Laser Irradiation of Monomeric Acetylene and the T-Shaped Acetylene Dimer in Xenon and Argon Matrices

Article abstract of DOI:10.1002/(sici)1099-0690(199805)1998:5<769::aid-ejoc769>3.0.co;2-s

Irradiation of acetylene (4) in an argon matrix with an ArF laser (λ = 193 nm) yields the ethynyl radical C₂H (5) and C₂ (6). If the same wavelength is used to irradiate 4 in a xenon matrix, only the infrared signal of the compound Xe-C₂ (7) can be detected. The same band is found on irradiation at λ = 248 nm (KrF laser) in a xenon matrix, despite the fact that acetylene (4) does not absorb light of this wavelength. Irradiation of the T-shaped acetylene dimer 9 in an argon or xenon matrix at λ = 193 nm yields butadiyne (11) and vinylacetylene (12). However, irradiation with a KrF laser (λ = 248 nm) in a xenon matrix additionally yields cyclobutadiene (13). The dependence of the mechanisms of the fragmentation and dimerization of acetylene (4) on the matrix material is discussed.

Full text of DOI:10.1002/(sici)1099-0690(199805)1998:5<769::aid-ejoc769>3.0.co;2-s

FULL PAPER
Laser Irradiation of Monomeric Acetylene and the T-Shaped Acetylene Dimer
in Xenon and Argon Matrices
Günther Maier* and Christian Lautz
Institut für Organische Chemie der Justus-Liebig-Universität,
Heinrich-Buff-Ring 58, D-35392 Gießen, Germany
Fax: (internat.) ϩ 49(0)641/99-34309
Received December 29, 1997
Keywords: Matrix isolation / IR spectroscopy / Photochemistry / Two-photon process / Excitons
Irradiation of acetylene (4) in an argon matrix with an ArF
Irradiation of the T-shaped acetylene dimer 9 in an argon or
xenon matrix at λ = 193 nm yields butadiyne (11) and vinyla-
cetylene (12). However, irradiation with a KrF laser (λ = 248
•
laser (λ = 193 nm) yields the ethynyl radical C
6). If the same wavelength is used to irradiate 4 in a xenon
matrix, only the infrared signal of the compound Xe–C
2 2
H (5) and C
(
(7) nm) in a xenon matrix additionally yields cyclobutadiene
(13). The dependence of the mechanisms of the fragmenta-
tion and dimerization of acetylene (4) on the matrix material
is discussed.
2
can be detected. The same band is found on irradiation at
λ = 248 nm (KrF laser) in a xenon matrix, despite the fact
that acetylene (4) does not absorb light of this wavelength.
The fundamental requirement for the use of photoexci- yields ethene (2) and 1-butene (3), although the molecule
tation in generating reactive species in inert rare-gas ma- does not absorb light of the wavelength λ ϭ 248 nm. If
trices is the absorption of light by the organic precursor. An argon is used as the matrix material, no fragmentation is
alternative to this direct photolysis is the use of an indirect detected because exciton formation with light of this wave-
[1]
method, in which the energy is absorbed by the matrix ma- length is impossible in an argon matrix .
terial. In this case, the stored energy has to be transferred
to the precursor in a second step in order to initiate a
chemical reaction.
When pure rare gases are used, very short wavelength
radiation must be employed to excite the matrix. With ar-
gon as the matrix material, light of wavelength λ ϭ 103 nm
(
1
12.1 eV) is necessary; a xenon matrix requires at least λ ϭ
48 nm (8.4 eV) to reach the exciton state .
The application of light sources producing these short
The goal of this investigation was to find out whether
indirect excitation using the absorption of the matrix mate-
rial rather than the absorption of the precursor itself can
also be used to induce dimerization instead of fragmen-
tation reactions. We chose acetylene (4) as the “test com-
pound”, because it shows no absorption above λ ϭ 237
[1]
wavelengths is difficult. Alternatively, the energy necessary
for exciton formation can be attained using a UV laser
through two-photon absorption. Böttcher and Schmidt[
detected the generation of charge carriers during the ir-
radiation of solid xenon with an excimer laser (KrF: λ ϭ
2]
[6]
nm in the gas phase and may perhaps lead to candidates
of the C H hypersurface. For comparison purposes, we
4
4
2
48 nm; ArF: λ ϭ 193 nm), whereas Schwentner et al.[3]
detected the solid-phase emission from Xe * excimers (λ ϭ
also studied the products arising from the direct excitation
at different wavelengths.
In addition, we studied the influence of the matrix mate-
rial on the fragmentation products of monomeric acetylene
2
1
72 nm) formed upon laser irradiation. This emission is
identical to that observed in a one-photon process when
solid xenon is irradiated at λ ϭ 148 nm .
[1]
(
4) and found astonishing differences in product formation
Lawrence and Apkarian[4] used this indirect method of upon laser irradiations of 4 in xenon and argon matrices.
excitation for the photochemically-induced fragmentation
of N O in a xenon matrix with the light of a KrF excimer
2
Irradiations of Monomeric Acetylene
laser (λ ϭ 248 nm; 5.0 eV). The two-photon-induced gener-
ation of excitonic states in the solid xenon and the sub-
sequent exciton-molecule encounters led to an efficient dis-
Irradiation at λ ؍ 193 nm (ArF Laser)
It is reported[7][8] that the ethynyl radical C H (5) is the
primary product of gas-phase photolysis of diluted acety-
•
2
sociation of N O.
2
[5]
As we have recently shown , the same kind of fragmen- lene (4)/rare gas mixtures at λ ϭ 193 nm. In a two-photon
tation can also be achieved in organic molecules. Irradiation sequential dissociation, the remaining hydrogen atom can
[
5]
of cyclobutane (1) in a xenon matrix with a KrF laser be split off and C (6) is also formed.
