Excited-state hydroxide ion release from a series of acridinol photobases

Article abstract of DOI:10.1021/acs.jpca.6b10980

The excited-state heterolysis of acridinol-based derivatives leads to the release of the OH-ion and the formation of the corresponding acridinium cations. To evaluate the parameters that control the reaction barriers, the kinetics of excited-state OH-release from a series of acridinol photobases were studied using transient absorption spectroscopy. The rate constants were obtained in three solvents (methanol, butanol, and isobutanol), and the data were modeled using Marcus theory. The intrinsic reorganization energies obtained from these fits were found to correlate well with the solvent reorganization energies calculated using dielectric continuum model, suggesting that the excited-state OH-release occurs along the solvent reaction coordinate. Furthermore, the ability of acridinol photobases to photoinitiate chemical reactions was demonstrated using the Michael reaction between dimethylmalonate and nitrostyrene.

Full text of DOI:10.1021/acs.jpca.6b10980

Subscriber access provided by University of Newcastle, Australia
Article
Excited State Hydroxide Ion Release From a Series of Acridinol Photobases
Yun Xie, Stefan Ilic, Sanja Skaro, Veselin Maslak, and Ksenija D. Glusac
J. Phys. Chem. A, Just Accepted Manuscript • DOI: 10.1021/acs.jpca.6b10980 • Publication Date (Web): 19 Dec 2016
Downloaded from http://pubs.acs.org on December 24, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
The Journal of Physical Chemistry A is published by the American Chemical Society.
1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 37
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
Excited State Hydroxide Ion Release From a Series
of Acridinol Photobases
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
Yun Xie, 1† Stefan Ilic,† Sanja Skaro,‡ Veselin Maslak,‡ Ksenija D. Glusac†*
†Department of Chemistry, Center for Photochemical Sciences, Bowling Green State University,
Bowling Green, Ohio 43403, United States
‡Faculty of Chemistry, University of Belgrade, Studentski Trg 12ꢀ16, Belgrade 11000, Serbia
ACS Paragon Plus Environment
1

The Journal of Physical Chemistry
Page 2 of 37
1
2
3
4
5
6
7
8
9
ABSTRACT. The excitedꢀstate heterolysis of acridinolꢀbased derivatives leads to the release of
the OH− ion and the formation of the corresponding acridinium cations. To evaluate the
parameters that control the reaction barriers, the kinetics of excitedꢀstate OH− release from a
series of acridinol photobases were studied using transient absorption spectroscopy. The rate
constants were obtained in three solvents (methanol, butanol and isobutanol), and the data were
modeled using Marcus theory. The intrinsic reorganization energies obtained from these fits
were found to correlate well with the solvent reorganization energies calculated using dielectric
continuum model, suggesting that the excitedꢀstate OH− release occurs along the solvent reaction
coordinate. Furthermore, the ability of acridinol photobases to photoꢀinitiate chemical reactions
was demonstrated using the Michael reaction between dimethylmalonate and nitrostyrene.
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
INTRODUCTION
Photoacids are a group of molecules, usually containing the aromatic alcohol motif, whose
acidity significantly increases upon electronic excitation (up to 32 pKa units).1ꢀ3 The excitation of
photoacids in the presence of appropriate base acceptors enables mechanistic studies of fast
proton transfer (PT) reactions in solution using timeꢀresolved laser techniques.4ꢀ9 These
experimental results provided a valuable platform to test and improve current theoretical models
for PT. For example, Nibbering discovered the transient signature of the solvated proton in the
midꢀIR range during the PT between the pyranine photoacid and the acetate base, which offered
evidence of a solventꢀmediated PT.7 Such solventꢀmediated mechanism is not modeled in the
traditional EigenꢀWeller theory for PT, which prompted Siwick and Bakker to develop an
improved model for PT that included a distribution of reactive acidꢀbase species separated by a
ACS Paragon Plus Environment
2

Page 3 of 37
The Journal of Physical Chemistry
1
2
varying number of solvent molecules.8 In addition to these basic studies of PT mechanism, the
photoacids have also been applied to: (i) generate and characterize shortꢀlived intermediates,
such as H2CO3;10 (ii) probe the microenvironment of the biological systems (protein pockets and
cells)11ꢀ17 and other confined media (reverse micelles, cyclodextrins and nafion membranes);18ꢀ20
(iii) study the protein folding kinetics using the pH jump technique.21ꢀ22
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
In analogy to the photoacids described above, the photobases are defined here as compounds
that release hydroxide ion (OH−) upon excitation. Several types of photobases were identified in
the ‘80s and were composed of triarylmethanol, xanthenol, fluorenol and other molecular
frameworks.23ꢀ24 These early studies treated photobases as suitable sources of shortꢀlived
carbocations, whose reactivity with various nucleophiles was studied using timeꢀresolved
nanosecond laser spectroscopy.25ꢀ26 More recently, the pseudobase derivatives of aryltropylium
and acridinium ions have been investigated as a new class of photoꢀswitches.27ꢀ29 The kinetics of
OH− release were not addressed in these studies, likely due to the insufficient time resolution of
the nanosecond transient absorption instrument used. Nowadays, the availability of ultrafast
lasers allows chemists to capture the kinetics of OH− release and investigate the mechanism of
OH− transport in various media, which will likely contribute to our understanding of OH−
solvation and diffusion. For example, recent findings have challenged the traditional view of
OH− solvation as [OH (H2O)3]− species, “mirror image” of proton solvation), and proposed a
“hypercoordinated” configuration [OH(H2O)4]−, which seems to agree with the experimental
evidence.30ꢀ31 With the unique lightꢀtriggered OH− release, photobases will likely serve as an
excellent probe for timeꢀresolved studies of OH− solvation. Furthermore, photobases have
already been utilized to achieve spatiotemporal control in biologically relevant systems, to
trigger DNA/RNA conformational changes and drug delivery.32ꢀ34
ACS Paragon Plus Environment
3

The Journal of Physical Chemistry
Page 4 of 37
1
2
We previously studied the kinetics of OH− release from an acridinol photobase using
femtosecond pumpꢀprobe spectroscopy,35 and found that the heterolysis is fast (108 ps lifetime)
in protic solvents (such as methanol), while no heterolysis occurs in aprotic solvent (such as
acetonitrile). This solventꢀdependent OH− release is quite similar to the previous studies of
excitedꢀstate PT,3 and is likely due to increased reactivity of the hydrogenꢀbonded complex
between the photoacid or the photobase with the accepting solvent molecules. To gain a more
quantitative comparison of the kinetics of H+/OH− release, the intrinsic barrier for OH− release
from a xanthenol photobase was evaluated in a series of solvent mixtures.36 Using the Marcus
model, the intrinsic barrier was found to be Δꢀꢁꢂ~10 kcal/mol, which is significantly higher than
the barriers found previously for the PT (Δꢀꢁꢂ~1ꢀ2 kcal/mol).3, 37 Such differences in the
photoreactivity of photoacids and photobases could be due to differences in solvation of released
ions. Alternatively, the lower reactivity of OH− release could be due to high reorganization
energy of the cation formed, or due to less efficient OH− release relative to H+.36
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
In this study, the role of the solvent reorganization on rates of excitedꢀstate OH− release was
evaluated using a series of acridinol photobases (Scheme 1). Specifically, the thermodynamics
(ꢁG) and kinetics (k) of the OH− release were studied using steadyꢀstate and timeꢀresolved
UV/VIS absorption spectroscopy. The experimental values were modeled using Marcus theory to
evaluate the intrinsic barrier Δꢀꢁꢂ for the OH− release. The intrinsic reorganization energies
obtained from the Marcus fits were found to correlate very well with the solvent reorganization
energies predicted by a dielectric continuum model. These results indicate that the OH− release
occurs along the solvent reaction coordinate and that the barrier height is controlled by the
solvent response to the formed charges.
ACS Paragon Plus Environment
4

