Pentabenzylcyclopentadienides of Lithium
Organometallics, Vol. 15, No. 22, 1996 4705
2.41 Å19). Especially if proper cations are present,
anionic arene complexes should be observed instead of
polymers. The formation of small neutral arene adducts
of alkaline cyclopentadienides such as [{Li(η5-C5Bz5)}2-
(C6D6)] seems to be restricted to special conditions in
which (i) the substituents of the ligand are bulky enough
to avoid the formation of polymers, (ii) no proper cation
can be generated (see below), and (iii) no strong donor
molecules are present, since they can stabilize mono-
meric species such as Li(C5H2(SiMe3)3)‚THF10,11
structural studies.22 [Li(η5-C5Bz5)2][Al(η5-C5Me5)2] is
insoluble in all common solvents.
2Li(C5Bz5) + 3Al(C5Me5) f Al0 + [Al(C5Me5)2]+ +
[Li(C5Bz5)2]- + Li(C5Me5) (4)
The crystallographic unit cell of [Li(η5-C5Bz5)2][Al(η5-
C5Me5)2] contains three almost identical entities of
[Al(η5-C5Me5)2]+ and [Li(η5-C5Bz5)2]- which are sepa-
rated by normal van der Waals distances. One of the
[Li(η5-C5Bz5)2]- species resides on a crystallographic
center of inversion.22 In addition, toluene molecules are
present. Since the experimental data are not sufficient
to solve the structure properly, we will not present
detailed structural data, besides the comments that the
structure of the cation [Al(η5-C5Me5)2]+ is comparable
to our former results23 and that the anion [Li(η5-
C5Bz5)2]- exhibits a slightly shorter Li to ring center
distance (1.90 Å) than was obtained for [Li(η5-C5H5)2]-
experimentally (2.01 Å).12 The distance between both
C5Bz5 rings is in the same range as in [{Li(η5-C5Bz5)}2-
(C6D6)] (3.80 Å; cf. Figure 3). In contrast to [{Li(η5-
C5Bz5)}2(C6D6)], the benzyl groups of [Li(η5-C5Bz5)2]- are
not oriented symmetrically.
In order to investigate the nature of Li(C5Bz5) in
solution, 7Li NMR spectra were recorded. When
Li(C5Bz5) is dissolved in benzene or toluene, only one
7Li NMR signal at -10.3 ppm is observed, indicating
either the presence of monomeric species or a fast
equilibrium between monomeric and dimeric species. In
view of the structure of solid Li(η5-C5Bz5) higher ag-
gregates seem to be unrealistic. Since the observed
shifts depend neither on the temperature (+60 to -70
°C for toluene) nor on the concentration (0.2 M solution,
melt) monomeric species are more likely.21 In OEt2 (0.2
7
M solution) again only one Li NMR signal is observed
(-7.6 ppm between +25 and -110 °C). The stronger
shielding of Li(η5-C5Bz5) in arene solutions might be a
result of the ring current of coordinated arene
molecules.17a
Exp er im en ta l Section
In THF below -10 °C a single resonance at -7.2 ppm
is accompanied by two new signals at -0.2 and -12.1
ppm. These results can be interpreted by the presence
of an equilibrium of THF-solvated, monomeric Li(η5-
C5Bz5) with Li+‚nTHF/[Li(η5-C5Bz5)2]-. Equilibrium
reactions of this type have been extensively studied for
Li(C5H5) and lithium isodicyclopentadienide.17 Further
cooling of the Li(η5-C5Bz5)/THF samples causes two
additional signals at -1.5 and -7.9 ppm, indicating a
more complex situation perhaps involving dimeric spe-
cies of Li(η5-C5Bz5).
Gen er a l Com m en ts. All operations were carried out
under dry nitrogen. Solvents were dried by conventional
procedures. NMR spectra were recorded by means of a Bruker
AC 250 (1H, 13C) or a Bruker AMX 300 spectrometer (7Li, 27Al).
Chemical shifts are reported in δ units (ppm) referenced to
C6D5H (7.15 ppm, 1H), C6D6 (128.0 ppm, 13C), external LiCl/
D2O (1 M, 0 ppm, 7Li) and external [Al(H2O)6]3+ (0.5 M, 0 ppm,
27Al). Mass spectra were obtained by electron impact. Melting
points are observed in sealed capillaries under vacuum and
are uncorrected. t-BuLi was obtained from Merck as a 1.6 M
hexane solution and used as purchased.
