Efficient oxidative cycloreversion reaction of photochromic dithiazolythiazole

Article abstract of DOI:10.1021/ja309275q

Electrochemical response of photochromic tearylenes was surveyed by means of cyclic voltammetry, DFT calculations, and spectroelectrochemistry. 4,5-Bis(2-phenyl-5-methylthiazolyl)-2-phenylthiazole was found to show electrochemical oxidative ring-cycloreversion reaction. The net current efficiency of the cycloreversion reaction under constant potential electrolysis was as high as 900%, which is ascribed to an electrochemical local-cell mechanism and a chain reaction mechanism. Electron transfer stopped-flow study using a chemical oxidant successfully identified radical cation intermediates of both closed- and open-ring isomers, involved in the oxidative cycloreversion process. The significantly long-lived radical cation of open-ring isomer with the lifetime of 33 s takes part in the indirect electron transfer process from the neutral closed-ring isomer to the radical cation of open-ring isomer in the chain reaction manner.

Full text of DOI:10.1021/ja309275q

Subscriber access provided by Milner Library | Illinois State University
Article
Efficient Oxidative Cycloreversion Reaction
of Photochromic Dithiazolythiazole
Takuya Nakashima, Yoshiyuki Kajiki, Sayo Fukumoto, Maki
Taguchi, Satoshi Nagao, Shun Hirota, and Tsuyoshi Kawai
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/ja309275q • Publication Date (Web): 19 Nov 2012
Downloaded from http://pubs.acs.org on November 21, 2012
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.
Journal of the American Chemical Society 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 21
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
Efficient Oxidative Cycloreversion Reaction of
Photochromic Dithiazolythiazole
Takuya Nakashima, Yoshiyuki Kajiki, Sayo Fukumoto, Maki Taguchi, Satoshi Nagao, Shun Hirota, and
Tsuyoshi Kawai*
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
Graduate School of Materials Science, Nara Institute of Science and Technology,
NAIST, Ikoma, Nara 630-0192, Japan
Author Information
*: corresponding author
E-mail: tkawai@ms.naist.jp
Notes
The authors declare no competing financial interest.
ACS Paragon Plus Environment
1

Journal of the American Chemical Society
Page 2 of 21
Abstract
1
2
3
4
5
6
7
8
9
Electrochemical response of photochromic tearylenes were surveyed by means of cyclic voltammetry,
DFT calculations and spectroelectrochemistry. 4,5-Bis(2-phenyl-5-methylthiazolyl)-2-phenylthiazole
was found to show electrochemical oxidative ring-cycloreversion reaction. The net current efficiency of
the cycloreversion reaction under constant potential electrolysis was as high as 900%, which are
ascribed to an electrochemical local-cell mechanism and a chain reaction mechanism. Electron transfer
stopped-flow study using a chemical oxidant successfuly identified radical cation intermediates of both
closed- and open-ring isomers, involved in the oxidative cycloreversion process. The significantly long-
lived radical cation of open-ring isomer with the lifetime of 33 s takes part in the indirect electron
transfer process from the neutral closed-ring isomer to the radical cation of open-ring isomer in the
chain reaction manner.
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
ACS Paragon Plus Environment
2

Page 3 of 21
Journal of the American Chemical Society
Introduction
1
2
3
4
5
6
7
8
9
Photoresponsive molecular materials have attracted much interest because their physicochemical
properties are modulated with precise spatio- and temporal- control in a remote and noninvasive manner,
which is required for a wide range of applications.1-6 Photoswitching systems based on the reversible
peri-cyclization reaction of hexatriene framework have been extensively studied, which is responsible
for the photochromism of fulgides,7 dihydropyrenes,8 diarylethenes9 and their related compounds.
Recently, we have proposed photochromic terarylenes composed of three heteroaryl groups to form a
hexatriene backbone, that undergo photo-cyclization and cycloreversion reactions in a similar manner to
diarylethenes.10,11 A variety of combinations of aromatic groups offered the modulation of thermal
bleaching of colored isomers in a wide dynamic range,11 and photo-switching of the properties relevant
to the central bridging unit including chemical reactivity,12,13 fluorescence intensity,14-16 and
luminescent properties of metal complexes.17,18 The precise placement of heteroatom in terarylene
molecules successfully controls the molecular folding via multiple weak interactions such as
CH/heteroatom, inter-heteroatom and CH/π interactions in intra-19 and intermolecular20 manners to be in
the conformation favorable for the contrarotatory cyclization reaction, leading to efficient photo-
cyclization reaction with quantum yield above 90%. Given both the photo-cyclization and -
cycloreversion reactions proceed via the identical conical intersection in the potential energy surfaces,
the efficiency of reactions should be dependent on the nature of conical intersection.21,22 Therefore, it
seems difficult to satisfy the high quantum yields for both of photo-cyclization and -cycloreversion
reactions. For instance, 2,3-dithiazolylbenzothiophene, which exhibits “photon-quantitative reaction”
with cyclization quantum yield of almost unity, indeed indicates the cycloreversion reaction quantum
yield smaller than 1%.19c Additional external stimuli such as multi-photon excitation,23,24 heat,25 or
electric potential to trigger the cycloreversion isomerization reactions would be desirable for efficient
reversible switching system. The last one is also expected to develop more complex and versatile dual-
mode hybrid optical and electronic materials.25
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
ACS Paragon Plus Environment
3

