The effect of the rate of photoinduced electron transfer on the photodecarboxylation efficiency in phthalimide photochemistry

Article abstract of DOI:10.1016/j.jphotochem.2020.113109

Reactivity in photoinduced electron transfer reactions (PET) has been investigated in a series of molecules possessing different distances between the electron donor (carboxylate or alkoxyphenyl) and the phthalimide as the electron acceptor. The molecules

Full text of DOI:10.1016/j.jphotochem.2020.113109

Journal of Photochemistry & Photobiology, A: Chemistry 408 (2021) 113109
Contents lists available at ScienceDirect
Journal of Photochemistry & Photobiology, A: Chemistry
journal homepage: www.elsevier.com/locate/jphotochem
The effect of the rate of photoinduced electron transfer on the
photodecarboxylation efficiency in phthalimide photochemistry
a,
b
a
b
c
a,
´
´
´
´
´
´
Leo Mandic , Margareta Sohora , Branka Mihaljevic , Laszlo Biczok , Nikola Basaric *
a
ˇ
´
ˇ
Department of Organic Chemistry and Biochemistry, Ruđer Boskovic Institute, Bijenicka cesta 54, 10 000 Zagreb, Croatia
b
ˇ
´
ˇ
Department of Material Chemistry, Ruđer Boskovic Institute, Bijenicka cesta 54, 10 000 Zagreb, Croatia
c Institute of Materials and Environmental Chemistry, Research Centre for Natural Sciences, 1519 Budapest, P.O. Box 286, Hungary
A R T I C L E I N F O
A B S T R A C T
Keywords:
Reactivity in photoinduced electron transfer reactions (PET) has been investigated in a series of molecules
possessing different distances between the electron donor (carboxylate or alkoxyphenyl) and the phthalimide as
the electron acceptor. The molecules were strategically designed to separate the donor and the acceptor by a
rigid adamantane spacer or through a peptide backbone with different number of amino acid residues. PET was
investigated by laser flash photolysis (LFP). Although previous reports demonstrated that the quantum yields of
the photodecarboxylation reaction (ΦR) depend on the distance between the donor and acceptor moieties, the
measured lifetimes of the triplet excited state did not provide information on the rates of PET. Due to reversible
PET and back electron transfer processes in compounds 1-5, the measured lifetimes correspond to the sum of the
rate constants for the disappearance of the charge transfer species. In the series of peptides, three different PET
processes take place and LFP provided rate constant for the reactions occurring subsequent to the intrastrand PET
between the carboxylate and the alkoxyphenyl radical cation. The understanding of the factors that affect
intramolecular PET processes is important for the rational design of different applications.
Laser flash photolysis
Peptides
Photoinduced
Electron transfer
Phthalimides
1. Introduction
Inspired by applications of phthalimide photochemistry in organic
synthesis, [25] we encountered photodecarboxylation reactions initi-
Photoinduced electron transfer (PET) is a fundamental reaction in
chemistry with numerous applications. [1] Due to its vital importance
and versatile potential utilizations, PET in DNA [2] and proteins [3] has
been intensively investigated and several times reviewed. [4] PET pro-
cesses were also used in organic synthesis for the preparation of complex
molecular structures and natural products. [5] Moreover, recent ad-
vances in photocatalysis largely include PET processes.[6]
ated by PET, which involved phthalimide chromophore. [26–28] The
PET phthalimide photochemistry was applied in the cyclizations with
memory of chirality [29] and diastereoselective peptide cyclizations.
[30] We studied photodecarboxylation efficiency in a series of phthali-
mide derivatives of adamantane amino acids 1-5 (Chart 1 ), where we
modified the distance between the electron donor (carboxylate) and the
acceptor (phthalimide). [26] Furthermore, decarboxylation reactions
were investigated in a series of peptides possessing a phthalimide moiety
(electron acceptor) at the N-terminus and different reducing amino acid
at the C-terminus. [30] A preliminary investigation of compound 4 by
laser flash photolysis (LFP) indicated that PET was followed by back
electron transfer, which was in competition with chemical reactions of
the phthalimide radical anion or carboxyl radical, determining the
overall efficiency of the decarboxylation process. [31] Herein we pre-
sent a thorough LFP study of the triplet excited state and the PET
reactivity for compounds with different distance between the electron
donor and acceptor moieties 1-10 to probe if the rate of PET determines
the overall photodecarboxylation efficiency.
Phthalimide is a chromophore undergoing a rich array of photo-
chemical reactions, [7,8] including PET. [9] Phthalimide derivatives in
the triplet excited state undergo PET, reacting with donors having
oxidation potential < 1.6 V vs. saturated calomel electrode. [10]
Therefore, phthalimide PET photochemistry has been used in organic
synthesis [11] to initiate elimination of silyl groups, [12] or photo-
decarboxylation. [9,10] These PET processes further enabled macro-
cyclizations, [13] photocyclization of peptides, [14, 15] photodecaging,
[16] enantioselective synthesis of benzodiazepines, [17] and cyclic aryl
ethers, [18] as well as acetate, [19,20] benzyl, [21–23] or α-amino acid
additions to phthalimides. [24].
* Corresponding author.
´
E-mail address: nbasaric@irb.hr (N. Basaric).
https://doi.org/10.1016/j.jphotochem.2020.113109
Received 13 September 2020; Received in revised form 6 December 2020; Accepted 21 December 2020
Available online 28 December 2020
1010-6030/© 2020 Elsevier B.V. All rights reserved.

