Synthesis of mixed alkali-metal - zinc enolate complexes derived from 2,4,6-trimethylacetophenone: New inverse crown structures

Article abstract of DOI:10.1021/om060878k

The solution and solid-state characterization of two new mixed alkali-metal - zinc enolate compounds is reported. These compounds are prepared by reaction of the relevant mixed-metal base [MZn(HMDS)₃] (M = Na, K; HMDS = 1,1,1,3,3,3-hexamethyldisilazide) with a stoichiometric amount of the sterically demanding ketone 2,4,6-trimethylacetophenone. Thus, the new mixed-metal enolate compounds [Na₂-Zn₂{OC(=CH₂)Mes} ₆{OC(CH₃)Mes}₂] (2) and [K₂Zn ₂{OC(=CH₂)Mes}₆(CH₃Ph)₂] (3) are obtained for M = Na, K, respectively. X-ray crystallographic studies reveal that both compounds adopt the same structural motif, which define them as inverse crown complexes, a cationic eight-membered [(MOZnO)₂] ²⁺ ring which hosts in its core two additional enolate ligands. Each Zn center is bonded to four anionic enolate ligands framing the structure, whereas the alkali metals form much weaker interactions with the oxygen atoms and complete their coodination sphere by bonding to a neutral molecule, an unenolised ketone for M = Na or toluene for M = K.

Full text of DOI:10.1021/om060878k

204
Organometallics 2007, 26, 204-209
Synthesis of Mixed Alkali-Metal-Zinc Enolate Complexes Derived
from 2,4,6-Trimethylacetophenone: New Inverse Crown Structures
Sharon E. Baillie, Eva Hevia,* Alan R. Kennedy, and Robert E. Mulvey*
WestCHEM, Department of Pure and Applied Chemistry, UniVersity of Strathclyde,
Glasgow, Scotland G1 1XL, U.K.
ReceiVed September 25, 2006
The solution and solid-state characterization of two new mixed alkali-metal-zinc enolate compounds
is reported. These compounds are prepared by reaction of the relevant mixed-metal base [MZn(HMDS)3]
(M ) Na, K; HMDS ) 1,1,1,3,3,3-hexamethyldisilazide) with a stoichiometric amount of the sterically
demanding ketone 2,4,6-trimethylacetophenone. Thus, the new mixed-metal enolate compounds [Na2-
Zn2{OC(dCH2)Mes}6{OC(CH3)Mes}2] (2) and [K2Zn2{OC(dCH2)Mes}6(CH3Ph)2] (3) are obtained for
M ) Na, K, respectively. X-ray crystallographic studies reveal that both compounds adopt the same
structural motif, which define them as inverse crown complexes, a cationic eight-membered [(MOZnO)2]2+
ring which hosts in its core two additional enolate ligands. Each Zn center is bonded to four anionic
enolate ligands framing the structure, whereas the alkali metals form much weaker interactions with the
oxygen atoms and complete their coodination sphere by bonding to a neutral molecule, an unenolised
ketone for M ) Na or toluene for M ) K.
are the unprecedented 1,1′,3,3′-fourfold deprotonation of fer-
rocene, ruthenocene, and osmocene by the tris(diisopropylamide)
[NaMg(NiPr2)3]6 or the regioselective metalation of toluene at
the meta ring position, leaving the more acidic Me site intact,
effected by the mixed-metal alkyl-amido base [(TMEDA)-
NaMg(Bu)(TMP)2] (TMP ) 2,2,6,6-tetramethylpiperidide).7
Previously we reported the synthesis and characterization of
mixed-metal sodium-magnesium enolates derived from the
ketone 2,4,6-trimethylacetophenone (1).8 In the present paper
we extend this study to zinc by investigating the reactivity of
the alkali-metal amidozincates [MZn(HMDS)3] (M ) Na, K;
HMDS ) 1,1,1,3,3,3-hexamethyldisilazide)9 toward 1. This
ketone is highly sterically demanding; thus, the possibility of
competitive addition reactions is reduced, favoring deprotonation
reactions to form enolates. A further incentive for employing
this particular ketone is the excellent crystallization properties
that enolate derivatives can have, which makes them amenable
to X-ray crystallographic study.10
Introduction
Selective deprotonation of ketones to generate enolate anions
is a key reaction in synthesis. Enolates participate in a wide
variety of fundamental organic processes which involve C-C
bond formation such as alkylations, aldol additions, Michael
reactions, and acylations.1 To shed some light on the factors
that control the stereochemistries and selectivities of these
reactions, the characterization and isolation of these intermediate
metal enolates has become an important issue for synthetic
chemists.2 Most of the metal enolates employed in synthesis
are (homometallic) alkali-metal enolates, particularly lithium
enolates,3 which are generally easily accessible by reaction of
the relevant ketone with sterically hindered lithium amides. The
latter are strong bases and poor nucleophiles, minimizing the
possibility of competitive addition reactions to the carbonyl
group. Recently new s-group metal reagents, such as magnesium
bis(amides), have started to be used as alternative bases, showing
in certain cases selectivities greater than those of related lithium
amides.4 In our research group we have developed a new kind
of mixed-metal amide of empirical formula MM′(NR2)3 (M )
Li, Na, K; M′ ) Mg, Zn) which at least notionally combine
the reactivity of a group 1 metal base with the selectivity of a
Mg or Zn bis(amide).5 These bases often display a unique
synergic reactivity. Thus, some of the most remarkable examples
that illustrate this cooperative effect between the two metals
Zinc enolates are useful intermediates in organic synthesis,11
as they can participate in a rich variety of important reactions
such as addition to carbonyl compounds,12 transition-metal-
catalyzed reactions with carbon electrophiles,13 and reactions
with imines to generate â-lactams,14 to name but a few. One of
(6) Andrikopoulos, P. C.; Armstrong, D. R.; Clegg, W.; Gilfillan, C. J.;
Hevia, E.; Kennedy, A. R.; Mulvey, R. E.; O’Hara, C. T.; Parkinson, J. A.;
Tooke, D. M. J. Am. Chem. Soc. 2004, 126, 11612.