Journal of the American Chemical Society
Page 4 of 21
Electrochemical ring-cyclization reaction were reported for hexatriene-based photochromic molecules
such as fulgides27 and dihydropyrenes.28 Oxidative cycloreversion reaction was first reported for a
closed-ring isomer of 1,2-dicyano-dithienylethene.29 Chemical modifications on the structure of
diarylethenes such as the substitution of the central bridging unit with hexahydrocyclopentene ring,30,31
the extension of π-conjugation system,32-34 and the introduction of various substituents30,31,35,36 have
afforded diverse electrochemical responses. Electrochemical isomerization reactions also undergo with
the aid of redox reactions of organometallic attached to diarylethenes as the side chain units.37,38 The
most fascinating property of electrochemical reaction is that it often involves indirect processes, which
increases the net current efficiency of the reaction system.39 For example, electrochemically generated
radical cation of closed-ring isomer spontaneously converts into more stable radical cation of open-ring
isomer, which has enough high oxidative power to withdraw an electron from neutral closed-ring isomer
to trigger the chain reaction.29,33
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
Here we first describe the electrochemical behavior of photochromic terarylenes (Scheme 1). Since
radical intermediates were reported to take part in the electrochemical isomerization processes, the
stabilization of radical intermediates of terarylenes with three heteroaryl groups is expected to modify
the electrochemical isomerization reaction. Spectro-electrochemical characterizations revealed that 4,5-
bis(2-phenyl-5-methyldithiazolyl)-2-phenylthiazole (1) showed an oxidative cycloreversion reaction
with a significantly high current efficiency. Two distinctive indirect electrochemical pathways are
responsible for the enhanced efficiency. We also successfully monitored the successive evolution of
both radical cation intermediates of closed- and open-ring isomers of dithiazolylthiazole with the aid of
an electron transfer stopped-flow method, which provided the detailed kinetics of oxidative
isomerization processes.40
ACS Paragon Plus Environment
4

Page 5 of 21
Journal of the American Chemical Society
1
2
3
Scheme 1
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
Results and Discussion
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
Survey of Electrochemical Properties. Electrochemical oxidative responses of open- and closed-
ring isomers of terarylenes 1–5 were surveyed with the cyclic voltammetry (CV). Terarylenes 1–5
possess a thiazolyl (1a, 3a–5a) or a benzothienyl (2a) rings as a central bridging unit with symmetric
side aromatics of phenylthiazole (1a, 2a), phenylthiophene (3a), benzothiophene (4a) or di-oxidized
benzothiophene (5a). As illustrated in Figure 1, open ring isomers of terarylenes except for 5a with
dioxidized-benzothienyl rings showed irreversible oxidation peaks above 0.9 V (vs Fc/Fc+), which are
similar to those observed for some open-ring isomers of dithienylethenes and are considered to include
both the first and second oxidation processes. 31-33 Probably due to the strong electron-accepting nature,
5a showed no oxidation peak below 1.5 V. Meanwhile, closed-ring isomer 5b showed an almost
reversible redox couple (E1/2 = 1.0 V). Closed-ring isomers 1b, 2b and 4b gave semi-reversible two
successive oxidation peaks. Especially, 1b and 4b showed the third oxidation peak at the same potential
with those of corresponding open-ring isomers 1a and 4a, respectively, suggesting the formation of
open-ring isomers with the electrochemical oxidation of the closed-ring isomers.
ACS Paragon Plus Environment
5

Journal of the American Chemical Society
Page 6 of 21
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 1. Cyclic voltammograms of (a) 1a (blue) and 1b (red); (b) 2a (blue) and 2b (red); (c) 3a (blue)
and 3b (red); (d) 4a (blue) and 4b (red); (e) 5a (blue) and 5b (red). Measurements were performed in
dry acetonitrile (1 × 10-3 M) with 0.1 M of TMAPF6 at scan rate of 50 mVs-1.
The direction of isomerization reaction subsequent to the electrochemical oxidation was deduced by
the relative potential energy of oxidized forms (i.e. radical cations) of open-(a-form) and closed-ring (b-
form) isomers calculated by the DFT method (Table 1, also see Supporting Information). The only small
energy differences between a·+ and b·+ were found for terarylenes 1–4. According to the hypothesis
reported by Matsuda et al.33 1, 2, 3 and 5 should undergo the anodic cyclization reaction, whereas 4 is
possible to show the cycloreversion reaction. However, the actual directions of the anodic reactions
were not accorded with those predicted based on the DFT calculation except for 4 (Table 1). In addition
to the small energy differences, the solute-solvent and the ionic interactions were not factored in the
calculations and therefore it was difficult to predict the direction of isomerization reaction for 1–4. The
anodic reaction was monitored by UV-vis absorption spectral change under varied potentials (Figure
S2–S6). As summarized in Table 1, only dithiazolylthiazole 1 showed the clear subsequent
cycloreversion isomerization reaction. Except for 3a, open-ring isomers showed no significant change in
ACS Paragon Plus Environment
6