´
L. Mandic et al.
Journal of Photochemistry & Photobiology, A: Chemistry 408 (2021) 113109
Chart 1. Investigated adamantylphthalimides and phthalimide peptides.
state population (for N-methylphthalimide, ΦISC = 0.8). [32] Lifetimes
2. Materials and methods
of the lowest-energy triplet excited states (τ) in CH3CN and relative ef-
Synthesis of compounds 1-10 has been described in literature pre-
cedent. [26,30] Compounds were dissolved in CH3CN or CH3CN-H2O
(1:1) in the presence of potassium phosphate buffer (the total phosphate
concentration 50 mM) at pH 3.8 (later denoted as pH 4) or 7.0. The
solutions were optically matched. The absorbances at the excitation
wavelength (λ =266 nm) were A266 = 0.30, which corresponded to the
concentrations of 0.13ꢀ 0.66 mM.
ficiencies of the intersystem crossing (ISC), obtained by comparison with
N-methylphthalimide, are reported in Table 1.
In CH3CN solution, the triplet excited state lifetimes of 1-5 are not
significantly different, with values of ≈ 1ꢀ 4 μs. However, the relative
efficiencies of ISC differ by more than an order of magnitude between 1
and 4, all of them being significantly lower than the ΦISC for N-meth-
ylphthalimide. ΦISC markedly increases when the carboxyl group is
moved from α, to β and to γ carbon atom relative to the imide nitrogen.
This finding suggests that the change of the relative orientation and
distance between the phthalimide chromophore and the carboxylic
functional groups affects the relative rate of nonradiative deactivation
from the singlet excited state, and therefore, alters the relative efficiency
of the ISC. A similar finding has been documented for N-alkylph-
thalimides when the change of N-substituent from methyl to i-propyl
and t-butyl decreased the ΦISC by an order of magnitude. [25]
3. Laser flash photolysis
The measurements were performed on a LP980 Edinburgh In-
struments spectrometer at 25 ◦C. The fourth harmonic of a Qsmart Q450
Quantel YAG laser was used for excitation. The pulse duration was 7 ns.
For the measurement of spectra, the energy of the laser pulse at 266 nm
was set to 15 or 20 mJ. For the monitoring of decay kinetics, it was set to
8 mJ to avoid two-photon processes and triplet-triplet annihilation. The
spectra were measured in flow cells, whereas the decay kinetics were
recorded in static cells, in which the solutions were frequently
exchanged to avoid potential problems originating from the formation of
photoproducts. The solutions were purged for 20 min with N2 or O2 prior
to the measurements.
It has been documented that carboxylates can be oxidized with
4. Results and discussion
4.1. Adamantylphthalimides
Irradiation of 1-5, and characterization of photoproducts have been
reported. [26] The photochemical reactions give simple decarboxylation
products, and in the case of 2 and 3, polycyclic molecules (Scheme 1).
Details on isolated yields can be found in literature precedent, [26]
whereas the reaction quantum yields are compiled in Table 2, together
with the LFP data (vide infra).
LFP experiments for compounds 1-5 were first conducted in CH3CN
solutions, where the characteristic spectra due to the triplet-triplet ab-
sorption of the phthalimide with a maximum at 330 nm were detected
(for 3 see Fig. 1 and for all LFP data see S1-S16 in the SI). Comparison of
the intensity of the transient absorption with the one of N-methyl-
phthalimide for the solutions optically matched at the excitation
wavelength provided approximation for the quantum yield of the triplet
Fig. 1. Transient absorption spectra of adamantylphthalimide 3 in N2-purged
CH3CN-H2O (1:1 v/v) in the presence of potassium phosphate buffer (50 mM) at
pH 4 and 7 collected 150 ns after the laser pulse (15 mJ/pulse). The spectrum
recorded in CH3CN is also displayed for the sake of comparison. The solutions
were optically matched at the excitation wavelength (A266 = 0.3).
Scheme 1. Photochemical reactions of 1-5 delivering simple decarboxylation products and polycyclic molecules.
2

´
L. Mandic et al.
Journal of Photochemistry & Photobiology, A: Chemistry 408 (2021) 113109
transient spectra at pH 7 does not reveal that one species is transformed
into another, both bands decay with the same kinetics. Furthermore, the
apparent rate constant for the quenching of the triplet excited state
(kobs = 1/τpH7ꢀ 1/τpH4) has values of 5 × 105 - 5 × 106 Mꢀ 1 sꢀ 1. Note
that the kobs rate constants do not depend on the distance between the
electron donor and the acceptor, and have too small values to be
correlated with PET. For example, the reported rate constants of intra-
Table 1
Data obtained by LFP of adamantylphthalimides 1-5 in CH3CN.
s a
ΦISC
b
τ
/
μ
1
3.3 ± 0.3
4.0 ± 0.2
1.9 ± 0.1
1.2 c
0.016 ± 0.004
0.08 ± 0.02
0.11 ± 0.03
0.22 d
2
3
4
5E
0.7 ± 0.1
0.06 ± 0.02
molecular PET in donor-acceptor systems separated by
σ bonds at the
distance of ≈ 7 Å have values > 1011 sꢀ 1. [34] On the other hand, the
largest distance between the donor and the acceptor in molecules 1-5 is
≈ 5.9 Å in 5E. Furthermore, the reported rate constant for PET in the
lowest singlet excited state (S1) between carboxylate and the phthali-
mide substituted with electron donating substituents OCH3 and NH2 for
the adamantyl derivative such as 4, have values of (2.0 ± 0.1) × 109 sꢀ 1
and (3.4 ± 1.0) × 107 sꢀ 1, respectively. [31] Therefore, it is plausible
that both rates for electron transfer, kET and k-ET (Scheme 2), have large
values, leading to the rapid equilibration of 5E (T1) and the corre-
sponding CT species. Since the spectra of the phthalimide triplet excited
state and the phthalimide radical anion have a significant overlap, the
observed transient absorption corresponds to both species and their
decay is primarily governed by the sum of rate constants leading to the
slow depopulation of the CT species (Σki). A similar kinetic scheme
involving fast equilibration between an excited state and an exciplex
followed by slow reactions from the exciplex has been reported, [35]
and the rate constant for the deactivation of the CT species in that case is
given by k = ket/k-etΣki.
