Synthesis of a four-coordinate titanium(IV) oxoanion via deprotonation and decarbonylation of complexed formate

Article abstract of DOI:10.1039/b504492h

Deprotonation of the titanium formate complex [Ar(t-Bu)N] ₃TiOC(O)H with LiN(i-Pr)₂ resulted in the release of free CO and the formation of a titanium(IV) oxoanion complex, isolated as its lithium salt. The Royal Society of Chemistry 2005.

Full text of DOI:10.1039/b504492h

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Synthesis of a four-coordinate titanium(IV) oxoanion via deprotonation
and decarbonylation of complexed formate{{
Arjun Mendiratta, Joshua S. Figueroa and Christopher C. Cummins*
Received (in Berkeley, CA, USA) 31st March 2005, Accepted 16th May 2005
First published as an Advance Article on the web 9th June 2005
DOI: 10.1039/b504492h
The successful deprotonation of 2 would provide a compound
in which the CO2 anion is stabilized by coordination to a Ti center.
Such a compound would be of considerable interest in that the
reduction potential of free CO2 is 21.90 V vs. NHE.16,17 In any
event, treatment of 2 with a slight excess of LiN(i-Pr)2 in Et2O
Deprotonation of the titanium formate complex
[Ar(t-Bu)N]3TiOC(O)H with LiN(i-Pr)2 resulted in the release
of free CO and the formation of a titanium(IV) oxoanion
complex, isolated as its lithium salt.
1
In recent years, both transition metal formates and oxos have been
studied intensively; the former as model compounds in the rich and
varied chemistry of syngas on metal surfaces,1–3 and the latter due
to their importance in biological4 and industrial oxidations.5
Historically, the oxo ligand has been encountered primarily in
conjunction with the middle metals of the transition series;6 this
observation has motivated a number of recent syntheses of both
early-7–9 and late-metal10,11 oxos. We report here on an
unprecedented, base-triggered formate decarbonylation, resulting
in the synthesis of the first anionic oxo of titanium(IV). DFT
calculations have been carried out in order to elucidate the
electronic structure of the new compounds.
resulted in the formation of a new crystalline product whose H
NMR spectrum displayed a single N(t-Bu)Ar environment. An
X-ray diffraction study revealed the product to be titanium(IV)
oxoanion 3, likely formed by facile CO ejection subsequent to
deprotonation. We have previously observed decarbonylation of
isoelectronic [(Ar[t-Bu]N)3NbLNLCLO]2 to form an anionic
nitridoniobium complex.18 Detection of liberated CO in the
present system was carried out by vacuum transfer of the volatiles
onto (Ph3P)3RhCl (Wilkinson’s catalyst), resulting in formation of
(Ph3P)2Rh(CO)Cl as documented in the literature.19 The genera-
tion of HN(i-Pr)2 was confirmed in a separate experiment.
Our synthesis commences with treatment of an emerald green
ethereal solution of Ti[N(t-Bu)Ar]3 (Ar 5 3,5-Me2C6H3, 1)12 with
1 equiv. of t-butyl formate at 25 uC, resulting in an immediate
color change to red-brown, followed by precipitation of a yellow
solid over the course of a few minutes. On the basis of 1H and 13
C
NMR spectroscopy, X-ray crystallography, and elemental analy-
sis, this solid is identified as [Ar(t-Bu)N]3TiOC(O)H, 2.§ Key
spectroscopic features are the formate proton resonance at
8.36 ppm and the OCO stretch observed by FTIR at 1685 cm21
.
The formation of formate 2 can be accounted for via initial
generation of a titanium-stabilized ketyl radical, followed by t-Bu
radical ejection to generate the observed product (see Scheme 1). A
similar sequence of events has been observed upon treatment of 1
with O2Mo(O–t-Bu)2.13 In both cases, the final product is the
result of formal displacement of CMe3 radical by 1.
The solid-state structure of formate 2 is shown in Fig. 1.14 The
molecule crystallizes on a crystallographic 3-fold axis, with the
result that the formate moiety is disordered over three positions.