Page 5 of 37
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
This manuscript also reports that the acridinol photobases can be utilized to drive chemical
reactions. Previous approaches to photobaseꢀdriven reactions have mostly relied on bases
protected with photoꢀlabile precursors.38ꢀ40 While these methods provide efficient base
generation, the process is irreversible. More recently, the reversibility was achieved using cis-
trans photoisomerization as a way to control the basicity of the photoactive molecule.41ꢀ42 Here,
we show that the irradiation of acridinol photobase can be utilized to drive a Michael addition
reaction. In specific, the reaction between dimethylmalonate and nitrostyrene was triggered using
AcrOMe photobase, which releases the methoxide ion upon excitation. The methoxide ion acts
as a base that deprotonates dimethylmalonate and initiates the addition to nitrostyrene. The
results of this study show that acridinol photobases can promote the baseꢀcatalyzed reactions,
which is expected to be useful in applications that require spatial or temporal control of the
chemical reactivity.
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
X
X
OH
h
ν
+
OH
N
N
XꢀAcr+
XꢀAcrOH
X=Me2N, MeO, H, CF3, CN
Scheme 1. The photochemical OH− release from model 9ꢀsubstituted acridinols.
EXPERIMENTAL SECTION
Syntheses. All chemicals were purchased from commercial suppliers and used without further
1
13
purification. H and C NMR spectra were recorded on a Bruker Avance III 500 MHz system.
ACS Paragon Plus Environment
5

The Journal of Physical Chemistry
Page 6 of 37
1
2
3
4
5
6
7
8
9
Electron impact ionization (EI) and MALDI mass spectra were measured on a Shimadzu
QP5050A and a Brucker Daltonics Omniflex mass spectrometer. 10ꢀMethylꢀ9ꢀphenylacridinium
perchlorate (Acr+) was purchased from TCI America. Arylꢀbromides (4ꢀbromoanisole, 4ꢀ
bromobenzonitrile and 4ꢀbromobenzotrifluoride) were purchased from Sigma. AcrOMe was
synthesized according to already reported procedure.35
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
10-methyl-9-phenyl-9,10-dihydroacridin-9-ol (AcrOH): AcrOH was synthesized according to
the literature.35 A 1 M solution of NaHCO3 (100 ml) was added to 10ꢀmethylꢀ9ꢀ
phenylacridinium perchlorate (Acr+, 225 mg, 0.6 mmol). Acetonitrile (~25 mL) was added to
dissolve Acr+ completely and the pH was then increased to 13 by addition of 30% aqueous
NaOH. Conversion was monitored using UV/Vis by observing the disappearance of the
absorption at 427 nm. When all Acr+ converted to AcrOH, the reaction mixture was stirred for 15
min in the dark. Acetonitrile was removed in vacuo and water solution was extracted with
dichloromethane (3x100 mL). Organic layers were combined, dried over anhydrous calcium
chloride and solvent evaporated. Cyclohexane was added to the residue causing the precipitation
of unreacted Acr+. Filtrate was collected and evaporated to give 95 mg of pure AcrOH (54%).
MSꢀMALDI: m/z 287 [M+ for C20H17NO], 1HꢀNMR (CD3CN, 500 MHz): δ 7.58 (d, 2H, J = 7.7
Hz), 7.35ꢀ7.31 (m, 2H), 7.22 (d, 4H, J = 4.3 Hz), 7.16ꢀ7.14 (m, 1H), 7.12 (d, 2H, J = 8.2 Hz),
13
7.01 (t, 2H, J = 7.5 Hz), 4.21 (br s, 1H), 3.49 (s, 3H). CꢀNMR (CD3CN, 125 MHz): δ 148.4,
140.0, 129.6, 128.1, 127.8, 126.7, 126.6, 125.6, 120.3, 112.4, 72.4, 32.7.
9-(4-(dimethylamino)phenyl)-10-methylacridin-10-ium perchlorate (Me2N-Acr+): Me2NꢀAcr+
was synthesized according to the literature under modified conditions.43 Under nitrogen, 10ꢀ
methylꢀ9(10H)ꢀacridone (1 g, 4.8 mmol), N,Nꢀdimethylaniline (1.36 mL, 10.7 mmol, 2.3 eq) and
phosphoryl chloride (2.43 mL, 26.1 mmol, 5.4 eq) were mixed and left to stir for 2 hours at 90
ACS Paragon Plus Environment
6

Page 7 of 37
The Journal of Physical Chemistry
1
2
oC. Concentrated aqueous sodium hydroxide solution was added carefully to quench the reaction.
The reaction mixture was then extracted with dichloromethane, organic layers were combined,
dried over anhydrous sodium sulfate and solvent evaporated under reduced vacuum. The oily
residue was dissolved in minimal volume of acetonitrile and perchloric acid was added until a
greenꢀyellowish color. Protonated Me2NH+ꢀAcr+ dication was then extracted with
dichloromethane and solvent was evaporated. Upon addition of water to the greenꢀyellow solid,
color changed to deep blueꢀpurple. Target compound was extracted with dichloromethane, dried
over anhydrous sodium sulfate and solvent was evaporated. The crude solid product was
recrystallized from acetonitrile/ether mixture. Formed precipitate was filtered and washed with
copious amount of ether to yield 1.78 g of deep blueꢀpurple solid (90%). 1HꢀNMR (CD3CN, 500
MHz): δ 8.54 (d, 2H, J = 9.2 Hz), 8.35 (dd, 1H, J1 = 6.8 Hz, J2 = 1.5 Hz), 8.33 (dd, 1H, J1 = 6.8
Hz, J2 = 1.5 Hz), 8.24 (dd, 2H, J1 = 6.8 Hz, J2 = 0.8 Hz), 7.85 (dd, 1H, J1 = 6.8 Hz, J2 = 0.8 Hz),
7.84 (dd, 1H, J1 = 6.8 Hz, J2 = 0.8 Hz), 7.48ꢀ7.45 (m, 2H), 7.13 (d, 2H, J = 7.9 Hz), 4.75 (s, 3H),
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
13
3.16 (s, 6H). CꢀNMR (CD3CN, 125 MHz): δ 162.9, 151.9, 141.7, 138.6, 132.7, 131.0, 127.2,
126.1, 119.6, 118.6, 111.7, 40.2, 39.0.
9-(4-(dimethylamino)phenyl)-10-methyl-4a,9,9a,10-tetrahydroacridin-9-ol
(Me2N-AcrOH):
Me2NꢀAcr+ (200 mg, 0.5 mmol) was dissolved in 10 mL of acetonitrile and pH was increased to
13 by addition of 5 M aqueous NaOH solution. The resulting solution was stirred for 2 h and
then extracted with dichloromethane. Organic layers were combined, dried over anhydrous
sodium sulfate and solvent evaporated. Crude product was dissolved in hexane/benzene mixture
(2:1) to precipitate the remaining starting material. Filtrate was collected and evaporated to give
1
85 mg of pure product (60%). MSꢀEI: m/z 312 [M+ calculated for C21H16N2O], HꢀNMR
(CD3CN, 500 MHz): δ 7.65 (dd, 2H, J1 = 7.7 Hz, J2 = 1.5 Hz), 7.34ꢀ7.31 (m, 2H), 7.10 (d, 2H, J
ACS Paragon Plus Environment
7