(1) Williard, P. G. In ComprehensiVe Organic Synthesis; Trost, B. M.,
Flemming, I. Eds.; Pergamon: Oxford, U.K., 1990; Vol. 1, Chapter 1.
(2) For an authoritative review on lithium enolate structures and their
influence on reactivity see: Seebach, D. Angew. Chem. Int. Ed. 1988, 27,
1624.
(3) See for example: (a) Hall, L. P.; Gilchrist, J. H.; Harrison, A. T.;
Fuller, D. J.; Collum, D. B. J. Am. Chem. Soc. 1991, 113, 9575. (b) Zhao,
P.; Condo, A.; Keresztes, I.; Collum, D. B. J. Am. Chem. Soc. 2004, 126,
3113.
(7) Andrikopoulos, P. C.; Armstrong, D. R.; Graham, D. V.; Hevia, E.;
Kennedy, A. R.; Mulvey, R. E.; O’Hara, C. T.; Talmard, C. Angew. Chem.,
Int. Ed. 2005, 44, 3459.
(8) Hevia, E.; Henderson, K. W.; Kennedy, A. R.; Mulvey, R. E.
Organometallics 2006, 25, 1778.
(9) The synergic metalation of toluene at the methyl position by the mixed
potassium-zinc amide base KZn(HMDS)3 has been the subject of a previous
communication: Clegg, W.; Forbes, G. C.; Kennedy, A. R.; Mulvey, R.
E.; Liddle, S. T. Chem. Commun. 2003, 406.
(4) (a) Zhang, M.; Eaton, P. E.; Angew. Chem. Int. Ed. 2002, 114, 2169.
(b) Bassindale, M. J.; Crawford, J. J.; Henderson, K. W.; Kerr, W. J.
Tetrahedron Lett. 2004, 45, 4175. (c) He, X.; Allan, J. F.; Noll, B. C.;
Kennedy, A. R.; Henderson, K. W. J. Am. Chem. Soc. 2005, 127, 6920.
(5) For a recent review on the reactivity and structures of these mixed-
metal reagents, see: Mulvey, R. E. Organometallics 2006, 25, 1060.
(10) He, X.; Noll, B. C.; Beatty, A.; Mulvey, R. E.; Henderson, K. W.
J. Am. Chem. Soc. 2004, 126, 7444.
(11) Nakamura, E. In Organometallics in Synthesis. A Manual, 2nd ed.;
Schlosser, M., Ed.; Wiley: Chichester, U.K., 2002; Chapter 5.
(12) Hansen, M. H.; Bartlett, P. A.; Heathcock, C. H. Organometallics
1987, 6, 2069.
10.1021/om060878k CCC: $37.00 © 2007 American Chemical Society
Publication on Web 12/05/2006

Mixed Alkali-Metal-Zinc Enolate Complexes
Organometallics, Vol. 26, No. 1, 2007 205
Scheme 1
the main advantages of using zinc enolates in synthesis is that
they possess a much higher functional group tolerance than is
the case for analogous alkali-metal enolates. However, only a
select few zinc enolate structures have been elucidated, as
illustrated by the fact that the first structure of a Reformatsky
reagent,15 one of the oldest and most versatile types of
organozinc reagent, an R-metalated ester (BrZnCH2COOR), was
elucidated as recently as 1984, nearly a century after the first
publication of this type of compound.16
Enolate ligands can adopt several coordination modes (O-
terminal, C-terminal, O-bridging or C,O-bridging) depending
on the metal they are coordinated to, the reaction solvent, or
the steric bulk of their organic substituents. For zinc there are
a few examples of ester enolates15 and amide enolates17 with
C,O-bridging structures, where the anionic charge of the enolate
is delocalized with the C-C double bond. On the other hand,
there are also some examples of O-bound compounds,18
including a series of enolates derived from simple ketones
reported in 2006 by Hagadorn.19
Herein we report the synthesis and characterization of the
mixed alkali-metal-zinc enolates [Na2Zn2{OC(dCH2)Mes}6-
{OC(CH3)Mes}2] (2) and [K2Zn2{OC(dCH2)Mes}6(CH3Ph)2]
(3), which are, to the best of our knowledge, the first reported
homoanionic mixed alkali-metal-zinc enolate compounds as
well as the first bimetallic enolates of this kind to contain zinc.20
reactions is the amine HMDS(H)), where each potassium is
bonded to a molecule of toluene, reflecting the higher affinity
of this heavier alkali metal to form metal-carbon bonds (and
metal-π-arene interactions in particular; vide infra) than metal-
oxygen bonds.21 In other words, in this system toluene is a better
donor for potassium than the unenolized neutral ketone mol-
ecule.
