Highly Polarized Coumarin Derivatives Revisited: Solvent-Controlled Competition Between Proton-Coupled Electron Transfer and Twisted Intramolecular Charge Transfer

Article abstract of DOI:10.1002/chem.202001079

Linking a polarized coumarin unit with an aromatic substituent via an amide bridge results in weak electronic coupling that affects the intramolecular electron-transfer (ET) process. As a result of this, interesting solvent-dependent photophysical properties can be observed. In polar solvents, electron transfer in coumarin derivatives of this type induces a mutual twist of the electron-donating and -accepting molecular units (TICT process) that facilitates radiationless decay processes (internal conversion). In the dyad with the strongest intramolecular hydrogen bond, the planar form is stabilized, such that twisting can only occur in highly polar solvents, whereas a fast proton-coupled electron-transfer (PCET process) occurs in nonpolar n-alkanes. The k_PCET rate constant decreases linearly with the energy of the fluorescence maximum in different solvents. This observation can be explained in terms of competition between electron- and proton-transfer from a highly polarized (ca. 15 D) and fluorescent locally excited (¹LE) state to a much less polarized (ca. 4 D) charge-transfer (¹CT) state, a unique occurrence. Photophysical measurements performed for a family of related coumarin dyads, together with results of quantum-chemical computations, give insight into the mechanism of the ET process, which is followed by either a TICT or a PCET process. Our results reveal that dielectric solvation of the excited state slows down the PCET process, even in nonpolar solvents.

Full text of DOI:10.1002/chem.202001079

A Journal of
Accepted Article
Title: Highly Polarized Coumarin Derivatives Revisited: Solvent-
controlled Competition Between Proton Coupled Electron
Transfer and Twisted Intramolecular Electron Transfer
Authors: Olaf Morawski, Łukasz Kielesiński, Daniel T Gryko, and
Andrzej Sobolewski
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.202001079
Link to VoR: http://dx.doi.org/10.1002/chem.202001079
Supported by

10.1002/chem.202001079
Chemistry - A European Journal
FULL PAPER
Highly Polarized Coumarin Derivatives Revisited: Solvent-
controlled Competition Between Proton Coupled Electron
Transfer and Twisted Intramolecular Electron Transfer
Olaf W. Morawski,*[a] Łukasz Kielesiński,[a,b] Daniel T. Gryko*[b] and Andrzej L. Sobolewski*[a]
Dedicated to Prof. Kyo Han Ahn on occasion of his 65th birthday
[a]
Dr. Olaf W. Morawski, Łukasz Kielesiński, Prof. Andrzej L. Sobolewski
Institute of Physics
Polish Academy of Sciences
Al. Lotników 32/46, 02-668 Warsaw, Poland
E-mail: morawo@ifpan.edu.pl, sobola@ifpan.edu.pl
Łukasz Kielesiński, Prof. Daniel T. Gryko
Institute of Organic Chemistry
[b]
Polish Academy of Sciences
Kasprzaka 44/52, 01-224 Warsaw, Poland
E-mail: dtgryko@icho.edu.pl
Supporting information for this article is given via a link at the end of the document.((Please delete this text if not appropriate))
Abstract:Linking
a
polarized coumarin unit with an aromatic
competition with radiative and radiationless deactivation
processes.[8-13]
substituent via an amide bridge results in weak electronic coupling
affecting the intramolecular electron transfer (ET) process. As a result
of this, interesting solvent-dependent photophysical properties can be
observed. In polar solvents, electron transfer in coumarin derivatives
of this type induces a mutual twist of electron-donating and electron-
accepting molecular units (TICT process) that opens radiationless
decay processes (internal conversion). In the dyad with the strongest
intramolecular hydrogen bond, the planar form is stabilized so twisting
can only occur in highly polar solvents, whereas fast proton coupled
electron transfer (PCET process) occurs in non-polar n-alkanes. The
kPCET rate constant decreases linearly with the fluorescence maximum
energy in different solvents. This observation is explained in terms of
competition between electron and proton transfer from a highly
polarized (~15 D) and fluorescent locally-excited (1LE) state to a much
Among the proposed intramolecular structural changes
related to the ET process, the twisted intramolecular charge-
transfer (TICT) hypothesis of Grabowski et al.[14] has attracted the
most attention (for a review see Ref. [15] and references therein).
Another intramolecular structural change related to this
phenomenon is the excited-state intramolecular proton transfer
(ESIPT) process, introduced originally by Weller decades ago.[16]
Both processes may result in
a
strongly red-shifted
fluorescence[17-21] and open competing radiationless decay
channels.[22] It has been shown that the radiationless decay
channel associated with the ESIPT process involves mutual
twisting of the electronically conjugated proton-donating (PD) and
proton-accepting (PA) molecular moieties in the S1 state which
less polarized (~4 D) charge-transfer (1CT) state, a unique occurrence. leads to a conical intersection (CI) with the ground state.[23] If,
Photophysical measurements performed for a family of related
coumarin dyads together with results of quantum-chemical
computations give insight into the mechanism of the ET process which
is followed by either a TICT or PCET process. Our results reveal that
dielectric solvation of the excited state slows down the PCET process,
even in non-polar solvents.
however, PD and PA moieties are not electronically conjugated to
each other or belong to different molecular systems, the ESIPT
process, or more precisely, the proton-coupled electron-transfer
(PCET) process, results in the formation of a biradical pair and
may lead directly to a CI (S1/S0) even for a coplanar orientation of
molecular sub-systems.[24,25]
In the majority of ET-relevant molecular systems, absorption
of a photon promotes the system to the first excited singlet state
(S1), the so-called locally-excited (LE) state, with a polarization
similar to or somewhat larger than that of the ground state (S0).
The photoinduced electron (or charge) transfer, and associated
conformational changes, result in the formation of a polarized
species. The ET reaction can thus be qualitatively described in
terms of the Marcus-parabola of free energy. Interaction of this
molecular system with a polar environment (solvent) may result in
the famous inverted-parabola behavior of the ET rate constant on
modulation of the free energy.[6]
There is a growing family of polarized molecular systems
where optical excitation to the 1LE state can significantly increase
the dipole moment of the system.[26-29] Among these, a class of
highly polarized molecular dyads composed of two coumarin
moieties linked in a (donor-acceptor)-linker-(donor-acceptor) (DA-
Introduction
Photoinduced inter- and intra-molecular electron transfer
(ET) processes play an important role in many areas of physics
and chemistry, ranging from charge separation in photovoltaics[1]
and photoluminescent sensors and switches,[2,3] to natural[4] and
artificial[5] photosynthesis. The theoretical basis for ET processes
was originally established in the seminal works of Marcus[6] and
Weller[7] more than half a century ago. The kinetics and
thermodynamics of the ET process is modulated by the
electrochemical and excited-state properties of molecular
systems, with accompanying structural changes, and is in
1
This article is protected by copyright. All rights reserved.

