Time-resolved and mechanistic study of the photochemical uncaging reaction of the o-hydroxycinnamic caged compound

Article abstract of DOI:10.1021/jp403323h

The o-hydroxycinnamic derivatives represent efficient caged compounds that can realize quantification of delivery upon uncaging, but there has been lack of time-resolved and mechanistic studies. We used time-resolved infrared (TRIR) spectroscopy to invest

Full text of DOI:10.1021/jp403323h

Article
pubs.acs.org/JPCA
Time-Resolved and Mechanistic Study of the Photochemical
Uncaging Reaction of the o‑Hydroxycinnamic Caged Compound
Youqing Yu, Lidan Wu, Xiaoran Zou, Xiaojuan Dai, Kunhui Liu, and Hongmei Su*
Beijing National Laboratory for Molecular Sciences (BNLMS), State Key Laboratory of Molecular Reaction Dynamics, Institute of
Chemistry, Chinese Academy of Sciences, Beijing 100190, China
*
S Supporting Information
ABSTRACT: The o-hydroxycinnamic derivatives represent
efficient caged compounds that can realize quantification of
delivery upon uncaging, but there has been lack of time-
resolved and mechanistic studies. We used time-resolved
infrared (TRIR) spectroscopy to investigate the photochemical
uncaging dynamics of the prototype o-hydroxycinnamic
compound, (E)-3-(2-hydroxyphenyl)-acrylic acid ethyl ester
(
HAAEE), leading to coumarin and ethanol upon uncaging.
Taking advantage of the specific vibrational marker bands and the IR discerning capability, we have identified and distinguished
two key intermediate species, the cis-isomers of HAAEE and the tetrahedral intermediate, in the transient infrared spectra, thus
providing clear spectral evidence to support the intramolecular nucleophilic addition mechanism following the trans−cis
photoisomerization. Moreover, the product yields of coumarin upon uncaging were observed to be greatly affected by the solvent
polarity, suppressed in CH Cl but enhanced in D O/CH CN with the increasing volume ratio of D O. The highly solvent-
2
2
2
3
2
dependent behavior indicates E1 elimination of the tetrahedral intermediate to give rise to the final uncaging product coumarin.
6
−1
The photorelease rate of coumarin was directly characterized from TRIR (3.6 × 10 s ), revealing the promising application of
such o-hydroxycinnamic compound in producing fast alcohol jumps. The TRIR results provide the first time-resolved detection
and thus offer direct dynamical information about this photochemical uncaging reaction.
1
. INTRODUCTION
achieved by analyzing the increase of fluorescence emission
after or during uncaging at the targeted site. Moreover, the o-
hydroxycinnamic group has sufficient sensitivity to both one-
Photolabile protecting groups and caged compounds have been
widely used in chemical and biological studies.
1
−4
Their
11,12
photon excitation (1PE) and two-photon excitation (2PE),
while chromophores that can act as a 2PE photoremovable
attractiveness comes from the possibility of controlling both
the location and the time of release of active chemicals by
exposure to light alone. A large variety of such protecting
groups have been developed, each of them tailored for specific
uses. Among them, developing phototriggers that can realize
quantification of delivery upon uncaging is of particular
importance when a critical molecule concentration is required
to induce a specific activity.
1
4,15
protecting group for biological use are rare.
Attributed to
these advantages, o-hydroxycinnamic platform is particularly
attractive in tailoring photoliable protecting groups that exhibit
large two-photon uncaging cross-sections and that release
strongly fluorescent coproducts upon uncaging, and thus holds
practical potential as an efficient cage for the liberation of
5
−8
9
−13
The o-hydroxycinnamic group to cage alcohols and amines
various biological stimulants.
9
−13
has received strong interest in this regard.
backbones in the caged precursor intrinsically do not fluoresce,
while the photoreleased coumarin coproduct is fluorescent
The cinnamate
For rational design of novel caging groups using o-
hydroxycinnamic derivatives, it is important to understand
the detailed mechanism concerning how the photochemical
uncaging reaction occurs. The products and conditions for the
o-hydroxycinnamic uncaging have been well established in
(
Scheme 1) and can serve as a fluorescent reporter.
Quantification of delivery upon uncaging could be then simply
9
−13
previous work.
In addition, an intramolecular nucleophilic
Scheme 1. Photochemical Uncaging Reaction of (E)-o-
Hydroxycinnamates to Coumarin and Alcohol
addition mechanism was proposed to account for the uncaging
process (Scheme 2), which involves the trans−cis photo-
isomerization followed by the nucleophilic addition of the o-
hydroxy group onto the ethyl ester moiety forming the
tetrahedral intermediate, prior to releasing ethanol and the
Received: October 11, 2012
Revised: July 12, 2013
Published: July 19, 2013
©
2013 American Chemical Society
7767
dx.doi.org/10.1021/jp403323h | J. Phys. Chem. A 2013, 117, 7767−7775

The Journal of Physical Chemistry A
Article
Scheme 2. Proposed Reaction Mechanism for the Photochemical Uncaging of HAAEE
coumarin coproduct. However, there is some uncertainty of
discerning tetrahedral intermediate from the cis-isomers in the
2. EXPERIMENTAL AND COMPUTATIONAL METHODS
.1. Experiments. 2.1.1. Materials. (E)-3-(2-Hydroxy-
phenyl)-acrylic acid ethyl ester (HAAEE) (>96%) and D O
2
1
11,16
H NMR spectra,
for both species have an identical C7
2
C8 double bond with (Z) stereochemistry. Unambiguous
identification of the reaction intermediates is thus desirable to
establish the uncaging mechanism. Meanwhile, the mechanisms
concerning the releasing of coumarin from tetrahedral
intermediate are required to be explored further.
