Towards an understanding of the conjugate addition of organolithium reagents to α,β-unsaturated ketones: The isolation and solid-state structure of a monomeric lithium aluminate with very short agostic LiHC interactions1

Article abstract of DOI:10.1016/S0022-328X(98)00876-6

Reaction of methylaluminium bis(2,6-di-tert-butyl-4-methylphenoxide) (MAD), 1, with alkyllithium reagents, R′Li (R′=Me, n-Bu or t-Bu), yields the solvent-dependent products lithium bis(2,6-di-tert-butyl-4-methylphenoxide)-THF complex, 2·THF, lithium dimethylbis(2,6-di-tert-butyl-4-methylphenoxide)aluminate, 3, a new type of lithium aluminate in which the lithium centre is stabilised by very short agostic LiH(t-Bu) interactions, and tris(alkyl)aluminium. The observation of these products suggests an explanation for the tendency of α,β-unsaturated ketones to undergo conjugate (rather than 1,2-) addition in the presence of MAD and organolithium reagents.

Full text of DOI:10.1016/S0022-328X(98)00876-6

Journal of Organometallic Chemistry 573 (1999) 305–312
Towards an understanding of the conjugate addition of organolithium
reagents to h,i-unsaturated ketones: the isolation and solid-state
structure of a monomeric lithium aluminate with very short agostic
Li···HC interactions1
William Clegg a, Elinor Lamb b, Stephen T. Liddle a, Ronald Snaith b,*,
Andrew E.H. Wheatley b
a Department of Chemistry, Uni6ersity of Newcastle, Newcastle upon Tyne, NE1 7RU, UK
b Department of Chemistry, Uni6ersity of Cambridge, Lensfield Road, Cambridge, CB2 1EW, UK
Received 21 May 1998
Abstract
Reaction of methylaluminium bis(2,6-di-tert-butyl-4-methylphenoxide) (MAD), 1, with alkyllithium reagents, R%Li (R%=Me,
n-Bu or t-Bu), yields the solvent-dependent products lithium bis(2,6-di-tert-butyl-4-methylphenoxide)-THF complex, 2·THF,
lithium dimethylbis(2,6-di-tert-butyl-4-methylphenoxide)aluminate, 3, a new type of lithium aluminate in which the lithium centre
is stabilised by very short agostic Li···H(t-Bu) interactions, and tris(alkyl)aluminium. The observation of these products suggests
an explanation for the tendency of h,i-unsaturated ketones to undergo conjugate (rather than 1,2-) addition in the presence of
MAD and organolithium reagents. © 1999 Elsevier Science S.A. All rights reserved.
Keywords: Aluminium; h,i-Unsaturated ketone; Conjugate addition; Lithium; Solid-state structure
methylaluminium bis(2,6-di-tert-butyl-4-methylphenox-
ide), MeAl(OR)2, 1, (MAD; R=t-Bu2C6H2Me;
1. Introduction
Scheme 1) [4,5]. Addition to this in non-polar media of
an h,i-unsaturated ketone is followed, upon subse-
quent treatment with an organolithium reagent, by 1,4-
rather than 1,2-addition. Why this is so has not been
While the synthetic utility of organolithium reagents
in 1,2-addition reactions to carbonyl groups is well
established [1], it has proven less easy to use such hard
nucleophiles in 1,4-conjugate additions to h,i-unsatu-
explained, other than to note that the oxygenophilic
rated ketones. Recent reports have detailed the devel-
aluminium species may electrophilically activate the
opment of new asymmetric hetero-bimetallic species [2]
enone. Given this uncertainty, we have investigated the
capable of catalytically promoting such reactions. Fur-
reactions between MAD and the organolithium
ther work has employed the non-catalytic addition of
reagents (R%=Me, t-Bu and n-Bu). In these instances
Lewis acids such as tri-coordinate aluminium species
[3], notably the sterically-crowded organoaluminium
common products, LiOR, 2, and Me2Al(v2-OR)2Li, 3,
are observed irrespecti6e of which organolithium is
used. More generally, the isolation of 3 suggests that
while the addition of R%Li to a system containing enone
and an aluminium oxygenophile yields a new hetero-
bimetallic species in situ, the persistent observation of a
dimethylaluminium moiety in 3 demonstrates that it is
* Corresponding author. Fax: +44 1223 336362; e-mail: cm-
cloo6@cam.ac.uk
1 Dedicated to Professor Brian Johnson on the occasion of his 60th
birthday and in recognition of his outstanding and ongoing contribu-
tions to organometallic chemistry.