10.1002/chem.202001079
Chemistry - A European Journal
FULL PAPER
L-DA) orientation, recently studied by us,[30,31] is particularly
interesting. However, flexibility of the amide linker used in these
studies results in very complex photophysical behavior, related to
both thermo- and photo-chemical populations of numerous
conformers. This observation prompted us to study the
photophysics of such systems at a more fundamental level. In this
respect, we designed two strongly polarized coumarin derivatives
possessing a CONH linker in position 3 to an aryl substituent of
differing electronic character i.e. one possessing an electron-
donating and one an electron-withdrawing substituent. The final
choice of substituents was governed by the desire to achieve
excellent solubility in a broad range of solvents, which was critical
for studies of solvatofluorochromism.
Scheme 1. Coumarins investigated in this study.
These molecules are denoted as Amide-PhF5, H-Amide-Ph and
D-Amide-Ph in Scheme 1. The fourth molecule Ester-Ph was
designed as a model lacking the ability to form an intramolecular
hydrogen bond. These compounds were synthesized using
standard coupling reagents from 7-(diethylamino)-2-oxo-2H-
chromene-3-carboxylic acid and the corresponding amines and
phenols.
Results and Discussion
Theoretical explorations
For all three investigated compounds the most stable
structure on the potential-energy (PE) surface of the ground state
has the trans conformation of the inter-ring amide or ester linker
(c.f. inserts in Figure 1). In these theoretically explored systems,
the terminal aliphatic groups were replaced by methyl groups.
This only has a marginal influence on the electronic properties of
the systems but significantly speeds up excited-state explorations
at the ab initio theoretical level.
Figure 1.Vertical absorption (blue arrows) and fluorescence (red arrows) transitions, and radiationless relaxation (dashed arrows) of investigated systems computed
at the ADC(2)/cc-pVDZ level of theory. Numbers denote the energy (in eV) and dipole moment (in Debye) computed for a given state. Solid lines denote adiabatic
(optimized) energy of a given state while dashed lines denote ‘vertical’ energy of the relevant state computed at that geometry.
Absorption spectra of the investigated systems are
dominated by the fully allowed (f = 0.8-0.9) transition to the first
excited singlet state of the ππ* electronic transition located
vertically around 3.4 eV at this theoretical level. Geometry
optimization of the S1 state resulted in stable trans structures in
the case of Ester-Ph and Amide-PhF5 which are predicted to
fluoresce at about 2.9 eV (Figure 1b and Figure 1c). Excited-state
geometry optimization of the H-Amide-Ph system, however,
resulted in a spontaneous excited-state intramolecular proton
transfer (ESIPT) process from the amide moiety to the carbonyl
group of coumarin (Figure 1a). As a result, the S1-S0 energy gap
is reduced to 0.36 eV which precludes observation of any
emission. Instead, internal conversion to the ground state is
predicted to be the dominant deactivation channel of optical
excitation of this system. All investigated systems are highly
polarized in the ground state (µ0 = 7-12 D, Figure 1), and
electronic excitation to the lowest 1ππ* state increases their
polarization significantly (µ1 = 15-21 D, Figure 1), but the ESIPT
process reduces the dipole moment of the H-Amide-Ph system
to 4.3 D in the S1 state (Figure 1a).
The radiationless decay channel associated with the ESIPT
process is not present in Ester-Ph because of the lack of a
‘mobile’ hydrogen-bonded proton. Interestingly, ESIPT does not
occur spontaneously in Amide-PhF5 which contains such a
proton. The electron withdrawing nature of the -PhF5 moiety
apparently acts as a barrier for this reaction. To investigate this
2
This article is protected by copyright. All rights reserved.

10.1002/chem.202001079
Chemistry - A European Journal
FULL PAPER
idea in more detail, calculations of the potential-energy (PE)
profiles along the PT reaction coordinate (NH distance) were
performed and the results are presented in Figure 2b. PE profiles
were optimized in the excited singlet state (S1) and the ‘vertical’
energy of the ground state was computed along such determined
minimum-energy reaction paths.
a ‘local’ excitation within the coumarin core (Table 1a), is an
increasing function of the proton-transfer reaction coordinate (NH
distance, squares in Fig. 1b) and at a certain value of the NH
distance (around 1.0-1.15 Å) it crosses with another MEP which
is a decreasing function of the coordinate (triangles in Fig. 2b).
Thus, transfer of a proton from the amide linker to the coumarin
core in Amide-PhF5 requires elongation of the NH bond and has
the energy barrier seen in Figure 2b. As a result, Amide-PhF5
should be fluorescent in a vacuum and solvation should result in
further stabilization of the highly polarized LE state and therefore
increase the barrier for a PCET process. Similarly to H-Amide-Ph
(Figure 1a), the proton transferred form of Amide-PhF5 shows a
very small energy gap with the ground state, and in effect the
ESIPT process provides a channel for radiationless deactivation
of an optical excitation. The crossing between MEP profiles of the
S1 state visible in Figure 2b is an apparent crossing since these
reaction paths differ in many other intramolecular coordinates as
it is reflected by different vertical PE profiles of the ground state
computed along these coordinates. This also indicates that proton
transfer is a non-adiabatic process induced by electron transfer
1LE
(a)
(b)
1LE
1PCET
1ET
between molecular moieties (Table 1b). This is
a rather
Figure 2. Potential-energy (PE) profiles of Amide-PhF5 computed in the S1
state (symbols connected by solid lines) along the minimum-energy path (MEP)
for electron transfer (ET, left panel) and for proton coupled electron transfer
(PCET, right panel). Circles connected by dashed lines denote vertical-energy
profiles of the ground state computed along the respective MEP. Squares
denote the locally-excited (LE) state; triangles denote the respective charge-
transfer (ET and PCET) states.
unexpected result for an ESIPT process which generally occurs
on an adiabatic PE surface of the excited state, i.e. electron
follows the proton adiabatically over the course of the reaction.[32]
A possible explanation for this effect is that the proton donating
(amide-NH) and proton accepting (coumarin-C=O) moieties are
not electronically conjugated to each other. In other words, a more
proper classification of this reaction is the electron-driven proton
transfer (EDPT)[32] or more commonly abbreviated as proton
coupled electron transfer (PCET).[33-38]
It can be noticed upon inspection of Figure 2 that the
minimum-energy PE profile (MEP) of the 1LE state, which involves
Table 1. Nuclear conformation and electronic structure of Amide-PhF5 computed at the most crucial points of the S1 PE landscape of Figure 2.
3
This article is protected by copyright. All rights reserved.