The geometry at titanium is approximately tetrahedral with an
O(1)–Ti(1)–N(1) angle of 110.7(47)u. The observed Ti–O distance
˚
of 1.868(4) A is similar to that observed in other compounds
containing the tris-t-butylanilide ligand set.13,15 The three-fold
disorder, which we have not been able satisfactorily to resolve,
prevents discussion of additional metrical parameters.
{ Electronic supplementary information (ESI) available: synthetic proce-
dures for 2 and 3, details of the electronic structure calculations. See http://
dx.doi.org/10.1039/b504492h.
{ This communication is dedicated to the memory of Ian P. Rothwell.
*ccummins@mit.edu
Scheme 1 Synthesis of 2 and 3. R 5 t-Bu, Ar 5 3,5-C6H3Me2.
Chem. Commun., 2005, 3403–3405 | 3403
This journal is ß The Royal Society of Chemistry 2005

View Article Online
These comparisons suggest that the Ti–O interaction in 3 is best
viewed as a multiple bond.
To further address this issue, we have carried out DFT
calculations on the model compounds H(O)COTi(NH2)3 (2-m)
and [Li(OMe2)2][OTi(NH2)3] (3-m)." The key features of the
experimental structures are satisfactorily reproduced at the BP86
level of density functional theory.23 Inspection of the Laplacian of
the electron density24 indicates that, in both cases, the Ti–O bond
is characterized by considerable ionic character, as anticipated
based on the large difference in electronegativities between Ti and
O. On moving from 2-m to 3-m, however, the value of the electron
density at the bond critical point increases from 0.1134 to
0.1815 a.u., consistent with a substantial increase in bond order
and, thus, covalency. Additionally, a pair of degenerate orbitals
corresponding to orthogonal Ti–O p bonds is observed in the DFT
analysis. Since the electron localization function (ELF) has been
particularly valuable for illuminating complex issues of chemical
bond multiplicity,25 we provide in Fig. 3 an ELF isosurface plot
for 3-m.26 In the ELF, triple bonds give rise to a toroidal basin
surrounding the bond axis. In Fig. 3 it is seen that such a toroidal
basin is present and is shifted close to the O atom, nearly merging
with the oxygen lone pair basin. By all accounts, the titanoxide
anion of interest herein manifests a quite polar triple bond.
The chemistry presented herein provides precedent for bound
formate decarbonylation as triggered by deprotonation. The
existence of such a transformation may be significant with regard
to the reactions of CO and CO2 on metal surfaces.1–3
Furthermore, oxoanion 3 is likely to be a potent nucleophile as
Fig. 1 ORTEP drawing of compound 2 with ellipsoids at the 50%
probability level. The formate ligand is positionally disordered about the
˚
C3 axis. Selected interatomic distances (A) and angles (u): Ti(1)–O(1)
1.874(4), Ti(1)–N(1) 1.928(3), O(1)–C(21) 1.285(8), C(21)–O(2) 1.282(13),
O(1)–Ti(1)–N(1) 110.74(7), O(1)–C(21)–O(2) 114.3(7).
The solid-state structure of 3 (see Fig. 2), reveals the expected
coordination of the titanoxide anion to the Li cation, with a Li–O
distance of 1.801(6) A.20 The Li+ ion is coordinated additionally by
˚
two molecules of Et2O, adopting an overall trigonal planar
˚
geometry. The Ti–O distance, at 1.717(2) A, is significantly
contracted relative to 2. While anionic titanium(IV) oxos have not
been previously structurally characterized, typical Ti–O distances
6
in neutral titanyls range from 1.61 to 1.68 A. Andersen et al.
˚
have prepared the anionic oxotitanium(III) compound
[Cp*2TiOLi(THF)]2, which features an average Ti–O distance of
21
1.787 A, while Hoskin and Stephan’s oxozirconium(IV) anion,
˚
22
˚
[Cp*2Zr(H)OLi(THF)]2, exhibits a Zr–O distance of 1.847 A.
