Matrix isolation of HGaX2 (X = Cl or Br): IR spectroscopy and ab initio calculations

Article abstract of DOI:10.1039/a905183j

High-vacuum thermolyses of the intramolecularly co-ordinated gallanes Me₂N(CH₂)₃GaX₂ with X = Cl 1 or Br 2 were investigated with matrix isolation techniques. Among the products, which have been identified with

Full text of DOI:10.1039/a905183j

View Article Online / Journal Homepage / Table of Contents for this issue
Matrix isolation of HGaX2 (X ؍ Cl or Br): IR spectroscopy and
ab initio calculations
Jens Müller* and Henning Sternkicker
Institut für Anorganische Chemie der RWTH Aachen, Professor-Pirlet-Str. 1, D-52056 Aachen,
Germany. Fax: ϩ49 (0)241/8888288; E-mail: jens.mueller@ac.rwth-aachen.de
Received 28th June 1999, Accepted 8th October 1999
High-vacuum thermolyses of the intramolecularly co-ordinated gallanes Me2N(CH2)3GaX2 with X = Cl 1 or
Br 2 were investigated with matrix isolation techniques. Among the products, which have been identified with
IR spectroscopy, ab initio calculations, and known literature data, are monomeric HGaCl2 and HGaBr2. The
experimental vibrational frequencies of these hydrides are compared with calculated harmonic frequencies at the
MP2(fc)/6-311ϩG(2d,p) and B3LYP/6-311ϩG(2d,p) level of theory. Beside monomeric HGaX2, the argon matrices
ؒ
of the thermolyses experiments contained CH , HCN, H C᎐CH , H C᎐NMe, [H CCHCH ] , H C᎐CHCH , HX,
᎐
᎐
᎐
2
4
2
2
2
2
2
3
and GaX (X = Cl or Br). In the case of GaBr the difference between the known gas-phase value and the measured
IR frequency in argon matrices is discussed and explained with the help of SCRF-B3LYP calculations.
Introduction
Recently, we reported on the high-vacuum thermolyses of
alanes, which were intramolecularly co-ordinated with the
dimethylaminopropyl ligand, eqn. (1).1 The first aim of this
investigation was to get an insight into the fragmentation
pattern of compounds of the type Me2N(CH2)3MX2, because
they had been used as single-source precursors for metal
organic chemical vapor deposition (MOCVD).2 In particular,
GaN, which is a promising material for new micro- and opto-
electronic devices, can be deposited from the diazidogallane
Fig. 1 Experimental and calculated IR spectra: A, products of
thermolysis of compound 2 at ca. 1000 ЊC trapped in argon at 15 K for
assignments ν1–ν6 see Table 1; for wavenumbers of compounds a–h see
Me2N(CH2)3Ga(N3)2.3
We investigated the thermolysis reactions (1) by the appli-
cation of matrix isolation techniques, i.e., volatile products
which emerged from the Al2O3 pyrolysis tube were trapped in
argon matrices at 15 K and identified by IR spectroscopy
with the help of ab initio calculations. Within this investigation
we identified the two monomeric hydrides HAlX2 (X = Cl or
Br); the chloride was already known by an early investigation
of Schnöckel,4 whereas the bromide was described for the
first time.1 Quite probably, these hydrides were formed by
β-hydrogen eliminations (1), but the second product, allyl-
dimethylamine, was not detected in matrices. Instead, we iden-
ؒ
Table 3: [HGaBr2 (ν1–ν6), GaBr (a), HCN (b), [H2CCHCH2] (c), Ga2O
(d), H2C᎐CHCH3 (e), H2C᎐CH2 (f), H2C᎐NMe (g), CH4 (h)]; B,
᎐
᎐
᎐
HGaBr2 calculated at the MP2(fc)/6-311ϩG(2d,p) level (half band
width of 2 cmϪ1).