The Journal of Physical Chemistry
Page 8 of 37
1
2
3
4
5
6
7
8
9
= 8.3 Hz), 7.03 (td, 2H, J1 = 7.7 Hz, J2 = 1.1 Hz), 6.98ꢀ6.94 (m, 2H), 4.02 (br s, 1H), 3.46 (s,
3H), 2.84 (s, 6H). 13CꢀNMR (CDCl3, 125 MHz): δ 162.8, 151.9, 141.6, 138.6, 132.7, 131.0,
127.2, 126.1, 119.6, 118.6, 113.3, 111.7, 77.2, 40.2, 39.0.
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
MeO-AcrOH, CN-AcrOH and CF3-AcrOH were synthesized according to the following
procedure: To a solution of corresponding arylꢀbromide (2 mmol) in dry THF (6 mL), 2.5 M
o
hexane solution of nꢀbutyllithium (0.96 mL, 2.4 mmol) was added dropwise at ꢀ78 C under
argon and stirred for 30 min. A suspension of 10ꢀmethylꢀ9(10H)ꢀacridone (0.33 g, 1.6 mmol) in
dry THF (14 mL) was then added and the resulting solution stirred for 12 h at room temperature.
The reaction was quenched by addition of 20 mL of saturated aqueous ammonium chloride
solution and extracted with dichloromethane (3x10 mL). Organic fractions were combined, dried
over anhydrous magnesium sulfate and solvent evaporated. Crude product was dissolved in
hexane/benzene mixture (2:1) causing the precipitation of the remaining starting material. The
filtrate containing desired product was collected and solvent removed in vacuo to give desired
acridinol.
9-(4-methoxyphenyl)-10-methyl-9,10-dihydroacridin-9-ol (MeO-AcrOH): Using the procedure
described above, 0.25 g of pure product (49%) was obtained. MSꢀEI: m/z 317 [M+ calculated for
C21H19NO2], 1HꢀNMR (CD3CN, 500 MHz): δ 7.62 (dd, 2H, J1 = 7.7 Hz, J2 = 1.5 Hz), 7.34 (dd,
1H, J1 = 7.0 Hz, J2 = 1.4 Hz), 7.33 (dd, 1H, J1 = 7.0 Hz, J2 = 1.4 Hz), 7.12 (d, 2H, J = 7.7 Hz),
7.10ꢀ7.07 (m, 2H), 7.03 (td, 2H, J1 = 7.7 Hz, J2 = 1.1 Hz), 6.77ꢀ6.74 (m, 2H), 4.11 (s, 1H), 3.72
13
(s, 3H), 3.48 (s, 3H). CꢀNMR (CD3CN, 125 MHz): δ 158.3, 140.5, 140.1, 129.8, 128.0, 126.9,
126.4, 120.2, 113.0, 112.3, 72.0, 54.8, 32.9.
4-(9-hydroxy-10-methyl-9,10-dihydroacridin-9-yl)benzonitrile (CN-AcrOH): Using the given
procedure, 0.31 g of pure product (48%) was obtained. MSꢀEI: m/z 312 [M+ calculated for
ACS Paragon Plus Environment
8

Page 9 of 37
The Journal of Physical Chemistry
1
2
C21H16N2O], 1HꢀNMR (CD3CN, 500 MHz): δ 7.56ꢀ7.53 (m, 2H), 7.51 (dd, 2H, J1 = 7.7 Hz, J2 =
1.4 Hz), 7.39 (d, 2H, J = 8.3 Hz), 7.35ꢀ7.30 (m, 2H), 7.12 (d, 2H, J = 8.3 Hz), 6.98 (t, 2H, J =
3
4
5
6
7
8
9
13
7.5 Hz), 4.49 (s, 1H), 3.48 (s, 3H). CꢀNMR (CD3CN, 125 MHz): δ 154.5, 140.7, 132.8, 129.5,
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
127.6, 127.2, 121.4, 113.6, 110.9, 72.9, 33.9.
10ꢀmethylꢀ9ꢀ(4ꢀ(trifluoromethyl)phenyl)ꢀ9,10ꢀdihydroacridinꢀ9ꢀol (CF3ꢀAcrꢀOH): Using the
given procedure, 0.39 g of pure product (55%) was obtained. MSꢀEI: m/z 355 [M+ calculated for
1
C21H16NF3], HꢀNMR (CD3CN, 500 MHz): δ 7.57ꢀ7.53 (m, 4H), 7.45 (d, 2H, J = 8.6 Hz), 7.36
(ddd, 2H, J1 = 8.7 Hz, J2 = 7.2 Hz, J3 = 1.6 Hz), 7.16 (d, 2H, J = 8.3 Hz), 7.02 (td, 2H, J1 = 7.8
Hz, J2 = 1.6 Hz), 4.40 (s, 1H), 3.52 (s, 3H). 13CꢀNMR (CD3CN, 125 MHz): δ 153.7, 140.7,
129.6, 129.4, 128.9 (q, JCF = 32 Hz), 127.7, 127.1, 125.7 (q, JCF = 3.8Hz), 121.2, 113.5, 73.0,
33.8.
Equilibrium constant determination. UV/Vis absorption spectra were recorded on Varian Cary
50 Bio spectrometer in a 1 cm quartz cell from 200 nm to 800 nm. The sample concentration was
controlled so that the absorption maximum was around 1. Fluorescence measurements were
performed on a single photon counting spectrofluorimeter from Edinburgh Analytical
Instruments (FL/FS 920) in a 1 cm quartz cell. The excitation wavelengths and the emission
spectral windows were chosen according to each sample. The concentration of the samples was
diluted 100 times from absorption measurements. The groundꢀstate pKOH values were
determined as described previously with a slight modification.35 The equilibrium constant KOH
was determined by titrating the solutions of acridinium cations (XꢀAcr+) with a solution of
sodium hydroxide with known concentration (it was assumed that NaOH fully dissociates in the
solvent) and the concentrations of XꢀAcr+ and XꢀAcrOH were determined using UV/Vis
ACS Paragon Plus Environment
9