(
>99%) was purchased from Sigma Aldrich and used as
received. Solvents of chromatographic grade (CH CN and
3
CH Cl ) were used.
2
2
2
.1.2. Time-Resolved Fourier Transform IR (TR-FTIR)
Experiment. Detailed experimental procedures for step-scan,
Another important issue that is critical to the use of the caged
time-resolved Fourier transform IR (TR-FTIR) absorption
2
1,22
compound is the rate of the photolytic release of the substrate
spectroscopy
have been described in our previous
8
23
from these inert precursors. Nevertheless, it has been lack of
publications. Briefly, the TR-FTIR instrument (setup shown
time-resolved measurements of the photorelease rate for the o-
hydroxycinnamic caged compounds. Transient UV−visible
absorption measurements could be a sensitive probe for this
in Figure 1) comprises a Nicolet Nexus 870 step-scan FTIR
1
7−19
rate.
However, this method suffers from an inability to
distinguish the photoreleased coumarin coproduct from its
cinnamate precursor because the two absorb UV−visible light
11
in the same wavelength range. Alternatively, the step-scan
8
,20
FTIR spectroscopy in the nanosecond time domain
offers a
direct, noninferential time-resolved method to determine the
rate of release where the photoreleased product exhibits specific
vibrational signatures. A time-resolved infrared (TRIR)
investigation using the vibrational marker bands would
therefore allow for the direct characterization of the photo-
release rate of the uncaged products.
In these regards, we report here a TRIR investigation on the
photochemical uncaging dynamics of the o-hydroxycinnamic
caging group represented by the prototype o-hydroxycinnamic
caged compound, the (E)-3-(2-hydroxyphenyl)-acrylic acid
ethyl ester (denoted as HAAEE in Scheme 1) leading to the
release of coumarin and ethanol upon uncaging. TRIR
experiments were performed to directly probe and determine
formation dynamics of the uncaged products from photoexcited
Figure 1. Schematic representation of the TR-FTIR experimental
setup.
spectrometer, Continuum Surelite II Nd YAG laser, and a pulse
generator (Stanford Research DG535) to initiate the laser pulse
and achieve synchronization of the laser with data collection,
two digitizers (internal 100kHz 16-bit digitizer and external 100
MHz 14-bit GAGE 14100 digitizer), which offer fast time
resolution and a wide dynamic range as needed, and a personal
computer to control the whole experiment. The detector used
in this work is the photovoltaic MCT (0.5 mm) equipped with
a fast internal preamplifier (50 MHz).
There are two outputs from the detector: output DC,
corresponding to the value of the static interferogram; and
output AC, corresponding to the time-resolved change of the
interferogram. The AC signal was then amplified by an external
preamplifier (Stanford Research, SRS560). The differential
absorbance spectra are calculated based on the equation where
HAAEE in D O/CH CN mixed solvents with varying water
2
3
concentration as well as in other typical solvents such as
CH Cl . It is expected that comparison of the real time reaction
2
2
dynamics in different solvents can provide important evidence
to help elucidate the nature of the photorelease mechanism of
the o-hydroxycinnamic caged compound. In addition, the key
transient intermediates can be identified in the TRIR spectra by
their characteristic vibrational marker bands, thus providing
spectral evidence to establish the reaction mechanism. To the
best of our knowledge, the TRIR results presented here provide
the first direct time-resolved detection of the photochemical
uncaging reaction for o-hydroxycinnamic phototrigger com-
pounds.
7
768
dx.doi.org/10.1021/jp403323h | J. Phys. Chem. A 2013, 117, 7767−7775

The Journal of Physical Chemistry A
Article
IDC and ΔI are the single-beam intensity spectra correspond-
AC
ing to static (DC) and dynamic (AC) channels. ΔI is
AC
calibrated before being used in equation because different gain
21,22
is applied to the AC channel.
The fourth harmonic of Nd YAG laser (266 nm) operating at
0 Hz repetition rate was used in the experiments. The laser
1
excitation beam was directed through an iris aperture (3 mm in
diameter) and then overlapped with the infrared beam in the
sample cell within the sample compartment of the FTIR
spectrometer. The laser beam energy after the aperture was 1
mJ per pulse. The IR spectra were collected with a spectral
−1
resolution of 8 cm . A Harrick flowing solution cell with 2 mm
thick CaF windows (path-length, 100 μm) was used for the
2
measurements. The closed flowing system is driven by a
peristaltic pump (ColeParmer Masterflex) to refresh the sample
before every laser pulse.
2
.1.3. Steady-State UV−Vis Absorption Experiment.
Absorption spectra were recorded with a UV−vis spectrometer
Model U-3010, Hitachi) in a 10 mm path length quartz
cuvette.
(
2.2. Computational Methods. We have calculated the
potential energy profile for the possible intramolecular
nucleophilic addition mechanism of HAAEE toward releasing
ethanol and coumarin. The geometries of the reactants,
products, intermediates, and transition states along the reaction
route are optimized using the hybrid density functional theory,
i.e., Becke’s three-parameter nonlocal exchange functional with
the nonlocal correlation functional of Lee, Yang, and Parr
Figure 2. (a) Steady-state IR spectrum of 8 mM HAAEE in
acetonitrile, (b) transient IR spectra of 8 mM HAAEE in acetonitrile
at selected time delays following 266 nm laser irradiation, and (c) the
steady-state difference spectrum of 8 mM HAAEE in acetonitrile after
one minute of 266 nm laser irradiation.
