Photolysis of a benzyne precursor studied by time-resolved FTIR spectroscopy

Article abstract of DOI:10.1021/acs.joc.5b02678

The 266 nm laser flash photolysis of phtaloyl peroxide (2) in liquid acetonitrile solution at room temperature has been investigated. Upon 266 nm laser irradiation, 2 is effectively photodecarboxylated leading to the formation of o-benzyne (1) and two equivalents of CO₂, yet a small fraction of photolyzed 2 follows a different pathway leading to 6-oxocyclohexa-2,4-dienylideneketene (3) and one equivalent of CO₂. Compound 3 is kinetically reactive and reacts in the microsecond time scale following a first-order kinetic law. The presence of 1 in the photolysis experiment is confirmed by trapping experiments with methyl 1-methylpyrrole-2-carboxylate (6). The Diels-Alder reaction between 1 and 6 occurs under the selected experimental conditions on a time scale shorter than 100 ms.

Full text of DOI:10.1021/acs.joc.5b02678

Article
pubs.acs.org/joc
Photolysis of a Benzyne Precursor Studied by Time-Resolved FTIR
Spectroscopy
Joel Torres-Alacan*
Institute for Physical and Theoretical Chemistry, Rheinische Friedrich-Wilhelms-Universtitat, Wegelerstraße 12, 53115 Bonn,
̈
Germany
S
* Supporting Information
ABSTRACT: The 266 nm laser flash photolysis of phtaloyl peroxide (2) in
liquid acetonitrile solution at room temperature has been investigated. Upon
266 nm laser irradiation, 2 is effectively photodecarboxylated leading to the
formation of o-benzyne (1) and two equivalents of CO2, yet a small fraction
of photolyzed 2 follows a different pathway leading to 6-oxocyclohexa-2,4-
dienylideneketene (3) and one equivalent of CO2. Compound 3 is kinetically reactive and reacts in the microsecond time scale
following a first-order kinetic law. The presence of 1 in the photolysis experiment is confirmed by trapping experiments with
methyl 1-methylpyrrole-2-carboxylate (6). The Diels−Alder reaction between 1 and 6 occurs under the selected experimental
conditions on a time scale shorter than 100 ms.
steps involving 1 or arynes in general proceed with a high
degree of regioselectivity. Although useful models able to
explain these regioselective trends are emerging,21 real-time
mechanistic studies are still missing.
When it comes to spectroscopy of functional groups,
Chapman and co-workers were the first to provide an infrared
spectrum of 1 in argon matrices at 8 K by photolysis of phtaloyl
peroxide 2.22 Chapman and co-workers observed that 1 reacts
with furan even at 50 K, forming 1,4-dihydronaphtalene-1,4-
endoxide (Scheme 2). In a remarkable work, Radziszewski et
INTRODUCTION
■
1,2-Dehydrobenzene or o-benzyne (1) was first proposed as a
reactive intermediate in 1927 by Bachmann and Clarke.1 In the
following years, Wittig2,3 greatly contributed to the under-
standing of its structure and reactivity. In 1953, Roberts et al.4
found, in an elegant experiment, that when [1-14C]-
chlorobenzene is reacted with potassium amide in liquid
ammonia, [1-14C]aminobenzene and [2-14C]aminobenzene are
formed in nearly the same proportions. This experiment
provided very strong evidence for the existence of 1 as a
reactive intermediate in polar liquid phases (Scheme 1). The
fascinating history of 1 is the subject of several review
articles.5−7
Scheme 2. Synthesis and Reactivity of 1 in Argon Matrices
Scheme 1
al.23 revisited the IR spectrum of 1 in inert cryogenic matrices
and finally located the weak (2.0
0.4 km/mol) CC
stretching vibration of 1 at 1846 cm−1, which is 120 cm−1 less
than that previously predicted by theoretical calculations.24
Because 1 is very reactive, the large body of spectroscopic
data about it has been collected in the gas25,26 or solid10 phase
at low temperatures. Spectroscopic data of 1 in liquid solution
at room temperature is very scarce. In this regard, the NMR
spectra of 1 trapped in the cavity of a hemicarcerand in liquid
[D8]-THF measured by Warmuth deserve special mention.27
Although much structural and spectroscopic information has
been collected on 1, temporal information is still missing at
least in the liquid phase. In this regard, time-resolved
spectroscopy, especially with IR detection, has proven to be a
On the basis of the large (37.5 0.2 kcal/mol)8,9 singlet−
triplet energy gap and short (124 2 pm)10 bond length, 1 is
better described as a strained alkyne than as a biradical.
