Phosphine and phosphido indium hydride complexes and their use in inorganic synthesis

Article abstract of DOI:10.1039/a908418e

Reaction of PR₃, R = cyclohexyl (Cy), cyclopentyl (Cy^p) or phenyl, with [lnH₃(NMe₃)] affords the 1:1 indium trihydride complexes, [InH₃(PR₃)]. The stabilities and spectroscopic properties of these complexes are described in terms of the phosphine ligands' steric bulk and nucleophilicity. Reaction of two equivalents of PCy₃ with [InH₃(NMe₃)] yields the complex [InH₃(PCy₃)₂] which has been characterised by X-ray crystallography. The first phosphido-indium hydride complex, [{InH₂(PCy₂)}₃], has been prepared by a novel synthetic route which involves treatment of [InH₃(NMe₃)] with LiPCy₂, Its crystal structure shows it to exist as a cyclic trimer in the solid state. The complex, [InH₃(PCy₃)] has been used to prepare a range of monomeric indium chalcogenolato complexes, [In(EPh)₃(PCy₃)], E = S, Se or Te, all of which have been structurally characterised. ' The Royal Society of Chemistry 2000.

Full text of DOI:10.1039/a908418e

View Article Online / Journal Homepage / Table of Contents for this issue
Phosphine and phosphido indium hydride complexes and their use
in inorganic synthesis
Marcus L. Cole, David E. Hibbs, Cameron Jones* and Neil A. Smithies
Department of Chemistry, University of Wales, Cardiff, PO Box 912, Park Place, Cardiff,
UK CF10 3TB
Received 21st October 1999, Accepted 22nd December 1999
Reaction of PR3, R = cyclohexyl (Cy), cyclopentyl (Cyp) or phenyl, with [InH3(NMe3)] affords the 1:1 indium
trihydride complexes, [InH3(PR3)]. The stabilities and spectroscopic properties of these complexes are described
in terms of the phosphine ligands’ steric bulk and nucleophilicity. Reaction of two equivalents of PCy3 with
[InH3(NMe3)] yields the complex [InH3(PCy3)2] which has been characterised by X-ray crystallography. The first
phosphido–indium hydride complex, [{InH2(PCy2)}3], has been prepared by a novel synthetic route which involves
treatment of [InH3(NMe3)] with LiPCy2. Its crystal structure shows it to exist as a cyclic trimer in the solid state.
The complex, [InH3(PCy3)] has been used to prepare a range of monomeric indium chalcogenolato complexes,
[In(EPh)3(PCy3)], E = S, Se or Te, all of which have been structurally characterised.
Introduction
The chemistry of both aluminium and gallium hydride com-
plexes is now well established.1 Over the years such compounds
have found a remarkable variety of applications in areas includ-
ing organic synthesis,2 inorganic synthesis3 and materials chem-
istry.4 It would perhaps be thought that the chemistry of indium
Scheme 1 i, PR3, Et2O.
hydride complexes would now be equally developed, especially
reaction solution and was readily recrystallised from DME in
considering the potential applications of these compounds.
This is, however, not the case and indeed after Wiberg’s original
synthesis of LiInH4 in 1957,5 and before 1998, only five struc-
turally characterised compounds containing In–H bonds had
been described, viz. [Li(thf)2][{(Me3Si)3C}2In2H5],6 K[H{In-
(CH2CMe3)3}2],7 K3[K(Me2SiO)7][InH(CH2CMe3)3]4,8 [InH-
{2-Me2NCH2(C6H4)}2]9 and [Me2InB3H8].10 We have since
reported the first indium trihydride complexes, [InH3{CN(Pri)-
C2Me2N(Pri)}] 111 and [InH3(PCy3)] 2,12 which are stabilised by
either a highly nucleophilic “Arduengo” carbene or a bulky
tertiary phosphine ligand. In addition, the ionic indium hydride
complex, [(Me3In)2H][Li(tmen)2], has been reported by us.13
Compound 2 possesses remarkable stability in that it does
not decompose in the solid state at temperatures below 50 ЊC
and is stable in air for days. This unexpected stability has
prompted us to begin a systematic study aimed at the prepar-
ation of a range of thermally stable indium hydride complexes
and comparing their chemistry to those of their aluminium and
gallium counterparts. Moreover, the ease of handling of 2
should lend it to an array of applications similar to those that
have already been found for aluminium and gallium trihydride
complexes. The work described herein addresses this potential
with the preparation of several monomeric indium chalcogenol-
ato complexes. In addition, the synthesis and characterisation
of several phosphine–InH3 adducts related to 2, and the first
phosphido–indium hydride complex are discussed.
moderate yield (39%). The reaction involving PPh3 also lead to
a precipitate forming which could be isolated at Ϫ20 ЊC but all
attempts to recrystallise this crude product led to its decom-
position to indium metal and PPh3 with the generation of a gas.
It is assumed that this precipitate is [InH3(PPh3)] 5 on the basis
of infrared data (see below). After treating [InH3(NMe3)] with
either PBut3, PMes3 or PEt3 at Ϫ30 ЊC no precipitate formed
and so volatiles were removed from the reaction mixture in
vacuo which resulted in decomposition, even at Ϫ30 ЊC. The
reaction precursor, 3, cannot be isolated in the solid state and
shows a similar instability in solution, and thus it is difficult to
be sure if the three adducts 6–8 have been formed or if no
reaction has occurred with these phosphines. Alternate routes
to [InH3(PR3)] were examined whereby LiInH4 was treated with
either PR3 or PR3ؒHCl at Ϫ20 ЊC in the expectation that LiH or
LiCl elimination would yield the target phosphine–indane
adducts. Similar routes have been successful in the formation of
related alane or gallane adducts, [MH3(PCy3)] M = Al,14 Ga,15
though in the present cases no reactions occurred. The most
likely reason for this lies with the inability to carry out these
reactions above Ϫ10 ЊC due to the thermal instability of
LiInH4.