10.1002/chem.202001079
Chemistry - A European Journal
FULL PAPER
To discuss this phenomenon in more detail, the electronic
structure of the LE and PCET states determined at the same
value of the PT coordinate (NH = 1.1 Å) are shown in Table 1. It
can be seen that for the 1LE state (Table 1a) both the singly
involved for these processes, complex photophysical behavior
depending on both chemical substitutions and on environment is
expected to be observed for this system. A particularly interesting
and intriguing aspect of the observed photophysical phenomenon
occupied HOMO and LUMO are localized on the coumarin moiety, is related to the fact that the 1LE state is highly polarized (µ1 = 15-
but in the 1PCET state (Table 1b) the HOMO is localized on the -
PhF5 moiety. In effect, a positive charge (hole) is transferred from
the coumarin to -PhF5 resulting in partial neutralization of the
charge distribution (µ1: 19.8 D → 3.2 D) which provides a driving
force for the PT reaction which leads to further reduction of µ1 to
20 D, Figure 1), and the competing TICT and PCET processes
result in formation of relatively non-polarized states. This is rather
exceptional for ET process, since an increase in solvent polarity
is expected to destabilize 1TICT (and 1PCET) states with respect
to the polarized S0 and 1LE states which may result in an increase
of the barrier for these processes and in effect may increase the
fluorescence yield vs. polarity of the solvent.
1
1.82 D at the minimum of the PCET state (Table ST12b of the
ESI). The S1-S0 energy gap is only 0.34 eV at the minimum of the
1PCET state (NH = 1.75 Å), and the system relaxes to the ground
state via internal conversion.
Experimental investigations
The last stable point along the proton transferring MEP was
obtained upon shrinking the NH distance to 1.05 Å. An
unconstrained geometry optimization starting from this point
resulted in another solution, which is illustrated in Figure 2a and
in Table 1c. In this geometry (C=O bond length around 1.33 Å)
the ground state (red circles in Figure 2a) is located slightly (0.07
eV) above the S1 state. This means a conical intersection
between the states is located near to this geometry. By inspecting
the electronic structure of the S1 state at its equilibrium geometry
(Table 1c) the same molecular orbitals as in the 1PCET state
(Table 1b) are involved, and the state has a pure electron-transfer
(ET) character, however, the nuclear geometries of these states
are essentially different. In the 1PCET state at NH = 1.1 Å a strong
hydrogen bond R(O…H) = 1.35 Å which facilitates the PT process
is formed. On the contrary, in the 1ET state the hydrogen bond is
broken (R(O…H) = 2.37 Å) and the amide moiety is strongly
twisted and pyramidized. However, a close inspection of the
equilibrium geometry of the 1ET state shows many intramolecular
coordinates are involved in the reaction. The most remarkable is
the perpendicular orientation of the coumarin core and the
benzene rings in the 1ET state vs. 53° at the 1LE equilibrium. Thus
formally, this state can be classified as a TICT state. There is also
a significant elongation of the coumarin C=O bond length from
Existence of an intramolecular hydrogen bond in H-Amide-
Ph and Amide-PhF5 was investigated in both the solid state and
in solution. Crystals of compounds H-Amide-Ph and Amide-
PhF5suitable for X-ray crystallography were obtained by slow
diffusion of hexane into a solution of the appropriate compound in
dichloromethane. X-ray analysis clearly proved the existence of a
hydrogen bonding interaction between the amide proton and
carbonyl oxygen of the coumarin moiety (Figure SF3 of the ESI).
The hydrogen bond length for Amide-PhF5 and H-Amide-Ph is
1.905 Å and 2.003 Å, respectively. The dihedral angle between
coumarin core and arylamide substituent for H-Amide-Ph is
23.84° while Amide-PhF5 is more twisted and the angle is 45.76°.
This difference is probably related to the fact that the
intramolecular hydrogen bond in Amide-PhF5 is weaker than in
H-Amide-Ph and hence the aniline moiety is more twisted with
respect to the coumarin core. The packing diagrams of Amide-
PhF5 and H-Amide-Ph indicate that the molecules are arranged
in head-to-tail manner (Figure SF4 of the ESI).
It is well known that the chemical shift of a proton is
proportional to the strength of a hydrogen bond it is involved
in,[39,40] so NMR can be used as
a technique for the
characterization of hydrogen bonding in solution. To investigate
the influence of solvents on hydrogen bonds in Amide-PhF5 and
H-Amide-Ph, we performed additional NMR measurements in
selected polar and non-polar solvents (Figure SF5). The chemical
shifts for both molecules behave similarly: in non-polar benzene
or solvents of low polarity (chloroform), the signals from the amide
protons are more upfield shifted, while in polar dimethyl sulfoxide
the signals remain in a downfield position. This reveals that the
intramolecular hydrogen bond is strong in non-polar solvents but
weakens in polar ones. The bond is stronger in H-Amide-Ph than
in Amide-PhF5 as chemical shifts for the former are larger than
for the latter (Figure S5b, lines from 10.5 to 11.25 and S5a, lines
around 10.5 respectively). The electron withdrawing -PhF5 group
reduces the electron density on the NH hydrogen and leads to the
weaker H-bonding. This is consistent with the observation that the
energy of the NH vibration in H-Amide-Ph, 3262 cm-1, is slightly
higher than in Amide-PhF5, 3245 cm-1, proving the larger strength
in the former molecule (Figure S6).
1
1
1.24 Å in the LE state to 1.33 Å in the ET state (Table 1). The
latter was chosen as the reaction coordinate for optimization of
the ET reaction pathway. PE profiles computed along this reaction
coordinate are shown in Figure 2a.
Inspection of the PE profiles of Figure 2a shows that the 1LE
state is crossed by the 1ET state close to the bottom of the former
state. Once again, this is only an apparent crossing since many
other intramolecular coordinates are involved in the reaction, and
relaxation along these coordinates results in a large shift of the
PE profile of the ground state along this reaction path. On the
other hand, the 1TICT/S0 crossing, visible in Figure 2a as red
circles/red triangles lines crossing, is a real crossing since both
