Dimethylaluminium enolates and alkoxides derived from trimethylaluminium and aromatic ketones: A synthetic, structural and theoretical investigation

Article abstract of DOI:10.1016/S0022-328X(00)00111-X

Reaction of Me₃Al with a series of aromatic ketones results in the precipitation of either dimethylaluminium enolates or alkoxides. In situ ¹H-NMR spectroscopic studies of the reaction between Me₃Al and acetophenone reveal a complex mixture of products whereas under the same conditions 2,4,6-trimethylacetophenone reacts cleanly to give the corresponding enolate. The enolate compounds [Me₂AlOC(2,4,6-Me₃-C₆H₂)=CH ₂] (2) and [Me₂AlOC(C₆Me₅)=CH₂] (4) were isolated and 2 as well as the representative alkoxide [Me₂AlOCMe₂Ph] (6) were characterised by X-ray crystallography. Both 2 and 6 form dimers with a central Al₂O₂ core. Ab initio molecular orbital calculations (HF/6-31G*) indicate that both 2 and 6 are the thermodynamic products of their reactions. For 2,4,6-trimethylacetophenone enolisation is preferred over alkylation by 4.70 kcal mol^-1 whereas for acetophenone alkylation is preferred by 25.39 kcal mol^-1 over enolisation. Disubstitution of the ortho positions on the aromatic ring by methyl groups results in the relative destabilisation of the alkoxide compared to the enolate due to steric crowding around the quaternary carbon atom.

Full text of DOI:10.1016/S0022-328X(00)00111-X

www.elsevier.nl/locate/jorganchem
Journal of Organometallic Chemistry 602 (2000) 15–23
Dimethylaluminium enolates and alkoxides derived from
trimethylaluminium and aromatic ketones: a synthetic, structural
and theoretical investigation
John F. Allan a, William Clegg b, Mark R.J. Elsegood b, Kenneth W. Henderson a,*,
Arlene E. McKeown a, Paul H. Moran a, Igor M. Rakov a
a Department of Pure and Applied Chemistry, Uni6ersity of Strathclyde, Glasgow, G1 1XL, UK
b Department of Chemistry, Uni6ersity of Newcastle upon Tyne, Newcastle, NE1 7RU, UK
Received 19 January 2000; received in revised form 9 February 2000
Abstract
Reaction of Me3Al with a series of aromatic ketones results in the precipitation of either dimethylaluminium enolates or
1
alkoxides. In situ H-NMR spectroscopic studies of the reaction between Me3Al and acetophenone reveal a complex mixture of
products whereas under the same conditions 2,4,6-trimethylacetophenone reacts cleanly to give the corresponding enolate. The
enolate compounds [Me2AlOC(2,4,6-Me3–C6H2)ꢀCH2] (2) and [Me2AlOC(C6Me5)ꢀCH2] (4) were isolated and 2 as well as the
representative alkoxide [Me2AlOCMe2Ph] (6) were characterised by X-ray crystallography. Both 2 and 6 form dimers with a
central Al2O2 core. Ab initio molecular orbital calculations (HF/6-31G*) indicate that both 2 and 6 are the thermodynamic
products of their reactions. For 2,4,6-trimethylacetophenone enolisation is preferred over alkylation by 4.70 kcal mol−1 whereas
for acetophenone alkylation is preferred by 25.39 kcal mol−1 over enolisation. Disubstitution of the ortho positions on the
aromatic ring by methyl groups results in the relative destabilisation of the alkoxide compared to the enolate due to steric
crowding around the quaternary carbon atom. © 2000 Elsevier Science S.A. All rights reserved.
Keywords: Enolates; Alkylations; Aluminium; Ab initio calculations; Crystal structure
pounds are important in the stereoselective formation
of carbon–carbon bonds and recently there has been
renewed interest in the chemistry of organoaluminiums
in combination with a second organometallic (hetero-
bimetallic species) to mediate a number of addition
reactions [2,3]. Rather surprisingly, there have been
relatively few studies into the mechanistic and struc-
tural aspects of enolisation reactions mediated by
organoaluminium compounds. In part, this is due to
the reported difficulty in the isolation and structural
identification of aluminium enolates [4]. Herein, we
report the reactions of Me3Al with a variety of aro-
matic ketones. In addition, the solid-state structures of
the enolate [Me2AlOC(2,4,6-Me3–C6H2)ꢀCH2] and the
alkoxide [Me2AlOCMe2Ph] have been elucidated and
will be discussed. Finally, an ab initio molecular orbital
computational study into the factors governing the
relative energies of enolisation and alkylation reactions
mediated by Me3Al is detailed [5].
1. Introduction
The reactions between trialkylaluminium compounds
(R3Al) and carbonyl containing species have been the
subject of intensive study over many years [1]. In sum-
mary, these reactions may result in adduct formation,
alkylation, reduction or enolisation, depending on the
nature of the ketone and also the alkyl group attached
to aluminium (Scheme 1). Moreover, aluminium com-
Scheme 1. Possible reactions between Et3Al and 2-propanone: (i)
adduct formation, (ii) alkylation, (iii) enolisation and (iv) reduction.
* Corresponding author. Tel.: +44-141-548 2351; fax: +44-141-
552 0876.
E-mail address: k.w.henderson@strath.ac.uk (K.W. Henderson)
0022-328X/00/$ - see front matter © 2000 Elsevier Science S.A. All rights reserved.