Results and discussion
The two starting gallanes 1 and 2 are volatile and moisture
sensitive compounds which can be purified by sublimation in
high vacuum. In accordance with the known method to prepare
the chloro compound 1,2a we synthesized analytically pure
bromogallane 2 in high yields starting from Li[(CH2)3NMe2]
and GaBr3.
A series of high-vacuum thermolyses of the precursors 1 and
2, respectively, were carried out. Between ambient temperature
and oven temperatures of ca. 700 ЊC the IR spectra of matrix
isolated precursors were unchanged. Around 800 ЊC new IR
bands appeared, indicating the beginning of fragmentation. At
1000 ЊC the typical IR absorptions of compound 2 could no
longer be detected in argon matrices, however small amounts
of 1 were still detectable after thermolysis. Fig. 1 shows a
typical IR spectrum recorded after a thermolysis of the
bromide 2 at 1000 ЊC. At first glance, the intense IR band at
1991.5 cmϪ1 hints to a species with a Ga–H bond (spectrum A);
this intense band lies in the well known region for gallium–
tified the carbon-containing molecules CH , HCN, H C᎐CH ,
᎐
4
2
2
ؒ
H C᎐NMe, [H CCH CH ] , and H C᎐CHCH . Furthermore,
᎐
᎐
2
2
2
2
2
3
the aluminium halogens AlX, AlX2, and AlX3 were found
among the thermolysis products, presumably as the result of
secondary reactions of HAlX2.
Besides being interested in the fragmentation of single-source
precursors, we began to explore the scope of synthesizing
monomeric hydrides by thermal β-hydrogen eliminations. In
this publication we report on thermolysis experiments with the
chloro- and bromo-gallane, 1 and 2, respectively.
J. Chem. Soc., Dalton Trans., 1999, 4149–4153
This journal is © The Royal Society of Chemistry 1999
4149

View Article Online
Table 1 Experimental IR frequencies and calculated harmonic frequencies of HGaCl2 and HGaBr2
MP2/6-311ϩG(2d,p)
B3LYP/6-311ϩG(2d,p)
Experimental
frequencya/
cmϪ1
frequencya/
cmϪ1
intensity/
km molϪ1
obs.:calc.
ratio
frequencya/
cmϪ1
intensity/
obs.:calc.
ratio
km molϪ1
b
HGaCl2
A1
ν1
ν2
ν3
ν4
ν5
ν6
ν(GaH)
2015.0
414.0
2101.5
420.8
133.7
605.2
454.3
443.2
96.1
28.7
8.9
122.4
141.0
60.1
0.9588
0.9838
2071.9
400.5
132.3
570.3
428.9
401.5
76.5
23.8
7.5
75.2
132.5
36.3
0.9725
1.0337
νsym(GaCl2)
δ(GaCl2)
δ(ClGaH)
νasym(GaCl2)
γ(H)
B1
B2
607.5
437.5
464.0
1.0038
0.9630
1.0470
1.0652
1.0201
1.1557
HGaBr2
A1
ν1
ν2
ν3
ν4
ν5
ν6
ν(GaH)
1991.5
287.5
2087.3
286.8
94.0
605.7
337.6
430.0
110.3
14.8
2.8
158.3
83.0
43.6
0.9541
1.0024
2050.9
272.5
92.2
566.6
322.3
380.0
90.3
12.8
2.2
121.6
79.8
25.4
0.9710
1.0550
νsym(GaBr2)
δ(GaBr2)
δ(BrGaH)
νasym(GaBr2)c
γ(H)
B1
B2
597.0
333.0
445.5
0.9856
0.9864
1.0360
1.0537
1.0332
1.1724
a Listed are the wavenumbers concerning the main isotopomers. b Known frequencies of HGaCl2 from ref. 6 are: 2015.3, 414.3, 437.3, 607.5, and
464.3 cmϪ1. Calculated frequencies of HGaCl2 at the B3LYP/6-31G(d,p) level, see ref. 20. c The νasym(GaBr2) results in two IR bands: the one at
333.0 cmϪ1 is caused by H69Ga79Br79Br, H69Ga79Br81Br, and H69Ga81Br81Br; that at 330.0 cmϪ1 is caused by H71Ga79Br79Br, H71Ga79Br81Br, and
H71Ga81Br81Br.
