Photophysical properties and electron transfer photochemical reactivity of substituted phthalimides

Article abstract of DOI:10.1039/d0nj03465g

The photochemical reactivity and photophysical and electrochemical properties of a series of N-adamantylphthalimides bearing carboxylic functional groups were investigated. Upon irradiation (with or without a triplet sensitizer), the compounds undergo decarboxylation via photoinduced electron transfer (PET) from the carboxylate to the phthalimide. UV-vis and fluorescence pH titrations were used to determine pKa values for the prototropic forms, which were put in connection with the quantum yields of the reaction (ΦR). The compounds bearing electron donors OH and OCH3 at the phthalimide 4 position are fluorescent (ΦF = 0.02-0.49) and PET takes place from both singlet and triplet excited states. The estimated rate constants for PET in the singlet excited states for the methoxy- A nd amino-substituted phthalimides are (2.0 ± 0.1) × 109 s-1 and (3.4 ± 1.0) × 107 s-1, respectively. Laser flash photolysis (LFP) was conducted to characterize the triplet excited states, which are populated less efficiently for compounds with electron donors. The PET is reversible and the overall ΦR depends on the rates for back electron transfer, protonation of the phthalimide radical anion and decarboxylation. The plausible photochemical and photophysical pathways depend on the phthalimide substituents, which is important for the use of phthalimide derivatives in organic synthesis and photocatalysis. This journal is

Full text of DOI:10.1039/d0nj03465g

View Article Online
View Journal
NJC
New Journal of Chemistry
Accepted Manuscript
A journal for new directions in chemistry
This article can be cited before page numbers have been issued, to do this please use: L. Mandic, I.
Dzeba, D. Jadresko, B. Mihaljevic, L. Biczok and N. Basari, New J. Chem., 2020, DOI:
10.1039/D0NJ03465G.
Volume 42
This is an Accepted Manuscript, which has been through the
Number
6
21 March 2018
Pages 3963-4776
Royal Society of Chemistry peer review process and has been
accepted for publication.
NJC
New Journal of Chemistry
rsc.li/njc
A journal for new directions in chemistry
Accepted Manuscripts are published online shortly after acceptance,
before technical editing, formatting and proof reading. Using this free
service, authors can make their results available to the community, in
citable form, before we publish the edited article. We will replace this
Accepted Manuscript with the edited and formatted Advance Article as
soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes to the
text and/or graphics, which may alter content. The journal’s standard
Terms & Conditions and the Ethical guidelines still apply. In no event
shall the Royal Society of Chemistry be held responsible for any errors
or omissions in this Accepted Manuscript or any consequences arising
from the use of any information it contains.
ISSN 1144-2546
PAPER
Chechia Hu, Tzu-Jen Lin et al.
Yellowish and blue luminescent graphene oxide quantum dots
prepared via a microwave-assisted hydrothermal route using
H2O2 and KMnO4 as oxidizing agents
rsc.li/njc

Please do not adjust margins
New Journal of Chemistry
Page 1 of 17
1
2
3
4
5
6
7
View Article Online
DOI: 10.1039/D0NJ03465G
ARTICLE
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Photophysical properties and electron transfer photochemical
reactivity of substituted phthalimides
Leo Mandić,a,b Iva Džeba,b Dijana Jadreško,c Branka Mihaljević,b László Biczókd and Nikola Basarić*a
Received 00th January 20xx,
Accepted 00th January 20xx
Photochemical reactivity, photophysical and electrochemical properties for a series of N‐adamantylphthalimides bearing
carboxylic functional groups were investigated. Upon irradiations (with or without a triplet sensitizer), compounds undergo
decarboxylation via a photoinduced electron transfer (PET) from the carboxylate to the phthalimide. UV‐Vis and
fluorescence pH titrations were used to determine pKa values for the prototropic forms, which were put in connection with
quantum yields of the reaction (ΦR). Compounds bearing electron donors OH and OCH3 at the phthalimide 4 position are
fluorescent (ΦF = 0.02‐0.49) and PET takes place from both singlet and triplet excited states. Estimated rate constants for
PET in the singlet excited states for methoxy‐ and amino‐substituted phthalimides are (2.0 ± 0.1) × 109 s‐1and (3.4 ± 1.0) ×
107 s‐1, respectively. Laser flash photolysis (LFP) was conducted to characterize triplet excited states, which are populated
less efficiently for compounds with electron donors. The PET is reversible and the overall ΦR depends on the rates for back
electron transfer, protonation of the phthalimide radical anion and decarboxylation. Plausible photochemical and
photophysical pathways depend on the phthalimide substituents, which is important for the use of phthalimide derivatives
in organic synthesis and photocatalysis.
DOI: 10.1039/x0xx00000x
drugs,20‐22
such
as
ketoprofen,23‐32
since
the
Introduction
photodecarboxylation of these drugs leads to photoallergic
responses.33 Photoinduced decarboxylation is also important in
the context of different photocatalytic processes.34‐41
Phthalimide is a chromophore whose photochemistry has been
investigated for decades since the pioneering work of Kanaoka
et al.1 A rich array of phthalimide photoreactions, involving H‐
abstractions, cycloadditions and photoinduced electron
transfer (PET),2 found widespread applications in the synthesis
of complex molecules and natural products.3 For example,
phthalimide derivatives in the triplet excited state undergo PET
whereby phthalimide is an oxidant capable of reacting with
donors having oxidation potential < 1.6 V vs. saturated calomel
electrode.4,5 These PET processes can initiate elimination of silyl
groups,6 or photodecarboxylation,4,5 both of which were used
in organic synthesis.7 Thus, photodecarboxylation initiated
macrocyclizations8 and photocyclization of peptides,9,10 photo‐
decaging,11 enantioselective synthesis of benzodiazepines,12
and cyclic aryl ethers,13 as well as acetate,14,15 benzyl,16‐18 or α‐
amino acid addition to phthalimides,19 were documented.
Furthermore, photodecarboxylations were also intensively
investigated in the series of nonsteroidal anti‐inflammatory
We have become interested in photodecarboxylation reactions
initiated by phthalimide chromophore42‐44 and applied them in
cyclizations with memory of chirality45 and diastereoselective
peptide cyclizations.46 Furthermore, the photodecarboxylation
efficiency was investigated in a series of phthalimide derivatives
of adamantane amino acids, where we modified the distance
between the electron donor (carboxylate) and the acceptor
(phthalimide).42 Thus, the decarboxylation efficiency was a
useful handle to indirectly probe the rate of PET. Herein we
describe the use of the same decarboxylation on adamantane
derivative
series of substituted phthalimide derivatives
Phthalimide was systematically modified by introducing
1 (Scheme 1) as a probe for the PET reactivity in a
1‐6.
1
electron‐withdrawing and electron‐donating substituents at the
4 position, which affected the photophysical properties of the
molecules, as well as their photochemical reactivity. Although
a. Department of Organic Chemistry and Biochemistry, Ruđer Bošković Institute,
Bijenička cesta 54, 10000 Zagreb, Croatia, NB E‐mail: nbasaric@irb.hr.
b. Department of Material Chemistry, Ruđer Bošković Institute, Bijenička cesta 54,
10000 Zagreb, Croatia.
c. Division for Marine and Environmental Research, Ruđer Bošković Institute,
Bijenička cesta 54, 10000 Zagreb, Croatia.
d. Institute of Materials and Environmental Chemistry, Research Centre for Natural
Sciences, 1519 Budapest, P.O. Box 286, Hungary.
Electronic Supplementary Information (ESI) available: details on photochemical
experiments, cyclic voltammetry, UV‐vis and fluorescence spectroscopy, laser flash
photolysis and MS and NMR spectra of pure compounds. See
DOI: 10.1039/x0xx00000x
Scheme 1. Photodecarboxylation of phthalimide derivatives.
Please do not adjust margins