0022-328X/99/$ - see front matter © 1999 Elsevier Science S.A. All rights reserved.
PII S0022-328X(98)00876-6

306
W. Clegg et al. / Journal of Organometallic Chemistry 573 (1999) 305–312
tert-butyl-4-methylphenoxide ligands [6].
Addition of t-BuLi to 1 at −80°C, followed by
warming to room temperature, gives a suspension
which can be dissolved in two ways. Treatment with
THF leads to colourless crystals which X-ray crystal-
lography shows to be the simple dimeric [7] lithium
2,6-di-tert-butyl-4-methylphenoxide-THF
2 · THF (Fig. 2 and Table 2).
complex,
Dissolution of the suspension resulting from the reac-
tion of 1 with t-BuLi can alternatively be effected by
the dropwise addition of toluene, storage at room tem-
perature affording lithium dimethylbis(2,6-di-tert-butyl-
4-methylphenoxide)aluminate, 3. Retention of the
Scheme 1.
1
aluminium centre is demonstrated by H-NMR spec-
not this species which directly causes 1,4-addition to the
enone. Nevertheless, an understanding of the structure
of 3 and how it comes about does shed light on the
probable route for conjugate addition to the enone,
and, furthermore suggests that 3 does, in fact, play an
indirect but crucial role in this process.
troscopy in [2H8]toluene, which shows a singlet at
−0.45 ppm, characteristic of an aluminium-attached
methyl group. Intriguingly, however, integrals show a
1:1 ratio of 2,6-di-tert-butyl-4-methylphenolate to alu-
minium-attached methyl. Unsurprisingly, the same re-
sult is observed if MeLi is employed in place of t-BuLi,
but the fact that the reaction undergone by t-BuLi is
not an anomaly is established by the isolation of 3 if
n-BuLi is employed instead. That 3 contains a dimethy-
lated aluminium centre is borne out by X-ray crystal-
lography which shows it to be the monomeric ate
2. Results and discussion
The 1:2 reaction of trimethylaluminium with 2,6-di-
tert-butyl-4-methylphenol in toluene at −80°C gener-
ates methylaluminium bis(2,6-di-tert-butyl-4-methyl-
phenoxide) (MAD), 1. Its crystal structure reveals a
monomer (Fig. 1 and Table 1) in which the aluminium
centre is tri-coordinate by virtue of the bulky 2,6-di-
complex
lithium
dimethylbis(2,6-di-tert-butyl-4-
methylphenoxy)aluminate (Fig. 3). At the core of the
structure is an AlO2Li four-membered ring (Al–O1=
˚
˚
1.8457(17) A, Al–O2=1.8425(17) A, Li–O1=1.848(5)
˚
˚
A, Li–O2=1.888(4) A). Although this motif has been
observed before, in all but one of these previous five
Fig. 1. Structure of the monomeric reagent MAD, 1, with key atoms labelled; hydrogen atoms omitted for clarity.

307
W. Clegg et al. / Journal of Organometallic Chemistry 573 (1999) 305–312
Table 1
had been consigned mostly, and rather vaguely, to that
of electrophilic activation of the enone. One study had
noted the possibility of lithium aluminate intermediates
during conjugate additions to cyclopentenones [12]. For
the specific case of MAD, 1, as the aluminium species
in question, however, experimental evidence was pro-
duced against the intermediacy of an ate complex [4].
