N -Chlorinative Ring Contraction of 1,4-Dimethoxyphthalazines via a Bicyclization/Ring-Opening Mechanism

Article abstract of DOI:10.1055/s-0040-1706639

An unprecedented N -chlorinative ring contraction of 1,2-diazines was discovered and investigated with an electrophilic chlorinating reagent, trichloroisocyanuric acid (TCICA). Through optimization and mechanistic analysis, the assisting role of n -Bu ₄NCl as an exogenous nucleophile was identified, and the optimized reaction conditions were applied to a range of 1,4-dimethoxyphthalazine derivatives. Also, an improvement of overall efficiency was demonstrated by the use of a labile O -silyl group. A bicyclization/ring-opening mechanism, inspired by the Favorskii rearrangement, was proposed and supported by the DFT calculations. Furthermore, the efforts on scope expansion as well as the evaluation of other electrophilic promoters revealed that the newly developed ring contraction reactivity is a unique characteristic of 1,4-dimethoxyphthalazine scaffold and TCICA.

Full text of DOI:10.1055/s-0040-1706639

SYNTHESIS
0
0
3
9
-
7
8
8
1
1
4
3
7
-
2
1
0
X
Georg Thieme Verlag KG Rüdigerstraße 14, 70469 Stuttgart
2
021, 53, 1760–1770
paper
1760
en
Synthesis
J. K. Im et al.
Paper
N-Chlorinative Ring Contraction of 1,4-Dimethoxyphthalazines
via a Bicyclization/Ring-Opening Mechanism
Jeong Kyun Im◊
Ilju Jeong◊
OMe
O
Cl+
Cl–
N
N
ByeongDo Yang
Hyeon Moon
N Cl
O
OMe
Jun-Ho Choi*
Won-jin Chung*
♦ Chlorenium-promoted ring contraction
♦
♦
Bicyclization/ring-opening mechanism
Elimination of a nitrogen fragment
Department of Chemistry, Gwangju Institute of Science and
Technology, 123 Cheomdangwagiro, Bukgu, Gwangju 61005,
Republic of Korea
junhochoi@gist.ac.kr
wjchung@gist.ac.kr
◊
These authors contributed equally
Corresponding Authors
Received: 26.10.2020
Accepted after revision: 14.11.2020
Published online: 21.12.2020
poor 1,2-diazine derivatives as well.4,5 On the other hand,
the complementary, oxidative ring contraction of electron-
rich 1,2-diazines has been investigated sporadically and
thus not found its synthetic utility, yet (Scheme 1b). Only a
few related examples with 2,3-dihydrophthalazine-1,4-di-
DOI: 10.1055/s-0040-1706639; Art ID: ss-2020-f0550-op
6
Abstract An unprecedented N-chlorinative ring contraction of 1,2-di-
azines was discovered and investigated with an electrophilic chlorinat-
ing reagent, trichloroisocyanuric acid (TCICA). Through optimization
one derivatives have been reported a while ago (Scheme
7
2
). In the presence of either oxygen gas or ceric ammoni-
and mechanistic analysis, the assisting role of n-Bu NCl as an exoge-
4
nous nucleophile was identified, and the optimized reaction conditions
were applied to a range of 1,4-dimethoxyphthalazine derivatives. Also,
an improvement of overall efficiency was demonstrated by the use of a
labile O-silyl group. A bicyclization/ring-opening mechanism, inspired
by the Favorskii rearrangement, was proposed and supported by the
DFT calculations. Furthermore, the efforts on scope expansion as well
as the evaluation of other electrophilic promoters revealed that the
newly developed ring contraction reactivity is a unique characteristic of
um nitrate (CAN), ring contraction took place to give
phthalimides in moderate yields. However, such reactivity
has not been systematically examined, and the contraction
mechanism is still unknown. Recently, our group has re-
ported a preliminary communication on a rare type of oxi-
dative ring contraction of electron-rich 1,4-dimethoxy-
phthalazines that is initiated by N-chlorination (Scheme
1
,4-dimethoxyphthalazine scaffold and TCICA.
8
1
b). The newly observed oxidative reaction is complemen-
tary to the well-known reductive ring contraction and ap-
plicable to a range of alkyl- and halogen-substituted
phthalazine derivatives although the reaction efficiency
was moderate. Herein, we report a detailed full account on
the investigation of this unusual mode of ring contraction.
Key words halogenation, heterocycles, trichloroisocyanuric acid, ring
contraction, phthalazine, phthalimide
Ring contraction of polyaza-heterocycles can serve as a
useful method for the construction of smaller aza-hetero-
1
(a) Reductive ring contraction of electron-poor 1,2-diazines
cycles. 1,2-Diazines are particularly suitable substrates for
-
well-developed under various conditions
such purpose because of the presence of a weak, easily
cleavable N–N bond. The loss of a nitrogen fragment typi-
cally takes place in the form of ammonia or ammonium ion.
Reductive ring contraction of electron-poor 1,2-diazine has
been the main focus of this field and investigated by many
research groups under a variety of reaction conditions
Zn, AcOH or TFA
EWG
chemical reduction
EWG
(
Kornfeld–Boger)
N
N
N
H
4
e– + 4 H+
EWG
EWG
electrochemical
reduction
2–4
(
Scheme 1a). For example, pyridazine 3,6-diester can be
contracted to pyrrole in the presence of chemical reductant
such as Zn/AcOH via reductive cleavage of the N–N bond
with high efficiency. This process has been effectively ad-
opted by the groups of Kornfeld and Boger in a sequential
tetrazine Diels–Alder cycloaddition/reductive ring contrac-
(
b) Oxidative ring contraction of electron-rich 1,2-diazines
-
rarely studied
OMe
O
Cl+
N
N
X
X
N Cl
2
c,2g–i
this work
tion strategy.
Similar transformations could also be
3
conducted under electrochemical conditions, in which
electricity serves as the source of electrons. Thermal and
photochemical activation effect the contraction of electron-
OMe
O
Scheme 1 Ring contraction of 1,2-diazine under (a) reductive and (b)
oxidative conditions
©
2020. Thieme. All rights reserved. Synthesis 2021, 53, 1760–1770
Georg Thieme Verlag KG, Rüdigerstraße 14, 70469 Stuttgart, Germany

1761
Synthesis
J. K. Im et al.
Paper
X
O
X
TCICA was increased up to eight equivalents, and the con-
version was gradually improved up to 89% (entries 3–5). At
this stage, it was found that the addition of exogenous chlo-
ride facilitates the ring contraction (see the mechanistic
O
R
H
DMF, O2, hν
N
N
N
R
H
R = H, Ph
X = H, NH2
27–33% yield
O
O
O
O
discussion below). In the presence of n-Bu NCl, a sizable en-
4
OR
OR
13
hancement of the conversion was observed (entry 3 vs 6).
H
H
CAN, MeCN, H2O
R = H, Me
N
N
However, loss of the product during repeated chromato-
graphic purification on silica gel led to erroneous isolated
yields. Interestingly, simple change of the reaction solvent
from MeCN to chlorinated hydrocarbon such as 1,2-DCE at-
tenuated the ring contraction even at reflux and resulted in
the formation 7 as the major product (entry 7). The low nu-
cleophilicity of 7 and the relative instability of the cationic,
chlorinated intermediate in nonpolar solvent are likely to be
responsible for the attenuation of the second N-chlorina-
tion as well as the subsequent ring contraction.