Page 7 of 21
Journal of the American Chemical Society
absorption spectra up to 1.5 V. Closed-ring isomers 1b, 2b and 4b, which gave similar CV responses,
exhibited continuous decrease in absorption band in visible region, suggesting cycloreversion reactions.
Dithienylthiazole 3 provided complex spectral changes both for open- and closed-ring isomers and the
spectra after the electrolysis from 3a and 3b were different from those of 3b and 3a, respectively.
Closed-ring isomer 5b also showed continuous decrease in visible band at 470 nm, while new band
appeared at 680 nm, which may correspond to the formation of radical cation of 5b. Eventually, only
dithiazolylthiazole 1 afforded a unidirectional specific subsequent reaction. We thus focus hereafter on
the oxidative cycloreversion reaction of 1b by means of electrochemical and chemical oxidation
reactions.
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
Table 1. Electrochemical Properties of Terarylenes, the Energy Difference of Radical Cations,a and
Reactivity.
Eox, onset / V vs Fc/Fc+
Subsequent
reaction
ꢀ
E(a·+ – b·+)
terarylenes
a-form
0.89
0.85
0.83
0.94
–
b-form
0.24
0.22
/ kcalmol-1
0.168
2.71
cycloreversion
1
2
3
4
5
cycloreversionc
d
–0.19
0.33
5.08
–2.00
16.85
–
cycloreversionc
0.89b
–
e
aTotal energy of radical cations in vacuum was calculated by DFT at UB3LYB/6-31+G*. Reversible
b
c
d
response was observed. The reaction was not completed above 1.3V. Decomposition reactions were
indicated both for a- and b-forms. e The formation of 5b·+ was indicated.
ACS Paragon Plus Environment
7

Journal of the American Chemical Society
Page 8 of 21
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. Change of absorption spectrum of closed-ring isomer 1b by electrolysis at 0.42 V (vs Fc/Fc+)
in acetonitrile (2.5 × 10-4 M).
Electrolysis of 1b. Since closed-ring isomer 1b showed oxidation peak at 0.35 V (vs Fc/Fc+), the
absorption spectral change was monitored under electrolysis condition at 0.42 V in acetonitrile. An
absorption band at around 600 nm typical for closed-ring isomer 1b continuously decreased during the
electrolysis. The electrolysis reaction of 1b was typically completed within 5 min. The absorption
spectrum after electrolysis was identical with that of open-ring isomer 1a. The spectral change with an
isosbestic point at 323 nm resembles to that observed for the photo-induced cycloreversion reaction
from 1b to 1a.11a The final product after the electrolysis was confirmed as open-ring isomer 1a by
HPLC and 1H NMR, and also showed a coloration reaction with UV irradiation to form 1b.
ACS Paragon Plus Environment
8

Page 9 of 21
Journal of the American Chemical Society
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 3. (a) Absorbance (at 587 nm) (solid line) and current (broken line) change and (b)
accumulations of reacted 1b (solid line) and quantity of electricity (broken line) as a function of time of
electrolysis. Inset shows the enlarged graph of (b); x and y-axes are identical with those of (b).
We then evaluated the efficiency of electrolysis in terms of molar-amount of converted 1b as a
function of quantity of electricity during the electrolysis. The consumption of 1b by the electrolysis was
monitored by the change in absorbance at 587 nm (Figure 3a). The absorbance at 587 nm of 1b
continuously decreased and finally converged to zero after the electrolysis for 150 s, which indicates
that all molecules of 1b in the electrolysis cell were consumed and converted to 1a by the 150s-
electrolysis. Meanwhile, the current flow also decreased and reached the background after the
electrolysis for 70 s; that is, after 70 s, the conversion reaction of 1b did not require current injection
from the working electrode, clearly demonstrating indirect processes. The current flow was converted to
the quantity of electricity, which corresponds to the integral electric consumption, using Faraday
constant together with the period of electrolysis after the subtraction of background current. The electric
ACS Paragon Plus Environment
9

Journal of the American Chemical Society
Page 10 of 21
quantity was compared with the amount of reacted 1b (Figure 3b). Usually, 1 mol of reactant requires
1F of electricity for the quantitative electrochemical reaction with a direct process. As shown in the
inset in Figure 3b, the reaction rate of 1b almost corresponded to the electric quantity at the early stage
of electrolysis within 5 s. After that, the consumption rate of 1b was much larger than the evolution of
electric quantity. Finally, the total quantity of electricity used was only 7.4 × 10-9 mol·electron, whereas
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
the amount of reacted 1b was 6.7× 10-8 mol. Therefore, more than 900 % of net current efficiency was
recorded for the electrolysis of 1b. The instantaneous current efficiency, which was calculated for every
5 seconds during the electrolysis, surprisingly reached over 2000% after 60 s (Figure S7).
Figure 4. Two plausible indirect reaction mechanisms; (a) chain reaction model and (b) local cell
reaction model.
The participation of indirect electrochemical processes is apparent since the decrease of absorbance at
587 nm proceeded even after the applied potential was removed (Figure S7). There would be two
plausible indirect electrochemical processes responsible for such an efficient electrochemical reaction.
One mechanism includes a chain reaction which takes place apart from the working electrode as already
demonstrated for oxidative cycloreversion of diarylethenes (Figure 4a).29,33 The radical cation 1b·+
generated by the electrode reaction of 1b spontaneously converts to more stable radical cation of open-
ring isomer 1a·+. Since the ionization potential of 1a was higher than that of 1b as demonstrated in CV
study (Table 1) by 0.65 eV, the radical cation 1a·+ has enough oxidation power to oxidize another
closed-ring isomer 1b to produce 1b·+ without the consumption of electricity. The kinetics of the
spontaneous conversions of 1b·+ to 1a·+ and 1a·+ to 1a, and the electron transfer from 1a·+ to 1b were
studied by the stopped-flow method using a chemical oxidant, which is described later. The other
ACS Paragon Plus Environment
10