At the outset of this study it was thought that the thermal
stability of any formed complexes would largely depend upon
two phosphine characteristics, their basicity and their steric
bulk. The more basic the phosphine, the stronger the bond
between it and the InH3 unit should be. In the case of Lewis
base adducts of AlH3 and GaH3 their primary decomposition
process is Lewis base loss1 and it would seem that this should
also be the case for InH3 adducts. Therefore, the stronger the
phosphine donor, the more stable should be the formed indane
complex. In terms of steric protection Downs and Pulham16
have suggested that the thermal instability of the binary heavier
Group 13 hydrides may not be solely thermodynamic in
nature but may have a kinetic aspect whereby decomposition
Results and discussion
In a preliminary communication12 we have described the syn-
thesis and characterisation of the indium trihydride complex, 2;
full synthetic details of which are included herein. In order
to examine the generality of formation of phosphine–InH3
adducts a series of other tertiary phosphines were reacted with
[InH3(NMe3)] 3 as outlined in Scheme 1. When PCyp , Cyp =
3
cyclopentyl, was employed the product, 4, precipitated from the
DOI 10.1039/a908418e
J. Chem. Soc., Dalton Trans., 2000, 545–550
This journal is © The Royal Society of Chemistry 2000
545

View Article Online
may occur through intermediates containing M–H–M bridges.
This could also be the case for complexes of the type
[InH3(PR3)], especially considering indium’s tendency to form
five- or six-coordinate complexes.17 Therefore it was hypoth-
esised that the greater the steric bulk of the phosphine ligand,
the greater the thermal stability of the resulting complex should
be, as the formation of intermolecular hydride bridges would be
circumvented.
This seemed to be borne out by the stability of 2 which has
been attributed to the 170Њ cone angle of the phosphine ligand.
However, its stability is also likely to result from the relatively
strong σ-donor characteristics of the ligand. There are some
surprises in the stabilities of the other complexes, 4–8, if these
are correlated to ligand cone angles [PMes3 (212Њ) > PBut3
(182Њ) > PCy3 (170Њ) > PPh3 (145Њ) > PEt3 (132Њ)]18 and ligand
basicities [PBut3 > PCy3 ≈ PEt3 > PMes3 > PPh3].19 Although
no cone angle or basicity data could be obtained for PCyp it
3
Fig. 1 Molecular structure of compound 9. Selected bond lengths (Å)
would seem likely that these are slightly less than for PCy3
which is in line with the lower thermal stability of 4 (decomp.
20 ЊC) relative to 2 and the fact that it is slightly air sensitive.
Interestingly, PBut3 is both bulkier and more nucleophilic than
PCy3 but no stable indane complex incorporating it could be
formed. It is unlikely that the ligand is too bulky to form a
complex as the corresponding gallane complex, [GaH3(PBut3)],
is known,20 though it is less stable than [GaH3(PCy3)]. The
PMes3 ligand is a weaker σ-donor but is very bulky and in this
case is perhaps too much so to displace NMe3 from 3 to form 7.
No stable complex could be formed with PEt3 which is a good
σ-donor, but has a relatively small cone angle. Finally, PPh3
has only a moderate cone angle and is a weak σ-donor, yet
forms a complex, 5, which can be isolated in the solid state
and is moderately thermally stable for an indane complex
(decomp. ca. 0 ЊC). It is probable that PPh3 forms an isolable
indane complex when ligands such as PBut3 do not because
the complex, 5, precipitates from the reaction mixture and is
considerably more stable in the solid state than in solution.
Indeed, when dissolution of 5 in thf is attempted at Ϫ70 ЊC
decomposition rapidly occurs.
and angles (Њ): In–P(1) 2.9869(5), P(1)–C(13) 1.8548(14), P(1)–C(7)
1.8567(14), P(1)–C(1) 1.8674(14); P(1)–In(1)–P(1Ј) 180.000(17).
Five-coordinate bis(phosphine) adducts of alane and gallane
are unknown though alane does form polymeric five-coordinate
complexes of the type [{AlH3(R2PCH2CH2PR2)}∞], R = alkyl,
with diphosphine ligands.14 As indium has a larger covalent
radius (1.55 Å) than aluminium or gallium (1.25 Å) it was
thought that it might be possible to form bis(phosphine)
adducts of indane. This was attempted by reacting either two
equivalents of PCy3 with in-situ generated [InH3(NMe3)] or an
equimolar amount of PCy3 or PCyp with 2 or 4 respectively
3
(Scheme 2). In each case the anticipated product, 9 or 10,
The infrared spectra of solid samples of the three isolated
phosphine–indane complexes show characteristic strong, broad
In–H stretching absorptions at 1661 cmϪ1 2, 1649 cmϪ1 4 and
1681 cmϪ1 5. Those for 2 and 4 are close to each other and to
that of the carbene–InH3 adduct 1 (1640 cmϪ1).11 All are at
considerably lower wavenumbers than either their alane or
gallane analogues, a fact which reflects the relative weakness
of the In–H bond. It is noteworthy that the In–H stretch for 5
is at a significantly higher frequency than in the other indane
adducts which is probably a result of the weak σ-donating
properties of PPh3.
Scheme 2 i, PCy3, Et2O, L = NMe3; ii, PR3, DME, L = PR3.
formed in moderate yield and could be recrystallised from
DME. An attempt was also made to prepare [InH3(PPh3)2] by
reaction of two equivalents of PPh3 with 3 at Ϫ30 ЊC, however
this led to the precipitation of 5 from solution in the early stages
of reaction and the second equivalent of PPh3 remained
unreacted.
Complex 9 is moderately stable at room temperature in the
solid state and does not decompose until 37 ЊC to indium metal
and the free phosphine. It is surprising that its decomposition
temperature is lower than that of 2 as it would be expected that
the primary decomposition process would be phosphine loss to
form 2 which is stable to 50 ЊC. Indeed, compound 9, unlike 2,
does fully decompose over several days at room temperature.
Not surprisingly, the thermal stability of 10 is slightly lower
than 9 and it decomposes at 15 ЊC.