PE profiles are computed along the same reaction path. Thus, a
non-adiabatic transition from the locally-excited S1 fluorescing
(1LE) state to the charge-transfer state, accompanied by either
transfer of the proton (1PCET state) or by structural relaxation
along other intramolecular coordinates (1TICT state), is
deleterious for fluorescence due to internal conversion to the
ground state.
The shape of the optical spectra of the three molecules in n-
hexane are similar and are positioned at comparable energies
(Figure 3). Small differences between the maxima of the
absorption spectra in n-hexane (24155, 24213 and 24570 cm-1 for
Amide-PhF5, H-Amide-Ph and Ester-Ph, respectively), reveal
that the hydrogen bond present in Amide-PhF5 and H-Amide-Ph
lowers the energy of the S1 state.
In summary, both TICT and PCET processes provide a
source of radiationless deactivation of an optical excitation and
compete with fluorescence as relaxation pathways for the system.
Since these states are of a different electronic nature than the
fluorescing 1LE state and different intramolecular coordinates are
4
This article is protected by copyright. All rights reserved.

10.1002/chem.202001079
Chemistry - A European Journal
FULL PAPER
The experimental energy values differ from the calculated ones
by ca. 0.3-0.4 eV, a typical overestimation of excitation energy at
this theoretical level.[41] The excited-state increase of dipole
moment calculated from solvatochromic slopes and expressed as
the µe/µg ratio (3.53, 2.8 and 2.9 - for Amide-PhF5, H-Amide-Ph
and Ester-Ph respectively; see ESI sections 4.3 to 4.5 for details)
as well as the calculated values (1.76, 1.7 and 2.05) point to a
large increase of µ upon photoexcitation. This corroborates the
fact that the three molecules studied are push-pull compounds,
with a remarkable charge shift due to an optical transition.
The influence of solvent on the fluorescence quantum yield
ΦF and decay time τF depends strongly on the molecule. Amide-
PhF5 and Ester-Ph have very high ΦF exceeding 0.9 and long τF
≈ 2.4 ns in several non-polar solvents (Tables ST4 and ST6 of the
ESI). In polar solvents, ΦF drops to 0.02 and τF shortens to below
100 ps. Very different and complex behaviour is observed for H-
Amide-Ph: in n-hexane it has very weak emission with low ΦF =
0.001 and short τF≈ 2 ps. In longer n-alkanes, toluene and
solvents of low dielectric constant (ε< 10) both parameters
increase to reach maximal values in dichloromethane and 1,2-
dichlorobenzene (ΦF = 0.22 and τF = 620 ps, Table ST5). In more
polar solvents (ε> 10) the fluorescence quantum yield decreases
to 0.063 and decay time to 214 ps. Fluorescence decays are
single exponential for all molecules (Figure SF8, SF13 and SF17),
so one may estimate the radiative kr and non-radiative knr rate
constants from the formulas: kr = ΦF/τF and knr = 1/τF - kr. In the
three molecules studied the radiative rate constant is nearly
constant in all solvents, while the non-radiative rate constant for
Amide-PhF5 and Ester-Ph grows with solvent polarity, in H-
Amide-Ph it has a more complex behaviour – strongly decreasing
with the solvent polarity parameter f’ in non-polar and polarizable
solvents and slightly growing with polarity in polar solvents (Figure
4 and Tables ST4 to ST6).
Figure 3. Absorption (solid lines) and normalized fluorescence spectra(dashed
lines) of Amide-PhF5, H-Amide-Ph and Ester-Ph in n-hexane recorded with
excitation at 390 nm. Legend specifies colours of lines.
The absorption spectra only shift slightly with regard to solvent
polarity. In contrast, the fluorescence spectra show much more
pronounced solvatochromism (Figure SF7, SF11 and SF15 of the
ESI), which points to a substantial increase of the electric dipole
moment in the excited state (µe) in comparison to its value in the
ground state (µg). From the slopes of the solvatochromic
dependencies of absorption spectra the energy of the first excited
state in the gas phase may be determined (see section 5.1 of ESI
for the method used and ESI sections 5.3 to 5.5 for data
elaboration). The experimental values (24160 cm-1 or 3.0 eV,
24290 cm-1 or 3.01 eV and 24613 cm-1 or 3.05 eV for Amide-PhF5,
H-Amide-Ph and Ester-Ph, respectively) can be compared with
those obtained through calculations (3.35 eV, 3.4 eV and 3.49 eV).
Figure 4. Semilogarithmic plot of radiative (kr, blue crosses) and non-radiative (knr, red squares) deexcitation rates of Amide-PhF5 (left), H-Amide-Ph (middle) and
Ester-Ph (right) versus modified Lippert-Mataga parameter f’= [(ɛ-1/(2ɛ+1) – 0.5×(n2-1)/(2n2+1)] in solvents. Solid lines represent linear fits to knr data.
The growth of knr observed for the Amide-PhF5 and Ester-Ph in
polar solvents (Figure 1) is assigned to an increase in the internal
conversion rate induced by nuclear rearrangement related to
electron transfer which leads to a conical intersection with the
ground state (see sections 5.3 to 5.6 of the ESI). Similar
phenomenon have previously been observed for other 7-
aminocoumarin derivatives and assigned to stabilization of a dark
TICT state.50 Indeed, in the solid state, where large amplitude
motions are hindered, fluorescence decays of all molecules
studied are hundreds of nanoseconds long (Figure SF20).
Amide-Ph, kDPCET, can be calculated as a difference between
non-radiative rates of these molecules, kH-Ph and kD-Ph
nr
nr
respectively, and the rate of Amide-PhF5, kPhF5nr, i.e. kHPCET = kH-
Ph
nr - kPhF5nr, and kDPCET = kD-Phnr - kPhF5nr. So obtained PCET rates
are presented in Figure 5, as semilogarithmic plot of PCET rate
constants depicted versus energy of the fluorescent LE state, the
initial state for the transfer process. A linear decrease of the PCET
rate constant is observed over a wide range of solvents (n-
alkanes, toluene, ethyl acetate, tetrahydrofuran and
dichloromethane) where PCET rate constants are higher than that
of the competing radiationless decay channel caused by benzene
ring twist. This result indicates that the barrier for the PCET
process is modulated by solvation of the S1(LE) state. Both theory
Amide-PhF5, a compound lacking proton transfer, may be
treated as a model molecule for H-Amide-Ph in evaluation of the
PCET rate constant. PCET rates for H-Amide-Ph, kHPCET, and D-
5
This article is protected by copyright. All rights reserved.