The Journal of Physical Chemistry
Page 10 of 37
1
2
absorption spectroscopy. The KOH was calculated from the solution in which 50% of Acr+ is
converted to AcrOH, as:
3
4
5
6
7
8
9
ꢏ
ꢕ
ꢐꢑꢒꢓꢔ
ꢆ
ꢉꢊ
ꢃꢄꢅ = ꢇꢈ = cꢋꢌꢍꢎ
−
ꢖ
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
Femtosecond Transient Absorption Experiments. The 800 nm laser pulses were produced at a
1 kHz repetition rate by a modeꢀlocked Ti:sapphire laser and regenerative amplifier (Hurricane,
SpectraꢀPhysics). The output from the Hurricane was split into pump (85%) and probe (10%)
beams. The pump beam was sent into an optical parametric amplifier (OPAꢀ400, Spectra
Physics) to obtain the desired excitation wavelength. The energy of the pump beam was <300
nJ/pulse. The probe beam was focused into a twoꢀdimensional translated 4 mm CaF2 crystal for
white light generation between 350 and 800 nm. The flow cell (Starna Cell Inc. 45ꢀQꢀ2, 0.9 mL
volume with 2 mm pathlength), pumped by a Variable Flow MiniꢀPump (Control Company),
was used to prevent photoꢀdegradation of the sample. After passing through the cell at the magicꢀ
angle geometry, the continuum was coupled into an optical fiber and input into a CCD
spectrograph (Ocean Optics, S2000). The data acquisition was achieved using inꢀhouse
LabVIEW (National Instruments) software routines. The group velocity dispersion of the
probing pulse was determined using nonresonant optical Kerr effect (OKE) measurements.
Sample solutions were prepared at a concentration needed to have absorbance of 1.0 at the
excitation wavelength.
PhotoꢀMichael addition. The model Michael addition was performed by preparing a solution of
10.5 mg of transꢀβꢀnitrostyrene (0.0706 mmol, 1 eq), 8.1 ꢂL of dimethylmalonate (0.0706 mmol,
1 eq) and 21.3 mg AcrOMe (0.0706 mmol, 1 eq) in 0.5 mL methanol. The protocol involved
either thermal or photochemical treatment. In the thermal treatment, the reaction mixture was
stirred for 24 h at 65 oC. In the photochemical treatment, the reaction mixture was irradiated (125
ACS Paragon Plus Environment
10

Page 11 of 37
The Journal of Physical Chemistry
1
2
o
3
W medium pressure mercury lamp) for 3 min and then stirred for 24 h at 65 C. After each
treatment, the reaction was quenched by addition of water and the reaction mixture was extracted
with dichloromethane. Organic layers were combined, dried over anhydrous sodium sulfate and
solvent evaporated. Oily crude product was purified with dryꢀflash chromatography
(hexane/ethyl acetate = 8/2) to yield the oily product (thermal treatment gave 8.3 mg of the
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
1
product or 38%, while the photochemical treatment gave 13 mg of the product or 60%). Hꢀ
NMR (CDCl3, 500 MHz): δ 7.34ꢀ7.28 (m, 3H), 7.24ꢀ7.22 (m, 2H), 4.95ꢀ4.86 (m, 2H), 4.24 (td,
1H J1 = 9.0 Hz, J2 = 5.1 Hz), 3.87 (d, 1H, J = 9.0 Hz), 3.76 (s, 3H), 3.56 (s, 3H). 13CꢀNMR
(CDCl3, 125 MHz): δ 168.0, 167.5, 136.4, 129.3, 128.6, 128.1, 77.6, 55.0, 53.2, 53.0, 43.2.
RESULTS AND DISCUSSION
Model compounds: The driving force for the excitedꢀstate OH− transfer was tuned by
preparing a series of model acridinol compounds with varying substituents on the phenyl ring
(Scheme 1). These acridinol derivatives were readily synthesized using a reaction between Nꢀ
methylꢀacridone and the corresponding arylꢀlithium reagent, as described in the Experimental
Section. The absorption and emission spectra of the model acridinol photobases XꢀAcrOH and
their corresponding acridinium salts XꢀAcr+ are shown in Figure 1. In general, the
absorption/emission maxima of all acridinol XꢀAcrOH derivatives appear at shorter wavelengths
than the bands of the corresponding XꢀAcr+ cations, which is consistent with the increased
conjugation of the XꢀAcr+ cations.
The redꢀshifts in the absorption/emission bands were observed for the model compounds that
can form chargeꢀtransfer excited states. For example, the emission bands of XꢀAcrOH
derivatives with electronꢀwithdrawing groups (478 nm for CNꢀAcrOH and 412 nm for CF3ꢀ
ACS Paragon Plus Environment
11

The Journal of Physical Chemistry
Page 12 of 37
1
2
3
4
5
6
7
8
9
AcrOH) appear at higher wavelengths than the emission bands of other acridinols (~350 nm),
which is likely due to the formation of charge transfer excited state, in which the electron density
migrates from the acridinol moiety to the substituted phenyl ring. Similarly, the emission band of
Me2NꢀAcr+ (680 nm) is redꢀshifted relative to emission bands of other acridinium cations (~510
nm), likely due to the formation of a charge transfer state in which the electron density moves
from the substituted phenyl ring to the acridinium moiety. Similar chargeꢀtransfer states were
reported in other acridinium salts.44
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
Figure 1. Absorption and emission spectra of A) XꢀAcrOH and B) XꢀAcr+. All of the spectra
were collected in acetonitrile at room temperature, except for Me2NꢀAcr+, whose emission
spectrum was collected in 2ꢀmethyltetrahydrofuran at 77 K). AcrOH/Acr+ (black), Me2Nꢀ
AcrOH/Me2NꢀAcr+ (purple), MeOꢀAcrOH/MeOꢀAcr+ (orange), CF3ꢀAcrOH/ CF3ꢀAcr+ (blue),
CNꢀAcrOH/CNꢀAcr+ (red); XꢀAcrOH absorption (dash line), XꢀAcrOH emission (+ marker), Xꢀ
Acr+ absorption (solid line), XꢀAcr+ emission (dotted line).
ACS Paragon Plus Environment
12

Page 13 of 37
The Journal of Physical Chemistry
1
2
3
4
THERMODYNAMICS
5
6
7
8
The thermodynamic driving force for the excitedꢀstate OH− release from model acridinol
derivatives was evaluated in four different solvents (water, methanol, butanol and isobutanol).
First, the groundꢀstate pKOH and ꢀGOH values for the reaction:
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
ꢕ
ꢑ
ꢆꢜꢉꢝꢞꢟ ꢊꢆꢄꢅ ꢊ
ꢆꢜꢉꢝꢞꢟꢄꢅꢊ
ꢗ − ꢘꢙꢚꢇꢈ ⇌ ꢗ − ꢘꢙꢚꢛ + ꢇꢈꢉ ꢃꢄꢅ
=
(1)
were obtained using pHꢀdependent UV/Vis absorption spectroscopy. The results indicate that
the OH− release from acridinol models in their ground state is thermodynamically unfavorable in
all solvents (Table 1). The thermodynamic parameters for CF3ꢀAcrOH and CNꢀAcrOH in n-
BuOH and iꢀBuOH were not obtained, because the excitedꢀstate OH− release was not observed
for these XꢀAcrOH/solvent systems. Across the acridinol series, the driving force increases with
the electronꢀdonating character of the substituent (for example, the lowest ꢀG value was
obtained for Me2NꢀAcrOH). This result is consistent with the fact that electronꢀdonating groups
stabilize the XꢀAcr+ cations formed upon OH− release by positive charge delocalization.
Similarly, previous studies have shown that the driving force for OH− release in other
pseudobases correlates well with factors that tend to stabilize the corresponding cations.45ꢀ47 For
example, the driving force for OH− release from xanthenol (XanOH) in water is ꢁG=18.8
kcal/mol,36 which is significantly higher than for acridinol AcrOH (ꢁG=3.9 kcal/mol), reflecting
the higher aromatic character of acridinium cation Acr+ relative to Xan+. The solvent effects on
the driving force is as expected: the high polarity solvents (water) are better at stabilizing the
ions formed upon heterolysis, which lowers the ꢀG values for OH− release. Similar solvent
polarity effects were observed in the photoacids.48
ACS Paragon Plus Environment
13