Table 2 Calculated bond lengths (Å) and angles (Њ) of HGaCl2 and
Table 3 Detected species from the thermolysis of compound 2 at
1000 ЊC (see Fig. 1)
a
HGaBr2
X = Cl
X = Br
Wavenumbers/cmϪ1
Assignment
Ga–H
Ga–X
1.550 (1.542)
2.156 (2.147)
1.553 (1.546)
2.300 (2.299)
2900.5, 2855.0, 1469.5,
1221.0, 1026.5
1440.5, 947.5
1305.0
H2C᎐NMe8
᎐
9,10
H2C᎐CH2
᎐
9
X–Ga–X
115.15 (115.40)
116.20 (116.85)
CH4
11
ؒ
12
801.0
909.5
[H2CCH2CH2]
a MP2(fc)/6-311ϩG(2d,p) and B3LYP/6-311ϩG(2d,p); results of the
B3LYP calculations in parentheses.
H2C᎐CHCH3
᎐
3305.5, 721.0
2568.5, 2453.5, 2431.0
822.5
HCN13
14
(HBr)x
Ga2O15
GaBr
hydrogen stretching vibrations of terminal GaH groups.5
Beside the experimental spectrum A, Fig. 1 shows a theoretical
IR spectrum of monomeric HGaBr2 (C2v symmetry), calculated
at the MP2(fc)/6-311ϩG(2d,p) level of theory. There is a good
match between the calculated and the measured IR spectra
depicted which applies to frequencies and intensities of the IR
bands and shows that monomeric HGaBr2 was formed by the
thermolysis. The experiments with the chloride 1 gave similar
results; i.e. monomeric HGaCl2 was formed and unequivocally
identified by comparison with literature data and ab initio cal-
culations. This gallium hydride was synthesized for the first time
by Schnöckel and co-workers6 photochemically from GaCl
with HCl in argon matrices. Within the experimental error, our
measured frequencies of HGaCl2 match with the published
ones. Table 1 shows the experimental and calculated IR data
of HGaCl2 and HGaBr2; for the prediction of the harmonic
frequencies we chose the MP2 and B3LYP method with the
6-311ϩG(2d,p) basis set.
251.5, 250.0, 248.5
data. The observed:calculated ratios vary in the range of
0.973–1.156 (HGaCl2) and 0.971–1.172 (HGaBr2). The most
problematic mode is the out-of-plane deformation mode ν6
which gave observed:calculated ratios of 1.156 for HGaCl2
and 1.172 for HGaBr2. The respective values of the two alu-
minum compounds HAlCl2 and HAlBr2 are 1.051 and 1.056.1
Even though it is known that the B3LYP method under-
estimates low frequencies,7 ratios of 1.156 and 1.172 are
unusually high. Keeping in mind that the respective values of
1.047 and 1.036 from the MP2 calculations are quite normal
(Table 1), we assume that the unusual ratios mainly result from
an inadequate description of the heavy dihalogenogallanes at
the B3LYP/6-311ϩG(2d,p) level of theory.
As in the case of the corresponding alanes, HAlCl2 and
HAlBr2, the ab initio calculations predict C2v-symmetrical equi-
librium geometries (Table 2), therefore one expects six normal
modes for the two gallium hydrides belonging to the irreducible
representations A1 (3), B1 (2), and B2 (1). Only five of the six
modes could be observed; the frequencies of the XGaX-
deformation vibration ν3 are out of the detectable range of
The first step of the thermolysis is probably a β-hydrogen
elimination, eqn. (2). Since the applied temperatures are quite
high, allyldimethylamine is not stable and fragments to several
240–4000 cmϪ1
.
