Visible Light and Hydroxynaphthylbenzimidazoline Promoted Transition-Metal-Catalyst-Free Desulfonylation of N-Sulfonylamides and N-Sulfonylamines

Article abstract of DOI:10.1021/acs.joc.8b01536

A visible light promoted process for desulfonylation of N-sulfonylamides and -amines has been developed, in which 1,3-dimethyl-2-hydroxynaphthylbenzimidazoline (HONap-BIH) serves as a light absorbing, electron and hydrogen atom donor, and a household white light-emitting diode serves as a light source. The process transforms various N-sulfonylamide and -amine substrates to desulfonylated products in moderate to excellent yields. The observation that the fluorescence of 1-methyl-2-naphthoxy anion is efficiently quenched by the substrates suggests that the mechanism for the photoinduced desulfonylation reaction begins with photoexcitation of the naphthoxide chromophore in HONap-BIH, which generates an excited species via intramolecular proton transfer between the HONap and BIH moieties. This process triggers single electron transfer to the substrate, which promotes loss of the sulfonyl group to form the free amide or amine. The results of studies employing radical probe substrates as well as DFT calculations suggest that selective nitrogen-sulfur bond cleavage of the substrate radical anion generates either a pair of an amide or amine anion and a sulfonyl radical or that of an amidyl or aminyl radical and sulfinate anion, depending on the nature of the N-substituent on the substrate. An intermolecular version of this protocol, in which 1-methyl-2-naphthol and 1,3-dimethyl-2-phenylbenzimidazoline are used concomitantly, was also examined.

Full text of DOI:10.1021/acs.joc.8b01536

Subscriber access provided by University of Winnipeg Library
Article
Visible light and hydroxynaphthylbenzimidazoline promoted transition-
metal-catalyst-free desulfonylation of N-sulfonylamides and N-sulfonylamines
Eietsu Hasegawa, Yuto Nagakura, Norihiro Izumiya, Keisuke Matsumoto, Tsukasa
Tanaka, Tomoaki Miura, Tadaaki Ikoma, Hajime Iwamoto, and Kan Wakamatsu
J. Org. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.joc.8b01536 • Publication Date (Web): 17 Jul 2018
Downloaded from http://pubs.acs.org on July 17, 2018
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 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 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.
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 53
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Visible light and hydroxynaphthylbenzimidazoline promoted
transition-metal-catalyst-free desulfonylation of N-sulfonylamides and
N-sulfonylamines
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
†
†
†
†
†
Eietsu Hasegawa, * Yuto Nagakura, Norihiro Izumiya, Keisuke Matsumoto, Tsukasa Tanaka,
†
†‡
†
§
Tomoaki Miura, Tadaaki Ikoma, Hajime, Iwamoto, Kan Wakamatsu
†
Department of Chemistry, Faculty of Science, Niigata University, 8050 Ikarashi-2, Nishi-ku,
Niigata 950-2181, Japan.
‡
Center for Coordination of Research Facilities, Niigata University, 8050 Ikarashi-2, Nishi-ku,
Niigata 950-2181, Japan.
§
Department of Chemistry, Faculty of Science, Okayama University of Science, 1-1 Ridaicho,
Kita-ku, Okayama 700-0005, Japan.
1
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 2 of 53
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ABSTRACT: A visible light promoted process for desulfonylation of N-sulfonylamides and
ꢀamines has been developed, in which 1,3ꢀdimethylꢀ2ꢀhydroxynaphthylbenzimidazoline
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(HONapꢀBIH) serves as a light absorbing, electron and hydrogen atom donor, and a household
white lightꢀemitting diode serves as a light source. The process transforms various N-sulfonylamide
and ꢀamine substrates to desulfonylated products in modest to excellent yields. The observation that
the fluorescence of 1ꢀmethylꢀ2ꢀnaphthoxy anion is efficiently quenched by the substrates suggests
that the mechanism for the photoinduced desulfonylation reaction begins with photoexcitation of
the naphthoxide chromophore in HONapꢀBIH, which generates an excited species via
intramolecular proton transfer between the HONap and BIH moieties. This process triggers single
electron transfer to the substrate, which promotes loss of the sulfonyl group to form the free amide
or amine. Results of studies employing radical probe substrates as well as DFT calculations suggest
that selective nitrogenꢀsulfur bond cleavage of the substrate radical anion generates either a pair of
an amide or amine anion and sulfonyl radical or that of an amidyl or aminyl radical and sulfinate
anion, depending on the nature of the N-substituent on the substrate. An intermolecular version of
this protocol, in which 1ꢀmethylꢀ2ꢀnaphthol and 1,3ꢀdimethylꢀ2ꢀphenylbenzimidazoline are used
concomitantly, was also examined.
2
ACS Paragon Plus Environment