Results and Discussion
New metal enolates 2 and 3 were synthesized according to
the reactions shown in Scheme 1. In the first part of the
synthesis, the mixed-metal amides were prepared in situ by
combining the alkali-metal amide MHMDS with the zinc bis-
(amide) Zn(HMDS)2 in hydrocarbon media. The addition of 3
equiv of 1 afforded the new mixed-metal enolates 2 and 3 as
colorless crystalline solids in reasonably good isolated yields
(55-65%). For 2, the yield of the reaction is increased
dramatically (from 23% to 55%) when 4 molar equiv of 1 is
employed, as one unenolized neutral ketone molecule remains
coordinated to each sodium. The fact that compound 2 is
obtained even when a deficiency of 1 is employed implies that
the rate of the formation of 2 is faster than the rate of
deprotonation of 1 by the mixed-metal base NaZn(HMDS)3 to
form the unsolvated mixed-metal enolate “NaZn(OR)3” (R )
C(dCH2)Mes). It suggests also that once 2 is formed the
unenolized neutral molecules of 1 coordinated to sodium do
not react over the time window studied (2 h) with the excess of
base still remaining in the solution. For M ) K, even when the
reaction is carried out using an excess of the ketone, compound
3 is obtained as a single metal product (the byproduct of these
As alluded to earlier, we have reported the synthesis and
characterization of a series of mixed sodium-magnesium
enolates, prepared by reaction of the stoichiometric variant
trialkyl (NaMgBu3) and tetraalkyl complexes (Na2MgBu4)
toward the ketone 1.8 Following the same procedure, we have
now prepared the corresponding trialkyl zincate NaZnBu3, by
mixing together equimolar amounts of BuNa and Bu2Zn.
However, compound 2 could not be obtained when this all-
alkyl zincate was treated with 3 or 4 equiv of the ketone 1. The
1H NMR spectrum of the crude reaction solution revealed a
mixture of two distinct enolate ligands and significant amounts
of unreacted ketone, together with a butyl ligand. This experi-
ment illustrates the higher covalent and carbophilic character
of Zn versus Mg. Thus, while NaMgBu3 is able to react cleanly
at room temperature with 3 equiv of 1 to form the desired mixed-
metal tris(enolate) complex, NaZnBu3 fails to react in the same
manner (or at least to completion), instead giving a mixture of
products resulting from the incomplete conversion of the butyl
groups to enolate ligands.
The structures of compounds 2 and 3 were successfully
determined by X-ray crystallographic studies. Figure 1 shows
the complete molecular structure of 2, while Figure 2 highlights
its inorganic core. Table 1 gives key bond lengths and angles
for 2. The structure of 2 can be viewed as a cationic eight-
membered [(NaOZnO)2]2+ ring hosting in its interior two
additional enolate ligands. This ring adopts a chair conformation,
with the sodium atoms displaced on either side of the plane
defined by O1Zn1O3‚‚‚O1*Zn1*O3 (Figure 2). The distortion
of the sodium atoms from planarity can be quantified by the
(13) (a) Nakamura, E.; Kuwajima, I. Tetrahedron Lett. 1986, 27, 83. (b)
Nakamura, E.; Sekiya, K. Kuwajima, I. Tetrahedron Lett. 1987, 28, 337.
(c) Tamaru, Y.; Ochiai, H.; Nakamura, T.; Yoshida, Z. Angew. Chem., Int.
Ed. 1987, 26, 1157. (d) Sekiya, K.; Nakamura, E. Tetrahedron Lett. 1988,
29, 5155. (e) Kinoshita, N.; Kawabata, T.; Tsubaki, K.; Bando, M.; Fuji,
K. Tetrahedron 2006, 62, 1756.
(14) van der Steen, F. H.; Jastrzebski, J. T. B. H.; van Koten, G.
Tetrahedron Lett. 1986, 27, 83.
(15) Dekker, J.; Budzelaar, P. H. M.; Boersma, J.; van der Kerk, G. L.
M.; Spek, A. L. Organometallics 1984, 3, 1403.
(16) Reformatsky, S. N. Ber. Dtsch. Chem. Ges. 1887, 20, 1210.
(17) Hlavinka, M. L.; Hagadorn, J. R. Organometallics 2005, 24, 4116.