2
Eur. J. Org. Chem. 1998, 769Ϫ776

WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
1434Ϫ193X/98/0505Ϫ0769 $ 17.50ϩ.50/0
769

G. Maier, C. Lautz
FULL PAPER
Figure 1. IR difference spectrum after the irradiation (ArF laser,
λ ϭ 193 nm, or KrF laser, λ ϭ 248 nm) of acetylene (4) in a xenon
matrix; the bands with negative values diminish, while those with
positive values are enhanced upon irradiation
The irradiation of acetylene (4) in an argon matrix [ratio
acetylene (4)/Ar ϭ 1:1000] with an ArF laser (λ ϭ 193 nm)
•
yields small amounts of C H (5) after several hours, in ac-
2
cordance with the results obtained in the gas phase. The
bands of 5 are found at ν˜ ϭ 1846.6, 2104.5, 4022.4, and
Ϫ1
4
132.8 cm , in agreement with the data published by Jacox
[
9]
et al. . Moreover, the UV/Vis spectrum exhibits an absorp-
tion band at λ ϭ 237 nm, which can be assigned to C
2
[
10]
(6) . Thus, the results of the experiment in an argon ma-
trix are in accordance with those of the gas-phase study.
More surprising is the fact that the irradiation of acety-
lene (4) in a xenon matrix with light from the same radi-
ation source leads to completely different spectra. The in-
Figure 2. UV/Vis spectrum after the irradiation (ArF laser, λ ϭ
frared spectrum shows one single band at ν˜ ϭ 1767.0 cmϪ1 193 nm, or KrF laser, λ ϭ 248 nm) of acetylene (4) in a xenon
matrix; the broken line indicates the base line before irradiation
and none of the aforementioned signals assigned to the rad-
•
ical C H (5) (Figure 1). The UV/Vis spectrum does not
2
show a band at λ ϭ 237 nm, but several new bands centered
around λ ϭ 198 nm and λ ϭ 423 nm instead (Figure 2).
With light of wavelength λ ϭ 193 nm, there are two pos-
sibilities for the excitation of acetylene (4) in a xenon ma-
trix: The direct absorption of the acetylene (4) molecule,
[6]
which absorbs strongly below λ ϭ 200 nm , or the absorp-
tion of the matrix material as described above (the mecha-
nism of energy transfer including exciton states will be dis-
cussed later).
Irradiation at λ ؍ 248 nm (KrF Laser)
If the emission from a KrF laser (λ ϭ 248 nm) is used,
only the absorption of the matrix material is possible, be-
cause acetylene (4) shows no absorption above λ ϭ 237 nm
seems to be too high merely for a change of the matrix
material. In order to unequivocally rule out that the signal
at ν˜ ϭ 1767.0 cm belongs to a molecule with at least one
Ϫ1
[
6]
(gas phase ). The transparency of acetylene (4) at λ ϭ 248
hydrogen atom, the deuterated compounds [D ]acetylene
nm could be verified. In an argon matrix, no conversion
could be detected even after several hours of irradiation.
1
(
[D ]-4) and [D ]acetylene ([D ]-4) were also irradiated. The
1 2 2
synthesis of the deuterated precursor molecules (see Exper-
To prove the hypothesis that the species formed upon ir-
radiation of 4 at λ ϭ 193 nm in xenon may also be obtained
by an indirect excitation of the matrix, the experiment was
repeated using a KrF laser (λ ϭ 248 nm). Indeed, the infra-
red (Figure 1) and the UV/Vis spectra (Figure 2) corre-
sponded to those obtained following irradiation at λ ϭ 193
imental Section) yielded a mixture of 4, [D ]-4 and [D ]-4,
1
2
and this mixture was used with xenon in a ratio of 1:1000
for the matrix experiments.
Figure 3. IR difference spectrum after the irradiation (KrF laser,
1 1
λ ϭ 248 nm) of a mixture of acetylene (4), [D ]acetylene ([D ]-4)
nm. Again, the infrared spectrum showed only one signal
2 2
and [D ]acetylene ([D ]-4) in a xenon matrix; the bands with nega-
Ϫ1
at ν˜ ϭ 1767.0 cm
.
tive values diminish, while those with positive values are enhanced
upon irradiation
The intensity of the signal at ν˜ ϭ 1767.0 cmϪ1 remains
unchanged on annealing of the matrix (stepwise annealing
at 38 K, 46 K, and 50 K; 30 minutes at each temperature),
but decreases on irradiation with light of λ Ն 254 nm (the
wavelengths λ ϭ 436 nm, λ Ն 385 nm, λ Ն 295 nm, and
λ ϭ 254 nm all lead to a decrease of the signal; λ ϭ 254
nm has the largest effect). No new signal appears in the
infrared spectrum during these secondary irradiations. The
groups of UV/Vis bands centered at λ ϭ 198 nm and λ ϭ
4
23 nm decrease accordingly as the IR band diminishes.
Ϫ1
A shift of almost ∆ ν˜ ϭ Ϫ80 cm (1767.0 vs. 1846.6
Ϫ1
•
cm ) for the proposed ethynyl radical C H (5) in xenon
2
770
Eur. J. Org. Chem. 1998, 769Ϫ776

Laser Irradiation of Monomeric Acetylene and the T-Shaped Acetylene Dimer
FULL PAPER
The result of the irradiation at λ ϭ 248 nm was unam- tween 300 and 750 cmϪ1 and cannot be assigned to any
Ϫ1
biguous: Only the signal at ν˜ ϭ 1767.0 cm appeared in vibration of a compound containing a C molecule (6), be-
2
the infrared spectrum, while all the signals for the acety- cause the progression for the CϭC stretching fundamental
lenes 4, [D ]-4, and [D ]-4 decreased upon irradiation (Fig- is completely absent. Therefore, the bands in the UV/Vis
1
2
ure 3). The UV/Vis spectrum showed no difference to that spectra cannot be assigned unambiguously to the com-
obtained from the photolysis of 4 alone (Figure 2). Thus, pound XeϪC (7).