a
Lifetime of the triplet excited state in the N2-purged solution. At least three
decays were detected at 330, 340 and 350 nm, and the average value was re-
ported. The quoted errors correspond to the maximum absolute deviation.
b
Relative efficiency of the triplet excited state population obtained by com-
parison of the transient absorption immediately after the laser pulse (ΔA) with
the one measured for the optically matched N-methylphthalimide solution
(ΦISC = 0.8). [32] (The absorbances at 266 nm were A266 = 0.3.).
c
Value taken from ref. [31].
d
Value taken from ref. [28].
triplet excited phthalimide, whereas carboxylic acids cannot,[9,10,21,
26,31] since carboxylates are better electron-donating groups than the
carboxylic acids (for example, the oxidation potentials in CH3CN are for
10-undecenoate Eox ≈ 1.38 V vs. Ag/AgCl, and for 10-undecenoic acid
Eox ≈ 1.85 V vs. Ag/AgCl). [33] To probe the PET reactivity of molecules
1-5, we performed LFP measurements in aqueous solution (CH3CN-H2O
1:1 v/v) in the presence of potassium phosphate buffer (50 mM) at pH 4
and 7, where the carboxylic functional group at the adamantane is
protonated, or not (pKa = 5.56 ± 0.01). [31] The transient absorption
spectra at pH 4 (Fig. 1) resemble those measured in CH3CN showing the
typical maximum at 330 nm, assigned to the phthalimide triplet excited
state. The decay of the transient absorption was fit to single exponential,
providing similar lifetimes of the triplet excited states (Table 2). The
triplet can be quenched by O2, and the rate constants for the quenching
are somewhat lower than the diffusion limit (0.8–2.0 × 109 Mꢀ 1 sꢀ 1), as
reported previously for N-alkylphthalimides. [25]
The decay of transient for compounds 3 and 4 at pH 7 was fit to a sum
of two exponentials. The faster decay was assigned to the triplet excited
state in equilibrium with the CT species. The origin of the slow decay
probably corresponds to the phthalimide ketyl radical formed by pro-
tonation of the radical anion, which absorbs in the same wavelength
region and has the lifetime of ≈ 50ꢀ 100 μs. [31] Such a long-lived
transient has not been revealed for substrates 1, 2 and 5 due to very
weak intensity of the transient absorption at long timescale.
At pH 7, the adamantane carboxylic acid is deprotonated allowing
PET from the carboxylate to the phthalimide to take place in the triplet
excited state. Transient absorption spectra measured in aqueous solution
at pH 7 have two absorption maxima at 320 and ≈400 nm (Fig. 1). The
absorbance decays were fit to one (compounds 1, 2 and 5) or sum of two
exponential functions (compounds 3 and 4). Short-lived transients
Particularly weak transient absorption was detected for 1, the com-
pound having a very low quantum yield of triplet formation (ΦISC). It is
known that phthalimide derivatives of α-amino acids undergo proton-
coupled electron transfer and decarboxylation in aprotic solvents and
deliver azomethine ylides, which can be trapped in [3 + 2] cycloaddi-
tions. [36] Indeed, such a process was taken into account for the CH3CN
solution of 1. [37] However, quantum yield for the decarboxylation (ΦR)
of 1 is an order of magnitude lower compared to those measured for 2-5,
in agreement with the low ΦISC (Tables 1 and 2). In any case, in aqueous
solution at pH 7, phthalimide 1 is present as carboxylate, which should
enable PET. However, the low intensity of the transient absorption did
(0.2ꢀ 3 μs) were detected for all compounds, which were tentatively
assigned to the phthalimide triplet excited state. Note that the decay is
faster at pH 7, in line with the anticipated quenching of the triplet
excited state by PET. It is plausible to assume that the long-lived tran-
sient corresponds to the phthalimide radical anion. Inspection of the
Table 2
Data obtained by LFP of adamantylphthalimides 1-5 in CH3CN-H2O (1:1 v/v) mixture and quantum yields for the decarboxylation reaction (ΦR).
sa
kO2 / 109 Mꢀ 1 s-1 b
τpH 7
/
μ
s c
kobs / 106 Mꢀ 1 s-1 d
ΦR
e
τpH 4
/
μ
1
2
3
3.3 ± 0.3
2.8 ± 0.2
2.7 ± 0.3
2.0 ± 0.6
3.3 ± 0.7
–
0.026 ± 0.002
0.41 ± 0.03
0.21 ± 0.02
0.78 ± 0.01
0.83 ± 0.01
1.2 ± 0.2
0.48 ± 0.02
0.2 ± 0.1
4 ± 2
(10.5 ± 0.7)
0.34 ± 0.04
(150 ± 50)
1.6 ± 0.4
4
1.8 ± 0.4
0.93 ± 0.05
2.5 ± 0.5
0.11
5Z
5E
2.7 ± 0.1
2.0 ± 0.2
0.93 ± 0.03
1.52 ± 0.07
0.26 ± 0.04
0.4 ± 0.1
0.19 ± 0.01
0.09 ± 0.01
1.2 ± 0.1
a
Lifetime of the triplet excited state in N2-purged solution in the presence of phosphate buffer (50 mM) at pH 3.8. Measurements were done in triplicate and the
average values are reported. The quoted errors correspond to the maximum absolute deviations.
b
c
Rate constant for the quenching of the triplet excited state by O2 in the presence of phosphate buffer (50 mM) at pH 3.8.
