HeI photoelectron spectroscopy of trialkylaluminum and dialkylaluminum hydride compounds and their oligomers

Article abstract of DOI:10.1021/om010994h

The HeI photoelectron spectra of trimethylaluminum, triethylaluminum, dimethylaluminum, and diethylaluminum hydrides were recorded as a function of temperature. From the spectra observed at different temperatures the spectra of the pure monomers and the trimethylaluminum, dimethylaluminum, and diethylaluminum hydride dimers and some of the bands related to the dimethylaluminum hydride trimer have been obtained. The spectra were interpreted with the aid of ab initio quantum chemical calculations, including Hartree-Fock/Koopmans, outer valence Green's function, and equation of motion coupled-cluster ionization energy calculations. The vertical ionization energies predicted by the latter two methods are in excellent agreement with the experimental values. Furthermore, the association capability of these compounds, the effects of the substituents, and the degree of association on the electronic structure are also discussed.

Full text of DOI:10.1021/om010994h

Organometallics 2002, 21, 2751-2757
2751
HeI P h otoelectr on Sp ectr oscop y of Tr ia lk yla lu m in u m
a n d Dia lk yla lu m in u m Hyd r id e Com p ou n d s a n d Th eir
Oligom er s
G a´ bor Vass, Gy o¨ rgy Tarczay, G a´ bor Magyarfalvi, Andr a´ s B o¨ di, and
L a´ szl o´ Szepes*
Department of General and Inorganic Chemistry, E o¨ tv o¨ s University, P.O. Box 32,
H-1518 Budapest 112, Hungary
Received November 15, 2001
The HeI photoelectron spectra of trimethylaluminum, triethylaluminum, dimethylalumi-
num, and diethylaluminum hydrides were recorded as a function of temperature. From the
spectra observed at different temperatures the spectra of the pure monomers and the
trimethylaluminum, dimethylaluminum, and diethylaluminum hydride dimers and some
of the bands related to the dimethylaluminum hydride trimer have been obtained. The spectra
were interpreted with the aid of ab initio quantum chemical calculations, including Hartree-
Fock/Koopmans, outer valence Green’s function, and equation of motion coupled-cluster
ionization energy calculations. The vertical ionization energies predicted by the latter two
methods are in excellent agreement with the experimental values. Furthermore, the
association capability of these compounds, the effects of the substituents, and the degree of
association on the electronic structure are also discussed.
In tr od u ction
Raman,25,26 NMR,3,27-29 and UV photoelectron spectro-
3
0
scopic (UPS) techniques, and quantum chemical cal-
The structure, bonding, and dissociation dynamics of
trialkylaluminum and dialkylaluminum hydride oligo-
mers (dimers and trimers), considered as simple model
systems of the electron-deficient, multicentered bond,
have received great and continuous attention since the
end of the 19th century. The investigations were based
3,31,32
culations.
Among the trialkylaluminum compounds trimethyl-
aluminum (TMA) has been studied most extensively and
14-17
for the longest time. According to X-ray
and recent
13
neutron diffraction measurements, as well as freezing
5-7
point depression measurements,
TMA forms dimers
1
,2
on a great variety of methods, including vapor density,
in the solid state and in solutions of benzene and
3
,4
vapor pressure depression, and freezing point de-
1
cyclopentane. Beyond the structure (Figure 1), H NMR
5
-7
3,8-12
13
pression
X-ray
measurements, electron,
diffraction, mass spectrometric,
neutron, and
27-29
investigations
have additionally confirmed that in
1
4-17
3,18-20
3,21-24
IR,
hydrocarbon solvents the interchange between the
methyl groups in terminal and in bridge position is fast,
comparable with the NMR time scale at -75 °C. In
contrast to the condensed phase experiments the fa-
vored structure of TMA in the vapor phase has not been
proved unambiguously. In 1941 Laubengayer and Gil-
(
1) Laubengayer, A. W.; Gilliam, W. F. J . Am. Chem. Soc. 1941, 63,
4
77.
(
(
(
2) Quincke, G. Ber. 1889, 22, 551.
3) Wartik, T.; Schlesinger, H. I. J . Am. Chem. Soc. 1953, 75, 835.
4) Downs, A. J .; Greene, T. M.; Collin, S. E.; Whitehurst, L. A.;
Brain, P. T.; Morrison, C. A.; Pulham, C. R.; Smart, B. A.; Rankin, D.
W. H.; Keys, A.; Barron, A. R. Organometallics 2000, 19, 527.
1
liam have demonstrated by vapor density measure-
(
(
(
(
5) Longuet-Higgins, H. C. J . Chem. Soc. 1946, 139.
ments that TMA forms dimers at 70 °C in the vapor
6) Pitzer, K. S.; Gutowsky, H. S. J . Am. Chem. Soc. 1946, 68, 2204.
7) Hoffmann, E. G. Liebigs Ann. Chem. 1960, 629, 104.