10.1002/chem.202001079
Chemistry - A European Journal
FULL PAPER
Transitions are sketched out by arrows: optical - absorption and emission
(dashed lines), radiationless – PCET and ET (dotted lines).
and solvatochromism of absorption and fluorescence spectra
(Figure SF11) set the 0-0 transition for the LE state in the gas
phase at 3.0 eV. Using the linear fit parameters of Figure 5 for this
energy, a gas phase PCET rate constant of 5.64×1012 s-1 can be
obtained.
This phenomenon occurs because the initial LE state has a much
higher dipole moment than the final CT state. The fluorescent LE
state enables monitoring of its solvation and PCET rate constant
in a wide range of solvents, which gives an interesting perspective
on the transfer mechanism and factors influenced by solvent.
Kinetic isotope effect decreases at high energy barrier
The theory for the PCET process was developed as an
extension of the Marcus electron transfer model by Cukier and
Nocera[51] and by Hammes-Schiffer and Soudackov.[52] In the
general PCET theory the transferring proton and the active
electrons are treated quantum mechanically, and the PCET
reaction is described in terms of nonadiabatic transitions between
mixed electron-proton vibronic states. The rate constant of the
PCET process between the initial µ and final ν vibronic states is
expressed as:[53]
Figure 5. Semilogarithmic plot of PCET rates of H-Amide-Ph,(kHPCET, blue
circles), D-Amide-Ph, (kDPCET, red crosses) left vertical axis, and kinetic isotope
effect rate (KIE =kHPCET/kDPCET, black stars) versus the energy of fluorescence
spectra maxima.
where Pµ is the Boltzmann population reactant state, Vel is the
electronic coupling, Sµν is the overlap integral between the proton
vibrational wave function of reactant and product states that
Discussion
depends on the proton donor-acceptor distance R, G0 is the
µν
For H-Amide-Ph in the gas phase, theory predicts a fast,
reaction free energy associated with vibronic state pair (µ, ν) and
λ is the total reorganization energy. T and kB denote the absolute
temperature and Boltzmann constant, respectively. The overall
PCET rate constant is obtained by thermal averaging of the rate
expressed in eq. 1 weighted by the probability of sampling a given
proton donor-aceptor distance R: kPCET = ∫ kPCET P(R)dR.[53] The
theory has been successfully applied to PCET processes in the
Marcus normal as well as in inverted regions,[53] and to describe
the kinetic isotope effect (KIE).[54] The theory reveals the effect of
the contribution to the rate constant is not only from the ground
vibronic state (0,0 pair) but also from excited vibronic states due
to larger overlap integrals of excited states.[53] This decreases KIE
because the excited vibronic states are associated with larger
overlaps and therefore a smaller ratio of hydrogen to deuterium
overlaps.[54] The effect is observed for H-Amide-Ph and D-
Amide-Ph in polar solvents, where the barrier increases and
PCET reaction slows down (Figure 5).
barrierless, nonadiabatic PCET (Figure
6 and 1a). Weak
fluorescence of H-Amide-Ph in n-pentane reveals that a small
barrier already exists in this inert solvent. This intuitively may be
understood, as the solvent increases the reorganization energy,
λ, and decreases the driving force, G. Indeed, in H-Amide-Ph
the initial state for PCET process, the LE state, has much higher
dipole moment, µe = 15 D, than the final CT state, µCT = 4.3 D, so
the LE state should be more stabilised by the solvent than located
at a lower energy CT state, so as a result G should decrease
with dielectric solvation. As a consequence of the solvent-induced
change of both factors, increasing λ and decreasing G, the
molecular system is detuned from the optimal ET condition of λ ≈
-G, which slows down the transfer rate in the normal Marcus
region.
PCET rate constant correlates with solvatochromism of
fluorescence spectrum
The PCET process in H-Amide-Ph occurs between the
initial higher and very polarized S1(LE) state and the final lower
and much less polarized CT state. Fluorescence from the S1(LE)
state to the less polarrized S0 state is also connected with a
remarkable change of charge distribution. Thus, a substantial
charge shift occurs in both transitions. The analogy between an
optical CT transition (radiative relaxation process) and electron
transfer process has been shown already by Marcus.[42] For a
single high-frequency intramolecular vibration (hν> 1000 cm-1) a
formula for the fluorescence spectrum of bridged DA molecules
was derived[43]
Figure 6. Simplified diagram of the photophysical processes that compete with
each other for the deactivation of optically populated 1LE state in the molecules
studied: H-Amide-Ph (left), Amide-PhF5 (middle), and Ester-Ph (right). For
transparency only the ground (lower) and 1LE (upper) states are shown.
6
This article is protected by copyright. All rights reserved.