Lifetime of the triplet excited state in N2-purged solution in the presence of phosphate buffer (50 mM) at pH 7.0. The long-lived lifetime reported in parenthesis is
tentatively assigned to the phthalimide radical anion or the ketyl radical.
d
1
1
Apparent rate constant for the quenching of the triplet excited state by the increase of basicity (kobs
=
ꢀ
) in the presence of phosphate buffer (50 mM).
τpH7 τpH4
The quoted errors correspond to the maximum absolute deviations.
e
Quantum yields for the decarboxylation reaction correspond to the disappearance of phthalimide 1-5 upon irradiation in CH3CN-H2O (2:1) in the presence of K2CO3
upon excitation at 254 nm, taken from ref. [26].
3

´
L. Mandic et al.
Journal of Photochemistry & Photobiology, A: Chemistry 408 (2021) 113109
Scheme 2. Reaction mechanism for PET promoted decarboxylation in 5E.
not permit the determination of kobs for 1.
Table 3
In the case of photochemistry of 1-5, we propose that the major
chemical deactivation routes for the disappearance of the triplet excited
state and the CT species are decarboxylation and protonation of the
phthalimide radical anion. Although the pKa for phthalimide radical
anion is not known (probably > 7), it is generally accepted that the
radical anion is basic and prone to protonation in neutral solution.
[9–12,31] Thus, the observed difference in the triplet lifetimes at two
different pH values, when PET can take place or not, probably corre-
sponds to kobs = ket/k-etΣki = ket/k-et(kH+[H+] + k-CO2 + k-et2).
Data obtained by LFP of phthalimide peptides 6-10 in CH3CN and CH3CN-H2O
(1:1 v/v) mixture.
s b
kO2 / 109
τpH 7
/
μ
s d
ΦR
e
τCH3CN
/
μs
τpH 4
/
μ
a
Mꢀ 1 s-1 c
6
0.15 ± 0.03
2.9 ± 0.2
55 ± 5
1.3 ± 0.2
0.67 ± 0.04
1.4 ± 0.2
2.7 ± 0.4
1.4 ± 0.2
0.16 ± 0.05
2.4 ± 0.3
55 ± 2
0.050 ± 0.005
7
0.19 ± 0.04
2.9 ± 0.4
32 ± 9
0.17 ± 0.03
2.5 ± 0.4
24.1 ± 0.5
0.24 ± 0.01
3.9 ± 0.2
100 ± 10
0.24 ± 0.05
2.6 ± 0.7
70 ± 20
0.057 ± 0.007
0.014 ± 0.003
0.027 ± 0.003
0.020 ± 0.006
The reported photodecarboxylation efficiencies for 1-4 in acetone
sensitized conditions have similar values ΦR = 0.4ꢀ 0.5, whereas it is
significantly lower for 5E (ΦR = 0.023) and for 5Z (ΦR = 0.15). [26] It is
interesting to put these values in connection with the measured rate
constants for the quenching of the triplet excited state (kobs). The
apparent quenching constants for 2 and 3 differ by an order of magni-
tude, but the efficiencies for the decarboxylation upon excitation at
254 nm have similar values and an opposite trend (Table 2). On the
contrary, the quenching rate constant has similar values for 5E and 5Z,
whereas the decarboxylation efficiency differs by an order of magnitude.
These findings suggest that ratio of ket and k-et probably determines the
efficiency of the whole decarboxylation process, and not the sum
8
0.21 ± 0.06
2.6 ± 0.7
240 ± 40
0.32 ± 0.04
3.4 ± 0.5
33 ± 10
9
0.31 ± 0.04
2.5 ± 0.3
30 ± 10
0.5 ± 0.1
6 ± 1
10
0.39 ± 0.04
4.1 ± 0.2
120 ± 10
0.33 ± 0.04
4.64 ± 0.02
50 ± 7
29 ± 6
a
Decay times in N2-purged CH3CN solution. All measurements were done in
triplicate and the average values are reported. The quoted errors correspond to
the maximum absolute deviations.
b
Lifetime of the triplet excited state in N2-purged CH3CN-H2O (1:1 v/v) in the
k
H+[H+] + k-CO2 + k-et2
.
presence of phosphate buffer 50 mM at pH 3.8.
c
Rate constant for the quenching of the triplet excited state (lifetime values ≈
2ꢀ 4 μs) by O2 in CH3CN-H2O (1:1 v/v) in the presence of phosphate buffer
4.2. Phthalimide peptides
50 mM at pH 3.8.
d
Decay times in N2-purged CH3CN-H2O (1:1 v/v) in the presence of phos-
Irradiation of peptides 6-10 gives rise to cyclic peptides (Scheme 3),
[30] where the photochemical reaction mechanism involves PET and
decarboxylation to produce radicals which cyclize, as reported in liter-
ature precedent. [14,15] Isolated yields of cyclic peptides were reported
[30] and the quantum yields for the disappearance of peptides are given
in Table 3.
phate buffer 50 mM at pH 7.0.
e
Quantum yields for the decarboxylation reaction correspond to the disap-
pearance of peptides 6-10 upon irradiation in CH3CN-H2O (1:2) in the presence
of potassium phosphate buffer 50 mM at pH 7.0 upon excitation at 254 nm,
taken from ref. 30.
