Single and double substrate insertion into the Ti=Nα bonds of terminal titanium hydrazides

Article abstract of DOI:10.1039/b920090h

Nitriles, CO₂ and isocyanates undergo net single or double insertion reactions into the Ti=N_α multiple bonds of terminal titanium hydrazides. These are the first such examples of this type of reactivity for any transition metal hydra

Full text of DOI:10.1039/b920090h

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Single and double substrate insertion into the TiQNa bonds of terminal
titanium hydrazidesw
Pei-Jen Tiong,a A. Daniel Schofield,a Jonathan D. Selby,a Ainara Nova,b Eric Clot*b and
Philip Mountford*a
Received (in Cambridge, UK) 28th September 2009, Accepted 2nd November 2009
First published as an Advance Article on the web 19th November 2009
DOI: 10.1039/b920090h
Nitriles, CO2 and isocyanates undergo net single or double
insertion reactions into the TiQNa multiple bonds of terminal
titanium hydrazides. These are the first such examples of this
type of reactivity for any transition metal hydrazide complex.
Group 4 hydrazides are closely connected with the important
area of catalytic N–C bond formation.11,23–25 Metallacycles
analogous to I are intermediates in the metal-catalysed
conversion of alkynes and hydrazines to hydrazones (i.e.,
hydrohydrazination: double N–H addition across a CRC
multiple bond11,23–25). The Na–Nb bond insertion compound
II and its homologues are intermediates in the diamination of
internal alkynes using Ph2NNH2 (N–N addition across a
CRC multiple bond).18,21
The fundamental paradigm of both hydrohydrazination and
diamination is net transfer of an ‘‘NNR2’’ moiety from a metal
hydrazide to an unsaturated substrate. In every case, the final
Ti–Na bond breaking step only occurs via protonolysis using
an incoming R2NNH2. As part of our continuing interest in
metal hydrazide and related chemistry, we have started to
develop new approaches to ‘‘NNR2’’ functional group transfer
reactions of these emerging new classes of early transition
metal compound. We report here the first MQNa bond
insertion reactions of any metal hydrazide compound.
Reaction of the recently reported18 terminal hydrazides
Cp*Ti{MeC(NiPr)2}(NNR2) (R = Ph (1a) or Me (1b))
with an excess of CO2 gave the dicarboxylate complexes
Cp*Ti{MeC(NiPr)2}{OC(O)N(NR2)C(O)O} (R = Ph (2a) or
Me (2b)) in good yields. The spectroscopic and other data for
2a and 2b are consistent with the Cs symmetric ‘‘double
insertion’’ products depicted in Scheme 1, and the X-ray
structure of 2a (Fig. 2) confirms this in the solid state.w
As Fig. 2 shows, two molecules of CO2 have effectively
inserted into the TiQNa (formally triple17) bond of 1a. This
type of transformation is very rare in metal–nitrogen multiple
Studies of the synthesis and reactivity of transition metal
hydrazides, (L)MQNNR2, have been of sustained interest
and activity for over 25 years. Historically, this is because of
their relevance to the biological conversion (fixation) of N2 to
ammonia.1–4 For this reason, much of the early work was
based around Group 6 systems4–6 for which MQNa–NbR2
group reactivity is characterised by reactions of the Nb atom
(quaternarisation or (for R = H) N–H bond chemistry) and/or
Na–Nb reductive cleavage. Although initial reactivity reports
of Group 4 hydrazides appeared some time ago,7,8 there has
recently been a surge in activity and output in this area,9–21
and a number of new reactions of the MQNNR2 functional
group have been discovered. These include the cycloaddition
of alkynes (as in I, Fig. 1) and heteroalkynes to the MQNa
multiple bond,21 and Na–Nb bond insertion reactions with
several substrates, including alkynes18,21 and isocyanides
(e.g. forming II and III, Fig. 1).16,22 The reactions of
Ti(N2NMe)(NNPh2)(py) (N2NMe = MeN(CH2CH2NSiMe3)2)
which lead to ‘‘insertion’’ products II and III can be viewed as
Na atom migration processes.
Just as Group 6 hydrazides are important because of their
structural and functional relevance to biological N2 fixation,
Fig. 1 Reaction products of TiQNNPh2 bonds with alkynes (I, II)
and XylNC (III): MQNa cycloaddition and Na–Nb bond cleavage.
a Chemistry Research Laboratory, University of Oxford,
Mansfield Road, Oxford, UK OX1 3TA.
