Decomposition of gaseous phthalic anhydride from a vibrationally hot molecule formed by ArF laser irradiation

Article abstract of DOI:10.1021/jp992469q

ArF laser flash photolysis was used to study the photodecomposition of gaseous phthalic anhydride (PA). The experimental conditions (concentration of the ground-state PA, the inert gas pressure, and the laser fluence) strongly influenced the transient absorption profiles. Both biphenylene and triphenylene were detected as the final products. The results indicated that benzyne (1,3-cyclohexadien-5-yne) formation occurred during the photolysis. The quenching of biphenylene formation by a foreign gas showed that biphenylene was derived from a vibrationally excited (hot) intermediate. The hot PA, the vibrational temperature of which was 2700 K, and benzyne were determined to be intermediates of the PA decomposition reaction.

Full text of DOI:10.1021/jp992469q

J. Phys. Chem. A 2000, 104, 203-208
203
Decomposition of Gaseous Phthalic Anhydride from a Vibrationally Hot Molecule Formed
by ArF Laser Irradiation
Tomoyuki Yatsuhashi* and Nobuaki Nakashima
Department of Chemistry, Graduate School of Science, Osaka City UniVersity, 3-3-138 Sugimoto,
Sumiyoshi, Osaka 558-8585, Japan
ReceiVed: July 20, 1999; In Final Form: October 20, 1999
The photodecomposition of gaseous phthalic anhydride (PA) has been studied by ArF laser photolysis. The
transient absorption profiles were strongly dependent on the concentration of the ground-state PA, the foreign
gas pressure, and the laser fluence. Both biphenylene and triphenylene were detected as the final products.
These results clearly indicated that benzyne (1,3-cyclohexadien-5-yne) was formed during the photolysis.
The quenching of biphenylene formation by a foreign gas clearly showed that biphenylene was derived from
a highly vibrationally excited (hot) intermediate. The hot PA and benzyne were found to be intermediates of
the PA decomposition reaction.
Introduction
conversion from an electronically excited state in the gas
phase.23 A hot molecule carries a very high vibrational energy
that nearly equals the absorbed photon energy. The equivalent
vibrational temperature is around 2000-4000 K for the benzene
derivatives. Many molecules such as aromatic hydrocarbons,24
olefins,25 and aliphatic hydrocarbons26 form hot molecules
during UV-VUV light irradiation. The hot molecule mechanism
for the chemical reactions has, however, been explored for only
a small variety of molecules. An accumulation of knowledge
of hot molecule chemistry will be a great help in understanding
high-temperature chemistry. In this study, decomposition of
gaseous PA was studied by ArF laser flash photolysis. The
influence of various experimental conditions such as the
concentration of PA, the laser fluence, and the inert gas pressure
on the transient species revealed that benzyne was formed from
vibrationally hot PA whose vibrational temperature was calcu-
lated to be 2700 K (Figure 1).
Benzyne, a very important intermediate in organic synthesis
processes,1 has attracted much attention for about 60 years since
the first demonstration of its existence.2 The existence of
benzyne was well proved by synthetic methods including isotope
labeling.3 The properties of benzyne were experimentally
examined and also discussed using theoretical calculations.