Page 11 of 21
Journal of the American Chemical Society
reaction involves the local cell action, where the electrode mediates both anodic and cathodic reactions
of substances. That is, the oxidation of 1b and the reduction of 1a·+ which is formed from 1b·+ are
coupled through the electrode. By coupling the two oxidation/reduction processes, faradaic current
should be cancelled and apparently no electricity is consumed as illustrated in Figure 4b. This local cell
mechanism seems rather complicated and is difficult to characterize precisely. An experiment to verify
this model was demonstrated as follows (see Supporting Information in detail). The electrolysis of 1b
was conducted to accumulate the population of 1a·+ in a cell (Cell A). This process corresponds to
charging. The Cell A was connected to another cell (Cell B) containing an ITO (In-Sn-oxide) electrode
and the same solution with Cell A via a cannula filled with the electrolyte solution. The absorbance of
1b (at 587 nm) was monitored in Cell B. During the electrolysis in the Cell A, no absorbance change
was observed in the Cell B. After the electrolysis in the Cell A for 60 s, the potential of ITO-working
electrode in the Cell B was set to 0 V against to the ITO-electrode in the Cell A. Absorbance of 1b in
the Cell B simultaneously decreased and faradaic current was observed, which corresponds to the
discharging current. The charged substance (1a·+) in the Cell A during electrolysis for 60 s was
consumed for the oxidation of 1b in the Cell B via electrodes. In this experiment, the two working
electrodes WE1 and WE2 were set at the same electrochemical potential, which is thus the model of the
different parts of same working electrode. 1b was consumed without external current flow, indicating
local cell effect. The coulombic efficiency for this reaction as high as 150% was evaluated from the
amount of consumed 1b in the Cell B and faradaic electricity between WE1 and WE2.
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
Chemical oxidation. The oxidative conversion of 1b to 1a was also studied with chemical oxidant
instead of applying voltage. Tris-(p-bromophenyl)ammonium hexachloroantimonate (TBPA·+), which is
one of stable radical ionic oxidants, was used as a chemical oxidant to trigger the oxidative
cycloreversion reaction. Preliminary test by mixing of 1/10 equimolar amount of TBPA·+ to the solution
of 1b resulted in the complete bleach of colored solution and conversion of 1b to 1a was confirmed by
HPLC. This cycloreversion reaction of 1b with TBPA·+ involved neither any side reactions nor
ACS Paragon Plus Environment
11

Journal of the American Chemical Society
Page 12 of 21
decomposition of 1, which was confirmed by HPLC measurements after the oxidation reaction. The
oxidant-based reaction yield of 1000% at the room temperature was readily observed and suggested the
chain reaction mechanism.32a The detailed kinetics of the chemical oxidative-cycloreversion reaction
was monitored by means of the stopped-flow measurements. The stopped-flow method is a useful tool
to spectroscopically monitor the evolution of unstable intermediates involved in multistep reaction
processes. The time resolution of electron transfer stopped-flow method was reported to correspond to
high speed CV measurement with the scan rate higher than 1000 Vs-1.41
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 5.Stopped-flow absorption spectral change after a 1b solution (1.2 × 10-5 M) was mixed with a
TBPA·+ solution (1.2 × 10-5 M) in acetonitrile. Time intervals are 0.1 s for (a) and 2 s for (b). The
recording of spectra started 5 ms after the injection of both solutions and stopped after 10 s (a) and 120 s
(b).
We first studied the kinetics of sequential spontaneous reactions, the conversion of 1b·+ to 1a·+ (k2 in
Figure 4a) followed by 1a·+ to 1a (k3), where the equimolar amount of 1b and TBPA·+ were mixed
(Figure 5). As soon as the both solutions were mixed, the strong absorption band appeared at 682 nm
(Figure 5a), which was different from both of 1b (λmax = 587 nm) and TBPA·+ (λmax = 708 nm). Since
the difference of the oxidation potential between TBPA (0.97 V vs Fc/Fc+) and 1b (0.32 V vs Fc/Fc+)
was large, the electron transfer from 1b to TBPA·+ was readily completed within the mixing time (< 5
ACS Paragon Plus Environment
12