Crucially, though, that study involved reacting R%Li,
MAD and enone in ethereal solution; in such cases 1,2-
and not 1,4-addition (as is observed in non-polar sol-
vents) was observed. In fact, that is perhaps unsurpris-
ing. If an ate complex such as 3, produced by reacting
MAD with R%Li, was involved directly in conjugate
addition, it would likely operate via the enone coordi-
nating to the formally two-coordinate Li+ centre of the
ate complex, displacing the t-Bu···Li+ interactions
noted above. The resulting ate-enone complex would
then be primed to deliver R% to the 4-position (Fig.
5(a)). In ethereal solvents, however, the ether would
compete with the enone for the Li+ coordination site,
thereby fully or partly blocking this mechanism. In fact,
there is some evidence that ethers are better Lewis bases
than enones. For example, we find that addition of
2-cyclohexenone (OꢀCHex) to MAD, 1, in toluene af-
fords the 1:1 complex 1 · OꢀCHex but that dissolution
of this complex by THF addition results in the etherate
complex 1 · THF. The suggestion is clearly that evi-
dence gained in polar, coordinating solvents against the
intermediacy of lithium aluminate species in 1,4-addi-
tions can be discounted. Our evidence, in contrast,
shows quite definitively that an ate complex cannot be
involved directly in such additions. Thus, we find that
MAD, 1, and R%Li reagents give one common product,
the lithium dimethylaluminate, 3, irrespective of
whether R% is Me or t-Bu or n-Bu. Obviously, this
product could deliver only MeLi in 1,4-fashion to an
enone, and yet it is proven that these mixtures do
deliver MeLi or t-BuLi or n-BuLi in such a way. There
seem to be two possible conclusions to be drawn from
these findings. The first, and rather negative one, is that
an ate complex such as 3 plays no role whatsoever in
these conjugate additions even though it is isolated in
reasonably high yields from R%Li and MAD mixtures.
The second possible conclusion is that 3 is involved in
an indirect (but crucial) way in accomplishing the 1,4-
addition of R%Li to an enone. In particular, when 3 is
produced from MAD, 1, and R%Li we have shown that
it is so, along with LiOR, 2, and, by implication, R%3Al
(Scheme 2). Clearly, when such a mixture is treated
with enone, the only possible sources of R% are unre-
acted R%Li (which, on its own, undergoes 1,2-addition)
˚
Selected bond lengths (A) and angles (°) for MAD, 1
Al(1)–O(1)
Al(1)–C(1)
1.695(2)
1.926(3)
Al(1)–O(2)
1.684(2)
O(2)–Al(1)–O(1) 111.97(11)
O(1)–Al(1)–C(1) 123.84(14)
C(17)–O(2)–Al(1) 147.3(2)
O(2)–Al(1)–C(1) 123.57(14)
C(2)–O(1)–Al(1) 140.0(2)
structures stabilisation of the lithium centre has been
effected by external solvation [5,8]; in the remaining
case [9] there is internal stabilisation via Li’F interac-
tions. It is, therefore, significant to note that 3 shows
stabilisation of the otherwise merely two-coordinate
lithium centre via extensive agostic Li···H(Me) bonding
which renders the overall coordination sphere of the
metal pseudo-octahedral (Fig. 4). Even though the Li–
O distances observed in 3 (mean=1.868 A) represent
the shortest of any yet observed in monomeric lithium
aluminates of this type, the four short Li···C distances
(Li···C8=2.732(7) A, Li···C9=2.458(7) A, Li···C28=
2.823(7) A, Li···C30=2.415(7) A) are similar to those
previously attributed to strong agostic interactions in
compounds containing formally two-coordinate lithium
(e.g. 2.787 A in [(Me3Si)2NLi]2 ([10]a) 2.482 A (mean)
in [(Me3Si)2NLi]3 ([10]a) and 2.82–3.05 A in anti-
[(Et2O)Li]2[(t-Bu)6Al6(O)6Me2] ([10]b)). Further, the
Li···H interactions seen in 3 are all exceptionally short
(Li···H8C=2.017 A, Li···H9C=1.868 A, Li···H28C=
2.127 A, Li···H30C=1.885 A; all other Li···H distances
˚
˚
˚
˚
˚
˚
˚
˚
˚
˚
˚
˚
˚
are ]2.35 A) ([10]b). For the purposes of computing
the agostic Li···distances relevant C–H bond lengths
were increased from 0.98 A, appropriate to X-ray dif-
fraction, to 1.08 A by simple displacement along the
˚
˚
bond (Table 3).
