Coordination chemistry of ene-1,1-diamines and a prototype "carbodicarbene"

Article abstract of DOI:10.1002/anie.200705798

(Chemical Equation Presented) Carbophilic Lewis acids can polarize a coordinated π-bond by a slippage mechanism. A series of stable ylid- or enolate gold complexes of ene-1,1-diamines not only emulate this property, but also reveal the exceptional donor capacity of such electron-rich olefin ligands. Moreover, the first metal complex of a tetraaminoallene is reported, which features a prototype "carbodicarbene" ligand bound to a transition-metal template.

Full text of DOI:10.1002/anie.200705798

Communications
DOI: 10.1002/anie.200705798
Ligand Design
Coordination Chemistry of Ene-1,1-diamines and a Prototype
“Carbodicarbene”**
Alois Fürstner,* Manuel Alcarazo, Richard Goddard, and Christian W. Lehmann
Dedicated to Professor Andreas Pfaltz on the occasion of his 60th birthday
Ene-1,1-diamines (ketene aminals, A) are a special class of
olefins distinguished by a very electron rich and strongly
polarized double bond.[1] This dipolar character is caused by
the significant contribution from the mesomeric extreme A’
to their ground-state structure (Scheme 1), and becomes
Scheme 2. Conditions and reagents: a) KH, Et2O, 87%; b) [AuCl-
(PPh3)], AgSbF6, THF, decomp; c) [AuCl(PPh3)], NaSbF6, THF, 89%.
Scheme 1. The two mesomeric extremes of a generic ene-1,1-diamine.
Structure of the well investigated imidazoline derivative 1.
olefin might reduce one (or both) of the noble-metal
components in the mixture, AgSbF6 was replaced by
particularly prominent if the nitrogen atoms are part of a
heterocyclic ring able to accommodate positive charge.
Likewise, coordination to a suitable (transition) metal
template enhances the inherent ylide character of compounds
such as 1[2,3] and its relatives. Surprisingly, however, few such
metal complexes of ene-1,1-diamines have been reported,[3,4]
and no attempts have been made to generalize the underlying
concept of charge separation. Outlined below are the results
of our initial foray into this promising territory, with emphasis
on the coordination chemistry of gold as a particularly
“carbophilic” and catalytically relevant Lewis acid.[5,6]
NaSbF6. This simple modification afforded the desired gold
complex 4 in the form of colorless crystals with excellent
yield. Its structure in the solid state confirms that the ligated
olefin behaves very much like a “carbon ylide”, as evident
from the end-on coordination to the metal center (Figure 1).
ꢀ
The Au C1 bond of 2.087(3) Š is even longer than the
corresponding single bond in [AuPh(PPh3)] (2.04 Š).[8] More-
over, the heterocyclic ring shows all the structural attributes
of an imidazolium cation.
Although compound 1 has been investigated in some
detail,[2,3,7] the parent 1,3-dimethyl-2-methyleneimidazoline
(3) itself has not been isolated in pure form. Gratifyingly, we
found that this particular compound is readily obtained as an
air-sensitive solid by deprotonation of imidazolium iodide 2
with KH in Et2O (Scheme 2). However, first attempts at
preparing the corresponding cationic gold complex through
reaction of 3 with [AuCl(PPh3)] and AgSbF6 in THF were
thwarted by the instantaneous formation of a black, intract-
able precipitate. On the assumption that the electron-rich
[*] Prof. A. Fürstner, Dr. M. Alcarazo, Dr. R. Goddard, Dr. C. W. Lehmann
Max-Planck-Institut für Kohlenforschung
45470 Mülheim an der Ruhr (Germany)
Fax: (+49)208-306-2994
Figure 1. Structure of the complex cation of 4 in the solidstate (the
SbF6 ion is removedfor clarity). [9]
ꢀ
E-mail: fuerstner@mpi-muelheim.mpg.de
[**] Financial support by the MPG, the Spanish Ministerio de Educación
y Ciencia (fellowship to M.A.), and the Fonds der Chemischen
Industrie is gratefully acknowledged. We thank Umicore AG & Co
KG, Hanau, for the generous gift of noble-metal salts.