Notably, metal complexes containing 1 forming part of the
metal coordination sphere have been crystallized, and their
structures have been resolved by X-rays.11−13
The strained triple bond present in 1 and in arynes in general
enables them to participate as reactive intermediates in
nucleophilic additions,14 pericyclic reactions such as Diels−
Alder cycloadditions,15 dipolar cycloadditions,16 metal-cata-
lyzed17 reactions, and others.18 Therefore, it is not surprising
that the search for easy-to-handle precursors of 1 still attracts
much attention.19 It is remarkable that more than 75
independent natural products have been synthesized using
aryne intermediates.20 A significant number of those synthetic
Received: November 23, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.joc.5b02678
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
good choice in the study of reactive intermediates and reaction
mechanisms.28−30
In this paper, I describe a systematic investigation of the
photolysis of 2 at room temperature in liquid acetonitrile using
time-resolved FTIR spectroscopy. The main goal is to
determine whether 1 can be generated from 2 by the action
of a single 266 nm photon.
electron-density between the initial and final states requires
consideration. Figure 1b shows the difference electron density
plot for the transition most resonant with the 266 nm
wavelength together with the Lewis structure of 2 for better
visualization. The blue and red regions in the difference
electron density plot represent negative and positive values,
respectively. The electron density plot suggests that, upon 266
nm laser excitation, electron density from ring-π electrons is
transferred directly into the carbonyl carbon atoms and oxygen
atoms of the peroxide unit simultaneously. One can anticipate
that the simultaneous allocation of electron density on the CO2
moieties in 2 might induce their desired coherent photo-
dissociation, leading to 1.
RESULTS AND DISCUSSION
■
1. UV−Vis Spectroscopy. The electronic spectrum of 2 in
acetonitrile at room temperature is located entirely in the UV
section, and it is asymmetrically divided into two regions by a
shallow minimum around 270 nm (solid curve, Figure 1a). The
2. Rapid-Scan Time-Resolved FTIR Spectroscopy.
Figure 2a shows the FTIR absorption spectrum of CO2 in
Figure 2. (a) Scaled and shifted FTIR spectrum of CO2 in CH3CN.
(b) Scaled and shifted FTIR spectrum of 2 in CD3CN (black) and
CH3CN (gray). (c) Difference absorption spectrum recorded 100 ms
after 266 nm pulse excitation of the sample of 2 in Ar-purged CD3CN
(black) and CH3CN (gray).
acetonitrile in the spectral region from 1900 to 2400 cm−1. The
strongest IR absorption of CO2 peaks at 2343 cm−1 as indicated
by the arrow. Figure 2b illustrates the scale and shifted FTIR
absorption spectrum of 2 recorded in deuterated (black curve)
and nondeuterated (gray curve) acetonitrile to circumvent the
lack of IR transparency of acetonitrile in the region from 1350
to 1600 cm−1. The strongest absorptions of 2 are located at
1287 and 1765 cm−1. A normal-mode analysis conducted at the
B3LYP/TZVP level of theory indicates that the 1287 cm−1
absorption is due to the coupling of ring stretching and in-plane
bending of the C−H and CO bonds along the entire
molecule, whereas the 1765 cm−1 is the result of out-of-phase
and in-phase stretching combinations of both CO groups in
2. Figure 2c shows a representative time-resolved difference
FTIR absorption spectrum recorded 100 ms after 266 nm pulse
excitation of 2 in acetonitrile under an inert argon atmosphere.
The bands pointing downward arise from the increased
transmission in the spectral region where 2 strongly absorbs
and indicates its clean photolysis. Concomitantly, a new strong
band at 2343 cm−1 due to CO2 appears (compare to Figure 2a).
The temporal formation of this band cannot be resolved with
the limited time resolution of our rapid-scan FTIR setup (∼60
ms).
Figure 1. (a) Electronic spectrum of phtaloyl peroxide 2 in
acetonitrile. The solid curve represents the experimental data.
Absorption maxima are indicated by the arrows. The bars represent
the electronic spectrum as calculated using TDDFT (see text for
details). The blue dashed line at 266 nm represents the wavelength of
the laser employed for photolysis. (b) Three dimensional plot
corresponding to the electron-density shift induced by the 266 nm
laser excitation. (c) Lewis structure of 2 with the same orientation as
that used for the difference electron density representation.
low energy region features two partially resolved absorptions
peaking at 297 and 289 nm and the high energy region a much
broader band with a maximum at 243 nm. For understanding
the underlying nature of those electronic transitions, time-
dependent density functional theory (TDDFT) was employed.