Page 13 of 21
Journal of the American Chemical Society
ms) to give 1b·+. During the waiting time of 5 ms, all neutral 1b and TBPA·+ were consumed to produce
1b·+. Therefore, the oxidation reaction rate constant (k1) of 1b by TBPA·+ was much larger than 200 s-1
(k1 >> 200 s-1). The absorption wavelength estimated by time-dependent DFT (TD-DFT) for 1b·+ (λmax
= 700 nm) agrees well with this experimental result (Figure S10). The decrease in the absorption band
of 1b·+ was accompanied with the increase at 510 nm with an apparent isosbestic point (Figure 5a),
which was assigned to the conversion of 1b·+ to 1a·+. Again, the TD-DFT calculation for 1a·+ also well
reproduced the experimental absorption band. By extending the monitoring period to 120 s, the
successive generation and decay processes of 1a·+ were recorded (Figure 5b).
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
The sequential reaction processes contains three components, 1b·+, 1a·+ and 1a. The mass balance
equations for these components are written as follows:
∙ꢄ
ꢁ
ꢅ
ꢀ ꢂꢃ
ꢁ
∙ꢋꢅ
ꢇ ꢈꢉꢊ ꢂꢃ (1)
ꢀꢆ
∙ꢄ
ꢀꢁꢂꢌ
ꢀꢆ
ꢅ
ꢁ
∙ꢋꢅ
ꢁ
∙ꢋꢅ
ꢇ ꢉꢊ ꢂꢃ ꢈ ꢉꢍ ꢂꢌ (2)
ꢀꢁꢂꢌꢅ ꢇ ꢉꢍꢁꢂꢌ∙ꢋꢅ ꢎ3)
ꢀꢆ
These equations lead to the equations for the evolutions of 1b·+ and 1a·+ as,
ꢂꢃ ꢇ ꢂꢃ ꢏꢐꢑꢒ ꢆ (4)
ꢓ
ꢁ
∙ꢋꢅ
ꢁ
∙ꢋꢅ
ꢒ
ꢓ
ꢓ
ꢒ ꢑꢒ
ꢔ
ꢔ
ꢁ
∙ꢋꢅ
ꢂꢌ ꢇ ꢂꢃ
ꢁ
∙ꢋꢅꢏ
ꢎ
ꢐ
ꢓ
ꢑꢒ ꢆꢕ
ꢑꢒ ꢆ ꢈ ꢐ
(5)
where [1b·+]0 is the initial concentration of 1b·+ which should be equal to the concentration of 1b
injected (1.2 × 10-5 M). Since 1b·+ and 1a·+ have non-zero absorbance at 510 nm, eq. (6) is regarded as,
ꢎ
ꢕ
ꢎ
∙ꢋꢕꢁ ∙ꢋꢅ
ꢂꢃ ꢗ ꢖ ꢂꢌ
ꢎ
∙ꢋꢕꢁ ∙ꢋꢅ
ꢂꢌ
A at 510 nm ꢇ ꢖ ꢂꢃ
(6)
where ε ε
(1b·+) and (1a·+) were molar absorption coefficients at 510 nm for 1b·+ and 1a·+, respectively.
By introducing eqs. (4) and (5) into (6), the temporal change of absorbance, A, at 510 nm is derived as
eq. (7).
ꢒ
ꢓ
∙ꢋ
∙ꢋ
ꢓ
ꢓ
ꢔ
ꢎ
ꢕ
ꢁ
∙ꢋꢅ
A at 510 nm ꢇ ꢖꢎꢂꢃ ꢕ ꢂꢃ ꢏꢐꢑꢒ ꢆ ꢗ ꢖꢎꢂꢌ ꢕ ꢂꢃ
ꢁ
∙ꢋꢅꢏ
ꢎ
ꢐ
ꢓ
ꢑꢒ ꢆꢕ
ꢑꢒ ꢆ ꢈ ꢐ
(7)
ꢒ ꢑꢒ
ꢔ
ACS Paragon Plus Environment
13

Journal of the American Chemical Society
Page 14 of 21
The fitting curve based on eq. (7) well traces the experimental data with a correlation coefficient of
1
2
3
0.999 with the assumed rate constants of k2=3.3±0.06 × 10-1 s-1 for and k3=3.0±0.009 × 10-2 s-1 together
4
5
6
with
ε
(1b·+) =9.4 × 103 cm-1M-1 for and
ε
(1a·+)=1.5 × 104 cm-1M-1 as shown in Figure 6. Difference in
7
8
-1
-1
the lifetimes of 1b·+ (k2 =3 s) and 1a·+ (k3 =33 s) gives rise to the apparent isosbestic points for the
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
early stage of sequential reaction (Figure 5a). It should be also noted the surprisingly long lifetime of
radical cation 1a·+ (
τ = 33 s) would be advantageous for the electron transfer in the chain reaction
(Figure 4a), which would be partly responsible for the high current efficiency of the electrolysis. It also
supports the results of dual-cell electrochemical experiments presented in Figures S8 and S9. Since k3
should correspond to reaction rate of reaction between 1a·+ and unknown reductant such as water, it
shall be sensitive to the experimental condition such as purity of solvents.
Figure 6. Temporal evolution of absorbance at 510 nm (circles) reproduced from Figure 5b and the
fitting curve based on eqn. 7 (solid line).
Finally, the kinetics of the electron transfer was studied under the condition that the chain reaction
takes dominant place. The varied amount of TBPA·+, corresponding to 2/3 eq., 1/2 eq., 1/3 eq. and 1/5
eq., were mixed with the definite quantity of 1b. When the solutions contain neutral 1b, the direct
photo-induced cycloreversion reaction of 1b with the rate constant of kL by the probe light for the
spectroscopy should be taken into account. We also confirmed that the radical cation state 1b·+ was
insensitive to light irradiation. Taking the reaction rate constants (k2–k4) depicted in Figure 4a as well as
ACS Paragon Plus Environment
14