The infrared spectra (Nujol mulls) of 9 and 10 display strong,
broad In–H stretches at 1666 and 1644 cmϪ1 respectively which
are close to those of their 1:1 analogues. Both complexes are
not stable in solution and will decompose at 25 ЊC in a matter
of minutes. However, at low temperature their 1H NMR spectra
exhibit broad resonances (δ 5.81 9 and 5.23 10) which corre-
spond to the hydride ligands in each complex. These are shifted
slightly downfield with respect to 2 and 4. Their 31P{1H} NMR
spectra display singlet resonances close to those of the free
phosphine ligands.
The crystal structure of 9 is depicted in Fig. 1. The indium
atom sits on a centre of inversion and consequently the hydride
ligands attached to it are necessarily disordered and thus could
not be located from difference maps. It is clear, however, that
the indium centre has a trigonal bipyramidal coordination
Although 2 is stable in the solid state for weeks at room
temperature, in benzene solutions it will decompose giving
indium metal, PCy3 and a gas at 25 ЊC over 1 hour. Compound
4 is less stable in solution and completely decomposes over 15
minutes at room temperature. Because of the thermal instability
of 5 in solution NMR data on only 2 and 4 could be obtained.
1
Those for 2 have been reported12 but most importantly its H
NMR spectrum shows a broad hydride resonance at δ 5.61
which is downfield of the related resonances in the spectra of
its alane, [AlH3(PCy3)], and gallane, [GaH3(PCy3)], counter-
parts, viz. δ 4.3214 and 4.2515 respectively. A similar pattern
was observed for the carbene–metal trihydride complexes,
[MH3{CN(Pri)C2Me2N(Pri)}] M = Al (δ 4.63), Ga (δ 4.48), In
(δ 5.58).11 Not surprisingly the hydride resonance of 4 (δ 5.14)
occurs in a similar position to 2. The signal integrates for 3
hydrogens and is very broad as a result of the influence of
the quadrupolar indium centre. Presumably this is also the
reason why no hydride–phosphorus coupling is observed in its
1H NMR spectrum. The 31P{1H} NMR spectrum of 4 is
unremarkable and exhibits a singlet (δ 3.45) close to that of the
free ligand (δ 1.38).
546
J. Chem. Soc., Dalton Trans., 2000, 545–550

View Article Online
environment with the two phosphine ligands in axial positions
and the hydride ligands in equatorial positions. The phosphine
ligands are staggered and, interestingly, the In–P bond lengths
[2.9869(5) Å] are almost 0.34 Å longer than in 2 [2.6474(6) Å].
Indeed, the lengths of these bonds are second only to the In–P
bonds (3.054 Å) in [{InI3(PPh3)}{InI3(PPh3)2}].21 It may be
thought that the very long In–P bond lengths in 9 are partly due
to steric buttressing of the two bulky phosphine ligands but this
is probably not the major reason. We have recently carried out
high level ab initio calculations (DFT/B3-LYP) on [InH3(PMe3)]
and [InH3(PMe3)2] which show similar differences between their
In–P bond lengths (2.755 Å and 3.038 Å respectively)22 as com-
pared to those of 2 and 9. However, in the model compounds
the ligands are not bulky and so there is most likely more of an
electronic rather than a steric basis to the lengths of the In–P
bonds in 9.
Considering the stability of 2 and 4 we thought it might be
possible to prepare phosphido–indium hydride complexes of
the type [{InH2(PR2)}n], R = alkyl, which could also be stable at
room temperature if bulky alkyl groups were employed. Com-
pounds of this type are worthy synthetic targets as related
Group 15–Group 13 hydride complexes have been used as fixed
stoichiometry single source precursors to III/V semiconducting
materials.23 It was hoped that if such compounds could be pre-
pared they would cleanly thermally decompose to InP via the
loss of alkane.
The dicyclohexylphosphide ligand was chosen for this study
because of its steric bulk which has previously been used to
stabilise the gallium hydride complex, [{GaH2(PCy2)}3],24 to
temperatures in excess of 150 ЊC. Initial attempts to form an
analogous indium hydride complex by reaction of LiInH4 with
a 1:1 mixture of PCy2H and anhydrous HCl, or PCy2H with 3
both led to decomposition. An alternative strategy was devised
whereby LiPCy2 was reacted with one equivalent of 3. This
proved successful and resulted in the formation of [{InH2-
(PCy2)}3] 11 in moderate yield (43%) via LiH elimination
(Scheme 3). Compound 11 is readily recrystallised from
Fig. 2 Molecular structure of compound 11. Selected bond lengths
(Å) and angles (Њ): P(1)–In(1) 2.5818(8), P(1)–In(2) 2.5985(8), P(2)–
In(2) 2.5778(8), P(2)–In(3) 2.5862(9), P(3)–In(1) 2.5666(8), P(3)–In(3)
2.6044(8), In(1)–H(1) 1.77(5), In(1)–H(2) 1.70(4), In(2)–H(3) 1.63(4),
In(2)–H(4) 1.74(5), In(3)–H(5) 1.79(5), In(3)–H(6) 1.64(5); P(3)–In(1)–
P(1) 103.95(2), P(3)–In(1)–H(1) 108.2(15), P(1)–In(1)–H(1) 107.0(15),
P(3)–In(1)–H(2) 108.4(14), P(1)–In(1)–H(2) 112.8(15), H(1)–In(1)–
H(2) 115.7(21), P(2)–In(2)–P(1) 108.83(3), P(2)–In(2)–H(3) 112.7(13),
P(1)–In(2)–H(3) 101.9(14), P(2)–In(2)–H(4) 110.0(15), P(1)–In(2)–H(4)
107.2(15), H(3)–In(2)–H(4) 115.6(20), P(2)–In(3)–P(3) 107.97(2), P(2)–
In(3)–H(5) 106.0(17), P(3)–In(3)–H(5) 105.4(17), P(2)–In(3)–H(6)
113.1(18), P(3)–In(3)–H(6) 107.2(17), H(5)–In(3)–H(6) 116.6(24),
In(1)–P(1)–In(2) 128.40(3), In(2)–P(2)–In(3) 121.08(3), In(1)–P(3)–
In(3) 119.53(3).
proved unfeasible due to the instability of the compound in
solution.