E-mail: philip.mountford@chem.ox.ac.uk
b Institut Charles Gerhardt Montpellier, CNRS 5253,
Universite´ Montpellier 2, cc 1501, Place Euge`ne Bataillon,
F-34095 Montpellier Cedex 5, France
E-mail: eric.clot@univ-montp2.fr
Scheme 1 Cycloaddition and cycloaddition–insertion reactions of
Cp*Ti{MeC(NiPr)2}(NNR2) with heterocumulenes. All reactions
are at room temperature in toluene or C6H6 with 1 atm CO2
(where relevant).
w Electronic supplementary information (ESI) available: Characteri-
sation and crystallographic data, and computational details. CCDC
752046 and 752047. For ESI and crystallographic data in CIF or other
electronic format see DOI: 10.1039/b920090h.
ꢀc
This journal is The Royal Society of Chemistry 2010
Chem. Commun., 2010, 46, 85–87 | 85

View Article Online
Fig. 2 Displacement ellipsoid plot (20%) of Cp*Ti{MeC(NiPr)2}-
{OC(O)N(NPh2)C(O)O} (2a). Ti(1)–O(1) 1.937(2), Ti(1)–O(3) 1.951(2),
Ti(1)–Cpcent 2.051, Ti(1)–N(3) 2.059(2), Ti(1)–N(4) 2.090(2), N(1)–N(2)
1.394(3), C(13)–O(2) 1.206(3), C(14)–O(4) 1.218(3) A.
Fig.
3
Displacement ellipsoid plot (20%) of Ti(N2NMe)-
{NC(ArF)NNPh2}(py) (7). Ti(1)–N(1) 1.781(1), Ti(1)–N(4) 1.964(1),
Ti(1)–N(5) 1.964(1), C(1)–N(1) 1.338(2), C(1)–N(2) 1.308(2), C(1)–C(2)
1.514(2), N(2)–N(3) 1.441(2) A.
bond chemistry in general,26,27 and without precedent in metal
hydrazide chemistry at all. While the OC(O)N(NR2)C(O)O
moiety in 2a and 2b is a new structural unit in transition metal
chemistry, we note Chirik’s recent reports of metal-bound
1,1- or 1,2-NN(CO2)2 ligands prepared from metallocene–N2
complexes and CO2.28,29
When the reaction between 2a and CO2 was followed by 1H
NMR an intermediate cycloaddition product 3a (Scheme 1)
was observed. Subsequent scale up led to isolated 3a in 54%
yield. Reaction of 3a with CO2 quantitatively forms 2a. No
intermediate analogous to 3a was observed for 1b at room
temperature. The stepwise formation of 2a from 1a via 3a
suggested that mixed TiQNa insertion products should be
accessible. Reaction of 1a with TolNCO gave 4a (Scheme 1),
and subsequent addition of CO2 to 4a gave the mixed double
insertion product 5a in 37% isolated yield (quantitative when
followed by 1H NMR). The IR spectrum of 5a shows n(CQO)
and n(CQNTol) bands at 1699 and 1628 cmꢁ1, respectively.
clusters. Since the metal-bound hydrazonamide moiety in 7
still possesses a TiQN multiple bond, further reaction
chemistry of this moiety should be possible, allowing devel-
opment of this hitherto unexplored functional group.
The reactions of metal hydrazido (or indeed imido) complexes
with nitriles are almost without precedent, and only one
previous system has been described (eqn (2)).21,32 This leads
to the dimeric products 9 on reaction of Ti(N2Npy)(NR)(py)
(8, R = tBu or NPh2) with MeCN. Terminal hydrazonamides
analogous to 7 were not observed. Interestingly, when 6 was
reacted with MeCN or PhCN equilibrium mixtures of unreacted
starting materials and unidentified product(s) were observed.
To probe the factors and mechanism leading to the formation
of the insertion product 7, we carried out DFT(B3PW91)
SiH3
calculationsw on the model system Ti(N2
N
Me)(NNMe2)(py)
SiH3 Me
(6Q, N2
N
= MeN(CH2CH2NSiH3)2) and ArFCN
(Fig. 4). All energies given in the text are Gibbs free energies
(T = 298 K, P = 1 atm) estimated from separated 6Q and nitrile.
ð1Þ
Compounds 2a, 2b and 5a form via a cycloaddition–insertion
mechanism leading to new homo-coupled and cross-coupled
products. As shown in eqn (1) we have also discovered another
type of TiQNa insertion reaction starting from the previously
reported Ti(N2NMe)(NNPh2)(py) (6) mentioned in the intro-
ductory paragraphs. In this case a single molecule of ArFCN
(ArF = C6F5) has inserted into the TiQNa bond of 6 forming
the hydrazonamide derivative 7. The solid state structure of 7
(Fig. 3) confirmed the formation of the new hydrazonamide
fragment NC(ArF)NNPh2. The Ti(1)–N(1) distance of 1.781(1) A
is consistent with substantial multiple bond character, and is
only slightly longer than TiQNa (1.733(5) A) in 6 itself.