Page 3 of 53
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
INTRODUCTION
Single electron transfer (SET) to LUMOs or from HOMOs of neutral organic molecules generates
radical ions, in which one or more covalent bonds are weakened. As a result, various types of bond
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
cleavage reactions of organic radical ions are promoted by SET. Some of these processes, which
produce free radical and ionic intermediates, take place efficiently and selectively. As a result, they
2
are valuable for the design of synthetically useful reactions. Because the sulfonyl group is useful
for protecting nitrogen containing compounds such as amines and amides, the development of
efficient, selective and mild deprotection (desulfonylation) protocols is an important goal in
3
synthetic organic chemistry. Consequently, various methods to promote desulfonylation reactions
4
of N-sulfonylamines and ꢀamides, which follow SET induced routes, have been described. The
most common approaches of this type include reduction reactions promoted by magnesium and
5
6
7
8
alkaline metals, samarium diiodide, low valent titanium, transition metal salts, organic electron
9
10
donors, and electrochemistry. Photoinduced electron transfer (PET) is another process that can be
11
employed to initiate reductive N-desulfonylation reactions. In seminal studies about three decades
ago, Yonemitsu et al uncovered a PET process for desulfonylation of sulfonylamines, which utilizes
organic electron donating photocatalysts such as methoxyarenes along with hydride donors such as
1
1a
NaBH
4
.
3
Padwa et al later described a direct UV irradiation protocol, in which NEt serves as both
1
1b
an electron and proton donor, and nBu
3
SnH as a hydrogen atom donor. Recently, Xiao et al
3
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 4 of 53
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
demonstrated that a process, utilizing an iridium complex as a visible light absorbing catalyst and
Hantzsch ester as both an electron and hydrogen atom donor, promotes desufonylation reactions of
1
1c
N-sulfonylbenzamides.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2ꢀArylꢀ1,3ꢀdimethylbenzimidazolines (ArꢀBIHs) are analogues of the reduced form (NADH)
+
of nicotinamide adenine dinucleotide (NAD ). As such, these substances serve as effective hydride,
12
hydrogen atom and electron donors.
Chikashita et al first recognized that
1,3ꢀdimethylꢀ2ꢀphenylbenzimidazoline (PhꢀBIH) donates a hydride ion to cationic intermediates,
13
generated by complexation of organic substrates with Lewis acids. Subsequently, Tanner et al
demonstrated that PhꢀBIH also participates in reactions as a sequential hydrogen atom and electron
14
donor. Based on this foundation, about two decades ago we initiated an investigation of reactions,
15
using PET and PhꢀBIH, that lead to reductive transformations of organic substances. Ensuing
studies by us and others demonstrated that PhꢀBIH and related substances do indeed act as effective
16,17
electron donating reductants in PET promoted reduction reactions.
Moreover, the redox activity
18
of ArꢀBIHs has been used advantageously in investigations related to hydrogen gas evolution,
19
20
21
22
2
artificial photosynthesis, solar cells, organic semiconductors, and SO activation.
Visible light promoted redoxꢀcatalyzed reactions have received recent attention in the area of
2
3
synthetic organic chemistry. Complexes of transition metals, such as ruthenium and iridium, have
been frequently used to promote these types of transformations. However, more economical as well
4
ACS Paragon Plus Environment

Page 5 of 53
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
as sustainable PET procedures, which do not use expensive transition metals, are also highly
attractive. Because of these features, our studies have focused on the design of visible light
promoted redox reactions that are analogous to those in which Ar–BIH serves as a reductant. In this
context, we developed a desulfonylation reaction of Nꢀsulfonylindoles, which utilizes substituted
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
6b
pyrenes as organic photocatalysts and ArꢀBIHs as reductants.
In recent investigations, we
designed Ar–BIH analogues, which contain arene or hydroxyarene chromophores connected to the
24
BIH framework to extend their light absorption profile into the visible light region. Thus,
investigations of PET promoted desulfonylation reactions of Nꢀtosylbenzamides independently
17a
11c
conducted by Yu and Xiao (Figure 1), particularly motivated us to apply the protocol utilizing
a visible light absorbing HOArꢀBIH to the desulfonyltaion reactions of their substrates and others.
In the study described below, we discovered that one substance of this type,
1,3ꢀdimethyꢀ2ꢀ(2ꢀhydroxynaphthyl)bennzimidazoline (HONapꢀBIH, 1a), serves as an effective
visible light absorbing reducing agent in desulfonylation reactions of Nꢀtosylbenzamides 2,
dibenzyl tosylamine 4, diphenyl tosylamine 6, Nꢀsulfonyl indols 8 to produce respective
desulfonylated compounds 3, 5, 7, and 9 by using a household 7.3 W white lightꢀemitting diode
(LED). The new protocol has the highly advantageous feature of not requiring a visible light
absorbing transition metal catalyst because HONapꢀBIH absorbs light in the visible region, and
5
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 6 of 53
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
HONapꢀBIH serves as a formal two hydrogen atom donor in a manner that is similar to that of
2
4a,24b
Hantzsch ester.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 1. PET promoted desufonylation reactions of sulfonamides.
RESULTS AND DISCUSSION
In the initial phase of this study, we assessed the viability of desulfonylation reactions of
N-sulfonylamides promoted by hydroxynaphthylbennzimidazoline 1a. For this purpose,
6
ACS Paragon Plus Environment