(18) van Vliet, M. R. P.; van Koten, G.; Buysingh, P.; Jastrzebski, T. B.
H.; Spek, A. L. Organometallics 1987, 6, 537.
(19) Hlavinka, M. L.; Hagadorn, J. R. Organometallics 2006, 25, 3501.
(20) We have recently reported the trapping and characterization of an
enolate anion as the result of a 1,6-nucleophilic addittion of a sodium
amidozincate to benzophenone; see: Hevia, E.; Honeyman, G. W.; Kennedy,
A. R.; Mulvey, R. E. J. Am. Chem. Soc. 2005, 127, 13106.
(21) The preference of potassium for carbon over oxygen coordination
has been recently highlighted in: Barnett, N. D. R.; Clegg, W.; Kennedy,
A. R.; Mulvey, R. E.; Weatherstone, S. Chem. Commun. 2005, 375.

206 Organometallics, Vol. 26, No. 1, 2007
Baillie et al.
Table 1. Key Bond Lengths (Å) and Bond Angles (deg)
within the Structures of 2 and 3
Compound 2a
Zn1-O3
Zn1-O1
Zn1-O2
Zn1-O2*
Na1-O4
1.9032(13)
1.9149(13)
2.0130(13)
2.0430(13)
2.2543(16)
Na1-O1
Na1-O3*
Na1-O2
Na1-C13
Na1C12
2.3468(15)
2.3501(15)
2.4574(15)
2.840(2)
2.941(3)
O3-Zn1-O1
O3-Zn1-O2
O1-Zn1-O2
O3-Zn1-O2*
O1-Zn1-O2*
O2-Zn1-O2*
O4-Na1-O1
O4-Na1-O3*
145.22(6)
114.26(5)
94.98(5)
94.80(5)
108.89(6)
80.43(5)
116.74(6)
105.75(6)
O1-Na1-O3*
O4-Na1-O2
O1-Na1-O2
O3*-Na1-O2
Zn1-O2-Zn1*
Zn1-O3-Na1*
Zn1-O1-Na1
128.51(6)
161.35(6)
74.11(5)
74.36(5)
99.57(5)
93.55(5)
94.68(5)
Compound 3b
Zn1-O2
Zn1-O1
Zn1-O3
Zn1-O3*
K1-O2*
1.9049(12)
1.9085(12)
2.0179(12)
2.0398(12)
2.6534(13)
K1-O1
K1-O3
K1-C24
K1-C23
2.7082(13)
2.9431(13)
3.1414(19)
3.167(2)
Figure 1. Molecular structure of 2 showing selected atom labeling,
with hydrogen atoms omitted for clarity.
O2-Zn1-O1
O2-Zn1-O3
O1-Zn1-O3
O2-Zn1-O3*
O1-Zn1-O3*
O3-Zn1-O3*
O2*-K1-O1
139.27(5)
115.24(5)
96.23(5)
99.48(5)
110.01(5)
82.80(5)
109.13(4)
O2*-K1-O3
O1-K1-O3
Zn1-O1-K1
Zn1-O3-K1
Zn1-O2-K1*
Zn1-O3-Zn1*
64.86(4)
62.16(4)
102.50(5)
92.31(4)
99.94(5)
97.20(5)
a The asterisk denotes the symmetry operation 2 - x, 1 - y, 1 - z.
b The asterisk denotes the symmetry operation -x, -y, 2 - z.
which indicates that in this compound the interactions between
the sodium atoms and the oxygens are approximately the same
for the anionic enolate ligands as for the neutral ketone. Williard
has reported the structure of a pinacolone-solvated sodium
pinacolate, finding a similar trend for the distinct Na-O bond
types.24 More recently, Henderson has described the synthesis
and characterization of a magnesium enolate derived from 1,
[Mg4{µ-OC(dCH2)Mes}6{OC(dCH2)Mes}2{OC(CH3)Mes}-
(CH3Ph)2], which exhibits a tetrameric OMg(µ-O)2Mg(µ-
O)2Mg(µ-O)2Mg(O) chain arrangement terminated at each end
by a neutral ketone molecule.25 This homometallic compound
also shows similar Mg-O distances for the enolate ligands and
the unenolized ketones, and it has been proposed by the authors
as a model for aldol addition reactions.
On the other hand, each sodium atom in 2 also π-engages
with the olefinic (CdCH2) carbons of one of the enolate ligands
encapsulated in the ring, as indicated by the short Na-C
distances (Na1-C12 ) 2.941(3) Å and Na1-C13 ) 2.840(3)
Å), in order to compensate for the comparative weakness of
the Na-O bonds in 2, as previously mentioned.