2
4,25
(
B3LYP) with the 6-311++G(d, p) basis sets.
Harmonic
vibrational frequencies, relative energies, and the zero-point
energies (ZPE) are calculated at the same level with the
optimized geometries. The intermediates are characterized by
all the real frequencies. The transition states are confirmed by
only one imaginary frequency. Connections of the transition
states between two local minima have been confirmed by
intrinsic reaction coordinate (IRC) calculations at the same
−1
bands were observed at 1604, 1635, and 1712 cm . To aid
with spectral assignment, IR frequencies and IR intensities were
calculated at the B3LYP/6-311++G(d,p) level for the possible
relevant species in the photochemical reaction of HAAEE, both
in vacuum and in different solvents with the solvation effect
simulated by PCM model (Table 1). The calculated IR
2
6
level. Bulk solvation effects are considered using the polarized
27,28
continuum model (PCM).
Geometry optimizations of ground-state HAAEE, cis-isomers
of HAAEE, tetrahedral intermediate, coumarin, and triplet-state
HAAEE were performed at the B3LYP/6-311++G(d,p) level of
theory. Tight optimization criteria were used in all calculations.
All stationary points were confirmed to be energy minima using
vibrational frequency calculations (B3LYP/6-311++G(d,p)),
which confirmed that all of the computed vibrational
frequencies were real (i.e., no imaginary vibrational frequen-
cies). To aid with spectral assignments, vibrational frequencies
were calculated in both the vacuum and in solvents under the
polarized continuum model. Using the ground-state stationary
structures, the vertical electronic excitations were calculated
with time-dependent DFT at the TD-B3LYP/6-311++G(d,p)
level of theory. Difference density plots were generated from
the computed electron densities of the ground and excited
states (TD-B3LYP/6-311++G(d,p)). All of the theoretical
calculations were performed with the Gaussian03 program
−1
frequencies of 1610, 1637, and 1693 cm for the stable
molecule HAAEE in CH CN are in good agreement with the
3
experimentally observed frequencies at 1604, 1635, and 1712
−1
cm , respectively, indicating that the current level of
calculation can predict reliable IR frequencies for spectral
assignment purpose, particularly for the obtained frequencies
under PCM model.
These bands, which arise from double-bond stretches
associated with the benzene groups and the carbonyl stretching
modes of HAAEE, bleached strongly after one minute of UV
exposure as shown in the difference spectra (Figure 2c).
−1
Meanwhile, a positive band at 1728 cm can be identified,
which is the characteristic CO stretching vibration of the
photolyzed coumarin product. To confirm this assignment, the
IR spectrum of the authentic sample of coumarin in CH CN
3
was recorded (Figure S1, Supporting Information) and found
−1
to show a strong CO absorption at 1733 cm and a weak
2
9
package.
−1
absorption at 1608 cm for the CC stretch of benzene
group. The spectral positions of coumarin match those
3. RESULTS AND DISCUSSION
observed in the IR difference spectra (Figure 2c) after UV
−
1
Steady-state infrared absorption spectra of HAAEE in CH CN
irradiation, which exhibits a strong positive band at 1728 cm
3
−1
were recorded before and after UV irradiation at 266 nm, in
order to locate IR marker bands that are indicative of the
photoreleased uncaging product. In the spectrum obtained
before UV irradiation (red curve in Figure 2a), three strong
and a discernible weak positive band at 1612 cm . The small
peak shift of 4 or 5 cm− in the IR difference spectra after UV
irradiation is resulted from the partial overlapping of the
positive coumarin band with the negative band tail of HAAEE.
1
7
769
dx.doi.org/10.1021/jp403323h | J. Phys. Chem. A 2013, 117, 7767−7775

The Journal of Physical Chemistry A
Article
−1
Table 1. B3LYP/6-311++G(d,p) Calculated IR Frequencies (cm ) and IR Intensities for the Possible Relevant Species in the
Photochemical Reaction of HAAEE, Both in Vacuum and in Different Solvents with the Solvation Effect Simulated by PCM
a
Model
31
aThe computed IR intensity in the unit of km mol−1 actually denotes the integrated IR absorption coefficient. Only the vibrational modes with
considerable IR intensities are listed here.
The band shapes for the CO absorption are also similar, with
Figure 2b displays the TRIR difference spectra of HAAEE in
a shoulder at the right-hand side, due to the pronounced
CH CN at typical delay times following 266 nm photolysis.
3
30a
anharmonicity of the CO stretching vibration in coumarin.
The difference spectra represent the difference between the
spectra obtained after photolysis and the spectrum before
photolysis. The depletion of reactant gives rise to negative
signals, and the formation of transient intermediates or
products leads to positive bands. Within the first 0.7 μs, a
prominent positive band at 1728 cm− is gradually formed, with
the negative bands of the HAAEE bleaching at 1635 and 1712
These results, together with the reported vibrational band of
−
1
30b
coumarin at 1722 cm in methylcyclohexane
and the
−1
theoretical prediction of 1729 cm in CH CN (Table 1), the
3
1
728 cm− band observed in the IR difference spectra can be
1
1
confidently assigned to coumarin. Obviously, the band at 1728
−1
cm for the photolyzed product coumarin can be distinguished
from its caged precursor HAAEE in the IR spectra, which
makes it an excellent probe to study the photochemical
uncaging reaction dynamics in the time-resolved experiments.