Please do not adjust margins
New Journal of Chemistry
Page 2 of 17
ARTICLE
Journal Name
1
2
3
4
5
6
7
8
9
photochemical transformations of phthalimides are very useful
in organic synthesis, not much is known about the
photochemistry of their substituted derivatives,47‐49 with a
notable exception of 4‐aminophthalimide derivatives, which are
known for their photochemical stability and high fluorescence
View Article Online
COCH3
DOI: 10.1039/D0NOJ03465G
O
OH
OOH
O
O
N
N
N
N
O
O
O
O
R
R
R
R
1H R = H
2COCH
3OH
4OH
5OH
3 R = COOH
R = OCH3
R = OH
R = NO2
4OOH
5OOH
R = OH
R = NO2
2H
3H
4H
5H
6H
R = COOH
R = OCH3
R = OH
R = NO2
R = NH2
3COCH3 R = OCH3
4COCH
quantum
yields.50‐56
Photochemical
reactivity
and
3 R = OH
photophysical properties for
1‐6
were investigated by
preparative irradiations, steady‐state and time‐resolved
fluorescence spectroscopy and laser flash photolysis (LFP). The
photochemical experiments allowed for a complete elucidation
of plausible pathways, particularly PET, which were put in
connection with electrochemical properties of the molecules
determined by cyclic voltammetry and pKa values measured by
UV‐Vis and fluorescence pH titrations. Our main finding is that
substituents on the phthalimide chromophore primarily affect
the efficiency of intersystem crossing (ISC), where electron
donors divert the photoreactions via singlet excited state
instead of triplet. The detailed mechanistic study presented
herein is important for the complete understanding of
phthalimide photochemistry and rational planning of
photoreactions for the use in organic synthesis.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Chart 1. Products of photodecarboxylation in CH3CN‐H2O.
Table 1. Irradiation conditions, conversions and product yields
(%) for photolysis of phthalimides
6 a
1‐ .
1H‐ 3OH‐
Compound
solvent
Conversionb
irradiation
time
45 min
100
4OOH‐
5OOH
2COCH3‐
4COCH3
6H
5OH
1
100
CH3CN‐H2O
2
81
30 min
66
20 min
88
30 min
86
30 min
33
1 h
70
4 h
81
43
88
70
33
42
8
n.d.
n.d.
n.d.
4
n.d.
n.d.
n.d.
n.d.
n.d.
8
n.d.
23c
n.d.
12
acetone‐H2O
2
CH3CN‐H2O
3
acetone‐H2O
3
CH3CN‐H2O
Results and discussion
Synthesis and preparative photochemistry
4
n.d.
16
n.d.
4
acetone‐H2O
Phthalimide derivatives
1
‐
6
were prepared according to
4
procedures published in literature precedent.56,57 The synthesis
involves condensation of 3‐aminoadamantane‐1‐carboxylic acid
with various 4‐substituted phthalic anhydrides, and some
subsequent transformations such as cleavage of the OMe ether
by BBr3, or reduction of the nitro group to amino.
CH3CN‐H2O
5
50
15
6
n.d.
n.d.
CH3CN‐H2O
6
20 min
22
22
n.d.
n.d.
CH3CN‐H2O
4 h
a
Preparative irradiations of 1‐6 were conducted in CH3CN‐H2O,
Irradiations were conducted in acetone‐H2O or CH3CN‐H2O (3:1 ‐ v/v) using 10
lamps with the maximum output at 300 nm. The concentration of phthalimides
was 1.5 mmol dm‐3 (≈ 100 mg/200 mL), whereas K2CO3 concentration was 0.75
mmol dm‐3. The solutions were purged with Ar prior to irradiation. The yields
correspond to isolated compounds, unless specified otherwise. b Conversion was
calculated from 1H NMR spectra of the crude irradiation mixture. c Yield obtained
by HPLC analysis. n. d. = not detected.
or acetone‐H2O mixtures (3:1 ‐ v/v), in the presence of
potassium carbonate to deprotonate the carboxylic functional
group at the adamantane moiety. Namely, carboxylate is a
better electron‐donating group than the carboxylic acid (for
example, 10‐undecenoate in CH3CN Eox ≈ 1.38 V vs. Ag/AgCl, and
10‐undecenoic acid in CH3CN Eox ≈ 1.85 V vs. Ag/AgCl).58
A
Simple decarboxylation products are expected but the
formation of alcohols, peroxides, and particularly ketones is
surprising at first sight. Peroxide and alcohol photoproducts are
probably formed by trapping of the radical (see Schemes 1 and
2) by traces of O2 impurities in the solution. We have previously
demonstrated that adamantyl radical formed in the
decarboxylation reaction can be trapped by O2 giving peroxides
and alcohols.42 Formation of ketones, on the other hand,
suggests that an intermediate in the process is adamantyl
carbanion that reacts with CH3CN, although the reaction of
adamantyl radical with CH3CN to give ketones cannot be
disregarded.
difference in the use of acetone or CH3CN is anticipated since
acetone acts as a triplet sensitizer with higher energy of the
triplet excited state (ET = 332 kJ mol‐1)59 than the one of N‐
alkylphthalimides (N‐methylphthalimide ET = 297 kJ mol‐1)59
where exergonic energy transfer to the phthalimide should be
feasible.
The irradiations (300 nm) in acetone‐H2O mixture were
generally selective, giving only decarboxylation products, 2H
‐
4H, analogous to the reaction shown in Scheme 1. On the other
hand, irradiations in CH3CN‐H2O gave, in addition to simple
decarboxylation products, alcohols 3OH
‐5OH, peroxides 4OOH‐
5OOH ketones 2COCH3 4COCH3 (Chart 1), and some
,
‐
To elucidate the reaction mechanism for the formation of
unidentified products in the case of nitro derivative. All types of
photoproducts were isolated and characterized by NMR and
MS, and the yields of the isolated photoproducts are given in
Table 1.
ketones, we performed irradiation of
3 in CH3CN‐H2O and
CH3CN‐D2O. After the irradiation under identical conditions, the
composition of the solution was analyzed by HPLC. Simple
decarboxylation photoproducts 3H and 3D were isolated and
2 | J. Name., 2012, 00, 1‐3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
New Journal of Chemistry
Page 3 of 17
1
2
3
4
5
6
7
View Article Online
DOI: 10.1039/D0NJ03465G
ARTICLE
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 2. Plausible mechanism for the formation of 3H, 3D and 3COCH3.
Table 2. Quantum efficiency for photoreaction (ΦR) of 1‐ .
6 a
Compound
CH3CN‐H2O (1:1 ‐ v/v)
acetone‐H2O (1:1 ‐ v/v)
pH = 4.5
n. d.
0.0083
± 0.0006
0.022
± 0.002
0.0018
± 0.0001
n. d.
pH = 6.4
n. d.
n. d.
pH = 8.3
0.1144
0.11 ± 0.01
pH = 4.5
n. d.
0.079
± 0.003
n. d.
pH = 6.4
n. d.
n. d.
pH = 8.3
0.544
0.34
± 0.01
0.35
± 0.02
0.0011
± 0.0001
0.10
1
2
3
4
5
6
n. d.
0.10 ± 0.01
n. d.
0.011
± 0.001
n. d.
0.0052
± 0.0005
0.035
± 0.002
0.0046
± 0.0001
0.00061
± 0.00003
n. d.
0.0032
± 0.0002
n. d.
± 0.01
0.0018
± 0.0001
0.00045 ± 0.00003
n. d.
0.0015
± 0.0001
n. d.
a Measured in CH3CN‐H2O or acetone‐H2O (1:1 ‐ v/v) in the presence of potassium phosphate buffer (c = 0.05 mol dm‐3). The samples were irradiated at 300 nm. The
values correspond to the average of the results of three measurements. The photochemical transformation of 7 into 8 and 9 in acetone‐H2O (1:1 ‐ v/v) or CH3CN‐H2O
(1:1‐ v/v) was used as a secondary actinometer (ΦR = 0.30 ± 0.03).42,60
Please do not adjust margins

Please do not adjust margins
New Journal of Chemistry
Page 4 of 17
1
2
3
4
View Article Online
DOI: 10.1039/D0NJ03465G
5
6
7
8
ARTICLE
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
analyzed by NMR and MS to determine the extent of different prototropic forms (vide infra), which should affect
deuteration. Upon irradiation of (10 min) in CH3CN‐H2O the their reactivity. Therefore, the photoreactivity of phthalimide
conversion to photoproducts was 80%, and the ratio of derivatives , and was investigated in solutions at different
3H 3COCH3 was ≈ 9:1 (see the Experimental Section). On the pH values (see Figs S1‐S3 in the ESI). Compound , which bears
other hand, under the same conditions in CH3CN‐D2O, the two carboxylic functional groups, undergoes more efficient
conversion was 13% and the ratio of 3H(3D) 3COCH3 was ≈ 4:1. photodecarboxylation in both solvents at pH > 6, whereas at pH
In addition, deuterium incorporation took place, with 12% D > 7, the efficiency slightly decreases. The results indicate that
incorporated according to MS. That is, 88% 3H and 12% 3D were the solution pH has to be basic enough to assure the
formed. A significant decrease of the photoreaction conversion deprotonation of the carboxylic functional group at the
in D2O was assigned to the equilibration of the excited state (3‐ adamantane (electron donor), which facilitates the PET.
3
2,
4
6
:
2
:
*
) with the charge transfer state formed by PET, where the However, formation of doubly charged anion by deprotonation
٭
latter
[
3‐CT
]
undergoes protonation vs. deuteration, of both carboxylic acids probably decreases the rate of PET.
is a phenolic compound that can be
. Thus, in the presence of D2O, the deprotonated at high pH values, which significantly decreases
slower D+ addition to [3‐CT]٭ changes the plausible pathways the photodecarboxylation efficiency. Thus,
undergoes the
and diminishes the importance of photoproduct formation most efficient photoreaction at pH 6‐7, when the carboxylic
compared to deactivation resulting in ground state group is deprotonated but the phenolic is not. On the other
Therefore, the deuterium effect provides a compelling evidence hand, phthalimide is non‐reactive (below pH 3) or very weakly
decarboxylation or electron back transfer leading to the ground Furthermore, phthalimide
state
, or excited state 3‐
4
3
٭
4
3.
6
that efficiency of the sequence that involves PET and reactive. Its reactivity increases very weakly in the pH region 3‐
photodecarboxylation does not depend on the rate of PET only, 7, in line with the anticipated pH region where the carboxylic
but it depends also on the subsequent reactions, protonation functional group exhibits protonation equilibrium.
and decarboxylation, that compete with the electron back Quantum yields for the photochemical reactions (ΦR) of
transfer. This is a new mechanistic detail that in the previous were measured in CH3CN‐H2O and acetone‐H2O (1:1 ‐ v/v)
reports has not been taken into account.42,60
(Table 2). Instead of the base, solutions were buffered at pH
Based on the deuterium incorporation in the molecule, values 4.5, 6.4 and 8.3 to have compounds present in different
plausible mechanism involves formation of carbanion 3‐AN prototropic forms (vide infra). The samples were excited at 300
which is deuterated or reacts with the solvent (CH3CN) to afford nm, and the photolysis of phthalimide was used as a
1‐6
,
7
imine 3‐IM, which subsequently hydrolyses to 3COCH3. On the secondary actinometer (ΦR = 0.3, Scheme 3).60 A general trend
other hand, since the extent of deuteration did not exceed 12%, appeared, that ΦR is higher at pH values when the adamantane
it is plausible that 3COCH3 was formed in a parallel reaction of carboxylic acid is deprotonated, in agreement with the
the adamantyl radicals 3‐BR
eventually react with CH3CN to yield the 3COCH3
formed from 3BR‐2. However, upon irradiation in CH3CN‐D2O, for
,
3‐BR2, and 3BR‐3, which assumption that decarboxylation takes place after the initial
3H can also be PET from carboxylate to the excited phthalimide. Furthermore,
, higher ΦR was observed for the acetone triplet‐
.
1‐3
incorporation of D is not expected since the only source of D is sensitized reactions.
D2O, and the homolytic cleavage of the H‐OH (or D‐OD) bond is
energetically very demanding (the bond dissociation energy for
H2O is 460 kJ/mol). Note the different ratio of 3H:3COCH3 which
changes from 9:1 in CH3CN‐H2O to 4:1 in CH3CN‐D2O. It indicates
the involvement of primary deuterium isotope effect with the
value of ≈ 2.2, due to competition of deuteration with the attack
of the solvent (CH3CN) to the carbanion 3‐AN. The key reaction
step in the carbanion reaction mechanism is back electron
transfer (BET) from the phthalimide radical anion to the
adamantyl radical (Scheme 2). To probe the feasibility of BET
and to get general knowledge about redox properties of the
molecules in PET, it is important to investigate their
electrochemical properties (vide infra).
Scheme 3. Photoreaction of phthalimide 7 used as a secondary actinometer for the
determination of reaction quantum yields.
Electrochemical properties
Cyclic voltammetry was conducted for 1‐6 to determine their
reduction potentials, which is important for the prediction of
their PET reactivity. The measurements were performed in
CH3CN in the presence of NBu4PF6 as an electrolyte in the
potential range from about 0.0 V to ‐2.0 V using Ag/AgCl (in 3
mol dm‐3 NaCl) reference electrode. The measured reduction
During the course of photolysis, the solution pH changes.
Moreover, the investigated phthalimide derivatives can exist in
Please do not adjust margins

Page 5 of 17
New Journal of Chemistry
Please do not adjust margins
Journal Name
ARTICLE
1
2
3
4
5
6
7
8
9
potentials expressed vs. ferrocene (Fc0/Fc+) are compiled in In the series of investigated phthalimides
1
‐
6
, only derivatives
View Article Online
DOI: 10.1039/D0NJ03465G
Table 3, whereas all details can be found in the ESI (Figs S4‐S9).
3, 4 and 6 exhibit a significant fluorescence, whereas for the
From the reduction potentials and estimated energies of the other molecules, fluorescence quantum yield (ΦF) was lower
singlet excited states (∆ꢀꢁ,ꢁ), obtained from the onsets of the
absorption spectra, Gibbs free energies for PET (∆ETG°) were
calculated. Previous studies demonstrated that the energy of S1
1.00
6
3
4
state, which corresponds to the location of the intersection of
the normalized absorption and fluorescence spectra, barely
differs from the energy of the onset of the absorption.61,62
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0.75
Therefore, the use of the latter quantity is
a good
approximation (see Figs S13 and S14 in the ESI). The obtained
values suggest that PET from the S1 state of all phthalimides
should be thermodynamically feasible except for amino
0.50
0.25
0.00
derivative 6.
Table 3. Reduction potentials for 1‐6,a energies for the
excitation to S1 and calculated Gibbs free energy for PET in S1.
350
400
450
500
550
600
650
Compound Ered / V vs. Fc0/Fc+
ES1 / kJ mol‐1 b
356 (336 nm)
345 (347 nm)
310 (386 nm)
308 (389 nm)
299 (400 nm)
∆ETG° / kJ mol‐1 c

