Synthesis, characterisation and theoretical studies of amidinato-indium(I) and thallium(I) complexes: Isomers of neutral group 13 metal(I) carbene analogues

Article abstract of DOI:10.1039/b507242e

The synthesis and characterisation of the monomeric amidinato-indium(I) and thallium(I) complexes, [M(Piso)]PisoH, M = In or Tl, Piso^- = [ArNC(Bu^t)NAr]^-, Ar = C₆H₃Pr ₂ⁱ-2,6, are reported. These complexes, in which the metal centre is chelated by the amidinate ligand in an N,η³-arene- fashion, can be considered as isomers of four-membered group 13 metal(I) carbene analogues. Theoretical studies have compared the relative energies of both isomeric forms of a model complex, [In{PhNC(H)NPh}]. The Royal Society of Chemistry 2005.

Full text of DOI:10.1039/b507242e

View Article Online / Journal Homepage / Table of Contents for this issue
C O M M U N I C A T I O N
Synthesis, characterisation and theoretical studies of
amidinato-indium(I) and thallium(I) complexes: isomers of neutral
group 13 metal(I) carbene analogues
Cameron Jones,*a Peter C. Junk,b Jamie A. Platts,a Daniel Rathmanna and Andreas Stascha
a Centre for Fundamental and Applied Main Group Chemistry, School of Chemistry,
Main Building, Cardiff University, Cardiff, UK CF10 3AT. E-mail: jonesca6@cardiff.ac.uk
b School of Chemistry, Monash University, PO Box 23, Victoria, 3800, Australia
Received 23rd May 2005, Accepted 13th June 2005
First published as an Advance Article on the web 21st June 2005
The synthesis and characterisation of the monomeric
amidinato-indium(I) and thallium(I) complexes, [M(Piso)]·
PisoH, M = In or Tl, Piso− = [ArNC(But)NAr]−, Ar =
C6H3Pri2-2,6, are reported. These complexes, in which the
metal centre is chelated by the amidinate ligand in an
13 metal(III) fragments (e.g. indium hydrides7), lent hope to our
cause. Initial efforts in this direction have led, unexpectedly, to
two monomeric N,arene-chelated isomers of 3, the synthesis,
characterisation and theoretical study of which are reported
herein.
3
N,g -arene-fashion, can be considered as isomers of four-
The reaction of the potassium amidinate, K[Piso], Piso−
=
membered group 13 metal(I) carbene analogues. Theoret-
ical studies have compared the relative energies of both
isomeric forms of a model complex, [In{PhNC(H)NPh}].
[ArNC(But)NAr]−, Ar = C6H3Pri2-2,6, with either InCl or TlBr
in a 1 : 1 stoichiometry led to the amidinato-metal(I) complexes,
4 and 5, in low to moderate yields (ca. 30%) after recrystallisation
from hexane (Scheme 1). It is believed that the PisoH of crystalli-
sation in these complexes arises from the partial decomposition
of [M(Piso)], M = In or Tl, via metal deposition and solvent
hydrogen abstraction. Similar decomposition processes have
been suggested for examples of the b-diketiminato complexes,
1.2 Although the partial decomposition of [M(Piso)] can be
avoided in these reactions, we have not yet been able to crystallise
[M(Piso)] free of PisoH. As a result, higher yields of 4 and 5 were
achieved with the 1 : 1 : 1 reactions of the metal halide, K[Piso]
and PisoH. Hexane solutions of both 4 and 5 decompose,
depositing metal, upon standing at 25 ◦C overnight, but the
complexes display differing thermal stabilities in the solid state
(4 decomp. 150–155 ◦C, 5 decomp. 50–52 ◦C).