Page 7 of 53
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
photoreactions of NꢀbenzylꢀNꢀtosylbenzamide (2a) (0.1 M) in various solvents were carried out in
the presence of 1a with irradiation using a white LED for 1 h (Table 1, and also see Figure S1).
Under these conditions, the expected desulfonylation product, N-benzylbenzamide (3a) is formed in
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
excellent yields when DMSO, DMF and PhCH
3
are used as solvents (entries 1, 2, 3 and 9).
Although reactions in MeOH, MeCN and CH
2
Cl
2
do not proceed to completion, the yields of 3a
based on the conversion of 2a are high (entries 6, 7, and 8). While the yields of 3a decrease as the
quantity of 1a is reduced for reactions of 2a in DMF, an excellent massꢀbalance is maintained
25
(entries 4 and 5).
a
Table 1. Photodesulfonylation of N-benzyl-N-tosylbenzamide (2a) in various solvents
b
b,c
entry
1a (equiv)
1.5
solvent
DMSO
DMF
conv of 2a (%)
yield of 3a (%)
d
1
2
3
4
5
6
7
8
9
100
100
100
87
97 [99]
1.5
99
1.2
DMF
97
1.0
DMF
81 (93)
49 (100)
30 (88)
83 (95)
93 (99)
0.5
DMF
49
1.5
MeOH
MeCN
34
1.5
87
1.5
CH
2
Cl
2
94
1.5
PhCH
3
100
98
a
b
2a (0.10 mmol), 1a (0.5~1.5 equiv vs 2a), solvent (1.0 mL); 7.3 W white LED, 1 h. Determined
1
c
d
by using H NMR. The yields in the parenthesis are based on conversion of 2a. Yield of 3a
isolated by column separation.
7
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 8 of 53
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
While conducting experiments described above, we found that reaction of 2a in the presence of
1a proceeds to some extent in the absence of LED irradiation. To obtain information about this
phenomenon, the irradiation time was shortened from 1 h to 20 min, at which time
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2ꢀbromoacetophenone was added to prevent further reaction of 2a caused by the remained 1a with a
26
ambient light during work up. Also, the effect of base on the process was examined by including
24b,24c,27
1,8ꢀdiazabicyclo[5.4.0]undecꢀ7ꢀene (DBU) in the reaction mixture.
As can be seen by
viewing the results in Table 2, the extent of the reaction promoted by 20 min LED irradiation
(LEDꢀON) of a mixture of 1a and 2a is significantly greater than that occurring without irradiation
28
(LEDꢀoff) in all solvents employed. Moreover, the presence of DBU enhances the progress of the
reaction conducted using either dark or irradiation conditions (compare entries 6 and 8 to 5 and 7
respectively).
Table 2. The effect of LED irradiation on photodesulfonylation of 2a in the presence or
a
absence of DBU in various solvents
b
b,c
entry
LED
ON
off
DBU
solvent
DMSO
DMSO
DMF
conv of 2a (%)
yield of 3a (%)
80 (96)
8 (89)
1
2
3
4
ꢀ
ꢀ
ꢀ
ꢀ
83
9
ON
off
100
12
94
DMF
9 (75)
8
ACS Paragon Plus Environment

Page 9 of 53
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
5
6
7
8
9
1
ON
ON
off
ꢀ
CH
CH
CH
CH
2
2
2
2
Cl
Cl
Cl
Cl
2
2
2
2
42
91
12
29
100
13
41 (98)
84 (92)
11 (92)
27 (93)
99
added
ꢀ
off
added
ON
off
ꢀ
ꢀ
PhCH
PhCH
3
0
3
11 (85)
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
2a (0.10 mmol), 1a (1.5 equiv vs 2a), DBU (1.5 equiv vs 2a ), solvent (1.0 mL); 7.3 W white LED
b
c
1
or in the dark, 20 min. Determined by H NMR. Yields in the parenthesis are based on the
conversion of 2a.
Next, this protocol was applied to desulfonylation of sufonamides 2, which contain various N-
and carbonyl substituents with differing electron accepting and donating abilities (Table 3). The
results show that the extents of substrate conversion and yields of these reactions directly correlate
with the electron accepting abilities of the R
2
group in the carbonyl moiety (except for the
red
p-nitrophenyl substituted amide 2c, entry 3), which are estimated using reduction potentials (E 1/2
,
see Figure S4 and Table S2) of the corresponding N-tosylamides (see entries 1, 2, 4, 6, 8, 10).
Although substrates possessing p-methoxyphenyl (2d) as well as alkyl (2f) or nonconjugated
alkenyl (2g) carbonyl substituents require longer irradiation times for completion, all substrates
except for 2c react to form the corresponding desulfonylated amides in >90% yields. The behavior
of 2c is interesting in that, although it is consumed, it does not produce the expected desulfonylated
product 3c under the conditions used to smoothly desulfonylate other N-tosylamides (entry 3).
Analysis of the reversible cyclic voltammogram (CV) of 2c (Figure 2) suggests that its radical anion
9
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 10 of 53
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
formed by SET is more stable than those of related amides 2 and, thus, this intermediate might be
29
less prone to NꢀS bond cleavage (see below). Also it should be noted that alkene tethered
substrates 2f and 2g, which are designed to undergo intramolecular amidyl radical trap via fast
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
0
5
ꢀexo cyclization, did not produce the expected cyclized products (Figure 3). This observation
suggests that the amidyl radicals would not be produced by fragmentation of the radical anions of 2f
and 2g (see below).
a
Table 3. Photodesulfonylation of various N-tosylamides 2 in DMF
red
E
time
conv of 2 yield of
1/2
1
2
entry
2
R
R
b
b,c
(
V vs SCE) (h)
(%)
100
100
23
3 (%)
97
d
1
2a PhCH
2b PhCH
2c PhCH
2d PhCH
2
2
2
2
Ph
ꢀ1.58
1
1
1
1
4
1
3
1
2
1
4
24
2
3
4
5
6
7
8
9
p-ClC
6
H
4
ꢀ1.52
91
p-NO
2
C
6
H
4
ꢀ0.87. ꢀ1.16
ꢀ1.69
0
p-MeOC
6
H
4
32
31 (97)
98
100
54
2e CH
3
2
(CH
2
)
3
Ph
ꢀ1.64
ꢀ1.66
ꢀ1.94
40 (74)
95
100
70
2f
CH
=CH(CH
2
)
3
Ph
65 (93)
97
100
17
10
2g PhCH
2
CH
2
=CH(CH
2
)
2
12 (71)
30 (94)
90
11
32
12
100
a
2 (0.10 mmol), 1a (1.2 equiv vs 2, 1.5 equiv and 2.0 equiv for entries 11 and 12), DMF (1.0 mL);
b
1
c
7
.3 W white LED. Determined by H NMR. Yields in the parenthesis are based on the conversion
d
of 2a. Same as entry 3 of Table 1.
10
ACS Paragon Plus Environment