10.1002/chem.202001079
Chemistry - A European Journal
FULL PAPER
Factors influencing electron transfer
where ΔG is the free energy change, h is the Planck constant, ν
is frequency of light, νi is the frequency of high frequency
intramolecular vibration, λS is the solvent reorganization energy,
and S=λi/hνi, λi is the intramolecular reorganization energy.
Following the above mentioned analogy, the rate constants for
radiative charge recombination in DA molecules,[44] as well as for
non-radiative electron transfer[45] were also proposed. This
allowed for the formulation of the charge recombination rate knr in
the form[46]
The free energy for the ET process can be calculated using
the Weller approximation for electron transfer[55]
where E0(A•+A) and E0(DD•-) are reduction potentials of acceptor
and donor parts of the dyad respectively, RAD is the distance of
ET and E0-0 approximates the free energy of difference of the
initial and final state, usually LE and ground state (GS), which
then can be assigned to the interception of absorption and
emission spectra. In our dyadsthe final state is the optically dark
CT state so this parameter cannot be directly observed.
Electrochemical potentials depend on solvent and for single
electron oxidation or reduction their values in solvents i and j of
static dielectric constant εi and εj are related by the simplified
formula[13]
where VAD is the electronic coupling element between the two
states involved in the transition. The fluorescence spectrum of H-
Amide-Ph in n-alkanes and toluene is dominated by the vibronic
progression at 1150 cm-1 (Figure SF12 and section 5.4 of ESI),
justifying usage of eq. 2. This feature and the similarity between
expressions (1), (2) and (3) suggested that kPCET values could be
plotted against the position of fluorescence spectrum maximum
ℎνmfl ax. The good (R-Square = 0.9945) linear dependence of kPCET
푚푎푥
푓푙
where F is the Faraday constant and R is the Born effective radius.
The reorganization energy, , is the sum of inner-sphere
(solute) i and outer-sphere (solvent) s, reorganization energies,
i.e.  = i + s, for which the Marcus two-sphere model is given by
on ℎ휈
observed in Figure 5 for several solvents from n-
pentane to dichloromethane confirms the relationship between
the two observables and justifies the approach. Indeed, knr in
푚푎푥
푓푙
Figure 4b and ℎ휈
presented in inset to Figure SF11a behave
in the same way with solvent polarity parameter f’ - they decrease
abruptly in non-polar solvents and much slower in polar ones.
Electron transfer triggers proton transfer or TICT depending
on energy barrier
where RA and RD are the Born radius of acceptor and donor
spheres respectively and RAD is distance between the centers of
them. The expression was derived with assumption of separated
spheres, i.e. RA + RD<RAD, but has also been applied for spheres
in contact as demonstrated for dyads of anthracene.[53] Let us
notice, that the reciprocal of dielectric constant, the 1/ factor, is
present in all equations (4) – (6), its value and change in non-polar
or medium polarity solvents is expected to be much more than in
polar solvents.
An alternative approach for the calculation of barriers and
solvent reorganization energy in optical and thermal ET
processes was proposed for a point dipole molecule in an
Onsager cavity using the thermodynamic approach proposed by
Marcus.[56] Two “Brunschwig” solvent polarity parameters F1 and
F2 were derived (equations SE4 and SE5 in section 5.7 of ESI) for
use in expressions of G change in solvent versus gas phase
G0()-G0( = ) (eq SE7) and solvent reorganization energy s
(eq SE6). Expressions for solvatochromism of absorption,
Eabs(), and emission, Eem(), were also derived.56 These
formulas are similar to those of the reaction field method by
Onsager or Lippert, but are obtained with a general method and
less restrictive assumptions.
Both non-radiative processes occurring in H-Amide-Ph are
associated with electron transfer. This is clear when one
compares the HOMO and LUMO orbitals involved in the LE state
(Table ST8 in analogy with Table 1a) with the relevant orbitals of
the TICT and PCET states where the electron density localized
on the coumarin rings of the former state moves to the benzene
ring in the latter states (Tables ST11a and ST11b). Optical
excitation shifts the charge from the electron donating moiety
(diethylamine) to the electron accepting coumarin, resulting in an
increase of the molecular dipole moment from 9 D in the S0 state
1
to about 15 D in the LE state (Table ST8). In the course of the
TICT and PCET processes, the electronic charge shifts in the
opposite direction partially neutralizing the polarization of the
system, so the final states have a particularly small dipole moment
of about 4-5 D (Tables ST11a and ST11b). Hence, one may
consider both radiationless transitions as electron transfer
processes. An electron transfer from the benzene to coumarin
moiety triggers either proton transfer from NH of the amide linker
onto the carbonyl oxygen (PCET) or results in elongation of the
C=O double bond and inter-ring twisting that stabilizes the TICT
state. Both processes lead the system to a CI with the ground
state which opens internal conversion channels. The ET triggered
twist occurs in polar solvents in all three dyeds. In acetonitrile the
process is the fastest in Ester-Ph which does not have an
intramolecular hydrogen bond (τF = 100ps) and slower in in H-
Amide-Ph (τF = 274ps) and D-Amide-Ph (τF = 288ps) where the
hydrogen bonds are stronger.
In both approaches, the total solvent relaxation energy may
be related to Stokes shift,  = (max
- maxflu)/2, and the free
abs
energy to the mean of optical transition energy, G = (max
+
abs
maxflu)/2.[46,56] In H-Amide-Ph the trend from solvatochromism of
spectra can be relevant to changes in G for PCET as both
processes originate from the highly polar 1LE state and end up in
much less polarized states. Figure SF25 of the ESI depicts Stokes
7
This article is protected by copyright. All rights reserved.

10.1002/chem.202001079
Chemistry - A European Journal
FULL PAPER
shift versus four different solvent polarity parameters. A linear
trend is observed only for the reciprocal of dielectric constant
(Figure SE25b). This points to the outer sphere dielectric
polarisation and Born cavity model rather than a point-dipole in
Onsager cavity approximation, (eq. SE1 and SE6). This is not
surprising as the 1/ factor is present in all 4-6 equations. The 0-
The above considerations provide, however, only a
phenomenological explanation of experimental observations,
discussed in an analogy to the Marcus theory of electron
transfer.[47] Since excited-state analytical gradients are not so-far
available for dielectric solvation models, in order to provide more
insight into the mechanistic aspects of the observed phenomena
we performed excited-state geometry optimization of the H-
Amide-Ph molecular system as a complex with a single molecule
of the polar solvents acetonitrile and methanol. The results of this
are presented in Tables ST13a and ST13b of the ESI,
respectively. It can be noticed upon inspection of the Table ST13a
that the interaction of H-Amide-Ph with a single acetonitrile
molecule hinders the PCET reaction by stabilization of the LE
state. A similar effect is also observed for the complex with a
methanol molecule (Table 13b). Additionally, the protic methanol
molecule competes with the intramolecular hydrogen bond. This
hinders the PCET process to a greater extent, but on the other
hand promotes the ET process by twisting the benzene ring with
respect to the coumarin core. Both polarized solvent molecules
reduce the S1/S0 energy gap, an effect which is stronger for the
complex with methanol, supporting the larger bathochromic shift
of fluorescence in this solvent and the increase of internal
conversion to the ground state. These effects are in full accord
with experimental results and provide a mechanistic explanation
of our observations.
1
0 transition to the LE state, important for G evaluation, shows
different trends in non-polar and polar solvents (Figure SE24).
The 1/ factor offers a monotonic trend, though the Brunschwig
F1-F2 parameter cannot be excluded. The changes in  and G,
even in non-polar solvents are clear, and both parameters lead to
1
a lowering in energy of the LE state and increase of the barrier
for the PCET reaction. The parameter (1/n2 – 1/) used for outer
sphere reorganization energy s leads to steep change of the
PCET rate constant, Stokes shift and E0-0 values in n-alkanes and
toluene (Figures SE23-25). The same is true for the amended
Marcus formula for s (eq. 7)[57]
which was derived to reduce overestimation of solvent relaxation
energy obtained with equation (6). It is important to notice that in
n-alkanes static≈optical = n2, which leadsto a very steep change in
quantities presented by the function of the (n-2-e-1) parameter. The
inclusion of optical frequency dielectric constant may account for
buildup of the barrier for ET already in n-alkanes, slowing down
the process in addition to the substantial solvatochromism in non-
polar solvents. However, inspection of figures SF23-SF25 reveals
that the best correlation is observed between the Stokes shift and
the 1/ parameter. This suggest that the slowing down of PCET
process and the solvatochromism of fluorescence may origin from
screening of electrostatic force in dielectric media.
Dielectric friction is another physical factor that may have an
impact on the fast PCET rate.[58] In n-alkanes at 20 °C the
dielectric relaxation time, D, changes from 2.4 ps in n-hexane
through 2.7 ps in n-nonane to 3.2 ps in n-decane and is related to
rotation of the ends of the alkyl chains because rotation of whole
molecules is an order of magnitude slower.[59] The longitudinal
relaxation time, L=D(optical/static) is very similar to D as in n-
alkanes static≈optical. Fluorescence decay times of H-Amide-Ph in
n-alkanes related to the PCET process, increase faster with chain
length than L values, and decay times of D-Amide-Ph are longer
than the longitudinal dielectric relaxation times (Table ST7), which
indicates dielectric friction is not the main factor limiting the PCET
rate. The next factor influencing the PCET process is the strength
of NH···O hydrogen bond which decreases with solvent polarity
as revealed in Figure SF5. A weaker HB requires larger distances
being penetrated for PCET to occur. This translates to a higher
barrier for the process. Solvent polarity affects the barrier in the
same direction through stabilisation of the 1LE state, so
disentanglement of the two factors influence remains a challenge.
Hydrogen bonding with protic solvent molecule provides another
channel competing with PCET and reducing its rate. The external
H-bond and other solvent – specific interactions, changing
photophysics of a molecule, were not a subject of this study.
Other non-radiative processes
For unsubstituted coumarin and 7-hydroxycoumarin two
other non-radiative mechanisms have been reported: breaking of
the lactone bond leading to ring opening and transition into a dark
nπ* state along the carbonyl stretching mode.[48]
Both
mechanisms were proposed for the ultrafast radiationless
relaxation observed in unsubstituted coumarin, where electron
density on the lactone ring is low.[48,49] The non-radiative
processes in coumarin derivatives studied in this work are slower
and do not lead to any photochemical products. Theoretical
calculations do not predict ring cleavage for H-Amide-Ph, Amide-
PhF5 or Ester-Ph. In light of both these results and in obvious
relation to the fast radiationless relaxation with the existence of a
NH – carbonyl oxygen hydrogen bond we excluded ring opening
from interpretation of our experimental results. Therefore, we
conclude that the radiationless relaxation observed in the
investigated coumarin derivatives is a manifestation of the
occurrence of two competing processes: PCET and TICT, which
through CI with the ground state, lead to efficient internal
conversion.
Conclusions
In conclusion, the slight structural modification of classical
push-pull coumarins possessing electron-donating groups at
position 7 and electron-accepting groups at position 3 leads to
significant changes in their photophysics through lowering of the
CT energy state. Replacing a -CO2Et moiety with a secondary
aromatic amide triggers proton-coupled electron transfer (PCET)
and twisted intramolecular charge transfer (TICT) processes.
Excited-state properties are strongly dependent on the electron-
withdrawing character of the aryl substituents located on the
nitrogen atom. If a -CONHPhF5 group is present, a high barrier for
Mechanistic aspects of electron transfer
8
This article is protected by copyright. All rights reserved.