PII: S 0 0 2 2 - 3 2 8 X ( 0 0 ) 0 0 1 1 1 - X

16
J.F. Allan et al. / Journal of Organometallic Chemistry 602 (2000) 15–23
Table 1
Solids isolated from the reaction of Me3Al with various ketones
2. Results and discussion
are listed in Table 1. As expected [1], the most common
reaction is direct alkylation across the carbonyl, resulting
in the formation of dimethylaluminium alkoxides. How-
ever, the reactions involving 2,4,6-trimethylacetophe-
none (1) and pentamethylacetophenone (3) yielded the
enolate compounds [Me2AlOC(2,4,6-Me3–C6H2)ꢀCH2]
(2) and [Me2AlOC(C6Me5)ꢀCH2] (4), respectively, as the
sole products [6]. It is interesting to note that the reaction
with 2,4,6-trimethoxyacetophenone (13) also results in
alkylation even though its steric properties resemble
those of 1.
2.1. Reacti6ity studies
Each reaction was carried out in a similar fashion. The
ketone was added to a toluene solution of Me3Al cooled
to −78°C, followed by slow warming to ambient tem-
perature. The solution was then heated to reflux for 3 h,
cooled to ambient temperature and the solvent removed
in vacuo and replaced by hexane. Each solution was then
cooled to −28°C to precipitate the products and these

J.F. Allan et al. / Journal of Organometallic Chemistry 602 (2000) 15–23
17
EtAl(BHT)2+2ButCOMe
A series of in situ 1H-NMR spectroscopic experi-
ments were conducted on the reaction mixtures ob-
tained from the addition of either 1 or acetophenone (5)
to one equivalent of Me3Al in toluene solution. In both
reactions the reagents were mixed at −78°C, followed
by slow warming to ambient temperature and then
stirred for 1 h. Analysis of the reaction mixtures at this
stage showed significant differences between the reactiv-
ity of the ketones. The mixture involving 1 contained
only Me3Al and the ketone, indicating the formation of
an adduct, [Me3Al·OꢀCMe(2,4,6-Me3–C6H2)]. In con-
trast, a complex mixture was formed in the reaction
involving 5, however, the major species present was
“[EtAl(BHT){ButCꢀOCH2C(Me)(But)O}]+BHT–H
(1)
No enolates were identified even when equimolar
amounts of reactants were used, suggesting fast reac-
tion of the enolate with ketone to form the thermody-
namically stable aldolates. The ability to prepare and
isolate enolate 2 is likely due to the difficulty in forming
the self-coupled aldol product derived from the steri-
cally hindered ketone 11.
Since only very limited information is available on
the solid-state structures of aluminium enolates, an
X-ray crystallographic study of 2, and for comparison
the alkylated compound 6, was undertaken [4,9]. Fur-
thermore, analysis of the products from the enolisation
and alkylation reactions may aid in rationalising the
reactivity differences between 1 and 52.
determined
to
be
the
alkylated
product
[Me2AlOCMe2Ph] (6) and no characteristic olefinic–
CH2 signals from an enolate were present. Both reac-
tion mixtures were heated to reflux in toluene solution
for 2.5 h and re-analysed. The reaction of 1 showed
almost quantitative conversion to the enolate 2 which is
consistent with the high yield (93.3%) of pure solid
precipitated from solution using this ketone. After heat-
ing the reaction containing 5 the alkylated product 6
was again determined to be present as part of a mix-
ture. Analysis of the remaining species present is consis-
tent with the formation of self-coupled aldolate from
identification of the characteristic gem-CH2 signals. On
precipitation of solids from the reactions involving 5
only the alkylated product 6 was obtained, which is
most likely a result of the relative solubilities of the
species present. Previously, Barron and co-workers
found that the bulky alkoxide EtAl(BHT)2 (BHT–H is
butylated hydroxytoluene) reacts with a variety of ke-
tones to yield aldol addition products (Eq. (1)) [7].
2.2. X-ray crystallography
To our knowledge, the only aluminium enolate to be
structurally characterised is the a-amino ester enolate
[{Me2Al(MeO)OCꢀ(H)CNMe2}2] (19) and its complex
with Me2AlCl [Me2Al(MeO)OCꢀ(H)CNMe2·Me2AlCl]
(20) [4]. These enolates were prepared by transmetalla-
tion of the corresponding lithium enolates with
Me2AlCl. Compound 19 crystallises as an oxygen-
bridged dimer, with each metal chelated by a dimethy-
lamido unit, resulting in two five-coordinate aluminium
centres. Complex 20 adopts a dinuclear geometry,
where the Me2AlCl unit binds to the enolate oxygen,
and internal chelation by the dimethylamido unit to the
remaining metal gives two four-coordinate aluminium
centres. Although a few simple aluminium enolates
have been characterised by spectroscopic techniques,
determination of their solid-state structures has proved
elusive [11].
As can been seen from Fig. 1, compound 2 adopts
dimeric aggregation in the solid-state by bridging
through the enolate oxygens. Each metal is four-coordi-
nate by binding to two methyl groups and two oxygens,
with the average of the angles at the metal being 108.1°,
which range between 78.87(7)° for O–Al–O, to
1
121.56(10)° for C(1)–Al–C(2). Both the H- and 13C-
NMR spectra of 2 in C6D6 solution are consistent with
symmetrical dimeric association.
In general, 2 and 19 display similar geometric
parameters (key bond lengths and angles are given in
Table 2). However, notable exceptions are the Al–O
1 Self-coupled aldolates of methyl ketones can be prepared through
the use of magnesium amides [8].