Peptides 6-10 are multichromophoric systems, and the light of
Scheme 3. Photochemical decarboxylative cyclization of peptides 6-10.
4

´
L. Mandic et al.
Journal of Photochemistry & Photobiology, A: Chemistry 408 (2021) 113109
Fig. 2. Transient absorption spectra of phthalimide peptides 6-10 in N2-purged CH3CN-H2O (1:1 v/v) in the presence of potassium phosphate buffer (50 mM) at pH 4
(left) and at pH 7 (right) collected with the delay of 10 ns after the laser pulse (20 mJ/pulse). The solutions were optically matched at the excitation wave-
length (A266 = 0.3).
266 nm wavelength, used in the LFP experiments, excites phthalimide,
as well as amino acids, Phe, TyrOCH3, or Tyr(OCH3)2. In particular, the
TyrOCH3 or Tyr(OCH3)2 are more absorptive compared to Phe, leading
to more efficient excitation of this amino acid residues, compared to the
phthalimide. Preliminary LFP measurements for phthalimide peptides 6-
10 were conducted in CH3CN, [30] and here we report LFP data in
CH3CN-H2O 1:1 v/v mixture at pH 4 and 7 when the carboxylic func-
tional group is protonated or not, to probe for PET (Fig. 2, Table 3 and
Figs. S17-S31 in the SI). In addition, for 9 and 10 we compare data
measured in CH3CN with those in aqueous solution at both pH values.
Transient absorption spectra for peptides 6 and 7 in CH3CN-H2O (1:1
v/v) solution at both pH values resemble those measured in CH3CN, [30]
or those of simple adamantylphthalimides 1-5 at pH 4, with the char-
acteristic maximum of absorption at 330 nm corresponding to the
phthalimide triplet excited state. Due to low molar absorptivity of Phe,
light of 266 nm wavelength excites primarily the phthalimide moiety of
these peptides, and after a rather inefficient ISC their triplet excited
states become populated (ΦISC = 0.05-0.06). [30] An energy transfer
from the phthalimide to Phe is energetically impossible since the triplet
energy of alkylphthalimides such as N-methylphthalimide, ET =297 kJ
molꢀ 1 is lower compared to the energy of Phe, ET =346 kJ molꢀ 1 (or
modified tyrosines ET =342 kJ molꢀ 1). [38] Furthermore, the charac-
teristic triplet-triplet absorption of Phe has a maximum at somewhat
[39] PET between phthalimide and Phe cannot occur due to the too high
oxidation potential of the peptide-bound Phe. Thus, the observed band
in the transient absorption spectra can confidently be assigned to the
phthalimide triplet. However, the decay was fit to a sum of three ex-
ponentials, where the decay time of 2ꢀ 3 μs most likely corresponds to
the phthalimide triplet, based on the quenching by O2 (Table 3). At pH 4,
where the PET is not anticipated, the long-lived species (30ꢀ 50 s)
μ
probably correspond to the phthalimide ketyl radicals, based on com-
parison with literature precedent. [31] The assignment of the short-lived
species is beyond the scope of this manuscript. Therefore, the proposed
assignments of two long-lived transients for peptides 6 and 7 at pH 4 are
the phthalimide triplet and the phthalimide ketyl radical. Increase of the
solution pH to 7 did not cause a significant difference in the transient
absorption spectra, except the fact that the signal of the transient ab-
sorption immediately after the laser pulse was weaker at pH 7 (Fig S29),
which may be due to a fast PET reactivity in the singlet excited state
giving phthalimide radical anion. Due to the subsequent back electron
transfer and protonation (Scheme 4), and the low efficiency of the whole
process (see Table 2, non-sensitized ΦR ≈ 0.05), [30] the phthalimide
radical anion was not detected. The plausible explanation for the lack of
detection of phthalimide radical anion and the low efficiency of the
decarboxylation process may be PET reaction taking place from the
singlet excited state, and fast subsequent back electron transfer which
shorter wavelength (at 310 nm), and the lifetime of the triplet is ≈ 3
μ
s.
repopulates the starting peptides 6 or 7 in the ground state.
Scheme 4. Reaction mechanism for PET promoted decarboxylation in 6 and 7.
5

´
L. Mandic et al.
Journal of Photochemistry & Photobiology, A: Chemistry 408 (2021) 113109
Scheme 5. Reaction mechanism for PET promoted decarboxylation in 10.
Consequently, for peptides 6 and 7 population of triplets is not efficient
and LFP did not reveal data about the PET reactivity.