Next, we prepared the corresponding rhodium complex 5,
as the IR stretching frequencies of the CO ligands in
compounds of type [RhCl(CO)2L] are commonly used to
assess the electronic properties of
a given ligand L
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
(Scheme 3).[10,11] According to this “rhodium scale”, even
3210
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 3210 –3214

Angewandte
Chemie
Figure 2. Structure of 7 in the solidstate. [9]
best described as a stable betaine. Likewise, the significant
torsion angle F = 38.38 is consistent with the notion that the
dipolar resonance form is structure determining.
Scheme 3. Preparation of the rhodium complex 5 andcomparison of
the donor capacity of its ene-1,1-diamine ligand with that of a cyclic
“carbodiphosphorane” and three different NHCs. The wavenumbers
(n˜, cmꢀ1) refer to the unsymmetrical stretching mode of the CO
ligands. Conditions and reagents: a) [{RhCl(CO)2}2], Et2O, RT, 59%.
Mes=mesityl=2,4,6-trimethylphenyl, R=iPr.
As a consequence, the corresponding gold adduct 8 can
not only be viewed as an ylide complex, but is more accurately
portrayed as a C-metalated noble-metal enolate.[17] The C1
ꢀ
ꢀ
C6 and the C6 C7 bond lengths fall within the range for
normal single bonds between sp2- and sp3-hybridized carbon
atoms (1.478(3) and 1.495(3) Š, respectively), whereas the
the most simple 2-methylene-imidazoline 3, outperforms the
standard N-heterocyclic carbenes (NHC) in terms of donor
capacity; these latter species are commonly employed wher-
ever electron-rich metal templates are desirable in catalytic
transformations.[12,13] The donor property of 3 is similar to that
of the much more elaborate cyclic carbodiphosphoranes
recently introduced by Kato, Baceiredo, and co-workers
(Scheme 3).[14]
ꢀ
C7 O1 bond (1.232(2) Š) corresponds to a normal carbonyl
group (Figure 3).[16] Notably, the aurated carbon atom in 8 is
tetrahedrally coordinated, and hence represents a chiral
center.
To probe the effect of functional groups, compound 7
bearing a ketone on the exocyclic double bond was targeted.
Not only is this derivative easy to prepare by deprotonation of
6[15] (Scheme 4), but any appreciable degree of charge
Figure 3. Structure of the goldenolate complex 8 in the solidstate. [9]
To further probe the enolate character, 7 was treated with
BF3(OEt2) in an attempt to address the oxygen terminus of
the ambidentate nucleophile with a “hard” Lewis acid. Again,
through X-ray analysis, the structure of the resulting O-
metalated adduct 9 was unambiguously confirmed (Figure 4).
Scheme 4. Conditions and reagents: a) KH, THF, RT, 79%; b) [AuCl-
(PPh3)], AgSbF6, THF, RT, 85%; c) BF3(OEt2), Et2O, RT, 92%.
separation should manifest itself in an enolate-like geometry
of the carbonyl subunit. The structure of 7 in the solid state
ꢀ
(Figure 2) nicely confirms this view, as the C4 C5 bond is
ꢀ
shorter than the formal C4 C1 “double” bond, whereas the
ꢀ
“carbonyl” bond length C5 O1 (1.2684(19) Š) clearly
exceeds that of a regular enone (for example, 1.24 Š in
chalcone).[16] As expected for a Hückel aromatic system, the
heterocyclic part is devoid of any significant bond alterations
at the symmetry-related positions. Compound 7 is therefore
Figure 4. Structure of the boron adduct 9 in the solidstate. [9]
Angew. Chem. Int. Ed. 2008, 47, 3210 –3214
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3211

Communications
Next, we investigated whether the imidazolium unit could
be replaced by other heterocyclic rings that are able to
stabilize positive charge, namely N-alkyl pyridinium cations.