The solid bars in Figure 1a represent the BP86/def2-aug-
TZVPP-TDDFT calculated electronic spectrum of 2 in which
solvent effects were considered. As anticipated, the lowest
energy transition, peaking at 297 nm, is mainly of HOMO →
LUMO character, whereas its neighbor band at 289 nm is
mostly the result of a HOMO-2 → LUMO transition (for
details, see the Supporting Information). Interestingly, the
TDDFT calculation shows that, close to the excitation
wavelength (blue dashed line, Figure 1a), there is a transition
that is mainly of HOMO-1 → LUMO character, although it
contains contributions of higher energy excitations. To take
into account the contribution of the complete set of individual
excitations to the final electronic state, the difference of
Remarkably, there is no formation of IR absorption bands in
the regions around 1287 and 1765 cm−1, where 2 strongly
absorbs. The absence of pump-induced absorptions around
1765 cm−1, where carbonyl-containing compounds significantly
absorb, is a strong indication that 2 is efficiently photo-
converted to 1 and 2 equiv of CO2 (path a, Scheme 3) in a
B
DOI: 10.1021/acs.joc.5b02678
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
and 80 μs after the laser excitation, it is no longer detectable
(blue spectrum, Figure 3a). A careful analysis performed on a
large set of experimental data shows there is an additional weak
absorption at 1636 cm−1 that decays with a time constant equal
to that measured for the 2140 cm−1 absorption. The
absorptions at 1636 cm−1 (carbonyl) and 2140 cm−1 (ketene)
are ascribed to 6-oxocyclohexa-2,3-dienylideneketene (3). This
assignment is supported by the similarity to the IR absorptions
(1650 and 2139 cm−1) of 3 isolated in inert cryogenic
matrices.31,32 Also, Liu et al.33 generated 3 by 308 nm laser
photolysis of 2-phenyl-3,1-benzoxan-4-one in acetonitrile at
room temperature and observed a transient absorption at 2135
2 cm−1 that decayed following first-order kinetics in the
microsecond time scale. Furthermore, the transient absorptions
of 3 are readily quenched if methanol is added to the
acetonitrile solution before photolysis.
The step-scan experiments suggest there are two competing
mechanisms in the photolysis of 2. On the one hand, the
resonance photoejection of two CO2 molecules from 2 leads to
the desired formation of 1 (path a, Scheme 3), and on the other
hand, the ejection of a single CO2 molecule from 2 leads to the
ultimate formation of 3 (path b, Scheme 3). For quantification
of the proportions of these competing photodissociation
pathways, spectral simulations were carried out. Figure 3b
shows the simulated transient spectrum at 200 ns using the
theoretical IR spectra of the species 1−3 and CO2 obtained
from DFT (B3LYP/TZVP) calculations (for details, see SI). In
the simulation, an 80% yield of 1 and a 20% yield of 3 is
assumed. The nice agreement between experiment and theory
suggests that the 266 nm photolysis of 2 in acetonitrile is a
feasible route to 1 in liquid solution.
Scheme 3. Photolysis at 266 nm of 2 in Liquid Acetonitrile at
Room Temperature
single photon process. Furthermore, bands around 1287 cm−1
due to photoproducts are expected to be less intense than that
of parent 2 because the coupling mechanism with the CO
oscillators is no longer operating and they might well be buried
in the baseline noise.
Organic reactive intermediates usually have lifetimes on the
microsecond time scale and therefore the temporal resolution
that rapid-scan FTIR spectroscopy provides is insufficient. In
the next section, to circumvent the intrinsic temporal
limitations of rapid-scan FTIR spectroscopy on the one hand
and to establish the presence of elusive intermediates in the
photolysis of 2 on the other, photolysis experiments on 2
monitored with nanosecond temporal resolution are described.