We found an acceptable fit of the MP2 data and the meas-
ured frequencies, which can be shown by the observed:calcu-
lated ratios of 0.959–1.047 and 0.954–1.036 for HGaCl2 and
HGaBr2, respectively (Table 1). The B3LYP values do not
match with the experimental frequencies as well as the MP2
carbon-containing molecules. We identified the well known
ؒ
species CH , HCN, H C᎐CH , H C᎐NMe, [H CCH CH ] , and
᎐
᎐
4
2
2
2
2
2
2
H C᎐CHCH by comparison with literature data (Table 3). The
᎐
2
3
relative amounts of these compounds are dependent on the
applied thermolysis conditions, but the intensity ratios of IR
4150
J. Chem. Soc., Dalton Trans., 1999, 4149–4153

View Article Online
Table 4 Calculated frequencies of the four isotopomers of GaBr
MP2(fc)/6-311ϩG(3df)
B3LYP/6-311ϩG(3df)
Abundance
(%)
frequency/
cmϪ1
intensity/
km molϪ1
frequency/
cmϪ1
intensity/
Species
km molϪ1
69Ga79Br
69Ga81Br
71Ga79Br
71Ga81Br
30.5
29.9
20.0
19.6
269.4
267.9
267.4
265.8
65.6
64.9
64.7
63.9
254.2
252.7
252.3
250.8
55.9
55.3
55.1
54.4
HCl using the B3LYP method with self consistent reaction
field (SCRF)21 calculations based on the Onsager22 model.20
The authors calculated that the GaCl stretching frequency
decreases by 6.5% on solvation. [The frequencies for GaCl are
352.2 (B3LYP/6-31G(d) level) and 329.2 cmϪ1 (SCRF-B3LYP/
6-31G(d) level; polar sphere with ε = 2; see ref. 20).] We
conducted SCRF calculations for GaBr at the B3LYP/6-
311ϩG(3df) level with a polar sphere with a relative permit-
tivity equal to that of solid argon (ε = 1.5).23 The SCRF
method predicts 246.2 cmϪ1 for the GaBr stretch which is
3.1% lower than the frequency of the undisturbed molecule
(69Ga79Br: 254.2 cmϪ1, Table 4). This shows that the GaBr
frequency is even sensitive to a slightly polar environment.
The calculated decrease of 3.1% should not be overestimated;
it is dependent on the size of the selected cavity and, of
course, on the relative permittivity, whose exact value is
unknown (see Experimental section). Moreover, the calcu-
lation is based on a simple theoretical model, which can not
predict the reality precisely.
Fig. 2 Experimental and calculated isotopic pattern of GaBr: A,
expanded region from Fig. 1; B, calculated isotopic pattern (B3LYP/6-
311ϩG(3df), see Table 4).
After a series of thermolyses experiments the pyrolyses oven
had to be cleaned. Therefore, we tempered the Al2O3 tube at ca.
1000 ЊC in high vacuum, purged it with argon, and trapped all
gases onto the CsI window at 15 K. The most intense signal in
these matrix IR spectra appeared at 822.5 cmϪ1. Through com-
parison with literature data we assigned this band to νasym
(GaO) of Ga2O.15 During a thermolysis experiment a part of
the gallium compounds must be deposited inside the therm-
olysis tube. When this pyrolysis oven is exposed to air all
substituents bound to gallium are probably hydrolysed leaving
a gallium oxide/hydroxide layer behind, which is subsequently
evaporated as Ga2O during the cleaning procedure described
above. The IR band at 822.5 cmϪ1, which was sometimes pres-
ent in matrix IR spectra of thermolysis experiments, can be
suppressed by a thorough preceding cleaning procedure.
bands for a given species are always preserved. The same carbon-
containing molecules were found in the thermolyses experi-
ments on the aluminium compounds mentioned before, eqn.