The Journal of Physical Chemistry
Page 14 of 37
1
2
3
4
Table 1. Groundꢀstate pKOH and ꢁGOH values for model XꢀAcrOH acridinols in different
5
6
solvents.
7
8
9
H2O
pKOH
2.3
MeOH
ꢁGOH pKOH ꢁGOH pKOH
nꢀBuOHa
iꢀBuOHa
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
X
ꢁGOH pKOH ꢁGOH
Me2N
MeO
H
3.2
3.7
3.9
4.4
4.5
4.0
4.5
4.7
5.6
5.8
5.5
6.1
6.4
7.7
8.0
5.1
4.9
5.2
ꢀ
7.0
6.8
7.2
ꢀ
4.4
5.0
6.1
ꢀ
6.1
6.8
8.3
ꢀ
2.7
2.8
CF3
CN
3.2
3.3
ꢀ
ꢀ
ꢀ
ꢀ
aThe calculations assume that NaOH is fully dissociates in nꢀBuOH and iꢀBuOH.
The groundꢀstate pKa values were then used to derive the excitedꢀstate thermodynamics using
the Förster cycle,35 as follows:
∆Gꢄ∗ ꢅ = ℎꢠꢜꢉꢝꢞꢟꢄꢅ − ∆Gꢄꢅ − ℎꢠꢜꢉꢝꢞꢟ
(2)
ꢕ
For this purpose, the frequencies for 0−0 transitions of XꢀAcrOH and XꢀAcr+ were derived
*
from the emission spectra of the model compounds (Figure 1). The resulting ꢁGOH values
(Table 2) show that the driving force for OH− release considerably increases upon excitation. For
example, the driving force for Me2NꢀAcrOH in methanol changes from 5.5 kcal/mol in the
ground state to −35.7 kcal/mol in the excited state. In other acridinols, similar but less drastic
changes in ꢁG values occur, and the excitedꢀstate OH− release is thermodynamically favored in
all compounds except for CNꢀAcrOH.
ACS Paragon Plus Environment
14

Page 15 of 37
The Journal of Physical Chemistry
1
2
3
4
Table 2. Excitedꢀstate ꢡꢃꢄ∗ꢅ and ∆Gꢄ∗ ꢅ values for model XꢀAcrOH acridinols in different
5
6
solvents.
7
8
9
H2O
MeOH
nꢀBuOH
iꢀBuOH
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
X
ꢡꢃꢄ∗ꢅ
∆Gꢄ∗ ꢅ
−37.3
−22.1
−20.6
−8.8
ꢡꢃꢄ∗ꢅ ∆Gꢄ∗ ꢅ ꢡꢃꢄ∗ꢅ
∆Gꢄ∗ ꢅ
ꢡꢃꢄ∗ꢅ
∆Gꢄ∗ ꢅ
Me2N −27.2
−25.5 −35.7 −25.1 −34.2
−14.4 −19.7 −13.9 −19.0
−13.2 −18.1 −12.7 −17.4
−25.8 −35.1
−13.8 −19.0
−11.8 −16.2
MeO
H
−16.1
−15.0
−6.4
0.7
CF3
CN
−3.9
3.3
−6.0
4.4
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
1.0
The experimental pKOH and ꢡꢃꢄ∗ꢅ values correlate linearly with the Hammett σ coefficients49
for the Xꢀsubstituents (Figure 2), demonstrating that the simple freeꢀenergy relationship concept
can be used efficiently to predict the driving force for the OH− release. Similar linear relationship
with a slope of opposite sign was observed for the photoacids,34 indicating that the driving force
for H+ or OH− release is controlled by stabilization of charges formed upon the ion release (RO−
for photoacids and R+ for photobases). Interestingly, the slope ρ is significantly steeper for the
excitedꢀstate ꢡꢃꢄ∗ꢅ values (ρ*=14.8 in water) than for the groundꢀstate pKOH values (ρ=0.6 in
water). Thus, the electronꢀdonating substituents in the Xꢀposition appear to stabilize more
efficiently the positive charge of XꢀAcr+ in its S1 state than in the S0 state. Similar effects were
observed in naphtholꢀbased photoacids, where the substituent effects were found to correlate
with the calculated charge densities of relevant atoms in their ground and excited states.3 It is
likely that a similar effect holds for XꢀAcrOH photobases.
ACS Paragon Plus Environment
15

The Journal of Physical Chemistry
Page 16 of 37
1
2
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
Figure 2. Hammet plots for ground (squares and solid lines) and excitedꢀstate (circles and
dashed lines) OH− release from XꢀAcrOH in water (blue) and methanol (red). The slopes are as
follows: for ground states ρ = 0.6 in water and 1.2 in methanol; for excitedꢀstates ρ* = 14.8 in
water and 17.6 in methanol.
KINETICS
The rates of OH− release from model acridinols were determined using pumpꢀprobe
spectroscopy. The spectral changes in the visible range were consistent with the photochemical
heterolysis of the C−O bond (Figure 3). For example, the transient absorption spectrum of the
methanol solution of Me2NꢀAcrOH at early times (red trace, Figure 3A) was assigned to its S1
state and exhibited broad features throughout the visible range with λmax=748 nm. The S1 signal
decayed within few hundreds picoseconds and was accompanied with a concomitant growth of a
new transient that absorbed at λmax=553 nm. The absorption spectrum of this growing transient
matched well to the spectrum of Me2NꢀAcr+ cation in the S0 state (Figure 3A, dashed line),
which is consistent with the excited state OH− release from Me2NꢀAcrOH. In a similar fashion,
the conversion from excited MeOꢀAcrOH, AcrOH and CF3ꢀAcrOH in methanol leads to the
formation of the corresponding cations (Figures 3BꢀD). In the case of CNꢀAcrOH, the excitedꢀ
state OH− release was not observed, which is consistent with the unfavorable thermodynamics
ACS Paragon Plus Environment
16

Page 17 of 37
The Journal of Physical Chemistry
1
2
(ꢁG* = + 1.0 kcal/mol, Table 2) and shows that the rates of the OH− release correlate with the
ꢁG* values calculated using simple Förster cycle.
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
Figure 3. Femtosecond transient absorption spectra of XꢀAcrOH in MeOH at several time delays
after the 310 nm excitation pulse: A) Me2NꢀAcrOH; B) MeOꢀAcrOH; C) AcrOH; D) CF3ꢀ
ACS Paragon Plus Environment
17