The required exo-alkylidene derivatives 11 and 14[18] were
prepared by the same, exceedingly simple route (Scheme 5).
Figure 6. Structure of complex 12 in the solidstate. [9]
Scheme 5. Conditions and reagents: a) KH, Et2O, RT, 84%; b) [AuCl-
(PPh3)], AgSbF6, THF, RT, 68%; c) NaOH (1m), THF, RT, 92%.
Whereas 11 is a pale orange liquid, 14 was obtained in
multigram quantities as a bright yellow crystalline solid upon
treatment of 13 with aqueous NaOH. However, its structure
in the solid state (Figure 5) reveals that 14 is clearly less
Scheme 6. Slippage mechanism responsible for the activation of
p bonds by carbophilic Lewis acids and comparison with the structure
of two representative ylide-type gold complexes escorted by a non-
nucleophilic counterion. Nuꢀ =nucleophile.
such compounds have been known for a long time,[20] no
transition-metal complex has ever been reported. A recent
theoretical analysis of the bonding situation in tetraaminoal-
lenes suggests that the mesomeric form B’’ makes a substan-
tial contribution to their ground-state structure (Scheme 7).[21]
Figure 5. Structure of 14 in the solidstate. [9]
“enolate-like” than 7, as evident from the pattern of
alternating bond lengths and the very small torsion angle F
of only 4.28 (Scheme 5). As a result, 14 does not form a stable
gold complex under the chosen conditions, whereas the
arguably more electron-rich parent compound 11 reacts
smoothly to give 12. This adduct again shows all the features
of an ylide complex, in which the metal center is bound to the
ꢀ
terminal carbon atom of the olefin; judging from the Au C2
length, the contribution of the alternative h2-coordination
mode must be marginal, if at all (Figure 6).
Gold catalysis relies, to a large extent, on the propensity of
this soft Lewis acid to enhance the electrophilicity of a bound
alkene or alkyne. This property is thought to arise from
“slippage” of the metal template along the axis of the p bond
(Scheme 6).[19] Complexes 4 and 12 described herein nicely
manifest this putative h2!h1 deformation, and are hence of
fundamental relevance for a better understanding of the
structural basis of “p acidity” as the characteristic trait of gold
and related noble-metal catalysts.[5,6]
Scheme 7. Schematic representation of the bonding situation in tetra-
aminoallenes, to which the mesomeric extreme of a “carbodicarbene”
makes a substantial contribution.
In the extreme, tetraaminoallenes can be considered to
consist of a formally zerovalent central carbon atom, which
is endowed with two pairs of electrons and flanked by two
strongly donating diamino-stabilized carbene entities (“car-
bodicarbene”). The lone pairs of electrons reside in an orbital
The coordination chemistry becomes even more intrigu-
ing if two ene-1,1-diamine units are formally combined into
the heterocumulene motif of a tetraaminoallene B. Although
3212
www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 3210 –3214

Angewandte
Chemie
of p symmetry (highest occupied molecular orbital; HOMO),
largely centered on the central carbon atom, and in an
s orbital (HOMO-1), thus potentially rendering tetraami-
noallene derivatives neutral, yet very basic net four-electron
donors.[21] The analogy to the more abundant carbodiphos-
phoranes C and related cumulated ylides is clear.[14,22–25] At
first glance, one might mistake “carbodicarbenes” for singlet
carbenes of the NHC type; note, however, that the latter are
formally divalent carbon species, with a pair of electrons in a
s orbital as well as an orthogonal but empty p orbital.[12] In
striking contrast, the two respective orbitals of a “carbodi-
carbene” are both filled, containing a pair of electrons each.[21]
Compound 18 is a representative example of this partic-
ular class of allene derivatives and was conveniently prepared
from commercially available 15 by following a reported route
(Scheme 8).[20,26] Again, reaction with [AuCl(PPh3)]/AgSbF6
atom exhibits the expected trigonal planar coordination
geometry, with the in-plane lone pair of electrons of the
ligand binding to the gold template. Each of the lateral
diamino-stabilized “carbene” moieties is also planar, but they
are tilted relative to each other to relieve allylic strain.