2
the signal in the IR spectrum, and probably the UV/Vis
absorptions as well, can be attributed to a species that con- Trapping Experiments with Xe؊C (7)
2
tains no hydrogen atoms. Secondary irradiation of the mix-
ture of isotopomers at λ ϭ 254 nm yielded the familiar
result, i.e. decrease of the signals in both the IR and UV/
Vis spectra.
In order to obtain a final proof for the structure of the
adduct 7 between Xe and C (6), we irradiated matrices
2
13
containing CO together with acetylene (4) or [1,2- C]ace-
tylene ([1,2- C]-4), in the hope of producing the cumulenes
C O (8) or [2,3- C]C O ([2,3- C]-8) as reaction products
13
13
13
3
3
from the photolytically generated XeϪC (7). The matrices
2
for these experiments were made by co-condensation of a
CO/xenon gas mixture (ratio 100:1000) with a xenon gas
13
mixture containing 4 or [1,2- C]-4 in a ratio of 3:1000.
Which molecule is responsible for the signal at ν˜ ϭ
Ϫ1
1
767.0 cm ? Because of the simplicity of the precursor
acetylene (4) and the fact that the produced molecule con-
tains no hydrogen atom, the photoinduced generation of C2
(
6) is the only plausible explanation. Since C (6) should
2
show no infrared band, we postulate that a new compound
Figure 5. IR difference spectrum after the irradiation (KrF laser,
XeϪC (7), formed by reaction of 6 with the matrix mate- λ ϭ 248 nm) of a matrix produced by co-condensation of acetylene
2
(
4)/xenon (3:1000) and CO/xenon (100:1000) gas mixtures; the
rial xenon, gives rise to the observed signal in the IR spec-
bands with negative values diminish, while those with positive
values are enhanced upon irradiation; the assignment of the pro-
[12]
1
3
trum. To verify this hypothesis, [1,2- C]acetylene ([1,2-
13
[11]
3
C]-4) was irradiated in a xenon matrix (ratio 1:1000) with duct bands was made with the aid of ref. for C O (8) and ref.
for the radical HCO•
light from a KrF laser.
Figure 4. IR difference spectrum after the irradiation (KrF laser,
13
13
λ ϭ 248 nm) of [1,2- C]acetylene ([1,2- C]-4) in a xenon matrix;
the bands with negative values diminish, while those with positive
values are enhanced upon irradiation
Ϫ1
Besides the signals at ν˜ ϭ 1767.0 cm in the case of
Ϫ1
13
acetylene (4) and ν˜ ϭ 1698.6 cm in the case of [1,2- C]-
13
acetylene ([1,2- C]-4), the two spectra clearly exhibit the
The irradiation again produced only one signal in the in- bands for the three most intense infrared transitions of C O
3
Ϫ1
13
13
[11]
frared spectrum, but at ν˜ ϭ 1698.6 cm rather than at (8) or [2,3- C]C O ([2,3- C]-8) , respectively (Figure 5).
3
Ϫ1
Ϫ1
ν˜ ϭ 1767.0 cm (Figure 4). The shift of ∆ ν˜ ϭ Ϫ68.4 cm
Clearly, the adduct 7 of C (6) and Xe reacts with carbon
2
corresponds very well with the theoretical value of ∆ ν˜ ϭ monoxide to give the more stable cumulene 8.
Ϫ1
12
13 1/2
Ϫ69.3 cm calculated from (m C/m C) ϭ 0.9608 for a
Moreover, the two most intense bands of the radical
1
3
•
Ϫ1
diatomic molecule containing two C atoms. Once again, HCO can be detected in both spectra at ν˜ ϭ 1856.3 cm
subsequent irradiation at λ ϭ 254 nm led to a decrease of and ν˜ ϭ 1076.6 cm
the signals in the IR and UV/Vis spectra. tion of CO with hydrogen atoms originating from the frag-
The UV/Vis spectrum exhibits the same features as in the mentation of 4 or [1,2- C]-4. The spectra for the two cu-
case of the 12C isotopomer of 4, with the absorptions mulenes were calculated at the BLYP/6-311ϩG* level of
shifted to longer wavelengths by only 1 nm. The pro- theory in order to allow a comparison of the experimentally
gressions of the UV/Vis bands formed upon irradiation of derived isotopic shifts with the theoretical values. The
Ϫ1[12]
. This species is formed by the reac-
13
1
3
13
acetylene (4) and [1,2- C]acetylene ([1,2- C]-4) are be- agreement is satisfactory (Table 1).
Eur. J. Org. Chem. 1998, 769Ϫ776
771

G. Maier, C. Lautz
FULL PAPER
Table 1. Experimental (xenon, 10 K) and calculated (BLYP/ the CϭC stretching fundamental, the stability of the com-
1
3
13
6
-311ϩG*) IR spectra of C
3 3
O (8) and [2,3- C]C O ([2,3- C]-8)
plex, and the charge transfer between the rare gas atom and
(
observed shifts on isotopic substitution are given in parentheses)
the C molecule (6). In order to permit description of such
2
a complex even with a xenon atom, the MP2 method was
used together with the LANL2DZ basis set (Table 2).
These calculations confirm the hypothesis of a linear
Experiment
ν˜ /cm
BLYP/6-311ϩG*
Ϫ1
Ϫ1
Int.
ν˜ /cm
Int.