It is not surprising that the reaction of 1 with MeLi
yields a product containing two aluminium-attached
methyl groups. However, that 3 results also from the
use of n- or t-BuLi is altogether more intriguing and is
best rationalised in terms of the isolation of 2 · THF.
The suggestion is clearly that in this instance BuLi
(n- or t-) undergoes a 3:1 reaction with 1 yielding
tris(butyl)aluminium and MeLi in situ, the combination
of the latter intermediate with a further equivalent of
MAD giving 3 (Scheme 2). The observation that MAD
is monomeric in solution and that it fails to act as a
precursor to Al(OR)3 [11] rules out the alternative
possibility that dimeric MAD acts as a source of
Me2AlOR and Al(OR)3 in situ, the latter species under-
going a 1:3 reaction with R%Li (R%=Me, n-Bu or t-Bu)
to afford LiOR which could then combine with
Me2AlOR.
and this R%Al co-product. Thus, one mechanistic sce-
3
Our findings may have significance regarding the
mechanism of the 1,4-addition of R%Li reagents (R%=
Me, n-Bu or t-Bu) to enones in the presence of alu-
minium species. Previously, the effect of such species
nario is that 3, formed in non-polar solvent from R%Li
and MAD, is coordinated at Li+ by subsequently
added enone. Such coordination would presumably dis-
place the (t-Bu)H···Li interactions found in 3 but

308
W. Clegg et al. / Journal of Organometallic Chemistry 573 (1999) 305–312
Fig. 2. The structure of one of the two crystallographically independent centrosymmetric dimeric molecules of 2 · THF with key atoms labelled;
hydrogen atoms omitted for clarity.
nonetheless the space around the Li+, the O of the
enone, and its attached carbon centre (C-2) would be
extremely crowded. This coordination and the result-
3. Experimental
3.1. General experimental
ing crowding might then direct the co-product R%Al
to attack the much more exposed C-4 position of the
enone (Fig. 5(b)).
3
Standard inert-atmosphere Schlenk techniques were
employed using a dual nitrogen/vacuum line. Schlenk
tubes were pre-dried at 180°C prior to evacuation to
less than 0.1 Torr three times, being filled with dry
nitrogen from the house supply between each evacua-
tion. Reagents were used as received from the Aldrich
Chemical Company. 2,6-Di-tert-butyl-4-methylphenol
We are exploring these mechanistic possibilities fur-
ther by NMR experiments designed to deduce the
solution species present in mixtures of MAD and
R%Li reagents and, in particular, by synthetic and
NMR studies in which isolated 3 is reacted with
enones and then the characterised and isolated prod-
ucts of these reactions are in turn reacted with
tris(alkyl)aluminium reagents. These results will be re-
ported in a subsequent paper.
Table 2
˚
Selected bond lengths (A) and angles (°) for the two crystallographi-
cally independent molecules of 2 · THF
O(1)–C(1)
O(1)–Li(1A)
O(2)–Li(2)
Li(1)–O(3)
1.3364(15) O(1)–Li(1)
1.874(3) O(2)–C(20)
1.852(3) O(2)–Li(2B)
1.883(3) Li(2)–O(4)
1.818(3)
1.3388(16)
1.834(3)
1.884(3)
C(1)–O(1)–Li(1)
158.48(11) C(1)–O(1)–Li(1A) 120.31(11)
Li(1)–O(1)–Li(1A) 80.77(12) O(1)–Li(1)–O(1A) 99.23(12)
C(20)–O(2)–Li(2) 132.24(11) C(20)–O(2)–Li(2B) 147.08(12)
Fig. 3. Structure of the monomeric lithium aluminate 3 with key
atoms labelled; hydrogen atoms (except for the methyl groups in-
volved in agostic interactions with the lithium centre) omitted for
clarity.
Li(2)–O(2)–Li(2B) 80.33(12) O(2)–Li(2)–O(2B)
99.67(12)
Symmetry operators: A, −x, 1−y, −z; B, 1−x, 2−y, −z.