N
68–77% yield
O
OMe
O
OMe
Scheme 2 Oxidative ring contraction of 2,3-dihydrophthalazine-1,4-
diones
Common electrophilic chlorinating reagents such as N-
chlorosuccinimide and N-chlorophthalimide have a polar-
ized N–Cl bond. The electron affinity of nitrogen is en-
hanced by the presence of one or more electron-withdraw-
ing groups on it, and the attached chlorine atom is thereby
rendered electron-deficient. We were curious about the un-
known, potentially interesting chlorinating reactivity of the
consecutively arranged N–Cl bonds. The desired structural
motif can be represented as N,N′-dichlorophthalazinedione
Even though the chlorinative ring contraction product
8a was generated in high conversion, its instability on silica
gel prevented quantitative recovery of pure material. This
problem could be alleviated by the reduction of the reactive
N–Cl bond with reducing agent prior to the purification (Ta-
1
4
ble 2). It was found that NaBH reduces 8a quantitatively.
4
(
1) (Scheme 3). Initially, direct, sequential N-chlorinations
With this reductive workup process, phthalimide (9a) could
be obtained in a reliable fashion (Table 2, entry 1). Again,
the ring contraction was assisted by the addition of n-
Bu NCl (entry 2). As the amount of n-Bu NCl was increased,
of 2,3-dihydrophthalazine-1,4-dione (2a) was attempted
with a variety of electrophilic chlorinating reagents includ-
ing trichloroisocyanuric acid (TCICA) in AcOH.9 However,
phthalic anhydride (3a) was obtained as a major product
along with insoluble polymeric materials, and no sign of
N-chlorination was detected (Scheme 3a). It is probable
that, after the first N-chlorination, the facile elimination of
HCl must have taken place prior to the second N-chlorina-
tion, thus preventing the formation of 1 (Scheme 3b). Then,
the resulting oxidized form, phthalazine-1,4-dione (5)
4
4
the isolated yield of 9a was improved a bit (entries 3 and
1
0
15
4). A large excess of TCICA was also beneficial (entries 5
and 6). However, the isolation procedure became cumber-
some. When the ring contraction was attempted at higher
(
(
(
a) Initial Attempt: Sequential, Direct Chlorinations of N–H Bonds
O
O
O
could have either polymerized by itself or reacted with sol-
H
H
Cl
Cl
TCICA, LiCl
AcOH, rt
N
N
N
N
vent (AcOH) to give 3a.11 To avoid the undesired HCl elimi-
+
O
nation of the putative intermediate 4, the N–H bonds need
to be removed. We hypothesized that an imidate moiety
may serve as a reactive nitrogen nucleophile that is absent
of N–H bonds (Scheme 3c). Therefore, 1,4-dimethoxy-
phthalazine (6a) was selected as a potential precursor to
the vicinal N,N′-dichloride.
O
O
O
2
a
1 (0%)
3a (56%)
desired product
b) Problem: Facile Elimination of HCl
O
O
O
H
H
Cl+
Cl
– HCl
N
N
N
N
N
N
undesired
elimination
N-Chlorination of 6a was examined with TCICA (Table
). At room temperature, the reaction was sluggish (Table 1,
H
1
O
a
O
O
2
4
5
entry 1). Only a small amount of monochloride 7 was pro-
duced, and no trace of dichloride was detected. The forma-
tion of 5 was prevented as proposed, but the reaction
stalled after the first N-chlorination. The transformation
was driven forward at higher temperature in refluxing
MeCN (entry 2). Unexpectedly, instead of the intended sec-
ond N-chlorination, an unprecedented type of electrophilic
halogenation-induced ring contraction took place to give N-
chlorophthalimide (8a) via elimination of one nitrogen at-
om. The ability of TCICA in promoting ring contraction of
unstable
c) Alternative Synthetic Plan: Removal of N–H bond
OMe
O
O
Cl+
Cl
Cl+
Cl
N
N
N
N
N
N
HCl elimination
impossible
Cl
OMe
OMe
O
6
a
7
1
no N–H bond
vicinal N–Cl bonds
Scheme 3 Synthetic approach toward the formation of vicinal N–Cl
bonds: (a) unsuccessful direct N–H chlorination, (b) decomposition of
intermediate via HCl elimination, (c) alternative reactant without N–H
bond.
6
a seems unique as other common electrophilic halogenat-
1
2
ing reagents are ineffective. Subsequently, the amount of
©
2020. Thieme. All rights reserved. Synthesis 2021, 53, 1760–1770

1762
Synthesis
J. K. Im et al.
Paper
Table 1 Initial Survey of Reaction Variablesa
cyclic structures (Scheme 5a). In addition, the installation
of a fluorine substituent was hampered by nucleophilic dis-
placement in the last step (Scheme 5b). From the reactions
OMe
O
O
Cl
8
TCICA (A equiv)
of 10j and 10h with NaOMe under the standard conditions,
N
N
N
N
+
N Cl
MeCN
the corresponding trimethoxyphthalazines 6j and 6k were
obtained. The relative reactivity of halogen substituents
could be estimated by stopping these reactions prematurely
and analyzing the remaining halogens. It turned out that
the 6-fluorine substituent of 10j is more labile than the
reflux, 48 h
O
OMe
OMe
6
a
7
8a
Entry
A (equiv)
6ab
7b
12
8ab
Yield (%) of 8a
1
- and 4-chlorine substituents. In contrast, the 5-fluorine
1c
2
1
1
2
4
8
2
4
88
14
0
0
36
51
75
89
82
20
0
substituent of 10h was found to be slightly less reactive
than the chlorine, and thus it was possible to produce 6h by
reducing the amount of NaOMe to two equivalents albeit
the low isolated yield (Scheme 4). A similar problem was
50
49
25
11
18
80
10
3
34
4
0
49
5
0
44d
37d
ND
also present when the introduction of a NO group was at-
2
tempted.
6e
0
The scope of N-chlorinative ring contraction was exam-
ined with these 1,4-dimethoxyphthalazines 6 under the op-
timized reaction conditions (Scheme 6). As observed with
7f
0
a
Reaction conditions: 1.0 mmol of 6a at 0.17 M concentration.
Approximate integration ratio by H NMR analysis of the crude mixture.
At room temperature.
Chromatographed twice on silica gel.
With 1 equiv of n-Bu NCl.
4
In 1,2-DCE.
b
c
d
e
f
1
6a, the addition of n-Bu NCl usually resulted in an im-
4
proved yield. On the other hand, the reductive NaBH work-
4
up was not always helpful as it often led to a diminished
mass recovery. Thus, the substrate survey was conducted
both with and without NaBH workup in some cases. The
4
temperature in refluxing EtCN, TCICA was consumed
through the chlorination of solvent, leading to a noticeably
decreased product formation (entry 7). Bromide turned out
to be an incompetent nucleophile, and its oxidation by
TCICA probably complicated the reaction progress (entry 8).