8) Brockway, L. O.; Davidson, N. R. J . Am. Chem. Soc. 1941, 63,
(21) Gray, A. P. Can. J . Chem. 1963, 41, 1511.
(22) Hoffmann, E. G.; Schomburg, G. Z. Elektrochem. 1957, 61, 1101.
(23) Schoetter, H. W.; Hoffmann, E. G. Ber. Bunsen-Ges. Phys.
Chem. 1964, 68, 627.
(24) Grady, A. S.; Puntambekar, S. G.; Russel, D. K. Spectrochim.
Acta 1991, 47A, 47.
(25) Feher, F.; Kolb, W. Naturwissenschaften 1939, 27, 616.
(26) Kohlrausch, K. W. F.; Wagner, J . Z. Physik. Chem. 1942, B52,
185.
(27) Muller, N.; Pritchard, D. E. J . Am. Chem. Soc. 1960, 82, 248.
(28) Groenewege, M. P.; Smidt J .; deVries H. J . Am. Chem. Soc.
1960, 82, 4425.
(29) Brownstein, S.; Smith B. C.; Erlich, G.; Laubengayer, A. W. J .
Am. Chem. Soc. 1960, 82, 1000.
(30) Barker, G. K.; Lappert, M. F.; Pedley, J . B.; Sharp, G. J .;
Westwood, N. P. C. J . Chem. Soc., Dalton Trans. 1975, 1765.
(31) Berthomieu, D.; Bacquet, Y.; Pedocchi, L.; Goursot, A. J . Phys.
Chem. A 1998, 102, 7821.
3
287.
(
(
(
9) Buraway, A. Nature 1945, 155, 269.
10) Skinner, H. A.; Sutton, L. E. Nature 1945, 156, 601.
11) Almenningen, A.; Halvorsen, S.; Haaland, A. Acta Chem. Scand.
1
971, 25, 1937.
12) Almenningen, A.; Anderson, G. A.; Forgaard, F. R.; Haaland,
A. Acta Chem. Scand. 1972, 26, 2315.
13) McGrady, G. S.; Turner, J . F. C.; Ibberson, R. M.; Prager, M.
Organometallics 2000, 19, 4398.
(
(
(
(
(
14) Lewis, P. H.; Rundle, R. E. J . Chem. Phys. 1953, 21, 986.
15) Vranka, R. G.; Amma, E. L. J . Am. Chem. Soc. 1967, 89, 3121.
16) Byram, S. K.; Fawcett, J . K.; Nyburg, S. C.; O’Brien, R. J . J .
Chem. Soc., Chem. Commun. 1970, 16.
17) Huffman, J . C.; Streib, W. E. J . Chem. Soc., Chem. Commun.
971, 911.
(
1
(
(
(
18) Winters, R. E.; Kiser, R. W. J . Organomet. Chem. 1967, 10, 7.
19) Tanaka, J .; Smith, S. R. Inorg. Chem. 1969, 8, 265.
20) Chambers, D. B.; Coates, G. E.; Glockling, F.; Weston, M. J .
(32) Balasubramanian, S.; Mundy, C. J .; Klein, M. L. J . Phys. Chem.
B 1998, 102, 10136.
Chem. Soc. A 1969, 1712.
1
0.1021/om010994h CCC: $22.00 © 2002 American Chemical Society
Publication on Web 05/30/2002

2
752 Organometallics, Vol. 21, No. 13, 2002
Vass et al.
gas phase electron diffraction investigations12 also sup-
ported the DMA dimer formation around this temper-
ature. At 83 °C the measured degree of association is
2
.44. On the basis of these facts the DMA has to be
considered as a mixture of dimers and trimers at lower
2
2-24
temperatures. Raman and infrared spectroscopy
4
also supported this observation. Downs et al. published
recently a comprehensive work of combined infrared
spectroscopic, NMR, mass spectrometric, gas phase
electron diffraction investigations together with vapor
pressure measurements, and ab initio calculations.
According to this work DMA exists in the form of dimers
and higher oligomers (e.g., tetramers) in hydrocarbon
solutions, while it is present as a mixture of dimers and
trimers (7:3) at 60 °C. At 200 °C the vapor phase
consists of dimers. It should also be noted that in
contrast to former mass spectrometric investigations1
they have observed the trimer ions with relatively high
intensity in the mass spectrum as well.
F igu r e 1. Geometry and the most important structural
parameters (as obtained at the HF/6-31G** level of theory)
of the (a) monomer and (b) dimer of trimethylaluminum
(TMA) and the (c) monomer, (d) dimer, and (e) trimer of
9,20
dimethylaluminum hydride (DMA). The corresponding
values of the ethyl analogues are given in parentheses.
phase, and by increasing the temperature the degree
of association decreases rapidly. In good agreement with
the age-old experiments of Quincke, they found the
average degree of association to be 1.6 at 140 °C.