Page 15 of 21
Journal of the American Chemical Society
the direct photochemical cycloreversion reaction constant kL into account, the evolutions of all
components, 1b, 1b·+, 1a·+ and 1a are written as follows:
1
2
3
4
5
ꢁ
ꢅ
ꢀ ꢂꢃ
ꢁ
ꢅꢁ ∙ꢋꢅ
ꢇ ꢈꢉꢘ ꢂꢃ ꢂꢌ ꢈ ꢉꢙꢁꢂꢃꢅ (8)
ꢀꢆ
6
7
8
∙ꢄ
ꢁ
ꢅ
ꢀ ꢂꢃ
ꢁ
∙ꢋꢅ
ꢇ ꢈꢉꢊ ꢂꢃ ꢗ ꢉꢘ ꢂꢃ ꢂꢌ (9)
ꢁ
ꢅꢁ ∙ꢋꢅ
9
ꢀꢆ
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
∙ꢄ
ꢁ
ꢅ
ꢀ ꢂꢌ
ꢁ
∙ꢋꢅ
ꢇ ꢉꢊ ꢂꢃ ꢈ ꢉꢍ ꢂꢌ ꢈ ꢉꢘ ꢂꢃ ꢂꢌ (10)
ꢁ
∙ꢋꢅ
ꢁ
ꢅꢁ ∙ꢋꢅ
ꢀꢆ
ꢁ
ꢅ
ꢀ ꢂꢌ
ꢁ
∙ꢋꢅ
ꢇ ꢉꢍ ꢂꢌ ꢗ ꢉꢘ ꢂꢃ ꢂꢌ ꢗ ꢉꢙꢁꢂꢃꢅ (11)
ꢁ
ꢅꢁ ∙ꢋꢅ
ꢀꢆ
The values obtained in the previously described experiment were applied for k2 and k3. The rate constant
of photo-cycloreversion reaction was separately determined in the absence of TBPA·+ to be 1.6 × 10-1 s-1
(Figure S11). The initial concentrations right after the mixing were zero for [1a] and [1a·+], whereas
those for [1b] and [1b·+] were estimated from the initial values of absorbance at 510 nm and 587 nm,
where the molar absorption coefficients for 1b and 1b·+ are known. The different amount of oxidant
gave different initial concentrations of [1b] and [1b·+]. By changing the second order rate constant k4 as
a parameter, the temporal change of concentrations of each component were simulated based on eqs.
(8)-(11). As a typical example, the temporal development of each component is illustrated in Figure 7a
with an assumed k4 value of 2 × 104 s-1M-1 for the condition of the addition 1/3 equimolar amount of
TBPA·+. These curves reconstructed the change in absorbance at 510 nm as follows:
ꢎ
ꢕ
ꢎ
ꢕꢁ
ꢅ
ꢎ
∙ꢋꢕꢁ ∙ꢋꢅ
ꢎ
∙ꢋꢕꢁ ∙ꢋꢅ
ꢂꢌ (12)
A at 510 nm ꢇ ꢖ ꢂꢃ ꢂꢃ ꢗ ꢖ ꢂꢃ ꢂꢃ ꢗ ꢖ ꢂꢌ
ACS Paragon Plus Environment
15

Journal of the American Chemical Society
Page 16 of 21
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 7. (a) Simulations of the evolution of each component during the cycloreversion reaction
triggered by the addition of TBPA·+ (1/3 equimolar amount). (b) Temporal change of absorbance at 510
nm together with the fitting curve (broken line) reproduced using the simulated values in (a) and molar
absorption coefficients for these components.
For this reconstruction, the values of 9.4 × 103 and 1.5 × 104 cm-1M-1 were used for
ε
(1b·+) and
ε
(1a·+),
respectively, which were obtained from the fitting in Figure 6, and 6.4 × 103 cm-1M-1 for
ε(1b). As
shown in Figure 7b, the reconstructed curve well accorded with the experimental plot. The k4 value of 2
× 104 s-1M-1 reasonably reproduced the experimental results for other conditions (Figures S12–S15).
Conclusion
We have investigated the electrochemical response of photochromic terarylenes. 4,5-
dithiazolylthiazole 1 demonstrated highly efficient oxidative cycloreversion reaction with the net current
efficiency as high as 900% and even more than 2000% provisionally. The two indirect electrochemical
processes, the chain reaction and the local cell reaction mechanisms, were considered responsible for
such a high current efficiency. Electron transfer stopped-flow measurements with the aid of a chemical
ACS Paragon Plus Environment
16

Page 17 of 21
Journal of the American Chemical Society
oxidant provided the direct evidence that two radical cation intermediates were involved in the oxidative
cycloreversion process; the closed-ring isomer was oxidized to radical cation which spontaneously
converted with τ = 3 sec to the radical cation of open-ring isomer with a long lifetime more than 30 s.
The variations of aryl units including polycyclic aromatics as well as the direction of their connections
to construct a terarylene structure would give a variety of electrochemical responses including
bidirectional reactions.35b Photo and electro dual-switching molecules which show efficient
photochemical cyclization and electrochemical cycloreversion reactions would serve as a versatile
material for future molecular devices.
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
Experimental Section
Materials and methods. Acetonitrile (dry grade) and TBPA·+ were purchased from Wako and
Aldrich, respectively, and used as received. Tetramethylammonium hexafluorophosphate (TMAPF6)
was purchased from Alfa Aesar and recrystallized from ethanol solution prior to use. Compounds 1–5
were synthesized according to the reported procedures.11a,14,19a,c UV–vis absorption spectra were
obtained on a JASCO V-670 spectrophotometer. HPLC was performed on a HITACHI LaChrom ELITE
HPLC system. Cyclic voltammetry (CV) measurements were carried out by using ꢁ-AUTOLAB
TYPE3 potentiostat/galvanostat equipped with electrochemical analysis system software. Stopped-flow
measurements were conducted with a rapid-scan stopped-flow spectroscopic system (Unisoku). DFT
calculations were performed with Gaussian 0342 at a B3LYP/6-31+G* for neutral compounds and at a
UB3LYP/6-31+G* level for radical cations.
Electrochemical measurements. CV measurements were conducted in a solution of 0.1 M TMAPF6
in dry acetonitrile with a scan rate of 50 mV/s at room temperature in an argon-filled glove box. A Pt
disk electrode and a Pt wire were used as a working and a counter electrodes, respectively. An Ag/Ag+
electrode or an Ag wire were used as reference electrodes, which were normalized with the half-wave
ACS Paragon Plus Environment
17