Additionally, the ease of dealkylation appeared critical as
the benzyl derivative 6′ was less reactive (entry 9). With the
reaction conditions in entry 4, we set out to survey sub-
strates.
derivatives with an alkyl group at the 6-position contract to
the corresponding N-chlorophthalimides 8b,c and
phthalimide 9c in moderate yields. The reaction conditions
also tolerate electron-withdrawing halogen substituents
(8e, 9d,e). Fortunately, steric hindrance at the 5-position
does not have noticeable influence on ring contraction. An
alkyl group as well as halogen substituents can be em-
Table 2 Reaction Condition Optimization with Reductive Workupa
A variety of 1,4-dimethoxyphthalazine derivatives 6
were prepared from readily available phthalic anhydride (3)
via a three-step sequence in a straightforward manner
OR
O
TCICA (A equiv)
n-Bu4NX (B equiv)
N
N
H
N
(
Scheme 4). First, condensation of 3 with hydrazine hydrate
MeCN, reflux, 48 h
then NaBH4
afforded 2,3-dihydrophthalazine-1,4-diones 2, which were
subsequently transformed to 1,4-dichlorophthalazines 10
upon treatment with POCl3.16 Then, the following displace-
ment of the chlorides with sodium methoxide provided 6 in
variable overall yields.17 This procedure generally tolerates
substituents at the 6-position. Weakly electron-donating
alkyl groups (b, c) as well as electron-withdrawing bromine
and chlorine substituents (d, e) could be installed at the 6-
position. The introduction of substituents at the sterically
hindered 5-position was accomplished in slightly lower
overall yields (f–h). In addition, a disubstituted substrate (i)
was also prepared with moderate efficiency.
Unfortunately, it was difficult to access higher substitu-
tion patterns. From tetrachloro- and tetrabromophthalic
anhydrides 3l and 3m, the corresponding N-aminoph-
thalimides 11l and 11m were produced as the major consti-
tutional isomers probably because sterically crowded envi-
ronment favors the formation of smaller five-membered
O
OR
6a (R = Me)
' (R = Bn)
9a
6
Entry
R
X
A (equiv)
B (equiv)
Yield (%)b
1
2
3
4
5
6
7c
8
Me
Me
Me
Me
Me
Me
Me
Me
Bn
Cl
4
0
1
2
4
1
2
1
4
1
35
49
52
54
44
54
32
11
28
Cl
Cl
Cl
Cl
Cl
Cl
Br
Cl
4
4
4
8
8
4
4
4
9
a
Reaction conditions: 1.0 mmol of 6a or 6′ at 0.17 M concentration.
Isolated yield after column chromatography.
In refluxing EtCN.
b
c
©
2020. Thieme. All rights reserved. Synthesis 2021, 53, 1760–1770

1763
Synthesis
J. K. Im et al.
Paper
O
O
Cl
Cl
OMe
O
N2H4•H2O
POCl3
(4.0 equiv)
NaOMe
(4.0 equiv)
(1.5 equiv)
NH
NH
N
N
N
N
X
O
X
X
X
10% aq HCl
neat
reflux, 3 h
MeOH
reflux, 24 h
reflux, 24 h
O
OMe
3
2
10
6
O
Cl
OMe
2
b (X = Me, 95%)
10b (X = Me, 85%)
10c (X = t-Bu, 91%)
10d (X = Br, 76%)
10e (X = Cl, 74%)
6b (X = Me, 83%)
6c (X = t-Bu, 69%)
6d (X = Br, 34%)
6e (X = Cl, 35%)
X
X
X
NH
NH
2c (X = t-Bu, 95%)
N
N
N
N
2
2
d (X = Br, 96%)
e (X = Cl, 97%)
O
O
Cl
Cl
OMe
OMe
X
X
X
2
2
f (X = Me, 79%)
g (X = Cl, 42%)
10f (X = Me, 89%)
10g (X = Cl, 74%)
10h (X = F, 76%)
6f (X = Me, 39%)
6g (X = Cl, 34%)
6h (X = F, 18%)
NH
NH
N
N
N
N
2h (X = F, 80%)
O
O
Cl
Cl
OMe
OMe
Cl
Cl
Cl
NH
NH
N
N
N
N
2i (84%)
10i (79%)
6i (25%)
Cl
Cl
Cl
O
Cl
OMe
Scheme 4 Preparation of 1,4-dimethoxyphthalazine substrates. Yield of the crude material is given for 2 and 10. Isolated yield after column chroma-
tography is given for 6. For 6h: The reaction was conducted with 2.0 equiv of NaOMe.
ployed with comparable efficiencies (8f–h). Strangely, 6,7-
dichlorinated substrate 6i is reluctant to the ring contrac-
tion. Upon the standard operation, the unreacted reactant
and the monochlorinated intermediate were produced as
the major components. Only a small amount of the ring
contraction product 8i could be obtained even after 5 days
of reaction time. Unfortunately, the current oxidative pro-
cess is not applicable to highly electron-rich substrates such
as 6j and 6k, which decompose under the reaction condi-
tions.
Additionally, several control experiments were carried
out (Scheme 7). A few readily available and structurally re-
lated heterocyclic compounds such as 2,3-dihydrophthala-
OMe
O
TCICA (4 equiv)
n-Bu NCl (0–4 equiv)
4
N
X
X
N
Z
N
MeCN, reflux, 48 h
with or without
NaBH workup
4
O
OMe
8 (Z = Cl)
(Z = H)
6
9
O
O
O
O
To probe the possibility of scope expansion, a monocy-
clic analogue, 3,6-dimethoxypyridazine (12) was examined
Me
t-Bu
Cl
Br
N
Z
N
N
Z
N
Z
(
Equation 1). N-Chlorination may take place, but it is gener-
ally more difficult to break the aromaticity of monocyclic -
system than that of polycyclic one. Consequently, instead of
the ring contraction, an electrophilic chlorination of the ar-
omatic ring was observed.
O
O
O
O
A: 8a (49%)
C: 9a (49%)
D: 9a (54%)
A: 8b (28%)
B: 8b (48%)
A: 8c (56%)
B: 8c (43%)
C: 9c (31%)
C: 9d (28%)
D: 9d (42%)
O
Me
O
Cl
O
F
O
Cl
(
a)
Z
Z
N
Z
O
O
N
Cl
N
Cl
N
Cl
N2H4•H2O (1.5 equiv)
EtOH, reflux, 30 min
Z
Z
Z
Z
O
O
O
O
O
N
NH2
then 10% aq HCl
reflux, 30 min
B: 8e (40%)
C: 9e (32%)
B: 8f (49%)
B: 8g (51%)
B: 8h (39%)
O
O
Z
Z
3
l (Z = Cl)
11l (Z = Cl, 60%)
11m (Z = Br, 54%)
O
3
m (Z = Br)
Cl
Cl
conditions A: n-Bu
conditions B: n-Bu
conditions C: n-Bu
conditions D: n-Bu
4
4
NCl (0 equiv) without NaBH
NCl (4 equiv) without NaBH
4
4
workup
workup
(
b)
Y
Cl
Y
OMe
N
Cl
NaOMe
4.0 equiv)
4
NCl (1 equiv) with NaBH
4
NCl (4 equiv) with NaBH
4
4
workup
workup
X
X
(
N
N
N
O
N
MeOH
reflux, 24 h
B: 8i (21%)
Cl
OMe
Scheme 6 N-Chlorinative ring contraction of 1,4-dimethoxyphthala-
zines. The reactions were conducted with 1.0 mmol of 6 at 0.17 M con-
centration. Isolated yields after column chromatography are shown in
the parenthesis. For 8i: Reaction time: 5 days.