Although there was no agreement concerning the mo-
lecular geometry of the TMA dimer for a long time,
dimers have been found by gas phase electron diffraction
The association capability of dialkylaluminum hy-
drides is higher than that of trialkylaluminum com-
pounds. This is due to two reasons. First, it is originated
from the smaller steric demand of the hydrogen atoms
in the bridge position. Second, the pyramidalization
around the Al atom, taking place during dimerization,
requires less energy in the case of dialkylaluminum
hydrides.
2
8
-11
25,26
measurements
and by Raman
and infrared
spectroscopy21 as well. Contrary to the above observa-
Diethylaluminum hydride (DEA) exists in the form
1
8
33,35
tions, Winters and Kiser as well as Tanaka and
of trimers in inert solvents,
while trimers in the gas
Smith,19 evaporating the sample at 190 °C and 60-100
phase have been observed only in a very low ratio,
among dimers and monomers measured by mass spec-
°
C, respectively, observed the dimers only in a negligible
1
9,20
ratio, <1%, by mass spectrometry. Concurrently Cham-
bers et al.20 performed a mass spectrometric investiga-
tion, in which the sample was evaporated at 44-50 °C.
Even in this experiment a monomer/dimer ratio of only
trometry.
Among the higher analogues di(tert-butyl)aluminum
hydride was investigated, which forms trimers both in
3
3
36
solution and in the solid phase.
9
0:3.5 was found. In the case of the mass spectrometric
The purpose of this paper was to study the electronic
structure of some trialkylaluminum and dialkylalumi-
num hydride compounds, namely, TMA, TEA, DMA,
and DEA, in the gas phase by HeI photoelectron
spectroscopy. An effort was made to observe not only
the monomers but the above-mentioned oligomers
(dimers, trimers) as well. The association capability of
the investigated compounds was studied by recording
the spectra as a function of temperature. Similarly to
investigation, the low dimer abundance can be the
consequence of dissociation caused by the low vapor
pressure and/or by the electron impact ionization.
Nevertheless, at higher sample pressure, used in HeI
photoelectron spectroscopy, Barker et al.30 could not
observe the dimer either.
Although the dissociation energy of triethylaluminum
(
TEA) dimer is significantly lower than that of TMA
3
7
dimer, similarly to TMA, TEA forms dimers in benzene
our recent study on alkyllithium clusters, the effects
of substituents and degree of association on the elec-
tronic structure have also been analyzed.
solutions.6 Vapor density measurements showed that
,7
1
the abundance of TEA dimer in the gas phase is only
1
2% at 150 °C. Using mass spectrometry no dimer could
2
0
be observed at all.
Exp er im en ta l Section
Considering the analogues with bulkier substituents,
the equilibrium between the dimers and monomers is
shifted to the monomers (even in solutions) in the case
of tripropyl- and tributylaluminum, while tri(isopropyl)-,
tri(isobutyl)-, and tri(tert-butyl)aluminum do not dimer-
A. Sa m p le P r ep a r a tion . TMA (97%) and TEA (93%) were
obtained from Aldrich and were used without further purifica-
tion. Dialkylaluminum hydrides were prepared by the method
3
of Wartik and Schlesinger, applying the modification by Grady
24
et al., namely, by the solvent-free heterogeneous reaction of
excess LiAlH and the corresponding trialkylaluminum at 90
4
5
,6,33,34
ize.
According to the vapor pressure depression measure-
ments of Wartik and Schlesinger,3 the degree of as-
sociation of dimethylaluminum hydride (DMA) in iso-
pentane at room temperature is between 3.55 and 3.13
depending on the concentration, while they found it to
be 2.04 by vapor density measurements at 167 °C. The
°C. The products were purified by trap-to-trap distillation
under vacuum and were collected at -19 °C (DMA) and 0 °C
(DEA), while the traces of the unreacted more volatile trialkyl-
aluminum compounds were entrapped by liquid nitrogen. All
operations were carried out in an inert atmosphere.
(
35) Coates, G. E.; Green, M. L. H.; Wade. K. Organometallic
Compounds; Methuen, 1967; Vol. 1.
(
33) Wilkinson, G., Stone, F. G. A., Abel, E. W., Eds. Comprehensive
Organometallic Chemistry; Elsevier Scientific Ltd., 1982; Vol. 1, p 583.
34) Smith, M. B. J . Organomet. Chem. 1970, 22, 273.
(36) Uhl, W. Z. Anorg. Allg. Chem. 1989, 570, 37.
(37) Tarczay, G.; Vass, G.; Magyarfalvi, G.; Szepes, L. Organome-
tallics 2000, 19, 3925.