The rapidly emerging importance of five-membered N-
heterocyclic carbenes (NHCs),:C{N(R)C(Rꢀ)}2, cannot be un-
derstated. Since they were first isolated in 1991, numerous
reports of their complexation to the majority of the p-, d-
and f-block metals have come forward.1 Much of the interest
in NHCs has derived from their special electronic properties
which have led to many of their transition metal complexes
exhibiting catalytic activity in a number of synthetic processes.1b
Not surprisingly, studies of heavier p-block element analogues
of these heterocycles have followed. In group 13, these include
the neutral, monomeric six-membered b-diketiminato-metal(I)
complexes, 1,2 the coordination and other chemistry of which
has begun to be examined.3 In addition, two direct valence iso-
electronic analogues of NHCs, namely the anions, 2, have been
reported by our group4 and that of Schmidbaur.5 Like NHCs,
one of these, 2, R = C6H3Pri2-2,6, has been shown to be very
nucleophilic and capable of stabilising p- and d-block metal
fragments which are normally thermally labile and/or contain
the metal in a low oxidation state.3
Considering the success had with these five and six-membered
heterocycles, it seemed appropriate to attempt the preparation
of neutral, four membered group 13 metal(I) heterocycles, 3, and
to subsequently develop their chemistry. Such a goal is especially
timely, as the first example of a four-membered NHC has been
very recently reported, and its use as a ligand in an olefin
metathesis catalyst described.6 Although the steric protection
afforded the metal centre by the N-substituents in 3 is likely to be
less than in 1 or 2, the fact that N,N-chelating amidinate ligands
have been shown to stabilise normally thermally labile group
Scheme 1 Reagents and conditions: i, InCl or TlBr, PisoH, THF.
The solution state 1H NMR spectra† of 4 and 5 are compli-
cated but similar to each other. It is, however, clear that the weak
interaction between the PisoH and [M(Piso)] fragments seen in
the solid state is not retained in solution. The reasoning behind
1
this conclusion is that the H NMR spectra of the complexes
dissolved in D8-toluene exhibit the characteristic pattern for
free PisoH, which is known to exist as a number of isomers in
solution8 and not solely as its Z-anti-isomer, as it does in the
structures of 4 and 5. In addition, there is little change in the
appearance of the spectra over the temperature range 298–213 K,
which suggests that there is no significant fluxional interaction
of the fragments in solution. From the available NMR data we
cannot be sure if the [M(Piso)] fragments of 4 and 5 retain their
solid state structure in solution, or if they are in equilibrium
with their heterocyclic form, cf. 3. However, considering the
likely high barrier to conversion between the two forms (vide
infra) and the fact that 4 and 5 can be recrystallised unchanged
from hexane/toluene solutions, it is probable that their N,arene-
amidinate chelated structures persist in solution.
T h i s j o u r n a l i s
T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 5
D a l t o n T r a n s . , 2 0 0 5 , 2 4 9 7 – 2 4 9 9
2 4 9 7
©

View Article Online
Compounds 4 and 5 are isomorphous‡ and thus only the
molecular structure for 4 is depicted in Fig. 1. Both complexes
are monomeric with the localised amidinate ligand chelating
In order to estimate the strength of the solid state in-
teraction between the PisoH and [In(Piso)] molecules in 4,
DFT calculations were carried out on the model complex,
[In{PhNC(H)NPh}]·{PhN(H)C(H)NPh}.15 Starting from sym-
metrised (Cs) and “cut-down” coordinates taken from the X-ray
crystal structure of 4, the model complex converged with an
optimised geometry similar to that observed in the solid state
structure of the indium complex. The magnitude of the binding
energy between the two molecular fragments was found to be
very weak (7.20 kJ mol−1), as expected.
We were also interested in examining the difference in energy
between the isomers of 4 in which the amidinate ligand chelates
the In centre in either an N,N- or N,arene-fashion. To this
end, the geometries of both isomers of the model complex,
[In{PhNC(H)NPh}] 6, were optimised using the same DFT
method as above. Graphical representations of these isomers,
and relevant geometrical parameters, are depicted in Fig. 2.
Isomer 6b converged with a similar geometry to the [In(Piso)]
fragment of 4, taking into account the steric differences between
the theoretical and experimental molecules. Isomer 6a converged
with a delocalised amidinate backbone and an essentially
planar InN2C four-membered ring. Not surprisingly, the In–
N bond lengths of 6a are significantly longer than that in 6b.
More surprising is the fact that 6a is more stable than 6b by
37.9 kJ mol−1, despite the apparent strain of its four-membered
ring. The barrier to conversion between these isomeric forms
was found to be significant at 85.0 kJ mol−1. An NBO charge
analysis of isomer 6b revealed the In–N bond to be largely
ionic (In charge +0.778) and the lone pair at the indium centre
to be essentially of s-character (5s1.92 5p0.30). In combination
with the aforementioned M · · · H separations, these observations
count against any significant M · · · H–N hydrogen bonding in 4
and 5.