significantly lower for tetrapeptide 10 (in acetone ΦR = 0.08). [30] On
the other hand, from the lifetimes of the CT species at pH 4 and 7 for
peptides 8 and 10, the apparent quenching rate constants are 5.8 × 103
Peptides 8-10 contain three chromophores, phthalimide, Phe and
TyrOCH3, or Tyr(OCH3)2, and contrary to 6 and 7, the light of 266 nm
wavelength excites both phthalimide and TyrOCH3, or Tyr(OCH3)2
moieties. Thus, in these peptides, PET may be initiated from the S1 state
of TyrOCH3, or Tyr(OCH3)2, and not just from the excited state of
phthalimide (Scheme 5). Indeed, the transient absorption spectra of 8-
10 at pH 4, as well as spectra in CH3CN, have two absorption maxima, at
300 and 400 nm, which can be attributed to the CT species formed by
initial PET. The assignment is based on the published spectra in litera-
ture precedent; the methoxybenzene radical cations absorb at
400ꢀ 430 nm, [40] as well as phthalimide radical anion. [25] For pep-
tides 8-10, PET can take place in CH3CN or in aqueous solution at pH 4
since it involves phthalimide and TyrOCH3, or Tyr(OCH3)2, and not the
carboxylate. Note that the spectra have the same shape over time,
indicating the lack of transformation from one species into another
(similar to 1-5). However, the decay of the transient absorption was fit
to a sum of three exponentials (Table 3). The tentative assignments of
the transients are phthalimide triplet, which is quenched or not, whereas
the long-lived species is probably the CT species. However, the
long-lived transient may also correspond to the phthalimide ketyl
radical. Note that the long-lived species for 8 and 10 (but not for 9) have
shorter decay times at pH 7 when the carboxylic functional group is
deprotonated. This quenching may originate from the equilibration of
the CT species due to intrastrand ET between the carboxylate and the
radical cation, followed by decarboxylation. The intrastrand ET in
phthalimide peptides has been reported by Mariano et al. (correspond-
ing to the rate constant kiset in Scheme 5), [30,41] and it is in accordance
with the well-accepted mechanism of PET in peptides involving super-
exchange over short distances and energy hopping over larger. [3,4] On
the other hand, the difference between peptides 8 and 9, where the CT
decay time has different dependence on pH may be due to different
conformational mobility. We have demonstrated different molecular
dynamics of tetra- and pentapeptides, resulting in cyclic peptides with
different stereochemistry. [30] Thus, different orientation between the
phthalimide and the TyrOCH3 in tetrapeptide and pentapeptide facili-
tates the kiset or k-et, respectively (Scheme 5). However, the efficiency for
the decarboxylation and cyclization of tetrapeptide 8 and pentapeptide
9 has similar values (in acetone ΦR = 0.3-0.4), [30] whereas it is
sꢀ 1 and 1.2 × 104
s
ꢀ 1. It is evident that these rate constants do not
correspond to kiset, but rather to the rate constants for subsequent re-
actions (tentatively, kH+[H+] + k-CO2). Consequently, different effi-
ciency for the decarboxylation in phthalimide peptides is most probably
related to the ratio of kiset and k-et or k-et2, and not to the actual decar-
boxylation, or primary PET between two amino acid residues.
5. Conclusion
LFP experiments for adamantylphthalimides and phthalimide pep-
tides with different distances between the electron donors (carboxylate)
and acceptor (phthalimide) unraveled the properties of their corre-
sponding triplet excited states. Although, the LFP data did not provide
direct rate constants for the PET taking place from the triplet excited
states, combination of photochemical reaction efficiencies and LFP data
indirectly point to the fact that overall photodecarboxylation processes
in these systems do not depend on the intrinsic rate of PET, or on the
rates of subsequent reactions of decarboxylation or phthalimide radical
anion protonation. Due to reversibility of PET and back electron transfer
in these systems, the overall efficiency is determined by the ratio of ket
and k-et or kiset and k-et. This fact has to be taken into account when
designing different donor-acceptor phthalimide systems for the appli-
cation in organic synthesis.
Declaration of Competing Interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
Acknowledgements
These materials are based on work financed by the Croatian Science
Foundation (HRZZ IP-2014-09-6312 and HrZZ-IP-2019-04-8008 for
NB). LB acknowledges the support of the National Research, Develop-
ment and Innovation Office of Hungary (Grant K123995).
6

´
L. Mandic et al.
Journal of Photochemistry & Photobiology, A: Chemistry 408 (2021) 113109
¨
[18] Y.-J. Lee, D.-H. Ahn, K.-S. Lee, A.R. Kim, D.J. Yoo, M. Oelgemoller, Photoinduced
Appendix A. Supplementary data
electron transfer cyclizations of aryl-linked phthalimides, Tetrahedron Lett. 52
(2011) 5029–5031.
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.jphotochem.2020.
113109.
¨
[19] F. Hatoum, S. Gallagher, M. Oelgemoller, Photodecarboxylative additions of
phenoxyacetates to N-methylphthalimide, Tetrahedron Lett. 50 (2009) 6593–6596.
¨
[20] F. Hatoum, J. Engler, C. Zelmer, J. Wißen, C.A. Motti, J. Lex, M. Oelgemoller,
Photodecarboxylative addition of carboxylates to phthalimides: A concise access to
biologically active 3-(alkyl and aryl)methylene-1H-isoindolin-1-ones, Tetrahedron
Lett. 53 (2012) 5573–5577.
References
¨
[21] A.G. Griesbeck, M. Oelgemoller, Photodecarboxylative addition of carboxylates
and
α
-keto carboxylates to phthalimides, Synlett (1999) 492–494.
[1] V. Balzani (Ed.), Electron Transfer in Chemistry, Wiley, VCH, Weinheim, 2001.
[2] (a) F.D. Lewis, X. Liu, J. Liu, S. E, Miller, R.T. Hayes, M.R. Wasielewski, Direct
measurement of hole transport dynamics in DNA, Nature 406 (2000) 51–53;
¨
[22] F. Hatoum, S. Gallagher, L. Baragwanath, J. Lex, M. Oelgemoller,
Photodecarboxylative benzylations of phthalimides, Tetrahedron Lett. 50 (2009)
6335–6338.