ν
ν
ν
1
2
3
[C
[C
[C
3
3
3
O (8)]
O (8)]
O (8)]
2237.2
1904.4
571.9
100.0
1.3
2259.9
1888.5
590.7
100.0
1.6
compound XeϪC (7) with only one xenon atom. The in-
2
crease in charge transfer, complex stabilization, and es-
pecially the increase of the infrared intensity of the CϭC
stretching fundamental on going from neon to xenon is
convincing. However, the calculated vibrational frequency
1.7
3.1
13
ν
ν
ν
1
([2,3- C]-8) 2219.0 (Ϫ18.2) 100.0 2236.1 (Ϫ23.8) 100.0
13
2
([2,3- C]-8) 1847.4 (Ϫ57.0)
([2,3- C]-8) 568.4 (Ϫ3.5)
1.3 1834.8 (Ϫ53.7)
2.8 586.1 (Ϫ4.6)
4.0
3.3
13
3
Ϫ1
Ϫ1
shift of ∆ ν˜ ϭ Ϫ16.8 cm (1798.7 vs. 1815.5 cm ) is still
Ϫ1
far from the observed shift of ∆ ν˜ ϭ Ϫ87.6 cm (1767.0 vs.
854.6 cm ) between the gas-phase and the matrix value.
The Infrared Spectrum
Ϫ1
1
1
3
13
The experiment with [1,2- C]acetylene ([1,2- C]-4) This indicates that XeϪC (7) cannot simply be described
2
clearly proves that the signal in the infrared spectrum be- as a complex of C (6) with Xe, but as a compound with a
2
longs to a compound containing a C (6) molecule. The relatively strong bond between Xe and C (6), which is not
2
2
presence of only one signal in the infrared spectrum reflects predicted by the MP2 (fc)/LANL2DZ calculations. If the
a preference for a compound with a fixed number of xenon polarizability of Xe, the electron affinity of C (6), and the
2
atoms directly adjacent to 6.
results of the calculations are taken into account, the com-
The CϭC stretching fundamental of the molecule C (6) pound XeϪC (7) is best described by two resonance struc-
2
2
Ϫ1
[13]
is found at ν˜ ϭ 1854.6 cm in the gas phase , whereas tures.
the experimental value for the infrared signal of the new
Ϫ1
compound 7 is ν˜ ϭ 1767.0 cm in a xenon matrix. This
Ϫ1
shift of ∆ ν˜ ϭ Ϫ87.6 cm is too high to be caused merely
by complexation of C (6) with the matrix material xenon.
2
[
14]
Jacox
studied the shifts in the ground-state vibrational
This assumption is supported by comparison with the
fundamental frequencies of diatomic molecules upon trap-
ping in argon matrices and found that the matrix shift rela-
Ϫ
CϭC stretching fundamental of C2 in the gas phase at
Ϫ1[19]
ν˜ ϭ 1781.2 cm
, which is close to the experimental
tive to the gas-phase position is less than 2%.
Ϫ1
value of ν˜ ϭ 1767.0 cm for XeϪC (7) in the xenon ma-
2
Furthermore, it is known[
15]
from the comparison of
trix.
measured vibrational frequencies in rare-gas matrices with
the corresponding gas-phase values that in rare gas/hydro-
gen halide complexes (RgϪHX) the correlation between the
vibrational shifts of the rare gas/hydrogen halide complex
and the polarizability of the rare-gas atoms is fairly linear.
In all cases, the stretching vibrations are red-shifted com-
pared to the free, gas-phase hydrogen halide fundamental.
The red-shift increases on going from neon to xenon as the
polarizability of the matrix material increases, but the shifts
are smaller than ∆ ν˜ ϭ Ϫ87.6 cmϪ1, even in the case of a
xenon matrix.
Despite the use of the basis set LANL2DZ, our postulate
of a linear compound 7 is in good agreement with a calcu-
lation at the high ab initio level CCSD-T/aug-cc-pVTZ for
[20]
the van der Waals complex ArϪC2 . The global minimum
of energy is predicted to occur for a linear ArϪC2 ge-
ometry, while a saddle point is predicted for a T-shaped
geometry. This T-shaped geometry represents the global
minimum for all other rare-gas complexes with heavier sym-
metric diatomic molecules of second-row atoms, e.g.
ArϪN .
2
Naumkin and McCourt[20] also performed calculations
The electron affinity of C (6) (EA ϭ 3.27 eV)[ is al-
16]
2
on complexes with more than one Ar atom and found min-
most as high as the electron affinity of the bromine atom
ima for the complexes Ar ϪC with n ϭ 2Ϫ4. The exper-
n
2
[
17]
(EA ϭ 3.36 eV)
and enables the molecule 6 to accept
imental spectra in our case, however, indicate that there is
a preference for one single compound, since only one signal
is observed in the infrared spectrum. On the basis of the
electron density. The LUMO, which is the acceptor orbital
for negative charge, is placed at Ϫ3.19 eV in the molecular
orbital scheme [calculated at the MP2 (fc)/6-311ϩG* level
of theory] and possesses the largest orbital coefficients out-
side the carbon atoms on the axis connecting the nuclei [see
electronic structure of the molecule C (6), a linear com-
2
pound XeϪC (7) seems to be preferred with respect to all
2
other species.
[
18]
ref.
regarding the molecular orbital scheme of C (6)].
2
This indicates a preference for a linear compound with one
xenon atom adjacent to each of the two carbon atoms of
C (6).
2
Irradiations of Dimeric Acetylene
IR Spectrum of the T-Shaped Acetylene Dimer 9
To examine the influence of various rare-gas atoms on
To answer the question whether indirect photolysis can
linear compounds with 6, we calculated the properties of also be used to induce dimerization of acetylene (4), rare-
the complexes RgϪC (Rg ϭ Ne, Ar, Kr, Xe) with respect gas matrices containing a higher concentration of 4 were
2
to the vibrational frequency shift, the infrared intensity of produced. The ratio acetylene (4)/rare gas was 30:1000.