/ nm
1
2
3
4
5
‐1.98
‐1.62 [d]
‐2.05
‐69
‐92
‐16
‐28
‐86
Figure 1. Normalized emission spectra of 3 (λexc = 310 nm), 4 (λexc = 310 nm) and 6 (λexc =
350 nm) in CH3CN at 25 °C.
‐1.90 [d]
‐1.21
than 10‐4. In addition, fluorescence was measured for 3H
, 4H
and 6H (See Fig S16). The fluorescence spectra (Fig 1) and ΦF
were determined in aprotic solvent (CH3CN) and in aqueous
solvent at two pH values (4.5 and 8.3), where the adamantane
‐1.70 [e]
‐2.17
6
260 (460 nm)
45
a
The measurements were conducted in CH3CN in the presence of 0.1 mol dm‐3 carboxylic acid should be protonated or not (Table 4). Under the
NBu4PF6 as electrolyte. The reported value is the mean between the reduction and
oxidation peak, unless stated otherwise. b The energy for the vertical excitation to
S1 (ES1), estimated from the onset of the absorption spectrum, the wavelength is
given in brackets. c Gibbs free energy for PET (∆ETG°) was calculated according to:
same conditions, time‐correlated single‐photon counting
measurements (TC‐SPC) were performed to reveal singlet
excited state lifetimes (τF) and potentially elucidate processes
that lead to the deactivation from the singlet excited state.
ꢋ∙
ꢂꢃꢄꢅ ꢆꢇꢈ ꢀ
D
/D ꢐ ꢀ A/A
ꢏ Dꢋ∙Aꢍ∙ ꢐ ∆ꢀꢁ,ꢁ
, where e is the
ꢊ ꢌ
ꢉ
ꢅꢊ
ꢌ
ꢅꢊ
ꢍ∙ꢌꢎ
Δ
elementary charge, e = 1.602 × 10‐19 C, NA is the Avogadro constant, NA = 6.022 × The most intensively fluorescent compound is
6, whose spectral
1023 mol‐1, E°(D+./D)/V is the standard electrode potential of the donor radical
and photophysical properties in CH3CN are analogous to those
reported for 6H 56
. Decay of fluorescence in CH3CN was fit to an
exponential function revealing similar τF for both compounds.
However, a small difference in ΦF at pH 4.5 and 8.3 was
cation formed in PET (0.96 V vs. Fc0/Fc+ value taken from the ref. 58), and E°(A/A–
.)/V is the standard electrode potential of the acceptor radical anion (the same as
the measured Ered), ꢏ Dꢋ∙Aꢍ∙ is the electrostatic work that accounts for the effect
ꢊ
ꢌ
of Coulombic attraction in the photoproducts (the estimated value is 3.4 kJ mol‐1,
see the SI), ∆ꢀꢁ,ꢁ is the vibrational zero electronic energy of the excited donor, observed for
6 in CH3CN‐H2O 1:1 v/v mixture, which was
d
e
approximated as ES1.63 Irreversible reduction peak. Two reduction processes,
attributed to the protonation equilibrium of the carboxylic acid.
At higher pH, a good electron donor carboxylate is formed but
PET to the phthalimide moiety is inefficient in the S1 excited
both reversible.
Photophysical properties
state due to a positive ∆ETG° of the process. The τF values for
6
Absorption spectra for phthalimides
CH3CN (Figs S10‐S15 in the ESI). Phthalimides
1
‐
6
were measured in
and exhibit the
in CH3CN‐H2O 1:1 v/v mixture diminished from 7.30 to 5.84 ns
when the pH was raised from 4.5 to 8.3, in line with the lowering
of the ΦF. Assuming that the τF decrease arises from PET
(possible only when the acid is deprotonated), ket = (3.4 ± 1.0) ×
107 s‐1 was estimated for the rate constant of this process at pH
8.3. At the low wavelength region of the fluorescence spectrum
(460 nm), a short‐lived emission of
appeared indicating the partial protonation of the amine moiety
in the S1 excited state.
1
2
typical absorption spectra of N‐alkylphthalimides with a low‐
energy maximum at ≈ 300 nm, whereas strong electron‐
donating and ‐withdrawing groups significantly affect the
spectra. Thus, absorption spectra of
3 and 4, compounds that
bear moderately electron‐donating groups, exhibit additional
low energy band stretching between 320 and 380 nm. The
effect is more pronounced with amino substitution, so that
≈ 0.4 ns lifetime also
absorption of 6 is characterized by a low‐energy maximum at ≈
Methoxy‐substituted compound
3 has a modest ΦF in CH3CN,
370 nm. Note that this band is solvatochromic, due to a CT
character of the aminophthalimide chromophore.50‐56 On the
other hand, electron‐withdrawing nitro group also affects the
spectrum by introducing a lower energy absorption band at
350‐400 nm with low absorptivity.
presumably due to competing intersystem crossing ISC (vide
infra). The decay of fluorescence was fit to a sum of two
exponentials (Table 4), with increasing contribution of a long‐
lived component at higher wavelengths (Table S1). The dual-
exponential fluorescence decay kinetics is probably attributed
to the two torsional isomers differing in the orientation of the
methoxy substituent relative to the phthalimide group.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1‐3 | 5
Please do not adjust margins

Please do not adjust margins
New Journal of Chemistry
Page 6 of 17
1
2
3
4
View Article Online
DOI: 10.1039/D0NJ03465G
5
6
7
8
ARTICLE
9
Table 4. Fluorescence quantum yields (ΦF)a and fluorescence decay times (τF)b
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Compound
In CH3CN
in CH3CN‐H2O (1:1 ‐ v/v)
pH = 4.5
τF / ns
pH = 8.3
ΦF
τF / ns
ΦF
ΦF
τF / ns
0.46 ± 0.01
23.7 ± 1.0
n.d.
3
3H
4
0.024 ± 0.002
1.68 ± 0.01
16.0 ± 0.4
n.d.
0.49 ± 0.05
4.6 ± 0.2
24.7 ± 1.0
n.d.
0.017 ± 0.006
0.008 ± 0.001
0.037 ± 0.003
0.30 ± 0.03
n. d.
2.33 ± 0.08
14.7 ± 0.3
0.013 ± 0.001
≤ 0.1
2.4 ± 0.3
22.0 ± 0.8 c
n.d.
0.018 ± 0.001
1.88 ± 0.01
n.d.
4H
6
0.027 ± 0.006
0.66 ± 0.03
n.d.
0.019 ± 0.001
0.11 ± 0.01
n. d.
17.2 ± 0.2
0.4 ± 0.1
0.087 ± 0.002
0.4 ± 0.1
7.32 ± 0.05
5.84 ± 0.07
6H
0.60 ± 0.06
18.98 ± 0.02 56
0.12 ± 0.01
n.d.
n. d.
n.d.
a Measured using quinine sulfate in 0.5 mol dm‐3 H2SO4 (ΦF = 0.55) or 9‐cyanoanthracene in cyclohexane (ΦF = 0.92) as a reference.59 The data represent the average of
the results obtained at three excitation wavelengths. The error corresponds to maximum absolute deviation. b Singlet excited state lifetimes were measured by time‐
correlated single photon counting (TC‐SPC). At least three decays were collected at three wavelengths and the average value is reported, and the error corresponds to
maximum absolute deviation Pre‐exponential factors depend on the detection wavelength and they are reported in Table S1 in the ESI. c The pre‐exponential factor of
this component has ≤ 0.7 % contribution to the sum of all pre‐exponential factors.
Similar behavior has been reported for 2‐methoxyanthracene,64 the monitoring wavelengths is raised, and the amplitude of a
and 6‐methoxy‐1‐methylquinolinium.65 In aqueous solvent, at longer‐lived emission (τF = 2.4 ns) gradually grows. The latter
pH 4.5 the ΦF of
whereas both decay components exhibit significantly longer fluorescence emerges at longer wavelengths. Similar behavior
lifetime. Red‐shifts in the absorption and fluorescence spectra was found at pH 6.4 where the COOH moiety of is
were also observed, implying that the energy of the S1 state deprotonated but the OH substituent remains unchanged
diminishes when CH3CN solvent is replaced by CH3CN‐H2O 1:1 (Table S1). At pH 8.3, the phenolate form of is produced in the
v/v mixture. This effect probably reduces the rate of triplet ground state. Hence, exponential fluorescence decay was
formation leading to higher ΦF and longer τF. The alteration of observed with 1.88 ns lifetime at all wavelengths.
pH from 4.5 to 8.3 results in considerable decrease in the
3 is about 20 times higher than in CH3CN, fluorescence is attributed to the phenolate form, whose
4
4
dominant fluorescence lifetime component of 3, probably due Acid‐base properties of substituted phthalimides
to PET from the deprotonated carboxylate to
methoxyphthalimide in the S1 excited state. From the difference
in lifetimes, ket = (2.0 ± 0.1) × 109 s‐1 was derived for the rate
Fluorescence measurements and the dependence of
photodecarboxylation efficiencies on pH suggested that
phthalimide derivatives exhibit different photophysical
properties and react differently depending on their prototropic
forms. Therefore, we investigated acid‐base properties of
molecules by UV‐Vis and fluorescence pH titrations.
constant of PET. The two order of magnitude higher ket for
than for
3
6
is in accordance with the smaller estimated ∆ETG° for
the former compound (Table 3).
In CH3CN, the OH‐substituted phthalimide
photophysical properties as the methoxy compound 3, with a
4
has similar
Fluorescence titration for 3 was conducted in the pH region 3.5‐
7.5, where a significant fluorescence quenching was observed
(Fig 2) with no change in the absorption spectra. The finding was
rationalized by quenching of fluorescence via electron transfer
in the S1 excited state from carboxylate, which is formed upon
deprotonation with increasing solution basicity (Eq 1), to the
methoxy‐substituted phthalimide moiety. The fluorescence
titration data was processed by nonlinear regression analysis
using the Specfit software, allowing for the estimation of the pKa
value for deprotonation of the adamantyl carboxylic functional
modest ΦF and dual‐exponential fluorescence decay. In
contrast, the addition of water leads to substantial ΦF decrease
in CH3CN‐H2O mixture at pH 4.5 because proton transfer from
the OH group to water becomes an efficient deactivation
pathway of the S1 excited state. The short‐lived fluorescence
component, whose lifetime is less than the time‐resolution of
our instrument (≤ 0.1 ns), indicates that the excited state proton
transfer (ESPT) is very fast. This decay component weakens as
Please do not adjust margins