Page 11 of 53
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 2. Cyclic voltammograms of 2a and 2c in MeCN.
Figure 3. Expected radical cyclization of Nꢀcentered radicals derived from 2f and 2g.
A plausible mechanism for the photoinduced desulfonylation reaction, which is consistent with
the observations described above, is shown in Scheme 1. It is known that photoexcitation of
naphthols enhances their acidities and that deprotonation of their excited states produces excited
31
1
24a
3
states of naphthoxides. In addition, the observation that the H NMR spectra of 1a in CDCl ,
11
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 12 of 53
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
CD
3
CN and DMSOꢀd
6
contain respective peaks at 10.43, 10.54, and 9.95 ppm (see Experimental
Section), suggests the existence of an intramolecular hydrogen bond between the hydroxyl
32
substituent of the naphthol group and nitrogen of the benzimidazoline moiety. As a result, a
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
suitable conformation appears to exist in the ground state of 1a for intramolecular proton transfer
(intraꢀPT) between hydroxynaphthyl and benzimidazoline moieties. Because the photoexcited
31b
2ꢀnaphthol is sufficiently acidic to donate proton to an aromatic amine, it is plausible to suggest
that photoexcited 1a (1a*) undergoes rapid intramolecular proton transfer to give the excited state
species 10 containing a protonated benzimidazoline group and naphthoxide moiety. Subsequent
SET from 10 to the tosylbenzamide 2 produces the corresponding naphthoxy radical 11 and radical
•– 33
•–
anion 2 . Nitrogenꢀsulfur (NꢀS) bond cleavage in 2 then occurs to form a benzamide anion and
sulfonyl radical. This proposal is also consistent with the observation that substrates 2f and 2g did
not produce the expected amidyl radical cyclization products (see above and below). Finally proton
transfer and hydrogen atom from 11 to the respective benzamide anion and sulfonyl radical
fragments produce amide 3 and p-toluenesulphenic acid along with benzimidazolium naphthoxide
3
4
1
2. When present in the reaction medium, DBU deprotonates 1a generating bennzimidazoline
2
4b,24c
nahthoxide 13, which serves as the light absorbing species.
In this case, photoexcitation of 13
•–
forms 13*, which donates an electron to 2 giving bennzimidazoline nahthoxyl radical 14 and 2 ,
12
ACS Paragon Plus Environment

Page 13 of 53
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
•
–
which follow a reaction pathway that is similar to that of the 11 and 2 pair to ultimately produce 3
3
3
and p-toluenesulphenic acid along with benzimidazolium naphthoxide 12.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 1. Plausible mechanism for the photodesulfonylation reaction of 2 promoted by
photoexcitation of 1a
As described above, photoexcitation of naphthoxy moiety of 1a is the initial step in the
mechanistic route for the desufonylation reaction. Moreover, we have previously developed the
13
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 14 of 53
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
photosensitization system in which organic sensitizers such as substituted pyrenes and anthracenes
1
6b,16c
cooperate with 1,3ꢀdimethylꢀ2ꢀphenylbenzimidazoline (1b) and its derivatives.
Above
discussion and our previous effort lead to the question of whether or not a combination of
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1ꢀmethylꢀ2ꢀnaphthol (15) and 1b, an intermolecular photoꢀreductant (interꢀPR) analog of 1a, would
promote photoinduced desulfonylation reaction of 2 upon visible light irradiation (see Figure S2).
The results of experiments designed to answer this question (Table 4) show that the interꢀPR system
comprised of 15 and 1b effectively promotes desulfonylation of 2a in DMF to form 3a in excellent
yields that are comparable to those for the reaction using 1a (see entries 1 and 2). Moreover, the
3
interꢀPRꢀsystem promotes desulfonylation reaction of 2a in nonpolar solvent such as PhCH , albeit
less efficiently than that induced by 1a (compare entry 6 to 5). Notably, reaction of 2a does occur
using 15 in the absence of 1b while 1b alone does promote the reaction inefficiently in DMF but
3
not in PhCH (entries 3 and 7).
Table 4. Comparison of photodesulfonylation of 2a promoted by 1a with that induced by the
a
intermolecular photo-reductant (inter-PR) comprised of Ph-BIH (1b) and MeNaphOH (15)
b
b,c
entry
1
15
-
solvent
DMF
conv of 2a (%)
yield of 3a (%)
d
1
1a
100
97
14
ACS Paragon Plus Environment