The Zn center in 2 is tetracoordinated in a distorted-tetrahedral
geometry, bonded to four oxygen atoms from the enolate
ligands. The Zn-O bond distances are significantly shorter for
the enolates that are part of the eight-membered ring (1.9032-
(13) and 1.9032(13) Å) than for the guest enolates (2.0130(13)
and 2.0430(13) Å), which can be can be attributed to the
different coordination numbers of both types of ligands; thus,
while the ring enolates connect one sodium and one zinc center,
the guest enolates bind to two zincs and one sodium. These
Zn-O(enolato) bond distances are shorter than those found in
other previously characterized zinc enolates. Thus, for example
in the enolate [Et4Zn4{O(OMe)Cd(H)N(tBu)Me}4], resulting
Figure 2. Inorganic core of 2 highlighting the chair conformation
of the [(NaOZnO)2] ring.
angle between the O1Na1O3* unit and the aforementioned
plane, 145.03(6)°. The sodium atoms achieve tetracoordination
by bonding to three enolate ligands, two of them being part of
the eight-membered ring, and the remaining enolate is one of
the guests of the ring (that is disposed syn to Na) and one
molecule of unenolized ketone, acting as a σ donor through the
oxygen atom. The Na-O(enolate) bond distances are modestly
shorter for the ligands that are part of the ring (average length
2.348 Å) than for the enolates hosted in the core of the ring
(2.4574(15) Å). These distances, although lying in a range
similar to those found for other Na-alkoxide compounds,22 are
moderately longer than those found for other sodium enolates
derived from ketone 1, such as in homometallic [Na2{OC(d
CH2)Mes}2(TMEDA)2] (2.225(3) Å)8 and heterometallic [Na2-
Mg2{OC(dCH2)Mes}6(TMEDA)2] (from 2.2901(18) to 2.3295-
(18) Å),8 which indicates that the Na-O(enolate) bonds in 2
are fairly weak, presumably as a result of the large steric
demands of the enolato R ligands.
A notable feature of 2 is the presence of a molecule of
unenolized ketone solvating each sodium atom. Although the
important role that solvated alkali-metal enolates may play in
aldol additions and related enolate reactions has been a matter
of much discussion,23 little definite structural information is
available on this type of solvated compound. The Na-O(ketone)
bond length in 2 is of the same order as or even slightly shorter
(2.2543(16) Å) than the lengths found for the enolate ligands,
(22) Bradley, D. C.; Mehrotra, R. C.; Rothwell, I. P.; Singh, A. Alkoxo
and Aryloxo DeriVatiVes of Metals, Academic Press: San Diego, CA, 2001.
(23) (a) Seebach, D.; Amstutz, R.; Dunitz, J. D. HelV. Chim. Acta 1981,
64, 2622. (b) Williard, P. G.; Liu, Q.-Y. J. Am. Chem. Soc. 1993, 115,
3380.
(24) Williard, P. G.; Carpenter, G. B. J. Am. Chem. Soc. 1986, 108, 462.
(25) Allan, J. F.; Henderson, K. W.; Kennedy, A. R.; Teat, S. L. Chem.
Commun. 2000, 1059.

Mixed Alkali-Metal-Zinc Enolate Complexes
Organometallics, Vol. 26, No. 1, 2007 207
Figure 3. ChemDraw representations of (a) [K2Ca2{OC(dCH2)Mes}6‚2THF] and (b) [Na2Mg2{OC(dCH2)Mes}6(TMEDA)2].
from the reaction of EtZnNiPr2 with an N,N-dialkylglycine
ester,26 where each enolate ligand links two zinc centers to give
rise to a tetrameric Zn4O4 cyclic arrangement, the Zn-O bond
lengths lie in the range 2.028(5)-2.076(6) Å. These facts show
that the enolate ligands in 2 are bonded to the zinc centers
through strong σ-covalent bonds while the sodium atoms interact
much more weakly with the anionic oxygens, forming additional
π-interactions with neighboring olefinic carbons.
The basic structural motif of 2 has been previously found
for the series of “inverse crown” complexes [{M1M2(Ni-
Pr2)2}2X2] (M1 ) Li, Na; M2 ) Mg; X ) OR, H),27 [{-
(TMEDA)MMg(Bu)2}2(OtBu)2] (M ) Na, K),21 and [K2Ca2-
{OC(dCH2)Mes}6‚2THF]10 but also extends to several other
homometallic and heterometallic systems.5 The aforementioned
potassium-calcium complex [K2Ca2{OC(dCH2)Mes}6‚2THF]
(Figure 3a) was the first example of a homoleptic inverse crown,
as enolate ligands occupy both host ring and guest sites, and it
is the closest precedent to 2, as it contains the same anionic
ligands. However, the main structural distinction is the reversal
Figure 4. Molecular structure of 3 showing selected atom labeling,
with hydrogen atoms omitted for clarity.
in the solvation of the metals. To explain, in this potassium/
covalent character exhibited by zinc in its bonding in comparison
calcium enolate, a molecule of THF is bonded to each divalent
to that of magnesium.
calcium center, whereas the potassium atoms are unsolvated
The potassium-zinc enolate 3 adopts a centrosymmetric
structure analogous to that of 2 (Figure 4) with a cationic
and form a series of π-stabilizing interactions with the neighbor-
ing phenyl rings and olefinic carbons from the enolate ligands.
This reversal in metal solvation can be rationalized by the larger
size and softer character of calcium in comparison to zinc.