1
1
0j (X = F, Y = H)
0h (X = H, Y = F)
6j (X = OMe, Y = H, 54%)
6k (X = H, Y = OMe, 43%)
Scheme 5 Problems encountered during substrate preparation
©
2020. Thieme. All rights reserved. Synthesis 2021, 53, 1760–1770

1764
Synthesis
J. K. Im et al.
Paper
OMe
OMe
these compounds was also challenging because these O–Si
bonds are hydrolytically unstable as expected. To avoid the
exposure of the silylated intermediates to moisture, a one-
pot operation was pursued (Scheme 8). Accordingly, the
treatment of 2a with TMSCl in the presence of 2,6-lutidine
afforded the monosilylated species 14, which was charac-
TCICA (4 equiv)
n-Bu4NCl (4 equiv)
Cl
N
N
N
N
MeCN, reflux, 48 h
OMe
OMe
1
2
13 (29%)
1
Equation 1 Attempt at scope expansion to monocyclic analogue
terized by H NMR and HRMS. Even when excess amounts
of TMSCl and base were employed, 14 was formed as a ma-
jor product. Therefore, the subsequent one-pot ring con-
traction was investigated with 14, hoping that the N-chlori-
nation would produce an intermediate analogous to 7. In-
terestingly, the reaction of 14 with TCICA proceeds
vigorously even at room temperature, and immediate gas
evolution was observed upon mixing. To control the high
reactivity, the chlorinative contraction step was performed
in a cooling bath. The reaction was still facile at 0 °C, and 14
was completely consumed within one hour. Unfortunately,
the crude mixture was complex, and the product 8a was
isolated in low yield. Further lowering the reaction tem-
perature to –40 °C did not lead to meaningful improvement.
Although the chemical yield of this process is not yet high
enough to be practical, the overall efficiency has been no-
ticeably enhanced through the modification of the O-pro-
tecting group. By utilizing the easily attachable and readily
cleavable TMS group, the previous lengthy reaction se-
quence became dramatically shortened, and much milder
reaction conditions were accomplished in terms of both re-
action time and temperature.
zine-1,4-dione (2a) and N-aminophthalimide (11a) were
used as potential alternative starting materials (Scheme
7
a). However, a complex mixture was obtained from 2a
without any sign of 8a, and only a small amount of 8a was
produced from 11a. Thus, it appears that the N-chlorinative
ring contraction reactivity is a characteristic property of
the electron-rich phthalazine scaffold. On the other hand,
other common oxidants including PhI(OAc) , ceric ammoni-
2
um nitrate (CAN), urea·H O (UHP), and 2,3-dichloro-5,6-
2
2
dicyano-1,4-benzoquinone (DDQ) are unable to promote
analogous oxidative ring contraction of 6a (Scheme 7b).
(
a) Alternative reactants
O
TCICA (4 equiv)
NH
NH
a complex mixture
MeCN, reflux, 48 h
O
2
a
O
O
TCICA (4 equiv)
N
NH2
N Cl
MeCN, reflux, 48 h
O
O
O
O
TMSCl
,6-lutidine
TCICA
n-Bu4NCl
1
1a
8a (10%)
2
NH
N
2
a
N Cl
(
b) Other oxidants
MeCN
MeCN
reflux, 12 h
temp, 1 h
OMe
O
PhI(OAc)2, CAN,
OTMS
urea•H2O2, or DDQ
14
H NMR & HRMS
8a
15% at 0 °C
16% at –40 °C
N
N
no ring contraction
1
MeCN, reflux
OMe
Scheme 8 Shorter sequence and milder conditions via the use of a la-
bile O-silyl protecting group
6
a
Scheme 7 Control experiments: (a) related heterocycles as reactants
and (b) examination of common oxidants.
Ring contraction mechanism can be categorized into
three types (Scheme 9). The transformation can be initiated
by cleavage of a bond within the ring, and then the subse-
quent cyclization onto a closer site leads to ring contraction
(Scheme 9a). Most of the reductive ring contractions of 1,2-
diazine were proposed to proceed via this type of ring-
Though an unprecedented reactivity has been discov-
ered, the ring contraction efficiency is only moderate. One
of the most difficult bond-forming/breaking events in the
transformation of 6 was thought to be the removal of two
O-methyl groups, which may, at least partly, be responsible
for the slow reaction rate. Thus, it was hypothesized that
the replacement of these robust methyl groups with more
labile protecting groups such as silyl groups may facilitate
the overall process. Moreover, the direct O-silylation of 2
will significantly decrease the number of steps, and thus
the step-economy could be improved as well. However,
whereas the first silylation took place readily, the installa-
tion of the second silyl group was sluggish, providing a mix-
ture of mono- and disilylated products. The purification of
2
,3
opening/ring-closure mechanism. Alternatively, bond for-
mation can take place first to generate a bicyclic intermedi-
ate (Scheme 9b). Then, the following opening of the
strained small ring completes the bicyclization/ring-open-
ing mechanism. A representative example of this category
18
is the well-known Favorskii rearrangement. Similar mech-
anisms have also been proposed for oxidative as well as re-
ductive ring contraction of polyaza-heterocycles.19 The
third pathway involves simultaneous migration of a -bond
©
2020. Thieme. All rights reserved. Synthesis 2021, 53, 1760–1770

1765
Synthesis
J. K. Im et al.
Paper
in a concerted fashion (Scheme 9c). After these ring con-
tractions, the resulting exocyclic part Y is often eliminated
as a small fragment.
sistent with the observed beneficial effect of n-Bu NCl (Ta-
4
bles 1 and 2). Finally, the extrusion of a small nitrogen frag-
ment from iii and the subsequent demethylation of iv
completes the N-chlorinative ring contraction to 8a. This
proposed mechanism is supported by density functional
theory (DFT) calculation (Figure 1b). The transition states
for the critical bicyclization/ring-opening sequence were
(
a) X–Y bond
cleavage
X–Z bond
formation
Z
Y
ring-opening
ring-closure
X
successfully located at the M06-2X-D3/6-311+G(d,p) level
(
b) X–Z bond
formation
X–Y bond
cleavage
21,22
of theory
with the implicit Solvation Model based on
Z
Y
Z
23
24–26
Z
Y
Density (SMD) by the GAMESS(US) software package.
bicyclization
X
Y
ring-opening
X
X
The precise single point energies were calculated with a
larger 6-311++G(2d,2p) basis set. The high activation barri-
er of the bicyclization step is probably responsible for the
moderate efficiency of overall ring contraction (from i to ii,
(
c)
σ-bond
migration
‡
Z
concerted
process
X
Y
‡
ΔG = 29.3 kcal/mol). The unassisted opening of diaziridini-
Scheme 9 Three types of ring contraction mechanism: (a) ring-open-
ing/ring-closure mechanism, (b) bicyclization/ring-opening mecha-
nism, and (c) concerted -bond migration mechanism.
um is energetically uphill (from ii to v, ΔG° = 27.6 kcal/mol),
+
27
and the resulting chloronitrenium (R–N=Cl ) species v re-
sides on a saddle point. On the other hand, the incorpora-
tion of a chloride nucleophile renders this process highly fa-
vorable (from ii to iii, ΔG° = –40.9 kcal/mol), and only a
The ring-opening/ring-closure mechanism would re-
quire a high activation energy because this process is initi-
ated by the cleavage of a strong bond.20 Thus, a pathway in-
volving a bond formation at an early stage is more probable,
and we propose a bicyclization/ring-opening mechanism
‡
small amount of activation energy is required (ΔG = 2.0
kcal/mol), thus supporting the beneficial role of n-Bu NCl.