Page 15 of 53
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
2
3
4
5
6
1b
1b
ꢀ
added
DMF
DMF
DMF
PhCH
PhCH
PhCH
100
37
3
98
ꢀ
35 (95)
added
0
e
1a
1b
1b
ꢀ
3
3
3
100
34
2
98
added
ꢀ
26 (76)
0
7
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
2
a (0.10 mmol), 1 (1.2 equiv vs 2a, 1.5 equiv of 1a for entry 5), 15 (1.2 equiv vs 2a), solvent (1.0
b
c
1
mL); 7.3 W white LED, 1 h. Determined by using H NMR. Yields in the parenthesis are based
d
e
on the conversion of 2a. Same as entry 3 of Table 1. Same as entry 9 of Table 3.
A possibility exists that the high reactivity of N-tosylamides 2 in photoinduced desulfonylation
reaction could be a consequence of the presence of an electron withdrawing aroyl substituents. To
investigate this issue, we explored the applicability of the protocol to desulfonylation reactions of
N-tosylꢀN,Nꢀdibenzylamine (4) which is a much weaker electron acceptor than the amide analog 2a
red
as demonstrated by its large reduction potential (E 1/2 = ꢀ2.39 V vs SCE). Photoreactions were
carried on mixtures of 1a and 4 in various solvents by utilizing not only LED but also Xeꢀlamp
irradiation. LED irradiation of a mixture of 1a and 4 in DMSO for 24 h does not lead to formation
of N,Nꢀdibenzylamine (5) (Table 5, entry 1). In contrast, irradiation of a mixture that also contains
24c
tꢀBuOK to activate 1a, leads to desulfonylation of 4 to give 5 although only in a modest yield
(entry 2). Irradiation using a Xe lamp for a shorter 6 h time period also promotes reaction of 4 to
give 5 in slightly higher yield (entry 7). Interestingly, the results of a solvent optimization study
3
(entries 5, 6 and 11–14) show that photoreaction in a 1:1 mixture of DMSO and PhCH occurs with
1
5
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 16 of 53
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
the highest conversion of 4 and yield of 5. Another notable observation is that the presence of
1,4ꢀcyclohexadiene, while decelerating the process, leads to a significant enhancement in the yield
of 5 (compare entries 3 and 9 to 1 and 8 respectively). On the other hand, styrene does not greatly
influence the conversion of 4 as well as the yield of 5 (entries 4 and 10). Finally, the
interꢀPRꢀsystem comprised of 1b and 15 does not promote desulfonylation of 4 (results not shown
in Table 5).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
Table 5. Photodesulfonylation of N,N-dibenzyl-N-tosylamine (4)
entry tꢀBuOK Additive
solvent
lamp time
(h)
conv of 4 yield of
b
b,c
(equiv)
(equiv)
(%)
11
5 (%)
1
2
3
4
5
6
7
8
9
1
ꢀ
-
DMSO
DMSO
DMSO
LED 24
0
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
2.0
2.0
1.5
ꢀ
LED 24
84
47 (56)
53 (95)
53 (60)
73
CHD (5.3)
LED 24
56
styrene (3.0) DMSO
LED 24
89
-
ꢀ
ꢀ
DMSO/PhCH
PhCH
3
(1/1) LED 24
LED 24
100
56
3
45 (80)
58 (77)
56
DMSO
Xe
(1/1) Xe
(1/1) Xe
(1/1) Xe
(1/4) Xe
(1/4) Xe
6
6
6
6
6
6
6
6
75
DMSO/PhCH
DMSO/PhCH
3
3
3
3
3
100
66
CHD (5.3)
59 (89)
69 (81)
62 (97)
81
0
styrene (1.0) DMSO/PhCH
DMSO/PhCH
85
11
64
12
DMSO/PhCH
100
45
13
DMSO/PhCH (1/9) Xe
44 (98)
10 (38)
3
14
PhCH3
Xe
26
16
ACS Paragon Plus Environment

Page 17 of 53
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
a
4 (0.10 mmol), 1a (1.5 equiv vs 4), tꢀBuOK (1.5–2.0 equiv vs 4), additive: 1,4ꢀcyclohexadiene
(
b
CHD) and styrene, solvent (1.0 mL); 7.3 W white LED and and 500 W Xeꢀlamp (λ > 390 nm).
1
c
Determined by using H NMR. Yields in the parenthesis are based on the conversion of 4.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Although it is difficult to explain some of these observations, the fact that the presence of
3
hydrogen atom donors such as CHD and PhCH lead to increased yields of 5 does suggest that
radical intermediates are involved in the process. Then, in order to gain the information of the
expected radical intermediate, we examined the reaction of alkene tethered substrate 16. However,
the expected radical cyclization product was not formed while the deprotected amine 17 was
obtained in modest yield as a major product (equation 1).
35
Since 5ꢀexo cyclization of aminyl radical is much slower process than that of amidyl radical, the
aminyl radical derived from 16, even if it is formed, would undergo a competitive hydrogen atom
abstraction from a appropriate donor, for example 14 in Scheme 1 (Scheme 2).
17
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 18 of 53
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Scheme 2. Competitive radical pathways of aminyl radical derived from 16
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The visible light promoted process, using 1a or the interꢀPRꢀsystem analog comprised of 15
red
and 1b, is applicable to the desulfonylation reaction of N,Nꢀdiphenylsulfonylamine 6 (E 1/2 = ꢀ2.07
red
V vs SCE) and the Nꢀsulfonylindoles 8a–8c (E 1/2 = ꢀ1.67 V, ꢀ1.74 V, and ꢀ1.92 V vs SCE for 8a,
8b, and 8c, respectively). The results (Table 6) demonstrate that 24 h irradiation of DMSO solution
of 6 containing 1a produces the secondary amine 7 in a good yield and that presence of DBU in the
reaction mixture enhances the efficiency of the process enabling 7 to be formed in high yield after
only a 2 h irradiation period (entries 2ꢀ6). Likewise, photodesulfonylation of 8 using 1a produces
the corresponding deprotected indoles 9 in good to excellent yields (entries 7, 8, and 9). The
interꢀPRꢀsystem of 1b and 15 was found to promote formation of 9 from 8 in modest to good yields
(entries 10ꢀ 14).
18
ACS Paragon Plus Environment

