compound is unambiguous based on
systematic absences, in addition to the facts that the molecule
is chiral, crystallises as its pure enantiomers [Flack parameter
0.01(6)] and sits on a 2-fold axis. The molecular structures of
the complexes are depicted in Figs. 1–9, showing ellipsoids at
the 30% probability level. CCDC reference numbers 195258–
66. See http://www.rsc.org/suppdata/nj/b2/b210043f/ for
crystallographic files in CIF or electronic format.
7
8
9
As determined from a survey of the Cambridge Crystallographic
Database.
10 O. Mundt, G. Becker, M. Rossler and C. Witthauer, Z. Anorg.
Allg. Chem., 1983, 506, 42.
11 The Chemistry of Metal CVD, eds. T. T. Kodas and M. J.
Hampden-Smith, VCH, Weinheim, 1994.
12 T. Hatanpa¨a¨, J. Kansikas, I. Mutikainen and M. Leskela¨, Inorg.
Chem., 2001, 40, 788.
Acknowledgements
13 M. L. H. Green, P. Mountford, G. J. Smout and S. R. Peel, Poly-
hedron, 1990, 9, 2763.
14 R. J. Baker, R. D. Farley, C. Jones, M. Kloth and D. M. Murphy,
J. Chem. Soc., Dalton Trans., 2002, 3844.
We gratefully acknowledge financial support from EPSRC
(studentships for RCT and TCW).
15 N. J. Hardman, B. E. Eichler and P. P. Power, Chem. Commun.,
2000, 1991.
16 R. J. Baker, R. D. Farley, C. Jones, M. Kloth and D. M. Murphy,
Chem. Commun., 2002, 1196.
17 G. Becker and C. Witthauer, Z. Anorg. Allg. Chem., 1982, 492, 28.
References
1
2
3
4
C. Jones, Coord. Chem. Rev., 2001, 215, 151 and references
therein.
C. Jones, P. C. Junk, J. W. Steed, R. C. Thomas and T. C.
Williams, J. Chem. Soc., Dalton Trans., 2001, 3219.
C. Jones, S. J. Black and J. W. Steed, Organometallics, 1998, 17,
5924.
18 G. Becker, A. Munch and C. Witthauer, Z. Anorg. Allg. Chem.,
1982, 492, 15.
¨
19 G. M. Sheldrick, SHELXL-97, University of Go¨ttingen,
Go¨ttingen, Germany, 1997.
20 G. M. Sheldrick, SHELXL-93, University of Go¨ttingen,
Go¨ttingen, Germany, 1993.
S. J. Black, D. E. Hibbs, M. B. Hursthouse, C. Jones and J. W.
Steed, Chem. Commun., 1998, 2199.
474
New J. Chem., 2003, 27, 466–474