4
On the basis of the calculation results, a few modifica-
tions of the reaction conditions that may accelerate the ring
contraction process were suggested and evaluated compu-
tationally in a similar manner as described above (Figure 2).
The introduction of a more electronegative element such as
for the newly developed N-chlorinative ring contraction
8
(
Figure 1a). Starting from the identified intermediate 7,
b
chlorination of the nucleophilic imidate nitrogen (N ) with
TCICA affords a delocalized cation i containing a highly elec-
trophilic low-lying -system, which can then be attacked by
b
fluorine in place of the chlorine (Cl ) in i will enhance the
electrophilicity of the cationic -system and potentially fa-
cilitate the diaziridinium formation. This argument is sup-
ported by the DFT calculation as a decreased activation en-
a
the adjacent nitrogen (N ) to form a [3,1,0] bicyclic species
ii. The opening of strained diaziridinium ring would be as-
sisted by exogenous nucleophile, and this proposal is con-
‡
ergy (ΔG = 25.6 kcal/mol) is found for the ring contraction
(
a)
O
O
O
O
Cla
Cl+
Cl–
Cl+
Cl–
Na
Nb
Cl
7
+N
N
Cl
Cl
N
Cl
8a
Clb
N
– NCl3
– MeCl
+
Cl
N
MeO
ii
MeO
OMe
+
MeO
Cl
i
iii
iv
(
b)
ΔG° = 27.6
no TS
ΔG‡ = 29.3
ΔG = 24.9
°
v
‡
ΔG‡ = 2.0
ΔG° = –40.9
Cl–
i
ii
‡
iii
TS-1 (TS-Cl)
TS-2
Figure 1 (a) Proposed bicyclization/ring-opening mechanism and (b) DFT calculation (M06-2X-D3/6-311++G(2d,2p)//6-311+G(d,p), SMD-MeCN). The
free energies are shown in kcal/mol.
©
2020. Thieme. All rights reserved. Synthesis 2021, 53, 1760–1770

1766
Synthesis
J. K. Im et al.
Paper
of the fluorinated analogue (X = F). On the other hand, the
presence of a more Lewis basic element such as oxygen is
expected to accelerate the N–N bond cleavage. Interestingly,
the internal nucleophilic assistance of the oxygen lone pair
UHP/MeReO , a loss of the N-chlorine substituent to 15 was
observed without any detectable sign of ring contraction.
3
With CF CO H, 7 remained mostly unreacted, and only a
3
3
small amount of 8a was produced. Moreover, the reaction
with N-fluorobenzenesulfonimide (NFSI) did not proceed at
all. Other fluorenium reagents such as Selectfluor could not
be examined properly because of their low solubility in 1,2-
DCE. The lack of oxidative ring contraction reactivity in 1,2-
DCE suggests that it is necessary to find oxidants that are
compatible with MeCN.
–
of the N-oxide analogue (X = O ) led to a dramatic change of
the reaction mechanism. In this case, the N–N bond cleav-
age takes place simultaneously during the C–N bond forma-
tion via a concerted -bond migration (see Scheme 9c), di-
rectly leading to an uncharged nitroso species without the
intermediacy of a bicyclic diaziridinium species. In conse-
quence, the transformation becomes thermodynamically
favorable (ΔG° = –12.5 kcal/mol). However, the activation
O
mCPBA or
H
‡
UHP/MeReO3
energy increases significantly (ΔG = 34.0 kcal/mol). The 3D
N
N
structures of these transition states from the DFT calcula-
tion are depicted in Figure 2b. To verify these computation-
al predictions, additional experiments were carried out un-
der the modified reaction conditions.
reflux, 24 h
(
32–100% conv.)
OMe
1
5
O
O
Cl
TCICA
N
N
CF3CO3H
6
a
N Cl
(
a)
‡
O
O
1,2-DCE
reflux, 24 h
reflux, 24 h
(12% conv.)
O
Cl
Cl
OMe
+
Na
Nb
+
N
8a
7
: used in situ
X = F
C
N
O
F
F
MeO
MeO
23.2
Cl
X
NFSI
N
N
TS-F: 25.6
no reaction
reflux, 24 h
‡
O
O
MeO
Scheme 10 Examination of other potential electrophilic promotors via
one-pot operation in 1,2-DCE
+
Cl
+ Na
Grel
0.0
N
Cl
O
X = O–
C
Nb
O–
N
MeO
MeO
After a brief screening, it was found that K S O , an inor-
2
2
8
TS-O: 34.0
–12.5
ganic oxidant, does not oxidize MeCN. Thus, the promoting
ability of K S O was evaluated in MeCN (Table 3). For this
2
2
8
(b)
study, 7 was isolated so that the unreacted TCICA would not
hamper the precise analysis of the next reaction. Disap-
pointingly, K S O was unable to trigger the ring contrac-
2
2
8
tion (Table 3, entry 1). Although K S O is a poor promoter
2
2
8
by itself, it may have an assisting effect as was observed
with n-Bu NCl. To test this hypothesis, K S O was em-
4
2
2
8
ployed in combination with TCICA, and a small amount of
8a was produced (entry 2). However, the isolated yield is
TS-Cl
TS-F
1.79 Å
1.53 Å
TS-O
2.37 Å
1.90 Å
a
1.87 Å
1.49 Å
N –C
N –N
a
b
much lower than those of the reactions with TCICA/n-Bu NCl
4
(
entry 3) and even TCICA alone (entry 4). Therefore, it ap-
Figure 2 The computational analysis on the effects of other N-substit-
uents on ring contraction process (M06-2X-D3/6-311++G(2d,2p)//6-
pears that K S O actually interferes with the ring contrac-
2
2
8
3
11+G(d,p), SMD-MeCN). (a) thermochemical data (the relative free
tion. Although the inferior promoting ability of oxygenating
reagents has been predicted by the DFT calculation (Figure
energies in kcal/mol), (b) transition states comparison.
2
), the inhibiting effect is unexpected, and the reasons are
The reasonably selective formation of the monochlori-
nated intermediate 7 in 1,2-DCE (Table 1, entry 7) allowed
the survey of other electrophilic promotors for the ring
contraction step. Unfortunately, a majority of these experi-
ments could not be conducted in refluxing MeCN because
most peroxides and fluorenium reagents react with the sol-
vent. Therefore, one-pot operation in 1,2-DCE was attempt-
ed instead (Scheme 10), even though 1,2-DCE is a poor sol-
vent for the ring contraction with TCICA. However, the imi-
date nitrogen of 7 is insusceptible to oxidation under these
reaction conditions. Upon treatment with mCPBA or
currently unknown.