Page 19 of 53
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Table 6. Photodesulfonylation reactions of 6 and 8 by using 1a and the inter-PR system
comprised of 1b and 15
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
entry substrate 1
15
ꢀ
DBU solvent time (h) conv (%) product yield (%)
1
2
6
1a
1a
1a
1b
1b
ꢀ
DMSO
2
2
38
7
32 (84)
66 (90)
79 (94)
28 (97)
32 (73)
79
6
ꢀ
added DMSO
73
7
3
6
ꢀ
DMSO 24
84
7
4
6
ꢀ
ꢀ
DMSO
2
2
29
7
5
6
ꢀ
added DMSO
44
7
6
6
1b added
DMSO 24
100
100
100
100
100
15
7
d
7
8a
8b
8c
8a
8a
8a
8b
8c
1a
1a
1a
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
3
3
3
3
3
3
3
3
9a
9a
9c
9a
9a
9a
9a
9c
97 [99]
98
8
9
83
10
1b added ꢀ
90
1
1
1
1
2
3
4
1b
-
ꢀ
ꢀ
0
added
5
0
1b added ꢀ
1b added ꢀ
100
79
76
1
55 (70)
a
6 or 8 (0.10 mmol), 1 (1.2 equiv vs substrate), 15 (1.2 equiv vs substrate), DBU (1.2 equiv vs 6),
b
c
1
solvent (1.0 mL); 7.3 W white LED. Determined by using H NMR. Yields in the parenthesis are
d
based on the conversion of 6 or 8. Yield of 9a isolated by column separation.
A key step in the proposed mechanism for the visible light promoted desulfonylation reaction
of N-sulfonlyamides and ꢀamines involves SET. The observation of quenching of emission
(fluorescence) from a photoexcited substance is a general method employed to determine if and
when it interacts with a substrate through SET. In the proposed mechanistic route (Scheme 1),
19
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 20 of 53
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
photoexcitation of 1a or its corresponding anion 13 produces the excited naphthoxide species,
which donates and electron to the sulfonamide substrate. Unfortunately, the use of emission
spectroscopy to gain evidence for the occurrence of this SET process is blocked by the fact that the
excited state of 1a is nonꢀemissive. However, we observed that 1ꢀmethylꢀ2ꢀnaphthoxy anion,
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
36
derived from the corresponding naphthol 15, emits. Specifically, a solution of DBU and 15 in
DMSO emits light with a wavelength maximum at 469 nm when excited at 400 nm, a wavelength at
which 15 does not absorb light (Figure S2 and Figure S6). As a result, fluorescence quenching
experiments were carried out using the DBU and 15 system and the representative sulfonylamides
and ꢀamines 2a, 4, 6, and 8a as quenchers (Figure S7). As expected based on the reduction
potentials of the quenchers (see above), fluorescence quenching occurs with efficiencies (k
q
) that
9
are directly dependent on the electron accepting abilities of N-sulfonylamides and ꢀamine (10 k
q
ꢀ1 ꢀ1
37
（
M s ） = 24.9 for 2a, 12.8 for 8a, 4.34 for 6, 3.28 for 4). The results suggest that the excited
napthoxide species formed from 1a participates in SET to the substrates in the photoreactions.
Radical anions derived from N-sulfonylamides and ꢀamines are proposed to be intermediates in
the visible light promoted desulfonylation process. N–S bond cleavage in these radical anions can
generate either Nꢀcentered anions or ꢀradicals in a manner that should be dependent on the nature of
6b,9b,10a,10c,11,16b,17a
the N-substituent.
Accordingly, radical anions derived from substrates with
electron withdrawing (EWG) N-substituents (eg., 2) should prefer fragmentation to give amide
20
ACS Paragon Plus Environment

Page 21 of 53
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
–
•
anions (R
1
R
2
N in Scheme 3) and sulfonyl radicals (RSO
2
), while those containing electron
•
donating alkyl substituents (eg., 4 and 6) should produce aminyl radicals (R
1
R
2
N ) and sulfinate
–
2
anions (RSO
). On the other hand, radical anions of 8 are expected to react by either route
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
depending on the nature of the C substituent.
Scheme 3. Hypothetical N–S bond cleavage reactions of N-sulfoyl-amide and -amine radical
anions.
In order to gain additional information about the electronic structures as well as the nature of
NꢀS bond cleavage in the N-sulfonylamide and ꢀamine radical anions, DFT calculations using
B3LYP/6ꢀ31+G(d) were conducted on neutral forms as well as radical anions of representative
38a
substrates. The LUMO electron distributions of 2a, 2c, 4, 6, 8a, and 8c are shown in Figure 4.
While LUMO coefficients exist only on nitrobenzoyl moiety of 2c, those of 2a, 4, 6, 8a and 8c are
located on the tosyl moieties and N–S bonds. Although several conformers with close energies exist
for all N-sulfonylamides and ꢀamines (see the cases of 2a and 2c in Table S5 and Table S7,
21
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 22 of 53
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
respectively), the LUMO distributions of these species are nearly identical for all conformers. The
geometry optimized radical anions, calculated starting with the initial geometries of the
corresponding neutral molecules, have structures containing elongated N–S bonds (see the case of
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2a in Table S6), which suggests that these bonds are significantly weakened. Another local energy
minimum conformer of the radical anion of 2a contains a normal N–S bond (ca. 1.7 Å) , but the
–1
barrier for its N–S cleavage was calculated to be small (< 2.82 kcal mol , see Table S9 and Figure
S8). This finding suggests that rapid dissociation to form two fragments takes place prior to a
conformational change in all cases except the radical anion of 2c (see below).
The optimized global minimum energy structures of the radical anions are given in Figure 5
(also see Table S10). The N–S atomic distances of radical anions of 2a (2.722 Å) and 4 (4.665 Å)
are significantly longer than those of their neutral forms, 2a (1.730 Å) and 4 (1.675 Å). The global
minimum structure of the radical anion of 2c has an elongated N–S bond (Figure 5) However,
careful optimization, using its neutral form (1.736 Å, conformer A in Table S7) as the initial
geometry, affords a structure having a slightly shortened N–S bond (1.699 Å, conformer A in Table
–1
S8) and an activation energy of N–S cleavage (5.32–9.24 kcal mol ), which is much greater than
that of 2a (see Table S9 and Figure S8). The results are consistent with the observations that 2c is
less reactive compared to other N-sulfonylamides and ꢀamines (see Table 3) and its CV displays
quasiꢀreversible redox behavior (Figure 2).
22
ACS Paragon Plus Environment