(

Trialkylaluminum Hydride Compounds
Organometallics, Vol. 21, No. 13, 2002 2753
Ta ble 1. Ca lcu la ted a n d Exp er im en ta l Ver tica l Ion iza tion En er gies (in eV) of th e Mon om er a n d Dim er of
a
Tr im eth yla lu m in u m (TMA)
calculated
statec
assignmentb
HF/Koopmans
OVGF
EOM-CCSD
experimental
Monomer (C3h)
E′
A′
σAl-C
σAl-C
σC-H
10.47
13.78
14.1-14.6
9.54
12.83
13.2-13.7
9.57
12.81
13.2-13.7
9.85
12.6
13.3-14.0
Dimer (C2v)
B2
A1
B1
A1
B2
σAl-C
10.49
10.51
11.08
11.22
12.05
14.1-16.3
9.39
9.44
9.94
10.09
10.96
13.1-15.0
9.42
9.46
9.98
9.75
σAl-C
σAl-µC-Al
σAl-C + σµC-µC
σAl-C
10.45
10.96
13.0-15.0
11.2
12.0-16.2
σC-H
a
The calculations have been performed for the HF/6-31G** optimized geometries, using the 6-31G** basis set for HF/Koopmans, OVGF,
and EOM calculations. Assignments are based on the population analysis of the HF orbitals. c Given for the lowest, experimentally
b
separable states only.
B. HeI P h otoelectr on Sp ectr oscop ic Mea su r em en ts.
atory calculations were performed using 6-31G**, cc-pVDZ,
and aug-cc-pVDZ basis sets and the HF/6-31G** and MP2/6-
31G** optimized structures. Since the maximum difference
between these results does not exceed 0.3 eV, calculations for
all other species were performed for the HF/6-31G** optimized
structures only. For the ionization energy calculations the
6-31G** basis set was used.
(As Supporting Information HOMO and LUMO energies,
which can be useful in order to estimate the gas phase Lewis-
acid character of the investigated compounds, are also pre-
sented.)
HeI photoelectron (PE) spectra were recorded on an ATOMKI
3
8
ESA-32 spectrometer using a pyrolizer inlet system heatable
up to 350 °C. The accuracy of the temperature measurement
is estimated to be (10 °C. The sample pressure at the
-
6
-5
ionization chamber was kept between 8 × 10 and 4 × 10
mbar in each experiment. The full width at half-maximum of
the P3/2 peak of argon was 20-50 meV during the measure-
2
ments of TMA and DMA and 50-100 meV in the case of TEA
and DEA. The maximum error in ionization energies is
estimated to be less than 0.05-0.1 eV. All spectra were
calibrated against argon or nitrogen as internal standards; the
background corrections were made by Shirley-type functions.
The spectra of the dimers have been obtained by subtracting
the spectra recorded at high temperatures from the low-
temperature ones.
Resu lts a n d Discu ssion
A. Tr im eth yla lu m in u m . Berthomieu et al.31 have
recently calculated the first vertical and adiabatic
ionization energy of the TMA monomer and dimer by
∆HF, ∆MP2, and ∆DFT methods. Despite the small
difference (<0.1 eV) between the first vertical ionization
energies obtained by ∆HF and ∆MP2 methods, they
have concluded on the basis of ∆DFT calculations that
C. Ab In itio Ca lcu la tion s. Ab initio geometry optimiza-
3
9
tions were carried out by the PQS 2.1 program system. Outer-
4
0
valence Green’s function (OVGF) ionization energies have
4
1
been obtained by Gaussian98, while the ACESII program
4
2
package was used to compute ionization energies in the
framework of the equation of motion (EOM) CCSD method.43
Geometry optimizations have been performed for the ex-
“the [first] ionization energy of the dimer is lower than
4
,12
4,31
perimentally found
and theoretically confirmed
global
that of the monomer, by about 1 eV, which indicates
minimum structures (Figure 1) only. The optimizations were
followed by second derivative calculations to determine whether
the obtained stationary points correspond to minima. The
dependence of the calculated ionization energies on the
geometry and on the basis set was verified by OVGF calcula-
tions for the monomer and the dimer of TMA. These explor-
+
+
that TMA and TMA2 could be separated by ionization
techniques”. Our ionization energy calculations (Table
1
) do not support their conclusion, but they are in good
agreement with their ∆HF and ∆MP2 results. Although
the first ionization energy of the monomer and the dimer
might not be significantly different, our calculations also
predict regions in the photoelectron spectrum where the
dimer could be distinguished from the monomer unam-
biguously. These characteristic bands of the dimer are
expected to appear at about 10.0, 11.0, and above 14.0
eV, and they are separated from the closest bands of
the monomer by about 0.5, 1.5, and 0.5-0.9 eV, respec-
tively.
(
38) Cs a´ kv a´ ri, B.; Nagy, A.; Zanathy, L.; Szepes, L. Magyar K e´ miai
Foly o´ irat 1992, 98, 415.
39) PQS version 2.1; Parallel Quantum Solutions: Fayetteville, AR,
998.
40) (a) Cederbaum, L. S. Adv. Chem. Phys. 1977, 36, 205. (b) von
Niessen, W.; Schirmer, J .; Cederbaum, L. S. Comput. Phys. Rep. 1984,
, 57. (c) Oritz, J . V. J . Chem. Phys. 1988, 89, 6348. (d) Zakrzewski, V.