In summary, a rarely observed oxidative ring contrac-
tion of 1,2-diazines was accomplished by N-chlorination.
TCICA promoted this unusual reactivity whereas other oxi-
dants were ineffective. On the basis of the mechanistic pro-
posal, the presence of a soluble chloride was found to facili-
tate the ring contraction, and the exogenous nucleophile
was expected to assist the opening of the putative diaziri-
dinium intermediate. The newly developed reactivity was
applied to a variety of 1,4-dimethoxyphthalazine deriva-
tives with the exception of highly electron-rich substrates.
©
2020. Thieme. All rights reserved. Synthesis 2021, 53, 1760–1770

1767
Synthesis
J. K. Im et al.
Paper
a
Table 3 Evaluation of K S O as an Electrophilic Promoter in MeCN
electric sector double focusing mass analyzer at Korea Basic Science
Institute (KBSI), Daegu Center. The preparations of 2b–i, 6b–k, 8b–i,
10b–h, and 9d have been described previously.8
2
2
8
O
O
TCICA
2.0 equiv)
reagents
(see Table)
Cl
(
N
N
6
a
N
Cl
1,4-Dibenzyloxyphthalazine (6′)
1
,2-DCE
reflux, 24 h
2%
MeCN
reflux, 24 h
To a stirred slurry of NaH (622 mg, 15.6 mmol, 4.0 equiv) in THF (22
mL) was added benzyl alcohol (1.62 mL, 15.6 mmol, 4.0 equiv) drop-
wise over 5 min at 0 °C in an ice bath. The mixture was stirred at rt for
O
8
OMe
: isolated
8a
7
1
5 min. Then, a solution of 1,4-dichlorophthalazine (774 mg, 3.89
mmol, 1.0 equiv) in THF (18 mL) was added dropwise over 5 min at 0
C, and the reaction mixture was refluxed in an oil bath for 24 h. After
Entry
K
2
S
2
O
8
(equiv) TCICA (equiv)
n-Bu
4
NCl (equiv)
Yield (%)b
°
1
2
3
2.0
2.0
–
–
–
0
14
45
26
cooling the mixture to rt, H O was added, and the aqueous layer was
extracted with CH Cl (3 ×). The combined organic layers were dried
2
2.0
2.0
2.0
–
2
2
(
anhyd MgSO ), filtered through a glass frit, and concentrated under
4
2.0
–
reduced pressure. The residue was purified by column chromatogra-
phy (SiO , R = 0.39 in EtOAc/hexanes 1:10) and bulb-to-bulb distilla-
4
–
2
f
a
Reaction conditions: 0.5 mmol of 7 at 0.17 M concentration.
Isolated yield after column chromatography.
tion (100 mTorr, ABT 280–300 °C) to give 6′ as a white solid; yield:
b
9
98 mg (75%); mp 104.1–108.1 °C.
1
H NMR (400 MHz, CDCl ):  = 8.21–8.16 (m, 2 H), 7.83–7.79 (m, 2 H),
3
Also, although further exploration is still needed, the over-
all sequence as well as the reaction conditions could be im-
proved by utilizing a more labile O-TMS group. For this un-
precedented transformation, a bicyclization/ring-opening
mechanism was proposed from the isolated monochlori-
nated intermediate and supported by DFT calculation. Ad-
ditional theoretical analysis suggested potential rate accel-
eration and even a change of the reaction pathway through
alteration of the electronic property of the second N-sub-
stituent. However, the examination of several other electro-
philic reagents only revealed the unique ring contraction-
promoting ability of TCICA. Nonetheless, whereas reductive
ring contractions of polyaza-heterocycles are well-docu-
mented, the complementary transformations under oxida-
tive conditions are far less developed. We believe that our
new finding will provide a valuable addition to the modern
synthetic toolbox in heterocyclic chemistry
7.59–7.56 (m, 4 H), 7.45–7.41 (m, 4 H), 7.39–7.35 (m, 2 H), 5.66 (s, 4
H).
13
C NMR (100 MHz, CDCl ):  = 157.9, 137.2, 132.1, 128.7, 128.4,
3
128.2, 123.3, 122.6, 68.9.
+
HRMS (ESI/Q-TOF): m/z [M + Na] calcd for C22H18N O Na: 365.1260;
2
2
found: 365.1263.
N-Chlorinative Ring Contraction of 1,4-Dimethoxyphthalazines 6
to Phthalimides 9; General Procedure I (GP I) (Scheme 6)
A stirred solution of 1,4-dimethoxyphthalazine 6 (1.0 equiv), n-Bu NCl
4
(
1.0 equiv), and TCICA (4.0 equiv) in MeCN was refluxed in an oil bath
for 48 h. Then, the reaction mixture was cooled to 0 °C in an ice bath
2
8
prior to the addition of NaBH (5.0 equiv) and EtOH. After 15 min,
4
sat. aq NH Cl was added, and the aqueous layer was extracted with
4
CH Cl (3 ×). The combined organic layers were dried (anhyd MgSO ),
2
2
4
filtered through a glass frit, and concentrated under reduced pressure.
The residue was purified by column chromatography (SiO ) to give
phthalimide 9.
2
4
-tert-Butylphthalimide (9c)
Following GP I, 6c (246 mg, 1.00 mmol), n-Bu NCl (278 mg, 1.00
mmol), and TCICA (930 mg, 4.00 mmol) were reacted in MeCN (6 mL)
4
All reactions were performed in oven-dried (140 °C) or flame-dried
glassware under an atmosphere of dry argon unless otherwise noted.
Purification of solvents and reagents are described in the Supporting
Information. Column chromatography was performed with Merck
to give 9c as a white solid after column chromatography (SiO , R =
2
f
0.27 in EtOAc/hexanes 1:5); yield: 46 mg (31%); mp 137.4–140.6 °C.
1H NMR (400 MHz, DMSO-d
):  = 11.23 (br s, 1 H), 7.84 (dd, J = 7.8,
Millipore silica gel (SiO ) 60 (0.040–0.063 mm). Analytical TLC was
6
2
performed on Merck Millipore TLC silica gel 60 F254. Visualization of
1.4 Hz, 1 H), 7.78 (d, J = 0.6 Hz, 1 H), 7.73 (d, J = 7.6 Hz, 1 H), 1.33 (s, 9
TLC was accomplished with UV (254 nm) and a KMnO staining solu-
H).
4
tion. 1H and C{ H} NMR spectra were recorded on a JEOL ECS400
13
1
13
C NMR (100 MHz, DMSO-d ):  = 169.4, 169.1, 157.9, 132.9, 131.2,
6
1
13
spectrometer (400 MHz, H; 100 MHz, C). Chemical shifts are refer-
1
30.1, 122.8, 119.6, 35.4, 30.8.
1
13
enced to residual CHCl (7.26 ppm for H; 77.23 ppm for C) and
3
1
13
DMSO; (2.50 ppm for H; 39.52 ppm for C). Chemical shifts are re-
ported in ppm. Multiplicities are indicated by standard abbreviations.