Numerous theoretical studies were carried out to investigate the
properties of benzyne, such as structure,4 aromaticity,5 thermo-
chemistry6 and the isomerization reaction.7 Moreover, not only
o-benzyne but also p-benzyne8 and m-benzyne9 were regarded
as remarkable intermediates. Recently, the observation of
benzdiyne (tetradehydrobenzene) was successful.10 Benzyne was
formed by the flash vacuum pyrolysis (FVP)11 or photolysis12
of the parent molecule, and then isolated in the cold matrix for
direct observations.13 In these cases, benzyne was characterized
by steady-state spectroscopic methods such as IR spectroscopy,14
photoelectron spectroscopy,15 microwave spectroscopy,16 and
NMR spectroscopy.17 However, time-resolved spectroscopy was
reported only in the 1960s for TOF-MS,18 and UV-vis
absorption spectroscopy.19 The observation of benzyne and its
dimerization process was successful in the microsecond region,
but flash photolysis in the nanosecond region was not achieved.18
The UV-vis absorption of benzyne was reinvestigated using a
steady-state matrix isolation technique.20 For the matrix isolated
condition, the photoirradiation of phthaloyl peroxide, benzocyclo-
butenedione,14b and phthalic anhydride (PA)21 resulted in the
formation of benzyne, and an interconversion among intermedi-
ates was observed. The matrix isolation is an effective method
for isolating and characterizing the active intermediate, but the
photoirradiation of matrix isolated materials may be complicated
because reactants were closely packed in the matrix.20 In the
case of FVP, a very high-temperature condition, benzyne was
also detected as an intermediate, but the interconversion among
intermediates never occurred. Biphenylene and triphenylene
were formed as final products by the FVP of PA.22
Experimental Section
Phthalic anhydride (PA, 99+%) was purchased from Aldrich
and purified by sublimation. Phthalic anhydride-d4 (PA-d4) was
synthesized by refluxing phthalic-d4 acid (Aldrich, 98% D) in
acetic anhydride, and then purified by recrystallization followed
by sublimation.27 Nitrogen gas was purchased from Osaka
Sanso, and the stated purity was 99.99%.
Crystals to be photolyzed were dissolved in benzene which
could be easily removed. About 100 µL of liquid was deposited
onto the bottom of the sample tube. The solvent was then
pumped carefully away until the background pressure was less
than 10-3 Torr, and the tube was sealed. No residue of benzene
was detected in the absorption spectra. The PA was vaporized
at 423 K in a homemade hot cell (temperature was controlled
by a chromel-alumel thermocouple and a 400 W heater). The
concentration of PA was 2.5 × 10-4 M (6.6 Torr). The accuracy
of the concentration was (5% because of the uncertainly of
the cell volume which was sealed off by hand. The maximum
concentration of PA in this experiment was well below the
calculated vapor pressure, at 423 K (18 Torr).28 A linear
correlation between the absorbance and the concentration of PA
was satisfied under the experimental condition. The ground-
The other very high-temperature condition, namely vibra-
tionally hot molecule formation, is achieved after a rapid internal
* To whom correspondence should be addressed. E-mail: tomo@
sci.osaka-cu.ac.jp.
10.1021/jp992469q CCC: $19.00 © 2000 American Chemical Society
Published on Web 12/14/1999

204 J. Phys. Chem. A, Vol. 104, No. 2, 2000
Yatsuhashi and Nakashima
Figure 1. Schematic description of the photodecomposition reaction of PA via hot state (S0**).
Figure 2. Absorption spectra of gaseous PA (423 K, 570 Torr N2,
thick line), biphenylene (423 K, 570 Torr N2, thin line), and triphenylene
(493 K, 650 Torr N2, dotted line). The absorption spectra of PA were
multiplied (× 5) to provide a clearer presentation.
state absorption of PA is too strong to observe the transient
absorption between 270 and 300 nm; therefore, the spectra of
this region were measured at a half-concentration of PA to obtain
the intensity of light transmittance. The reaction cell was kept
for over 30 min at the experimental temperature for the complete
vaporization of PA. The decomposition of PA was not observed
for several hours under this condition.
The absorption spectra were measured by a spectrophotometer
(Shimadzu UV-2400). Gaseous PA was irradiated by an ArF
laser (Lambda Physik LPF 205, 193.3 nm, fwhm 20 ns, 150
Figure 3. (a) Transient decay profiles of PA observed at 260 nm in
the absence of nitrogen (dotted line) and in the presence of 570 Torr
of nitrogen (solid line). Laser fluence was 15 mJ cm-2. (b) Stern-
Volmer plot of the transient absorption at 400 ns after the laser pulse
observed at 355 nm in the presence of 570 Torr of nitrogen. Laser
mJ/pulse) in an area of 0.5 cm × 3.0 cm. Laser fluence was
controlled by a NaCl aqueous solution filter and by adjusting
the applied voltage of the laser. Laser flurence was monitored
by a pyroelectric joule meter (Genetec ED 200). Transient
species were monitored by a pulsed Xe flash lamp (EG&G FX
425) with a perpendicular beam against the laser beam. The
transmittance light was focused into a monochromator (Acton
Research SpectraPro 150) which was connected to a photomul-
tiplier tube (Hamamatsu Photonics R758). Data were collected
by a digital oscilloscope (Sony Tektronix TDS 620B) and
analyzed on a Power Macintosh. The sample and cell were
renewed at every laser shot, so the error in the optical density
may be relatively large because of the difference in the cells.