−1
cm . After reaching its maximum intensity at 0.7 μs, the
transient positive band at 1728 cm− does not decay (Figure 3),
indicating its identity as a stable photoproduct. This band is a
1
7
770
dx.doi.org/10.1021/jp403323h | J. Phys. Chem. A 2013, 117, 7767−7775

The Journal of Physical Chemistry A
Article
S (ππ*) states, on the basis of the TD-DFT calculated vertical
3
−1
excitation energies of 95.2 and 106.5 kcal mol (shown in
Table 2) along with their respective oscillator strengths of 0.27
Table 2. TD-B3LYP/6-311++G(d,p) Calculated Vertical
Excitation Energies, Oscillator Strengths (f), and the
Dominant Occupied and Virtual Orbitals Contributing to
the Three Lowest Energy Singlet Excitations of HAAEE
a
character
energy
(eV)
energy
(nm)
oscillator
strength
state
(% contribution)
Figure 3. Kinetics traces and the single-exponential fitting curves for
S1
S2
S3
51 → 52 (62%)
49 → 52 (69%)
50 → 52 (56%)
4.14
4.37
4.63
300
283
268
0.27
0
−1
the coumarin band at 1728 cm and the HAAEE bleaching band at
−
1
1
712 cm .
0.44
a
Orbital 51 is the highest occupied molecular orbital.
characteristic vibrational marker band of coumarin, according to
its spectral position and band shape. In fact, the excellent
agreement between the TRIR spectra and the steady-state IR
difference spectrum confirms further that the 1728 cm−1 band
in the TRIR spectra is ascribed to the stable photoproduct,
coumarin, but not transient species such as triplet state of
HAAEE or the proposed intermediates of cis-isomer and
tetrahedral intermediate. The tetrahedral intermediate simply
has no absorption in this spectral range, and the only IR band
with considerable intensity for the tetrahedral intermediate is
and 0.44. Thus, the excitation wavelength used in this work,
66 nm, corresponds to the strongly dipole allowed transition
2
32
to S (ππ*) state. According to Kasha’s rule, the subsequent
3
photochemical or photophysical processes are favorable to take
place at the lowest excited state of a given multiplicity no
matter which excited state is initially populated. So the highly
excited S state is assumed to rapidly internally convert to the
3
lowest lying singlet state S (ππ*), from which the photo-
1
chemical uncaging of HAAEE is initiated.
The calculated electron density difference plots (Figure 4)
−1
anticipated to be present at 1139 cm as shown in Table 1.
The expected two pairs of characteristic absorptions for the cis-
isomer intermediates, as will be discussed later in the spectra of
HAAEE in CH Cl , are at the noise level here, indicating
suggested that S state corresponds to the ππ* excitation of the
1
2
2
negligible amount of cis-isomers present in the spectra of
HAAEE photolyzed in CH CN.
3
In addition, all of the transient features observed in the TRIR
spectra are shown not be affected by whether the solution is in
air condition, oxygen-saturated, or deaerated with nitrogen.
This observation suggests that the photochemical uncaging
reaction occurs from a HAAEE singlet excited state, and there is
no involvement of the triplet state.
Figure 4. Electron density difference plots for the S , S , and S states
1
2
3
of HAAEE calculated at the TD-B3LYP/6-311++G(d,p) level of
theory. Isovalue: ±0.0004. The green contours represent the depletion
of electron density in the ground-state, while the red contours
represent the accumulation of electron density in the specific excited
state.
To investigate the rate of release of coumarin due to
photolysis of the caged HAAEE, the time evolution of the
−1
intensity changes for the 1728 cm band was examined
shown in Figure 3 as open triangles). The increase in the
(
−1
intensity of the marker band of coumarin at 1728 cm as a
function of time is well represented by a single exponential
increase, with a rate constant of 3.6 × 10 s , corresponding to
the photorelease rate of coumarin from HAAEE. Concom-
conjugated benzene and C7C8 moiety such that there is a
change in the electron density along the conjugated moiety.
6
−1
The ππ* excitation of S state is expected to result in a
1
lengthening of the CC double bond and contractions of CC
single bonds, a feature that has been proved to allow the trans−
cis isomerization to proceed for such kind of ethylenic
−1
itantly, the bleaching band of HAAEE at 1712 cm follows an
almost identical rate of depletion (shown in Figure 3 as open
squares), with a rate constant of 3.7 × 10 s obtained from
the single exponential fitting. Here, the TRIR experiment
measures directly the dynamic process of product formation
after one laser pulse and thus determines the microscopic
uncaging rate noninferentially. It differs from previous steady-
state experiments also in the aspect that the caged precursor
monitored here is the simplest o-hydroxycinnamic compound
without functional group substitution in the benzene ring and
the double bond (Scheme 1).
6
−1
33
systems. This corroborates also the observation that there is
no triplet state involved in the photochemical reaction of
HAAEE.
The trans−cis photoisomerization has been proposed to play
key roles in initiating the photochemical uncaging reactions of
1
1
9,11
the o-hydroxycinnamic series of compounds.
The ground
state cis-isomers formed after the trans−cis photoisomerization
was proposed to be followed by an intramolecular nucleophilic
addition of the o-hydroxy group onto the ethyl ester moiety,
34,35
It shows here in the TRIR spectra that the photochemical
uncaging of HAAEE releases coumarin and ethanol rapidly and
that the reaction is associated to a singlet excited state. How
does the reaction take place? HAAEE represents the typical
caged compound containing the o-hydroxycinnamic chromo-
phore. It exhibits two absorption bands peaked at 320 and 270
nm (shown in Figure S2, Supporting Information), which can
leading to a tetrahedral intermediate.