772
Eur. J. Org. Chem. 1998, 769Ϫ776

Laser Irradiation of Monomeric Acetylene and the T-Shaped Acetylene Dimer
FULL PAPER
˚
2
Table 2. Comparison of the calculated (MP2 (fc)/LANL2DZ) properties of complexes RgϪC (distances d in A; q denotes the charge at
Ϫ1
Ϫ1
the atoms, calculated from an NPA; CϭC stretching fundamental is given in cm , the intensity of this vibration in km·mol ; Estab is
Ϫ1
the stabilization of the complex in kcal·mol compared to its fragments, which were calculated at the same level of theory)
Complex
d(RgϪC
1
)
d(C
1 2
ϭC )
q(Rg)
1
q(C )
q(C
2
)
ν˜ CϭC
Int.
E
stab
C
1
ϭC
2
Ϫ
1.295
1.295
1.295
1.296
1.297
Ϫ
0.0
0.0
1815.5
1815.4
1814.6
1811.5
1798.7
0.0
0.5
Ϫ
NeϪC
1
ϭC
2
3.157
3.505
3.378
3.317
ϩ0.004
ϩ0.007
ϩ0.018
ϩ0.047
Ϫ0.005
Ϫ0.011
Ϫ0.029
Ϫ0.069
ϩ0.001
ϩ0.004
ϩ0.011
ϩ0.023
0.26
0.21
0.31
0.46
ArϪC
KrϪC
XeϪC
1
ϭC
2
1.4
1
ϭC
ϭC
2
2
10.9
91.8
1
The infrared spectra of these matrices show the intense Figure 7. Calculated (B3LYP/6-311G**) structures for the acety-
lene dimer 9 (C ) and the acetylene trimer 10 (C ); distances are
2
v
3h
signals of monomeric acetylene (4) along with new signals
that can only be attributed to acetylene oligomers (Figure
˚
given in A
6
).
Figure 6. IR spectrum before irradiation of a matrix produced from
an acetylene (4)/xenon gas mixture with a ratio of 30:1000
Table 3. Comparison of the calculated (B3LYP/6-311G**) IR spec-
The acetylene dimer 9 has a T-shaped geometry with C2v
tra for the acetylene dimer 9 and trimer 10 with the experimental
symmetry. According to ab initio calculations, this weakly (xenon, 10 K) IR spectrum from an acetylene (4)/xenon gas mixture
with a ratio of 30:1000
bound van der Waals complex 9 is the global minimum on
the (C H ) potential energy surface[ . The T-shaped
21]
2
2 2
Acetylene dimer 9
Acetylene trimer 10
Experiment
structure could be confirmed through high-resolution vi-
bration-rotation spectra in a pulsed supersonic free jet[
and through microwave spectra obtained for the ground vi-
brational state using a pulsed-nozzle Fourier transform mi-
(
C
2v
)
(C3h)
22]
ν˜ /cmϪ1
ν˜ /cmϪ1
ν˜ /cmϪ1
Int.
Int.
Int.
6
6
6
46.2
54.2
64.2
1.9
606.5
616.2
<0.1
1.6
2.2
1.2
654.1
673.3
2.2
[23]
crowave spectrometer . The most intense infrared signals
2.6
622.1
1.2
of the dimer 9 were obtained by Collins[24] upon irradiation
[a]
772.3
781.7
19.0
68.2
87.9
43.1
2.1
729.0
Ϫ
99.3
100.0
86.3
0.4
of 2-butyne in a xenon matrix, and by Bothur[25] upon pho-
736.2
7
88.1
799.7
801.9
2062.0
50.4
62.1
0.9
744.0
tolysis of cyclobutadiene (13).
797.6
063.7
066.8
401.5
751.3
2
2
3
1963.4
1967.8
3248.6
3251.5
3345.5
3358.1
The acetylene trimer 10 was calculated to have C sym-
3
h
<0.1
100.0
57.8
0.7
0.4
[
21]
metry . From the experimentally derived rotation-vi-
3397.3
3508.0
100.0
0.3
24.2
90.2
<0.1
<0.1
bration spectrum one could not discriminate between C - 3418.0
3h
[26]
3511.5
521.0
and D -symmetric structures
.
3h
3
<0.1
For a comparison of the experimental spectrum with the
theoretical ones, we recalculated acetylene (4), the acetylene
dimer 9, and the trimer 10 at the B3LYP/6-311G** level signal of the monomeric acetylene (4) and thus cannot be inte-
of theory. Figure 7 shows the optimized structures for the
[a]
The most intense signal strongly overlaps with the most intense
grated.
acetylene dimer 9 and trimer 10 with C2v and C3h sym-
metry, respectively. The calculated binding energies of these of the trimer 10 in the experimental spectrum (Figure 6).
Ϫ1
van der Waals complexes are E ϭ 0.69 kcal·mol and E ϭ This assumption can be confirmed by comparing the calcu-
Ϫ1
2
.16 kcal·mol (zero-point energy corrected values).
lated infrared spectra for 9 and 10 with the experimental
The calculated infrared frequencies of both complexes one. A preference for the acetylene dimer 9 in comparison
are listed in Table 3, together with the band positions from with the acetylene trimer 10 is indicated simply by the num-
the matrix experiment. ber of signals in the experimental infrared spectrum. The
On the basis of the ratio acetylene (4)/xenon available, intensities of the signals, however, show no good agreement
the concentration of the dimer 9 should be higher than that between experiment and theory. The reason for this weak-
Eur. J. Org. Chem. 1998, 769Ϫ776
773

G. Maier, C. Lautz
FULL PAPER
ness is the problem of integration in the experimental spec- with a low-pressure mercury lamp at a wavelength of λ ϭ
trum, because most of the bands strongly overlap, especially 185 nm.
the intense ones.