¨
(b) B. Giese, J. Amaudrut, A.-K. Kohler, M. Spormann, S. Wessely, Direct
[23] V. Belluau, P. Noeureuil, E. Ratzke, A. Skvortsov, S. Gallagher, C.A. Motti,
observation of hole transfer through DNA by hopping between Adenine bases and
by tunnelling, Nature 412 (2001) 318–320;
¨
M. Oelgemoller, Photodecarboxylative benzylations of phthalimide in pH 7 buffer:
a simple access to 3-arylmethyleneisoindolin-1-ones, Tetrahedron Lett. 51 (2010)
(c) B. Giese, Electron transfer in DNA, Curr. Opin. Chem. Biol. 6 (2002) 612–618.
[3] (a) H.B. Gray, J.R. Winkler, Electron transfer in proteins, Annu. Rev. Biochem. 65
(1996) 537–561;
4738–4741.
¨
[24] S. Gallagher, F. Hatoum, N. Zientek, M. Oelgemoller, Photodecarboxylative
additions of N-protected α-amino acids to N-methylphthalimide, Tetrahedron Lett.
(b) B. Giese, M. Graber, M. Cordes, Electron transfer in peptides and proteins,
Curr. Opin. Chem. Biol. 12 (2008) 755–759.
51 (2010) 3639–3641.
¨
¨
´
[25] M. Horvat, H. Gorner, K.-D. Warzecha, J. Neudorfl, A.G. Griesbeck, K. Mlinaric-
[4] (a) B. Giese, Electron transfer through DNA and peptides, Bioorg. Med. Chem. 14
(2006) 6139–6143;
´
Majerski, N. Basaric, Photoinitiated domino reactions: N-(adamantyl)phthalimides
and N-(adamantylalkyl)phthalimides, J. Org. Chem. 74 (2009) 8219–8231.
(b) K. Siriwong, A.V. Voityuk, Electron transfer in DNA, WIREs Comput. Mol. Sci.
2 (2012) 780–794;
´
´
´
[26] L. Mandic, K. Mlinaric-Majerski, A.G. Griesbeck, N. Basaric, Photodecarboxylation
of adamantane amino acids activated by phthalimide, Eur. J. Org. Chem (2016)
4404–4414.
(c) D.N. Beratan, Why are DNA and protein electron transfer so different? Annu.
Rev. Phys. Chem. 70 (2019) 71–97.
ˇ
´
´
[27] M. Sohora, T. Sumanovac Ramljak, K. Mlinaric-Majerski, N. Basaric,
Photodecarboxylation of N-adamantyl and N-phenylphthalimide dipeptide
derivatives, Croat. Chem. Acta 87 (2014) 431–446.
[5] A.G. Griesbeck, N. Hoffmann, K.-D. Warzecha, Photoinduced Electron transfer
chemistry: from studies on PET processes to applications in natural product
synthesis, Acc. Chem. Res. 40 (2007) 128–140.
´
´
[28] M. Horvat, K. Mlinaric-Majerski, A.G. Griesbeck, N. Basaric, Photoinduced
decarboxylation of 3-(N-phthalimido)adamantane-1-carboxylic acid and radical
addition to electron deficient alkenes, Photochem. Photobiol. Sci. 10 (2011)
610–617.
[6] (a) V. Srivastava, P.P. Singh, Eosin Y catalysed photoredox synthesis: a review,
RSC Adv. 7 (2017) 31377–31392;
(b) J.-R. Chen, X.-Q. Hu, L.-Q. Lu, W.-J. Xiao, Visible light photoredox-controlled
reactions of N-radicals and radical ions, Chem. Soc. Rev. 45 (2016) 2044–2056;
(c) N.A. Romero, D.A. Nicewicz, Organic photoredox catalysis, Chem. Rev. 116
(2016) 10075–10166;
ˇ
´
´
[29] T. Sumanovac Ramljak, M. Sohora, I. Antol, D. Kontrec, N. Basaric, K. Mlinaric-
Majerski, Memory of chirality in the phthalimide photocyclization of adamantane
dipeptides, Tetrahedron Lett. 55 (2014) 4078–4081.
(d) K.L. Skubi, T.R. Blum, T.P. Yoon, Dual catalysis strategies in photochemical
synthesis, Chem. Rev. 116 (2016) 10035–10074;
́
́
́
[30] M. Sohora, M. Vazdar, I. Sovic, K. Mlinaric-Majerski, N. Basaric, Photocyclization
of tetra- and pentapeptides containing adamantylphthalimide and phenylalanines:
reaction efficiency and diastereoselectivity, J. Org. Chem. 83 (2018) 14905–14922.
(e) J.W. Beatty, C.R.J. Stephenson, Amine functionalization via oxidative
photoredox catalysis: methodology development and complex molecule synthesis,
Acc. Chem. Res. 48 (2015) 1474–1484;
´
ˇ
ˇ
´
´
´
[31] L. Mandic, I. Dzeba, D. Jadresko, B. Mihaljevic, L. Biczok, N. Basaric, Photophysical
properties and electron transfer photochemical reactivity of substituted
phthalimides, New J. Chem. 2020, 44, 17252-17266.
¨
(f) D.P. Hari, B. Konig, Synthetic applications of eosin Y in photoredox catalysis,
Chem. Commun. (Camb.) 50 (2014) 6688–6699;
(g) J. Xuan, W.-J. Xiao, Visible-light photoredox catalysis, Angew. Chem. Int. Ed.
(51) (2012) 2–13.