Journal of the American Chemical Society
Page 18 of 21
potential of ferrocene/ferrocenium+ (Fc/Fc+) redox couple. Typically, the concentration of
dithiazolylarylenes was adjusted to be 1 mM for CV measurements. Constant potential spectro-
electrochemical experiments were performed using a quartz glass cell with an indium tin oxide (ITO)-
coated quartz glass plate as a working and an Ag wire as a reference electrode. A separated sample
solution with a Pt wire counter electrode was connected with the quartz glass cell by a cannula, which
was filled with sample solution. The applied potential was fixed to 0.42 V vs Fc/Fc+ for electrolysis
condition. The oxidative cycloreversion process was monitored by UV-vis absorption spectroscopy.
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
Reaction kinetics study. Oxidative cycloreversion reaction kinetics with an oxidant was studied at
room temperature using stopped-flow measurements. The solutions of closed-ring isomer and oxidant
TBPA·+ were charged to reservoirs separately and mixed in equal amount in a reaction cell. Dynamic
transformation of absorption spectra associated with the chemical reactions in the cell was recorded with
the minimized time interval of 1.0 ms from 5 ms after mixing two solutions with a photodiode array
detector. The temporal evolutions of absorbance at wavelength specific for reaction components were fit
with reaction models to determine the rate constant of each reaction process.
Acknowledgement
This study was partly supported by NAIST, under the projects for “Bio-oriented Photoreactive
Molecules” and “Green Photonics Research”.
Supporting Information
Supplementary DFT calculations, UV-vis absorption spectral change during electrolysis, experiment for
indirect electrochemical processes, photo-induced cycloreversion reaction kinetics of 1b, and stopped-
flow study with varied 1b-TBPA·+ ratios. This material is available free of charge via the Internet at
http://pubs.acs.org.
ACS Paragon Plus Environment
18

Page 19 of 21
Journal of the American Chemical Society
References
1
2
3
(1) (a) Molecular Switches; Feringa, B. L. Eds.; Wiley-VCH, Weinheim, 2001; (b) Special Issue on
Photochromism: Irie, M.
4
5
(2) Photochromism:Memories and Switches: Irie, M. Chem. Rev. 2000, 100, 1683.
(3) Willner, I. Acc. Chem. Res. 1997, 30, 347 – 356.
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
(4) Yildiz, I.; Deniz, E.; Raymo, F. M. Chem. Soc. Rev. 2009, 38, 1859–1867.
(5) Stoll, R. S.; Hecht, S. Angew. Chem. Int. Ed. 2010, 49, 5054 – 5075.
(6) Gorostiza, P.; Isacoff, E. Y. Science 2008, 322, 395 –399.
(7) Yokoyama, Y. Chem. Rev. 2000, 100, 1717–1739.
(8) Mitchell, R. H. Eur. J. Org. Chem. 1999, 2695–2703.
(9) Irie, M. Chem. Rev. 2000, 100, 1685–1716.
(10) Kawai, T.; Iseda, T.; Irie, M. Chem. Commun. 2004, 72–73.
(11) (a) Nakashima, T.; Atsumi, K.; Kawai, S.; Nakagawa, T.; Hasegawa, Y.; Kawai, T. Eur. J. Org.
Chem. 2007, 3212–3218. (b) Kawai, S.; Nakashima, T.; Atsumi, K.; Sakai, T.; Harigai, M.;
Imamoto, Y.; Kamikubo, H.; Kataoka, M. Chem. Mater. 2007, 19, 3479–3483.
(12) Nakashima, T.; Goto, M.; Kawai, S.; Kawai, T. J. Am. Chem. Soc. 2008, 130, 14570-14575.
(13) (a) Neilson, B. M.; Lynch, V. M.; Bielawski, C. W. Angew. Chem. Int. Ed. 2011, 50, 10322–
10326. (b) Neilson, B. M.; Bielawski, C. W. J. Am. Chem. Soc. 2012, 134, 12693-12699.
(14) Taguchi, M.; Nakagawa, T.; Nakashima, T.; Kawai, T. J. Mater. Chem. 2011, 21, 17425–17432.
(15) Suzuki, K.; Ubukata, T.; Yokoyama, Y. Chem. Commun. 2012, 48, 765–767.
(16) Pang, S. C.; Hyun, H.; Lee, S.; Jang, D.; Lee, M. J.; Kang, S. H.; Ahn, K. H. Chem. Commun.
2012, 48, 3745-3747.
(17) (a) Nakagawa, T.; Atsumi, K.; Nakashima, T.; Hasegawa, Y.; Kawai, T. Chem. Lett. 2007, 36,
372–373. (b) Nakagawa, T.; Hasegawa, Y.; Kawai, T. J. Phys. Chem. A 2008, 112, 5096–5103.
(18) Yam, V. W. W.; Lee, J. K. W.; Ko, C. C.; Zhu, N. Y. J. Am. Chem. Soc. 2009, 131, 912–913.
(19) (a) Kawai, S.; Nakashima, T.; Kutsunugi, Y.; Nakagawa, H.; Nakano. H.; Kawai, T. J. Mater.
Chem. 2009, 19, 3606–3611. (b) Morinaka, K.; Ubukata, T.; Yokoyama, Y. Org. Lett. 2009, 11,
3890–3893. (c) Fukumoto, S.; Nakashima, T.; Kawai, T. Angew. Chem. Int. Ed. 2011, 50, 1565–
1568.
(20) (a) Nakashima, T.; Fujii, R.; Kawai, T. Chem. Eur. J. 2011, 17, 10951–10957. (b) Wu, Y.; Chen,
S. J.; Yang, Y. H.; Zhang, Q.; Xie, Y. S.; Tian, H.; Zhu, W. H. Chem. Commun. 2012, 48, 528-530.
(21) Boggio-Pasqua, M.; Ravaglia, M.; Bearpark, M. J.; Garavelli, M.; Robb, M. A. J. Phys. Chem. A
2003, 107, 11139–11152.
(22) Ishibashi, Y.; Umesato, T.; Kobatake, S.; Irie, M.; Miyasaka, H. J. Phys. Chem. C 2012, 116,
4862–4869.
ACS Paragon Plus Environment
19