(1).1 Furthermore, we found HBr and monomeric GaBr in the
argon matrices among the thermolysis products of 2 (Table 3);
one feasible reaction path to these two species is the fragmen-
tation of HGaBr2. Similar results were obtained starting with
the dichlorogallane 1. [GaCl was identified by its IR band at
342.0 cmϪ1 (see ref. 16); we tentatively assigned the IR band
at 2786.0 cmϪ1 to HCl (see ref. 14(b).]
It must be pointed out that the thermolyses of compounds 1
and 2 seams to be quite complex. For example, the fragmen-
tation of the intramolecularly co-ordinated alanes, eqn. (1),
yielded beside the hydrides HAlX2 the three aluminium halogens
AlX, AlX2, and AlX3, with the latter resulting in an intense IR
band.1 In the case of gallium we could only identify GaX;
in particular, there were no IR absorptions of the well known
GaX3.17 One might suppose that if GaX3 was formed in the gas
phase it fragmented, quite probably to a large extent, to GaX
and X2, a reaction which was used to synthesize GaX for gas-
phase spectroscopy.18 Hence, we attribute the main differences
between the results of the thermolyses of intramolecularly co-
ordinated alanes and gallanes to the higher stability of the
oxidation state ϩ for gallium.
Conclusion
Recently, we reported on the syntheses of monomeric hydrides
of aluminium starting from intramolecularly co-ordinated
alanes. Within this paper we have shown that this method can
be transferred to gallium chemistry, i.e. the monomeric hydrides
HGaCl2 and HGaBr2 were obtained thermolytically from the
3-dimethylaminopropylgallanes 1 and 2. So far, only three
monomeric gallanes of the type HaGaX3 Ϫ a have been character-
ized, namely GaH3,24 HGaCl2,6 and H2GaCl;25 with HGaBr2
we characterized a new compound of this class. Surprisingly,
the gallium hydrides HGaCl2 and HGaBr2 withstand the rather
harsh thermolyses conditions.
This report is not the first one concerning β-hydrogen elimin-
ations to synthesize gallium hydrides; e.g., Russel26 investigated
intensively the fragmentation of trialkylgallanes with the infra-
red laser powered homogeneous pyrolysis (IR LPHP) tech-
nique. With this method GaEt3 was thermolysed to give either
(HGaEt2)x or (H2GaEt)x and ethane. The pure hydrides as well
as their NMe3 adducts were identified by FTIR and NMR
spectroscopy. The gallanes are oligomeric, however monomeric
species have not been detected.
A comment is needed on the GaX species. While the matrix
IR spectrum of GaCl is well known,16 the respective data for
GaBr are to our knowledge not published. We unequivocally
identified GaBr in argon matrices by its split IR band at
251.5/250.0/248.5 cmϪ1. This splitting is due to the presence
of four isotopomers. Table 4 shows calculated harmonic
frequencies of GaBr, and Fig. 2 illustrates the fit between the
predicted and the measured isotopic pattern. The measured
frequency of 250.0 cmϪ1 (69Ga81Br) is shifted by 4.9% to lower
frequencies in comparison with the gas-phase value of 263.0
19
cmϪ1
;
a similar trend is known for 69Ga35Cl, 342.9 cmϪ1 in
argon matrices and 365.3 cmϪ1 in the gas phase.20 These are
unusually large differences between gas-phase IR data and a
respective value in a rare gas matrix. Recently, this was
pointed out in a reinvestigation of the reaction of GaCl with
Currently, we are exploring the scope of using β-hydrogen
J. Chem. Soc., Dalton Trans., 1999, 4149–4153
4151

View Article Online
eliminations to synthesize new monomeric hydrides of Group
13 elements.
increased and new broad IR bands appeared indicating that
notable parts of the molecule were no longer matrix isolated
under these conditions.