2 For information on the related structural studies of lithium
enolates, see Ref. [10].
Fig. 1. Molecular structure of [{Me2AlOC(2,4,6-Me3C6H2)ꢀCH2}2]
(2).

18
J.F. Allan et al. / Journal of Organometallic Chemistry 602 (2000) 15–23
CꢀC bond lengths in 2 conform to those expected for
such species [4,12].
,
The separation of 3.292 A between C(4) of the alkene
group in 2 and the metal appears too long to involve
any significant bonding interaction [13]. Such interac-
tions have been found for a variety of metal enolates
but are usually only present in coordinatively unsatu-
rated species [14], for example, in lithium pinacolate,
[{LiO(But)CꢀCH2}6], where the distance between the
lithium atoms and the terminal carbon of the olefins
,
range between 2.3 and 2.6 A [15]. In 2 the olefin bond
sits almost in the same plane as the dimeric Al2O2 ring,
,
with C(3) and C(4) sitting 0.064 and 0.310 A, respec-
tively out of the plane; the torsion angle Al–O–C(3)–
C(4) is 23.3°. Similarly, the ipso-carbon of the aromatic
group is also close to the Al2O2 ring plane, with the
torsion angle Al(1%)–O(1)–C(3)–C(5) being 19.5°, and
Fig. 2. Molecular structure of [Me2AlOCMe2Ph] (6).
,
the Al%–C(5) distance is 3.216 A.
For comparison, the crystal structure of the alkyla-
tion product 6 was determined and is shown in Fig. 2
(key bond lengths and angles are given in Table 3). In
contrast to the scant information on aluminium eno-
lates, the structural chemistry of dialkylaluminium
alkoxides is relatively well established3. Again the ex-
pected dimeric structural motif is seen which is consis-
Table 2
,
Key bond lengths (A) and angles (°) for [{Me2AlOC(2,4,6-Me3–
C6H2)ꢀCH2}2] (2) a
Al(1)–O(1)
Al(1)–C(1)
O(1)–C(3)
C(3)–C(5)
1.8720(15)
1.945(2)
1.386(2)
1.488(3)
Al(1)–O(1%)
Al(1)–C(2)
C(3)–C(4)
1.8638(14)
1.947(2)
1.312(3)
1
tent with the H- and 13C-NMR spectroscopic solution
O(1)–Al(1)–O(1%)
78.87(7)
C(1)–Al(1)–O(1%)
C(2)–Al(1)–O(1%)
C(1)–Al(1)–C(2)
Al(1)–O(1)–C(3)
O(1)–C(3)–C(4)
O(1)–C(3)–C(5)
115.00(8)
111.79(8)
121.56(10)
128.56(12)
121.2(2)
data. The two independent Al–O distances in the dimer
O(1)–Al(1)–C(1) 112.89(8)
O(1)–Al(1)–C(2) 108.68(8)
C(3)–O(1)–Al(1%) 130.20(12)
Al(1)–O(1)–Al(1%) 101.13(6)
,
are of similar length at 1.8445(5) and 1.8464(5) A, these
,
are shorter by ꢀ0.02 A than those in 2 (typically such
4
,
Al–O distances are approximately 1.87 A) .
C(4)–C(3)–C(5)
125.3(2)
113.46(15)
Internal Al2O2 ring angles are smaller at the metals
and larger at the oxygens, which is consistent with sp2
hybridisation for the bridging oxygens. Both methyl
groups and the ipso-carbon of the aromatic ring are
staggered with respect to the central Al2O2 ring, with
torsion angles of 39.0, 82.5 and 159.1° for Al–O–C(3)–
a Symmetry operator for primed atoms: 1−x, 1−y, 1−z.
Table 3
Key bond lengths (A) and angles (°) for [Me2AlOCMe2Ph] (6)
,
C(6),
respectively.
Al–O–C(3)–C4
and
Al–O–C(3)–C(5),
Al(1)–O(1)
Al(1)–C(2)
O(1)–C(3)
C(3)–C(5)
1.8445(5)
1.9415(9)
1.4608(9)
1.5233(12)
Al(1)–O(1%)
Al(1)–C(1)
C(3)–C(4)
C(3)–C(6)
1.8464(5)
1.9456(9)
1.5163(13)
1.5252(11)
2.3. Theoretical calculations
O(1)–Al(1)–O(1%)
80.28(2)
O(1)–Al(1)–C(2)
O(1)–Al(1)–C(1)
C(1)–Al(1)–C(2)
Al(1%)–O(1)–C(3)
O(1)–C(3)–C(4)
C(4)–C(3)–C(5)
C(4)–C(3)–C(6)
113.94(3)
113.61(4)
117.68(4)
130.16(4)
107.78(7)
109.89(8)
112.86(7)
C(2)–Al(1)–O(1%) 112.70(3)
C(1)–Al(1)–O(1%) 112.93(3)
Al(1)–O(1)–C(3) 129.17(4)
Al(1)–O(1)–Al(1%) 99.72(2)
O(1)–C(3)–C(5)
O(1)–C(3)–C(6)
C(5)–C(3)–C(6)
An ab initio molecular orbital computational study
(HF/6-31G*) was undertaken to investigate the ener-
getics of the enolisation and alkylation reactions be-
tween Me3Al and ketones 1 and 55. Scheme 2 shows the
alkylation and enolisation reactions for the ketones and
the relative energies involved for each transformation
(dimeric aggregation is assumed for the organometallic
species calculated)6. Fig. 3 shows the optimised geo-
107.60(6)
106.59(6)
111.86(7)
bond distances. The two independent Al–O bond
,
lengths in 2 are similar at 1.8638(14) and 1.8720(15) A,
whereas in 19 one short and one long Al–O distance of
1.853(3) and 2.055(3) A are found. This feature in 19 is
a consequence of steric repulsions induced by neigh-
bouring methoxy and dimethylamido units adjacent to
the dimethylaluminium centres. The Al–C, O–C and
3 For related structures, see Ref. [16].
,
4 As found from a search of the Cambridge Structural Database
[17].