Page 7 of 17
New Journal of Chemistry
Please do not adjust margins
Journal Name
ARTICLE
1
2
3
4
5
6
7
8
9
Table 5. pKa values for phthalimide derivatives determined by group (Table 5). Furthermore, it is reasonable to assume that all
View Article Online
UV‐Vis or fluorescence pH titration.a
compounds 1‐6
probably have similar vaDluOeI:s1f0o.1r03th9/eDp0NKJa0o3f46th5Ge
adamantane carboxylic group since the changes at the distant
phthalimide site should not affect its acidity.
For compound 4 at low pHs, the fluorescence is very weak in the
400‐500 nm range, where its emission is expected, due to the
very rapid ESPT bringing about singlet excited phenolate form.
The increase of the solution basicity enhances the phenolate
fluorescence appearing at longer wavelengths as a result of
Compound / equilibrium
Spectroscopic
pKa
method
Fluorescence
UV‐Vis
Fluorescence
UV‐Vis
Fluorescence
Fluorescence
UV‐Vis
Fluorescence
UV‐Vis
3 / RCOOH→RCOO‐ + H+
4 / ROH→RO‐ + H+
5.56 ± 0.01
7.36 ± 0.01
7.32 ± 0.01
7.36 ± 0.02
7.28 ± 0.02
6.0 ± 0.2
1.65 ± 0.02
2.11 ± 0.03
2.02 ± 0.04
2.30 ± 0.04
4H / ROH→RO‐ + H+
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
6 / RCOOH→RCOO‐ + H+
concomitant
4 deprotonation in the ground and singlet excited
+
RNH3 →RNH2 + H+
states (Fig 3). The alteration of the UV‐Vis spectra (Fig S17)
implies the formation of the phenolate absorbing at higher
wavelengths in line with the equilibrium for the phenol
deprotonation in the ground state. Processing of the
fluorescence and UV‐Vis data by nonlinear regression analysis
+
RNH3 →RNH2 + H+
+
6H / RNH3 →RNH2 + H+
+
RNH3 →RNH2 + H+
Fluorescence
a The titrations were conducted in CH3CN‐H2O in the presence of KCl (c = 2 mol dm‐
3) (for 6 and 6H) or phosphate buffer (c = 0.05 mol dm‐3) (for 3, 4, and 4H) at 25 °C.
provided similar pKa values for the deprotonation of
4 in the
ground state, which are somewhat lower than for the parent
phenol 4H (Table 5, Figs S18 and S19) because the electron‐
withdrawing effect of the imide stabilizes the double negative
charge formed under basic conditions (Eq 2).
300
250
100
80
200
150
100
50
buffer pH value increase
60
buffer pH
value increase
40
0
400
440
480
520
560

/ nm
20
0
300
250
200
150
100
50
Experiment
Model
450
500
550
600
650
700

/ nm
100
75
50
25
0
Experiment
Model
0
3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5
pH
Figure 2. Emission spectra of 3 (λexc = 350 nm, c = 9.13 × 10‐6 mol dm‐3) in CH3CN‐H2O (1:1
‐v/v) in the presence of phosphate buffer (c = 0.05 mol dm‐3) at different buffer pH values
(top) and dependence of the fluorescence intensity at 435 nm on pH (bottom); black dots
are the measured values and the red line corresponds to the calculated values obtained
by nonlinear regression analysis with the Specfit software. The measurements were
conducted at 25 °C.
6.0
6.5
7.0
7.5
8.0
8.5
9.0
pH
Figure 3. Emission spectra of 4 (λexc = 380 nm, c = 2.72 × 10‐5 mol dm‐3) in CH3CN‐H2O (1:1
‐ v/v) in the presence of phosphate buffer (c = 0.05 mol dm‐3) at different buffer pH values
(top) and dependence of the fluorescence intensity at 560 nm on pH (bottom); black dots
are the measured values and the red line corresponds to the calculated values derived
by nonlinear regression analysis with the Specfit software. The measurements were
conducted at 25 °C.
COOH
COO-
O
O
pKa
H+
(1)
+
N
N
O
O
3-
OCH3
OCH3
3
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1‐3 | 7
Please do not adjust margins

Please do not adjust margins
New Journal of Chemistry
Page 8 of 17
ARTICLE
Journal Name
1
2
3
4
5
6
7
8
6H for which similar pKa values were revealed from the
titrations. The increase of the solution pH above 4.5, very
weakly quenched fluorescence for 6, but no change was found
for 6H, confirming that the quenching originates from the
deprotonation of the carboxylic group (Eq 3), which was also
indicated by TC‐SPC (vide supra).
COO-
COO-
View Article Online
DOI: 10.1039/D0NJ03465G
O
O
pKa
H+
(2)
+
N
N
O
4-
O
42-
OH
O-
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Protonation of the NH2 group of
fluorescence (Fig S20 in the ESI) and induces pronounced
changes in the absorption spectra (Fig 4). Processing of the UV‐
6 significantly quenches
Vis and fluorescence titration data for 6 and 6H measured in the
acidic region allowed for the estimation of the amine pKa. The
estimated values from the UV‐Vis and fluorescence data are
somewhat different (Table 5), and imperfect fit was obtained
from the fluorescence titration. The discrepancy may be due to
the different pKa values of the ground and excited singlet states.
It is known that ammonium ions become more acidic in the
excited S1 state.66‐68 The same trend in the UV‐Vis and
Laser flash photolysis (LFP) and properties of triplet excited states
Phthalimides substituted with electron‐donating substituents
participate in PET reactions from S1 state, as indicated by
fluorescence measurements. However, triplet excited state
reactions for imides are more ubiquitous.1,2 Furthermore,
regardless of the phthalimide substitution, acetone
sensitization should give rise to phthalimide triplet excited
states that undergo decarboxylation. To investigate triplet
fluorescence
spectra
was
observed
for
excited state properties of
1‐6, LFP measurements were
conducted. Similar to the fluorescence measurements, LFP was
conducted for CH3CN and CH3CN‐H2O solutions. The samples
were excited at 266 nm, and the solutions were purged with N2
or O2, where a difference was anticipated since O2 quenches
triplet excited states, radicals and radical anions. For all data see
Figs S24‐S41 in the ESI.
0.8
0.6
0.4
0.2
0.0
solution pH value increase
Transient absorption spectra measured in CH3CN are shown in
Fig 5. For
exhibited 30‐50 nm batochromically shifted maxima compared
to unsubstituted and the typical N‐alkylphthalimides (λmax
2‐4, triplet excited states were detected, which
1
=
330 nm).42,69 The assignment to triplet excited states was based
on the quenching with O2 (Table 6). The efficiency for the
formation of triplet states was estimated by comparing the
intensity of transients immediately after the laser pulse with the
one of N‐methylphthalimide (ΦISC = 0.8)70 for the optically
matched solution at the excitation wavelength. However, due
250
300
350
400
450

/ nm
to batochromically shifted maxima for
approximation. Triplet excited state for
about 50% less efficiently than for
intersystem crossing (ISC) is almost three times less efficient
than for . Furthermore, the triplet excited state lifetime (τ) for
and is about 30 times shorter than for methoxy‐substituted
phthalimide . Although the reason for the shorter lifetime of
triplet excited state of
difference between the τ of
2
‐
3
4
, it is only a coarse
and are formed
the
0.08
0.06
0.04
0.02
0.00
4
1, whereas for
2
Experiment
Model
1
2
4
3
2
(compared to
and
1
) is not clear, the
3
4 may be due to
photoinduced intermolecular H‐atom transfer from phenolic
group to carbonyl, an ubiquitous quenching mechanism for
triplet excited carbonyl compounds.71 Transient absorption
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5
data for nitro compound
5, reveals a completely different
pH
behavior. Only a very short‐lived transient was detected,
absorbing at ≈ 330 and 350‐450 nm, most probably
corresponding to the triplet excited state of 5, in line with the
spectra of the triplet excited states of nitrobenzene derivatives.
Thus, the nitro group significantly changed the photophysical
properties of the molecule. On the other hand, the amino group
Figure 4. Absorption spectra of 6 (c = 1.75 × 10‐5 mol dm‐3) in CH3CN‐H2O (1:9 ‐ v/v) in the
presence of KCl (c = 2 mol dm‐3) at different pH values (left) and dependence of
absorbance at 370 nm on pH (right); black dots are the measured values and the red line
corresponds to the data calculated by nonlinear regression analysis with Specfit
software. The measurements were conducted at 25 °C.
8 | J. Name., 2012, 00, 1‐3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 9 of 17
New Journal of Chemistry
Please do not adjust margins
Journal Name
ARTICLE
1
2
3
4
5
6
7
8
9
in
6
gave rise to the formation of S1 CT state, as revealed from decay is pH dependent, the observed quenching rate constant
View Article Online
DOI: 10.1039/D0NJ03465G
the fluorescence measurement, that very inefficiently may tentatively be assigned to the sum of rate constants for the
undergoes ISC. Nevertheless, the triplet state of 6 was detected reaction of protonation and decarboxylation of the radical
(broad band 400‐600 nm), assigned based on literature anion, which are subsequent to fast equilibration of the triplet
precedent,72 albeit its signal was weak due to inefficient excited state and the triplet charge transfer species formed by
formation.
PET (see Scheme 2).
T
able 6. Properties of the triplet excited state obtained by LFP.
b
Comp.
τ (CH3CN)
/ μs a
13 f
ΦISC
kq O2 / dm3
τ (pH 4) / τ (pH 7) /
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
mol‐1 s‐1 c
μs d
μs e
0.34 ±
0.04
0.020
0.015
2
3
4
5
6
f
1
0.22 f
2×109
1.8 ± 0.4
1.2
150 ± 50
0.4 ± 0.3
6 ± 1
30‐100
0.023 ±
0.006
6.0 ± 0.6
20‐30
7 ± 1
2
3
0.62 ±
0.02
0.08 ±
0.02
(5.5
0.90 ±
0.09
± 0.7) × 108
0.010
A
17 ± 1
0.15 ±
0.05
(3.1
± 0.1) × 109
0.020 ±
0.002
12 ± 1
30‐60
0.30 ±
0.05
0.005
0.000
4
0.45 ±
0.05
0.15 ±
0.05
(3.6
± 0.2) × 109
300
350
400
450
500
550
600
20‐100
 / nm
6.5 ± 0.3
‐
5
6
< 0.01
0.29 ±
0.01
‐
‐
‐
‐
‐
Figure 5. Transient absorption spectra of 2‐6 in N2‐purged CH3CN, recorded immediately
after the laser pulse, within 10 ns (λ = 266 nm).
(1.2
± 0.1) ×1010
‐
6.4 ± 0.4
To probe for the PET from the triplet excited state, LFP
measurements were conducted for 1‐4, at pH 4 and 7, where
the adamantane carboxylic acid is protonated or not. We
anticipated that the formation of the carboxylate would quench
the triplet excited state of the phthalimide group, and result in
a
Lifetime of the triplet excited state in N2‐purged CH3CN solution. The average
value from at least three decays measured at three wavelengths is reported, and
b
the errors correspond to the maximum absolute deviation. Quantum yield of
intersystem crossing in CH3CN estimated by comparison of the intensity of the
transient absorption immediately after the laser pulse of the optically matched
solution at the excitation wavelength (266 nm) with the one of N‐
methylphthalimide (ΦISC = 0.8).70 c Rate constant for the quenching of the triplet
excited state by O2 in CH3CN. d Decay times for the transients detected in N2‐purged
CH3CN‐H2O (1:1 ‐ v/v) in the presence of potassium phosphate buffer (c = 0.05 mol
dm‐3) at pH = 3.8. e Decay times for the transients detected in N2‐purged CH3CN‐
H2O (1:1‐ v/v) in the presence of potassium phosphate buffer (c = 0.05 mol dm‐3)
at pH = 7.0. f Taken from reference 44.
the formation of phthalimide radical anion. For
1 we detected
weak transients, which did not provide spectra with a clear
pattern for two anticipated species. However, the decay of the
triplet at pH 7 was faster compared to the one detected at pH
4. An additional long‐lived transient was detected (τ = 150 μs)
and tentatively assigned to the radical anion or ketyl radical
based on comparison with the literature precedent.44,68 From
the two different rate constants for the decay of the
phthalimide triplet at pH 7 and 4 (See Fig S25 and S26), the
quenching rate constant can be calculated kq = (2.5 ± 0.5) × 106
s‐1, which is too small for PET.
Compound 2 has two carboxylic functional groups. Therefore, it
is plausible that efficient PET takes place only if the adamantane
carboxylic acid is deprotonated, and the one at the phthalimide
is not. Thus, at pH 7 it is likely that
2 exists in an equilibrium
Phthalimides are known to undergo H‐abstraction reactions
giving rise to ketyl‐like radicals.1 To rule out that the detected
transients in the photochemistry of
between mono‐ and dianionic species, that may have different
triplet excited state lifetimes because the monoanionic species
undergoes PET and the dianionic does not. Indeed, at pH 7, we
detected three transients with the lifetimes of 0.4, 6 and 30‐100
μs (Figure S31). Tentatively, the shorter lived two transients
may be assigned to the triplet excited states (quenched or not),
whereas the longest lifetime probably corresponds to the
phthalimide radical anion or ketyl radical. From the two
lifetimes at pH 4 and 7 (0.9 and 0.4 μs, respectively), the
estimated quenching rate constant is kq ≈ 1 × 106 s‐1, but the
weakness of the transient signals precluded the precise
determination of decay kinetics. Note the small value of the
estimated rate constant, which cannot be attributed to PET, and
1 correspond to a ketyl
radical, which undergoes deprotonation equilibrium, and thus,
exhibit different lifetime depending on pH, we conducted
measurements for N‐methylphthalimide in CH3CN‐H2O without
buffer and in the presence of phosphate buffer at pH 4 and 7
(See Figs S42‐S44). The transient absorption spectra at two pH
values look the same with the maximum at 330 nm. The decay
of the transient was fit to a sum of two exponentials with τ ≈ 10
μs and τ ≈ 50 μs, giving similar values for the decay times
regardless of conditions. The short decay time (10 μs) was
assigned to the triplet excited state of the phthalimide, whereas
the long one (50 μs) probably corresponds to the ketyl radical
that is not acidic enough to be deprotonated at pH 7, and
therefore, does not exhibit pH‐dependent decay. Consequently,
which is in accord with the above discussion for 1. However, the
transient absorption spectra show different maxima depending
on pH (Fig 6), in agreement with the detection of the triplet
based on the results for N‐methylphthalimide and
1 where the
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1‐3 | 9
Please do not adjust margins