[(KOZnO)2]2+ eight-membered ring hosting an additional two
enolate ligands. However, in this case, the alkali metal is
solvated by a different donor molecule. Thus, in 3 a molecule
of toluene binds to each potassium center through its π-system.28
The six K-C(toluene) distances are within a short range (from
3.307(2) to 3.449(2) Å), which shows that although individually
these interactions are weak, collectively they must be significant,
as revealed by the relative shortness of the K-C(centroid)
distance (3.081 Å).29 The potassium center also interacts with
three enolate ligands, two of them being part of the host ring,
whereas the third occupies a “guest” position. However, these
interactions are relatively weak, as indicated by a comparison
of the K-O bond distances found for 3 (2.6534(13), 2.7082-
(13), and 2.9431(13) Å) with those in the homometallic enolate
[K4{OC(dCH2)Mes}4(CH3Ph)4]10 (range 2.528(6)-2.706(6) Å;
mean 2.631 Å). The potassium centers in 3 complete their
coordination sphere by forming a series of π-contacts with the
olefinic carbons of the guest enolate ligands (K1-C24 )
3.1414(193) Å and K1-C23 ) 3.167(2) Å). As previously
discussed for compound 2, the structure of 3 is similar to that
of the potassium-calcium inverse crown [K2Ca2{OC(dCH2)-
Mes}6‚2THF].10 The K-O distances in the latter (2.633(2),
2.702(2), and 2.842(2) Å) are in the same range as those in 3;
As previously mentioned, the number of metal enolates that
have been structurally characterized is relatively scarce.2 To our
knowledge, the only examples of bimetallic homoanionic
enolates that combine one alkali metal and one divalent metal
are the K/Ca compound mentioned above10 and a series of Na/
Mg enolates obtained by reaction of either NaMgBu3 or Na2-
MgBu4 with variable amounts of 1, as for example [Na2Mg2{µ-
OC(dCH2)Mes}6(TMEDA)2] (Figure 3b).8 The latter compound
has a structural arrangement significantly different from that of
2. Thus, the sodium/magnesium enolate presents a linear
tetranuclear Na‚‚‚Mg‚‚‚Mg‚‚‚Na chain arrangement connected
by six enolate ligands. Thus, while both compounds possess a
{(OR)2M(µ-OR)2M(OR)2}2- ate core, in the sodium/magnesium
enolate, each sodium is coordinated to two terminal enolates,
positioned in the same plane as the Mg centers of the Mg(µ-
O)2Mg inner ring, giving rise to a linear structure; however, in
contrast in 2, each sodium is bonded to three enolate ligands
and is situated above and below the plane defined by OZnO‚
‚‚O*Zn*O* adopting the cyclic inverse crown structure. Given
the similar sizes of magnesium and zinc, this fundamental
difference in the structures of these mixed-metal enolates must
be at least in part attributed to electronic factors and the greater
(28) For examples of metal-π-arene interactions see: (a) Gokel, G. W.;
De Wall, S. L.; Meadows, E. S. Eur. J. Org. Chem. 2000, 2967. (b) Forbes,
G. C.; Kennedy, A. R.; Mulvey, R. E.; Roberts, B. A.; Rowlings, R. B.
Organometallics 2002, 21, 5115.
(29) This K-C(centroid) bond length is within the range found for other
compounds crystallographically characterized that contain a potassium center
solvated by a neutral molecule of toluene: Andrikopoulos, P. C.; Armstrong,
D. R.; Kennedy, A. R.; Mulvey, R. E.; O’Hara, C. T.; Rowlings, R. B.
Eur. J. Inorg. Chem. 2003, 3354.
(26) van der Steen, F. H.; Boersma, J.; Spek, A. L.; van Koten, G.
Organometallics 1991, 10, 2567.
(27) (a) Drewette, K. J.; Henderson, K. W.; Kennedy, A. R.; Mulvey,
R. E.; O’Hara, C. T.; Rowlings, R. B. Chem. Commun. 2002, 1176. (b)
Gallagher, D. J.; Henderson, K. W.; Kennedy, A. R.; O’Hara, C. T., Mulvey,
R. E. Rowlings, R. B. Chem. Commun. 2002, 376.

208 Organometallics, Vol. 26, No. 1, 2007
Baillie et al.
Table 2. Selected 13C NMR Chemical Shifts (δ in ppm) for
Table 3. Key Crystallographic and Refinement Parameters
for Compounds 2 and 3
Enolate Ligands
compd
δ(dCO) δ(dCH2)
∆δa
solvent
2
3
2
3
166.59
168.64
169.94
83.98
84.94
79.60
82.61 d5-pyridine
83.70 d5-pyridine
90.34 d6-benzene
empirical formula
Mr
cryst syst
space group
a/Å
C88H106Na2O8Zn2
1468.45
triclinic
P1h
11.0238(3)
12.6380(3)
15.7638(4)
80.211(1)
76.930(2)
68.384(1)
1979.71(9)
1
C80H946K2O6Zn2
1360.50
monoclinic
P21/c
11.0836(2)
20.0998(3)
16.3947(3)
90
101.590(1)
90
3577.91(11)
2
[Na2{µ-OC(dCH2)Mes}4-
(TMEDA)2]
[K2{OC(dCH2)Mes}4‚
4(toluene)]
171.23
74.58
96.65 d6-benzene
b/Å
c/Å
a ∆δ corresponds to the difference in chemical shift between the two
13C resonances arising from the carbon-carbon double bond, dCO and
dCH2.