Data were not averaged, and a single shot measurement was
carried out.
fluence was 14.8 ( 0.4 mJ cm-2
.
difficulty in transient absorption measurement of gaseous PA
due to the significant influence of photoproducts on transient
time profiles even in the second laser shot. The transient decay
profiles of the first and second laser shots were dramatically
different. A significant difference in the decay profiles was
clearly observed at 230 and 260 nm. The sample and cell were
then renewed for every laser shot in order to obtain results in
this study. However, successive laser irradiations were also
tested. After tens of laser shots, no transient was observed
because of the consumption of vapor material under the
experimental conditions. The pale yellow products, which were
insoluble in acetonitrile, were observed on the wall of the cell.
These insoluble materials were not analyzed, but would be a
high polymer of benzyne or secondary products of biphenylene.
Foreign Gas Pressure and Concentration Effects. The
effects of foreign gas and the concentration of the ground-state
PA on the transient absorption were examined. Both factors were
found to be very important. Figure 3a shows the effect of the
addition of nitrogen as the foreign gas. Zero s was defined as
the time where the transient absorption at 260 nm in the presence
of 570 Torr of nitrogen has a sharp hump as shown in Figure
3a. In the absence of nitrogen, the transient profile at 260 nm
Results and Discussion
Laser Flash Photolysis of PA and Final Products. The
absorption spectra of gaseous PA (423 K, thick line) and
biphenylene (423 K, thin line) are shown in Figure 2. PA has
a strong absorption at the laser wavelength. An ArF laser (193.3
nm) pumps PA to the S3 state. The photolyzate was dissolved
in acetonitrile, and the absorption spectra of biphenylene and a
small amount of triphenylene were clearly observed. The
formation of biphenylene and triphenylene is strong evidence
of benzyne formation as an intermediate during the photode-
composition process of PA. We may first point out some

Decomposition of Gaseous Phthalic Anhydride
J. Phys. Chem. A, Vol. 104, No. 2, 2000 205
Figure 4. Transient decay profiles at relatively low (thick line, [PA]
) 2.5 × 10-5 M) and high (thin line, [PA] ) 1.0 × 10-4 M)
concentrations of PA observed at 230 nm. Laser fluence was 15 mJ
cm-2
.
consisted of two rise components. The fast-rise component
appearing before 0 s was identical whether the nitrogen gas was
added or not. On the contrary, the slow-rise component
appearing after 0 s was very much suppressed with nitrogen,
and a stable transient was observed over several hundreds of
nanoseconds. The fast-rise component would correspond to the
excited singlet state of PA and the formation of vibrationally
hot PA. The slow-rise component would imply (1) a relaxation
process of vibrationally hot PA and hot benzyne in the early
stage, and (2) a dimerization or trimerization process of benzyne
in the latter stage. Even in the absence of a foreign gas, the
collisional relaxation of hot molecules will occur because the
concentration of the ground state of PA is high enough (6.6
Torr). Assuming that the collisional frequency is 5 × 107 Torr-1
s-1, PA collides every 3 ns.
The sharp decay, which appeared in the presence of sufficient
nitrogen, would originate in the collisional deactivation of the
vibrationally hot molecules (PA and benzyne) and the stable
absorption appearing after a sharp hump would be that of
benzyne. The decarbonylation and decarboxylation reaction of
hot PA would compete with the collisional deactivation rate by
nitrogen due to sufficient initial internal energy, unconfined
conditions, and a large enthalpy change during the decomposi-
tion reaction.29
The quenching of product by a foreign gas was then
examined. Figure 3b shows the Stern-Volmer plot of the
transient absorption with nitrogen. Biphenylene would be the
only observable product at 400 ns after the laser pulse observed
at 355 nm. The formation of biphenylene was significantly
suppressed by the addition of a foreign gas, and a fairly linear
Stern-Volmer plot was obtained. The Stern-Volmer constant
was 4.3 × 10-3 Torr-1. The dimerization of benzyne may occur
by a three-body collision mechanism. The increase of foreign
gas will enhance the dimerization reaction if the benzyne
formation was not affected by foreign gas. However, the yield
of biphenylene decreases with increasing nitrogen gas pressure
due to the collisional deactivation of vibrationally hot phthalic
anhydride, the source of benzyne. All the results indicate that
the biphenylene should be derived from the vibrationally excited
molecule.