Page 23 of 53
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
3
8b
Finally, the results of atomic charge and spin by natural population analysis
reveal that the
–
radical anion of 2a should undergo preferential fragmentation to give benzamide anionꢀ (BzBnN ,
•
fragment charge –0.62) and N-tosyl radical (TS , fragment spin 0.61) species while the aminyl
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
•
–
radicalꢀ (Bn N , fragment spin 0.97) and tosyl anion (TS , fragment charge –0.92) are generated
2
preferentially from the radical anion of 4 (details see Table S10 and Table S11). Also, the results
suggest that cleavage of the radical anion of the sulfonylindole, which lacks a methoxycarbonyl
substitutent, slightly favors formation of the aminyl radical (compare 8c to 8a).
2a
2c
4
6
8a
8c
Figure 4. LUMO distributions of 2a, 2c, 4, 6, 8a, and 8c.
23
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 24 of 53
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
a
2c
4
6
8a
8c
Figure 5. Optimized global minimum structures of radical anions of 2a, 2c, 4, 6, 8a, and 8c.
DFT calculation of the radical probe substrates 2f, 2g and 16 show that they have LUMO
distributions that are similar to those of 2a and 4 respectively (Figure 6). As a result, 2f and 2g
should undergo fragmentation in a similar manner to form amide anions and N-tosyl radicals
(
Figure 7, also see Table S10 and Table S11). This prediction is consistent with the reaction of 2f
and 2g because the amide anion derived from the radical anions is not expected to undergo
cyclization. Moreover, the calculation suggest that cyclization of the corresponding amidyl radical
•
‡
(
H C=CH(CH ) BzN ) would be energetically feasible (ꢀG = –6.75 kcal/mol, ꢀG = 9.70 kcal.mol)
2
2 3
24
ACS Paragon Plus Environment

Page 25 of 53
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
•
(
see Table S12). On the other hand, aminyl radical (H C=CH(CH ) BnN ) generated from radical
2 2 3
anion of 16 is less likely to cyclize because this process is slightly endergonic (ꢀG = 2.68 kcal/mol)
‡
and has a high activation energy (ꢀG = 16.32 kcal/mol) (see Table S12).
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2f
2g
16
Figure 6. LUMO distributions of 2f, 2g, and 16.
2
f
2g
16
Figure 7. Optimized structures of radical anions of 2f, 2g, and 16.
CONCLUSION
In the effort described above, we developed a visible light promoted desulfonylation process
utilizing 1,3ꢀdimethylꢀ2ꢀhydroxynaphthylbenzimidazoline (1a), which is applicable to a variety of
25
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 26 of 53
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
N-sulfonylamides and ꢀamines. Although only briefly explored,
a
combination of
1ꢀmethylꢀ2ꢀnaphthol (15) and 1,3ꢀdimethylꢀ2ꢀphenylbenzimidazoline (1b) was also found to
promote the photoꢀdesulfonylation reaction. In the initial step in the mechanism proposed for the
process promoted by 1a, photoꢀexcited 1a donates an electron to the substrates. Support for this
proposal came from the observation that the fluorescence of 2ꢀmethylꢀ1ꢀnaphthoxy anion derived
from 15 was efficiently quenched by representative sulfonylamide and ꢀamine substrates. These
observations suggest that an excited naphthoxide species generated from 1a acts as an electron
donor in the reaction. This proposal is supported by the wellꢀknown photoinduced deprotonation
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
31
reactions of 2ꢀnaphthol derivatives. DFT calculations provided information about the selectivity
of NꢀS bond cleavage in the radical anions of the substrates for this reaction, which is consistent
with the results obtained from studies of desulfonylation reactions of radical probe substrates.
Although several desulfonylation reactions that take place via PET induced pathways have been
11c
previously developed, they often require expensive transition metal photocatalysts, harmful UV
17a
11b
9b
irradiation, toxic hydrogen atom donors and/or extremely strong electron donors. Therefore,
the new visible light promoted, transition metal free protocol has several attractive features,
including that a preparation of 1a requires only short sequence, 1a is easily handled under air at
ambient temperature, and the process is activated by using a household white LED. Further
26
ACS Paragon Plus Environment