G.; von Niessen, W. J . Comput. Chem. 1993, 14, 13.
41) Frisch, M. J .; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
(
1
(
1
(
Robb, M. A.; Cheeseman, J . R.; Zakrzewski, V. G.; Montgomery, J . A.,
J r.; Stratmann, R. E.; Burant, J . C.; Dapprich, S.; Millam, J . M.;
Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J .;
Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo,
C.; Clifford, S.; Ochterski, J .; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J . B.; Cioslowski, J .; Ortiz, J . V.; Baboul, A. G.; Stefanov,
B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.;
Martin, R. L.; Fox, D. J .; Keith, T.; Al-Laham, M. A.; Peng, C. Y.;
Nanayakkara, A.; Gonzalez, C.; Challacombe, M.; Gill, P. M. W.;
J ohnson, B.; Chen, W.; Wong, M. W.; Andres, J . L.; Gonzalez, C.; Head-
Gordon, M.; Replogle, E. S.; Pople, J . A. Gaussian 98, Revision A.7;
Gaussian, Inc.: Pittsburgh, PA, 1998.
The experimental photoelectron spectra of TMA re-
corded at different temperatures are shown in Figure
2
. In the low-temperature spectra the above-mentioned
bands, characteristic for the dimer, can be observed at
only slightly higher ionization energies than predicted
by our calculations. Furthermore, the intensity of these
bands decreases rapidly with the increase of tempera-
ture, indicating the dissociation of the dimers. At about
75 °C the dissociation of the dimers is completed (i.e.,
the dimer bands are not recognizable in the spectrum,
which means that the dimer concentration is below a
few percent). Above this temperature, up to 200 °C, the
(
42) Stanton, J . F.; Gauss, J .; Watts, J . D.; Lauderdale, W. J .;
Bartlett, R. J . Int. J . Quantum Chem., Quantum Chem. Symp. 1992,
6, 897.
43) Stanton, J . F.; Bartlett, R. J . J . Chem. Phys. 1993, 98, 7029.
2
(

2
754 Organometallics, Vol. 21, No. 13, 2002
Vass et al.
F igu r e 4. HeI photoelectron spectrum of the monomer of
triethylaluminum (TEA).
compounds of group 13 elements (B, Al, Ga).30 In the
C2v symmetry dimer two of the degenerate orbitals of
the monomer units form four nondegenerate orbitals.
Three of these are well recognizable in the low-energy
region of the dimer spectrum. The fourth band is faded
into the higher energy broad band assigned to the
ionization of C-H bonds, which is wider by about 1-2
eV than the corresponding band of the monomer.
As it can be seen from Table 1, the HF/Koopmans
calculations predict correctly the main spectral features.
Furthermore, as far as absolute values are concerned,
both the OVGF and the EOM-CCSD calculations are
in very good agreement with the experimental data; the
maximum deviation from the measured ionization ener-
gies is less than 0.4 eV. The difference between the first
vertical ionization energy of the monomer and dimer is
only about 0.1 eV. This experimental result clearly
shows that the performance of the ∆DFT method to
predict this energy difference is much worse than that
of any other methods, including ∆HF, ∆MP2, OVGF,
EOM-CCSD, and even HF/Koopmans theories. There-
fore, the above-mentioned conclusion of Berthomieu et
F igu r e 2. Temperature-dependent HeI photoelectron
spectrum of trimethylaluminum (TMA).
3
1
al. based on ∆DFT method is essentially erroneous.
B. Tr ieth yla lu m in u m . To obtain reasonable signal-
to-noise ratio, the sample has to be heated at least to
5
0-60 °C. By heating the path between the sample and
the ionization chamber up to 200 °C no change of the
photoelectron spectrum was experienced. This observa-
tion confirms that already at 50-60 °C the pure
monomer spectrum was recorded (Figure 4). This is also
supported by the fact that the calculations predict a
band at 10.1 eV for the dimer, which cannot be observed
in the experimental spectrum. Furthermore, the calcu-
lated ionization energies obtained by the OVGF and the
EOM-CCSD methods for the monomer are in excellent
accordance with the experimental spectrum (Table 2).
Similarly to TMA, the maximum error in calculated
ionization energies is no more than 0.4 eV.
The main difference between the spectrum of TMA
and TEA monomers is that the first ionization energy
of TEA is lower almost by 1 eV, which is a consequence
of the larger electron-releasing effect of the ethyl group.
Furthermore, the broad band in the higher energy
region of the spectrum of TMA, corresponding to the
ionization of CH3 groups, is split into two parts in the
case of TEA. At the higher energy region, the struc-
tureless part between 13.8 and 16.0 eV belongs to the
ionization of the CH3 groups (C-H bonds). In the lower
energy part, between 11.0 and 13.5 eV, three bands can
be identified at 11.4, 11.9, and 12.65 eV (see Table 2).