5 For other ab initio studies on aluminium systems, see Ref. [18].
6 For reviews on aluminium structural chemistry, see Ref. [19].

J.F. Allan et al. / Journal of Organometallic Chemistry 602 (2000) 15–23
19
for ketone V. Therefore, the calculations indicate that
both III and VII, representing 2 and 6, are the thermo-
dynamic products of their respective reactions, i.e. the
outcome of the reaction (alkylation or enolisation) is
not necessarily controlled by kinetic factors.
Significantly, there is also a large difference in the
relative energies within each type of reaction for the
ketones. While the energy difference between the enoli-
sation reactions is only 6.27 kcal mol−1, the alkylation
reactions differ by 23.82 kcal mol−1. The large differ-
ence in the relative energies of the alkylation reactions
can be attributed to the increased steric bulk of the
alkoxide in IV compared to that in VII. Steric crowding
around the quaternary carbon (OCMe2Ar) is signifi-
cantly greater in IV compared to VII, as a consequence
of the methyl groups in the 2- and 6-positions of the
aromatic ring. The difference between the energies of
the enolisation reactions can mainly be attributed to the
higher Bro¨nsted acidity of II compared to V. The
energy required to deprotonate ketones II and V to
form the corresponding enolate anions [2,4,6-Me3–
C6H2C(O)ꢀCH2]− (IX) and [PhC(O)ꢀCH2]− (X) was
Scheme 2. Energetics of enolisation and alkylation reactions for
ketones 1 and 5. Energies are quoted in kcal mol−1 and Mes=2,4,6-
Me3–C6H2.
metries for the alkoxides and enolates III, IV, VI and
VII. The geometric parameters within III and VII com-
pare favourably with the experimentally determined
crystal structures of 2 and 6 [18].
Both the alkylation and enolisation reactions are
highly exothermic for each ketone. However, the enoli-
sation route is preferred by 4.70 kcal mol−1 for ketone
II, whereas alkylation is preferred by 25.39 kcal mol−1
,
Fig. 3. Geometry optimised structures of the dimethylaluminium alkoxides and enolates. Important bond lengths (A) and angles (°) are included.
Hydrogen atoms, except those of the enolate CH2, are omitted for clarity.

20
J.F. Allan et al. / Journal of Organometallic Chemistry 602 (2000) 15–23
calculated to be 389.06 and 392.98 kcal mol−1, respec-
tively, i.e. methylation of the aromatic ring increases
the acidity of the ketone.
Comparing the dimensions between the alkoxides IV
and VII reveals slightly longer bond lengths associated
with the atoms in close proximity to the aromatic ring,
i.e. the difference between the C–C bonds joining the
available ketone, as has been seen previously for the
reactions of EtAl(BHT)2 [7]. In contrast, alkylation is
suppressed in the reactions of Me3Al with 1 and 3
which yield the enolate products 2 and 4, respectively.
No trace of aldolate was formed in these reactions
which is a consequence of the steric bulk of the anion.
The calculations confirmed that the enolate and alkox-
ide compounds 2 and 6, formed in the reactions using 1
and 5, respectively, are indeed the thermodynamic
products of their reactions. Although the electronics of
the ligand may play a role, it appears that steric influ-
ences are the major contributing factor in determining
the outcome of the reaction between the ketones and
Me3Al. It also appears from the reactions of 9 and 11
that substitution by methyl groups on both ortho posi-
tions of the aromatic ring is necessary to lead to
significantly different reactivity for these ketones.
Finally, the 2,4,6-trimethoxy substituted ketone 13
reacts with Me3Al to give the alkylated product 14. The
reactivity of 13 may be due to the Lewis base nature of
the attached methoxy substituents which in conjunction
with the carbonyl initially coordinate to the aluminium
of Me3Al to give a pentacoordinate metal geometry,
and direct the transformation in favour of alkylation
[16b,20].
,
aromatic ring and the quaternary carbon is 0.027 A,
,
and between the C–O bonds is 0.011 A. Other dimen-
sions within the structures, including the bond angles,
are similar. Almost no differences in dimensions are
discernible between the enolates III and VI.
The bond lengths associated with the a-carbon (OC)
decrease significantly comparing the alkoxides and the
enolates. As expected the formation of the olefinic bond
shortens one C–C length. However, the O–C and the
C–C(ipso) bonds also shorten significantly. It is perti-
nent to note that this contraction is greater for the
,
more substituted anion, i.e. O–C shortens by 0.078 A
,
for III and by 0.064 A for VI, and C–C(ipso) shortens
,
,
by 0.063 A for III and by 0.045 A for VI. These results
support the assessment that steric crowding induced by
the ortho methyl groups is an important factor in
determining the stability of complexes III and IV.