3. Step-Scan FTIR Spectroscopy. Representative step-
scan time-resolved difference absorption spectra recorded 200
ns (red), 4 μs (orange), 16 μs (green), and 80 μs (blue) after
the absorption of a single 266 nm laser pulse by 2 in liquid
acetonitrile are illustrated in Figure 3a. All transient spectra
The limited time-resolution (200 ns) of the step-scan
spectrometer does not allow one to resolve the formation of
3. The formation of 3 can take place by incomplete
photodecarboxylation of 2 coupled with an oxygen atom
transfer (path b, Scheme 3). This incomplete photodecarbox-
ylation pathway might occur in an excited or a ground potential
energy surface with or without benzopropiolactone 4 as reactive
intermediate. Alternatively, 1 could react with CO2 (in cage),
leading to 3 with 4 as a reactive intermediate (Scheme 4).
Scheme 4
Figure 3. (a) Step-scan FTIR difference absorption spectra recorded
for 200 ns (red), 4 μs (orange), 16 μs (green), and 80 μs (blue) after
266 nm laser-pulsed photolysis. The arrow indicates the spectral
evolution of the transient IR absorption band with time. (b) DFT
simulation of the 200 ns transient absorption spectrum (see text for
details).
show a negative absorption at 1767 cm−1, indicating the net
photolysis of 2 and a gigantic positive absorption at 2343 cm−1
due to the simultaneous generation of CO2. Even at the
shortest delay time, 200 ns, both the transient depletion of 2
and the formation of CO2 measured by their corresponding IR
resonances are in their corresponding stationary values,
indicating that the primary process leading to the actual
dissociation of CO2 from 2 and its thermal equilibration occur
in time scales well below those accessible by the step-scan
spectrometer. In contrast with the rapid-scan experiments, the
step-scan difference absorption spectra show a previously
undetected transient absorption at 2140 cm−1. The species
responsible for the 2140 cm−1 absorption reacts following a
(pseudo)first-order kinetic law with a time constant of 22.5 μs,
If the latter pathway is operating, conducting the photolysis
in a CO2 atmosphere should enhance the formation of 4 and 3.
The CO group in 4 gives rise to a strong IR resonance at 1904
cm−1 in Ar matrices at 8 K.31 Therefore, the presence of 4
could be readily observed in the time-resolved FTIR experi-
ments. Furthermore, DFT calculations suggest the trans-
formation 4 → 3 is exergonic (ΔG = −9.0 kcal/mol), but it
is hampered by a thermal barrier of 5.4 kcal/mol. For
establishing whether 1 reacts with CO2 leading to the ultimate
formation of 3, the step-scan experiments on 2 were carried out
in CO2 atmosphere. These experiments showed that the yield
of 3 is unaffected by the presence of CO2 in the reacting
system, and therefore 3 should be formed directly from 2 by
incomplete photodecarboxylation. Because the absorptions
C
DOI: 10.1021/acs.joc.5b02678
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
bands of 4 were not detected, it can be concluded that 1 does
not react with CO2 in liquid acetonitrile.
octyl azide36 and 2 form acetonitrile solutions that are stable for
indefinite periods of time. Octyl azide is photo inert to 266 nm
laser light; however, the much anticipated [3 + 2] cyclo-
addition16 reaction with 1 does not take place in acetonitrile
solvent. DFT calculations suggest the [3 + 2] cycloaddition
between 1 and alkyl azides is hampered by a significant thermal
barrier.21 However, it is important to point out that [3 + 2]
cycloadditions between 1 and azides have been conducted via
photolysis of 2 in dichloroethane solvent at room temper-
ature.16 Conducting the photolysis of 2 in CO-saturated
acetonitrile gives rise to experimental data that is identical to
that obtained when experiments are carried out in Ar
atmosphere, thus confirming that a reaction between 1 and
CO giving rise to benzocyclopropenone requires activation.37
Inspired by the quenching experiment on 1 with furan at 50
K in an argon matrix by Chapman,22 I decided to use the
Diels−Alder reactivity of 1 as a probe of its presence. For
enhancement of the IR sensitivity of the corresponding Diels−
Alder reaction, methyl 1-methylpyrrole-2-carboxylate (6) was
selected. This quencher has the benefit of containing a built-in
IR-CO chromophore whose temporal detection along the
quenching reaction should in principle be feasible. It can be
anticipated that the CO stretching frequency should shift to
higher frequencies in the Diels−Alder adduct 7 with respect to
that in 6 due to the missing conjugation in the final product.
Compound 6 has the disadvantage that its UV−vis spectrum
has a sharp maximum at 267 nm, which is very close to the
excitation wavelength. Because the opacity of quencher 6 is
high at 266 nm, “background” flash photolysis experiments on
6 alone in acetonitrile were carried out. The laser energy used
in the photolysis experiments with 6 alone was at least equal to
that employed in similar experiments with 2. Under these
experimental conditions, 6 does not undergo net photolysis,
which means all photoexcited molecules of 6 return back to the
ground state without experiencing any chemical transformation.