The Journal of Physical Chemistry
Page 18 of 37
1
2
3
4
5
6
AcrOH; E) CNꢀAcrOH. The dashed lines represent the groundꢀstate absorption spectra for the
corresponding XꢀAcr+ cations.
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
Two similarities can be drawn between the photoꢀheterolysis observed in photoacids37, 50 and
the photobases studied here: (i) the rates of heterolysis correlate well with the driving force for
the adiabatic process (ꢁG*), as discussed in the previous paragraph; (ii) the excitedꢀstate
heterolysis occurs predominantly in protic solvents (such as alcohols). For example, MeOꢀ
AcrOH undergoes efficient excitedꢀstate OH− release in nꢀBuOH (Figure 4A), while S1→T1 ISC
occurs in acetonitrile (Figure 4B). Such dependence is suggestive of the important role that the
solvent reorganization energy plays in the H+ and OH− release from organic alcohols.2, 48 One
notable difference exists between the excited state behavior of photoacids and photobases: the
PT generally occurs adiabatically, while the OH− transfer occurs nonꢀadiabatically in most
photobases. For example, the fluorescence spectra of photoacids in protic solvents generally
contain a variable degree of the redꢀshifted emission due to the conjugate base formed upon the
PT, confirming its adiabatic nature.51 On the other hand, the transient absorption spectra of
X−AcrOH derivatives (Figure 3 and ref. 35) show that X−Acr+ cations are formed in the ground
state, suggesting a nonꢀadiabatic process. To the best of our knowledge, only two previous
reports showed adiabatic OH− release from a xanthenol photobase,23, 52 but even in this case the
adiabatic process was a minor heterolysis channel, while most of the OH− release occurred nonꢀ
adiabatically.
ACS Paragon Plus Environment
18

Page 19 of 37
The Journal of Physical Chemistry
1
2
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
Figure 4. Femtosecond transient absorption data for MeOꢀAcrOH in A) nꢀBuOH and B) ACN.
The samples were excited at 310 nm.
The kinetics of S1 states in all model X−AcrOH derivatives can be readily obtained by probing
the red portion of the visible spectrum (for example λ=750 nm, Figure 5). The observed rate of
S1 decay increases along X−AcrOH series (Figure 5A), with faster decay occurring for electronꢀ
donating groups (such as Me2N−AcrOH). The result is consistent with the ꢁG* values listed in
Table 2: the photoꢀheterolysis rate increases as the process becomes more thermodynamically
favorable. The solvent effects on the rate of OH− release are shown in Figure 5B: the faster
heterolysis is observed in higher polarity solvent (the solvent dielectric constants are as follows:
εMeOH=33.0 εnꢀBuOH=17.8, εiꢀBuOH=17.953). The solvent effects are consistent with increased
stability of product ions by polar solvents, and this stabilization is reflected in the
thermodynamic driving forces (Table 2). The polarities of nꢀBuOH and i-BuOH are similar, but
ACS Paragon Plus Environment
19

The Journal of Physical Chemistry
Page 20 of 37
1
2
3
4
iꢀBuOH is bulkier. The observed rate does not change significantly in the two solvents,
5
6
indicating that the solvent polarity plays the deciding role on the rate of OH− release.
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
Figure 5. A) Normalized transient absorption traces (λprobe=750 nm) for XꢀAcrOH solutions in
MeOH (λpump=310 nm). Lifetime: Me2NꢀAcrOH 53 ps; MeOꢀAcrOH 76 ps; AcrOH 108 ps; CF3ꢀ
AcrOH 1.4 ns; CN−AcrOH 6.3 ns; B) The kinetic trace for AcrOH (λpump=310 nm; λprobe=750
nm) in MeOH (108 ps), nꢀBuOH (357 ps) and i-BuOH (319 ps).
In all model compounds, the 750 nm signal exhibited monoꢀexponential decay with rate
constants kexp (Table 3). As the rate of OH− release slows down across the XꢀAcrOH series, the
process starts to compete with the S1→T1 IC. Our earlier work on AcrOH35 has shown that the
excitedꢀstate OH− release occurs only in protic solvents (methanol), while ISC dominates the
excitedꢀstate deactivation in aprotic solvents (acetonitrile). Similar behavior in aprotic solvents
was observed for most XꢀAcrOH derivatives (data not shown), which allowed the determination
of ISC rates for each acridinol (kISC). The competition between heterolysis and ISC can be
visualized in the case of CF3ꢀAcrOH (Figure 3D). In addition to the growth of CF3ꢀAcr+
ACS Paragon Plus Environment
20

Page 21 of 37
The Journal of Physical Chemistry
1
2
3
4
5
6
absorption bands (λABS=425 nm), additional absorption appears in the 500ꢀ600 nm and is
assigned to the T1 state of CF3ꢀAcrOH. Thus, the rates for OH− release were obtained as follows:
7
8
9
kOH = kexp − kISC
(3)
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
The resulting kOH values are listed in Table 3. The use of equation (3) assumes that the rate of
ISC does not change significantly when going from acetonitrile to the alcohols studied here.
Such assumption likely introduced an error in the estimated kOH values, since previous reports
indicate that some compounds exhibit relatively strong solvent effects on their ISC rates.54ꢀ55
However, since the triplet excited state absorption energy does not change significantly in
different solvents (for example, triplet excited state of MeOꢀAcrOH absorbs at 550 nm in both
ACN and nꢀBuOH, Figure 4), it is likely that the above assumption is justified.
Table 3. Experimental rate constants (kexp) for the singlet excited state decays of XꢀAcrOH
models and the derived rates for OH− release (kOH) determined as described in the text.
MeOH
nꢀBuOH
iꢀBuOH
kOH/.109 sꢀ kexp/.109 sꢀ kOH/.109 sꢀ
X
kexp/.109 sꢀ1 kOH/.109 sꢀ1 kexp/.109 sꢀ1
1
1
1
Me2N 0.2
MeO 0.1
0.2
0.1
9.3
53.3
ꢀ
3.7
3.4
3.1
−
3.0
2.4
2.4
−
4.3
3.8
3.1
−
3.6
2.9
2.4
−
H
9.3
CF3
CN
70.9
15.9
−
−
−
−
The obtained kOH values were used to construct their dependence on the thermodynamic
driving force (Figure 6). To model the rate constant dependence on the thermodynamic driving
ACS Paragon Plus Environment
21