Figure 5. (a) Normalized transient absorption profiles at different laser
fluences. Trace (I) was measured at 35 mJ cm-2; trace (II) was observed
at 15 mJ cm-2. Each profile was normalized at 0 s. (b) Laser fluence
dependency of the transient absorption at 400 ns after the laser
irradiation in the presence of 570 Torr of nitrogen observed at 260 nm
(O) and at 355 nm (b). The solid lines have a slope of 1.0 (O) and 2.0
(b), respectively.
was calculated to be 2.9 × 1010 s-1, assuming the collisional
rate constant of 5 × 107 Torr-1 s-1 30
. Therefore, a hot molecule
should not live for a long time. The influence of the concentra-
tion on the transient profiles, even in the presence of nitrogen,
clearly shows the involvement of the intermolecular process in
the transient profiles. The transient decay profile at 230 nm looks
complicated, but it can be easily explained. The fast rise and
decay around 0 s would imply at least three processes, such as
(1) the depletion of the ground-state PA, (2) the formation of
vibrationally hot PA, and (3) the collisional deactivation of the
vibrationally hot PA with N2. The shape of the hump depends
on the molar extinction coefficient of these related species. As
clearly shown in Figure 4, the subsequent slow rise was strongly
dependent on the concentration of PA. At 230 nm, the dominant
species at 400 ns after a laser pulse would be biphenylene, which
has a strong absorption, as shown in Figure 2. The rise was
attributed to the dimerization process of benzyne, and the initial
concentration of benzyne was proportional to that of PA. On
the contrary, a significant concentration effect was not observed
at 290 nm (not shown), where the ground-state PA has a strong
absorption. These results indicate that the dimerization of
benzyne, rather than the reaction between PA and benzyne, is
important.
Benzyne Formation by Single-Photon Process. The laser
fluence dependencies of the transient absorption were examined.
A definite difference was observed in the transient profile
depending on the laser fluences. Figure 5a shows the normalized
transient profiles at 260 nm. Trace (I) was measured at a
relatively high laser fluence (35 mJ cm-2), and trace (II) was
observed at 15 mJ cm-2. In the low-laser fluence region,
transient time profiles consisted of a sharp hump, which was
assigned to be the formation and collisional deactivation of the
Figure 4 shows the transient absorption profiles at 230 nm
obtained at relatively low (thick line, 2.5 × 10-5 M) and at
high (thin line, 1.0 × 10-4 M) concentrations of PA. In the
presence of 570 Torr of nitrogen, the frequency of collision

206 J. Phys. Chem. A, Vol. 104, No. 2, 2000
Yatsuhashi and Nakashima
Figure 6. Transient absorption spectra of PA at 0 s (b) and 400 ns (O) after the laser excitation. (a) Low-laser fluence condition in the presence
of 570 Torr of nitrogen (15.1 ( 0.9 mJ cm-2). (b) Low-laser fluence condition in the absence of nitrogen (15.0 ( 0.6 mJ cm-2). (c) High-laser
fluence condition in the presence of 570 Torr of nitrogen (36.1 ( 1.5 mJ cm-2). (d) Corrected transient absorption spectra of Figure 6a. The solid
line on the corrected spectrum is the simulated spectrum of hot molecule by modified Sulzer-Wieland model calculation (see the text).
vibrationally excited molecule, and a long-lived absorption. In
the high-laser fluence region, the rise component after the small
hump was clearly observed even in the presence of nitrogen.
This behavior should not originate in the hot species. The laser
fluence dependency on the optical density was then examined
(Figure 5b). The slope of the plot (open circle, 260 nm) was
greater than 1 in the high-fluence region (>10 mJ cm-2). These
results indicated that the observed transient at the high-laser
fluence condition was formed by an intermolecular reaction.