Coupling constants, J, are reported in hertz (Hz). Kugelrohr distillation
was performed by a Büchi B585 glass oven with a Büchi bulb-to-bulb
distillation apparatus, and the temperatures reported are air bath
temperatures (ABT). ESI-HRMS was performed on an Agilent 6520
quadrupole time-of-flight (Q-TOF) mass spectrometer at Environ-
mental OMICS Laboratory, GIST, and EI-HRMS was performed on a
JEOL JMS-700 MStation mass spectrometer with magnetic sector-
4
-Chlorophthalimide (9e)
Following GP I, 6e (225 mg, 1.00 mmol), n-Bu NCl (278 mg, 1.00
mmol), and TCICA (930 mg, 4.00 mmol) were reacted in MeCN (6 mL)
to give 9e as a white solid after column chromatography (SiO , R =
4
2
f
0.35 in EtOAc/hexanes 1:5); yield: 58 mg (32%); mp 206.4–213.9 °C.
1H NMR (400 MHz, DMSO-d
13
):  = 11.49 (s, 1 H), 7.86–7.80 (m, 3 H).
6
C NMR (100 MHz, DMSO-d ):  = 168.3, 167.9, 139.1, 134.7, 134.1,
31.2, 124.7, 123.1.
6
1
©
2020. Thieme. All rights reserved. Synthesis 2021, 53, 1760–1770

1768
Synthesis
J. K. Im et al.
Paper
+
35
37
Attempted Preparation of 2,3-Dihydrophthalazine-1,4-diones 2
Resulting in the Formation of N-Aminophthalimides 11; General
Procedure II (GP II)29 (Scheme 5)
HRMS (ESI/Q-TOF): m/z [M + H] calcd for C H ClN O , C H ClN O :
6
8
2
2
6
8
2
2
175.0274 (100.0%), 177.0245 (32.0%); found: 175.0264 (100.0%),
177.0237 (28.1%).
A stirred solution of phthalic anhydride 3 (1.0 equiv) and N H ·H O
2
4
2
(
1.5 equiv) in EtOH was refluxed in an oil bath for 30 min. Then, aq
One-Pot Ring Contraction of 2,3-Dihydrophthalazine-1,4-dione
(2a) via a Silylated Intermediate (Scheme 8)
HCl (10 wt%) was added, and the reaction mixture was refluxed for an
additional 30 min. The mixture was cooled to rt and filtered through
filter paper. The filter cake was washed with EtOH and dried under
reduced pressure to give N-aminophthalimide 11. The crude material
was characterized without further purification.
A stirred solution of 2,3-dihydrophthalazine-1,4-dione (2a; 162 mg,
1
.00 mmol, 1.0 equiv), TMSCl (253 L, 2.00 mmol, 2.0 equiv), and 2,6-
lutidine (232 L, 2.00 mmol, 2.0 equiv) in MeCN (1.5 mL) was refluxed
in an oil bath under argon for 12 h to form 14. Then, the reaction mix-
ture was cooled to 0 °C or –40 °C prior to the addition of a solution of
N-Aminotetrachlorophthalimide (11l)
n-Bu NCl (1.11 g, 4.00 mmol, 4.0 equiv) in MeCN (1.5 mL). Then, a
4
solution of TCICA (930 mg, 4.00 mmol, 4.0 equiv) in MeCN (3.0 mL)
was added dropwise over 10 min. After 50 min, the mixture was
warmed to rt and concentrated under reduced pressure. The residue
was purified by column chromatography (SiO , R = 0.19 in CH Cl /
hexanes 1:1) to give N-chlorophthalimide 8a as a white solid; yield:
27 mg (15%) at 0 °C and 29 mg (16%) at –40 °C.
Following GP II, 3l (2.86 g, 10.0 mmol), aq HCl (10 wt%, 10 mL), and
N H ·H O (728 L, 15.0 mmol) were reacted in EtOH (20 mL) to give
2
4
2
1
1l; yield: 1.79 g (60%); ivory solid.
1
2
f
2
2
H NMR (400 MHz, DMSO-d ):  = 5.10 (br s, 2 H).
6
1
3
C NMR (100 MHz, DMSO-d ):  = 162.6, 138.0, 127.9, 127.0.
6
HRMS (EI): m/z [M] calcd for C H Cl N O , C H235Cl337ClN O ,
+
35
8
2
4
2
2
8
2
2
C8H235Cl2 Cl N O : 297.8870 (77.6%), 299.8841 (100.0%), 301.8811
37
2
2
2
4-[(Trimethylsilyl)oxy]phthalazin-1(2H)-one (14)
(
(
48.5%); found: 297.8870 (75.1%), 299.8850 (100.0%), 301.8818
50.5%).
1
H NMR (400 MHz, CDCl ):  = 9.70–9.67 (br s, 1 H), 8.30 (d, J = 7.6 Hz,
H), 7.91 (d, J = 7.3 Hz, 1 H), 7.80–7.71 (m, 2 H), 0.34 (s, 9 H).
3
1
HRMS (EI): m/z [M]+ calcd for C11H14N O Si: 234.0825; found:
2
2
N-Aminotetrabromophthalimide (11m)
2
34.0825.
Following GP II, 3m (4.64 g, 10.0 mmol), aq HCl (10 wt%, 10 mL), and
N H ·H O (728 L, 15.0 mmol) were reacted in EtOH (20 mL) to give
2
4
2
Electrophilic Promotor Survey via One-Pot Operation; General
Procedure III (GP III) (Scheme 10)
1
1m; yield: 2.59 g (54%); ivory solid.
1
H NMR (400 MHz, DMSO-d ):  = 5.05 (br s, 2 H).
6
A stirred solution of 1,4-dimethoxyphthalazine (6a; 0.2 mmol, 1.0
equiv) and TCICA (0.8 mmol, 4.0 equiv) in 1,2-DCE (1.2 mL) was re-
fluxed in an oil bath for 24 h to form the monochlorinated intermedi-
ate 7. Then, a solution of either oxygenating or fluorinating reagent in
1
3
C NMR (100 MHz, DMSO-d ):  = 163.0, 136.2, 129.7, 120.3.
6
HRMS (EI): m/z [M] calcd for C H 79Br BrN O , C H 79Br Br N O ,
+
81
81
8
2
3
2
2
8
2
2
2
2
2
C8H279Br2 Br N O : 475.6829 (68.3%), 477.6809 (100.0%), 479.6788
81
2
2
2
1
,2-DCE was added at rt, and the reaction mixture was refluxed for an
(
(
65.3%); found: 475.6826 (65.5%), 477.6810 (100.0%), 479.6786
65.3%).
additional 24 h. After the mixture was cooled to down rt and concen-
1
trated under reduced pressure, the crude mixture was analyzed by H
NMR spectroscopy.
3
,6-Dimethoxypyridazine (12)30
A stirred solution of 3,6-dichloropyridazine (500 mg, 3.36 mmol, 1.0
equiv) and NaOMe (725 mg, 13.4 mmol, 4.0 equiv) in MeOH (17 mL)
was refluxed in an oil bath for 24 h. Then, the reaction mixture was
cooled to rt and concentrated under reduced pressure. The residue
was diluted with CH Cl and washed with H O (3 ×). The combined
mCPBA
Following GP III, 6a (38 mg, 0.20 mmol) and TCICA (190 mg, 0.80
mmol) in 1,2-DCE (1.2 mL) and then mCPBA (49 mg, 0.20 mmol, 1.0
equiv) in 1,2-DCE (0.9 mL) were reacted together. A partial loss of the
N-chlorine substituent in 7 was observed (7:15 = 68:32).