W. Clegg et al. / Journal of Organometallic Chemistry 573 (1999) 305–312
309
Scheme 2.
which was dissolved in hexane (3 ml) and the mini-
mum of hot toluene. Storage at room temperature for
12 h afforded large colourless crystals of 1, m.p.,
171–173°C; yield, 89%. Found: C 76.48, H 9.97. Calc.
for C31H49AlO2: C 77.50, H 10.21. 1H-NMR spec-
troscopy (400.137 MHz, 25°C, [2H6]benzene), l 7.18
(s, 4H, Ar), 2.30 (s, 6H, p-CH3), 1.58 (s, 36H, o-
CCH3), −0.26 (s, 3H, AlCH3).
Fig. 4. The core of 3 emphasising the pseudo-octahedral coordination
of the lithium centre.
was weighed into the Schlenk tube prior to purging,
while toluene (freshly distilled and maintained at
reflux over sodium), hexane (freshly distilled and
maintained at reflux over sodium–potassium amal-
gam), trimethylaluminium, methyllithium, n-butyl-
lithium, t-butyllithium and 2-cyclohexenone were
added direct to the nitrogen-filled Schlenk tube using
standard syringe techniques.
3.3. Synthesis of MAD-(2-cyclohexenone) complex,
1 · OꢀCHex
Trimethylaluminium (1.5 ml, 2.0 M in toluene, 3.0
mmol) was added to a stirred suspension of 2,6-di-
tert-butyl-4-methylphenol (1.32 g, 6.0 mmol) in
toluene (2 ml) under nitrogen at −78°C and the re-
sulting solution was stirred for 1 h at room tempera-
ture. The solution was cooled to −78°C and
2-cyclohexenone (0.29 ml, 3.0 mmol) was added,
yielding a deep orange suspension. Warming to room
temperature afforded a yellow suspension which was
dissolved by gentle heating. Storage at +5°C for 1
day yielded bright yellow micro-crystals of
1 · OꢀCHex, m.p., 250–251°C (dec.); yield, 24%.
Found: C 77.87, H 9.62. Calc. for C39H57AlO3: C
All 1H-NMR spectra were recorded at room tem-
perature using a Bruker AM 400 FT-NMR spectrom-
eter.
3.2. Synthesis of methylaluminium
bis(2,6-di-tert-butyl-4-methylphenoxide) (MAD), 1
Trimethylaluminium (1.5 ml, 2.0 M in toluene, 3.0
mmol) was added to a stirred suspension of 2,6-di-
tert-butyl-4-methylphenol (1.32 g, 6.0 mmol) in
toluene (2 ml) under nitrogen at −78°C and the re-
sulting solution was stirred for 1 h at room tempera-
ture. Removal of the solvent afforded a white powder
1
78.00, H 9.50. H-NMR spectroscopy (400.137 MHz,
25°C, [2H6]benzene), l 7.25 (s, 4H, 1, Ar), 7.13–7.00
(m, 3H, 2/3tol.), 6.23 (dt, 1H, OꢀCHex, 3-CH,
3JC(H)ꢀCH=9.95 Hz, 3JC(H)–CH=4.02 Hz), 6.07 (dt,
Table 3
˚
Selected bond lengths (A) and angles (°) for 3
3
1H, OꢀCHex, 2-CH, JC(H)ꢀCH=9.97 Hz), 2.33 (s, 6H,
Al–O(1)
Al–C(31)
Li(1)–O(1)
O(1)–C(1)
Li···C8
1.8457(17) Al–O(2)
1.8425(17)
1.968(3)
1.888(4)
1.384(3)
2.458(7)
2.415(7)
1, p-CH3), 2.11 (s, 2.08H, 2/3tol.), 2.07 (t, 1H,
1.969(3)
1.848(5)
1.390(3)
2.732(7)
2.823(7)
Al–C(32)
Li(1)–O(2)
O(2)–C(16)
Li···C9
OꢀCHex, 6-CH2, 3JC(H)–CH=6.81 Hz), 1.67 (s, 36H,
3
1, o-CCH3), 1.20 (quart., 1H, OꢀCHex, 4-CH2, JC(H)–
CH=5.60 Hz), 0.97 (dt., 1H, OꢀCHex, 5-CH2,
Li···C28
Li···C30
3JC(H)–CH=6.40 Hz), 0.00 (s, 3H, 1, AlCH3).