Although complex 19 engages only one of the two lone
pairs of electrons proposed to reside on the central carbon
atom of the tetraaminoallene 18 in bonding to the transition-
metal center, it remains to be seen if a dimetalation of this
position can also be achieved.[24] To this end, it will be
interesting to extend this study to metals other than gold,
which was chosen for our preliminary investigation because of
its pronounced carbophilicity. Equally promising is the out-
look in structural terms, as different types of cumulated ylides
are known which might serve as ligands with unusual
structural characteristics and donor properties; compound
20 is an obvious candidate amongst the many conceivable
choices (Scheme 9).[27] Investigations along these lines are
Scheme 9. Conditions and reagents: a) 3, KH, THF, 72%.
being actively pursued by our research group and will be
reported in due course.
Received: December 18, 2007
Published online: March 17, 2008
Scheme 8. Conditions and reagents: a) DMA, CH2Cl2, reflux;
b) 1. Me2NH (neat); 2. sat. aq NaClO4, CH2Cl2, 89% (over 2 steps);
c) nBuLi, THF, RT, 76%; d) [AuCl(PPh3)], NaSbF6, THF, RT, 72%.
DMA=N,N-dimethylacetamide.
Keywords: allenes · carbenes · enamines · gold· ylides
.
[1] General reviews on alkene-1,1-diamines: a) W. Kantlehner in
Science of Synthesis Vol. 24 (Ed.: A. de Meijere), Thieme,
Stuttgart, 2006, pp. 571 – 705; b) P. A. Keller, J. Morgan in
Science of Synthesis Vol. 24 (Ed.: A. de Meijere), Thieme,
Stuttgart, 2006, pp. 707 – 746.
[2] N. Kuhn, A. Abu-Rayyan, A. Al-Sheikh, K. Eichele, C. Maichle-
Mößmer, M. Steimann, K. Sweidan, Z. Naturforsch. B 2005, 60,
294 – 299.
failed to afford a stable complex, but we were pleased to see
that the use of NaSbF6 as a halide scavenger worked nicely to
provide product 19 in 72% yield. Its remarkable structure in
the solid state is depicted in Figure 7. The metalated carbon
[3] a) N. Kuhn, H. Bohnen, J. Kreutzberg, D. Bläser, R. Boese, J.
Chem. Soc. Chem. Commun. 1993, 1136 – 1137; b) N. Kuhn, H.
Bohnen, G. Henkel, J. Kreutzberg, Z. Naturforsch. B 1996, 51,
1267 – 1278.
[4] a) P. P. Ponti, J. C. Baldwin, W. C. Kaska, Inorg. Chem. 1979, 18,
873 – 875; b) N. Kuhn, H. Bohnen, D. Bläser, R. Boese, Chem.
Ber. 1994, 127, 1405 – 1407; c) H. Schuhmann, M. Glanz, J.
Winterfeld, H. Hemling, N. Kuhn, H. Bohnen, D. Bläser, R.
Boese, J. Organomet. Chem. 1995, 493, C14 – C18.
[5] A. Fürstner, P. W. Davies, Angew. Chem. 2007, 119, 3478 – 3519;
Angew. Chem. Int. Ed. 2007, 46, 3410 – 3449.
[6] a) D. J. Gorin, F. D. Toste, Nature 2007, 446, 395 – 403; b) E.
JimØnez-Nfflæez, A. M. Echavarren, Chem. Commun. 2007, 333 –
346; c) A. S. K. Hashmi, G. J. Hutchings, Angew. Chem. 2006,
118, 8064 – 8105; Angew. Chem. Int. Ed. 2006, 45, 7896 – 7936.