1
3
the metal in an g -N,g -arene fashion, rather than acting
as a delocalised N,N-chelating ligand. It is noteworthy that
a formamidinate anion, [ArNC(H)NAr]− (Fiso−), which is
closely related to Piso− has recently been reported to act as
a similar N,arene-chelating ligand in its potassium complex,
[K(THF)3(Fiso)].9 The fact that 5 is monomeric is of interest, as
recent work on thallium(I) amides has revealed their propensity
to aggregate through intermolecular Tl–Tl and/or Tl–arene
interactions,10 e.g. as in [{TlN(SiMe3)(C6H3Pri2-2,6)}4].11 Saying
that, monomeric thallium amides are accessible if bulky amide
substituents are employed. Examples here include the two-
coordinate systems, 1d,f and g,2d,e and the essentially one-
coordinate complex, [TlN(Me){C6H3(C6H2Me3-2,4,6)2-2,6}].12
˚
Fig. 1 Molecular structure of 4. Selected bond lengths (A) and
angles (◦): In(1)–N(1) 2.329(5), N(1)–C(1) 1.379(8), C(1)–N(2) 1.286(7),
N(2)–C(18) 1.406(6), In(1)–C(18) 2.822(6), In(1)–C(19) 2.968(6),
In(1)–C(23) 3.038(6), N(3)–C(30) 1.387(8), N(4)–C(30) 1.269(7),
In(1)–N(1)–C(1) 126.3(4), N(1)–C(1)–N(2) 122.2(5), C(1)–N(2)–C(18)
◦
˚
119.9(5). Selected bond lengths (A) and angles ( ) for 5: Tl(1)–N(1)
2.445(7), N(1)–C(1) 1.366(11), C(1)–N(2) 1.299(10), N(2)–C(18)
1.412(8), Tl(1)–C(18) 2.882(7), Tl(1)–C(19) 3.015(7), Tl(1)–C(23)
3.087(7), N(3)–C(30) 1.352(10), N(4)–C(30) 1.283(10), Tl(1)–N(1)–C(1)
125.6(5), N(1)–C(1)–N(2) 122.9(7), C(1)–N(2)–C(18) 120.4(6).
Fig.
2
Calculated and fully optimised geometries of (a) the
N,N-amidinate chelating and (b) N,arene-amidinate chelating isomers
The Tl–N distance in 5 is in the normal range for thallium(I)
amides,13 though because of a strong dependence of Tl–N
bond lengths on steric and electronic effects, this range is wide.
In contrast, there are very few examples of indium(I) amides
to compare to 4, but its In–N separation is not dissimilar
˚
of [In{PhNC(H)NPh}] 6. Selected bond lengths (A) and angles
(◦) for (a): In–N 2.397 (avge.), N–C(amid.) 1.321 (avge.), N–In–N
56.02, In–N–C(amid.) 93.43 (avge.), N–C–N 116.95; (b): In–N 2.281,
=
N–C(amid.) 1.351, N C(amid.) 1.295, In–phenyl plane 2.850, N–C–N
127.02, In–N–C(amid.) 125.26.
2c
2d
˚
˚
to those of 1c [2.272 A avge.] and 1e [2.361 A avge.].
3
In both 4 and 5 one arene substituent is approximately g -
coordinated to the metal centre with M-arene least squares
In conclusion, the synthesis and characterisation of
monomeric amidinato-indium(I) and thallium(I) complexes has
been reported. The former represents a rare example of an
indium(I)–amide, and both can be considered as isomeric
forms of four-membered group 13 metal(I) carbene analogues.
Theoretical studies have suggested that the heterocyclic isomer
of the indium amide, 4, should be stable and perhaps inter-
convertible with its experimentally observed form. Our further
efforts to stabilise four-membered group 13 metal(I) carbene
analogues will be documented in a forthcoming publication.