Fig. 2 ORTEP drawing of ion pair [Li(OEt2)2] 3 with ellipsoids at the
˚
50% probability level. Selected interatomic distances (A) and angles (u):
Ti(1)–O(1) 1.717(2), Ti(1)–N(1) 1.986(3), Ti(1)–N(2) 1.989(2), Ti(1)–N(3)
Fig. 3 A plot of the isosurface corresponding to ELF 5 0.84 for 3-m.
1.990(3), O(1)–Li(1) 1.801(6), Ti(1)–O(1)–Li(1) 167.8(2).
Color scheme: green, Ti; dark blue, N; light blue, C; red, O; white, H.
3404 | Chem. Commun., 2005, 3403–3405
This journal is ß The Royal Society of Chemistry 2005

View Article Online
13 J. C. Peters, A. R. Johnson, A. L. Odom, P. W. Wanandi, W. M. Davis
and C. C. Cummins, J. Am. Chem. Soc., 1996, 118, 10175.
well as an interesting metallo-ligand, prospects currently under
investigation. Finally, the valence-isoelectronic relationship of
[OTi(N[t-Bu]Ar)3]12 to the previously described neutral
[OV(N[t-Bu]Ar)3] and cationic [OMo(N[t-Bu]Ar)3]1+ complexes
reveals 3 to be the newest member of an intriguing series of
four-coordinate oxometal entities.27,28
14 Crystal data. C37H54N3O2Ti,
a 5 15.7365(7), b 5 15.7365(7), c 5 24.686(2) A, U 5 5294.1(6) A ,
M 5 620.73, rhombohedral,
3
˚
˚
T 5 193 K, space group R3, Z 5 6, m(Mo-Ka) 5 0.277 mm21, 4300
¯
reflections measured, 1423 unique (Rint 5 0.043) which were used in all
calculations. The final wR(F2) was 0.134 (all data). Due to the disorder,
the formate hydrogen atom was not included in the refinement. CCDC
267713. See http://dx.doi.org/10.1039/b504492h for crystallographic
data in CIF or other electronic format.
We are grateful to the National Science Foundation for their
support of this work (CHE-9988806 and a predoctoral fellowship
for A.M.).
15 T. Agapie, P. L. Diaconescu, D. J. Mindiola and C. C. Cummins,
Organometallics, 2002, 21, 1329.
16 H. A. Schwarz and R. W. Dodson, J. Phys. Chem., 1989, 93, 409.
17 P. S. Surdhar, S. P. Mezyk and D. A. Armstrong, J. Phys. Chem., 1989,
93, 336.
18 M. G. Fickes, A. L. Odom and C. C. Cummins, Chem. Comm., 1997,
1993.
Arjun Mendiratta, Joshua S. Figueroa and Christopher C. Cummins*
Department of Chemistry, Massachusetts Institute of Technology, 77
Massachusetts Ave, Room 2-227, Cambridge, MA 02139, USA.
E-mail: ccummins@mit.edu; Fax: +1 617 258 5700; Tel: +1 617 253 5332
19 J. A. Osborn, F. H. Jardine, J. F. Young and G. Wilkinson, J. Chem.
Soc. A, 1966, 1711.
Notes and references
20 Crystal data. C44H68LiN3O3Ti,
a 5 13.3553(10), b 5 8.3843(9), c 5 25.8835(18) A, U 5 4626.7(6) A ,
T 5 193 K, space group P212121, Z 5 4, m(Mo-Ka) 5 0.222 mm21
M 5 741.85, orthorhombic,
3
˚
˚
§ All manipulations were carried out under an atmosphere of dry nitrogen
using solvents purified by standard methods. The new compounds 2 and 3
gave satisfactory elemental analyses (C, H, and N).
,
30248 reflections measured, 8968 unique (Rint 5 0.039) which were used
in all calculations. The final wR(F2) was 0.1573 (all data). CCDC
267714. See http://dx.doi.org/10.1039/b504492h for crystallographic data
in CIF or other electronic format.
" Calculations were carried out using the ADF2004.0129 software package.