Page 27 of 53
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
investigations aimed at expanding the substrate tolerance and more fully understanding the reaction
mechanism are underway in our laboratory.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
EXPERIMENTAL SECTION
1
13
1
General Methods. H and C{ H} NMR spectra were recorded on CDCl
3
with
1
tetramethylsilane (Me
4
Si) as an internal standard, and DMSOꢀd
6
and CD
3
CN at 400 MHz for H
1
3
13
NMR and 100 MHz for C NMR. Protonꢀdecoupled carbon data of C NMR are reported. High
resolution mass spectra (HRMS) were recorded on a doubleꢀfocusing mass spectrometer by using
Atmospheric Pressure Chemical Ionization (APCI). Oxidation and reduction potentials in MeCN
16b
were measured using cyclic voltammetry and a previously described procedure. Calibration of
the potentials were performed using the formal potentials of ferrocene/ferrocenium couple, which
3
are 0.067 V and 0.442 V versus Ag/AgNO and SCE, respectively. Halfꢀwave potentials (E1/2)
reported in the manuscript were obtained from the peak potentials by subtracting or adding 0.029 V.
Light sources for photoreactions were a 500 W Xe lamp with glass filter Lꢀ42 (λ > 390 nm) and a
7.3 W white LED. Column chromatography was performed with silica gel. Anhydrous solvents for
2 2 3
photoreactions were obtained as follows. CH Cl and PhCH were purified in a same manner by the
2 4 2 2
treatment with H SO , water, 5% NaOH, water, and CaCl and then distilled over CaH . MeCN was
2 5 2 3
distilled over P O and subsequently distilled with K CO . Anhydrous DMF, DMSO, and MeOH
27
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 28 of 53
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
were purchased and used without distillation. Other reagents and solvents were used without further
purification.
11c
11c
11c
39
24c,40
24a,41
24a,42
24a,43
44
Substrates 2a, 2e, 2f, 4, 6,
8a,
8b,
8c,
and 16, which are known
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
compounds, were prepared by using reported procedures. Preparation procedure and spectroscopic
data of substrates 2b-2d, 2g are described below. 1ꢀMethylꢀ2ꢀhydroxy naphthalene 15 is also
45
known, and was prepared from 2ꢀnaphthol.
24a
Preparation of benzimidazolines. 1,3ꢀDimethylbenzimidazoline (BIH) derivatives 1a,
16a
16a
16a
1
b, 2ꢀ(2ꢀhydroxyphenyl)ꢀBIH and 2ꢀ(4ꢀhydroxyphenyl)ꢀBIH were prepared using previously
reported procedures. While all of these substances were already characterized, chemical shifts of
1
24a
HꢀNMR for 1a in CDCl
3
,
CD
3
CN and DMSOꢀd
6
are described below for the discussion in the
1
text. 2-(2-hydroxynaphthyl)-1,3-dimethybennzimidazoline (1a). HꢀNMR (400 MHz, CDCl )
δ
3
10.43 (br s, 1H), 8.02 (d, J = 8.8 Hz, 1H), 7.85–7.80 (m, 2H), 7.51–7.45 (m, 1H), 7.35 (t, J = 7.4 Hz,
1H), 7.20 (d, J = 9.2 Hz, 1H), 6.90–6.84 (m, 2H), 6.68–6.63 (m, 2H), 5.84 (s, 1H), 2.68 (s, 6H);
1
HꢀNMR (400 MHz,CD CN)
δ 10.54 (br s, 1H), 8.15 (d, J = 8.0 Hz, 1H), 7.90–7.85 (m, 2H), 7.52–
3
7
2
.48 (m, 1H), 7.38ꢀ7.34 (m, 1H), 7.14 (d, J = 8.8 Hz, 1H), 6.83 (s, 2H), 6.70 (s, 2H), 5.91 (s, 1H),
1
.62 (s, 6H). HꢀNMR (400 MHz,DMSOꢀd )
δ 9.95 (br s, 1H), 8.53ꢀ8.51 (m, 1H), 7.80–7.76 (m,
6
2H), 7.25ꢀ7.19 (m, 3H), 6.65–6.63 (m, 2H), 6.49ꢀ6.47 (m, 2H), 6.01 (s, 1H), 2.46(s, 6H).
28
ACS Paragon Plus Environment

Page 29 of 53
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Preparation of
N-Benzyl-N-tosylbezamides 2b, 2c, 2d, and 2g. Synthesis of
N-Benzyl-N-tosylamine. To benzylamine (2.2 ml, 20 mmol) and pꢀtoluenesulfonyl chloride (4.577 g,
23.9 mmol) was added pyridine (20 mL) with cooling in iceꢀwater bath. The resulting mixture was
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
purged with N for 10 min and subsequently heated at 50 ºC for 20 h. Then, after addition of water
(
2
50 mL), extraction with Et O (50 mL × 3) was performed. The extract was washed with 2M HCl
(
4
50 mL), 2M NaOH (50 mL), and brine (50 mL), and then dried over anhydrous MgSO . Addition
of EtOH to the residue obtained after concentration gave NꢀbenzylꢀNꢀtosylamine as a colorless solid
(
4.185 g, 16.0 mmol, 80%).
Synthesis of N-Benzyl-N-tosyl-4-chlorobenzamide (2b). To NꢀbenzylꢀNꢀtosylamine (785 mg,
3.0 mmol) in DMF(11 ml) were added NaH (~60%, 204 mg, 5.1 mmol) and pꢀchlorobenzoyl
chloride (0.42 mL, 3.3 mmol) with cooling in iceꢀwater bath. This mixture was stired at room temp
2
for 6 h and then extracted with Et O (40 mL × 2) after addition of water (40 mL). The extract was
4
washed with water (40 mL x 2) and brine (40 ml), and then dried over anhydrous MgSO . The
residue obtained by concentration was subjected to column chromatography using EtOAc and
nꢀC
6
H14 (1/4). The obtained solid was crystalized from EtOH to give 2b (720 mg, 1.8 mmol, 60%).
1
Colorless solid; mp 99.5–100.0 °C; HꢀNMR (400 MHz, CDCl )
δ 7.58 (d, J = 8.0 Hz, 2H), 7.41 (d,
3
13
J = 8.0 Hz, 2H), 7.30 (d, J = 8.0 Hz, 2H), 7.27–7.20 (m, 7H), 4.93 (s, 2H), 2.41 (s, 3H); C NMR
(100 MHz, CDCl ) δ 170.7, 145.0, 138.2, 136.1, 153.7, 133.6, 129.9, 129.7, 128.7, 128.6, 128.5,
3
29
ACS Paragon Plus Environment