˚
˚
plane distances of 2.734(3) A and 2.786(4) A, respectively. The
latter distance is within the normal range for intramolecular
arene–Tl(I) interactions.10 To the best of our knowledge there
are no structurally characterised examples of intramolecular
In–arene interactions in indium(I) amide complexes to compare
with 4, however, reported interactions between arenes and the
+
˚
In ion appear to be of a similar distance, e.g. 2.83 and 2.89 A in
[In(g -mesitylene)2]+.14 Although the arene substituent attached
6
to N(4) in both structures appears to be orientated correctly
with respect to the metal centres to act as a p-donor ligand,
the large distances from the metal to the least squares plane of
˚
˚
that substituent [4 3.978(4) A and 5 3.830(6) A] suggest that any
intermolecular M–arene interactions must be weak. In addition,
the large separations between the amidine proton, H(3), and
Acknowledgements
We gratefully acknowledge financial support from The Lever-
hulme Trust (fellowship for AS) and the ERASMUS scheme of
the European Union (travel grant for DR). Thanks also go to
the EPSRC Mass Spectrometry Service.
˚
˚
the metal centres [4 3.684(4) A, 5 3.583(5) A] are probably not
indicative of any unconventional hydrogen bonding interaction
between the metal lone pair and N–H fragments.
2 4 9 8
D a l t o n T r a n s . , 2 0 0 5 , 2 4 9 7 – 2 4 9 9

View Article Online
See http://dx.doi.org/10.1039/b507242e for crystallographic data in
CIF or other electronic format.
Notes and references
† Synthesis and selected data for 4: A solution of K[Piso] (0.64 g,
1.40 mmol, prepared from the reaction of PisoH8 and KN(SiMe3)2)
and PisoH (0.59 g, 1.40 mmol) in THF (30 mL) was added over
5 min to a suspension of InCl (0.20 g, 1.32 mmol) in THF (20 mL)
at −90 ◦C. The reaction mixture was warmed to 0 ◦C overnight in
the absence of light, whereupon it was filtered and volatiles removed
from the filtrate in vacuo. Extraction of the residue into hexane (30 mL)
and concentration to ca. 6 mL afforded pale yellow crystals of 4 after
placement at −30 ◦C overnight. Yield: 0.92 g (72 %); Mp 150–155 ◦C
(dec.); 1H NMR (400 MHz, C7D8, 298 K): d 0.72–1.55 (overlapping
unresolved multiplets, 48 H, CH(CH3)2), 1.27 (br. s, 9 H, But), 1.41 (s,
9 H, But), 3.02–3.45 (overlapping septets, 8 H, CH(CH3)2), 5.15 and
1 (a) e.g. N. Kuhn and A. Al-Sheikh, Coord. Chem. Revs., 2005, 249,
829; (b) W. A. Herrmann, Angew. Chem., Int. Ed., 2002, 41, 1291;
(c) C. J. Carmalt and A. H. Cowley, Adv. Inorg. Chem., 2000, 50, 1;
(d) D. Bourissou, O. Guerret, F. P. Gabai and G. Bertrand, Chem.
Rev., 2000, 100, 39, and references therein.
2 (a) C. Cui, H. W. Roesky, H.-G. Schmidt, M. Noltemeyer, H. Hao
and F. Cimpoesu, Angew. Chem., Int. Ed., 2000, 39, 4274; (b) N. J.
Hardman, B. E. Eichler and P. P. Power, Chem. Commun., 2000,
1991; (c) M. S. Hill and P. B. Hitchcock, Chem. Commun., 2004,
1818; (d) M. S. Hill, P. B. Hitchcock and R. Pongtavornpinyo, Dalton
Trans., 2005, 273; (e) Y. Cheng, P. B. Hitchcock, M. F. Lappert and
M. Zhou, Chem. Commun., 2005, 752.
3 R. J. Baker and C. Jones, Coord. Chem. Rev.,
DOI: 10.1016/j.ccr.2004.12.016.
4 R. J. Baker, R. D. Farley, C. Jones, M. Kloth and D. M. Murphy,
J. Chem. Soc., Dalton Trans., 2002, 3844.
5 E. S. Schmidt, A. Schier and H. Schmidbaur, J. Chem. Soc., Dalton
Trans., 2001, 505.
6 (a) E. Despagnet-Ayoub and R. H. Grubbs, J. Am. Chem. Soc.,
2004, 126, 10198; (b) E. Despagnet-Ayoub and R. H. Grubbs,
Organometallics, 2005, 24, 338.