The olefinic group of the enolate in III sits almost
perfectly in the same plane as the dimeric ring (with a
torsion angle of 0.0° for Al–O–C–C between the metal
centre and the enolate group). No local minimum was
found for a geometry where the enolate ligands are
orthogonal to the Al2O2 plane. Placing the enolates in a
pseudo-syn orientation with respect to each other re-
sulted in the location of an energy minimum, XI, with
essentially the same absolute energy as III (with a
torsion angle of 19.7° for Al–O–C–C between the
metal centre and the enolate group). No minimum was
located for an ‘in-plane’ orientation for VI. In this case
the ligands rotate to adopt staggered conformations
with respect to the Al2O2 ring (with a torsion angle of
59.5° for Al–O–C–C between the metal centre and the
enolate group). In combination, these results suggest
that any agostic interactions between the enolate anion
and the metal are small, if present at all, and that the
conformation adopted by the ligands is mainly gov-
erned by steric factors.
4. Experimental
4.1. General experimental conditions
All syntheses were conducted in Schlenk-type glass-
ware under a blanket of argon gas. Glassware was dried
in an oven overnight and flame-dried under vacuum
before use. All metallated compounds isolated are
highly air- and moisture-sensitive, and were handled in
an argon-filled glove box fitted with a recirculating
column [21]. Ketones were distilled over CaH2 and
,
stored over 4 A molecular sieves before use. All sol-
vents were distilled over sodium/benzophenone prior to
use and used directly from the still. Trimethylalu-
minium was supplied as a 2 M solution in toluene from
Aldrich and used as received. NMR spectroscopy was
run on a Bruker AMX 400 MHz spectrometer at 300 K
in C6D6 solutions unless otherwise stated.
4.2. Synthesis of the complexes
3. Conclusions
Each compound was prepared in a similar manner.
The synthesis of [Me2AlOC(2,4,6-Me3–C6H2)ꢀCH2] (2)
is given as an example. In most instances the products
were highly soluble in hydrocarbon solvents, leading to
low yields of isolated solid precipitates. The samples
using ketones 9, 11 and 15 were contaminated by a
small quantity (B5%) of the corresponding enolate.
Purification of the mixtures was achieved through frac-
tional crystallisation.
The most common reaction between the aryl ketones
and Me3Al is alkylation across the carbonyl to give
dimethylaluminium alkoxides. The 1H-NMR spectro-
scopic studies of the reaction between Me3Al and 5
reveal that a complex mixture of products are formed
and although alkylation is the major reaction, enolisa-
tion is a competitive side reaction. Formation of aldo-
late indicates that any enolate formed reacts with

J.F. Allan et al. / Journal of Organometallic Chemistry 602 (2000) 15–23
21
4.2.1. [Me2AlOC(2,4,6-Me3–C6H2)ꢀCH2] (2)
4.2.5. [Me2AlOC(2-Me–C6H4)Me2] (10)
1
2,4,6-Trimethylacetophenone (1.46 g, 0.009 mol) was
added dropwise to a −78°C cooled solution of Me3Al
(0.011 mol) in 12 ml of toluene. The reaction mixture
was allowed to warm slowly to ambient temperature
with constant stirring and subsequently heated to
reflux for 3 h. Solvents were removed in vacuo, the
residue dissolved in 5 ml of hexane and a crystalline
product was obtained on cooling the mixture to −
28°C for 24 h. Yield, 93.3%; m.p., 131–132°C. 1H-
NMR spectrum, l −0.51 (s, 6H, AlCH3), 2.02 (s, 3H,
Yield, 26.3%; m.p., 158–160°C. H-NMR spectrum,
l −0.68 (s, 6H, AlCH3), 1.61 (s, 6H, OCCH3), 2.52
(s, 3H, o-CH3), 6.88 (m, 3H, m- and p-H, C6H4), 7.19
3
(d, J=7.7 Hz, 1H, o-H, C6H40). 13C-NMR spectrum,
l −5.30 (AlCH3), 23.50 (C6H4–CH3), 31.30 (OC–
CH3), 78.92 (OCCH3), 126.40, 126.79, 129.24, 134.01,
138.64 and 142.19 (o-, m-, p- and i-C, C6H4).
4.2.6. [Me2AlOC(2,4-Me2-C6H3)Me2] (12)
1
Yield, 50.9%; m.p., 118–120°C. H-NMR spectrum,
2
p-CH3), 2.33 (s, 6H, o-CH3), 4.04 (d, J=2.3 Hz, 1H,
l −0.57 (s, 6H, AlCH3), 1.70 (s, 6H, OCCH3), 2.03
(s, 3H, o-CH3), 2.60 (s, 3H, p-CH3), 6.78 (d, J=9.4
3
CꢀCH2), 4.80 (d, 1H, CꢀCH2), 6.68 (s, 2H, m-H,
C6H2). 13C-NMR spectrum, l −9.05 (AlCH3), 20.90
(o, m-CH3), 21.66 (p-CH3), 96.35 (CꢀCH2), 129.28
(m-C, C6H2), 153.68 (OCꢀCH2), 133.72, 137.65 and
139.78 (i-, o- and p-C, C6H2). X-ray quality crystals
were prepared by sublimation of the sample under
high vacuum (B0.1 Torr) using a 2 cm diameter
Schlenk tube. A temperature gradient between 5 and
82°C was established along 20 cm of the tube by
means of a copper jacket fitted with a water cooling
element at one end and immersing the other end in a
heated oil bath.