The molecular structure of 11 (Fig. 2) confirmed that it exists
as a cyclic trimer in the solid state as does its gallium analogue.
The geometry of the In3P3 core is best described as a flattened
or twisted boat conformation which, although unusual, has
been seen in related complexes, e.g. [{GaH2(PCy2)}3]24 and
[{Me2Ga(PPri2)}3].25 The In–P bond lengths are all similar
(2.596 Å ave.) and lie in the normal region for such interactions.
The quality of the X-ray data was sufficient for the hydride
ligands to be located from difference maps and refined iso-
tropically. Although not accurate, the In–H distances (1.71 Å
ave.) are in the expected region for terminal In–H bonds (cf.
1.68 Å ave. in 212).
Lewis base adducts of AlH3 and GaH3 have proved to be
valuable in a range of inorganic syntheses.1 It seemed logical
that this would also be the case for thermally stable InH3
adducts. In the final part of this study we wished to demon-
strate this with the preparation of a series of monomeric tris-
(chalcogenolato)indium complexes. There were a number of
reasons for choosing these systems which include their potential
use as precursors to III/VI materials, e.g. InxSey, that have
useful opto-electronic properties.26 In addition, monomeric
tris-thiolato- and -selenolato-indium complexes are rare, whilst
structurally characterised tris(tellurolato)indium complexes are
unknown.
Scheme 3 i, LiPCy2, Et2O, ϪLiH.
diethyl ether to give colourless crystals which are stable in air
for hours and decompose at 64 ЊC. Although this is a higher
decomposition temperature than for 2 crystals of 11 will
slowly decompose at room temperature over a period of
days. Solutions of 11 are stable at room temperature for only 15
minutes. Unfortunately, in both solution and the solid state this
compound does not decompose to InP but to indium metal and
PCy2H with the evolution of a gas, presumably H2.
As with the phosphine–InH3 adducts, complex 11 exhibits a
strong broad In–H absorption in its infrared spectrum, but at
significantly higher frequency, 1686 cmϪ1. The reason for this
shift is probably an inductive effect of the anionic phosphido
ligand which strengthens its In–H bonds relative to those of 2
and 4. In solution the 31P NMR spectrum of 11 displays a
singlet at δ Ϫ47 (cf. δ Ϫ33 in [{GaH2(PCy2)}3]24). Its 1H NMR
spectrum shows only a single broad hydride resonance at δ 5.88
over a range of temperatures which is in contrast to the hydride
resonance of [{GaH2(PCy2)}3]24 which exists as a broad triplet
at δ 4.68 due to coupling to the two equivalent P-centres
adjacent to each In centre. This difference is not surprising as
P–H coupling through the more quadrupolar indium centre
would be difficult to observe. Although 11 is trimeric in the
solid state it would be interesting to carry out a solution state
molecular weight determination to shed light on the nuclearity
of this complex in solution. Unfortunately such an experiment
A synthetic route that has been shown to be facile for alane
and gallane complexes27 was utilised whereby complex 2 was
reacted with 3/2 equivalents of the diphenyldichalcogenides,
E2Ph2, E = S, Se or Te, which afforded 12–14 in moderate yields
(Scheme 4) via elimination of dihydrogen. All prepared com-
plexes are thermally robust and appear to be stable to the
atmosphere indefinitely.
In solution 12–14 display similar NMR spectra and in the
case of their 31P{1H} NMR spectra singlets were observed at
J. Chem. Soc., Dalton Trans., 2000, 545–550
547

View Article Online
Fig. 5 Molecular structure of compound 14. Selected bond lengths
Fig. 3 Molecular structure of compound 12. Selected bond lengths
(Å) and angles (Њ): P(1)–In(1) 2.6109(8), S(1)–In(1) 2.4456(10), S(2)–
In(1) 2.4544(9), S(3)–In(1) 2.4384(9); S(3)–In(1)–S(1) 119.96(4), S(3)–
In(1)–S(2) 105.79(4), S(1)–In(1)–S(2) 109.48(3), S(3)–In(1)–P(1)
97.85(3), S(1)–In(1)–P(1) 112.56(3), S(2)–In(1)–P(1) 110.48(3), C(1)–
S(1)–In(1) 108.92(12), C(7)–S(2)–In(1) 110.45(12), C(13)–S(3)–In(1)
101.50(12).
(Å) and angles (Њ): In(1)–P(1) 2.642(3), In(1)–Te(1) 2.7301(13), In(1)–
Te(2) 2.7435(13), In(1)–Te(3) 2.7681(13); P(1)–In(1)–Te(1) 118.06(8),
P(1)–In(1)–Te(2) 98.05(8), Te(1)–In(1)–Te(2) 115.66(4), P(1)–In(1)–
Te(3) 105.72(8), Te(1)–In(1)–Te(3) 109.96(4), Te(2)–In(1)–Te(3)
108.31(4), C(19)–Te(1)–In(1) 101.6(3), C(25)–Te(2)–In(1) 96.4(3),
C(31)–Te(3)–In(1) 98.6(3).
complex, [Ga(TePh)3(PCy3)].27 Despite this, the geometries of
all three compounds are similar and the indium centre in each
sits in a heavily distorted tetrahedral environment which pre-
sumably results from the steric congestion within the molecule.
The In–E distances in each are in the normal regions and
average 2.446 Å (12), 2.563 Å (13) and 2.747 Å (14). Finally the
In–P distances in all compounds are similar and close to that in
2 [2.6474(6) Å].12
Conclusions
A series of phosphine–indium trihydride complexes has been
prepared and found to have varying thermal stabilities depend-
ing on the choice of phosphine ligand. The first examples of
2:1 phosphine–InH3 and phosphido–indium hydride com-
plexes have been prepared and structurally characterised. The
utility of phosphine–InH3 complexes in inorganic synthesis has
been partly addressed with the preparation of a series of
indium chalcogenolato complexes. Overall, the study shows
that relatively stable and easily handled indium hydride com-
plexes are accessible and suggests they will find further uses in
synthesis and as materials precursors. These applications will be
the subject of a series of forthcoming papers.