Having established 6 is resistant to the action of 266 nm laser
light, laser flash experiments with acetonitrile solutions of 2 and
6 (2:1) were carried out. Because of the inconvenient
absorption of 266 nm laser light by 6, the step-scan difference
absorption spectra are significantly perturbed by thermal
artifacts. Therefore, the millisecond rapid-scan time-resolved
technique was the only alternative left to circumvent these
experimental issues (a rapid-scan experiment requires approx-
imately 16 laser shots; a step-scan experiment requires
approximately 10600 laser shots).
For further investigation of the possible participation of
elusive radical intermediates in the formation of 3, independent
step-scan experiments were carried out in the presence of
molecular oxygen. Surprisingly, the time-resolved difference
absorption spectra obtained in this way were formally identical
to those obtained in experiments under inert argon atmosphere,
which strongly suggests there are no radical intermediates on
the pathway leading to 3. From the body of these time-resolved
FTIR experiments, one can conclude that the primary
mechanistic steps leading to 3 occur on time scales shorter
than those available to the step-scan regime (<200 ns).
Before closing this section, it is necessary to clarify why 1 is
not directly observed in the nanosecond time-resolved FTIR
experiments. The distinctive IR absorption of 1, that is the C
C stretching, peaks at 1846 cm−1 in Ne matrices at 4 K.23
However, this absorption is extremely weak and therefore, as
expected, is not observed in the step-scan time-resolved FTIR
experiments. Other absorptions bands of 1 in the IR window
available in the experimental setup (1050−2800 cm−1) include
ring stretchings with or without C−H bending contributions,
which are also too weak to be detected or masked by other
product absorptions. Therefore, for finally establishing the
presence of 1 in the 266 nm photolysis of 2 in acetonitrile at
room temperature, appropriate trapping experiments are
mandatory. In the next section, the necessary trapping
experiments on 1 that give rise to detectable IR spectral
changes are described.
4. Quenching Experiments. In neat solvent, 1 dimerizes
forming biphenylene 5 (path a, Scheme 5).34,35 The IR
Scheme 5. Trapping Experiments on 1
The scaled and shifted IR spectra of 2 alone, and a mixture of
2 and 6 (2:1) in acetonitrile, are illustrated in Figure 4a. The IR
absorption band of 2 at 1710 cm−1 is clearly resolved from that
of 6 at 1765 cm−1. Figure 4b illustrates the transient difference
absorption spectrum recorded 100 ms after exciting with a 266
nm laser pulse, an acetonitrile solution containing 2, and
quencher 6. The 100 ms difference absorption spectrum shows
a photoinduced strong positive absorption band at 2343 cm−1
due to CO2 formation, indicating the successful decarboxylation
of 2 under these experimental conditions and negative
absorptions at 1710 and 1765 cm−1, indicating that both 6
and 2 undergo net photolysis. In addition, the 100 ms rapid-
scan difference absorption spectrum shows a weak positive
absorption at 1731 cm−1 (asterisk, Figure 4b). Although weak,
the 1731 cm−1 absorption is reproducible on an experiment-to-
experiment basis. For better understanding of the results of
these rapid-scan experiments, the transient spectrum at 100 ms
was simulated using the calculated IR spectra of the species
spectrum of authentic samples of 5 in acetonitrile shows
resonances at 1130, 1151, and 1261 cm−1. The characteristic IR
bands of 5 could not be clearly observed in the time-resolved
FTIR experiments, presumably due to their intrinsic weakness
(ε < 300 L mol−1cm−1), the insufficient signal-to-noise ratio, or
because they might be simply masked by other product bands.
At any rate, this channel is clearly invisible in our experiments
and cannot account for the presence of 1.
The strategy to confirm the presence of 1 requires an
efficient trapping reaction easy to follow by IR spectroscopy.
Strong nucleophiles such as N3―, SCN―, and NCO―
were considered first. These quenchers are strong IR
chromophores and are practically transparent to 266 nm
light; however, they react readily with 2 leading to the
instantaneous formation of phtaloyl anhydride. Therefore, they
had to be disqualified as trapping reagents. Alkyl azides such as
D
DOI: 10.1021/acs.joc.5b02678
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
I am currently engaged in the application of the current
protocol to asymmetric derivatives of 2 with the intention of
detecting arynes in solution directly using time-resolved IR
spectroscopy.