R/deg
â/deg
γ/deg
V/Å3
Z
T/K
however, the potassium centers in this potassium-calcium
enolate form more stabilizing π-interactions involving not only
the olefinic carbons of the enolates but also the aromatic rings.
123(2)
9056
1.020
0.0395
123(2)
8168
1.017
0.0344
no. of indep rflns
goodness of fit
R1 (I > 2σ(I))
wR2
On the other hand, the Zn-O bond distances are nearly
identical with those found within compound 2, showing again
the strong σ-coordination of the enolates to the zinc centers,
which frame the inverse crown ate motif while the softer
potassium cations form much weaker interactions with the
oxygen atoms and adhere to the framework through additional
π-contacts. At this point it should be noted that this σ- and
π-bonding distinction for the alkali metal and its Zn (or Mg)
copartner has become a signature feature of inverse crown
complexes and other related mixed-metal compounds and clearly
plays a major part in the often special reactivity that such
“synergic” mixtures can display.5
0.0917
0.0801
characterized in solution and in the solid state. Their crystal
structures show a common motif, a cationic eight-membered
ring (MOZnO)2 hosting two enolate ligands, which defines them
as inverse crowns. These compounds have strong, anchoring
Zn-O bonds that frame an anionic ate structure, onto which
the alkali metals cling through weaker bonds to the anionic
oxygens and through ancillary π-bonds to olefinic carbon atoms.
Experimental Section
General Considerations. All reactions were performed under a
protective argon atmosphere using standard Schlenk techniques.
Hexane and toluene were dried by heating to reflux over sodium
benzophenone ketyl and distilled under nitrogen prior to use. ZnCl2
was purchased from Aldrich Chemicals as a 1 M solution in diethyl
Compounds 2 and 3 have also been characterized in solution
using NMR spectroscopy (see hte Experimental Section);
however, given the low solubility of these mixed-metal enolates
in arene solvents such as toluene or benzene, their H and 13C
1
31
ether. Butylsodium30 and Zn(HMDS)2 were prepared according
NMR spectra have had to be recorded in the more polar solvent
deuterated pyridine. As evidenced by these spectra, the solution
structures of 2 and 3 in this solvent are different from those in
the solid state. Thus, for both compounds, the solvating ligands
within the solid state (1 and toluene for 2 and 3, respectively)
appear free in solution, whereas only one single set of enolate
signals is observed. Furthermore, the chemical shifts of the
resonances attributed to the mesityl and the dCH2 moieties in
to literature methods. NMR spectra were recorded on a Bruker DPX
400 MHz spectrometer, operating at 400.13 MHz for 1H and 100.62
MHz for 13C. Elemental analyses of compounds 2 and 3 could not
be obtained, due to their instability.
X-ray Crystallography. Single-crystal diffraction data were
recorded by a Nonius Kappa CCD diffractometer using graphite-
monochromated Mo KR radiation (λ ) 0.710 73 Å). The structures
were refined by full-matrix least squares and against F2 to
convergence using the SHELXL-97 program.32 Specific crystal-
lographic data and refinement parameters are given in Table 3.
Synthesis of [Na2Zn2{OC(dCH2)Mes}6{OC(CH3)Mes}2] (2).
Zn(HMDS)2 (1.8 mL, 5 mmol) was added to a suspension of
NaHMDS (prepared in situ by reaction of BuNa (0.8 g, 5 mmol)
and HMDS(H) (1.05 mL, 5 mmol)) in hexane (15 mL), affording
a white suspension. 2,4,6-Trimethylacetophenone (3.34 mL, 20
mmol) was then added. The mixture was heated under reflux for
30 min. Toluene (5 mL) was introduced at this stage, and the
mixture was gently heated until all the white solid had dissolved,
affording a pale orange solution. Allowing this solution to cool
slowly to room temperature produced a crop of colorless crystals
(yield 2.27 g, 63%). 1H NMR (400 MHz, 25 °C, C5D5N): 6.84 (s,
12H, m-H Mes, enolate), 6.79 (s, 4H, m-H Mes, ketone), 4.90 (s,
broad, 6H, CdCHH′), 3.97 (s, broad, 6H, CdCHH′), 2.65 (s, 36H,
o-CH3, Mes, enolate), 2.44 (s, 6H, C(O)CH3, ketone), 2.26 (s, 18H,
p-CH3, Mes, enolate), 2.21 (s, 18H, o-CH3 and p-CH3, Mes, ketone).
13C{1H} NMR (100.63 MHz, 25 °C, C5D5N): 208.31 (CdO),
the H and 13C NMR spectra are nearly identical for 2 and 3.