The most likely species that appears at the high-laser fluence
condition would be triphenylene, which has a strong absorption
at 260 nm (2 × 104 M-1 cm-1), as shown in Figure 2. After a
single-shot laser irradiation at 35 mJ cm-2, the formation of
triphenylene was clearly observed. The formation of tri-
phenylene contributes to the transient absorption at 260 nm in
the high-laser fluence region.
Figure 5b also shows the laser fluence dependencies observed
at 355 nm. The slope (2.0) indicated that the transient was
formed by a bimolecular reaction or a two-photon reaction. The
latter possibility will be neglected because the transient has a
rise profile at 355 nm. In the region of 355 nm at 400 ns after
laser irradiation, neither hot PA nor benzyne will have a
significant absorption. At 400 ns after the laser pulse, biphen-
ylene would be a dominant product. Biphenylene is formed by
the dimerization of benzyne, and benzyne should be formed by
a single-photon reaction. Porter et al. reported the photodecom-
position of PA and the formation of benzyne and its dimerization
by a microsecond flash lamp excitation.19b The PA could be
decomposed by a single-photon process with ArF laser excita-
tion.
spectra were recorded under different conditions. The spectra
exhibited strong bleaching between 270 and 300 nm under all
conditions, which corresponded to the depletion of the ground-
state PA. The absorbance at 0 s monotonically decreased with
increasing wavelength, and no vibrational structure was observed
in the region of 300-370 nm under all conditions, where the
ground-state PA did not have significant absorption. The
monotonic spectra were understood to be those of a hot molecule
by analogy to the other hot molecules.31 The highly vibrationally
excited PA, hot PA, has broader and structureless spectra. The
absorption spectrum of the hot molecule was successfully
simulated for many molecules such as benzene32 and naphtha-
lene,33 which have high vibrational temperatures of 3390 and
2320 K, respectively. The transient spectra of PA were then
adjusted until there was no spectral structure of the ground-
state PA. The corrected transient absorption spectrum of PA is
shown in Figure 6d. The other spectra taken under different
conditions were also corrected, and they were essentially
identical at 0 s. The solid line on the corrected spectrum in
Figure 6d was the theoretical simulation curve of the hot
molecule spectrum by modified Sulzer-Wieland model.34 PA
hardly emits fluorescence above room temperature.35 The triplet
yield of PA was not definitively determined,36 but the internal
conversion may be a dominant process by analogy, with many
aromatic molecules whose internal conversion yield is nearly
unity when they are excited by a VUV laser in the gas phase.23
Based on these facts, it is reasonable that the hot PA, which
was instantaneously formed with a laser pulse, is a dominant
absorber at 0 s.
Benzyne, however, also has a broadened spectrum between
250 and 400 nm in a matrix isolated condition.20 It is difficult
to distinguish the vibrationally hot PA from benzyne because
Benzyne Formation from Hot Phthalic Anhydride. The
transient absorption spectra of PA are shown in Figure 6. The

Decomposition of Gaseous Phthalic Anhydride
J. Phys. Chem. A, Vol. 104, No. 2, 2000 207
the absorption spectra may overlap in the entire region. The
clue to distinguishing a hot spectrum from that of a vibrationally
ground state was the rapid collisional decay (sharp hump)
induced by a foreign gas, as seen in Figure 3a. The rapid decay
was observed up to 350 nm, depending on the experimental
conditions. At longer wavelengths, the identification of the sharp
hump induced by a foreign gas was difficult because of the
low signal-to-noise value. After 400 ns, the structured spectrum
of biphenylene (340-370 nm) clearly appeared under the high
laser fluence conditions as shown in Figure 6c.
were then compared after the laser experiments. Again, no
significant difference was observed. The deuterium isotope effect
was, however, seen in the dissociation process of the alkyl
benzene derivatives.39 The difference between the alkyl benzene
derivatives and PA could be attributed to the position of the
deuteration. The hydrogen was directly attached to the C-C
bond, which will be cleaved in the case of alkyl benzene
derivatives, whereas hydrogen was not attached to the quaternary
benzenoid carbon and the carbonyl carbon, which will be
cleaved.