Irradiation at λ ؍ 248 nm (KrF Laser)
The most obvious reason for the preference of the dimer
9
is the observed photochemistry, which will be described
The transparency of acetylene (4) at λ ϭ 248 nm was
again confirmed upon irradiation of the acetylene dimer 9;
no conversion was detected in an argon matrix. However, a
change of the matrix material from argon to xenon made
the indirect photolysis possible, and reactions of the dimer
below. Nevertheless, traces of benzene can also be detected
among the products of the irradiations, and thus it is clear
that small amounts of the acetylene trimer 10 are also pre-
sent, despite the still low concentration of acetylene (4)/
xenon ϭ 30:1000. Hence, it cannot be excluded that signals
of the dimer 9 are overlapped by the intense signals of the
trimer 10.
9
could be initiated.
Besides traces of benzene, vinylacetylene (12), cyclobuta-
diene (13), and butadiyne (11) were detected as the main
products of the indirect photolysis (Figure 8, upper spec-
trum).
Irradiation at λ ؍ 193 nm (ArF Laser)
Butadiyne (11) is the main product of photolysis of gase-
Since the overall conversion of the acetylene dimer 9 is
ous acetylene (4) at λ ϭ 193 nm; ethene (2), 1,3-butadiene, low even after irradiation for 10 h, only the two most in-
and vinylacetylene (12) are formed as by-products[ . The tense signals of the seven expected fundamentals of cyclo-
27]
mechanisms of product formation in this gas-phase reaction butadiene (13) could be detected. As an additional proof
have been studied in detail by Seki and Okabe[
28]
.
for the formation of 13, the matrix was irradiated with light
For a comparison of the products from the direct pho- of the wavelengths λ ϭ 250Ϫ420 nm. This light is most
tolysis with those from the indirect photolysis, irradiations effective in inducing the fragmentation of 13 into two acety-
of the matrices containing the acetylene dimer 9 were first lene (4) molecules[ , and, in fact, only the intensities of
30]
Ϫ1
Ϫ1
carried out with light from an ArF laser (λ ϭ 193 nm), the two signals at ν˜ ϭ 573.3 cm and ν˜ ϭ 1237.7 cm
[6]
since acetylene (4) absorbs strongly below λ ϭ 200 nm . were seen to decrease.
In this case, the products of the photolyses were found to
All products appear simultaneously in the IR spectrum,
be independent of the matrix material, i.e. the results were so each molecule has to be considered as a product of the
the same for an argon and a xenon matrix. The main prod- primary photochemically induced chemistry, despite the
ucts were butadiyne (11) and vinylacetylene (12). Figure 8 fact that vinylacetylene (12) is a known product of the ir-
lower spectrum) shows the difference spectrum from the radiation of cyclobutadiene (13) at λ ϭ 248 nm[ . Buta-
31]
(
irradiation in a xenon matrix. Two additional signals can diyne (11), however, is only produced photolytically from
be detected in this infrared spectrum: The signal at ν˜ ϭ vinylacetylene (12) at wavelengths below λ ϭ 200 nm[
32]
.
Ϫ1
1
767.0 cm , assigned to XeϪC (7) (this band is missing
Comparison with the irradiation at λ ϭ 193 nm shows a
new product formed only at λ ϭ 248 nm, namely cyclobuta-
diene (13) (Figure 8), and a lower concentration of buta-
2
Ϫ1
if an argon matrix is used), and a signal at ν˜ ϭ 890.9 cm
The latter could be assigned to the vinyl radical C H3
.
,
•
[29]
2
but no further investigations were made to prove this as- diyne (11), whereas the concentration of vinylacetylene (12)
signment.
compared to that of 11 increases. Hence, the products must
originate from different mechanisms.
Figure 8. IR difference spectra obtained after laser irradiation of
acetylene dimer 9 in xenon matrices; lower spectrum: photolysis of
9
9
with an ArF laser (λ ϭ 193 nm); upper spectrum: photolysis of
with a KrF laser (λ ϭ 248 nm); all bands shown are enhanced
upon irradiation
Discussion
The irradiation of monomeric acetylene (4) with light of
the wavelength λ ϭ 193 nm proceeds differently in xenon
•
and argon matrices. In argon, C H (5) and C (6) are the
2
2
•
products, whereas in the xenon matrix, no C H (5) can be
2
detected and a new compound, XeϪC (7), is formed as the
2
The products from the irradiation of the acetylene dimer sole product. This different behaviour can clearly only be
9
at λ ϭ 193 nm in a rare-gas matrix clearly correspond to attributed to the matrix material. The formation of the
those formed in the photochemical excitation of gaseous stable compound 7 is only possible in xenon, and for this
acetylene (4). The photochemistry in the two “media” is reason the elimination of the second hydrogen atom from
•
therefore very similar. The same results can be obtained C H (5) is favoured and no signal for 5 can be detected in
2
774
Eur. J. Org. Chem. 1998, 769Ϫ776

Laser Irradiation of Monomeric Acetylene and the T-Shaped Acetylene Dimer
FULL PAPER
this matrix material. The exceptional feature of this ir- self-trapping, forming an excimer center Xe *, which decays
2
[1]
radiation is that the polarizability of the matrix material with emission of radiation (λ ϭ 172 nm) .
xenon allows the formation and infrared detection of the
species XeϪC (7).
2
The preferred elimination of both hydrogen atoms from
acetylene (4) in xenon could be confirmed by irradiation of
4
in the presence of CO. In this case, C O (8) was formed
3
•
as a trapping product and again no signal for C H (5)
2
could be detected in the infrared spectrum.
If the matrix is further doped with molecules such as the
Compound XeϪC (7) is not only formed by irradiation acetylene dimer 9, exciton-molecule encounters are highly
2
at λ ϭ 193 nm in a xenon matrix, but also upon irradiation probable because of the large mobility of the free excitons
˚
[1]
with the light of a KrF laser (λ ϭ 248 nm), even though (25Ϫ260 A) . The energy stored in the exciton can be re-
acetylene (4) shows no absorption at this wavelength. This leased through this process and transferred to the embed-
can clearly be explained in terms of the different mecha- ded molecule. The possible mechanisms for this energy
nisms of excitation, namely by the absorption of the matrix transfer are discussed in a review article by Schwentner et
material in the case of xenon with light of the wavelength al.[
33]
.