The intermediate
then eliminated ethanol and released the coumarin coproduct,
completing the uncaging process. Efforts have been devoted to
1
11,16
identify key intermediates with H NMR spectroscopy,
and
an intermediate containing a double bond with a (Z)
stereochemistry has been located. However, the identity of
1
this long-lived intermediate captured with H NMR remains
be assigned, respectively, to the excitation to the S (ππ*) and
unclear because the tetrahedral intermediate and the cis-isomer
1
7
771
dx.doi.org/10.1021/jp403323h | J. Phys. Chem. A 2013, 117, 7767−7775

The Journal of Physical Chemistry A
Article
intermediates both have identical C7C8 bond with a (Z)
these positions. According to the spectral position, the four
positive bands newly observed in CH Cl are ascribed to the
stereochemistry, which could not be easily distinguished from
2
2
1
each other with H NMR spectroscopy. So we attempted to
cis-isomers of HAAEE formed after the trans−cis photo-
isomerization. Our B3LYP/6-311++G(d,p) calculations (Table
1) show that the cis-isomer of HAAEE has two conformations,
cis-1 and cis-2. The calculated vibrational frequencies of C7
examine this issue from the TRIR spectra, taking advantage of
the IR discerning capability for these two intermediate species.
As shown in Table 1 for the predicted vibrational frequencies,
the tetrahedral intermediate possesses a vibrational marker
−
1
C8 and C9O10 stretching modes are 1678 and 1736 cm
for cis-1 and 1637 and 1680 cm− for cis-2, in agreement with
the four positive bands in Figure 5b,c and thus allow the
assignment of the observed bands at 1658 and 1735 cm−
1
−1
band at 1137 or 1139 cm (in CH CN or CH Cl ) different
3
2
2
from those of the cis-isomer intermediates, though with a weak
IR intensity.
1
(
labeled # in Figure 5) to cis-1 and the bands at 1620 and 1689
In the polar solvent of CH CN, the cis-isomers and the
3
−1
cm (labeled * in Figure 5) to cis-2. The identification of the
cis-isomers in TRIR spectra confirms their existence and
validates the key roles of the trans−cis photoisomerization in
initiating the photochemical reaction.
tetrahedral intermediate are subject to fast further uncaging
6
−1
reactions, releasing coumarin with a rate of 3.6 × 10 s as has
been detected in Figure 3. So the intermediates can not be
accumulated with a detectable concentration and can not be
easily observed experimentally in this case. In fact, a close
However, the prominent coumarin band at 1728 cm−
1
observed in CH CN is not present in the spectra collected
3
inspection of the TRIR spectra for the photolysis in CH CN
3
for the photolysis in CH Cl , showing that in the less polar
2
2
reveals no discernible IR band ascribed to the tetrahedral
−1
solvent CH Cl , the trans−cis photoisomerization is not
2 2
intermediate in the spectral range around 1137 cm , and the
expected characteristic absorptions for the cis-isomer inter-
mediates are at the noise level. Only the end product coumarin
followed by a further release of the final uncaging products of
coumarin and ethanol as in the case of CH CN. Obviously, the
3
uncaging reaction subsequent to the trans−cis photoisomeriza-
is observed in the TRIR spectra for the photolysis in CH CN.
3
tion is greatly suppressed in less polar solvents of CH Cl .
2
2
When the photolysis of HAAEE was performed in the apolar
Meanwhile, the proposed tetrahedral intermediate is captured
in the TRIR spectra for the HAAEE photolysis in CH Cl . As
solvent CH Cl , distinctive results were obtained. In CH Cl ,
2
2
2
2
2
2
the photolysis of HAAEE leads to three ground state bleaching
can be seen in Figure 6a, a close inspection of the TRIR spectra
−1
bands at 1603, 1635, and 1712 cm together with formation of
four positive bands at 1620, 1658, 1689, and 1735 cm− in the
TRIR spectra (Figure 5b). The positive bands are ascribed to
stable products but not transient species since the steady-state
difference spectra (Figure 5c) also reveal identical bands at
−1
in the range of 1100−1240 cm reveals two positive bands
1
along with a negative band. With the ground state depletion
−1
band of HAAEE at 1180 cm , two positive bands are present.
−1
The one at 1214 cm is in excellent agreement with the
−1
calculated vibrational frequency for cis-2 (1219 cm ) and thus
is assigned to the cis-2, which originates from trans−cis
photoisomerization of HAAEE. Same as the other two stronger
−1
−1
bands at 1620 and 1689 cm , this weak band at 1214 cm for
the cis-2 isomer does not decay after reaching its maximum
intensity at 0.7 μs, exhibiting the characteristic as a stable
product. Noticeably, the other weak positive band at 1141 cm−
1
agrees very well with the predicted vibrational frequency of
1
1
139 cm− for the tetrahedral intermediate. In addition, the
−1
band at 1141 cm is short-lived. The decay rate of the 1141
−
1
5
−1
cm band is obtained to be 3.0 × 10
s
from the
biexponential fitting for its kinetic trace, as shown in Figure
1
6
b. So the 1141 cm− band should be ascribed to a transient
unstable intermediate, most likely, the tetrahedral intermediate,
according to its IR spectral position. This assignment is in line
11
with the previous assumption that the tetrahedral inter-
mediate is prone to afford cis-isomers by backward reaction and
can be further supported by our calculated results that cis-1 and
cis-2 are 9.5 and 14.1 kcal mol− more stable than the
tetrahedral intermediate, respectively (Figure S3, Supporting
Information). To the best of our knowledge, this is the first
direct spectroscopic evidence revealing the existence of the
tetrahedral intermediate, providing support for the previously
1
9
−13
proposed intramolecular nucleophilic addition mechanism.