Fig. 4 Molecular structure of compound 13. Selected bond lengths
(Å) and angles (Њ): In(1)–P(1) 2.6145(11), In(1)–Se(1) 2.5760(7), In(1)–
Se(2) 2.5537(8), In(1)–Se(3) 2.5604(5); Se(2)–In(1)–Se(3) 116.76(3),
Se(2)–In(1)–Se(1) 110.736(19), Se(3)–In(1)–Se(1) 106.60(3), Se(2)–
In(1)–P(1) 117.28(4), Se(3)–In(1)–P(1) 98.66(4), Se(1)–In(1)–P(1)
105.47(3), C(1)–Se(1)–In(1) 101.50(11), C(7)–Se(2)–In(1) 105.21(10),
C(13)–Se(3)–In(1) 97.49(11).
Experimental
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, hexane, toluene and
DME were distilled over either potassium or Na/K alloy then
freeze/thaw degassed prior to use. 1H, 13C and 31P NMR spectra
were recorded on a JEOL FX90, Bruker AMX360 or Bruker
DPX400 spectrometer in C6D6 or C7D8 and were referenced
Scheme 4 i, 3/2 E2Ph2, DME, ϪH2.
1
δ 18.61, 10.01 and Ϫ7.10 respectively (cf. δ 7.43 for 212). It is
noteworthy that the related signal for the gallium analogue of
14, viz. [Ga(TePh)3(PCy3)],27 has been reported to occur at
significantly lower field (δ 68.6) though no explanation for this
difference can be offered here.
The molecular structures of 12–14 (Figs. 3–5) show all these
compounds to be monomeric. Compounds 13 and 14 are iso-
morphous with each other but not with the thiolato complex,
12. Interestingly, this complex is isomorphous to the related
to the residual H or 13C resonances of the solvent used, or to
external 85% H3PO4, δ 0.0 (31P NMR). Mass spectra were
recorded using a VG Fisons Platform II instrument under EI
or APCI conditions. Melting points were determined in sealed
glass capillaries under argon, and are uncorrected. Elemental
analyses were carried out at either the Chemistry Department
at Cardiff or the Warwick Analytical Service. Analyses for 4, 5,
9, 10 and 11 could not be obtained due to the thermal instab-
ility of these compounds. The starting material [InH3(NMe3)]
548
J. Chem. Soc., Dalton Trans., 2000, 545–550

View Article Online
was prepared by a literature procedure.11 All other reagents
were used as received.
[InH3(PCyp )2] 10. PCyp3 (0.23 cm3, 0.96 mmol) was added to
3
a stirred solution of [InH3{PCyp }] (0.35 g, 0.98 mmol) in DME
3
(ca. 40 cm3) at Ϫ50 ЊC. The resulting solution was warmed to
Ϫ30 ЊC and stirred for 2 h whereupon the volatiles were
removed in vacuo to yield a white solid. This was washed with
cold diethyl ether (ca. 5 cm3) and extracted with cold (Ϫ20 ЊC)
DME (ca. 30 cm3), concentrated in vacuo, filtered and the fil-
trate placed at Ϫ30 ЊC to yield 10 as colourless plates (0.12 g,
21%), decomp. ϩ15 ЊC. 1H NMR (400 MHz, C6D5CD3, 300 K):
δ 1.17–1.74 (m, 54H, C5H9), 5.23 (br s, 3H, In–H). 13C NMR
Syntheses
[InH3(PCy3)] 2.12 PCy3 (0.72 g, 2.6 mmol) in diethyl ether (20
cm3) was added to a solution of [InH3(NMe3)] (2.6 mmol) in
diethyl ether (40 cm3) at Ϫ50 ЊC forming a white precipitate.
The reaction mixture was allowed to warm to Ϫ20 ЊC and
stirred for a further 2 h. After this time the solution was filtered
to leave a white solid, which was washed with cold hexane (20
cm3). Extraction with toluene (30 cm3) at Ϫ20 ЊC and cooling
to Ϫ35 ЊC overnight afforded colourless cubic crystals. Yield
1
(100.6 MHz, C6D5CD3, 300 K): δ 26.7 (d, CH, JPC 19.8 Hz),
30.6 (s, CH2), 31.9 (d, CH2, 2JPC 15.0 Hz). 31P NMR (36.3 MHz,
C6D5CD3, 85% H3PO4, 233 K): δ 3.45 (s). MS EI: m/z (%) 238
[P(C5H9)3ϩ, 15], 169 [P(C5H9)2ϩ, 63], 117 [InH2ϩ, 7], 116 [InHϩ,
6], 100 [P(C5H9)ϩ, 100], 69 [C5H9ϩ, 63]. IR (Nujol) ν/cmϪ1: 1644
(s, br, In–H str.).
1
0.71 g, 71%; mp 50 ЊC (decomp.). H NMR (400 MHz, C6D6,
298 K): δ 0.92–1.75 [mult, 33H, C6H11], 5.61 [br s, 3H, In-H].
13C NMR (100.6 MHz, C6D6, 298 K): δ 26.6 [s, CH2], 27.7 [d,
2
1
CH2, JPC 10.2 Hz], 30.5 [s, CH2], 31.8 [d, CH, JPC 13.0 Hz].
31P NMR (36.3 MHz, C6D6, 85% H3PO4, 298 K): δ 7.43. MS
EI: m/z (%) 280 [PCy3ϩ, 63], 197 [PCy2ϩ, 100], 114 [PCyϩ, 78].
IR (Nujol) ν/cmϪ1: 1661 (s, br, In–H str).