[7] For pioneering studies on other well characterized ketene-N,X-
acetals, see a) H. Böhme, F. Soldan, Chem. Ber. 1961, 94, 3109 –
3119; b) A. I. Meyers, N. Nazarenko, J. Am. Chem. Soc. 1972, 94,
Figure 7. Structure of the gold“carbodicarbene” complex 19 in the
solidstate. [9]
Angew. Chem. Int. Ed. 2008, 47, 3210 –3214
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3213

Communications
3243 – 3245; c) H. Quast, L. Bieber, Angew. Chem. 1975, 87,
[15] Prepared by the method described in: L. A. M. Bastiaansen,
A. A. Macco, E. F. Godefroi, J. Chem. Soc. Chem. Commun.
1974, 36.
422 – 423; Angew. Chem. Int. Ed. Engl. 1975, 14, 428 – 429; d) C.
Reichardt, N. Kaufmann, Chem. Ber. 1985, 118, 3424 – 3427;
e) H. Quast, M. Ach, M. K. Kindermann, P. Rademacher, M.
Schindler, Chem. Ber. 1993, 126, 503 – 516; f) U. Gruseck, M.
Heuschmann, Chem. Ber. 1987, 120, 2053 – 2064.
[16] F. H. Allen, O. Kennard, D. G. Watson, L. Brammer, A. G.
Orpen, R. Taylor, J. Chem. Soc. Perkin Trans. 2 1987, S1 – S19.
[17] Gold enolates: a) Y. Ito, M. Inouye, M. Suginome, M. Mur-
akami, J. Organomet. Chem. 1988, 342, C41 – C44; b) M.
Murakami, M. Inouye, M. Suginome, Y. Ito, Bull. Chem. Soc.
Jpn. 1988, 61, 3649 – 3652; c) A. N. Nesmeyanov, É. G. Pereva-
lova, V. P. Dyadchenko, K. I. Grandberg, Izv. Akad. Nauk SSSR
Ser. Khim 1974, 2872.
[8] For a compilation of structural data, see E. J. Fernµndez, A.
Laguna, M. E. Olmos, Adv. Organomet. Chem. 2005, 52, 77 – 141.
[9] Anisotropic displacement parameters are drawn at the 50%
probability level. Hydrogen atoms have been omitted for clarity,
except those residing on the aurated carbon atoms.
CCDC 670968 (8), 670969 (14a), 670970 (7), 670971 (12),
670972 (4), 670973 (19), and 671017 (9) contain the supplemen-
tary crystallographic data for this paper. These data can be
obtained free of charge from The Cambridge Crystallographic
Data Centre via www.ccdc.cam.ac.uk/data_request/cif. Short
crystallographic summaries are found in the Supporting
Information.
[10] a) A. R. Chianese, X. Li, M. C. Janzen, J. W. Faller, R. H.
Crabtree, Organometallics 2003, 22, 1663 – 1667; b) W. A. Herr-
mann, J. Schütz, G. D. Frey, E. Herdtweck, Organometallics
2006, 25, 2437 – 2448.
[11] For a comparison of the different “scales” commonly used to
measure the donor properties of ligands, and a compilation of IR
data for complexes of type [RhCl(CO)2L], see: A. Fürstner, M.
Alcarazo, H. Krause, C. W. Lehmann, J. Am. Chem. Soc. 2007,
129, 12676 – 12677 and its Supporting Information.
[12] a) A. J. Arduengo, Acc. Chem. Res. 1999, 32, 913 – 921; b) D.
Bourissou, O. Guerret, F. P. Gabbaï, G. Bertrand, Chem. Rev.
2000, 100, 39 – 91; c) W. A. Herrmann, Angew. Chem. 2002, 114,
1342 – 1363; Angew. Chem. Int. Ed. 2002, 41, 1290 – 1309; d) F. E.
Hahn, Angew. Chem. 2006, 118, 1374 – 1378; Angew. Chem. Int.