O(1)–Al–O(2)
Al–O(1)–Li(1)
O(1)–Li(1)–O(2)
C(16)–O(2)–Al
89.40(7)
91.92(15)
87.96(19)
127.40(14)
C(31)–Al–C(32) 115.39(13)
Al–O(2)–Li(1)
C(1)–O(1)–Al
90.72(15)
125.16(16)
3.4. Synthesis of MAD-THF complex, 1 · THF
C(1)–O(1)–Li(1) 129.7(2)
An identical procedure as for 1 · OꢀCHex was used
except that dissolution was effected by the dropwise
C(16)–O(2)–Li(1) 130.86(19)

310
W. Clegg et al. / Journal of Organometallic Chemistry 573 (1999) 305–312
Fig. 5. (a) Schematic representation of a complex between enone and an ate complex such as 3. (b) Schematic representation of a
tris(alkyl)aluminium at the C-4 position of an enone coordinated to an ate complex such as 3.
addition of THF (ca. 1 ml). Storage at +5°C for 1 day
afforded colourless crystals of 1 · THF, m.p.,
176–178°C, yield, 59%. Found: C 75.87, H 10.44. Calc.
for C37H57AlO3: C 77.08, H 9.90. 1H-NMR spec-
troscopy (400.137 MHz, 25 °C, [2H8]toluene), l 7.21 (s,
4H, 1, Ar), 3.69 (m, 4H, THF), 2.32 (s, 6H, 1, p-CH3),
1.52 (s, 36H, 1, o-CCH3), 0.98 (4H, THF), 0.03 (s, 3H,
1, AlCH3).
H 10.40, Li 2.35. 1H-NMR spectroscopy (400.137
MHz, 25°C, [2H8]toluene), l 7.24 (s, 2H, Ar), 3.18 (m,
4H, THF), 2.39 (s, 3H, p-CH3), 1.66 (s, 18H, o-CCH3),
1.07 (m, 4H, THF).
3.6. Syntheses of lithium dimethylbis(2,6-di-tert-butyl-
4-methylphenoxide)aluminate, 3
(a) An identical procedure as for 2 · THF was used
except that dissolution was effected by the dropwise
addition of toluene at room temperature (ca. 5 ml).
Storage at room temperature for 4 days afforded
colourless crystals of 3, m.p., 232–234°C, yield, 53%
(with respect to t-BuLi, see Scheme 2). Found: C 75.48,
H 10.18, Li 1.28. Calc. for C32H52AlLiO2: C 76.49, H
3.5. Synthesis of lithium bis(2,6-di-tert-butyl-4-methyl-
phenoxide)-THF complex, 2 · THF
Trimethylaluminium (1.5 ml, 2.0 M in toluene, 3.0
mmol) was added to a stirred suspension of 2,6-di-tert-
butyl-4-methylphenol (1.32 g, 6.0 mmol) in toluene (2
ml) under nitrogen at −80°C and the resulting solution
was stirred for 1 h at room temperature. To this
solution of methylaluminium bis(2,6-di-tert-butyl-4-
methylphenoxide), t-BuLi (1.77 ml, 1.7 M in hexane,
3.0 mmol) was added at −80°C. Warming to room
temperature afforded a suspension which was dissolved
at reflux by the dropwise addition of THF (ca. 3 ml).
Storage at room temperature for 2 days afforded
colourless crystals of 2 · THF, m.p., 304–306°C; yield,
89% (with respect to t-BuLi, see Scheme 2). Found: C
75.64, H 10.35, Li 2.10. Calc. for C19H31LiO2: C 76.51,
1
10.36, Li 1.39. H-NMR spectroscopy (400.137 MHz,
25°C, [2H8]toluene), l 7.11 (s, 2H, Ar), 2.25 (s, 3H,
p-CH3), 1.46 (s, 18H, o-CCH3), –0.45 (s, 3H, AlCH3).