Here, in the apolar solvent of CH Cl , the subsequent reaction
2
2
of the tetrahedral intermediate releasing coumarin is obviously
prohibited, so the tetrahedral intermediate can be detected in
the TRIR spectra, unlike the case in the polar solvent of
CH CN. Because of its instability, the tetrahedral intermediate
3
Figure 5. (a) Steady-state IR spectrum of 8 mM HAAEE in CH Cl ,
2
2
decays to form the more stable cis-isomers by backward
reactions, eventually, if not undergoing the forward uncaging of
coumarin, as indicated in the calculated energetics diagram
(Figure S3, Supporting Information). Therefore, for the
(
b) transient IR spectra of 8 mM HAAEE in CH Cl at selected time
2 2
delays following 266 nm laser irradiation, and (c) steady-state
difference spectrum of 8 mM HAAEE in CH Cl after one minute
2
2
of 266 nm laser irradiation.
7
772
dx.doi.org/10.1021/jp403323h | J. Phys. Chem. A 2013, 117, 7767−7775

The Journal of Physical Chemistry A
Article
Figure 6. (a) Transient IR spectra of 8 mM HAAEE in CH Cl at selected time delays following 266 nm laser irradiation, in the spectral range of
2
2
−
1
−1
1
100−1240 cm . The overscaled signal above 1220 cm is due to the solvent absorption. (b) Kinetics traces and the biexponential fitting curves for
−1
the tetrahedral intermediate band at 1141 cm .
photolysis in CH Cl when uncaging of coumarin is prohibited,
2
2
the tetrahedral intermediate are captured as transient unstable
intermediate and the cis-isomers are observed as stable
photoproducts, whereas no coumarin are detected in the
TRIR spectra.
What is the mechanism for the tetrahedral intermediate to
release coumarin? It shows in the TRIR experiments that the
coumarin yields are greatly affected by the solvent polarity,
suppressed greatly in less polar solvents of CH Cl (Figure 5)
2
2
compared to those in the polar solvent of CH CN (Figure 2).
3
Figure 7. (a) Transient IR spectra of 8 mM HAAEE in D O and
2
Such a solvent polarity-dependent behavior indicates that the
tetrahedral intermediate most likely undergoes an E1
elimination mechanism, to give rise to the final uncaging
product, coumarin. In the E1 elimination, a complex of charge
separated ion pair could be produced and serve as an
intermediate to produce the final uncaging products (shown
in Scheme 2). It is known that the E1 elimination involving
charge separation and the intermediate formation of a
carbocation could be facilitated in polar solvents, due to the
CH CN mixed solvent (VD2O/Vtotal = 0.1) at selected time delays
3
following 266 nm laser irradiation; the corresponding steady-state
difference spectrum is shown in Figure S5, Supporting Information;
(b) reaction rates of coumarin formation plotted as a function of water
concentration.
interesting to note that, for solvents with water content larger
than 20%, the increase of the reaction rates becomes much
more modest upon further addition of water to the solvent. The
measured reaction rates follow a curved dependence on water
concentration (Figure 7b). This indicates the predominant role
of the solvent water molecules in the inner solvation shell to
promote the reaction at large water concentrations.
36
strong effect of solvation on the reaction energetics, but in
apolar solvent such as CH Cl , the E1 elimination is greatly
2
2
prohibited. The E1 elimination mechanism can explain the
observed suppression of coumarin yields in CH Cl compared
2
2
to those in CH CN.
To gain further evidence supporting the E1 elimination
mechanism of the tetrahedral intermediate, the photolysis of
3
4. CONCLUSIONS
In this work, we have performed time-resolved infrared
detection of the photochemical uncaging reaction for the o-
hydroxycinnamic phototrigger compound HAAEE. By taking
advantage of the specific vibrational marker bands and the IR
discerning capability, we have identified and distinguished two
key intermediate species, the cis-isomers of HAAEE and the
tetrahedral intermediate, in the transient infrared spectra, thus
providing clear spectral evidence to support the previously
proposed intramolecular nucleophilic addition mechanism
following the trans−cis photoisomerization. Moreover, by
detecting the photolysis in different solvents, we observed
that the product yield of coumarin upon uncaging is greatly
affected by the solvent polarity, suppressed in CH Cl but
HAAEE was performed in the mixed solvents of D O/CH CN
2
3
with varying water concentration (Figure S4−S7, Supporting
Information). The good ionizing and dissociating capability
makes water an ideal solvent for the efficient charge separation
36
to occur and thus facilitates the E1 elimination. Together with
its larger polarity, the addition of water to the solvent is
expected to lead to increased reaction rates for the E1
elimination. Indeed, with the addition of 10% D O in the mixed
2
D O/CH CN solvent (Figure 7a), the amplitude of the
2
3
−1
coumarin band at 1728 cm relative to the HAAEE bleaching
band at 1712 cm− greatly increases compared to that in pure
1
CH CN, indicating an increased uncaging product yields at a
3
2
2
certain delay time and thus an increased reaction rate. Here, the
enhanced in D O/CH CN with the increasing volume ratio of
2 3
−1
positive band at 1728 cm is partially overlapped with the
D O. The highly solvent-dependent behavior indicates an E1
2
−1
bleaching band at 1712 cm . The real band intensities are
therefore obtained by Lorentzian fitting as shown in Figure S8−
S9, Supporting Information. The reaction rates can be obtained
from the time evolution of the band intensities, by the method
of single-exponential fitting as in Figure 3.
elimination mechanism of the tetrahedral intermediate to give
rise to the final uncaging product coumarin.