Please do not adjust margins
New Journal of Chemistry
Page 10 of 17
ARTICLE
Journal Name
1
2
3
4
5
6
7
8
9
excited state (at pH 4) and the phthalimide radical anion or the photochemical reactivity of the chromophore. VNiewitArorticlegrOonulinpe
ketyl radical (at pH 7).
changed the pathways, making the DmOIo: 1le0c.1u0l3e9/Dm0NoJr0e34l6i5kGe
nitrobenzene derivative that has a very short triplet excited
state lifetime due to the fast radiationless deactivation
producing ground state molecules and the intrinsic reactivity of
the nitro group. Therefore, in addition to 5H which was formed
inefficiently, some unidentified products resulting from
nitrobenzene photochemistry were detected, whose isolation
was impossible due to instability. The measured quantum yield
0.006
0.005
0.004
pH 4
pH 7
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
for
5 reported in Table 2 corresponds to both processes,
0.003
A
decarboxylation and those typical for the nitrobenzene
chemistry. On the other hand, the strong electron‐donating
amino group in
that PET from the carboxylate was not thermodynamically
feasible. The OH group in also has an electron‐donating
0.002
0.001
0.000
6 decreased the energy of the triplet state so
4
300
350
400
450
500
550
600
character, but the PET is still feasible. However, PET initiated
processes were inefficient due to concomitant H‐transfer
 / nm
Figure 6. Transient absorption spectra for 2, detected 200 ns after the laser pulse in N2‐ reactions from the phenolic OH that quenches triplet excited
purged CH3CN‐H2O (1:1 ‐ v/v) in the presence of phosphate buffer at pH 4 and 7.
states.
Lifetimes of triplet excited states at different pH values where
It is plausible that compounds 3 and 4 undergo PET from S1 and
PET was anticipated (or not) were analyzed providing apparent
rate constants, which are too small to correspond to PET.
Therefore, we propose that equilibration of the excited state
with the CT species by PET and back electron transfer takes
place, and the observed rate constant corresponds to the decay
of the radical anion by sum of rate constants for protonation
and decarboxylation. Such a mechanistic scenario is also in
accord with the observed primary deuterium isotope effect on
the photochemical conversion (see Scheme 3).
Upon direct excitation of 1‐6 at 300 nm, the phthalimide S1
states are populated, that contrary to N‐alkylphthalimides do
not undergo very efficient ISC. Properties of the excited states
for 1 3, 4 and 6 are given in Table 7. Nevertheless, triplet excited
state can be populated to some extent after direct excitation,
which is followed by decarboxylation, but it is plausible that the
reaction would be less efficient. Indeed, for
lower in CH3CN than in the presence of acetone. The same trend
probably does not hold for , that is prone to H‐transfer
reactions. This process significantly quenches triplets, and less
likely singlet excited states. Moreover, the charge‐transfer
triplet state of
inefficiently (or not at all) the PET with the carboxylate.
Phthalimides with electron‐donating substituents and 6 are
fluorescent, and due to low ΦISC in aqueous solvent the
photoreactions from S1 compete with other deactivation
pathways. Note that ΦF for 3, and somewhat for 6, decreases in
the pH region where the carboxylic group is deprotonated.
Concomitantly, there is an increase in the photodecarboxylation
efficiency for these two compounds, strongly indicating that the
PET and decarboxylation dominantly take place from S1.
from the triplet excited state. Therefore, it is anticipated that
the phthalimide radical anion and ketyl radicals can be formed
by two rate constants, fast from S1 and slower from the triplet
state. However, comparison of the transient absorption spectra
at two pH values for
3 is not very informative, although the
transient assigned to the triplet excited state decays faster at
pH 7. The estimated lifetimes of 12 or 6 μs correspond to kq = (8
± 2) × 104 s‐1, which is too slow for PET, similar to rates estimated
for
detected at both pH values for
the scope of this manuscript. On the other hand, a pronounced
difference between the spectra for at two pH values was
1
and
2
. In addition, a very short‐lived transient (≈ 20 ns) was
3, whose assignment is beyond
4
observed, where a new absorption band at ≈ 475 nm was
detected at pH 7, tentatively attributed to the phthalimide
radical anion or ketyl radicals. However, no significant
difference in decay kinetics for the triplet excited state was
observed, which may be due to inefficient decarboxylation
1‐3 and 5 the ΦR is
4
photochemical reaction for triplet
4 (Table 2).
Photochemical reaction mechanisms
6 has too low energy and undergoes very
Preparative irradiations, fluorescence and LFP measurements
provided a full description of photophysical and photochemical
pathways for differently substituted phthalimides, whose
understanding is important for the rational applications in
synthesis. Results obtained from the experiments using acetone
as a solvent will be discussed first. Upon irradiation in acetone,
sensitization takes place and phthalimide triplet excited states
are populated that react in intramolecular PET reactions with
the deprotonated carboxylic group, leading efficiently to the
3,
4
Shortening
of
the
singlet
excited
state
decarboxylation products 1H
compounds , where the substituents on the phthalimide
significantly affected the photophysical properties and
‐
6H
.
The exceptions are
4‐6
10 | J. Name., 2012, 00, 1‐3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 11 of 17
New Journal of Chemistry
Please do not adjust margins
Journal Name
Table 7. Photophysical properties of
ARTICLE
1
2
3
4
5
6
7
8
1
,
3
,
4
, and
6
View Article Online
DOI: 10.1039/D0NJ03465G
Comp.
ES1 / kJ mol‐1 a
ΦF (CH3CN) b
< 10‐4
0.024 ± 0.002
τF (CH3CN) /ns c
n.d.
1.68 ± 0.01
16.0 ± 0.4
2.33 ± 0.8
14.7 ± 0.3
17.2± 0.2
ΦISC (CH3CN) d
0.22
0.15 ± 0.05
τT (CH3CN) /μs e
1
3
356
310
1.2
17 ± 1
4
6
308
260
0.037 ± 0.003
0.66 ± 0.03
0.15 ± 0.05
n.d.
0.45 ± 0.05
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0.29 ± 0.01
6.4 ± 0.4
a Energy of the singlet excited state. b Quantum yield of fluorescence in CH3CN. c Lifetime of singlet excited state. d Quantum yield of intersystem crossing. e Lifetime of
triplet excited state.
lifetimes allowed for the estimation of the rate constants for
PET, which is for
that PET occurs also from the S1 state of
3
two order of magnitudes larger. It is plausible
, but the reaction may
Conclusions
4
proceed efficiently only in the narrow pH region when the
carboxylic functional group is deprotonated and the phenol is
not. Formation of the phenolate probably significantly hampers
the PET due to increase of the electron density on the
phthalimide. LFP data allowed for the detection of triplet
Photophysical properties and photochemical reactivity of
adamantane carboxylic acid covalently attached to differently
substituted phthalimides were investigated and put in
connection with redox properties. Quantum yields of the
reaction (ΦR) greatly depend on pH, being much higher when
the carboxylic acid is deprotonated, in line with the
photoinduced electron transfer (PET) taking place from the
carboxylate to the phthalimide in the excited state. Compounds
bearing electron donor groups NH2, OH and OCH3 are
fluorescent (ΦF = 0.02‐0.66). For the compounds substituted
with OH and OCH3, the PET takes place from both singlet and
triplet excited states. The quantum yield of the triplet
population by intersystem crossing (ΦISC) depends on the
substituent at the phthalimide, and is much lower than for N‐
methylphthalimide. Therefore, the phthalimide substituents
have a profound effect on the photophysics and spin selectivity
in the photochemical reactions. LFP measurements in aqueous
solutions at different pH values allowed for the detection of the
triplet excited states, whose lifetimes depend on pH. Based on
LFP and deuterium labeling experiments in preparative
irradiation, a photochemical reaction mechanism was proposed
that includes equilibration of the excited state (singlet or triplet)
with the CT state by PET and back electron transfer,
respectively. The shortening of the triplet lifetimes at pH 7
provided quenching constants which were assigned to the sum
of rate constants corresponding to the decay of phthalimide
radical anion by protonation and decarboxylation.
Consequently, we have compelling evidences that after initial
PET, back electron transfer is in competition with the
subsequent reactions involving decarboxylation or protonation
of the phthalimde radical anion, determining the overall
excited states of
3 and 4, which can also undergo PET, as
indicated from the irradiation experiments with acetone as
sensitizer. The spin selectivity for the PET reaction in these
compounds depends on the rate constants for fluorescence,
PET and ISC. Apparently, electron donors on the phthalimide
affect the photophysics in a way to make the ISC inefficient
(probably by slowing down the nonradiative deactivation) and
thus, facilitate the reactions from singlet excited states. Note
that the photoreactions from S1 still involve PET and back
electron transfer because 3‐* (S1), [3‐CT]* and 3‐AN are not
mesomeric structures since the system is not conjugated.
Photochemical reactions in CH3CN gave in addition to simple
decarboxylation products 1H‐6H, alcohols, peroxides, and
ketones. Whereas alcohols and peroxides are formed by
trapping the adamantyl radical by O2, the mechanism for the
formation of ketones needs further discussion. We postulated
two pathways for the formation, that involve back electron
transfer from the phthalimide radical anion to adamantyl
radical giving anion, or radical pathways (Scheme 3). Taking the
potential of the adamantyl radical reduction to anion (Ered = ‐
2.19 V vs. Fc0/Fc+),73,74 and the measured reduction potentials
for phthalimides (Table 3), it is unlikely that back electron
transfer takes place, except for those that have electron‐
donating substituent. Thus, the carbanion mechanism is
plausible only for derivatives with electron donating
substituents, such as OCH3. Nevertheless, finding pathways that
involve PET, subsequent chemical transformations and back
electron transfer, which can initiate more chemical steps may
have important impact in the design of new photocatalytic
pathways, where phthalimide derivatives would be used as
catalysts.
efficiency of the decarboxylation process.
A thorough
understanding of different pathways in phthalimide
photophysics and photochemistry presented herein highlights
plausible processes that largely depend on the phthalimide
substituents and the solution pH, understanding of which is
important for the rational use of phthalimide derivatives in
organic synthesis and photocatalytic processes.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1‐3 | 11
Please do not adjust margins