λ ϭ 248 nm, and the direct absorption of the organic pre-
We assume that the transfer of energy through exciton-
cursor in an argon or xenon matrix at λ ϭ 193 nm. Two- molecule encounters in the host lattice is sufficient to bring
photon-induced generation of excitonic states in solid the dimer 9 to a highly excited state (C H ) * (electronically
2
2 2
xenon followed by exciton-molecule encounters can there- or vibrationally).
fore also be used to generate the compound XeϪC (7).
2
As far as the photochemistry of the acetylene dimer 9
upon irradiation at λ ϭ 193 nm is concerned, a comparison
between the gas-phase chemistry of 4[
27][28]
and the matrix
experiments can be drawn. The formation of butadiyne (11)
upon irradiation of gaseous acetylene (4) at λ ϭ 193 nm is
explained by a two step process.
Our proposed mechanism for this photochemically in-
duced reaction involves a 1,4-diradical 14 as an intermedi-
ate. Cyclobutadiene (13) can be formed by ring closure,
while vinylacetylene (12), which is the source of the third
product butadiyne (11), can be formed by a 1,3-hydrogen
shift. The addition of a third acetylene (4) molecule to the
diradical 14 in the same matrix cage readily explains the
formation of benzene from the acetylene trimer 10.
The same mechanism can also be applied to the matrix
experiments. However, since the fragments are not mobile
in the matrix as they are in the gas phase, the formation
of the products can alternatively be explained in terms of
disproportionation of the acetylene dimer 9 to form an
•
•
ethynyl radical C H (5) and a vinyl radical C H , which
2
2
3
can subsequently form vinylacetylene (12) by recombina-
tion.
Bally et al.[34] explain their ab initio results on the reac-
tion of acetylene (4) with its radical cation using a very
Vinylacetylene (12) is not stable under the experimental similar reaction scheme.
[32]
photolytic conditions and forms butadiyne (11) . This ex-
plains the low concentration of 12 obtained in the matrix ton formation, light of the wavelength λ ϭ 193 nm (6.4 eV)
experiments (Figure 8, lower spectrum). should be able to induce the exciton state of solid xenon
None of these intermediates can, however, explain the (8.4 eV) in a two-photon process even more readily[1][2][3]
Considering the energy necessary for the process of exci-
.
product formation and distribution observed in the experi- At this wavelength, however, the two processes of direct and
ments at λ ϭ 248 nm. Acetylene (4) does not absorb light indirect absorption compete with each other. An esti-
of this wavelength and thus the energy has to be transferred mation, using a formula given by Lawrence and Ap-
[4]
to the organic precursor through exciton-molecule encoun- karian , together with the absorption cross-section of
[
4]
[28]
ters .
acetylene (4) and the two-photon excitation cross-section
The irradiation of a xenon matrix with the light of a KrF of solid xenon[ at λ ϭ 193 nm, clearly shows that the di-
2]
laser (λ ϭ 248 nm) generates delocalized excitons Xe * by rect absorption of the precursor acetylene (4) predominates
n
two-photon absorption. These excitons can be localized by at this wavelength. Consequently, the experimental results
Eur. J. Org. Chem. 1998, 769Ϫ776
775

G. Maier, C. Lautz
FULL PAPER
[
8]
B. A. Balko, J. Zhang, Y. T. Lee, J. Chem. Phys. 1991, 94,
of irradiations of the acetylene dimer 9 at λ ϭ 193 nm in
xenon and argon matrices are the same.
7
958Ϫ7966.
[9]
D. Forney, M. E. Jacox, W. E. Thompson, J. Mol. Spectrosc.
1
995, 170, 178Ϫ214.
Part of this work was presented as a poster at the Conference
[10]
D. E. Milligan, M. E. Jacox, L. Abouaf-Marguin, J. Chem.
Phys. 1967, 46, 4562Ϫ4570.
11]
“Structural and Mechanistic Organic Chemistry: An International
[
P. Botschwina, H. P. Reisenauer, Chem. Phys. Lett. 1991, 183,
Conference in Honor of Norman L. Allinger”, Athens, Georgia,
USA, in June 1997. C. L. wishes to thank the Gesellschaft Deut-
scher Chemiker for financial support to attend this conference. We
also wish to thank the Volkswagenstiftung and the Deutsche For-
schungsgemeinschaft for their support and Dr. H. P. Reisenauer for
most helpful discussions.
2
17Ϫ222.
[12]
[13]
D. E. Milligan, M. E. Jacox, J. Chem. Phys. 1969, 51, 277Ϫ288.
S. P. Davis, M. C. Abrams, J. G. Phillips, M. L. P. Rao, J. Opt.
Soc. Am. B 1988, 5, 2280Ϫ2285.
[
[
14]
15]
M. E. Jacox, Chem. Phys. 1994, 189, 149Ϫ170.
F. Huisken, M. Kaloudis, A. A. Vigasin, Chem. Phys. Lett.
1
997, 269, 235Ϫ243.
[16]
K. M. Ervin, W. C. Lineberger, J. Phys. Chem. 1991, 95,
167Ϫ1177.
1
Experimental Section
[17]
T. M. Miller in CRC Handbook of Chemistry and Physics (Ed.:
D. R. Lide), 72nd ed., CRC Press, Boca Raton, Florida, 1991.
J. E. Huheey, Anorganische Chemie: Prinzipien von Struktur und
Reaktivität, Walter de Gruyter, Berlin, 1988, pp. 117Ϫ122.