Additionally, we show here that photoexcitation of HAAEE
gives rise to good yields of ethanol with coumarin coproduct in
polar solvents of CH CN or D O/CH CN rapidly (rate
3
2
3
6
−1
The reaction rates of coumarin formation were measured at
several water concentrations and are found to increase with the
measured to be 3.6 × 10 s by the direct, noninferential TRIR
method). It is thus anticipated that the o-hydroxycinnamic
caged compounds such as HAAEE hold promise for use in
volume ratio of D O in the mixed solvents (Figure 7b). It is
2
7
773
dx.doi.org/10.1021/jp403323h | J. Phys. Chem. A 2013, 117, 7767−7775

The Journal of Physical Chemistry A
Article
producing fast and robust alcohol jumps with fluorescence
reporting. The dependence of the product yields and reaction
rates on the solvent polarity would implicate that the quantity
of uncaged alcohols could be modulated by the polarity of the
local environment of biological cells or tissues, which should
also be noteworthy in biological applications.
(12) Gagey, N.; Neveu, P.; Jullien, L. Two-Photon Uncaging with the
Efficient 3,5-Dibromo-2,4-Dihydroxycinnamic Caging Group. Angew.
Chem., Int. Ed. 2007, 46, 2467−2469.
(13) Gagey, N.; Emond, M.; Neveu, P.; Benbrahim, C.; Goetz, B.;
Aujard, I.; Baudin, J.-B.; Jullien, L. Alcohol Uncaging with
Fluorescence Reporting: Evaluation of o-Acetoxyphenyl Methylox-
azolone Precursors. Org. Lett. 2008, 10, 2341−2344.
(14) Goppert, M. The Probability of Two Photons Cooperating in
ASSOCIATED CONTENT
Supporting Information
■
One Elementary Act. Naturwissenschaften 1929, 17, 932−932.
*
S
(15) Goppert-Mayer, M. Elementary File with Two Quantum
Additional computational and experimental results. This
material is available free of charge via the Internet at http://
pubs.acs.org.
Fissures. Ann. Phys. 1931, 9, 273−294.
(16) Duan, X. Y.; Zhai, B. C.; Song, Q. H. Water-Soluble o-
Hydroxycinnamate as an Efficient Photoremovable Protecting Group
of Alcohols with Fluorescence Reporting. Photochem. Photobiol. Sci.
2012, 11, 593−598.
AUTHOR INFORMATION
Corresponding Author
(H.S.) E-mail: hongmei@iccas.ac.cn.
■
(
17) Lee, Y. P. State-Resolved Dynamics of Photofragmentation.
Annu. Rev. Phys. Chem. 2003, 54, 215−244.
18) Conrad, P. G.; Givens, R. S.; Hellrung, B.; Rajesh, C. S.;
*
(
Notes
Ramseier, M.; Wirz, J. p-Hydroxyphenacyl Phototriggers: The Reactive
Excited State of Phosphate Photorelease. J. Am. Chem. Soc. 2000, 122,
The authors declare no competing financial interest.
9
346−9347.
ACKNOWLEDGMENTS
(19) Hess, G. P.; Grewer, C. Development and Application of Caged
Ligands for Neurotransmitter Receptors in Transient Kinetic and
■
This work is financially supported by the National Natural
Science Foundation of China (Grant No. 21073201) and the
National Basic Research Program of China (No.
Neuronal Circuit Mapping Studies. Methods Enzymol. 1998, 291, 443−
4
73.
(
20) Gee, K. R.; Carpenter, B. K.; Hess, G. P. Synthesis,
2
013CB834602).
Photochemistry, and Biological Characterization of Photolabile
Protecting Groups for Carboxylic Acids and Neurotransmitters.
Methods Enzymol. 1998, 291, 30−50.
(21) Uhmann, W.; Becker, A.; Taran, C.; Siebert, F. Time-Resolved
FT-IR Absorption-Spectroscopy Using a Step-Scan Interferometer.
Appl. Spectrosc. 1991, 45, 390−397.
(22) Drapcho, D. L.; Curbelo, R.; Jiang, E. Y.; Crocombe, R. A.;
McCarthy, W. J. Digital Signal Processing for Step-Scan Fourier
Transform Infrared Photoacoustic Spectroscopy. Appl. Spectrosc. 1997,
REFERENCES
■
(
1) Ellis-Davies, G. C. R. Caged Compounds: Photorelease
Technology for Control of Cellular Chemistry and Physiology. Nat.
Methods 2007, 4, 619−628.
2) Klan, P.; Pelliccioli, A. P.; Pospisil, T.; Wirz, J. 2,5-
Dimethylphenacyl Esters: A Photoremovable Protecting Group for
Phosphates and Sulfonic Acids. Photochem. Photobiol. Sci. 2002, 1,
20−923.
3) Mayer, G.; Heckel, A. Biologically Active Molecules with a ″Light
Switch″. Angew. Chem., Int. Ed. 2006, 45, 4900−4921.
4) Sankaranarayanan, J.; Muthukrishnan, S.; Gudmundsdottir, A. D.
(
9
(
5
(
1, 453−460.
23) Wu, W. Q.; Liu, K. H.; Yang, C. F.; Zhao, H. M.; Wang, H.; Yu,
Y. Q.; Su, H. M. Reaction Mechanisms of a Photo-Induced [1,3]
Sigmatropic Rearrangement Via a Nonadiabatic Pathway. J. Phys.
Chem. A 2009, 113, 13892−13900.
(
Photoremovable Protecting Groups Based on Photoenolization. In
Advances in Physical Organic Chemistry; Academic Press Ltd−Elsevier
Science Ltd: London, U.K., 2009; Vol. 43, pp 39−77.