EXPERIMENTAL SECTION
■
General Methods. Acetonitrile was obtained from a commercial
supplier, and it was stored for at least 2 days under molecular sieves (3
Å) prior to use. Phtaloyl peroxide was prepared according to a
literature method.38 Biphenylene and methyl 1-methylpyrrole-2-
carboxylate were purchased from commercial providers and used as
supplied.
Step-Scan Time-Resolved FTIR. Our step-scan instrument has
been described elsewhere;39 therefore, only a brief description is
presented here. Stock solutions (250 mL) of phtaloyl peroxide (1−3
mmol L−1) were circulated through a CaF2 cell (0.2 mm path length)
by means of a peristaltic pump. Photolysis took place inside the probe
chamber of the step-scan spectrometer by the fourth harmonic pulses
(266 nm, 7 ns) of a Nd:YAG laser. For avoiding thermal artifacts in
the transient spectra, the energy of the 266 nm pulses was attenuated
to 7 mJ/pulse before the sample cell. For circumventing extended
irradiation times, the measurements were conducted with 6 cm−1
spectral resolution.
Figure 4. Quenching benzyne 1 with 6. (a) Black: scaled and shifted
FTIR spectrum of 2. Blue: scaled and shifted FTIR spectrum of a
solution containing 2 and 6 (2:1). (b) Difference absorption spectrum
recorded 100 ms after 266 nm laser photolysis of a sample solution
containing 2 and 6 in a molar ratio of 2:1. The asterisk marks the IR
absorption of the reaction product between 1 and 6. (c) A simulation
of the difference absorption spectrum illustrated in (b) using the DFT
theoretical IR spectra of all of the species involved (see text for
details).
DFT Calculations. DFT calculations were performed with
ORCA.40 Geometry optimizations were carried out in two stages to
obtain the best compromise between accuracy and computing time. In
the first stage, the Becke−Perdew functional BP8641,42 was used in
conjunction with the Ahlrichs-VDZ43,44 basis sets by invoking the
resolution of the identity (RI)45 approximation. In the second stage,
the geometric parameters, previously obtained, were further refined
using the hybrid B3LYP46,47 functional in conjunction with the
Ahlrichs-TZVP44 basis set by invoking the RIJCOSX48 approximation.
Solvent effects were taken into account through the conductor-like
screening model (COSMO).49,50 For characterizing the nature
(minimum or saddle point) of the optimized geometries, numerical
frequencies were computed by two-sided numerical differentiation
using step increments of 0.002 bohr.
involved in this quenching experiment (path b, Scheme 5).
Figure 4c shows the result of such a simulation (see Supporting
Information). The excellent agreement between experiment
and theory is very gratifying and indicates the quenching of 1
was finally successful.
An upper bound for the photolytic yield of 1 can be
calculated from the magnitudes of the negative absorptions of
precursor 2 (5.57 mOD) and quencher 6 (1.65 mOD) at 1708
and 1768 cm−1, respectively (Figure 4b). Now, taking into
account the extinction coefficients of 6 and 2 at the peak of the
CO stretching maxima, they are equal to 0.76 and 1.93 × 103 L
mol−1 cm−1, respectively. The desired limiting yield for the
formation of 1 is 75%. In the previous section, an 80% yield was
calculated based on DFT simulated time-resolved step-scan
FTIR spectra. It is remarkable that both calculated photolytic
yields agree. This coincidence suggests the DFT calculated IR
spectra can be employed for quantitative estimation and
simulation of time-resolved FTIR spectra. Previous synthetic
work has shown that Diels−Alder reactions with 1 are
quantitative.19 The experiments presented here confirm the
synthetic findings. Before closing this section, I would like to
mention that when methyl furan-2-carboxylate is employed as a
quencher similar results are obtained, indicating that Diels−
Alder reactions provide a convenient way to trap 1 in liquid
acetonitrile.
The electronic spectra were calculated within the framework of the
time-dependent density functional theory (TDDFT).51,52 The first 16
excited states were computed using the BP86 functional in conjunction
with the def2-aug-TZVPP53 basis sets.
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.joc.5b02678.