The first of these (11.4 eV), together with the first band
F igu r e 3. HeI photoelectron spectrum of the (a) monomer
and (b) dimer of trimethylaluminum (TMA).
spectrum does not change any more; consequently at
higher temperatures the spectrum of the pure monomer
was obtained. This spectrum is in good agreement with
the previously published photoelectron spectrum of
3
0
TMA monomer. The spectra recorded below 75 °C can
be related to a mixture of monomers and dimers.
The pure dimer spectrum was obtained by subtracting
a high-temperature spectrum from a low-temperature
one. This dimer spectrum together with the spectrum
of the monomer is presented in Figure 3; the experi-
mental vertical ionization energies are collected in Table
1
. The spectrum of the monomer consists of three bands.
The first two bands correspond to the ionization of the
Al-C bonds, while the third band, which overlaps with
the second one, is originated from the ionization of the
C-H bonds. As can be seen from Table 1, the first band
of the monomer corresponds to the ionization from a
degenerate orbital. As a consequence, this band is
slightly asymmetrical due to J ahn-Teller splitting. The
same structure is evident in the spectra of trimethyl

Trialkylaluminum Hydride Compounds
Organometallics, Vol. 21, No. 13, 2002 2755
Ta ble 2. Ca lcu la ted a n d Exp er im en ta l Ver tica l Ion iza tion En er gies (in eV) of th e Mon om er a n d Dim er of
a
Tr ieth yla lu m in u m (TEA)
calculated
state
assignment
HF/Koopmans
OVGF
EOM-CCSD
8.82
experimental
Monomer (C3h)
E′
A′
A′′
E′′
E′
A′
σAl-C
9.82
12.49
12.93
12.95
13.47
13.42
15.3-16.1
8.82
11.58
12.07
12.07
12.26
12.27
14.2-14.9
9.05
11.4
11.9
σAl-C
σC-H
12.12
12.12
12.28
12.28
σC-H
σC-C
12.65
σC-C + σAl-C
σC-H
14.1-14.9
13.8-16.0
Dimer (C2)
B
A
B
A
B
σAl-C
9.76
9.88
10.33
10.52
11.29
12.7-17.3
8.58
8.72
9.11
9.31
10.14
11.7-15.7
not
σAl-C
σAl-µC- Al
σAl-C + σAl-µC- Al
σAl-C
observed
σC-H, σC-C
a
See footnotes of Table 1.
the bands at about 10.5, 11.3, and 12.1 eV. Since all
these bands, within reasonable accuracy, are predicted
for the dimer by our calculations, consequently this
change is due to the dissociation of the dimers. Al-
though, this is in contrast to the observation of Downs
4
et al., who, as it was mentioned above, found predomi-
nantly dimers in the vapor phase at 200 °C. This is
probably due to the different conditions (e.g., lower
pressure) applied in our experiment.
The photoelectron spectra of the monomer and the
dimer of DMA, obtained from the temperature-depend-
ent spectra, are shown in Figure 6. (Due to the complex-
ity of the low-temperature spectra, it was not possible
to get the trimer spectrum.) The spectrum of the
monomer looks very similar to the TMA monomer
spectrum; all of the bands are shifted only by 0.1-0.3
eV to higher energy. A minor difference is observed for
the first band, which is less symmetrical in the case of
DMA. This is due to the Al-H bond and the lower
symmetry of the DMA monomer as compared to TMA.
This results in a splitting (predicted to be 0.4-0.5 eV
by the calculations) of the ionic states corresponding to
the ionization of Al-C bonds, which were degenerate
in the case of TMA.
F igu r e 5. Temperature-dependent HeI photoelectron
spectrum of dimethylaluminum hydride (DMA). Arrows on
the spectrum recorded at 25 °C indicate the change as
compared to higher temperature spectra due to the decom-
position of the trimers.
The difference is appreciably larger in the case of the
dimers, which is a result of the much stronger interac-
tion between the two monomer units in the DMA dimer.
The splitting of the first band corresponding to the
ionization of the Al-C (and Al-H) bonds is larger by
more than 0.5 eV in the DMA dimer spectrum compared
to that of TMA dimer.
The OVGF and EOM-CCSD ionization energies agree
very well with the experimental values. The only
exception is the value obtained for the ionization of the
Al-H bond, which is higher by 0.6 eV than the experi-
mental value.
D. Dieth yla lu m in u m Hyd r id e. The calculated and
experimental ionization energies of DEA are presented
in Table 4. (The photoelectron spectrum of DEA as a
function of temperature can be found in the Supporting
Information. Due to the low volatility of DEA, the
sample holder had to be heated at least to 40 °C.)
Similarly to the observations made in the case of TMA
and DMA, the theoretically predicted bands character-
istic for the dimers can be identified in the spectra
of the spectrum, corresponds to the ionization of the
Al-C bonds, while the second two belong to the ioniza-
tion of the ethyl groups (C-C and C-H bonds).