(24) Becke, A. D. Density-Functional Thermochemistry 0.3. The
Role of Exact Exchange. J. Chem. Phys. 1993, 98, 5648−5652.
(
5) Ackmann, A. J.; Frechet, J. M. J. The Generation of Hydroxide
(
25) Lee, C. T.; Yang, W. T.; Parr, R. G. Development of the Colle−
and Methoxide Ions by Photo-Irradiation: Use of Aromatization to
Stabilize Ionic Photo-Products from Acridine Derivatives. Chem.
Commun. 1996, 605−606.
Salvetti Correlation-Energy Formula into a Functional of the Electron-
Density. Phys. Rev. B 1988, 37, 785−789.
(26) Gonzalez, C.; Schlegel, H. B. Reaction-Path Following in Mass-
(
6) Suzuki, A. Z.; Watanabe, T.; Kawamoto, M.; Nishiyama, K.;
Weighted Internal Coordinates. J. Phys. Chem. 1990, 94, 5523−5527.
Yamashita, H.; Ishii, M.; Iwamura, M.; Furuta, T. Coumarin-4-
Ylmethoxycarbonyls as Phototriggers for Alcohols and Phenols. Org.
Lett. 2003, 5, 4867−4870.
(27) Tomasi, J.; Persico, M. Molecular-Interactions in Solution: An
Overview of Methods Based on Continuous Distributions of the
Solvent. Chem. Rev. 1994, 94, 2027−2094.
(
7) Ma, C. S.; Kwok, W. M.; Chan, W. S.; Du, Y.; Kan, J. T. W.; Toy,
(28) Cramer, C. J.; Truhlar, D. G. Implicit Solvation Models:
P. H.; Phillips, D. L. Ultrafast Time-Resolved Transient Absorption
and Resonance Raman Spectroscopy Study of the Photodeprotection
and Rearrangement Reactions of p-Hydroxyphenacyl Caged Phos-
phates. J. Am. Chem. Soc. 2006, 128, 2558−2570.
Equilibria, Structure, Spectra, and Dynamics. Chem. Rev. 1999, 99,
161−2200.
29) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
2
(
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A.; et al. Gaussian 09, revision A.02; Gaussian, Inc.:
Wallingford, CT, 2009.
(
8) Cheng, Q.; Steinmetz, M. G.; Jayaraman, V. Photolysis of
Gamma-(alpha-carboxy-2-nitrobenzyl)-L-glutamic Acid Investigated in
the Microsecond Time Scale by Time-Resolved FTIR. J. Am. Chem.
Soc. 2002, 124, 7676−7677.
(30) (a) Kus, N.; Breda, S.; Reva, I.; Tasal, E.; Ogretir, C.; Fausto, R.
FTIR Spectroscopic and Theoretical Study of the Photochemistry of
(
9) Turner, A. D.; Pizzo, S. V.; Rozakis, G.; Porter, N. A.
Matrix-Isolated Coumarin. Photochem. Photobiol. 2007, 83, 1541−
Photoreactivation of Irreversibly Inhibited Serine Proteinases. J. Am.
Chem. Soc. 1988, 110, 244−250.
10) Li, H.; Yang, J. H.; Porter, N. A. Preparation and Photo-
chemistry of o-Aminocinnamates. J. Photochem. Photobiol., A 2005,
69, 289−297.
11) Gagey, N.; Neveu, P.; Benbrahim, C.; Goetz, B.; Aujard, I.;
Baudin, J.-B.; Jullien, L. Two-Photon Uncaging with Fluorescence
Reporting: Evaluation of the o-Hydroxycinnamic Platform. J. Am.
Chem. Soc. 2007, 129, 9986−9998.
1
542. (b) Wolff, T.; Goerner, H. Photocleavage of Dimers of
(
Coumarin and 6-Alkylcoumarins. J. Photochem. Photobiol., A 2010, 209,
219−223.
(31) Hess, B. A.; Schaad, L. J.; Carsky, P.; Zahradnik, R. Ab Initio
Calculations of Vibrational-Spectra and Their Use in the Identification
of Unusual Molecules. Chem. Rev. 1986, 86, 709−730.
(32) Kasha, M. Characterization of Electronic Transitions in
Complex Molecules. Discuss. Faraday Soc. 1950, 14−19.
1
(
7
774
dx.doi.org/10.1021/jp403323h | J. Phys. Chem. A 2013, 117, 7767−7775

The Journal of Physical Chemistry A
Article
(
33) Gonzalez-Luque, R.; Garavelli, M.; Bernardi, F.; Merchan, M.;
Robb, M. A.; Olivucci, M. Computational Evidence in Favor of a Two-
State, Two-Mode Model of the Retinal Chromophore Photo-
isomerization. Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 9379−9384.
(
34) Hershfie., R.; Schmir, G. L. Lactonization of Coumarinic Acids:
Kinetic Evidence for Three Species of Tetrahedral Intermediate. J. Am.
Chem. Soc. 1973, 95, 8032−8040.
(
35) Hershfie., R.; Schmir, G. L. Lactonization of Ring-Substituted
Coumarinic Acids: Structural Effects on Partitioning of Tetrahedral
Intermediates in Esterification. J. Am. Chem. Soc. 1973, 95, 7359−
7369.
(
36) Reichardt, C. Solvent and Solvent Effect in Organic Chemistry;
VCH Verlagsgesellschaft mbH: Weinheim, Germany, 1988.
7
775
dx.doi.org/10.1021/jp403323h | J. Phys. Chem. A 2013, 117, 7767−7775