Hz, 1H, m-H, C6H4), 6.85 (s, 1H, o%-H, C6H4), 7.18
(d, 1H, o-H, C6H4). 13C-NMR spectrum, l −5.32
(AlCH3), 21.28 (o-CH3), 23.44 (p-CH3), 31.45
(OCCH3), 78.78 (OCCH3), 126.89, 127.02, 134.81,
138.50, 138.64, 139.47 (o-, m-, p- and i-C, C6H4).
4.2.7. [Me2AlOC(2,4,6-OMe–C6H2)Me2]
1
Yield, 40.1%; m.p., 138–140°C. H-NMR spectrum,
l −0.47 (s, 6H, AlCH3), 2.14 (s, 6H, OCCH3), 3.34
(s, 3H, p-OCH3), 3.40 (s, 6H, o-OCH3, C6H4), 6.05 (s,
2H, m-H, C6H2). 13C-NMR spectrum,
l −5.82
(AlCH3), 33.60 (C–CH3), 54.95 (p-OCH3), 55.22 (o-
OCH3), 78.15 (C–CH3), 92.44 (m-C, C6H2), 113.70
(i-C, C6H2), 161.21 and 161.50 (o- and p-C, C6H2).
4.2.2. [Me2AlOC(C6Me5)ꢀCH2] (4)
Yield, 60.0%; m.p., 210–214 °C with decomposition.
1H-NMR spectrum, l −0.54 (s, 6H, AlCH3), 1.97 (s,
3H, p-CH3), 2.02 (s, 6H, m-CH3), 2.35 (s, 6H, o-CH3),
4.13 (d, 2J=2.2 Hz, 1H, CꢀCH2), 4.83 (d, 1H,
CꢀCH2). 13C-NMR spectrum, l −9.20 (AlCH3),
17.02 (m-CH3), 17.31 (p-CH3), 19.22 (o-CH3), 96.01
(CꢀCH2), 132.96, 133.46 (o- and m-C, C6Me5), 134.59,
136.71 (i- and p-C, C6Me5), 155.35 (OCꢀCH2).
4.2.8. [Me2AlOC(1-C10H7)Me2] (16)
Yield, 13.0%; slowly decomposes on heating. 1H-
NMR spectrum, l −0.70 (s, 6H, AlCH3), 1.87 (s, 6H,
OCCH3), overlapping signals covering the region
7.14–8.92 (m, 7H, naphthyl). 13C-NMR spectrum in
d8-toluene, l −6.10 (AlCH3), 31.46 (OCCH3), 78.69
(OCCH3), overlapping signals 123.65–139.83 (aro-
matic naphthyl).
4.2.3. [Me2AlOCMe2Ph] (6)
Yield, 8.9%; m.p., 82–84°C. H-NMR spectrum, l
1
4.2.9. [Me2AlOC(2-C10H7)Me2] (18)
−0.58 (s, 6H, AlCH3), 1.61 (s, 6H, OCCH3), 7.07 (t,
1H, p-H, Ph), 7.17 (s, 2H, m-H, Ph), 7.41 (d, 2H,
o-H, Ph). 13C-NMR spectrum, l −5.29 (AlCH3),
31.62 (C–CH3), 77.50 (C–CH3), 126.68, 129.28,
129.88 (o-, m- and p-C, Ph), 145.75 (i-C, Ph). X-ray
quality crystals were grown from a hexane solution of
the sample which was cooled to −28°C for 24 h.
1
Yield 13.9%; m.p. 134–136°C. H-NMR spectrum,
l −0.67 (s, 6H, AlCH3), 1.60 (s, 6H, OCCH3), over-
lapping signals covering the region 7.07–7.75 (m, 7H,
naphthyl). 13C-NMR spectrum, l −5.41 (AlCH3),
31.48 (OCCH3), 77.59 (OCCH3), overlapping signals
124.93–143.08 (aromatic naphthyl).
1
4.3. H-NMR spectroscopic experiments
4.2.4. [Me2AlOC{2-(MeO)–C6H4}Me2]
1
Yield, 37.2%; m.p., 112–114°C. H-NMR spectrum,
Ketone 1 or 5 (1.5 mmol) was added dropwise to a
−78°C cooled solution of Me3Al (1.7 mmol) in 3 ml
of toluene. The reaction mixture was allowed to warm
slowly to ambient temperature and stirred for 1 h. All
solvents were removed in vacuo and a portion of the
oily residue removed for spectroscopic analysis. The
remaining oil was dissolved in 2 ml of toluene and the
solution was heated to reflux for 3 h. Solvents were
removed in vacuo and a portion removed for spectro-
l −0.54 (s, 6H, AlCH3), 1.85 (s, 6H, OCCH3), 3.42
(s, 3H, OCH3), 6.53 (d, 1H, m%-H, C6H4), 6.80 (t, 1H,
m-H, C6H4), 7.07 (t, 1H, p-H, C6H4), 7.36 (d, 1H,
o-H, C6H4). 13C-NMR spectrum, l −5.55 (AlCH3),
30.58 (OCCH3), 54.59 (OCH3), 77.24 (C-CH3), 112.12
(m%-C, C6H4), 120.50 (m-C, C6H4), 127.50, (o-C,
C6H4), 129.92 (p-C, C6H4), 132.96 (i-C, C6H4), 158.57
(o%-C, C6H4).

22
J.F. Allan et al. / Journal of Organometallic Chemistry 602 (2000) 15–23
Table 4
Crystallographic data
All calculations used the general basis set 6-31G* [26].