Please do not adjust margins
New Journal of Chemistry
Page 12 of 17
ARTICLE
Journal Name
1
2
3
4
5
6
7
8
9
(d, J = 8.2 Hz, 1H), 7.22 (d, J = 2.1 Hz, 1H), 7.13 (dd, J = 8.2, 2.1
Experimental
General
View Article Online
Hz, 1H), 3.91 (s, 3H), 2.53‐2.59 (m, 4H), 2D.4O5I:(1d0,.1J0=391/2D.01NHJ0z3,426H5G),
2.32 (br s, 2H), 2.14 (s, 3H), 1.86 (d, J = 12.1 Hz, 2H), 1.76‐1.82
(m, 3H), 1.66 ppm (d, J = 12.5 Hz, 1H); 13C NMR (150 MHz,
CDCl3): δ = 212.5 (s), 169.7 (s), 169.6 (s), 164.7 (s), 134.5 (s),
124.5 (d), 124.0 (s), 120.2 (d), 107.1 (d), 60.3 (s), 56.2 (q), 48.7
(s), 40.5 (t), 39.5 (t, 2C), 37.2 (t, 2C), 35.4 (t), 29.5 (d, 2C), 24.7
ppm (q); IR (neat): ν = 2909, 2853, 1762, 1696, 1620, 1490,
1355, 1283, 1079, 745 cm‐1; HRMS (MALDI): m/z calcd for
C21H23NO4+H+: 354.1705; found: 354.1706.
1H and 13C NMR spectroscopic data were recorded at room
temperature on Bruker Avance 300 MHz or Bruker Avance 600
MHz spectrometer. CDCl3, DMSO‐d6 or CD3OD was used as
deuterated solvent. TMS (1H NMR) or deuterated solvent itself
(13C NMR) was used as internal reference. Chemical shifts were
reported in ppm. 13C NMR spectra were recorded in the APT
technique or fully decoupled (denoted as COM in the ESI). IR
spectra were recorded with a FT‐IR ABB Bomem MB102
(samples as KBr pellets) or FT‐IR ATR PerkinElmer UATR Two
spectrometer (neat samples). The frequencies of the
characteristic bands are reported in cm‐1. Melting points were
measured on an Original Kofler Mikroheiztisch apparatus and
were not corrected. HRMS data were obtained on Applied
Biosystems DE STR MALDI‐TOF/TOF instrument. Irradiation
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Preparative photochemistry of 4 without a sensitizer
Following the general procedure irradiation of
4 followed by
purification gave 4H (37 mg, 42%), 4COCH3 (4 mg, 4%), 4OH (15
mg, 16%) and 4OOH (8 mg, 8%) as colorless solids, as well as
recovered
3‐Acetyl‐1‐[
4
(30 mg, 29% of starting amount).
‐(4'‐hydroxy)phthalimido]adamantane (4COCH3)
N
m.p. 128‐131 °C; 1H NMR (300 MHz, CD3OD): δ = 7.55‐7.61 (m,
1H), 7.03‐7.11 (m, 2H), 2.53 (br s, 2H), 2.43‐2.57 (m, 4H), 2.28
(br s, 2H), 2.15 (s, 3H), 1.76‐1.90 (m, 5H), 1.71 ppm (d, J = 12.4
Hz, 1H); 13C NMR (75 MHz, CD3OD): δ = 215.0 (s), 171.1 (s), 170.9
(s), 164.8 (s), 135.8 (s), 125.6 (d), 123.6 (s), 121.6 (d), 110.0 (d),
61.1 (s), 49.9 (s), 41.5 (t), 40.5 (t, 2C), 38.3 (t, 2C), 36.3 (t), 30.9
(d, 2C), 24.7 ppm (q); IR (neat): ν = 3331, 2915, 2852, 1761,
1683, 1605, 1467, 1333, 1315, 1301, 1213, 1075, 978, 749, 637
cm‐1; HRMS (MALDI) m/z calcd for C20H21NO4+H+: 340.1549;
found: 340.1541.
experiments were performed in
a
Rayonet RPR‐100
photoreactor equipped with 10 lamps with the maximum
output at ≈ 300 nm (1 lamp 8 W). During the irradiations, the
irradiated solutions was placed in a quartz vessel (φ = 3.0 cm),
that was placed in the center of the irradiated chamber, 7.5 cm
separated from each lamp surrounding the vessel. The
irradiated solutions were continuously purged with Ar and
cooled by a tap water finger condenser. Solvents for the
irradiations were of HPLC purity. Chemicals were purchased
from the usual commercial sources and used as received.
Solvents for chromatographic separations were used as is from
the supplier (p.a. or HPLC grade) or purified by distillation
(CH2Cl2).
3‐[
(
N‐(4'‐Hydroxy)phthalimido]adamantane‐1‐hydroperoxide
4OOH) m.p. 218‐220 °C; 1H NMR (300 MHz, CD3OD): δ = 7.53‐
7.61 (m, 1H), 7.02‐7.10 (m, 2H), 2.56 (br s, 2H), 2.29‐2.51 (m,
6H), 1.76 (d, J = 11.1 Hz, 2H), 1.66‐1.80 (m, 3H), 1.60 ppm (d, J =
12.7 Hz, 1H); 13C NMR (150 MHz, CD3OD) δ = 171.0 (s), 170.8 (s),
164.8 (s), 135.8 (s), 125.6 (d), 123.6 (s), 121.6 (d), 110.0 (d), 81.0
(s), 62.6 (s), 43.9 (t), 40.4 (t, 2C), 40.0 (t, 2C), 36.3 (t), 32.0 ppm
(d, 2C); IR (neat): ν = 3176, 2915, 2852, 1759, 1682, 1607, 1464,
1336, 1304, 1263, 1081, 978, 745, 639 cm‐1; MS (+ESI): m/z 330;
MS (‐ESI): m/z 328; HRMS (MALDI) m/z calcd for C18H19NO5+Na+:
352.1161; found: 352.1152.
Preparative photochemistry of substituted phthalimide
derivatives ‐ general procedure
A glass vessel (200 mL) was charged with the compound
(0.30 mmol) and K2CO3 (0.30 mmol, 1 equiv. for , and 0.15
2‐6
2
mmol, 0.5 equiv. for other compounds); dissolved in CH3CN or
acetone (150 mL) and H2O (50 mL). The resulting solution was
purged with Ar for 30 min and then irradiated in a reactor (10
lamps, 300 nm). H2O (150 mL) was added, and an extraction
with CH2Cl2 (5 × 100 mL) was carried out. The organic layers
were dried over anhydrous MgSO4 and filtered, and the solvent
was removed on a rotary evaporator. Chromatography of the
remaining residue on a silica gel column using Et2O in CH2Cl2 (0
→ 100%) or EtOAc in CH2Cl2 (0 → 100%) as eluent afforded pure
photoproducts.
Preparative photochemistry of 5 without a sensitizer
Following the general procedure for preparative
photochemistry of substituted phthalimide derivatives,
irradiation of
5OH (15 mg, 15%) and 5OOH (6 mg, 6%) as colorless solids, as
well as recovered (56 mg, 50% of starting amount).
Characterization of
5H and 5OH was reported.57
3‐[ ‐(4'‐Nitro)phthalimido]adamantane‐1‐hydroperoxide
5OOH) m.p. 128‐131 °C; H NMR (300 MHz, CDCl3): δ = 8.52‐
5, followed by purification, gave 5H (8 mg, 8%),
5
5
,
Preparative photochemistry of 2 and 3 in acetone as a sensitizer
N
1
Following the general procedure, irradiation of
purification gave 2H (81 mg, 81%) as colorless solid. Irradiation
of , followed by purification, gave 3H (88 mg, 88%) as colorless
solid. Characterization of 2H
and 3H was reported.57
Preparative photochemistry of 3 without a sensitizer
2 followed by
(
8.62 (m, 2H), 7.97 (d, J = 7.9 Hz, 1H), 7.61 (s, 1H), 2.64 (s, 2H),
2.34‐2.54 (m, 6H), 1.94 (d, J = 11.7 Hz, 2H), 1.67‐1.83 (m, 3H),
1.61 ppm (d, J = 12.7 Hz, 1H); 13C NMR (75 MHz, CDCl3): δ = 167.5
(s), 167.2 (s), 151.9 (s), 136.2 (s), 133.2 (s), 129.2 (d), 124.2 (d),
118.4 (d), 80.8 (s), 62.9 (s), 42.5 (t), 39.2 (t, 2C), 38.6 (t, 2C), 35.0
(t), 30.5 ppm (d, 2C); IR (KBr): ν = 3443, 2918, 2862, 1712, 1542,
1344, 1326, 1309, 1089, 720 cm‐1; HRMS (MALDI): m/z calcd for
C18H18N2O6+H+: 359.1243; found: 359.1243.
3
2
,
, 3
Following the general procedure irradiation of
purification gave 3H (66 mg, 70%), 3COCH3 (13 mg, 12%) and
3OH (4 mg, 4%), all as colorless solids. Characterization of
and 3OH was reported.57
3 followed by
3, 3H
3‐Acetyl‐1‐[
(
N‐(4'‐methoxy)phthalimido]adamantane
3COCH3) m.p. 102‐104 °C; 1H NMR (600 MHz, CDCl3): δ = 7.66
12 | J. Name., 2012, 00, 1‐3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 13 of 17