Cartesian coordinates and electronic energies for 1−7
and the transition structure for the transformation of 4
toward 3, details of the calculated electronic spectrum of
2 by TDDFT, brief description of the methodology to
simulate the difference time-resolved FTIR spectra,
kinetic profiles at 2140 and 2343 cm−1 corresponding
to step-scan experiments on 2 under inert argon
atmosphere, and step-scan difference absorption FTIR
spectra conducted on 2 in CO and O2 atmospheres and
with added methanol in argon-saturated acetonitrile
(PDF)
CONCLUSIONS
■
o-Benzyne 1 has found numerous synthetic applications;
however, real-time mechanistic studies on 1 in liquid solution
are scarce and restricted to low-temperature conditions. I have
confirmed that phtaloyl peroxide 2 is a suitable precursor of 1
in liquid acetonitrile solution when irradiated with 266 nm laser
light. Reactant 1 is formed after the simultaneous loss of two
CO2 molecules from 2, a process that cannot be temporally
resolved with a step-scan FTIR spectrometer. In another
competing, but less efficient, photolytic pathway, 2 ejects a
single CO2 molecule, and ketoketene 3 is formed. As expected,
3 reacts readily with methanol. The existence of 1 was
confirmed by means of its Diels−Alder reactivity with 1-
methylpyrrole-2-carboxylate 6.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: joel.torresalacan@pc.uni-bonn.de.
Notes
The authors declare no competing financial interest.
E
DOI: 10.1021/acs.joc.5b02678
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
(38) Yuan, C. X.; Axelrod, A.; Varela, M.; Danysh, L.; Siegel, D.
Tetrahedron Lett. 2011, 52, 2540.
ACKNOWLEDGMENTS
■
Financial support by the Deutsche Forschungsgemeinschaft
through the Collaborative Research Center “Chemistry at Spin
Centers” is gratefully acknowledged.
(39) Torres-Alacan, J.; Krahe, O.; Filippou, A. C.; Neese, F.;
Schwarzer, D.; Vohringer, P. Chem. - Eur. J. 2012, 18, 3043.
(40) Neese, F. ORCA-An ab Initio, Density Functional, and
Semiempirical Program Package Version 2.8; University of Bonn,
Bonn, Germany, 2011.
(41) Perdew, J. P.; Yue, W. Phys. Rev. B: Condens. Matter Mater. Phys.
1986, 33, 8800.
(42) Becke, A. D. Phys. Rev. A: At., Mol., Opt. Phys. 1988, 38, 3098.
(43) Schafer, A.; Horn, H.; Ahlrichs, R. J. Chem. Phys. 1992, 97, 2571.
(44) Eichkorn, K.; Weigend, F.; Treutler, O.; Ahlrichs, R. Theor.
Chem. Acc. 1997, 97, 119.
REFERENCES
■
(1) Bachmann, W. E.; Clarke, H. T. J. Am. Chem. Soc. 1927, 49, 2089.
(2) Wittig, G. Naturwissenschaften 1942, 30, 696.
(3) Wittig, G.; Pohmer, L. Angew. Chem. 1955, 67, 348.
(4) Roberts, J. D.; Simmons, H. E.; Carlsmith, L. A.; Vaughan, C. W.
J. Am. Chem. Soc. 1953, 75, 3290.
(5) Heaney, H. Chem. Rev. 1962, 62, 81.
(6) Wenk, H. H.; Winkler, M.; Sander, W. Angew. Chem., Int. Ed.
2003, 42, 502.
(45) Kendall, R. A.; Fruchtl, H. A. Theor. Chem. Acc. 1997, 97, 158.
(46) Lee, C. T.; Yang, W. T.; Parr, R. G. Phys. Rev. B: Condens. Matter
Mater. Phys. 1988, 37, 785.
(7) Wentrup, C. Aust. J. Chem. 2010, 63, 979.
(8) Leopold, D. G.; Miller, T. M.; Lineberger, W. C. J. Am. Chem. Soc.
1986, 108, 178.
(47) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
(48) Neese, F.; Wennmohs, F.; Hansen, A.; Becker, U. Chem. Phys.
2009, 356, 98.
(9) Wenthold, P. G.; Squires, R. R.; Lineberger, W. C. J. Am. Chem.
Soc. 1998, 120, 5279.
(49) Klamt, A.; Schuurmann, G. J. Chem. Soc., Perkin Trans. 2 1993,
799.
(50) Klamt, A. J. Phys. Chem. 1995, 99, 2224.
(51) Gross, E. K. U.; Dobson, J. F.; Petersilka, M. Density Functional
Theory II 1996, 181, 81.