The IR spectra of the matrices (15 K) were recorded on a
Perkin-Elmer FTIR 1720x instrument from 240 to 4000 cmϪ1
Experimental
General remarks
with a resolution of 1 cmϪ1
.
All procedures for syntheses were carried out under dry nitro-
gen atmospheres with standard Schlenk techniques. Solvents
were dried by standard procedures, distilled, and stored under
nitrogen and molecular sieves (4 Å). The reagents Me2N(CH2)3-
GaCl2 12a and Li(CH2)3NMe2 27 were prepared according to the
literature procedures. NMR: Varian Unity 500 (ambient tem-
perature; 499.843 MHz for 1H and 125.639 MHz for 13C),
calibrated against residual protons of the deuteriated solvents;
1H and 13C chemical shifts are reported relative to TMS. Elem-
ental analysis (C, H, N): Carlo-Erba elemental analyzer, Model
1106. MS: Finnigan MAT 95.
Ab initio calculations
The GAUSSIAN 94 package,28 run on a cluster of work-
stations (Rechenzentrum der RWTH Aachen), was applied
for all ab initio calculations. The total energies Eh (in
Hartrees) and the zero point vibrational energy (in kJ molϪ1; in
parentheses) are as follows: HGaCl2, Ϫ2843.298259 (24.88) for
MP2(fc)/6-311ϩG(2d,p), Ϫ2846.028291 (23.96) for B3LYP/
6-311ϩG(2d,p); HGaBr2, Ϫ7068.959265 (22.98) for MP2(fc)/
6-311ϩG(2d,p), Ϫ7073.862273 (22.04) for B3LYP/6-311ϩ
G(2d,p); GaBr, Ϫ4495.872086 (1.61) for MP2(fc)/6-311ϩG-
(3df), Ϫ4499.089983 (1.52) for B3LYP/6-311ϩG(3df).
The SCRF calculations for GaBr were conducted at the
B3LYP/6-311ϩG(3df) level with a polar sphere of ε = 1.523 and
a radius of 3.45 Å. This radius was obtained by adding 0.50 Å
to the radius gained from the calculated molecular volume
(keyword: volume = tight). The total energy Eh (in Hartrees)
and the ZPVE (in kJ molϪ1; in parentheses) for GaBr with the
reaction field is Ϫ4499.090559 (1.47); the GaBr distance is
2.408 Å.
Dibromo[3-(dimethylamino)propyl]gallium 2
The compound Li(CH2)3NMe2 (0.70 g, 7.52 mmol) was added
to a freshly prepared solution of 2.20 g (7.11 mmol) of GaBr3 in
30 ml of diethyl ether at Ϫ78 ЊC. The mixture was allowed to
warm to room temperature and stirred overnight. After
removal of the solvent in high vacuum, sublimation (80–96 ЊC,
10Ϫ3 mbar) gave 1.85 g (82%) of 2 as a colorless solid, mp 88–
1
3
90 ЊC. H NMR (C6D6): δ 0.71 (t, J = 7.6 Hz, 2 H, GaCH2),
1.06 (m, 2 H, GaCH2CH2), 1.51 (t, 3J = 6.1 Hz, 2 H, NCH2) and
1.75 (s, 6 H, CH3). 13C-{1H} NMR (C6D6): δ 12.72 (GaCH2),
21.45 (GaCH2CH2), 45.99 (CH3) and 61.11 (NCH2). MS (70
eV): m/z (%) 315 (5) [Mϩ], 287 (5) [Mϩ Ϫ C2H4], 236 (16)
[Mϩ Ϫ Br] and 58 (100) (C3H8N). Calc. for C5H12Br2GaN
(315.68): C, 19.02; H, 3.83; N, 4.44. Found: C, 18.70; H, 4.09;
N, 4.39%.