New Journal of Chemistry
Please do not adjust margins
Journal Name
ARTICLE
1
2
3
4
5
6
7
8
Irradiation of 3 in CH3CN‐D2O
photomultiplier, which was connected to a Picoquant Timeharp
100 electronics having 36 ps/channel time resolution. For the
instrument response function, silica in H2O was used as
scatterer. The decays were fit with a sum of exponentials
according to Eq S4 using Picoquant FluoFit software.
UV‐Vis pH titration for 4 and 4H
View Article Online
DOI: 10.1039/D0NJ03465G
A glass vessel (200 mL) was charged with
3 (27.4 mg, 0.077
mmol), K2CO3 (5.40 mg, 0.039 mmol, 0.5 equiv.), dissolved in
CH3CN (37.5 mL) and H2O (12.5 mL) or D2O (12.5 mL). The
resulting solution was purged with Ar for 30 min and then
irradiated in a reactor (15 lamps, 300 nm) for 10 minutes. HPLC
9
analysis of both solutions was carried out after the irradiation. A set of quartz cells (2.5 mL) were charged with the solution of
Removal of the solvent by evaporation and chromatography of
(c = 2.72 × 10‐5 mol dm‐3) or 4H (c = 3.30 × 10‐5 mol dm‐3) in
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4
the remaining residue on a silica gel column by using CH3CN‐H2O (1:9 ‐ v/v) mixture in the presence of phosphate
Et2O/CH2Cl2 (0 → 100%) as eluent gave 3H and 3D. The amount buffer (KH2PO4‐K2HPO4, c = 0.05 mol dm‐3). Prior to the UV‐Vis
of deuterium in 3D was determined by HPLC‐MS analysis.
Photolyses of 2, 4 and 6 at different pH values
measurements, the pH of the solution was measured by a pH
meter (Mettler, Toledo, Seven Multi). The measurements were
conducted at 25 °C. The data were processed by multivariate
nonlinear regression analysis by fitting to one‐step protonation
model with the Specfit software.
Quartz test tubes were charged with the solution (10 mL) of the
phthalimide derivatives (c = 10‐3 mol dm‐3) in CH3CN‐H2O (1:1 ‐
v/v) or acetone‐H2O (1:1 ‐ v/v) in the presence of phosphate
buffer (H3PO4‐KH2PO4, KH2PO4‐K2HPO4 or K2HPO4‐K3PO4) with
6
different pH values (total phosphate c = 0.05 mol dm‐3). The A vial (10 mL) was charged with the solution of (c = 1.75 × 10‐
resulting solutions were purged with N2 for 30 minutes and then 5 mol dm‐3) or 6H (c = 1.84 × 10‐5 mol dm‐3) in CH3CN‐H2O (1:9 ‐
irradiated in a reactor (8 lamps, 300 nm). The composition of v/v) in the presence of KCl (c = 2 mol dm‐3). After measurement
UV‐Vis and fluorescence titration of 6 and 6H
the irradiated solution was analyzed by HPLC.
Photodecarboxylation quantum yields
of the solution pH by a pH meter, a part of the solution was
transferred to a quartz cell (2.5 mL), absorption and emission
spectra were recorded, and then the solution was returned to
the vial. To the solution aliquots of H2SO4 (c = 0.5 mol dm‐3) were
added, and after each addition pH was measured, as well as
absorption and emission spectra. For the fluorescence
measurements, excitation slit was set to a bandpass of 5 nm and
emission slit to 10 nm. The excitation was 370 nm, and the
emission was collected in the range 380‐730 nm. The
measurement was conducted at 25 °C. The data was processed
by multivariate nonlinear regression analysis by fitting to one‐
step protonation model with the Specfit software.
Quartz test tubes were charged with the solution of phthalimide
1‐6
(c = 10‐3 mol dm‐3, 10 mL) in CH3CN‐H2O (1:1 ‐ v/v) or
acetone‐H2O (1:1 ‐ v/v) in the presence of phosphate buffer
(KH2PO4‐K2HPO4, total phosphate c = 0.05 mol dm‐3) at pH 4.5
or pH 8.3. Phthalimide
quantum yield (ΦR = 0.3)42,60 was used as a secondary
actinometer. To the solution of
(c = 10‐3 mol dm‐3, 10 mL),
7 with the known photodecarboxylation
7
K2CO3 (0.5 equiv.) was added. The resulting solutions were
purged with N2 for 30 min and then irradiated in a reactor (8
lamps, 300 nm in acetone or 254 nm in CH3CN). The
composition of the irradiated solutions was determined by
Fluorescence pH titration of 3 or 6
HPLC. The photodecarboxylation quantum yields were A set of quartz cells (2.5 mL) were charged with the solution of
calculated according to Eq S1.
3 6
(c = 9.13 × 10‐6 mol dm‐3) or (c = 1.76 × 10‐5 mol dm‐3) in
UV‐Vis and fluorescence measurements
CH3CN‐H2O (1:9 ‐ v/v) in the presence of phosphate buffer
(KH2PO4‐K2HPO4, c = 0.05 mol dm‐3). Prior to the fluorescence
measurements, the pH of the solution was measured by a pH
meter. For the measurement of emission spectra, the excitation
slit was set to a bandpass of 5 nm, and the emission slit to 10
UV‐Vis and fluorescence measurements were conducted in
CH3CN or in CH3CN‐H2O (1:1 ‐ v/v) or (1:9 ‐ v/v) in the presence
of potassium phosphate buffer (c = 0.05 mol dm‐3) at pH 4.5 and
8.3. For the fluorescence measurements, the solutions were
purged with Ar for 20 min. Emission spectra were recorded in
the range 360‐690 nm by exciting samples at three different
wavelengths (330, 340 and 350 nm). The excitation slit was set
to the bandpass of 2.5 or 5 nm, and the emission to 5 nm.
Fluorescence quantum yields were determined using quinine
sulfate in 0.5 mol dm‐3 H2SO4 (ΦF = 0.55) or 9‐cyanoanthracene
in cyclohexane (ΦF = 0.92) as reference.59 Quantum yields were
calculated according to Eq S3. The measurements were
conducted at 25 °C.
nm, or both slits to 5 nm in case of
350 nm, and the emission was collected in the range 360‐610
nm. The excitation for was 370 nm, and the emission was
3. The excitation for 3 was
6
collected in the range 380‐730 nm. The measurements were
conducted at 25 °C. The data was processed by multivariate
nonlinear regression analysis by fitting to one‐step protonation
model with the Specfit software.
Laser flash photolysis
The measurements were performed on a LP980 Edinburgh
Instruments spectrometer. For the excitation the fourth
harmonic of a Qsmart Q450 Quantel YAG laser was used. The
pulse duration was 7 ns. For the measurement of spectra, the
energy of the laser pulse at 266 nm was set to 20 mJ, and in
some cases it was lower (8 mJ). Decay traces were measured
with lower energy of the pulse (8 mJ) to avoid quenching by
triplet‐triplet annihilation. Absorbances of the solutions at the
Time‐resolved fluorescence measurements
Time‐resolved fluorescence was measured with a time‐
correlated single‐photon counting instrument described in
literature previously.75 The samples were excited with a pulsed
diode laser at 368 nm. The pulse width was
fluorescence signals were monitored at several wavelengths
with Hamamatsu R3809U‐51 microchannel plate
∼70 ps. The
a
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1‐3 | 13
Please do not adjust margins