[{InH2(PCy2)}3] 11. A solution of [Li{PCy2}(thf)n] (2.50
mmol), prepared in situ from HPCy2 and BunLi, in thf (ca. 20
cm3) was added to a solution of [InH3(NMe3)] (≈2.54 mmol) in
diethyl ether (ca. 70 cm3) at Ϫ70 ЊC. The solution was warmed
to Ϫ30 ЊC and stirred for 2 h whereupon volatiles were removed
in vacuo to yield an oily off-white solid. This was extracted with
cold (Ϫ40 ЊC) toluene (ca. 50 cm3), filtered and the filtrate dried
in vacuo prior to further extraction with cold (Ϫ15 ЊC) diethyl
ether (ca. 30 cm3). Concentration, filtration and placement at
Ϫ30 ЊC yielded 11 as colourless blocks (0.33 g, 43%), decomp.
[InH3(PCyp3)] 4. PCyp (0.65 g, 2.54 mmol) was added to a
3
solution of [InH3(NMe)3] (≈2.54 mmol) in diethyl ether (ca. 70
cm3) at Ϫ70 ЊC. The resulting solution was warmed to Ϫ30 ЊC
and stirred for 2 h whereupon it was cooled to Ϫ70 ЊC to yield a
colourless precipitate. This was washed with cold diethyl ether
(ca. 10 cm3) and dried in vacuo. Subsequent extraction with cold
(Ϫ25 ЊC) DME (ca. 40 cm3), filtration and placement at Ϫ30 ЊC
yielded 4 as colourless needles (0.36 g, 39%), decomp. ϩ20 ЊC.
1H NMR (400 MHz, C6D5CD3, 300 K): δ 1.12–1.56 (m, 27H,
C5H9), 5.14 (br s, 3H, In–H). 13C NMR (100.6 MHz, C6D5CD3,
1
ϩ64 ЊC. H NMR (400 MHz, C6D5CD3, 300 K): δ 1.00–1.94
(m, 66H, C6H11), 5.88 (br s, 6H, In–H). 13C NMR (100.6 MHz,
C6D5CD3, 300 K): δ 27.0 (s, CH2), 27.8 (s, CH2), 33.3 (d, CH,
2
1JPC 21.4 Hz), 33.9 (d, CH2, JPC 10.3 Hz). 31P NMR (145.8
1
300 K): δ 26.7 (d, CH, JPC 20.8 Hz), 30.6 (s, CH2), 31.8 (d,
MHz, C6D5CD3, 85% H3PO4, 298 K): δ Ϫ47.12 (s). MS EI: m/z
(%) 198 [P(C6H11)2ϩ, 14], 117 [InH2ϩ, 57], 83 [C6H11ϩ, 96]; IR
(Nujol) ν/cmϪ1: 1686 (s, br, In–H str.).
2
CH2, JPC 14.3 Hz). 31P NMR (36.3 MHz, C6D5CD3, 85%
H3PO4, 233 K): δ 3.45 (s). MS EI: m/z (%) 169 [P(C5H9)2ϩ, 4],
100 [P(C5H9)ϩ, 100], 69 [C5H9ϩ, 79]. IR (Nujol) ν/cmϪ1: 1649 (s,
br, In–H str.).
[In(SPh)3(PCy3)] 12. S2Ph2 (0.20 g, 0.93 mmol) in DME
(10 cm3) was added to a solution of [InH3(PCy3)] (0.25 g,
0.62 mmol) at Ϫ50 ЊC. The resulting solution was allowed to
slowly warm to room temperature and stirred for 4 h. Filtration
afforded a colourless solution, which was reduced in volume
and placed at Ϫ35 ЊC overnight yielding cubic colourless crys-
tals. Yield 0.23 g, 51%; mp 140 ЊC. 1H NMR (400 MHz, C6D6,
298 K): δ 1.0–2.0 [mult, 33H, C6H11], 6.9–7.4 [aromatics, mult,
15H]. 13C NMR (100.6 MHz, C6D6, 298 K): δ 25.9 [s, CH2], 27.4
[InH3(PPh3)] 5. A solution of PPh3 (0.67 g, 2.55 mmol) in
diethyl ether (30 cm3) was slowly added to a solution of [InH3-
(NMe3)] (≈2.54 mmol) in diethyl ether (70 cm3) at Ϫ70 ЊC. The
solution was warmed to Ϫ30 ЊC and stirred for 2 h whereupon a
precipitate of 5 deposited. This was washed with cold diethyl
ether (ca. 10 cm3) and dried in vacuo (0.67 g, 69%), decomp.
ca. 0 ЊC. IR (Nujol) ν/cmϪ1: 1681 (s, br, In–H str.).
2
1
[d, CH2, JPC 10.2 Hz], 29.9 [s, CH2], 31.85 [d, CH, JPC 13.0
Hz], 125.4, 128.8, 134.8, 136.3 [aromatics]. 31P (36.3 MHz,
C6D6, 298 K): δ 18.61. MS EI: m/z (%) 280 [PCy3ϩ, 100]. IR
(Nujol) ν/cmϪ1: 1573w, 1296m, 1080m, 1022m, 850m (Found:
C, 59.22; H, 6.81. Calc. for C36H48InPS3: C, 59.82; H, 6.69%).
[InH3(PCy3)2] 9. Method (i). A solution of PCy3 (1.43 g, 5.10
mmol) in diethyl ether (ca. 30 cm3) was slowly added to a solu-
tion of [InH3(NMe3)] (≈2.54 mmol) in diethyl ether (ca. 70 cm3)
at Ϫ70 ЊC. The resulting solution was warmed to Ϫ30 ЊC and
stirred for 2 h whereupon volatiles were removed in vacuo to
yield a white solid. This was washed with cold diethyl ether (ca.
10 cm3) and extracted with cold (Ϫ20 ЊC) DME (ca. 50 cm3),
filtered and the filtrate placed at Ϫ30 ЊC to yield 9 as colourless
needles (1.09 g, 63%).