C. Dim eth yla lu m in u m Hyd r id e. The photoelectron
spectra of dimethylaluminum hydride recorded at dif-
ferent temperatures are presented in Figure 5. There
are two temperature regions where spectrum changes
are visible. The first one is between 25 and 55 °C. This
change is not dramatic; it can be imperceptibly seen on
the side of some bands (see arrows on spectrum recorded
at 25 °C). Since the calculations (Table 3) predict only
marginal differences between the spectrum of the dimer
and trimer of DMA, this change is most likely due to
the decomposition of the trimers. This is also in agree-
3
,4
ment with experiments available from literature,
which indicate the dissociation of the trimers around
this temperature.
The second, much more pronounced change in the
spectrum starts at about 125 °C, and it lasts out even
to 180-200 °C. This change includes the vanishing of

2
756 Organometallics, Vol. 21, No. 13, 2002
Vass et al.
Ta ble 3. Ca lcu la ted a n d Exp er im en ta l Ver tica l Ion iza tion En er gies (in eV) of th e Mon om er , Dim er , a n d
a
Tr im er of Dim eth yla lu m in u m Hyd r id e (DMA)
calculated
state
assignment
HF/Koopmans
OVGF
EOM-CCSD
experimental
Monomer (C2v)
B2
A1
A1
σAl-C
10.67
10.94
14.28
14.4-14.8
9.80
10.30
13.50
13.5-13.9
9.83
10.23
13.41
13.5-13.9
9.9
10.2
12.9
12.2-15.0
σAl-C
σAl-H + σAl-C
σC-H
Dimer (D2h)
B1g
B2u
Ag
B1u
B3u
σAl-C
10.63
10.89
11.34
12.20
13.01
14.3-17.6
9.62
9.92
10.36
11.40
12.02
13.3-16.2
9.64
9.93
10.35
11.28
12.00
13.3-16.1
9.7
10.1
10.5
11.3
12.05
12.6-15.3
σAl-C
σAl-C + σAl-µH-Al
σAl-µH-Al
σAl-C
σC-H (σAl-µH-Al)
Trimer (D3h)
E′′
A2′′
A1′
E′
σAl-C
10.74
10.91
11.34
11.75
14.31
14.4-17.9
14.44
9.67
9.86
σAl-C
σAl-C + σAl-µH-Al
σAl-µH-Al + σAl-C
σAl-µH-Al + σC-H
σC-H (σAl-µH-Al)
σC-H
10.28
10.73
13.29
13.9-16.5
13.46
[10.3]
E′
a
See footnotes of Table 1.
Ta ble 4. Ca lcu la ted a n d Exp er im en ta l Ver tica l Ion iza tion En er gies (in eV) of th e Mon om er a n d Dim er of
a
Dieth yla lu m in u m Hyd r id e (DEA)
calculated
state
assignment
HF/Koopmans
OVGF
EOM-CCSD
experimental
Monomer (C2v)
B2
A1
σAl-C
9.97
10.63
12.9-13.8
15.3-16.0
9.04
9.88
12.1-12.7
14.3-15.0
9.05
9.82
12.1-12.7
14.3-14.8
9.0
9.6
11.8-12.8
14.3-15.0
σAl-C
σC-C, σAl-H
σC-H
Dimer (D2)
B1
B3
A
B2
B1
σAl-C
9.93
10.18
10.70
11.98
12.17
13.0-13.6
15.3-17.6
8.88
9.16
9.65
11.00
11.24
12.1-12.5
14.2-16.1
9.0
9.5
10.0
10.7
σAl-C
σAl-C + σµH-µH
σAl-C
σAl-µH-Al
σC-C, σC-H
σC-H (σAl-µH-Al)
11.8-12.6
14.1-15.0
a
See footnotes of Table 1.
F igu r e 7. HeI photoelectron spectrum of the (a) monomer
and (b) dimer of diethylaluminum hydride (DEA).
F igu r e 6. HeI photoelectron spectrum of the (a) monomer
and (b) dimer of dimethylaluminum hydride (DMA).
(11.8-12.8 and 14.3-15 eV) corresponding to the ethyl
recorded at lower temperatures. Changes in the spec-
trum can be tracked up to 60-70 °C; at this temperature
the dissociation seems to be complete. The photoelectron
spectrum of the DEA monomer and dimer, obtained
from the temperature-dependent spectrum, is presented
in Figure 7. The two-band structure at higher energies
groups and the splitting of the first band of the monomer
are well recognizable in the monomer spectrum. Due
to the larger difference in the inductive effect of hydro-
gen and the ethyl group, this splitting is larger in this
case than the case of DMA; it is about 0.6-0.8 eV.
Comparing the spectrum of DEA to that of DMA, the

Trialkylaluminum Hydride Compounds
Organometallics, Vol. 21, No. 13, 2002 2757
shift of the bands (about 1 eV) due to the larger
inductive effect of the ethyl groups is also conspicuous.