The Journal of Physical Chemistry
Page 22 of 37
1
2
3
4
5
6
force, the wellꢀknown transitionꢀstate theory expression was used. It relates the rate constants
with the corresponding energy barriers:
7
8
9
ꢦ
ꢤꢥ
ꢢ = ꢘ ∙ ꢣꢉ
(4)
ꢧ∙ꢨ
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
where A is the preꢀexponential factor and ꢀG≠ is the Gibbs free energy for the transition state.
Two theoretical models were then used to correlate the dependence of ꢀG≠ on the
thermodynamic driving force, ꢀG0. One treatment utilized the Marcus model,56ꢀ59 which has
been successfully applied to group transfers in which reaction rate is controlled by the response
of the solvent to the charge redistribution. In this case, the reaction proceeds along the solvent
reaction coordinate and the free energy correlation is:
ꢖ
ꢫ
∆ꢪ
∆ꢀꢂ = ꢩ1 + ꢬ∙∆ꢪꢦꢭ ∙ ∆ꢀꢁꢂ
(5)
ꢫ
where ∆ꢀꢁꢂ is the activation energy for the symmetric group transfer (ꢀG0=0). The second
treatment utilized the bondꢀenergy bondꢀorder (BEBO) theory,60ꢀ62 which models reactions that
occur in a strong coupling regime, where the bond stretching and compression define the
reaction coordinate. The BEBO model evaluates the energy of a reacting system at any point
along the reaction coordinate using a linear combination of reactant and product energies, which
are scaled by their bond order at a given point of the reaction coordinate. According to BEBO
model, the free energy correlation is expressed as:
ꢦ
ꢫ
ꢫ
∆ꢀꢂ = ∆ꢪ + ∆ꢀꢁꢂ + ∆ꢪ ∙ ꢰꢱ ꢲꢙꢳꢴℎ ꢩ∆ꢪ ∙ꢮꢯꢖꢭꢵ
(6)
ꢫ
ꢦ
ꢖ
ꢮꢯꢖ
ꢖ∙∆ꢪ
ꢫ
At lower driving forces, both Marcus and BEBO models predict similar rate dependence on the
driving force, where the rate increases with increasing driving force. At large driving forces,
however, the two models predict different behavior: while the Marcus model predicts a decrease
in the rate at high driving forces (inverted region), the BEBO model predicts the rates to level off
ACS Paragon Plus Environment
22

Page 23 of 37
The Journal of Physical Chemistry
1
2
at high driving forces (Figure 6). In the case of OH− release from photobases, it is not known
whether the rateꢀdetermining process involves predominantly the solvent reorganization or the
reorganization of the molecular bonds; therefore the data were evaluated using both models.
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
Figure 6. The dependence of kOH values on thermodynamic driving forces for the excitedꢀstate
OH− release from model XꢀAcrOH derivatives. The fits were obtained using either Marcus (solid
lines) or BEBO (dashed lines) model. The results of a previous study involving a xanthenol
(XanOH) photobase36 in acetonitrile/water mixtures are included for comparison.
Unfortunately, the limited range and number of experimental data points precluded us from
evaluating whether Marcus or BEBO theory is better at modeling the excitedꢀstate OH− release
from our model compounds. However, it is interesting to note that the reorganization energies λ
obtained using the Marcus model (Table 4) for AcrOH photobases in alcoholꢀbased solvents
match reasonably well with the expected solvent reorganization energies λs at the contact
distance calculated using the dielectric continuum expression:63ꢀ64
ꢹ
ꢸ
ꢼ
ꢼ
ꢼ
ꢼ
ꢶꢷ = ꢬꢺꢻ ꢩꢖꢟ
+
− ꢼꢟꢭ ꢿꢻ − ꣃ ∙ ꣄ꢝ
(7)
ꢖꢟ
ꢻ
ꣂ
ꢫ
ꢽ
ꢾ
ꣀꣁ
ACS Paragon Plus Environment
23

The Journal of Physical Chemistry
Page 24 of 37
1
2
3
4
5
6
7
8
9
where εop and εs are optical and static dielectric constants of the solvent, rD and rA are the
donor and acceptor radii (we assumed that the radii are 3 Å each) and r is the distance between
donor and acceptor. Based on these values, it appears that the excited state OH− release from
model acridinol photobases occurs along the reaction coordinate involving solvent responses and
can be predicted well using the Marcus model.
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
Table 4. Summary of the parameters obtained from Marcus and BEBO analyses.
Marcus
A
BEBO
A
solvent
λb
kcal/mol
Δꢀꢁꢂkcal/mol
Δꢀꢁꢂkcal/mol
λs kcal/mol
ACN/H2Oa 11.5 10.8
43.2
28.3
26.6
26.6
29.1ꢀ30.5
29.8
11.5 10.2
10.3 4.5
MeOH
10.6 7.1
9.9 6.7
10.0 6.7
nꢀBuOH
iꢀBuOH
25.2
9.5
9.6
3.7
3.6
25.3
a. For XanOH photobase, taken from ref. 36.
b. ꢶ = 4Δꢀꢁꢂ
The above results obtained in alcohol solvents are in contrast with the results reported in our
previous study involving xanthenol (XanOH) photobase in acetonitrile/water mixtures.36 The
Marcus fit from this study provided a value for reorganization energy of λ=43.2 kcal/mol, which
is significantly higher than the solvent reorganization energy of λs=29.1ꢀ30.5 kcal/mol calculated
using the dielectric continuum model for acetonitrile and water (Table 4). Since the structures of
XanOH and AcrOH photobases are quite similar, the differences in reorganization energies are
likely due to solvents used. The dielectric continuum model is known to fail in predicting the
ACS Paragon Plus Environment
24

Page 25 of 37
The Journal of Physical Chemistry
1
2
solvent reorganization energies in cases when there are specific reactantꢀsolvent interactions.65ꢀ67
When such nonꢀcontinuum factors are present, the experimentally observed rates are generally
slower than those predicted by the continuum model, which is consistent with our results for
XanOH in water/ACN mixtures.36
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
PHOTOꢀMICHAEL ADDITION
The utility of photobases to drive chemical reaction has been reported previously, namely to
drive Henry41 and thiolꢀMichael additions.68ꢀ69 In both of these examples, the photochemical
event generates an amine, which serves as a base to drive the chemical reaction of interest. In a
similar fashion, the acridine photobases could be utilized to drive photochemically the reactions
that require either hydroxide ions or other anionic species (such as alkoxyde ions). In order to
investigate this possibility, AcrOMe derivative was used as a system that, upon excitation,
releases methoxyde (MeO−) ion. The photogenerated MeO− ions were successfully utilized in a
model Michael addition between dimethylmalonate 1 and nitrostyrene 3 (Scheme 2), in a
reaction that involves three individual steps: (i) deprotonation of malonate 1 using the
photogenerated MeO− ions; (ii) nucleophilic attack of the enolate 2 to the electronꢀdeficient C=C
bond of nitrostyrene 3; (iii) protonation of the intermediate 4 to give the final product 5.
O
O
O
O
AcrOMe, hν
O
O
O
O
NO2
MeO
OMe
NO2
MeO
OMe
NO2
H+
∆
+
MeO
OMe
MeO
OMe
AcrOMe, ∆
step I
1
2
3
4
5
step III
step II
Scheme 2. Model PhotoꢀMichael addition between nitrostyrene and dimethylmalonate using a
lightꢀtriggered base AcrOMe.
While Michael additions often requires only catalytic amounts of the base, the model reaction
presented in Scheme 2 requires stoichiometric amounts of the MeO− ion. Experiments that
ACS Paragon Plus Environment
25