Syn th esis of C5Bz5H. The compound was synthesized by
a modification of the method by Hirsch and Bailey.24 The
solvents diisopropylbenzene and benzene were replaced by
isopropylbenzene and toluene, respectively. Originally the
final purification was carried out by distillation of the crude
product. Instead of this procedure, the resulting viscous
orange crude product was suspended with hot methanol and
treated with ultrasonic waves to dissolve impurities. After
cooling of the suspension to room temperature, the methanol
was decanted. This process was repeated several times until
removal of the last solvent by vacuum (10-3 mbar) left colorless
C5Bz5H (63% yield). Mp: 72-74 °C. MS (m/ e, ion, relative
intensity): 516, [M]+, 39%; 425, [M - Bz]+, 18%;, 91, [Bz]+,
100%.
Rea ction of Li(C5Bz5) w ith Al(C5Me5). Using the
lattice energy of Li(C5Me5), which is insoluble in aro-
matic hydrocarbons, as a driving force, the synthesis of
Al(C5Bz5) was attempted according to eq 3.
Li(C5Bz5) + Al(C5Me5) f Al(C5Bz5) + Li(C5Me5)V
(3)
The products of the reaction were separated by virtue
of their differing solubility in toluene, the soluble
compounds being Al(C5Bz5) and residual Li(C5Bz5).
Unfortunately, to date it has proved impossible to isolate
Al(C5Bz5) as a pure compound. However, NMR and
mass spectroscopic data give evidence that Al(C5Bz5) is
monomeric both in solution and in the gas phase.
Syn th esis of Li(C5Bz5). C5Bz5H was dissolved in toluene
to give a 0.2 M solution. An equimolar amount of a hexane
solution of t-BuLi was added (the reaction flask was connected
to a mercury overpressure valve). The slow metalation was
Li(η5-C5Me5) and the side products [Li(η5-C5Bz5)2][Al(η5-
C5Me5)2] and aluminum metal (maximum 3% of the
total Al content) are insoluble in toluene. [Li(η5-
C5Bz5)2][Al(η5-C5Me5)2], presumably formed as shown
in eq 4, is obtained as a colorless crystalline compound.
The identity of this product was confirmed by X-ray
(22) The experimental data obtained are not sufficient to solve the
structure properly. Data for the X-ray structural study of
[Li(η5-C5Bz5)2][Al(η5-C5Me5)2]: Mr ) 1335.7, triclinic, space group P1h,
a ) 13.712(10) Å, b ) 16.443(10) Å, c ) 29.86(3) Å, R ) 98.46(6)°, â )
99.63(7)°, γ ) 98.31(6)°, Z ) 3, V ) 6466(9) Å3, F(calcd) ) 1.100 g cm-3
,
F(000) ) 2298, µ ) 0.070 mm-1, θ range 1.53-22.50°, 15 857 unique
reflections, 6436 observed reflections, 1126 parameters, GOF ) 1.036,
R1 (I > 2σ(I)) ) 0.0951, wR2 ) 0.2881, w ) σ-2|F|, largest difference
peak 1.068 e Å-3, deepest difference peak -0.611 e Å-3
. While one
(21) The degree of aggregation (n) was also measured by cryoscopic
and osmometric molecular weight determinations in benzene (n ) 1.04
for a 0.013 M solution (cryoscopically), n ) 1.12 for a 0.036 M solution
(cryoscopically), and n ) 1.08 for a 0.04 M solution (osmometrically)).
These results confirm monomeric Li(η5-C5Bz5) to be the dominant
species over the measured concentration range.
[Al(η5-C5Me5)2]+ is normal, the second is split over two positions
(occupation number 0.25). Whether this is a disorder or an overstruc-
ture problem could not be decided because of the low crystal quality.
(23) Dohmeier, C.; Schno¨ckel, H.; Robl, C.; Schneider, U.; Ahlrichs,
R. Angew. Chem., Int. Ed. Engl. 1993, 32, 1655-1657.
(24) Hirsch, S. S.; Bailey, W. J . J . Org. Chem. 1978, 43, 4090-4094.