Acknowledgements
We are grateful to the Deutsche Forschungsgemeinschaft and
to the Fonds der Chemischen Industrie for generous financial
support and to the Rechenzentrum der RWTH Aachen for pro-
viding generous computer time. Especially, we would like to
thank T. Eifert and J. Risch for their support, and P. Geisler for
the construction of the thermolysis oven.
Matrix isolation
References
The matrix apparatus consisted of a vacuum line (Leybold
Turbovac 151; Leybold Trivac D4B) and a Displex CSW
202 cryogenic closed-cycle system (APD Cryogenics Inc.) fitted
with CsI windows. In a typical experiment the starting com-
pound was kept in a small metal container in high vacuum at
constant temperature (10Ϫ6 to 10Ϫ7 mbar; compound 1, 55–
65 ЊC; 2, 80–90 ЊC), while a flow of argon was conducted over
the sample (Linde 6.0; flow = 1.0 ml minϪ1 (0 ЊC, 1.013 bar);
MKS mass flow controller type 1179). Subsequently, this gas-
eous mixture was passed through an Al2O3 tube (inner diameter
of 1 mm; heated by tungsten wire coiled around the last 10
mm). The hot end of the pyrolysis tube was just 25 mm away
from the cooled CsI window to assure that the maximum
amount of volatile fragments emerging from the oven was
trapped in the matrix. For the temperature determination of the
pyrolysis oven a current-to-temperature relation was measured
with a thermocouple (Thermocoax: NiCr/NiAl) inside the
Al2O3 tube. For the experiments with compound 2 an oven
fitted with an Al2O3 tube with two parallel, inner canals was
used (outer diameter 4 mm; inner diameter 1 mm each; last 10
mm heated with a tungsten wire), with one of the inner canals
equipped with a thermocouple (Thermocoax: NiCrSi/NiSi)
and through the other canal the argon/compound mixture was
conducted. With this set-up reliable thermolysis temperatures
could be measured during the experiment without contact
between the substance and the thermocouple.
1 J. Müller and B. Wittig, Eur. J. Inorg. Chem., 1998, 1807.
2 (a) L. Pohl, M. Hostalek, H. Schumann, U. Hartmann, W.
Wassermann, A. Brauers, G. K. Regel, R. Hövel, P. Balk and
F. Scholz, J. Cryst. Growth, 1991, 107, 309; (b) H. Schumann,
U. Hartmann, W. Wassermann, O. Just, A. Dietrich, L. Pohl,
M. Hostalek and M. Lokai, Chem. Ber., 1991, 124, 1113.
3 A. Miehr, M. R. Mattner and R. A. Fischer, Organometallics, 1996,
15, 2053; A. Miehr, O. Ambacher, W. Rieger, T. Metzger, E. Born
and R. A. Fischer, Chem. Vap. Deposition, 1996, 2, 51; R. A.
Fischer, A. Miehr, E. Herdtweck, M. R. Mattner, O. Ambacher,
T. Metzger, E. Born, S. Weinkauf, C. R. Pulham and S. Parsons,
Chem. Eur. J., 1996, 2, 1353.
4 H. Schnöckel, J. Mol. Struct., 1978, 50, 275.
5 A. J. Downs and C. R. Pulham, Adv. Inorg. Chem., 1994, 41, 171.
6 R. Köppe, M. Tacke and H. Schnöckel, Z. Anorg. Allg. Chem., 1991,
605, 35.
7 A. Y. Timoshkin, H. F. Bettinger and H. F. Schaefer III, J. Am.
Chem. Soc., 1997, 119, 5668.
8 J. Hinze and R. F. Curl, Jr., J. Am. Chem. Soc., 1964, 86, 5068;
I. Stolkin, T.-K. Ha and H. H. Günthard, Chem. Phys., 1977, 21,
327.
9 G. Herzberg, Molecular Spectra and Molecular Structure, D. Van
Nostrand Reinhold Company, Inc., Princeton, NJ, 1st edn., 1945,
vol. II.