The Journal of Physical Chemistry
Page 26 of 37
1
2
utilized a lower number of MeO− equivalents yielded less product, indicating that the turnover
frequency (TOF) of the catalytic cycle is not sufficiently high to compete with other unwanted
chemical reactions. The likely reason for low catalytic TOF is the low basicity of the
intermediate 4, which prevents it from abstracting a proton from malonate 1 (a step that is
needed to close the catalytic cycle): the acidity of compound 5 is expected to be similar to that of
nitromethane (pKa=10),70 while the acidity of malonate 1 is pKa=13.71 However, this model
reaction is expected to serve as a proofꢀofꢀprinciple study for development of future reactions in
which the pKa values of relevant reactants are tailored towards catalysis.
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
When the reaction mixture containing AcrOMe was photoꢀirradiated, the product yield was
o
found to be 60%. The reaction mixture was heated to 65 C, since the addition step II does not
occur at room temperature. When the same reaction was performed in the absence of light, the
product yield was only 38%, indicating that the AcrOMe performance can be significantly
improved using photochemical methods. However, the results show that reaction occurs even in
the absence of irradiation, possibly due to the thermal heterolysis of AcrOMe bond. This finding
is consistent with the fact that the Gibbs free energy for AcrOMe heterolysis is expected to be
ꢁGOMe~6.4 kcal/mol (similar to that of AcrOH, Table 1). At the reaction temperature of 65 oC, at
least 1% of the AcrOMe is expected to thermally dissociate into Acr+ and MeO− ions, which can
drive the ‘dark’ Michael addition. The contrast in the reaction yields between the dark and
photochemical Michael reactions can likely be increased if the choice of reactants was made to
perform the reaction at room temperature.
CONCLUSIONS
The excitedꢀstate OH− transfer has been studied to a lesser degree relative to the corresponding
PT reactions. This manuscript aims to bridge the gap by investigating the parameters that affect
ACS Paragon Plus Environment
26

Page 27 of 37
The Journal of Physical Chemistry
1
2
the kinetics of the OH− release from model acridinol photobases. The thermodynamic driving
force for the reaction was tuned using a series of acridinol derivatives with varying substituents
on the phenyl ring, while the reaction rates were determined using timeꢀresolved laser
spectroscopy. The experimentally obtained rate constants and Gibbs free energies were fit using
the Marcus model for group transfer reactions. The fitted data provided an intrinsic
reorganization energy, which matches surprisingly well with the solvent reorganization energies
calculated using a simple dielectric continuum model. The results of this study indicate that the
solvent reorganization plays a major role in the kinetics of the excitedꢀstate OH− release and can
be efficiently modeled using the Marcus model. These results are in contrast to our previous
study involving a xanthenol photobase in acetonitrile/water mixtures,36 where the experimentally
obtained reorganization energy was significantly higher than the solvent reorganization energy
predicted by the dielectric continuum model. The observed discrepancy could be due to strong
specific solventꢀsolute interactions present in aqueous environment.
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
This manuscript also evaluates the utilization of acidinol photobases in driving useful chemical
reactions. Specifically, the Michael addition between dimethylmalonate and nitrostyrene was
successfully performed using MeO− base that was photochemically generated from the
corresponding AcrOMe derivatives. While the reported reaction also occurred thermally (with a
lower yield), the results indicate that the photoꢀMichael addition reported here can be further
optimized and developed into a useful photochemical tool for applications where spatialꢀ
temporal control is needed.
AUTHOR INFORMATION
*Corresponding author eꢀmails: kglusac@bgsu.edu (K. D. G.)
ACS Paragon Plus Environment
27

The Journal of Physical Chemistry
Page 28 of 37
1
2
1Current address: Department of Petroleum Engineering, University of Wyoming, Laramie,
Wyoming 82071, United States
3
4
5
6
7
8
9
ACKNOWLEDGEMENTS
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
KDG thanks National Science Foundation (1055397) and Petroleum Research Fund (54436ꢀ
ND4) for the financial support.
REFERENCES
1. Arnaut, L. G.; Formosinho, S. J., ExcitedꢀState Proton Transfer Reactions I.
Fundamentals and Intermolecular Reactions. J. Photochem. Photobiol. A 1993, 75 (1), 1ꢀ20.
2. Tolbert, L. M.; Solntsev, K. M., ExcitedꢀState Proton Transfer: From Constrained
Systems to “Super” Photoacids to Superfast Proton Transfer. Acc. Chem. Res. 2002, 35 (1), 19ꢀ
27.
3. Agmon, N., Elementary Steps in ExcitedꢀState Proton Transfer. J. Phys. Chem. A 2005,
109 (1), 13ꢀ35.
4. Huppert, D.; Kolodney, E.; Gutman, M.; Nachliel, E., Effect of Water Activity on the
Rate of Proton Dissociation. J. Am. Chem. Soc. 1982, 104 (25), 6949ꢀ6953.
5. Agmon, N.; Huppert, D.; Masad, A.; Pines, E., ExcitedꢀState Proton Transfer to
MethanolꢀWater Mixtures. J. Phys. Chem. 1991, 95 (25), 10407ꢀ10413.
6. Rini, M.; Magnes, B.ꢀZ.; Pines, E.; Nibbering, E. T., RealꢀTime Observation of Bimodal
Proton Transfer in AcidꢀBase Pairs in Water. Science 2003, 301 (5631), 349ꢀ352.
ACS Paragon Plus Environment
28

Page 29 of 37
The Journal of Physical Chemistry
1
2
3
4
7. Mohammed, O. F.; Pines, D.; Dreyer, J.; Pines, E.; Nibbering, E. T., Sequential Proton
5
6
Transfer Through Water Bridges in AcidꢀBase Reactions. Science 2005, 310 (5745), 83ꢀ86.
7
8
9
8. Siwick, B. J.; Bakker, H. J., On the Role of Water in Intermolecular ProtonꢀTransfer
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
Reactions. J. Am. Chem. Soc. 2007, 129 (44), 13412ꢀ13420.
9. Liu, W.; Han, F.; Smith, C.; Fang, C., Ultrafast Conformational Dynamics of Pyranine
during Excited State Proton Transfer in Aqueous Solution Revealed by Femtosecond Stimulated
Raman Spectroscopy. J. Phys. Chem. B 2012, 116 (35), 10535ꢀ10550.
10. Adamczyk, K.; PrémontꢀSchwarz, M.; Pines, D.; Pines, E.; Nibbering, E. T., RealꢀTime
Observation of Carbonic Acid Formation in Aqueous Solution. Science 2009, 326 (5960), 1690ꢀ
1694.
11. Das, R.; Mitra, S.; Nath, D.; Mukherjee, S., Excited State Proton Transfer Reaction as a
Probe for the Microenvironment of a Binding Site of Bovine Serum Albumin: Effect of Urea. J.
Phys. Chem. 1996, 100 (34), 14514ꢀ14519.
12. Nachliel, E.; Pollak, N.; Huppert, D.; Gutman, M., TimeꢀResolved Study of the Inner
Space of Lactose Permease. Biophys. J. 2001, 80 (3), 1498ꢀ1506.
13. Cohen, B.; Martin Álvarez, C.; Alarcos Carmona, N.; Organero, J. A.; Douhal, A.,
ProtonꢀTransfer Reaction Dynamics within the Human Serum Albumin Protein. J. Phys. Chem.
B 2011, 115 (23), 7637ꢀ7647.
14. Shvadchak, V. V.; FalomirꢀLockhart, L. J.; Yushchenko, D. A.; Jovin, T. M., Specificity
and Kinetics of αꢀSynuclein Binding to Model Membranes Determined with Fluorescent Excited
ACS Paragon Plus Environment
29