(b) A solution of methylaluminium bis(2,6-di-tert-
butyl-4-methylphenoxide) was prepared as for 2 · THF.
To this was added n-BuLi (1.88 ml, 1.6 M in hexane,
3.0 mmol) at −80°C. Warming to room temperature
afforded a suspension which was dissolved at reflux.
Storage of the resultant colourless solution at room
temperature for 1 day afforded colourless micro-crys-
tals of 3, yield, 45% (with respect to n-BuLi).

W. Clegg et al. / Journal of Organometallic Chemistry 573 (1999) 305–312
311
Table 4
Crystallographic data for 1, 2 · THF and 3
1
2 · THF
38H62Li2O4
3
Formula
Mr
C31H49AlO2
480.68
C
C32H52AlLiO2
502.66
596.76
(
(
Space group
P1
P21/n
P1
˚
a (A)
12.144(2)
12.846(2)
11.014(2)
102.63(1)
110.52(1)
68.18(1)
1486.2(4)
2
11.5446(6)
15.7174(9)
21.3907(11)
90
103.607(2)
90
3772.4(4)
4
1.051
10.8610(2)
11.6770(11)
14.3324(13)
93.616(2)
108.845(2)
112.793(2)
1548.6(2)
2
˚
b (A
˚
c (A)
h (°)
i (°)
k (°)
3
˚
V (A )
Z
Dc (g cm−3
)
1.074
1.078
Crystal size (mm)
0.40×0.30×0.20
Mo–Kh, 0.71069
0.092
0.80×0.60×0.30
Mo–Kh, 0.71073
0.065
0.56×0.38×0.37
Mo–Kh, 0.71073
0.090
˚
Radiation (A)
v (mm−1
F(000)
)
528
1312
552
T (K)
180
160
160
Scan mode
ꢀ-2q
ꢀ
ꢀ
2q range (°)
5.50–54.98
7144
6817
3.70–57.74
23149
8870
4.30–57.68
11455
6933
Measured reflections
Unique reflections
Rint
0.0643
0.0282
0.0373
Reflections with I\2|(I)
6816
0.0627, 0.1994
0.991
6096
0.0527, 0.1514
1.033
4257
0.0641, 0.1789
0.974
Final R(F), wR(F2)
Goodness-of-fit
Max. peak, hole (eA
−3
˚
)
0.298, −0.394
0.350, −0.274
0.549, −0.299
(c) A solution of methylaluminium bis(2,6-di-tert-
butyl-4-methylphenoxide) was prepared as for 2 · THF.
To this was added MeLi (3.00 ml, 1.0 M in 90%
cumene/10% THF, 3.0 mmol) at −80°C. Warming to
room temperature afforded a suspension which was
dissolved at reflux. Toluene (2 ml) was added and the
colourless solution was stored at +5°C for 2 days,
whereupon colourless crystals of 3 were deposited, yield,
38% (with respect to MeLi).
Acknowledgements
Thanks go to the UK EPSRC (W.C., A.E.H.W) and
to the University of Cambridge (E.L.) for financial
support.
References
[1] (a) B. Weidmann, D. Seebach, Angew. Chem. Int. Ed. Engl.
22 (1983) 31. (b) M.T. Reetz, Angew. Chem. Int. Ed. Engl. 23
(1984) 556.
[2] T. Arai, H. Sasai, K. Aoe, K. Okamura, T. Date, M.
Shibasaki, Angew. Chem. Int. Ed. Engl. 35 (1996) 104.
[3] (a) K. Maruoka, M. Sakurai, H. Yamamoto, Tetrahedron
Lett. 26 (1985) 3853. (b) K. Maruoka, T. Itoh, H. Yamamoto,
J. Am. Chem. Soc. 107 (1985) 4573. (c) K. Maruoka, T. Itoh,
M. Sakuria, K. Manoshita, H. Yamamoto, ibid. 110 (1988)
3588.