[In(SePh)3(PCy3)] 13. Se2Ph2 (0.15 g, 0.9 mmol) in DME (10
cm3) was added to a solution of [InH3(PCy3)] (0.25 g, 0.62
mmol) at Ϫ50 ЊC. The resulting solution was allowed to slowly
warm to room temperature and stirred for 4 h. Filtration
produced a yellow solution, which was reduced in volume
and placed at Ϫ35 ЊC overnight yielding cubic yellow crystals.
Yield 0.27 g, 49%; mp 143–145 ЊC. 1H NMR (400 MHz, C6D6,
298 K): δ 0.9–2.0 [mult, 33H, C6H11], 6.9–8.0 [aromatics, mult,
15H]. 13C NMR (100.6 MHz, C6D6, 298 K): δ 25.99 [s, CH2],
27.49 [d, CH2, 2JPC 11.0 Hz], 29.86 [s, CH2], 31.94 [d, CH, 1JPC
Method (ii). A solution of PCy3 (0.32 g, 1.14 mmol) in DME
(ca. 20 cm3) was added slowly to a stirred solution of [InH3-
{PCy3}] (0.44 g, 1.10 mmol) in DME (ca. 30 cm3) at Ϫ70 ЊC.
The resulting solution was warmed to Ϫ30 ЊC and stirred for
2 h whereupon the volatiles were removed in vacuo to yield a
white solid. This was washed with cold diethyl ether (ca. 5 cm3)
and extracted with cold (Ϫ20 ЊC) DME (ca. 30 cm3), concen-
trated in vacuo, filtered and the filtrate placed at Ϫ30 ЊC to yield
9 as colourless needles (0.57 g, 76%), decomp. ϩ37 ЊC. 1H
NMR (400 MHz, C6D6, 300 K): δ 1.23–1.97 (m, 66H, C6H11),
5.81 (br s, 3H, In–H). 13C NMR (100.6 MHz, C6D6, 300 K):
13.0 Hz], 126.17, 128.50, 128.90, 136.83 [aromatics]. 31P (36.3
ϩ
MHz, C6D6, 298 K): δ 10.01. MS APCI: m/z (%) 280 [PCy3
,
100], 198 [PCy2ϩ, 48]. IR (Nujol) ν/cmϪ1: 1572w, 1296m, 1261m,
1114m, 1020m, 850m (Found: C, 48.96; H, 5.50. Calc. for
C36H48InPSe3: C, 50.08; H, 5.60%).
3
2
δ 27.3 (s, CH2), 28.4 (d, CH2, JPC 9.1 Hz), 32.0 (d, CH2, JPC
12.6 Hz), 32.6 (d, CH, JPC 18.6 Hz). 31P NMR (36.3 MHz,
[In(TePh)3(PCy3)] 14. Te2Ph2 (0.39 g, 0.93 mmol) in DME (10
cm3) was added to a solution of [InH3(PCy3)] (0.25 g, 0.6 mmol)
at Ϫ50 ЊC. The resulting solution was allowed to slowly warm
to room temperature and stirred for 4 h. Filtration afforded an
1
C6D6, 85% H3PO4, 298 K): δ 7.32 (s). MS EI: m/z (%) 280
[P(C6H11)3ϩ, 23], 198 [P(C6H11)2 ϩ Hϩ, 74], 116 [InHϩ, 100], 83
[C6H11ϩ, 65]. IR (Nujol) ν/cmϪ1: 1666 (s, br, In–H str.).
J. Chem. Soc., Dalton Trans., 2000, 545–550
549

View Article Online
Table 1 Crystal data for compounds 9 and 11–14
9
11
12
13
14
Chemical formula
M
C36H69InP2
678.67
C36H72In3P3
942.31
C36H48InPS3
722.71
C36H48InPSe3
863.41
C36H48InPTe3
1009.33
Crystal system
Space group
Triclinic
P1
Monoclinic
P21/n
Monoclinic
P21/c
Triclinic
P1
Triclinic
P1
¯
¯
¯
a/Å
b/Å
c/Å
α/Њ
8.036(2)
9.8530(10)
12.361(2)
80.611(9)
93.060(12)
68.470(7)
896.9(3)
1
10.2825(9)
30.403(2)
13.5672(10)
11.6650(10)
33.359(2)
19.122(2)
10.4085(7)
11.6417(9)
15.5724(11)
85.84(3)
82.96(3)
70.98(3)
1769.4(2)
2
10.3689(10)
12.0296(19)
15.862(2)
85.164(9)
82.62(2)
72.022(10)
1864.2(4)
2
β/Њ
94.14(2)
107.258(10)
γ/Њ
V/Å3
4230.3(6)
4
7106.0(11)
4
Z
T/K
150(2)
7.70
5411
5411
0.0373
0.0275
0.0754
0.0731
150(2)
17.57
8567
293(2)
9.10
150(2)
38.22
7434
7159
0.0623
0.0300
0.0755
0.0676
150(2)
30.00
7958
7524
0.1498
0.0653
0.1700
0.1487
µ(Mo-Kα)/cmϪ1
Reflections collected
No. unique reflections
R (all data)
R (I > 2σ(I))
wRЈ (all data)
wRЈ (I > 2σ(I))
15141
14420
0.0816
0.0352
0.0954
0.0822
8567
0.0440
0.0281
0.0698
0.0665
3 J. L. Atwood, S. G. Bott, C. Jones and C. L. Raston, Inorg. Chem.,
1991, 30, 4868.
4 M. G. Simmonds and W. L. Gladfelter, in The Chemistry of Metal
CVD, ed. T. Kodas and M. Hampden-Smith, VCH, Weinheim, 1994
and references therein.
5 E. Wiberg and M. Schmidt, Z. Naturforsch., Teil B, 1957, 12, 54.
6 A. G. Avent, C. Eaborn, P. B. Hitchcock, J. D. Smith and A. C.
Sullivan, J. Chem. Soc., Chem. Commun., 1986, 988.
7 O. T. Beachley, S. L. Chao, M. R. Churchill and R. F. See,
Organometallics, 1992, 11, 1486.