Visualization of the Laplacian and determination of bond critical points
were performed using the Xaim software package.30 The ELF isosurface
was generated using the DGRID software package.31
21 W. W. Lukens, P. T. Matsunaga and R. A. Andersen, Organometallics,
1998, 17, 5240.
22 A. J. Hoskin and D. W. Stephan, Organometallics, 1999, 18, 2479.
23 (a) A. D. Becke, Phys. Rev. A, 1988, 38, 3098; (b) J. P. Perdew, Phys.
Rev. B, 1986, 33, 8822.
1 D. H. Gibson, Coord. Chem. Rev., 1999, 186, 335.
2 I. Wender, Fuel Process. Technol., 1996, 48, 189.
3 Formates are also intermediates in the hydrogenation of CO2. See:
P. G. Jessop, F. Joo and C. C. Tai, Coord. Chem. Rev., 2004, 248, 2425.
4 P. R. Ortiz de Montellano, Cytochrome P-450: Structure, Mechanism
and Biochemistry, Plenum, New York, 1995.
24 R. F. W. Bader, Chem. Rev., 1991, 91, 893.
25 H. Grutzmacher and T. F. Fassler, Chem. Eur. J., 2000, 6, 2317.
26 (a) A. D. Becke and K. E. Edgecombe, J. Chem. Phys., 1990, 92, 5397;
(b) B. Silvi and A. Savin, Nature, 1994, 371, 683; (c) A. Savin, R. Nesper,
S. Wengert and T. F. Fassler, Angew. Chem., Int. Ed. Engl., 1997, 36,
1809.
5 G. Centi, E. Trefiro, J. R. Ebner and V. Franchetti, Chem. Rev., 1999,
88, 55.
6 W. A. Nugent and J. M. Mayer, Metal–Ligand Multiple Bonds, Wiley,
New York, 1988.
27 J. K. Brask, V. Dura-Vila, P. L. Diaconescu and C. C. Cummins, Chem.
Commun., 2002, 902.
7 M. J. Carney, P. J. Walsh, F. J. Hollander and R. G. Bergman, J. Am.
Chem. Soc., 1989, 111, 8751.
8 M. R. Smith, III, P. T. Matsunaga and R. A. Andersen, J. Am. Chem.
Soc., 1993, 115, 7049.
9 J. E. Hill, P. E. Fanwick and I. P. Rothwell, Inorg. Chem., 1989, 28,
3602.
10 R. S. Haymotherwell, G. Wilkinson, B. Hussainbates and
M. B. Hursthouse, Polyhedron, 1993, 12, 2009.
28 A. R. Johnson, W. M. Davis, C. C. Cummins, S. Serron, S. P. Nolan,
D. G. Musaev and K. Morokuma, J. Am. Chem. Soc., 1998, 120, 2071.
29 (a) G. te Velde, F. M. Bickelhaupt, S. J. A. van Gisbergen,
C. Fonseca Guerra, E. J. Baerends, J. G. Snijders and T. Ziegler,
J. Comput. Chem., 2001, 22, 931; (b) C. Fonseca Guerra, J. G. Snijders,
G. te Velde and E. J. Baerends, Theor. Chem. Acc., 1998, 99; (c)
ADF2004.01, SCM, Theoretical Chemistry, Vrije Universiteit,
Amsterdam, The Netherlands, http://www.scm.com.
11 P. Barthazy, M. Worle and A. Mezzeti, J. Am. Chem. Soc., 1999, 121,
480.
12 P. W. Wanandi, W. M. Davis, C. C. Cummins, M. A. Russell and
D. E. Wilcox, J. Am. Chem. Soc., 1995, 117, 2110.
30 J. C. Ortiz Alba and C. Bo, XAIM-1.0, Universitat Rovira i Virgili,
Tarragon, Spain, 1998, http://www.quimica.urv.es/XAIM.
31 M. Kohout, DGRID, edn. 2.4, Max Planck Institute for Chemical
Physics of Solids, Dresden, 2004.
This journal is ß The Royal Society of Chemistry 2005
Chem. Commun., 2005, 3403–3405 | 3405