The estimation of the vibrational temperature of the hot
molecule is important when considering the chemical reaction
of the hot molecule. The vibrational temperature of the hot PA,
q, was then calculated using the equation below:33
Acknowledgment. The present research was partially sup-
ported by a Grant-in-aid (No. 11750720) from the Ministry of
Education, Science, Sports and Culture, Japan.
References and Notes
E0 (ini) ) hυ (193.3 nm) + EkT (423 K) )
(1) Levin, R. H. In ReactiVe Intermediates, Vol. 1; Jones, M., Moss,
R. A., Eds.; Wiley and Sons: New York, 1978; Chapter 1. Isaacs, N. S.
ReactiVe Intermediates in Organic Chemistry; Wiley and Sons: New York,
1974; Chapter 7.
(2) Wittig, G.; Pieper, G.; Fuhrmann, G. Chem. Ber. 1940, 73, 1193.
Wittig, G.; Pieper, G.; Fuhrmann, G. Ber. Dtsch. Chem. Ges. 1940, 73,
193.
(3) Heaney, H. Chem. ReV. 1962, 62, 81.
(4) Langgenaeker, W.; Proft, F. D.; Geerlings, P. J. Phys. Chem. A
1998, 102, 5944.
(5) Jiao, H.; Shleyer, P. R.; Beno, B. R.; Houk, K. N.; Warmuth, R.
Angew. Chem., Int. Ed. Engl. 1997, 36, 2761.
(6) Cramer, C. J.; Nash, J. J.; Squires, R. R. Chem. Phys. Lett. 1997,
277, 311.
(7) Cioslowski, J.; Piskorz, P.; Moncrieff, D. J. Am. Chem. Soc. 1998,
120, 1695. Burton, N. A.; Quelch, G. E.; Gallo, M. M.; Shaefer, H. F. J.
Am. Chem. Soc. 1991, 113, 764.
(8) Marquardt, R.; Balster, A.; Sander, W.; Kraka, E.; Cremer, D.;
Radziszewski, J. G. Angew. Chem., Int. Ed. Engl. 1998, 37, 955.
(9) Marquardt, R.; Sander, W.; Kraka, E. Angew. Chem., Int. Ed. Engl.
1996, 35, 746. Kraka, E.; Cremer, D.; Bucher, G.; Wandel, H.; Sander, W.
Chem. Phys. Lett. 1997, 268, 313.
(10) Moriyama, M.; Ohana, T.; Yabe, A. Chem. Lett. 1995, 557; J. Am.
Chem. Soc. 1997, 119, 10229.
(11) Brown, R. F. C. Pyrolytic Methods in Organic Chemistry; Academic
Press: New York, 1980.
(12) Givans, R. S.; Levi, N. In Supplement B: The Chemistry of Acid
DeriVatiVes. Part 1; Patai, S., Ed.; Wiley and Sons: Chichester, 1979;
Chapter 11.
(13) Seybold, G. Angew. Chem., Int. Ed. Engl. 1977, 16, 365.
(14) (a) Nam, H. H.; Leroi, G. E. J. Mol. Struct. 1987, 157, 301. (b)
Chapman, O. L.; Mattes, K.; McIntosh, C. L.; Pacansky, J.; Calder, G. V.;
Orr, G. J. Am. Chem. Soc. 1973, 95, 6134. Chapman, O. L.; Chang, C. C.;
Kolc, J.; Rosenquist, N. R.; Tomioka, H. J. Am. Chem. Soc. 1975, 97, 6586.
(15) Wenthold, P. G.; Squires, R. R.; Lineberge, W. C. J. Am. Chem.
Soc. 1998, 120, 5279. Werstiuk, N. H.; Roy, C. D.; Ma, J. Can. J. Chem.
1995, 146, 146. Leopold, D. G.; Miller, A. E. S.; Lineberger, W. C. J. Am.
Chem. Soc. 1986, 108, 1379. Dewar, M. J. S.; Tien, T. J. Chem. Soc., Chem.
Commun. 1985, 1243.