Absolute energies (Hartrees) for the calculated species:
I, −721.51640; II, −499.57495; III, −1640.42162; IV,
−1720.80446; V, −382.47637; VI, −1406.21446; VII,
−1486.64526; VIII, 40.19517; IX, −498.95493; X, −
381.85011; XI, −1640.42166.
2
6
Formula
Mr
Crystal system
Space group
C
26H38Al2O2
C22H34Al2O2
384.45
Monoclinic
P21/c
436.52
Monoclinic
P21/c
,
a (A)
7.320(2)
15.091(4)
12.153(3)
102.447(16)
1310.9(6)
2
1.106
1.131
160
4.74–67.33
4112
2108
1875
1.071
0.0991
0.0454
0.1180
9.0532(9)
15.2132(15)
8.7909(9)
109.281(2)
1142.8(2)
2
1.117
0.140
160
2.38–28.69
15 277
5. Supplementary material
,
b (A)
,
c (A)
i (°)
V (A )
Crystallographic data (excluding structure factors)
for the structural analysis have been deposited with the
Cambridge Crystallographic Data Centre, CCDC nos.
139425 and 139426. Copies of the data can be obtained
free of charge from The Director, CCDC, 12 Union
Rd., Cambridge CB2 1EX, UK (fax: +44-1223-
336033; e-mail: deposit@ccdc.com.ac.uk or www: http:/
/www.ccdc.cam.ac.uk).
3
,
Z
Dcalc (g cm−3
)
v (mm−1
T (K)
)
q Range (°)
Measured reflections
Unique reflections
Reflections with I\2|(I)
Goodness-of-fit (S)
Rint
11 628
1.062
R (F, F2\2|)
0.0428
0.1297
Acknowledgements
Rw (F2, all data)
We would like to thank the Royal Society for fund-
ing an Exquota Fellowship (I.R.) and a University
Research Fellowship (K.W.H.). Thanks also to the
EPSRC and Bede Scientific Instruments Ltd (micro-
source division) for equipment funding (W.C.).
scopic analysis. The gem-CH2 signals assigned to the
self-coupled aldol reaction of 5 with Me3Al were dou-
blets located at l 2.99 and 2.63 with J=17.1 Hz.
2
4.4. X-ray crystallographic studies
References
Crystals of 2 and 6 were mounted onto glass fibres in
an oil drop. For 2, data were collected on a Stoe-
Siemens four-circle diffractometer equipped with a
novel microsource X-ray generator, tube and focussing
[1] J.R. Zietz, G.C. Robinson, K.L. Lindsay, Chapter 46, in: G.
Wilkinson, F.G.A. Stone, E.W. Abel (Eds.), Comprehensive
Organometallic Chemistry, vol. 7, Pergamon, Oxford, 1983.
[2] H. Yamamoto, K. Maruoka, Pure Appl. Chem. 60 (1988) 21.
[3] (a) S. Saito, M. Shiozawa, M. Ito, H. Yamamoto, J. Am. Chem.
Soc. 120 (1998) 813. (b) D.A. Cogan, J.A. Ellman, J. Am. Chem.
Soc. 121 (1999) 268. (c) S. Saito, T. Sone, K. Shimada, H.
Yamamoto, Synlett (1999) 81. (d) S. Saito, H. Yamamoto,
Chem. Commun. (1997) 1585.
,
optics [22]; the radiation was Cu–Ka (u=1.54184 A).
Intensities were measured with on-line profile fitting
[23]. For 6, data were collected on a Bruker AXS
SMART CCD diffractometer with Mo–Ka radiation
,
(u=0.71073 A). Both data sets were measured at 160 K
[4] F.H. van der Steen, G.P.M. van Mier, A.L. Spek, J. Kroon, G.
van Koten, J. Am. Chem. Soc. 113 (1991) 5742.
by use of a Cryostream cooler [24]. The crystal of 6 was
found to be non-merohedrally twinned, and intensities
were derived for both twin components, which were
subsequently refined to a ratio close to 70:30. Crystal
data are in Table 4; merging of symmetry equivalent
reflections is not performed in the case of non-mero-
hedral twinning, in order to retain the separate compo-
nents for use in refinement. The structure solutions
were by direct methods, with refinement on F2 for all
measured data [25].
[5] M.J. Frisch, G.W. Trucks, H.B. Schlegel, P.M.W. Gill, B.G.
Johnson, M.A. Robb, J.R. Cheeseman, T. Keith, G.A. Pe-
tersson, J.A. Montgomery, K. Raghavachari, M.A. Al-Laham,
V.G. Zakrzewski, J.V. Ortiz, J.B. Foresman, J. Cioslowski, B.B.
Stefanov, A. Nanayakkara, M. Challacombe, C.Y. Peng, P.Y.
Ayala, W. Chen, M.W. Wong, J.L. Andres, E.S. Replogle, R.
Gomperts, R.L. Martin, D.J. Fox, J.S. Binkley, D.J. Defrees, J.
Baker, J.P. Stewart, M. Head-Gordon, C. Gonzalez, J.A. Pople,
Gaussian Inc., Pittsburgh, PA, 1995.
[6] (a) S. Pasynkiewicz, E. Silwa, J. Organomet. Chem. 3 (1965) 121.
(b) E.A. Jeffery, A. Meisters, T. Mole, J. Organomet. Chem. 74
(1974) 365.
[7] M.B. Power, A.W. Apblett, S.G. Bott, J.L. Atwood, A.R. Bar-
ron, Organometallics 9 (1990) 2529.