5.31 (2 br. s, 1 H, NH from PisoH isomers), 6.70–7.25 (overlapping m,
12 H, ArH); MS/APCI m/z (%): 533 (InPiso − H+, 8), 422 (PisoH2
,
+
100); IR m/cm−1 (Nujol): 3324(m) (NH), 1653(m), 1616(m), 1456(s),
1388(m), 1210(m), 932(m), 800(m), 767(m); accurate mass MS/EI calc.
for [In(Piso)] C29H43N2In: 534.2460, found: 534.2459; 5: A similar
procedure as was used to◦prepare 4 gave compound 5 as pale yellow
crystals. (74%); Mp 50–52 C (dec.); 1H NMR (400 MHz, C7D8, 298 K):
d 0.78–1.43 (overlapping unresolved multiplets, 48 H, CH(CH3)2), 1.27
(br. s, 9 H, But), 1.47 (s, 9 H, But), 3.04–3.47 (overlapping septets, 8 H,
CH(CH3)2), 5.13 and 5.31 (2 br. s, 1 H, NH from PisoH isomers), 6.65–
+
7.26 (overlapping m, 12 H, ArH); MS/APCI m/z (%): 422 (PisoH2
,
7 R. J. Baker, C. Jones, P. C. Junk and M. Kloth, Angew. Chem., Int.
Ed., 2004, 43, 3852.
8 A. Xia, H. M. El-Kaderi, M. J. Heeg and C. H. Winter, J. Organomet.
Chem., 2003, 682, 224.
9 M. L. Cole and P. C. Junk, J. Organomet. Chem., 2003, 666, 55.
10 (a) L. H. Gade, Dalton Trans., 2003, 267; (b) C. Janiak, Coord. Chem.
Rev., 1997, 163, 107, and references therein.
100); IR m/cm−1 (Nujol): 3331(m) (NH), 1651(m), 1614(m), 1455(s),
1376(m), 1211(m), 930(m), 801(m), 760(m). The 13C{ H} NMR spectra
1
of both compounds displayed many broad, overlapping signals which
could not be confidently assigned.
‡ Crystal data for 4: C58H87InN4 M = 955.14 triclinic, space group
˚
P1, a = 10.436(2), b = 10.670(2), c = 14.277(3) A, a = 102.258(10),
3
˚
b = 109.421(10), c = 105.790(10)o, V = 1360.1(5) A , Z = 1, Dc =
11 S. D. Waezsada, T. Belgardt, M. Noltemeyer and H. W. Roesky,
Angew. Chem., Int. Ed. Engl., 1994, 33, 1351.
1.166 g cm−3, F(000) = 512, l(Mo-Ka) = 0.473 mm−1, 150(2) K,
10385 unique reflections [R(int) 0.0630], R (on F) 0.0529, wR (on F2)
0.1261 (I > 2rI); 5: C58H87TlN4 M = 1044.69 triclinic, space group
12 R. J. Wright, M. Brynda and P. P. Power, Inorg. Chem., 2005, 44,
3368.
˚
P1, a = 10.398(2), b = 10.654(2), c = 14.201(3) A, a = 102.53(3), b =
13 As determined by a survey of the Cambridge Crystallographic
Database, May 2005.
14 J. Ebenho¨ch, G. Mu¨ller, J. Riede and H. Schmidbaur, Angew. Chem.,
Int. Ed. Engl., 1984, 23, 386.
15 A full description of the theoretical methods employed and
the results obtained in this study are included as ESI. See
http://dx.doi.org/10.1039/b507242e.
3
−3
˚
108.39(3), c = 105.77(3), V = 1355.5(5) A , Z = 1, Dc = 1.280 g cm
,
F(000) = 544, l(Mo-Ka) = 3.017 mm−1, 120(2) K, 10072 unique
reflections [R(int) 0.0928], R (on F) 0.0525, wR (on F2) 0.1353 (I >
2rI). During the course of refinement it was found that the metal atoms
in both structures were disordered over two sites. This disorder was
successfully modelled. CCDC reference numbers 272677 and 272648.
D a l t o n T r a n s . , 2 0 0 5 , 2 4 9 7 – 2 4 9 9
2 4 9 9