10 L. Andrews, G. L. Johnson and B. J. Kelsall, J. Chem. Phys., 1982,
76, 5767.
11 G. Maier, H. P. Reisenauer, B. Rohde and K. Dehnicke, Chem. Ber.,
1983, 116, 732; J. W. Huang and W. R. M. Graham, J. Chem. Phys.,
1990, 93, 1583.
The concentrations of the samples in the matrices are
unknown. Therefore, several experiments under various condi-
tions were carried out to assure that the molecules were indeed
matrix isolated. The concentration of the samples decreases
significantly at lower sublimation temperatures (40 ЊC for 1 and
70 ЊC for 2), but no differences in the IR spectra were observed.
At higher concentrations the half band widths of IR bands
12 L. Andrews, G. L. Johnson and B. J. Kelsall, J. Am. Chem. Soc.,
1982, 104, 6180.
13 D. E. Milligan and M. E. Jacox, J. Chem. Phys., 1963, 39, 712;
D. V. Lanzisera and L. Andrews, J. Phys. Chem. A, 1997, 101,
824.
14 (a) M. T. Bowers and W. H. Flygare, J. Chem. Phys., 1966, 44, 1389;
(b) A. J. Barnes, H. E. Hallam and G. F. Scrimshaw, Trans. Faraday
Soc., 1969, 65, 3150.
4152
J. Chem. Soc., Dalton Trans., 1999, 4149–4153

View Article Online
15 D. M. Makowiecki, D. A. Lynch, Jr. and K. D. Carlson, J. Phys.
Chem., 1971, 75, 1963; A. J. Hinchcliffe and J. S. Ogden, J. Phys.
Chem., 1973, 77, 2537; T. R. Burkholder, J. T. Yustein and
L. Andrews, J. Phys. Chem., 1992, 96, 10189.
16 E. D. Samsonova, S. B. Osin and V. F. Pevel’kov, Russ. J. Inorg.
Chem., 1988, 33, 1598.
24 P. Pullumbi, Y. Bouteiller, L. Manceron and C. Mijoule, Chem.
Phys., 1994, 185, 25.
25 R. Köppe and H. Schnöckel, J. Chem. Soc., Dalton Trans., 1992,
3393.
26 D. K. Russell, (a) Coord. Chem. Rev., 1992, 112, 131; (b) Chem. Vap.
Deposition, 1996, 2, 223.
17 R. G. S. Pong, R. A. Stachnik, A. E. Shirk and J. S. Shirk,
J. Chem. Phys., 1975, 63, 1525.
18 A. Petrikaln and J. Hochberg, Z. Phys., 1933, 86, 226.
19 K. P. Huber and G. Herzberg, Molecular Spectra and Molecular
Structure, D. Van Nostrand Reinhold Company, Inc., Princeton,
NJ, 1979, vol. IV.
20 M. Tacke, J. P. Dunne, El-Gamati, S. Fox, A. Rous and L. P. Cuffe,
J. Mol. Struct., 1999, 477, 221.
21 J. B. Foresman and A. E. Frisch, in Exploring Chemistry with
Electronic Structure Methods, Gaussian Inc., Pittsburgh, PA, 2nd
edn., 1996.
27 K.-H. Thiele, E. Langguth and G. E. Müller, Z. Anorg. Allg. Chem.,
1980, 462, 152.
28 GAUSSIAN 94, 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.
Petersson, 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 and J. A.
Pople, Gaussian, Inc., Pittsburgh, PA, 1995.
22 L. Onsager, J. Am. Chem. Soc., 1936, 58, 1486.
23 Handbook of Chemistry and Physics, ed. R. C. Weast, CRC Press,
Cleveland, OH, 52 edn., 1971.
Paper 9/05183J
J. Chem. Soc., Dalton Trans., 1999, 4149–4153
4153