B. D. Rehfuss, D.-J. Liu, B. M. Dinella, M.-F. Jagod, W. C. Ho,
M. W. Crofton, T. Oka, J. Chem. Phys. 1988, 89, 129Ϫ137.
F. Y. Naumkin, F. R. W. McCourt, J. Chem. Phys. 1997, 107,
Cryostat for Matrix Isolation: Closed-cycle compressor unit
RW2 with coldhead base unit 210 and extension module ROK from
Leybold. Ϫ Spectrometers: IR: FT-IR spectrometer IFS 85 from
[
[
[
18]
19]
20]
Ϫ1
Bruker, resolution 1 cm ; UV/Vis: Diode array spectrometer HP
8453 from Hewlett-Packard, resolution 0.5 nm. Ϫ Light Sources:
1185Ϫ1194.
Excimer laser LPX 105 from Lambda Physik, mercury high-press-
ure lamp HBO 200 from Osram, and mercury low-pressure spiral
lamp from Gräntzel.
[21]
I. L. Alberts, T. W. Rowlands, N. C. Handy, J. Chem. Phys.
1988, 88, 3811Ϫ3816.
[
[
22]
23]
M. Takami, Y. Ohshima, S. Yamamoto, Y. Matsumoto, Faraday
Discuss. Chem. Soc. 1988, 86, 1Ϫ12.
1
[D ]Acetylene ([D
1
]-4) and [D
2
]Acetylene ([D
2
]-4): At a tem-
G. T. Fraser, R. D. Suenram, F. J. Lovas, A. S. Pine, J. T.
Hougen, W. J. Lafferty, J. S. Muenter, J. Chem. Phys. 1988,
perature of Ϫ30°C, acetylene (4) was passed through 200 ml of a
stirred solution of nBuLi in hexane (ca. 1.5 ) for 15 min. The
reaction mixture was then heated to reflux for 2 h, cooled to Ϫ30°C
once more, and then 15 ml of D O was added by means of a drop-
2
ping funnel. The gaseous products were collected in a cold trap (77
K) by passing a gentle stream of helium through the apparatus.
8
9, 6028Ϫ6045.
[24]
S. Collins, J. Phys. Chem. 1990, 94, 5240Ϫ5243.
A. Bothur, Dissertation, Universität Gießen, 1997.
D. Prichard, J. S. Muenter, B. J. Howard, Chem. Phys. Lett.
[
[
25]
26]
1
987, 135, 9Ϫ15.
[27]
K. Seki, N. Nakashima, N. Nishi, M. Kinoshita, J. Chem. Phys.
1
986, 85, 274Ϫ279.
[
[
28]
29]
The synthesis yielded a mixture of 4, [D
purities, which was used without further purification.
1
]-4 and [D
2
]-4 free of im-
K. Seki, H. Okabe, J. Phys. Chem. 1993, 97, 5284Ϫ5290.
R. A. Shepherd, T. J. Doyle, W. R. M. Graham, J. Chem. Phys.
1988, 89, 2738Ϫ2742.
13
13
[30]
[
1,2- C]Acetylene ([1,2- C]-4) was purchased from Promo-
G. Maier, Angew. Chem. 1988, 100, 317Ϫ341; Angew. Chem.
Int. Ed. Engl. 1988, 27, 309Ϫ332.
chem.
[
[
[
31]
32]
33]
C. Lautz, Diplomarbeit, Universität Gießen, 1996.
K. Lanz, Dissertation, Universität Gießen, 1985.
Calculations: The calculations were performed with the GAUS-
N. Schwentner, E.-E. Koch, J. Jortner in Energy Transfer Pro-
cesses in Condensed Media (Eds.: B. Di Bartolo, A. Karipidou),
NATO ASI Series B, vol. 114, Plenum Press, New York, 1984,
SIAN-94 package of programs[35]
.
417Ϫ469.
[
1]
[34]
N. Schwentner, E.-E. Koch, J. Jortner, Electronic Excitations in
Condensed Rare Gases, Springer Tracts in Modern Physics 107,
Springer-Verlag, Berlin, 1985.
V. Hrouda, M. Roeselov a´ , T. Bally, J. Phys. Chem. A 1997,
101, 3925Ϫ3935.
35]
[
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. Peters-
son, J. A. Montgomery, K. Raghavachari, M. A. Al-Laham, V.
G. Zakrzewski, J. V. Ortiz, J. B. Foresman, J. Cioslowski, B. B.
Stefanov, A. Nanayakkara, M. Challacombe, C. Y. Peng, P. Y.
Ayala, W. Chen, M. W. Wong, J. L. Andres, E. S. Replogle, R.
Gomperts, R. L. Martin, D. J. Fox, J. S. Binkley, D. J. Defrees,
J. Baker, J. P. Stewart, M. Head-Gordon, C. Gonzalez, J. A.
Pople, GAUSSIAN-94, Revision B.1, Gaussian Inc., Pittsburgh,
Pennsylvania, 1995.
[
[
[
2]
3]
4]
E.-H. Böttcher, W. F. Schmidt, Phys. Stat. Sol. B 1984, 126,
K 165Ϫ169.
T. Kessler, R. Markus, H. Nahme, N. Schwentner, Phys. Stat.
Sol. B 1987, 139, 619Ϫ625.
W. G. Lawrence, V. A. Apkarian, J. Chem. Phys. 1992, 97,
6
199Ϫ6207.
[
[
5]
6]
G. Maier, S. Senger, Liebigs Ann. 1996, 45Ϫ47.
H. Okabe, Photochemistry of Small Molecules, John Wiley &
Sons, New York, 1978.
[
7]
A. M. Wodtke, Y. T. Lee, J. Phys. Chem. 1985, 89, 4744Ϫ4751.
[97413]
776
Eur. J. Org. Chem. 1998, 769Ϫ776