Similarly to the observations made in the case of the
TEA and DMA monomer, and in line with the theoreti-
cal predictions, the first ionization energy of the DEA
dimer is lower only by about 0.2 eV than that of the
DEA monomer. In this case too, the difference between
experimental data and calculated ionization energies
resulting from OVGF as well as EOM-CCSD methods
is less than 0.4 eV. Similarly to DMA, the largest
disagreement is found in the case of the Al-H bond.
significant in the dimer spectrum. This is a consequence
of the increased interaction between the monomer units
due to the change of methyl bridges for hydrogen
bridges.
The substitution of the methyl groups for ethyl groups
results in a decrease of the ionization energies corre-
sponding to the Al-C bonds by about 1 eV.
Calculations related to the ionization energies of
trimethylaluminum monomer and dimer clearly dem-
onstrate that the performance of the ∆DFT method is
much worse than that of any other methods, including
∆
HF, ∆MP2, OVGF, EOM-CCSD, and even HF/Koop-
mans theories.
Con clu sion s
As far as the relative order and absolute values of
ionization energies are concerned, HF/Koopmans theory
predicts correct order of the ionization energies. The
OVGF and EOM-CCSD results are even better; the
absolute ionization energies obtained by these methods
agree mostly with the experimental values within 0.3-
In this study we have presented the HeI photoelectron
spectrum of trimethylaluminum, triethylaluminum,
dimethylaluminum, and diethylaluminum hydride mono-
mers and trimethylaluminum, dimethylaluminum, and
diethylaluminum hydride dimers, all of which, except
the spectrum of the trimethylaluminum, have not been
published before. On the basis of these experimental
results the following conclusions can be drawn concern-
ing the behavior of the title compounds in the gas phase
under high-vacuum conditions.
0
.4 eV even using a medium-quality basis set (6-31G**).
The only exceptions are the values obtained for the
ionization of the Al-H bond of the dialkylaluminum
hydride monomers, which are higher by 0.6 eV than the
experimental values.
The complete dissociation of the trimethylaluminum
dimer was observed at about 75 °C, while the dissocia-
tion of the triethylaluminum dimer takes place below
Ack n ow led gm en t. We are grateful to PQS llc. for
permission to use their code. This work has been
supported by the Hungarian Scientific Research Fund
5
0-60 °C. In the case of dimethylaluminum hydride
even some signal due to the trimer was observable at
room temperature. The dissociation of its dimer is
completed above 180-200 °C only, while the dissociation
of diethylaluminum hydride dimer already occurs at
about 70 °C. From these observations, in accordance
with the previous experimental and theoretical results
available from the literature, the following order of the
association capability of alkylaluminum compounds can
be established: dimethylaluminum hydride > trimethyl-
ma l ui n mu imn u. m ∼ diethylaluminum hydride > triethylalu-
The dimerization of the compounds results in a
splitting of the bands corresponding to the ionization
of the Al-C bonds. This splitting is larger in the case
of dialkylaluminum hydrides than in the case of tri-
alkylaluminum compounds.
Both the experimental observations and theoretical
calculations indicate only marginal difference between
the ionization energies of the dialkylaluminum hydride
dimers and trimers. This is due to the fact that both
molecules are closely related as far as chemical bonds
and structure are concerned (see geometry parameters
in Figure 1).
The substitution of one alkyl group of trialkylalumi-
num compounds with hydrogen does not result in a
remarkable change in the photoelectron spectrum of the
monomer. In contrast to this, the change is much more
(
grant nos. OTKA T032489 and OTKA F033002).
Note Ad d ed in P r oof. After the submission of the
corrected manuscript we have realized that some ad-
ditional papers were formerly published on the HeI
photoelectron spectrum of the TMA monomer (Wang,
D.; Li, S.; Li, Y.; Zheng, S.; Chen, B.; Ding, C.; Gao, Y.
Chem. Phys. Lett. 1996, 260, 95) and the spectrum of
the TMA dimer (Wang, D.; Qian, X.; Zheng, S.; Shi, Y.
Chem. Phys. Lett. 1997, 277, 502). The monomer spec-
trum recorded by this group is in good agreement with
our measurements. Nevertheless, the dimer spectra
recorded by the two groups show significant discrepan-
cies, including the value of the first vertical ionization
energy. Since our measurements are supported by the
sophisticated EOM-CCSD and OVGF calculations (also
consistent with our results obtained for the dialkylalu-
minum hydrides), we believe that the spectrum recorded
by us is more reliable. This statement will be elaborated
in a follow-up paper or communication after further
theoretical investigations.
Su p p or tin g In for m a tion Ava ila ble: Table of HOMO and
LUMO energies and figure of HeI photoelectron spectrum of
DEA. This material is available free of charge via the Internet
at http://pubs.acs.org.
OM010994H