Compound 7 is the first example of a terminal hydrazon-
amide complex. Mono- and di-metallated hydrazonamides
have been reported by sequential reaction of AlMe3 or dialkyl
zincs with Me2NNH2 and MeCN,30,31 but formed polynuclear
ð2Þ
The reaction of 6Q with ArFCN starts with dissociation
of py leading to Int1. This process, unfavorable by DG =
9.4 kcal molꢁ1, generates a vacant site for coordination
of ArFCN. Two different orientations are possible for the
Z2-ArFCN adduct, namely Int2 (with ArF towards NNMe2,
see Fig. 4) and Int2b (with N toward NNMe2, not shown in
Fig. 4). Int2 is more stable than Int2b by 6.0 kcal molꢁ1, and
only Int2 was considered in the mechanism. From this inter-
mediate, [2+2] cycloaddition to form the diaza-metallacycle
Int3 is thermodynamically favored (DG = ꢁ7.3 kcal molꢁ1
)
with a low activation barrier (DG# = 6.5 kcal molꢁ1). The
Ti–Nnitrile distance in Int3 is 0.22 A shorter than Ti–Na (Fig. 4),
while C–Na and C–Nnitrile distances have similar values (1.35 and
ꢀc
This journal is The Royal Society of Chemistry 2010
86 | Chem. Commun., 2010, 46, 85–87

View Article Online
In conclusion, we have reported the first MQNa insertion
reactions of any transition metal hydrazide. Two different
mechanisms have been observed: (i) cycloaddition–insertion
leading to homo- or cross-coupled bis(heterocumulene)
derivatives; (ii) ‘‘arrested’’ metathesis giving single-substrate
insertion, again with TiQNa bond cleavage. In this latter case
(7, eqn (1)), a new TiQN multiple bond is formed which could
in principle be a site of additional functionalisation. These
results will be of benefit in developing Group 4 based substrate
functionalisation chemistry via hydrazide intermediates.
We thank the EPSRC for scholarships to A.D.S. and J.D.S.,
the Malaysian Higher Education Ministry for a scholarship to
P.-J.T., and the Spanish MICINN for a MEC postdoctoral
fellowship to A.N.
Notes and references
1 R. R. Schrock, Angew. Chem., Int. Ed., 2008, 47, 5512.
2 R. R. Schrock, Acc. Chem. Res., 2005, 38, 955.
3 B. A. MacKay and M. D. Fryzuk, Chem. Rev., 2004, 104, 385.
4 M. Hidai, Coord. Chem. Rev., 1999, 185–186, 99.
5 M. D. Fryzuk, Chem. Rec., 2003, 3, 2.
N
Me)(NNMe2)(py)
SiH3
Fig. 4 Energy profile for reaction of Ti(N2
(6Q) with ArFCN. All energies are DG in kcal molꢁ1. [Ti] =
SiH3
N
Me).
6 M. P. Shaver and M. D. Fryzuk, Adv. Synth. Catal., 2003, 345,
1061.
Ti(N2
1.32 A, respectively). Ring-opening of Int3 (via Int4) and
coordination of pyridine is exoergic by DG = ꢁ3 kcal molꢁ1
and yields 7Q (a model for the product 7 observed experimentally).
The reactions of 6Q with MeCN and PhCN were also
examined computationally. In contrast to the situation with
the fluorinated substrate, the products of MeCN (7Q_Me) and
PhCN (7Q_Ph) insertion into the TiQNa bond were computed
to be endoergic (DG = 3.1 kcal molꢁ1 for MeCN; 5.1 kcal molꢁ1
for PhCN). The Z2-RCN adducts (Int2_Me and Int2_Ph)
and the cycloaddition products (Int3_Me and Int3_Ph) have
also energies above those of the reactants (DG = 6.3 kcal molꢁ1
for Int2_Me; DG = 8.5 kcal molꢁ1 for Int2_Ph; DG =
0.2 kcal molꢁ1 for Int3_Me; DG = 3.2 kcal molꢁ1 for Int3_Ph).
There is thus no thermodynamic driving force towards forma-
tion of the insertion products 7Q_Me or 7Q_Ph. This could
explain the experimental observation of equilibrium mixtures of
products and starting materials in the reactions of the real
system 6 with MeCN and PhCN. Note that self-trapping
(dimerisation) of species of the type Int3 would lead to dimeric
compounds such as 9 (eqn (2)) observed experimentally. An
NBO analysis of the electronic structure of 7Q indicated that
the electron-withdrawing character of the ArF group plays a
crucial role in formation of the insertion product through
stabilisation of the Na lone pair. Accordingly, the insertion
product with CF3CN (7Q_CF3) was computed to lie below
7 P. J. Walsh, M. J. Carney and R. G. Bergman, J. Am. Chem. Soc.,
1991, 113, 6343.