¨
[32] A.G. Griesbeck, H. Gorner, Laser flash photolysis of N-alkylated phthalimides,
J. Photochem, Photobiol. A: Chemistry. 129 (1999) 111–119.
[33] R.J. Perkins, H.-C. Xu, J.M. Campbell, K.D. Moeller, Anodic coupling of carboxylic
acids to electron-rich double bonds: a surprising non-Kolbe pathway to Lactones,
Beilstein J. Org. Chem. 9 (2013) 1630–1636.
[7] Y. Kanaoka, Photoreactions of cyclic imides. Examples of synthetic organic
photochemistry, Acc. Chem. Res. 11 (1978) 407–413.
´
´
[8] M. Horvat, K. Mlinaric-Majerski, N. Basaric, Photochemistry of N-alkyl and N-aryl
substituted phthalimides: H-Abstractions, single Electron transfer and
cycloadditions, Croat. Chem. Acta 83 (2010) 179–188.
[34] J.W. Verhoeven, Through-bond charge transfer interaction and photoinduced
charge separation, Pure Appl. Chem. 58 (1986) 1285–1290.
[35] M. Van der Auweraer, A. Gilbert, F.C. De Schryver, Influence of starting
¨
[9] M. Oelgemoller, A.G. Griesbeck, Photoinduced electron-transfer processes of
conformations on intramolecular exciplex formation in ω-phenyl-α-N,N-
phthalimides, in: W. Horspool, F. Lenci (Eds.), CRC Handbook of Organic
Photochemistry and Photobiology, 2nd ed., CRC Press, Boca Raton, 2004.
dimethylaminoalkanes, J. Am. Chem. Soc. 102 (1980) 4007–4017.
[36] U.C. Yoon, D.U. Kim, C.W. Lee, Y.S. Choi, Y.-J. Lee, H.L. Ammon, P.S. Mariano,
Novel and efficient azomethine ylide forming photoreactions of N-(silylmethyl)
phthalimides and related acid and alcohol derivatives, J. Am. Chem. Soc. 117 (10)
(1995) 2698–2710.
¨
[10] M. Oelgemoller, A.G. Griesbeck, Photoinduced Electron transfer chemistry of
phthalimdes: an efficient tool for C–C-Bond Formation, J. Photochem. Photobiol. C:
Photochem. Rev. 3 (2002) 109–127.
[11] U.C. Yoon, P.S. Mariano, The synthetic potential of phthalimide SET
photochemistry, Acc. Chem. Res. 34 (2001) 523–533.
´
´
[37] M. Sohora, L. Mandic, N. Basaric, [3+2] cycloadditions with photogenerated
azomethine ylides in β-cyclodextrine, Beilstein J. Org. Chem. 16 (2020)
1296–1304.
[12] U.C. Yoon, P.S. Mariano, The photochemistry of silicon-substituted phthalimides,
in: W. Horspool, F. Lenci (Eds.), CRC Handbook of Organic Photochemistry and
Photobiology, 2nd ed., CRC Press, Boca Raton, 2004.
[38] M. Montalti, A. Credi, L. Prodi, M.T. Gandolfi, Handbook of photochemistry, CRC
Taylor and Francis, Boca Raton, 2006.
[13] A.G. Griesbeck, A. Henz, K. Peters, E.-M. Peters, H. G.v on schnering, photo
[39] D.V. Bent, E. Hayon, Excited state chemistry of aromatic amino acids and related
peptides. II. phenylalanine, J. Am. Chem. Soc. 75 (1965) 2606–2612.
[40] P. O’Neill, S. Steenken, D. Schulte-Frohlinde, Formation of radical cations of
methoxylated benzenes by reaction with OH radicals, Tl2+, Ag2+, and SO4⋅ꢀ in
aqueous solution. An optical and conductometric pulse radiolysis and in situ
radiolysis electron spin Rrsonance study, J. Phys. Chem. 79 (1975) 2773–2779.
[41] U.C. Yoon, H.C. Kwon, T.G. Hyung, K.H. Choi, S.W. Oh, S. Yang, Z. Zhao, P.
S. Mariano, The photochemistry of polydonor-substituted phthalimides: curtin-
hammett-Type control of competing reactions of potentially interconverting
zwitterionic biradical intermediates Curtin-Hammet-type, J. Am. Chem. Soc. 126
(2004) 1110–1124.
electron transfer induced macrocyclization of N-Phthaloyl-ω-aminocarboxylic
acids, Angew. Chem. Int. Ed. 34 (1995) 474–476.
¨
[14] A.G. Griesbeck, T. Heinrich, M. Oelgemoller, J. Lex, A. Molis, A photochemical
route for efficient cyclopeptide formation with a minimum of protection and
activation chemistry, J. Am. Chem. Soc. 124 (2002) 10972–10973.
[15] U.C. Yoon, Y.H. Jin, S.W. Oh, C.H. Park, J.H. Park, C.F. Campana, X. Cai, E.
N. Duesler, P.S. Mariano, A synthetic strategy for the preparation of cyclic peptide
mimetics based on SET-Promoted photocyclization processes, J. Am. Chem. Soc.
125 (2003) 10664–10671.
[16] A. Soldevilla, A.G. Griesbeck, Chiral photocages based on phthalimide
photochemistry, J. Am. Chem. Soc. 128 (2006) 16472–16473.
[17] A.G. Griesbeck, W. Kramer, J. Lex, Diastereo- and enantioselective synthesis of
pyrrolo[1,4]benzodiazepines through decarboxylative photocyclization, Angew.
Chem. Int. Ed. 40 (2001) 577–579.
7