Journal of the American Chemical Society
Page 20 of 21
(23) (a) Miyasaka, H.; Murakami, M.; Itaya, A.; Guillaumont, D.; Nakamura, S.; Irie, M. J. Am.
Chem. Soc. 2001, 123, 753–754. (b) Murakami, M.; Miyasaka, H.; Okada, T.; Kobatake, S.; Irie, M.
J. Am. Chem. Soc. 2004, 126, 14764–14772.
1
2
3
4
5
(24) Tsuboi, Y.; Shimizu, R.; Shoji, T.; Kitamura, N. J. Am. Chem. Soc. 2009, 131, 12623-12627.
6
7
8
9
(25) (a) Kobatake, S.; Yamashita, I. Tetrahedron 2008, 64, 7611–7618. (b) Morimitsu, K.; Shibata,
K.; Kobatake, S.; Irie, M. J. Org. Chem. 2002, 67, 4574–4578.
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
(26) Mialane, P.; Zhang, G. J.; Mbomekalle, I. M.; Yu, P.; Compain, J. D.; Dolbecq, A.; Marrot, J.;
Secheresse, F.; Keita, B.; Nadjo, L. Chem. Eur. J. 2010, 16, 5572-5576.
(27) Fox, M. A.; Hurst, J. R. J. Am. Chem. Soc. 1984, 106, 7626–7627.
(28) Mitchell, R. H.; Brkic, Z.; Sauro, V. A.; Berg, D. J. J. Am. Chem. Soc. 2003, 125, 7581–7585.
(29) (a) Koshido, T.; Kawai, T.; Yoshino, K. J. Phys. Chem. 1995, 99, 6110–6114. (b) Koshido, T.;
Kawai, T.; Yoshino, K. Chem. Express., 8, 1993, 553-556.
(30)
(a) Browne, W. R.; de Jong, J. J. D.; Kudernac, T.; Walko, M.; Lucas, L. N.; Uchida, K.; van
Esch, J. H.; Feringa, B. L. Chem. Eur. J. 2005, 11, 6414–6429. (b) Browne, W. R.; de Jong, J. J. D.;
Kudernac, T.; Walko, M.; Lucas, L. N.; Uchida, K.; van Esch, J. H.; Feringa, B. L. Chem. Eur. J.
2005, 11, 6430–6441.
(31) Guirado, G.; Coudret, C.; Hliwa, M.; Launay, J. P. J. Phys. Chem. B 2005, 109, 17445–17459.
(32) (a) Peters, A.; Branda, N. R. J. Am. Chem. Soc. 2003, 125, 3404–3405. (b) Peters, A.; Branda, N.
R. Chem. Commun. 2003, 954-955.
(33) Moriyama, Y.; Matsuda, K.; Tanifuji, N.; Irie, S.; Irie, M. Org. Lett. 2005, 7, 3315–3318.
(34) Staykov, A.; Areephong, J.; Browne, W. R.; Feringa, B.; Yoshizawa, K. ACS Nano 2011, 5,
1165–1178.
(35) (a) Gorodetsky, B.; Samachetty, H. D.; Donkers, R. L.; Workentin, M. S.; Branda, N. R. Angew.
Chem. Int. Ed. 2004, 43, 2812–2815. (b) Gorodetsky, B.; Branda, N. R. Adv. Funct. Mater. 2007, 17,
786–796. (c) Léaustic, A.; Anxolabéhère-Mallart, E.; Maurel, F.; Midelton, S.; Guillot, R.; Métivier,
R.; Nakatani, K.; Yu, P. Chem. Eur. J. 2011, 17, 2246–2255.
(36) Lee, S.; You, Y.; Ohkubo, K.; Fukuzumi, S.; Nam, W. Org. Lett. 2012, 14, 2238-2241.
(37) Guirado, G.; Coudret, C.; Launay, J. P. J. Phys. Chem. C 2007, 111, 2770–2776.
(38) (a) Tanaka, Y.; Ishisaka, T.; Inagaki, A.; Koike, T.; Lapinte, C.; Akita, M. Chem. Eur. J. 2010,
16, 4762–4776. (b) Akita, M. Organometallics 2011, 30, 43–51.
(39) Andrieux, C. P.; Hapiot, P.; Saveant, J. M. Chem. Rev. 1990, 90, 723–738.
(40) Oyama, M.; Higuchi, T.; Okazaki, S. Electrochem. Commun. 2000, 2, 675–678.
(41) Oyama, M.; Matsui, J.; Park, H. Chem. Commun. 2002, 604-605.
(42) Gaussian 03, revision D.01, Gaussian, Inc., Wallingford CT, 2004.
ACS Paragon Plus Environment
20

Page 21 of 21
Journal of the American Chemical Society
Figure for Table of Contents
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
ACS Paragon Plus Environment
21