39
hυi
hυ (193.3 nm) +
∑
hυi
i)1
exp
- 1
(
)
kBT
39
hυi
hυi
E (vib) )
∑
0
i)1
exp
- 1
(
)
kBθ
where E0 is the internal energy of the molecule, EkT is the
vibrational energy at the experimental temperature, h is Planck’s
constant, T is the experimental temperature (423 K), ni is the
normal vibrational mode of PA, and kB is the Boltzmann
constant. If the initial energy, E0 (ini), is distributed to the
vibrational modes of PA (υi, i ) 1 ∼ 39), E0 (ini) should equal
the vibrational energy, E0 (vib). PA has 39 normal vibrational
modes, including 33 IR active modes and 6 IR inactive modes.
All the vibrational modes were not experimentally obtained.
An ab initio calculation (Gaussian 94, HF/6-31G(dp))37 was then
carried out to theoretically obtain all the modes. The scaling
factor (0.91) was used for adjusting the calculated vibrational
frequencies to the experimental results, and used for equivalent
vibrational temperature calculation.38 Thus the vibrational
temperatures were 2694 K (PA-h4, single-photon absorption)
and 2596 K (PA-d4, single-photon absorption). These high-
temperature conditions may be enough to induce the decom-
position reaction of PA, which releases carbon dioxide and
carbon monoxide.
(16) Brown, R. D.; Godfrey, P. D.; Rodler, M. J. Am. Chem. Soc. 1986,
The influence of deuteration on the benzenoid ring was
examined. It was assumed that the degree of deuteration in the
PA-d4 was the same as in the deuterated phthalic acid (98%)
from which the PA-d4 was synthesized. The transient absorption
profiles were compared at 260 nm. The laser flash photolysis
experiments were performed using the saturated PA concentra-
tion. No significant difference between the deuterated and
nondeuterated PA was observed in either the initial optical
density or rise profiles at different temperatures of 393 and 403
K. The former result (initial optical density) meant that the
internal conversion yields of PA-h4 and PA-d4 were identical.
When the molecules were excited to the higher energy state,
the internal conversion rate became so fast23 that it diminished
the effect of the deuteration, because the density of states was
very high whether molecules were deuterated or not. The latter
result (rise profile) indicated that the collisional deactivation
of the hot PA with the ground-state PA was not affected by the
deuteration on the benzenoid ring. The yields of biphenylene
108, 1296.
(17) Warmuth, R. Chem. Commun. 1998, 59. Angew. Chem., Int. Ed.
Engl. 1997, 36, 1347. Orendt, A. M.; Facelli, J. C.; Radziszewski, J. G.;
Horton, W. J.; Grant, D. M.; Michl, J. J. Am. Chem. Soc. 1996, 118, 846.
(18) Berry, R. S.; Clardy, J.; Schafer, M. E. J. Am. Chem. Soc. 1964,
86, 2738.
(19) (a) Schafer, M. E.; Berry, R. S. J. Am. Chem. Soc. 1965, 87, 4497.
Berry, R. S.; Spokes, G. N.; Stiles, M. J. Am. Chem. Soc. 1962, 84, 3570.
Berry, R. S.; Spokes, G. N.; Stiles, R. M. J. Am. Chem. Soc. 1960, 82,
5240. (b) Porter, G.; Steinfeld, J. I. J. Chem. Soc. A 1968, 877.
(20) Mu¨nzel, N.; Schweig, A. Chem. Phys. Lett. 1988, 147, 192. Simon,
J. G. G.; Mu¨nzel, N.; Schweig, A. Chem. Phys. Lett. 1990, 170, 187.
(21) Radziszewski, J. G.; Hess, B. A.; Zahradnik, R. J. Am. Chem. Soc.
1992, 114, 52.
(22) Fields, E. K.; Meyerson, S. Chem. Commun. 1965, 474. Brown,
R. F. C.; Garedner, D. V.; McOmie, J. F. W.; Solly, R. K. Chem. Commun.
1966, 408.
(23) Nakashima, N.; Yoshihara, K. J. Phys. Chem. 1989, 93, 7763.
(24) Shimizu, S.; Nakashima, N.; Sakata, Y. Chem. Phys. Lett. 1998,
284, 396.
(25) Nakashima, N.; Ikeda, N.; Shimo, N.; Yoshihara, K. J. Chem. Phys.
1987, 87, 3471.