4.5. Computational details
[8] J.F. Allan, K.W. Henderson, A.R. Kennedy, Chem. Commun.
(1999) 1325.
[9] R.M. Waymouth, B.D. Santarsiero, R.H. Grubbs, J. Am. Chem.
Soc. 106 (1984) 4050.
[10] (a) D. Seebach, R. Amstutz, J.D. Duntiz, Helv. Chim. Acta 64
(1981) 2622. (b) D. Seebach, Angew. Chem. Int. Ed. Engl. 27
The GAUSSIAN 94 program (revision E.2), run on a
Silicon Graphics Origin 200 workstation, was used for
the calculations [5]. No symmetry constraints were im-
posed and all molecules were allowed to freely optimise.

J.F. Allan et al. / Journal of Organometallic Chemistry 602 (2000) 15–23
23
[18] (a) W. Clegg, S.T. Liddle, K.W. Henderson, F.E. Keenan, A.R.
Kennedy, A.E. McKeown, R.E. Mulvey, J. Organomet. Chem.
572 (1999) 283. (b) D.R. Armstrong, F.J. Craig, A.R. Kennedy,
R.E. Mulvey, J. Organomet. Chem. 550 (1998) 355. (c) D.R.
Armstrong, F.J. Craig, A.R. Kennedy, R.E. Mulvey, Chem. Ber.
129 (1996) 1293. (d) I. Demachy, F. Volatron, Eur. J. Inorg.
Chem. (1998) 1015. (e) M.G. Gardiner, G.A. Koutsantonis, S.M.
Lawrence, F.-C. Lee, C.L. Raston, Chem. Ber. 129 (1996) 545.
[19] (a) J.J. Eisch, Chapter 6, in: G. Wilkinson, F.G.A. Stone, E.W.
Abel (Eds.), vol. 1, Pergamon Press, Oxford, 1982. (b) G.H.
Robinson, Coordination Chemistry of Aluminium, VCH, New
York, 1993.
[20] (a) J. Lewinski, J. Zachara, I. Justyniak, Organometallics 16
(1997) 4597. (b) J. Lewinski, J. Zachara, K.B. Starowieyski, J.
Chem. Soc. Dalton Trans. (1997) 4217.
[21] D.F. Schriver, M.A. Drezdon, The Manipulation of Air-Sensi-
tive Compounds, Wiley, New York, 1986.
[22] U.W. Arndt, J.V.P. Long, P. Duncumb, J. Appl. Crystallogr. 31
(1999) 936.
[23] W. Clegg, Acta Crystallogr. Sect. A 37 (1981) 22.
[24] J. Cosier, A.M. Glazer, Appl. Crystallogr. 19 (1986) 105.
[25] G.M. Sheldrick, SHELXTL Version 5, Bruker AXS Inc, Madison,
WI, USA, 1997.
[26] (a) W.J. Hehre, R. Ditchfield, J.A. Pople, J. Chem. Phys. 56
(1972) 2257. (b) P.C. Hariharan, J.A. Pople, Theor. Chim. Acta
28 (1973) 213. (c) J.D. Dill, J.A. Pople, J. Chem. Phys. 62 (1975)
2921.
(1988) 1624. (c) K.W. Henderson, A.E. Dorigo, Q.-Y. Liu, P.G.
Williard, P.v.R. Schleyer, P.R. Bernstein, J. Am. Chem. Soc. 118
(1996) 1339. (d) P.G. Williard, Q.Y. Liu, J. Am. Chem. Soc. 115
(1993) 3380.
[11] (a) E.A. Jeffery, A. Meisters, T. Mole, J. Organomet. Chem. 74
(1974) 373. (b) E.A. Jeffery, A. Meisters, J. Organomet. Chem.
82 (1974) 307.
[12] P. Veya, P. Giorgio Cozzi, C. Floriani, A. Chiesi-Villa, C.
Rizzoli, Organometallics 13 (1999) 4939.
[13] G. Erker, M. Albrecht, Organometallics 12 (1993) 4969.
[14] (a) W. Clegg, E. Lamb, S.T. Liddle, R. Snaith, A.E.H. Wheatley,
J. Organomet. Chem. 573 (1999) 305. (b) R.J. Wehmschulte,
W.J. Grigsby, B. Schiemenz, R.A. Bartlett, P.P. Power, Inorg.
Chem. 35 (1996) 6694. (c) K.W. Henderson, A.E. Dorigo, P.G.
Williard, P.R. Bernstein, Angew. Chem. Int. Ed. Eng. 35 (1996)
1322. (d) L.M. Jackman, M.M. Petrei, B.D. Smith, J. Am.
Chem. Soc. 113 (1991) 3451. (e) M. Westerhausen, W. Schwarz,
Z. Anorg. Allg. Chem. 619 (1993) 1053.
[15] P.G. Williard, G.B. Carpenter, J. Am. Chem. Soc. 108 (1986)
462.
[16] (a) M.L. Sierra, V.S.J. de Mel, J.P. Oliver, Organometallics 8
(1989) 2486. (b) H. Schumann, M. Frick, B. Heymer, F. Girgs-
dies, J. Organomet. Chem. 512 (1996) 117. (c) P.B. Hitchcock,
H.A. Jasim, M.F. Lappert, H.D. Williams, Polyhedron 9 (1990)
245.
[17] F.H. Allen, O. Kennard, Chem. Des. Autom. News 8 (1993) 31.
.