10.1002/chem.202001079
Chemistry - A European Journal
FULL PAPER
column chromatography (silica, DCM). Crystallization from DCM-hexane
afforded product of analytical purity. Yield: 0.260g, 79%;m.p. 207-208
C;1H NMR (500 MHz, CDCl3, 25 C, TMS): δ = 10.39 (s, 1H, NH), 8.77 (s,
1H, CH), 7.48 (d, J = 9.0 Hz, 1H, Ar), 6.72 (dd, J = 9.0, 2.5 Hz, 1H, Ar),
6.56 (d, J = 2 Hz, 1H, Ar), 3.49 (q, J = 7.0 Hz, 4H, NCH2CH3), 1.27 (t, J =
7.0 Hz, 6H, NCH2CH3); IR (KBr, cm-1): 791, 1009, 1135, 1205, 1354, 1493,
1512, 1580, 1619, 1712, 2979, 3213; HRMS (ESI-TOF): m/z calculated for
C20H15F5N2O3[M+H]+ = 449.0896; found: 449.0901 (+Na); Elemental
analysis (%):calculated for C20H15F5N2O3: C, 56.34; H, 3.55; N, 6.57;
found: C, 56.26; H, 3.38; N, 6.50.
PCET occurs in all solvents resulting in strong fluorescence and
long lifetimes in non-polar and moderately-polar solvents.
Replacing -PhF5 with the electron-donating -C6H4C6H13 causes
PCET to be the dominant process in non-polar solvents, while in
polarizable toluene and in polar solvents, the observed increase
in ΦF and τF indicate a slowdown of the PCET which points to an
increase in the barrier for the process. Replacing amides with a
phenol-derived ester linker removes the deexcitation channels
related to PCET, and fluorescence from a locally excited-state is
strong in non-polar and in weakly polar solvents.
Synthesis of H-Amide-Ph. To the solution of 7-(diethylamino)-2-oxo-2H-
chromene-3-carboxylic acid 1 (1.0 mmol) in dry DCE, EDC (1.5 mmol),
DMAP (cat. amount) and 4-n-hexylaniline 3 (1.0 mmol) were added and
the reaction mixture was stirred at room temperature for 12 h under argon.
The mixture was diluted with water and extracted with DCM. The combined
organic layers were dried over Na2SO4 and concentrated under reduced
pressure. The crude product was purified by column chromatography
(silica, DCM). Crystallized from DCM-hexane afforded product of analytical
purity. Yield: 0.380g, 83%;m.p. 168-169 C;1H NMR (500 MHz, CDCl3, 25
C, TMS): δ = 10.79 (s, 1H, NH), 8.79 (s, 1H, CH), 7.63 (d, J = 8.0 Hz, 2H,
Ar), 7.47 (d, J = 9.0 Hz, 1H, Ar), 7.16 (d, J = 8 Hz, 2H, Ar), 6.71 (dd, J =
8.0, 2.0 Hz, 1H, Ar), 6.57 (d, J = 2.0 Hz, 1H, Ar), 3.47 (q, J = 8.0 Hz, 4H,
NCH2CH3), 2.58 (t, J = 8.0 Hz, 2H, CH2(CH2)4CH3), 1.63-1.57 (m, 2H,
CH2CH2(CH2)3CH3), 1.35-1.28 (m, 6H, CH2CH2(CH2)3CH3), 1.25 (t, J = 8.0
Hz, 6H, NCH2CH3), 0.88 (t, J = 7.0 Hz, 3H, CH2(CH2)4CH3);13C NMR (125
MHz, CDCl3, 25 C, TMS): δ = 12.4, 14.1, 22.6, 28.9, 31.5, 31.7, 35.4, 45.4,
97.0, 108.9, 110.4, 110.7, 120.3, 128.8, 131.2, 135.9, 138.9, 148.3, 152.4,
157.6, 160.8, 163.0; HRMS (EI): m/z calculated for C26H32N2O3[M·+] =
420.2413; found: 420.2406.
Computational studies have revealed that the photophysics
of the investigated molecular systems are determined by
competition between the PCET and TICT processes. Both
processes are initiated by electron transfer from the excited
coumarin core to the molecular moiety substituted at position 3. If
both molecular subsystems are tightly hydrogen bonded to each
other, the CT state is populated and stabilized by ET triggering
the PCET process. If, however, the hydrogen bond does not exist,
like in Ester-Ph, or is weakened due to interaction with polar or
protic solvent, the CT state is stabilized by a mutual twist of the
electron-donating and electron-accepting moieties (TICT
process). Both photophysical processes lead to conical
intersections with the ground state and provide channels for
radiationless decay that compete with fluorescence. In all cases,
since dipole moments of charge-separated TICT and PCET
states formed in the photoreaction are much smaller than dipole
moments of the LE state they proceed through, the barrier for
these processes is sensitive to interaction with the environment.
The experimentally elucidated photophysical properties,
including the strong dependence of the fluorescence quantum
yield on solvent polarity fully corroborate these theoretical findings.
This unusual phenomenon is strongly associated with the fact that
in both TICT and PCET processes, the electronic charge shifts in
an opposite direction to that induced by an optical excitation
partially neutralizing polarization of the system, so the final states
have a particularly small dipole moment of about 4-5 D. The PCET
process is controlled by the solvent through dielectric solvation of
the highly polarized S1 state, and shows a moderate kinetic
isotope effect (< 3) that decreases at higher barriers in agreement
with the general theory developed for proton transfer as an
extension of the Marcus ET model.
Synthesis of Ester-Ph. To the solution of 7-(diethylamino)-2-oxo-2H-
chromene-3-carboxylic acid 1 (1.25 mmol) in dry DCE, 3,5-dimethylphenol
4 (1.25 mmol), EDC (1.875 mmol) and DMAP (1.25 mmol) were added.
The reaction mixture was stirred at room temperature for 12 h under argon.
The mixture was diluted with water and extracted with DCM. The combined
organic layers were dried over Na2SO4 and concentrated under reduced
pressure. The crude product was purified by column chromatography
(silica, hexane-ethyl acetate 1:1). Yield: 0.380g, 42%;m.p. 116-118 oC;1H
NMR (500 MHz, CDCl3, 25 C, TMS): δ = 8.57 (s, 1H, CH), 7.40 (d, J = 9.0
Hz, 1H, Ar), 6.87 (s, 1H, Ar), 6.84 (s, 2H, Ar), 6.65 (dd, J = 9.0, 2.5 Hz, 1H,
Ar), 6.52 (d, J = 2.5 Hz, 1H, Ar), 3.47 (q, J = 7.0 Hz, 4H, NCH2CH3), 2.33
(s, 6H, CH3), 1.25 (t, J = 7.0 Hz, 6H, NCH2CH3);13C NMR (125 MHz, CDCl3,
25 C, TMS): δ = 12.4, 21.2, 45.3, 97.0, 108.0, 108.2, 109.9, 119.4, 127.5,
131.3, 139.1, 150.0, 150.7, 153.0, 157.9, 158.7, 162.9; HRMS (EI): m/z
calculated for C22H23NO4[M·+] = 365.1627; found: 365.1634.
Synthesis of D-Amide-Ph. The H-Amide-Ph (0.24 mmol, 100 mg) was
dissolved in dry THF (20 ml) and to the mixture was added D2O (3 ml). The
reaction mixture was heated at 40 oC for 8 h under argon. Then the mixture
was cooled to room temperature, THF was evaporated and the product
was filtered.
Experimental Section
The description of methods and tools used in this study as well as detailed
results are presented in Supporting Information (synthesis, experimental
and spectroscopic methods, optical results, computational methods and
theoretical results, IR and NMR spectra, crystallographic and NMR data).
In particular, sections 5.3 ÷ 5.7 contain detailed presentation and
discussion of optical spectroscopy results for all molecules (H-Amide-Ph,
D-Amide-Ph, Amide-PhF5 and Ester-Ph) in solvents, in anhydrous
solvents and in powder.
Acknowledgements
This work was supported by the National Science Centre (Grants
PRELUDIUM UMO-2017/25/N/ST5/00227 and HARMONIA
2016/22/M/ST5/00431) The authors would like to thank the
National Science Centre under QuantERA programme (project
2017/25/Z/ST2/03038) and Global Research Laboratory Program
(2014K1A1A2064569) through the National Research Foundation
(NRF) funded by Ministry of Science, ICT & Future Planning
(Korea).
Synthesis of Amide-PhF5. 7-(Diethylamino)-2-oxo-2H-chromene-3-
carboxylic acid 1 (1.0 mmol), PyClU (1.15 mmol) and dry DCE were placed
in a dry Schlenk flask under argon. N,N-Diisopropylethylamine (3.45 mmol)
was subsequently added and the reaction mixture was stirred for 30 min
at room temperature. Then, pentafluoroaniline 2 (0.77 mmol) was added
and the mixture was heated at 85 °C for 24 h under argon. Then, the
mixture was cooled to room temperature, diluted with water and extracted
with DCM. The combined organic layers were dried over Na2SO4 and
concentrated under reduced pressure. The crude product was purified by
9
This article is protected by copyright. All rights reserved.