Ed. 2006, 45, 1348 – 1352; e) N-Heterocyclic Carbenes in Syn-
thesis (Ed.: S. P. Nolan), Wiley-VCH, Weinheim, 2006; f) “N-
Heterocyclic Carbenes in Transition Metal Catalysis”: Topics in
Organometalic Chemistry, Vol. 21 (Ed.: F. Glorius), Springer,
Berlin, 2007; g) for an extensive literature overview, also see: D.
Kremzow, G. Seidel, C. W. Lehmann, A. Fürstner, Chem. Eur. J.
2005, 11, 1833 – 1853.
[13] This implies that 3 also constitutes a better electron donor than
trialkylphoshines, because N-heterocyclic carbenes are known to
be more strongly electron releasing than R3P. [12]
[14] S. Marrot, T. Kato, H. Gornitzka, A. Baceiredo, Angew. Chem.
2006, 118, 2660 – 2663; Angew. Chem. Int. Ed. 2006, 45, 2598 –
2601.
[18] R. M. Acheson, J. Woollard, J. Chem. Soc. Perkin Trans. 1 1975,
744 – 748.
[19] a) The concept of “slippage” was introduced in: O. Eisenstein,
R. Hoffmann, J. Am. Chem. Soc. 1981, 103, 4308 – 4320; b) for
recent Car–Parinello molecular dynamics calculations, see H. M.
Senn, P. E. Blöchl, A. Togni, J. Am. Chem. Soc. 2000, 122, 4098 –
4107; c) for an early structural manifestation of “slippage” in
complexes bearing enamine ligands, see T. C. T. Chang, B. M.
Foxman, M. Rosenblum, C. Stockman, J. Am. Chem. Soc. 1981,
103, 7361 – 7362; d) R. McCrindle, G. Ferguson, G. J. Arsenault,
A. J. McAlees, D. K. Stephenson, J. Chem. Res. Miniprint 1984,
3301 – 3344; e) S. L. Doran, D. B. Brown, C. W. Allen, J. Organo-
met. Chem. 1988, 340, 111 – 118.
[20] H. G. Viehe, Z. Janousek, R. Gompper, D. Lach, Angew. Chem.
1973, 85, 581 – 582; Angew. Chem. Int. Ed. Engl. 1973, 12, 566 –
567.
[21] R. Tonner, G. Frenking, Angew. Chem. 2007, 119, 8850 – 8853;
Angew. Chem. Int. Ed. 2007, 46, 8695 – 8698.
[22] R. Tonner, F. ꢀxler, B. Neumüller, W. Petz, G. Frenking, Angew.
Chem. 2006, 118, 8206 – 8211; Angew. Chem. Int. Ed. 2006, 45,
8038 – 8042.
[23] S. Braverman, M. Cherkinsky, M. L. Birsa in Science of Syn-
thesis, Vol. 18 (Ed.: J. G. Knight), Thieme, Stuttgart, 2005,
pp. 282 – 286.
[24] For a gem-dimetalated complex of a carbodiphosphorane, see
a) J. Vicente, A. R. Singhal, P. Jones, Organometallics 2002, 21,
5887 – 5900; b) I. Romeo, M. Bardaji, M. C. Gimeno, M. Laguna,
Polyhedron 2000, 19, 1837 – 1841.
[25] S. Pascual, M. Asay, O. Illa, T. Kato, G. Bertrand, N. Saffon-
Merceron, V. Branchadell, A. Baceiredo, Angew. Chem. 2007,
119, 9236 – 9238; Angew. Chem. Int. Ed. 2007, 46, 9078 – 9080.
[26] Z. Janousek, H. G. Viehe, Angew. Chem. 1971, 83, 615 – 616;
Angew. Chem. Int. Ed. Engl. 1971, 10, 574 – 575.
[27] For examples, see a) N. Kuhn, H. Bohnen, T. Kratz, G. Henkel,
Liebigs Ann. Chem. 1993, 1149 – 1151; b) N. Kuhn, M. Göhner,
M. Steimann, Z. Anorg. Allg. Chem. 2002, 628, 896 – 900.
3214
www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 3210 –3214