[4] K. Maruoka, K. Nonoshita, H. Yamamoto, Tetrhedron Lett.
28 (1987) 5723.
[5] M.B. Power, S.G. Bott, J.L. Atwood, A.R. Barron, J. Am.
Chem. Soc. 112 (1990) 3446.
[6] C.E. Holloway, M. Melnik, J. Organomet. Chem. 543 (1997) 1.
[7] For other dimeric lithium phenolates see: (a) B. Cetinkaya, I.
Gu¨mru¨kcu¨, M.F. Lappert, J.L. Atwood, R. Shakir, J. Am.
Chem. Soc. 102 (1980) 2086; (b) J.C. Huffman, R.L. Geerts,
K.G. Caulton, J. Crystallogr. Spectrosc. Res. 14 (1984) 541; (c)
G. Kociok-Ko¨hn, J. Pickardt, H. Schumann, Acta Crysallogr.,
Sect. C 47 (1991) 2649; (d) L. Matilainen, M. Klinga, M.
Leskela¨, Polyhedron 14 (1995) 635.
3.7. X-ray crystallography
Essential crystallographic details are given in Table 4.
Data for 1 were collected on a Rigaku AFC5R four-circle
diffractometer with data for 2 · THF and 3 being col-
lected on a Bruker AXS SMART CCD diffractometer.
All three structures were solved using direct methods and
subsequent Fourier difference syntheses and refined by
full-matrix least-squares on F2 with anisotropic displace-
ment parameters for all non-hydrogen atoms [13]. All
hydrogen atoms were placed in geometrically-idealised
positions based on peaks found in a difference synthesis
and refined using a riding model including free torsional
rotation about X–Me bonds. Two-fold disorder was
resolved for the THF ligands in one of the two indepen-
dent molecules of 2 · THF. Crystallographic data (ex-
cluding structure factors) for the structures reported in
this paper have been deposited with the Cambridge
Crystallographic Data Centre.
[8] For solvated AlO2Li rings see Ref. [5] and: (a) W.J. Evans,
T.J. Boyle, J.W. Ziller, Polyhedron 11 (1992) 1093; (b) D.A.
Atwood,

312
W. Clegg et al. / Journal of Organometallic Chemistry 573 (1999) 305–312
D. Rutherford, Inorg. Chem. 34 (1995) 4008; (c) A.G. Avent,
(1995) 6465.
W.-Y. Chen, C. Eaborn, I.B. Gorrell, P.B. Hitchcock, J.D.
Smith, Organometallics 15 (1996) 4343; (d)W.-Y. Chen, C.
Eaborn, I.B. Gorrell, P.B. Hitchcock, M. Hopman, J.D. Smith,
J. Chem. Soc., Dalton Trans., (1997) 4689.
[11] A.P. Shreve, R. Mulhaupt, W. Fultz, J. Calabrese, W. Robbins,
S.D. Ittel, Organometallics 7 (1988) 409.
[12] (a) M.B. Floyd, M.J. Weiss, J. Org. Chem. 44 (1979) 71. (b) K.F.
Bernady, M.B. Floyd, J.F. Poletto, M.J. Weiss, ibid. 44 (1979)
1438.
[13] (a) G.M. Sheldrick, SHELXS and SHELXL, University of Go¨t-
tingen (1986, 1993 and 1997). (b) G.M. Sheldrick, Acta Crystal-
logr. Sect. A 46 (1990) 467. (c) G.M. Sheldrick, SHELXTL User
Manual, Version 5, Bruker AXS, Madison, WI, USA (1994 and
1998).
[9] T.J. Barbarich, S.T. Handy, S.M. Miller, O.P. Anderson, P.A.
Grieco, S.H. Strauss, Organometallics 15 (1996) 3776.
[10] For recent examples of very short Li···HC agostic bonds in
di-coordinate lithium compounds see: (a) I. Hemme, B. Tecklen-
burg, M. Noltmeyer, U. Klingebiel, Chem. Ber. 128 (1995) 351;
(b) C.J. Harlan, S.G. Bott, A.R. Barron, J. Am. Chem. Soc. 117
.