Please do not adjust margins
New Journal of Chemistry
Page 14 of 17
ARTICLE
Journal Name
1
2
3
4
5
6
7
8
9
6
U. C. Yoon, P. S. Mariano, in CRC HandbooVkiewofArtOiclregOannlinice
Photochemistry and Photobiology, 2nd Ed. (Eds.: W. Horspool,
excitation wavelength were set to 0.3. For the measurements of
decay kinetics, static cells were used, which were frequently
exchanged to assure no absorption of light by photoproducts.
The spectra were measured in flow cells. The solutions were
purged for 20 min with N2 or O2 prior to the measurements.
Cyclic voltammetry
F. Lenci), CRC Press, Boca Raton, 2004DOI: 10.1039/D0NJ03465G
.
7
8
U. C. Yoon, P. S. Mariano, Acc. Chem. Res., 2001, 34, 523‐533.
A. G. Griesbeck, A. Henz, K. Peters, E.‐M. Peters, H. G. von
Schnering, Angew. Chem. Int. Ed., 1995, 34, 474‐476.
A. G. Griesbeck, T. Heinrich, M. Oelgemöller, J. Lex, A. Molis,
J. Am. Chem. Soc., 2002, 124, 10972‐10973.
9
Cyclic voltammetric measurements were carried out using the
computer‐controlled electrochemical system Autolab PGSTAT
30 (Eco‐Chemie, Utrecht, Netherlands) equipped with GPES
software. A three‐electrode system (Methrom, Switzerland)
with glassy carbon electrode (GCE) of 3 mm in diameter as a
working electrode, Ag/AgCl (in 3 mol dm‐3 NaCl) as a reference
electrode and a platinum wire as a counter electrode were
used. All potentials were expressed versus the Ag/AgCl (in 3 mol
dm‐3 NaCl) reference electrode. For the supporting electrolyte,
0.1 mol dm‐3 NBu4PF6 in CH3CN was used. The supporting
electrolyte was placed in the electrochemical cell and the
required aliquot of the standard analyte solution was added.
The solution in the electrochemical cell was deaerated with high
purity nitrogen for 10 min before measurement, and the
nitrogen blanket was maintained thereafter. Before each run,
the glassy carbon (GC) working electrode was polished with
diamond suspension in spray (grain size 6 µm) and rinsed with
ethanol and deionised water. Cyclic voltammetry on GC
electrode was performed for all compounds using a potential
step increment of 2 mV and a scan rate of 500 mV s‐1 (except for
10 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, J. Am. Chem.
Soc., 2003, 125, 10664‐10671.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
11 A. Soldevilla, A. G. Griesbeck, J. Am. Chem. Soc., 2006, 128
,
16472‐16473.
12 A. G. Griesbeck, W. Kramer, J. Lex, Angew. Chem. Int. Ed.,
2001, 40, 577‐579.
13 Y.‐J. Lee, D.‐H. Ahn, K.‐S. Lee, A. R. Kim, D. J. Yoo, M.
Oelgemöller, Tetrahedron Lett., 2011, 52, 5029‐5031.
14 F. Hatoum, S. Gallagher, M. Oelgemöller, Tetrahedron Lett.,
2009, 50, 6593‐6596.
15 F. Hatoum, J. Engler, C. Zelmer, J. Wißen, C. A. Motti, J. Lex, M.
Oelgemöller, Tetrahedron Lett., 2012, 53, 5573‐5577.
16 A. G. Griesbeck, M. Oelgemöller, Synlett, 1999, 492‐494.
17 F. Hatoum, S. Gallagher, L. Baragwanath, J. Lex, M.
Oelgemöller, Tetrahedron Lett., 2009, 50, 6335‐6338.
18 V. Belluau, P. Noeureuil, E. Ratzke, A. Skvortsov, S. Gallagher,
C. A. Motti, M. Oelgemöller, Tetrahedron Lett., 2010, 51
,
4738‐4741.
19 S. Gallagher, F. Hatoum, N. Zientek, M. Oelgemöller,
Tetrahedron Lett., 2010, 51, 3639‐3641.
20 F. Boscá, M. A. Miranda, J. Photochem. Photobiol. B: Biol.,
1998, 43, 1‐26.
21 F. Boscá, M.‐L. Marín, M. A. Miranda, Photochem. Photobiol.,
2001, 74, 637‐655.
22 F. Boscá, M.‐L. Marín, M. A. Miranda, in CRC Handbook of
2
where scan rate was 100 mV s‐1). All experiments were
performed at room temperature and each individual
measurement was repeated three times.
Organic Photochemistry and Photobiology, 2nd Ed. (Eds.: W.
Horspool, F. Lenci), CRC Press, Boca Raton, 2004
.
23 S. Monti, S. Sortino, G. De Guidi, G. Marconi, J. Chem. Soc.,
Faraday Trans., 1997, 93, 2269‐2275.
24 L. J. Martínez, J. C. Scaiano, J. Am. Chem. Soc., 1997, 119
Conflicts of interest
There are no conflicts to declare.
,
11066‐11070.
25 G. Cosa, L. J. Martínez, J. C. Scaiano, Phys. Chem. Chem. Phys.,
1999, 1, 3533‐3537.
26 M. Laferrière, C. N. Sanram, J. C. Scaiano, Org. Lett., 2004, 6,
Acknowledgements
873‐875.
27 J. A. Blake, E. Gagnon, M. Lukeman, J. C. Scaiano, Org. Lett.,
These materials are based on work financed by the Croatian
Science Foundation (HRZZ IP‐2014‐09‐6312 for NB). The authors
thank Dr. Blaženka Gašparović for the support and help in
conducting cyclic voltammograms. LB acknowledges the
support of the National Research, Development and Innovation
Office of Hungary (Grant K123995 and BIONANO GINOP‐2.3.2‐
15‐2016‐00017).
2006, 8, 1057‐1060.
28 Y. P. Chuang, J. Xue, Y. Du, M. Li, H.‐Y. An, D. L. Phillips, J. Phys.
Chem. B, 2009, 113, 10530‐10539.
29 M.‐D. Li, Y. Du, Y. P. Chuang, J. Xue, D. L. Phillips, Phys. Chem.
Chem. Phys., 2010, 12, 4800‐4808.
30 Y. Xu, X. Chen, W.‐H. Fang, D. L. Phillips, Org. Lett., 2011, 13
,
5472‐5475.
31 M.‐D. Li, C. S. Yeung, X. Guan, J. Ma, W. Li, C. Ma, D. L. Phillips,
Chem. Eur. J., 2011, 17, 10935‐10950.
32 M.‐D. Li, T. Su, J. Ma, M. Liu, H. Liu, X. Li, D. L. Phillips, Chem.
Eur. J., 2013, 19, 11241‐11250.
33 F. Boscá, M. A. Miranda, G. Carganico, D. Mauleón,
Photochem. Photobiol., 1994, 60, 96‐101.
34 H. Huang, C. Yu, Y. Zhang, Y. Zhang, P. S. Mariano, W. Wang, J.
Am. Chem. Soc. 2017, 139, 9799‐9802.
35 J. A. Kautzky, T. Wang, R. W. Evans, D. W. C. MacMillan, J. Am.
Chem. Soc. 2018, 140, 6522‐6526.
36 N. A. Till, R. T. Smith, D. W. C. MacMillan, J. Am. Chem. Soc.
2018, 140, 5701‐5705.
37 T. Yamamoto, T. Iwasaki, T. Morita, Y. Yoshimi, J. Org. Chem.
2018, 83, 3702‐3709.
38 Q.‐Q. Ge, J.‐S. Qian, J. Xuan, J. Org. Chem. 2019, 84, 8691‐
8701.
Notes and references
1
2
Y. Kanaoka, Acc. Chem. Res., 1978, 11, 407‐413.
M. Horvat, K. Mlinarić‐Majerski, N. Basarić, Croat. Chem. Acta,
2010, 83, 179‐188.
3
4
A. G. Griesbeck, N. Hoffmann, K.‐D. Warzecha, Acc. Chem.
Res., 2007, 40, 128‐140.
M. Oelgemöller, A. G. Griesbeck, in CRC Handbook of Organic
Photochemistry and Photobiology, 2nd Ed. (Eds.: W. Horspool,
F. Lenci), CRC Press, Boca Raton, 2004
M. Oelgemöller, A. G. Griesbeck, J. Photochem. Photobiol. C:
Photochem. Rev., 2002, , 109‐127.
.
5
3
14 | J. Name., 2012, 00, 1‐3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 15 of 17
New Journal of Chemistry
Please do not adjust margins
Journal Name
ARTICLE
1
2
3
4
5
6
7
8
9
39 S. Senaweera, K. C. Cartwright, J. A. Tunge, J. Org. Chem. 2019, 70 A. G. Griesbeck, H. Görner, J. Photochem. Photobiol. A: Chem.,
View Article Online
84, 12553‐12561.
1999, 129, 111‐119.
DOI: 10.1039/D0NJ03465G
40 K. C. Cartwright, S. B. Lang, J. A. Tunge, J. Org. Chem. 2019, 84
,
71 W. J. Leigh, E. C. Lathioor, M. J. St. Pierre, J. Am. Chem. Soc.,
1996, 118, 12339‐12348.
72 S. Aich, C. Raha, S. Basu, J. Chem. Soc., Faraday Trans., 1997,
93, 2991‐2996.
2933‐2940.
41 X. Zhang, P. Zhu, R. Zhang, X. Li, T. Yao, J. Org. Chem. 2020, 85
,
9503‐9513.
42 L. Mandić, K. Mlinarić‐Majerski, A. G. Griesbeck, N. Basarić, 73 D. Occhialini, J. S. Kristensen, K. Daasbjerg, H. Lund, Acta
Eur. J. Org. Chem., 2016, 4404‐4414.
43 M. Sohora, T. Šumanovac Ramljak, K. Mlinarić‐Majerski, N. 74 T. Kjærsbo, K. Daasbjerg, S. U. Pedersen, Electrochim. Acta,
Basarić, Croat. Chem. Acta, 2014, 87, 431‐446. 2003, 48,1807‐1816.
Chem. Scand., 1992, 46, 474‐481.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
44 M. Horvat, K. Mlinarić‐Majerski, A. G. Griesbeck, N. Basarić, 75 M. Megyesi, L. Biczók, J. Phys. Chem. B 2010, 114, 2814‐2819.
Photochem. Photobiol. Sci., 2011, 10, 610‐617.
45 T. Šumanovac Ramljak, M. Sohora, I. Antol, D. Kontrec, N.
Basarić, K. Mlinarić‐Majerski, Tetrahedron Lett., 2014, 55
,
4078‐4081.
46 M. Sohora, M. Vazdar, I. Sović, K. Mlinarić‐Majerski, N.
Basarić, J. Org. Chem., 2018, 83, 14905‐14922.
47 A. G. Griesbeck, S. Schieffer, Photochem. Photobiol. Sci., 2003,
2, 113‐117.
48 V. Wintgens, P. Valat, J. Kossanyi, A. Demeter, L. Biczók, T.
Bérces, New J. Chem., 1996, 20, 1149‐1158.
49 M. Atar, B. Öngel, H. Riedasch, T. Lippold, J. Neudörfl, D.
Sampedro, A. G. Griesbeck, ChemPhotoChem, 2020, 4, 89‐97.
50 T. Soujanya, R. W. Fessenden, A. Samanta, J. Phys. Chem.,
1996, 100, 3507‐3512.
51 G. Saroja, T. Soujanya, B. Ramachandram, A. Samanta, J.
Fluoresc., 1998, 8, 405‐410.
52 A. Morimoito, T. Yatsuhashi, T. Shimada, L. Biczók, D. A. Tryk,
H. Inoue, J. Phys. Chem. A, 2001, 105, 10488‐10496.
53 D. E. Wetzler, C. Chesta, R. Fernández‐Prini, P. F. Aramendía,
J. Phys. Chem. A, 2002, 106, 2390‐2400.
54 V. Wintgens, C. Amiel, J. Photochem. Photobiol. A: Chem.,
2004, 168, 217‐226.
55 D. C. Khara, S. Banerjee, A. Samanta, ChemPhysChem, 2014,
15, 1793‐1798.
56 P. Benčić, L. Mandić, I. Džeba, I. Tartaro Bujak, L. Biczók, B.
Mihaljević, K. Mlinarić‐Majerski, I. Weber, M. Kralj, N. Basarić,
Sens. Actuators: B: Chem., 2019, 286, 52‐61.
57 L. Mandić, P. Benčić, K. Mlinarić‐Majerski, S. Liekens, R.
Snoeck, G. Andrei, M. Kralj, N. Basarić, Arch Pharm., 2020,
e2000024. https://doi.org/10.1002/ardp.202000024.
58 R. J. Perkins, H.‐C. Xu, J. M. Campbell, K. D. Moeller, Beilstein
J. Org. Chem., 2013,
59 M. Montalti, A. Credi, L. Prodi, M. T. GandolfiIn Handbook of
Photochemistry; CRC Taylor and Francis: Boca Raton, 2006
9, 1630‐1636.
.
60 H. Görner, M. Oelgemöller, A. G. Griesbeck, J. Phys. Chem A,
2002, 106, 1458‐1464.
61 A. Reiffers, C. Torres Ziegenbein, L. Schubert, J. Diekmann, K.
A. Thom, R. Kühnemuth, A. G. Griesbeck, O. Weingart, P. Gilch,
Phys. Chem. Chem. Phys., 2019, 21, 4839‐4853.
62 V. Wintgens, P. Valat, J. Kossanyi, L. Biczók, A. Demeter, T.
Bérces, J. Chem. Soc., Faraday Trans., 1994, 90, 411‐421.
63 S. E. Braslawsky, Pure Appl. Chem., 2007, 79, 293‐465.
64 M. Albrecht, C. Bohne, A. Graznhan, H. Ihmels, T. C. S. Pace, A.
Schnurpfeil, M. Waidelich, C. Yihwa, J. Phys. Chem. A, 2007,
111, 1036‐1044.
65 Z. Miskolczy, J. G. Harangozó, L. Biczók, V. Wintgens, C.
Lorthioir, C. Amiel, Photochem. Photobiol. Sci., 2014, 13, 499‐
508.
66 D. W. Ellis, L. B. Rogers, Spectrochim. Acta, 1964, 20, 1709‐
1720.
67 J. F. Ireland, P. A. H. Wyatt, Adv. Phys. Org. Chem., 1976, 12
,
131‐221.
68 L. G. Arnaut, S. J. Formosinho, J. Photochem. Photobiol. A:
Chem., 1993, 75, 1‐20.
69 M. Horvat, H. Görner, K.‐D. Warzecha, J. Neudörfl, A. G.
Griesbeck, K. Mlinarić‐Majerski, N. Basarić, J. Org. Chem.,
2009, 74, 8219‐8231.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1‐3 | 15
Please do not adjust margins

Please do not adjust margins
New Journal of Chemistry
Page 16 of 17
1
2
ARTICLE
Journal Name
View Article Online
DOI: 10.1039/D0NJ03465G
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Please do not adjust margins
16 | J. Name., 2012, 00, 1‐3
This journal is © The Royal Society of Chemistry 20xx

Page 17 of 17
1
New Journal of Chemistry
2
3
4
TOC Entry
View Article Online
DOI: 10.1039/D0NJ03465G
5
6
7
8
COOH
O
O
h
-CO2
N
N
CH3CN-H2O
buffer, pH = 7
O
O
R
R
1
2
3
4
5
6
R = H
1H
2H
3H
R = H
9
R = COOH
R = OCH3
R = OH
R = NO2
R = NH2
R = COOH
R = OCH3
R = OH
photodecarboxylation
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
singlet vs. triplet reactivity 4H
5H
6H
R = NO2
R = NH2
Substituents on the phthalimide affect its photophysics and photochemical reactivity. Electron donors generally result in low quantum
yields of intersystem crossing and reactivity from singlet excited states.