8 M. R. Churchill, C. H. Lake, S. L. Chao and O. T. Beachley,
J. Chem. Soc., Chem. Commun., 1993, 1577.
orange solution, which was reduced in volume and placed at
Ϫ35 ЊC overnight yielding cubic orange crystals. Yield 0.26 g,
41%; mp 127–128 ЊC. 1H NMR (400 MHz, C6D6, 298 K): δ 1.1–
2.1 [mult, 33H, C6H11], 6.9–8.3 [aromatics, mult, 15H]. 13C
NMR (100.6 MHz, C6D6, 298 K): δ 24.58 [s, CH2], 26.05 [d,
CH2, 2JPC 10.1 Hz], 28.54 [s, CH2], 31.83 [d, CH, 1JPC 12.9 Hz],
125.59, 127.66, 128.26, 139.53 [aromatics]. 31P (36.3 MHz,
C6D6, 298 K): δ Ϫ7.1. MS EI: m/z (%) 280 [PCy3ϩ, 100], 198
[PCy2, 52]. IR (Nujol) ν/cmϪ1: 1569w, 1261m, 1101m, 804m
(Found: C, 41.65; H, 4.75. Calc. for C36H48InPTe3: C, 42.84; H,
4.79%).
9 C. Kümmel, A. Meller and M. Noltemeyer, Z. Naturforsch., Teil B,
1996, 51, 209.
10 S. Aldridge, A. J. Downs and S. Parsons, Chem. Commun., 1996,
2055.
Structure determinations
11 (a) D. E. Hibbs, M. B. Hursthouse, C. Jones and N. A. Smithies,
Chem. Commun., 1998, 869; (b) M. D. Francis, D. E. Hibbs, M. B.
Hursthouse, C. Jones and N. A. Smithies, J. Chem. Soc., Dalton
Trans., 1998, 3249.
12 D. E. Hibbs, C. Jones and N. A. Smithies, Chem. Commun., 1999,
185.
13 D. E. Hibbs, M. B. Hursthouse, C. Jones and N. A. Smithies,
Organometallics, 1998, 17, 3108.
14 F. R. Bennett, F. M. Elms, M. G. Gardiner, G. A. Koutsantonis,
C. L. Raston and N. K. Roberts, Organometallics, 1992, 11, 1457.
15 J. L. Atwood, K. D. Robinson, F. R. Bennett, F. M. Elms, G. A.
Koutsantonis, C. L. Raston and D. J. Young, Inorg. Chem., 1992, 31,
2673.
16 A. J. Downs and C. R. Pulham, Chem. Soc. Rev., 1994, 175.
17 Chemistry of Aluminium, Gallium, Indium and Thallium, ed. A. J.
Downs, Blackie, Glasgow, 1993.
18 D. White and N. J. Coville, Adv. Organomet. Chem., 1994, 36, 95.
19 R. C. Bush and R. J. Angelici, Inorg. Chem., 1988, 27, 681.
20 F. M. Elms, M. G. Gardiner, G. A. Koutsantonis, C. L. Raston,
J. L. Atwood and K. D. Robinson, J. Organomet. Chem., 1993, 449,
45.
Crystals of 9, 11, 13 and 14 suitable for X-ray structure deter-
mination were mounted in silicone oil and a crystal of 12 was
mounted in contact cement. All crystallographic measurements
were made using an Enraf-Nonius CAD4 diffractometer. The
structure was solved by direct methods and refined on F2 by full
matrix least squares (SHELX97)28 using all unique data. All
non-hydrogen atoms are anisotropic with H-atoms included in
calculated positions (riding model) except the hydride ligands
of 11 for which the positions and isotropic displacement
parameters were refined. Empirical absorption corrections were
carried out by the DIFABS method29 on 9 and 11, and using
ψ scans for 13 and 14. Crystal data, details of data collections
and refinement are given in Table 1. Compound 12 crystallised
with 2 crystallographically independent molecules in the
asymmetric unit. There are no significant geometric differences
between the 2 molecules so the molecular structure of only
one is shown in Fig. 3. The molecular structures of the other
complexes are depicted in Figs. 1, 2, 4 and 5.
21 S. M. Godfrey, K. J. Kelly, P. Kramkowski, C. A. McAuliffe and
R. G. Pritchard, Chem. Commun., 1997, 1001.
22 S. T. Howard and C. Jones, unpublished work.
23 J. J. Jeiger, S. McKernan and W. L. Gladfelter, Inorg. Chem., 1999,
38, 2726.
CCDC reference number 186/1782.
See http://www.rsc.org/suppdata/dt/a9/a908418e/ for crystal-
lographic files in .cif format.
24 F. M. Elms, G. A. Koutsantonis and C. L. Raston, J. Chem. Soc.,
Chem. Commun., 1995, 1669.
Acknowledgements
25 A. H. Cowley, R. A. Jones, M. A. Mardones and C. M. Nunn,
Organometallics, 1991, 10, 1635.
26 H. J. Gysling, A. A. Wernberg, T. N. Blanton, Chem. Mater., 1992, 4,
900.
We gratefully acknowledge financial support from EPSRC
(studentships for N. A. S. and M. L. C.).
27 W. J. Grigsby, C. L. Raston, V. A. Tolhurst, B. W. Skelton and
A. H. White, J. Chem. Soc., Dalton Trans., 1998, 2547.
28 G. M. Sheldrick, SHELX97, University of Göttingen, 1997.
29 N. P. C. Walker and D. Stuart, Acta Crystallogr., Sect. A, 1983, 39,
158.
References
1 (a) C. Jones, G. A. Koutsantonis and C. L. Raston, Polyhedron,
1993, 12, 1829; (b) M. G. Gardiner and C. L. Raston, Coord. Chem.
Rev., 1997, 166, 1 and references therein.
2 C. L. Raston, A. F. H. Siu, C. J. Tranter and D. J. Young,
Tetrahedron Lett., 1994, 35, 5915.
Paper a908418e
550
J. Chem. Soc., Dalton Trans., 2000, 545–550