10.1002/chem.202001079
Chemistry - A European Journal
FULL PAPER
[32] A. L. Sobolewski, W. Domcke, J. Phys. Chem. A 2007, 111, 11725-11735.
[33] M. H. V. Huynh, T. J. Meyer, Chem. Rev. 2007, 107, 5004-5064.
[34] D. R. Weinberg,C. J. Gagliardi, J. F. Hull, C. F. Murphy, C. A. Kent, B. C.
Westlake, A. Paul, D. H. Ess, D. G. McCafferty, T. J. Meyer, Chem. Rev.
2012, 112, 4016-4093.
Keywords: absorption • chromophores • donor-acceptor
systems • fluorescence • coumarins
[1]
[2]
[3]
M. Grꢀtzel, J. Photochem. Photobiol. C 2003, 4, 145-153.
B. Daly, J. Ling, A. P. De Silva, Chem. Soc. Rev. 2015, 44, 4203-4211.
E. M. S. Stennett, M. A. Ciuba, M. Levitus, Chem. Soc. Rev. 2014, 43,
1057-1075.
[35] M. Natali, S. Campagna, F. Scandola, Chem. Soc. Rev. 2014, 43, 4005-
4018.
[36]
K. K. Mentel, A. Serra, P. E. Abreu, L. G. Arnaut, Nat. Commun. 2018,
9, 2903-2912.
[4]
[5]
H. B. Gray, J. R. Winkler, Biochim. Biophys. Acta Bioenerg. 2010, 1797,
1563-1572.
[37] C. F. Murphy, P. Dongare, S. C. Weatherly, C. J. Gagliardi, H. Holden
(a) N. V.Tkachenko, H. Lemmetyinen, J. Sonoda, K. Ohkubo, T. Sato,
H.Imahori, S.Fukuzumi, J. Phys. Chem. A 2003,107,8834-8844; (b) N. V.
Tkachenko, L. Rantala, A. Y. Tauber, J. Helaja, P. H. Hynninen, H.
Lemmetyinen, J. Am. Chem. Soc. 1999, 121, 9378-9387; (c) L.
Hammarstrꢁm, Acc. Chem. Res. 2015, 48, 840-850; (d) L. Flamigni, B.
Ventura, M. Tasior, T. Becherer, H. Langhals, D. T. Gryko, Chem. Eur.
J. 2008, 14, 169-183; (e) B. Ventura, A. Barbieri, A. Zanelli, F. Barigelletti,
J. B. Seneclauze, S. Diring, R. Ziessel, Inorg. Chem. 2009, 48, 6409-
6416; (f) L. Flamigni, B. Ventura, C.-C. You, C. Hippius, F. Würthner J.
Phys. Chem. C 2007, 111, 622-630; (g) A. Purc,E. M. Espinoza, R. Nazir,
J. J. Romero, K. Skonieczny, A. Jeżewski, J. M. Larsen, D. T. Gryko, V.
I. Vullev, J. Am. Chem. Soc., 2016, 138, 12826-12832.
Thorp, T. J. Meyer, J. Phys. Chem. C 2018122, 24830-24837.
[38]
E. Odella, S. J.Mora, B. L.Wadsworth, M. T. Huynh, J. J. Goings, P. A.
Liddell, T. L. Groy, M. Gervaldo, L. E.Sereno, D.Gust, T. A.Moore, G.
F.Moore, S.Hammes-Schiffer, A. L. Moore, J. Am. Chem. Soc. 2018, 140,
15450-15460.
[39] N. Y. Gorobets, S. A. Yermolayev, T. Gurley, A. A. Gurinov, P. M. Tolstoy,
I. G. Shenderovich, N. E. Leadbeater, J. Phys. Chem. 2012, 25, 287-295.
[40]
[41]
L. Yao, A. Grishaev, G. Cornilescu, A. Bax, J. Am. Chem. Soc. 2010,
132, 10866-10875.
N. O. C. Winter, N. K. Graf, S. Leutwyler, C. Hꢀttig, Phys. Chem. Chem.
Phys. 2013, 15, 6623-6630.
[42] R. A. Marcus, J. Phys. Chem. 1989, 93, 3078-3086.
[6]
[7]
[8]
[9]
[10]
R. A. Marcus, Angew. Chem. Int. Ed. 1993, 32, 1111-1121.
A. Weller, Z. Phys. Chem. 1982, 133, 93-98.
[43] J. Cortes, H. Heitele, J. Jortner, J. Phys. Chem. 1994, 98, 2527-2535.
[44]
M. Bixon, J. Jortner, J. W. Verhoeven, J. Am. Chem. Soc. 1994, 116,
D. Gust, T. A. Moore, A. L. Moore, Acc. Chem. Res. 1993, 26, 198-205.
H. B. Gray, J. R. Winkler, Proc. Nat. Acad. Sci. 2005, 102, 3534-3539.
D. N. Beratan, J. N. Onuchic, J. R. Winkler, H. B. Gray, Science1992,
258, 1740-1741.
7349-7355.
[45] S. Efrima, M. Bixon, Chem. Phys. 1976, 13, 447-460.
[46]
A. Kapturkiewicz, J. Herbich, J. Karpiuk, J. Nowacki, J. Phys. Chem.
1997, 101, 2332-2344.
[11] G. Jones II, L. N. Lu, V. Vullev, D. J. Gosztola, S. R. Greenfield, M. R.
[47] R. A. Marcus, N. Sutin, Biochim. Biophys. Acta 1985, 811, 265-322.
[48] C. M. Krauter, J. Mohring, T. Buckup, M. Pernpointner, M. Motzkus, Phys.
Chem. Chem. Phys. 2013, 15, 17846-17861.
Wasielewski, Bioorg. Med. Chem. Lett. 1995, 5, 2385-2390.
[12]
C. Shih, A. K. Museth, M. Abrahamsson, A. M. Blanco-Rodriguez, A.
J. Di Bilio, J. Sudhamsu, B. R. Crane, K. L. Ronayne, M. Towrie, A. Vlcek
Jr., J. H. Richards, J. R. Winkler, H. B. Gray, Science 2008, 320, 1760-
1762.
[49] D. Murdock, R. A. Ingle, I. V. Sazanovich, I. P. Clark, Y. Harabuchi, T.
Taketsugu, S. Maeda, A. J. Orr-Ewing, M. N. R. Ashfold, Phys. Chem.
Chem. Phys. 2016, 18, 2629-2638.
[13] M. Krzeszewski, E. M. Espinoza, C. Červinka, J. B. Derr, J. A. Clark, D.
Borchardt, G. J. O. Beran, D. T. Gryko, V. I. Vullev, Angew. Chem. Int.
Ed. 2018, 57, 12365-12369.
[50] G. Jones II, W. R. Jackson, Ch. Y. Choi, W. R. Bergmark, J. Phys. Chem.
1985, 89, 294-300.
[51] R. I. Cukier, D. G. Nocera, Annu. Rev. Phys. Chem. 1998, 49, 337-369.
[52] A. Soudackov, S. Hammes-Schiffer, J. Chem. Phys. 2000, 113, 2385-
2396.
[14] K. Rotkiewicz, K. H. Grellmann, Z. R. Grabowski, Chem. Phys. Lett. 1973,
19, 315-318.
[15] Z. R. Grabowski, K. Rotkiewicz, W. Rettig, Chem. Rev. 2003, 103, 3899-
4031.
[53] (a) A. Soudackov, E. Hatcher, S. Hammes-Schiffer, J. Chem. Phys. 2005,
122, 014505; (b) G. A. Parada, Z. G. Goldsmith, S. Kolmar, B. Petterson-
Rimgard, B. Q. Mercado, L. Hammarstrꢁm, S. Hammes-Schiffer, J. M.
Mayer, Science 2019, 364, 471-475.
[16] A. Weller, Z. Electrochem. 1956, 60, 1144-1147.
[17] A. J. Stasyuk, P. Cywiński, D. T. Gryko, J. Photochem. Photobiol. C:
Reviews 2016, 28,116-137.
[54] S. J. Edwards, A. V. Soudackov, S. Hammes-Schiffer, J. Phys. Chem. A
2009, 113, 2117-2126.
[18] J. E. Kwon, S. Y. Park, Adv. Mat. 2011, 23, 3615-3642.
[19]
J. Zhao, S. Ji, Y. Chen, H. Guo, P. Yang, Phys. Chem. Chem. Phys.
2012, 14,8803-8817.
[55] A. Weller, Pure Appl. Chem. 1968, 16, 115-124.
[56] B. S. Brunschwig, S. Ehrenson, N. Sutin, J. Phys. Chem. 1987, 91, 4714-
4723.
[20] P. Zhou, K. Han, Acc. Chem. Res. 2018, 51, 1681-1690.
[21]
A.Brenlla, M.Veiga, J. L.Pérez Lustres, M. C.Ríos Rodríguez,
[57] H.-Y. Wu, H.-S. Ren, Q. Zhu, X.-Y. Li, Phys. Chem. Chem. Phys. 2012,
14, 5538-5544.
F.Rodríguez-Prieto, M. Mosquera, J. Phys. Chem. B 2013, 117, 884-896.
[22] H. Shizuka, K. Matsui, T. Okamura, I. A. Tanaka, J. Phys. Chem. 1975,
79, 2731-2734.
[58] G. F. Manbeck, E. Fujita, J. J. Concepcion, J. Am. Chem. Soc. 2016, 138,
11536−11549.
[23] A. L. Sobolewski, W. Domcke, Phys. Chem. Chem. Phys. 2006, 8, 3410-
3417.
[59] O. Gottmann, U. Stumper, Chem. Phys. Lett. 1973, 22, 387-389.
[24] T. Schultz, E. Samoilova, W. Radloff, I. V. Hertel, A. L. Sobolewski, W.
Domcke, Science2004, 306, 1765-1768.
[25] A. L. Sobolewski, W. Domcke, Chem. Phys. Chem. 2006, 7, 561-564.
[26]
B. Valeur, M. N. Berberan-Santos, Molecular Fluorescence Principles
and Applications, 2nd ed., Wiley-VCH Verlag & Co. KGaA, 2012, pp. 380-
384.
[27] F. Bures, RSC Adv. 2014, 4, 58826-58851.
[28] A. M. Moran, D. S. Egolf, M. Blanchard-Desce, A. M. Kelley, J. Chem.
Phys.2002, 116, 2542-2555.
[29] O. Morawski, B. Kozankiewicz, A. Miniewicz, A. L. Sobolewski, Chem.
Phys. Chem. 2015, 16, 3500-3510.
[30] Ł. Kielesiński, O. W. Morawski, Ł. Dobrzycki, A. L. Sobolewski, D. T. Gryko,
Chem. Eur. J. 2017, 23, 9174-9184.
[31]
Ł. Kielesiński, D. T. Gryko, A. L. Sobolewski, O. W. Morawski, Phys.
Chem. Chem. Phys. 2018, 20, 14491-14503.
10
This article is protected by copyright. All rights reserved.

10.1002/chem.202001079
Chemistry - A European Journal
FULL PAPER
Entry for the Table of Contents
((max. width: 5.5 c; max.
height: 5.0 cm))
Please delete this box prior to
submission
A solvent-modulated competition between PCET and TICT processes, that both dump fluorescence of the LE state, occurs in amido-
coumarins. This stems from the combination of the following facts: (1) ET occurring in excited state neutralizes the charge distribution
resulting in CT state possessing low dipole moment; (2) CT state is stabilized either by proton transfer of by mutual twist of electron-
donating and electron-accepting molecular moieties.
Institute and/or researcher Twitter usernames: ((optional))
11
This article is protected by copyright. All rights reserved.