1
This suggests that, in the presence of a strong coordinative
solvent such as pyridine, these mixed-metal compounds undergo
cleavage, probably to generate the solvent-separated ion pair
[M(pyridine)x]+[Zn(enolate)3]-, which would explain the ob-
servations mentioned above.
As recently described by Hagadorn, the difference between
the chemical shifts in the 13C NMR spectrum of the C-O and
dCH2 resonances can be considered as a good indication of
the degree of polarization of the CdC bond and the nucleo-
philicity of the enolate ligand.19 The ∆δ values of 2 and 3 (see
Table 2) are significantly lower than those obtained for the
analogous monometallic sodium and potassium enolates, con-
sistent with the reduced polarization of the M-enolate bonds
in the mixed-metal compounds and the greater covalent character
of zinc. On the other hand, the chemical shift differences
between 2 and 3 are very similar, showing that the alkali metal
in these compounds has little influence in solution, which
supports the idea that in pyridine solution 2 and 3 are present
as the solvent-separated ion pair compounds [M(pyridine)x]+-
[Zn(enolate)3]-.
166.59 (OCdCHH′), 143.61 (Cipso, Mes, enolate), 141.14 (Cipso
Mes, ketone), 138.77 (Cpara, Mes, ketone), 135.99 (Cortho, Mes,
,
(30) Schade, C.; Bauer, W.; Schleyer, P. v. R. J. Organomet. Chem.
1985, 295, C25.
(31) Bu¨rger, H.; Sawodny, W.; Wannagat, U. J. Organomet. Chem. 1965,
3, 113.
(32) Sheldrick, G. M. SHELXL-97, Program for Crystal Refinement;
University of Go¨ttingen, Go¨ttingen, Germany, 1997.
To conclude, two mixed alkali-metal-zinc enolate inverse
crowns derived from 2,4,6-trimethylacetophenone have been
synthesized using the mixed-metal tris(amides) [MZn(HMDS)3]
as the base. These first compounds of their type have been fully

Mixed Alkali-Metal-Zinc Enolate Complexes
Organometallics, Vol. 26, No. 1, 2007 209
enolate), 135.40 (Cpara, Mes, enolate), 132.98 (Cortho, Mes, ketone),
129.31 (Cmeta, Mes, ketone), 128.57 (Cmeta, Mes, enolate), 83.98
(OCdCHH′), 32.58 (C(O)CH3), 21.51 (p-CH3, enolate), 21.39 (p-
CH3, ketone), 20.98 (o-CH3, enolate), 19.54 (o-CH3, ketone).
Synthesis of [K2Zn2{OC(dCH2)Mes}6(CH3Ph)2] (3). K(H-
MDS) (1.0 g, 5 mmol) was suspended in hexane (15 mL). Zn-
(HMDS)2 (1.8 mL, 5 mmol) was added, and the mixture was stirred
for 30 min, affording a white suspension. 2,4,6-Trimethylacetophe-
none (2.5 mL, 15 mmol) was then introduced, and the mixture was
heated under reflux for 30 min. The solution changed from colorless
to pale orange, and most of the white solid remained suspended.
At this stage toluene (5 mL) was added and the mixture was gently
heated until all the solid had dissolved, affording an orange solution.
Cooling this solution slowly to room temperature produced a crop
NMR (100.63 MHz, 25 °C, C6D6): 168.64 (OCdCHH′), 145.50
(Cipso, Mes),140.12 (Cipso, CH3Ph), 137.35 (Cortho, Mes), 136.63
(Cpara, Mes), 131.42, 130.67 (Cortho and Cmeta, CH3Ph), 130.09 (Cmeta
,
Mes), 121.77 (Cpara, CH3Ph), 84.94 (OCdCHH′), 23.34 (CH3, CH3-
Ph), 23.03 (p-CH3, Mes), 22.61 (o-CH3, Mes).
Acknowledgment. We thank the EPSRC (Grant Award No.
GR/R81183/01), the EU (Marie Curie Fellowship to E.H.), the
Carnegie Trust (through a vacation scholarship to S.B.), and
the Royal Society/Leverhulme Trust (Fellowship to R.E.M.) for
generously sponsoring this research.
Supporting Information Available: CIF files giving crystal
data and figures giving NMR spectra for compounds 2 and 3. This
material is available free of charge via the Internet at http://
pubs.acs.org.
1
of colorless crystals (yield 1.85 g, 54%). H NMR (400 MHz, 25
°C, C5D5N): 7.32 (m, 2H, Hpara, CH3Ph), 7.21 (m, 8H, Hortho and
Hmeta, CH3Ph), 6.81 (s, 12H, m-H Mes), 4.91 (s, broad, 6H, Cd
CHH′), 3.94 (s, broad, 6H, CdCHH′), 2.69 (s, 36H, o-CH3, Mes),
2.44 (s, 6H, CH3, CH3Ph), 2.23 (s, 18H, p-CH3, Mes). 13C{1H}
OM060878K