(52) Dreuw, A.; Head-Gordon, M. Chem. Rev. 2005, 105, 4009.
(53) Weigend, F.; Ahlrichs, R. Phys. Chem. Chem. Phys. 2005, 7, 3297.
(10) Orendt, A. M.; Facelli, J. C.; Radziszewski, J. G.; Grant, D. M.;
Michl, J. J. Am. Chem. Soc. 1996, 118, 846.
(11) Mclain, S. J.; Schrock, R. R.; Sharp, P. R.; Churchill, M. R.;
Youngs, W. J. J. Am. Chem. Soc. 1979, 101, 263.
(12) Buchwald, S. L.; Watson, B. T.; Huffman, J. C. J. Am. Chem. Soc.
1986, 108, 7411.
(13) Buchwald, S. L.; Nielsen, R. B. Chem. Rev. 1988, 88, 1047.
(14) Stiles, M.; Miller, R. G.; Burckhardt, U. J. Am. Chem. Soc. 1963,
85, 1792.
(15) Jones, M.; Decamp, M. R. J. Org. Chem. 1971, 36, 1536.
(16) Chang, D.; Zhu, D.; Shi, L. J. Org. Chem. 2015, 80, 5928.
(17) Bennett, M. A.; Wenger, E. Chem. Ber. 1997, 130, 1029.
(18) Pellissier, H.; Santelli, M. Tetrahedron 2003, 59, 701.
(19) Kitamura, T.; Yamane, M. J. Chem. Soc., Chem. Commun. 1995,
983.
(20) Tadross, P. M.; Stoltz, B. M. Chem. Rev. 2012, 112, 3550.
(21) Medina, J. M.; Mackey, J. L.; Garg, N. K.; Houk, K. N. J. Am.
Chem. Soc. 2014, 136, 15798.
(22) Chapman, O. L.; Mattes, K.; McIntosh, C. L.; Pacansky, J.;
Calder, G. V.; Orr, G. J. Am. Chem. Soc. 1973, 95, 6134.
(23) Radziszewski, J. G.; Hess, B. A., Jr.; Zahradnik, R. J. Am. Chem.
Soc. 1992, 114, 52.
(24) Scheiner, A. C.; Schaefer, H. F.; Liu, B. J. Am. Chem. Soc. 1989,
111, 3118.
(25) Berry, R. S.; Stiles, M.; Spokes, G. N. J. Am. Chem. Soc. 1962, 84,
3570.
(26) Diau, E. W. G.; Casanova, J.; Roberts, J. D.; Zewail, A. H. Proc.
Natl. Acad. Sci. U. S. A. 2000, 97, 1376.
(27) Warmuth, R. Angew. Chem., Int. Ed. Engl. 1997, 36, 1347.
(28) Bucher, G.; Halupka, M.; Kolano, C.; Schade, O.; Sander, W.
Eur. J. Org. Chem. 2001, 2001, 545.
(29) Colley, C. S.; Grills, D. C.; Besley, N. A.; Jockusch, S.;
Matousek, P.; Parker, A. W.; Towrie, M.; Turro, N. J.; Gill, P. M. W.;
George, M. W. J. Am. Chem. Soc. 2002, 124, 14952.
(30) Torres-Alacan, J.; Vohringer, P. Int. Rev. Phys. Chem. 2014, 33,
521.
(31) Chapman, O. L.; McIntosh, C. L.; Pacansky, J.; Calder, G. V.;
Orr, G. J. Am. Chem. Soc. 1973, 95, 4061.
(32) Koch, R.; Blanch, R. J.; Wentrup, C. J. Org. Chem. 2014, 79,
6978.
(33) Liu, R. C. -Y.; Lusztyk, J.; McAllister, M. A.; Tidwell, T. T.;
Wagner, B. D. J. Am. Chem. Soc. 1998, 120, 6247.
(34) Logullo, F. M.; Seitz, A. H.; Friedman, L. Org. Synth. 1968, 48,
12.
(35) Dvorak, V.; Kolc, J.; Michl, J. Tetrahedron Lett. 1972, 13, 3443.
(36) Alvarez, S. G.; Alvarez, M. T. Synthesis 1997, 1997, 413.
(37) Simon, J. G. G.; Munzel, N.; Schweig, A. Chem. Phys. Lett. 1990,
170, 187.
F
DOI: 10.1021/acs.joc.5b02678
J. Org. Chem. XXXX, XXX, XXX−XXX