8 A. J. Blake, J. M. McInnes, P. Mountford, G. I. Nikonov,
D. Swallow and D. J. Watkin, J. Chem. Soc., Dalton Trans.,
1999, 379.
9 D. J. Mindiola, Angew. Chem., Int. Ed., 2008, 47, 1557.
10 Y. Li, Y. Shi and A. L. Odom, J. Am. Chem. Soc., 2004, 126, 1794.
11 S. Banerjee and A. L. Odom, Organometallics, 2006, 25, 3099.
12 S. Patel, Y. Li and A. L. Odom, Inorg. Chem., 2007, 46, 6373.
13 K. Weitershaus, H. Wadepohl and L. H. Gade, Organometallics,
2009, 28, 3381.
14 T. Gehrmann, J. Lloret Fillol, H. Wadepohl and L. H. Gade,
Angew. Chem., Int. Ed., 2009, 48, 2152.
15 H. Herrmann, J. L. Fillol, T. Gehrmann, M. Enders, H. Wadepohl
and L. H. Gade, Chem.–Eur. J., 2008, 14, 8131.
16 H. Herrmann, J. L. Fillol, H. Wadepohl and L. H. Gade, Angew.
Chem., Int. Ed., 2007, 46, 8426.
17 T. B. Parsons, N. Hazari, A. R. Cowley, J. C. Green and P. Mountford,
Inorg. Chem., 2005, 44, 8442.
18 J. D. Selby, C. D. Manley, M. Feliz, A. D. Schwarz, E. Clot and
P. Mountford, Chem. Commun., 2007, 4937.
19 A. J. Clulow, J. D. Selby, M. G. Cushion, A. D. Schwartz and
P. Mountford, Inorg. Chem., 2008, 47, 12049.
20 J. D. Selby, C. D. Manley, A. D. Schwarz, E. Clot and P. Mountford,
Organometallics, 2008, 27, 6479.
21 J. D. Selby, C. Schulten, A. D. Schwarz, A. Stasch, E. Clot,
C. Jones and P. Mountford, Chem. Commun., 2008, 5101.
22 A. D. Schofield and P. Mountford, unpublished results.
23 C. Cao, Y. Shi and A. L. Odom, Org. Lett., 2002, 4, 2853.
24 J. T. Ciszewski, J. F. Harrison and A. L. Odom, Inorg. Chem.,
2004, 43, 3605.
25 A. L. Odom, Dalton Trans., 2005, 225.
26 D. S. Glueck, J. Wu, F. J. Hollander and R. G. Bergman,
J. Am. Chem. Soc., 1991, 113, 2041.
separated 6Q and CF3CN by DG = ꢁ11.6 kcal molꢁ1
.
The proposed mechanism is reminiscent of a classical
Chauvin-type metathesis reaction,33 in that it proceeds
through [2+2] cycloaddition (Int3) and reverse cycloaddition
steps, leading to 7Q. However, whereas in alkyne or alkene
metathesis (mediated through metal alkylidyne or alkylidene
species), separation of the bond-exchanged partners can occur,
this does not happen in the reactions of hydrazides and nitriles
since the required species, Ti(N2NMe)(py)(QNꢂ) and
Me2NNQCꢂArF, are evidently radical-like in nature. Thus
eqn (1) and Fig. 4 can be thought of as an ‘‘arrested’’ or
‘‘frustrated’’ TiQN/CRN metathesis reaction.
27 A. E. Guiducci, C. L. Boyd, E. Clot and P. Mountford, Dalton
Trans., 2009, 5960.
28 W. H. Bernskoetter, E. Lobkovsky and P. J. Chirik, Angew. Chem.,
Int. Ed., 2007, 46, 2858.
29 D. J. Knobloch, H. E. Toomey and P. J. Chirik, J. Am. Chem. Soc.,
2008, 130, 4248.
30 V. C. Gibson, C. Redshaw, A. J. P. White and D. J. Williams,
Angew. Chem., Int. Ed., 1999, 38, 961.
31 M. R. Elsegood and C. Redshaw, Chem.–Eur. J., 2008, 14, 3530.
32 S. M. Pugh, D. J. M. Trosch, D. J. Wilson, A. Bashall, F. G. N. Cloke,
¨
L. H. Gade, P. B. Hitchcock, M. McPartlin, J. F. Nixon and
P. Mountford, Inorg. Chem., 2000, 19, 3205.
33 Y. Chauvin, Angew. Chem., Int. Ed., 2006, 45, 3740.
ꢀc
This journal is The Royal Society of Chemistry 2010
Chem. Commun., 2010, 46, 85–87 | 87