208 J. Phys. Chem. A, Vol. 104, No. 2, 2000
Yatsuhashi and Nakashima
(26) Shimada, T. Thesis, Osaka University, 1993.
Lett. 1996, 263, 197.
(27) Mayo, D. W.; Pike, R. M.; Butcher, S. S. Microscale Organic
Laboratory, 2nd ed.; Wiley and Sons: New York, 1986; p 273.
(28) Dean, J. A., Ed. Lange’s Handbook of Chemistry, 13th ed.;
McGraw-Hill: New York, 1985; Section 10.
(34) Hippler, H.; Troe, J.; Wendelken, H. J. J. Chem. Phys. 1983, 78,
5351. Astholz, D. C.; Brouwer, L.; Troe, J. Ber Bunsen-Ges. Phys. Chem.
1981, 85, 559. Reference 31.
(35) Hardy, G. E.; Baldwin, J. C.; Zink, J. I.; Kaska, W. C.; Liu, P.;
Dubois, L. J. Am. Chem. Soc. 1977, 99, 3552. Zink, J. I. J. Am. Chem. Soc.
1974, 96, 6775.
(36) Sano, M.; Narisawa, T.; Matuka, M.; I’Haya, Y. J. Bull. Chem.
Soc. Jpn. 1977, 50, 0, 2266. Iwata, S.; Tanaka, J.; Nagakura, S. J. Chem.
Phys. 1967, 47, 2203.
(37) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Gill, P. M. W.;
Johnson, B. G.; Robb, M. A.; Cheeseman, J. R.; Keith, T.; Petersson, G.
A.; Montgomery, J. A.; Raghavachari, K.; Al-Laham, M. A.; Zakrzewski,
V. G.; Ortiz, J. V.; Foresman, J. B.; Cioslowski, J.; Stefanov, B. B.;
Nanayakkara, A.; Challacombe, M.; Peng, C. Y.; Ayala, P. Y.; Chen, W.;
Wong, M. W.; Andres, J. L.; Replogle, E. S.; Gomperts, R.; Martin, R. L.;
Fox, D. J.; Binkley, J. S.; Defrees, D. J.; Baker, J.; Stewart, J. P.; Head-
Gordon, M.; Gonzalez, C.; Pople, J. A. Gaussian 94, SGI-G94 Rev. E.1;
Gaussian, Inc.: Pittsburgh, PA, 1995.
(38) Biernacki, P. R.; Kaszynski, P.; Andes Hess, B., Jr.; Thulstrup, E.
W.; Radziszewski, J. G. J. Phys. Chem. 1995, 99, 6309.
(39) Kajii, Y.; Obi, K.; Tanaka, I.; Ikeda, N.; Nakashima, N.; Yoshihara,
K. J. Chem. Phys. 1987, 86, 6115.
(29) We have not succeeded in measuring the rise time of benzyne.
Benzyne, which showed absorption in the UV region shorter than 400 nm,
was decreased by adding foreign gases; therefore, hot PA does not produce
immediately hot benzyne. Benzyne was still observable even in the presence
of 570 Torr of nitrogen (Figure 3a). The collision time is estimated to be
35 ps, and every collision does not always quench the formation of benzyne.
The rise time can be estimated to be an order of 100 ps on the basis of the
foreign gas effects and is beyond our time resolution. The calculated enthalpy
change of PA to benzyne was 313 kJ mol-1. The enthalpies of formation
are: phthalic anhydride (-371 kJ/mol), carbon dioxide (-393 kJ/mol),
carbon monoxide (-110 kJ/mol), benzyne (443 kJ/mol). These values were
taken from ref 8 and Wenthold, P. G.; Paulino, J. A.; Squires, R. R. J. Am.
Chem. Soc. 1991, 113, 7414.
(30) Hipper, H.; Troe, J.; Wendelken, H. J. J. Chem. Phys. 1983, 78,
6709.
(31) Sulzer, P.; Wieland, K. HelV. Phys. Acta 1952, 25, 653.
(32) Nakashima, N.; Yoshihara, K. J. Chem. Phys. 1983, 79, 2727.
(33) Suzuki, T.; Ichimura, T.; Kusaba, M.; Nakashima, N. Chem. Phys.