2
2
2
aqueous layers were extracted with CH Cl (2 ×). The combined or-
2
2
ganic layers were dried (anhyd MgSO ), filtered through a glass frit,
and concentrated under reduced pressure. The residue was purified
4
UHP/MeReO
3
by column chromatography (SiO , R = 0.28 in CH Cl /EtOAc 30:1) to
Following GP III, 6a (38 mg, 0.20 mmol) and TCICA (190 mg, 0.80
2
f
2
2
give 12 as a white solid; yield: 271 mg (58%).
mmol) in 1,2-DCE (1.2 mL) and then UHP (56 mg, 0.60 mmol, 3.0
1
equiv) and MeReO (5 mg, 0.02 mmol, 0.1 equiv) were reacted togeth-
er. A complete loss of the N-chlorine substituent in 7 was observed
H NMR (400 MHz, DMSO-d ):  = 7.19 (s, 2 H), 3.93 (s, 6 H).
3
6
(7:15 = 0:100).
4
-Chloro-3,6-dimethoxypyridazine (13) (Equation 1)
A stirred solution of 12 (140 mg, 1.00 mmol, 1.0 equiv), n-Bu NCl
4
CF CO H
3
3
(1.11 g, 4.00 mmol, 4.0 equiv), and TCICA (930 mg, 4.00 mmol, 4.0
A stock solution of CF CO H was prepared by UHP (94 mg, 1.0 mmol,
equiv) in MeCN (6 mL) was refluxed in an oil bath for 48 h. Then, the
reaction mixture was cooled to rt and concentrated under reduced
pressure. The residue was purified by column chromatography (SiO2,
R = 0.38 in CH Cl ) to give 13 as a white solid; yield: 50 mg (29%); mp
3
3
5
.0 equiv) and TFAA (153 L, 1.10 mmol, 5.5 equiv) in 1,2-DCE (1.0
3
1
mL). Following GP III, 6a (38 mg, 0.20 mmol) and TCICA (190 mg,
.80 mmol) in 1,2-DCE (1.2 mL) and then a portion of the CF CO H
0
3
3
f
2
2
solution (0.5 mL, 0.5 mmol, 2.5 equiv) and Na CO (150 mg, 1.4 mmol,
79.5–83.9 °C.
2
3
7
.0 equiv) were reacted together. Approximately 12% conversion to 8a
1
H NMR (400 MHz, CDCl ):  = 7.05 (s, 1 H), 4.13 (s, 3 H), 4.05 (s, 3 H).
3
was observed.
1
3
C NMR (100 MHz, CDCl ):  = 162.1, 158.3, 130.0, 120.4, 55.7, 55.2.
3
©
2020. Thieme. All rights reserved. Synthesis 2021, 53, 1760–1770

1769
Synthesis
J. K. Im et al.
Paper
NFSI
Supporting Information
Following GP III, 6a (38 mg, 0.20 mmol) and TCICA (190 mg, 0.80
mmol) in 1,2-DCE (1.2 mL) and then NFSI (63 mg, 0.20 mmol, 1.0
equiv) in 1,2-DCE (0.9 mL) were reacted together. No reaction was ob-
served.
Supporting information for this article is available online at
https://doi.org/10.1055/s-0040-1706639. Su
p
p
orting Inform atio
n
Su
p
p
orting Inform atio
n
References
4-Methoxyphthalazin-1(2H)-one (15)
1
H NMR (400 MHz, CDCl ):  = 9.21 (s, 1 H), 8.41–8.38 (m, 1 H), 8.01–
.99 (m, 1 H), 7.85–7.78 (m, 2 H), 3.99 (s, 3 H).
(1) Reviews: (a) Joshi, U.; Pipelier, M.; Naud, S.; Dubreuil, D. Curr.
Org. Chem. 2005, 9, 261. (b) Csende, F.; Miklós, F.; Porkoláb, A.
Curr. Org. Chem. 2010, 14, 745.
3
7
HRMS (ESI/Q-TOF): m/z [M + H]+ calcd for C H N O : 177.0664;
found: 177.0659.
9
9
2
2
(
2) Chemical reduction: (a) Thompson, T. W. J. Chem. Soc., Chem.
Commun. 1968, 532. (b) Longridge, J. L.; Thompson, T. W.
J. Chem. Soc. C 1970, 1658. (c) Bach, N. J.; Kornfeld, E. C.; Jones,
N. D.; Chaney, M. O.; Dorman, D. E.; Paschal, J. W.; Clemens, J. A.;
Smalstig, E. B. J. Med. Chem. 1980, 23, 481. (d) Gewald, K.; Hain,
U. Synthesis 1984, 62. (e) Somei, M.; Inoue, S.; Tokutake, S.;
Yamada, F.; Kaneko, C. Chem. Pharm. Bull. 1981, 29, 726.
8
2
-Chloro-4-methoxyphthalazin-1-one (7) (Table 3)
A stirred solution of 1,4-dimethoxyphthalazine (6a; 151 mg, 0.79
mmol, 1.0 equiv) and TCICA (369 mg, 1.59 mmol, 2.0 equiv) in 1,2-
DCE (4.8 mL) was refluxed in an oil bath for 24 h. Then, the reaction
mixture was cooled to rt and concentrated under reduced pressure.
The residue was purified by column chromatography (SiO , R = 0.16
(f) Bruce, J. M. J. Chem. Soc. 1959, 2366. (g) Sugimoto, A.; Tanaka,
2
f
H.; Eguchi, Y.; Ito, S.; Takashima, Y.; Ishikawa, M. J. Med. Chem.
1984, 27, 1300. (h) Brown, G. R.; Foubister, A. J.; Wright, B.
J. Chem. Soc., Chem. Commun. 1984, 1373. (i) Boger, D. L.;
Coleman, R. S.; Panek, J. S.; Yohannes, D. J. Org. Chem. 1984, 49,
4405. (j) A representative example of total synthesis: Oakdale, J.
S.; Boger, D. L. Org. Lett. 2010, 12, 1132. (k) Silva, P. J. J. Org.
Chem. 2012, 77, 4653.
in CH Cl /hexanes 1:1) to give 7 as a white solid; yield: 136 mg (82%);
mp 156.9–162.2 °C.
2
2
1
H NMR (400 MHz, CDCl ):  = 8.42–8.38 (m, 1 H), 8.00–7.96 (m, 1 H),
3
7
.86–7.79 (m, 2 H), 4.02 (s, 3 H).
Electrophilic Promotor Survey with Isolated 7; General Procedure
IV (GP IV) (Table 3)
(
(
3) Electrochemical reduction: (a) Lund, H.; Lunde, P. Acta Chem.
Scand. 1967, 21, 1067. (b) Lund, H. Discuss. Faraday Soc. 1968,
A stirred solution of 2-chloro-4-methoxyphthalazin-1-one (7; 0.5
mmol, 1.0 equiv) and required reagents (1.0 mmol, 2.0 equiv each) in
MeCN (3.0 mL) was refluxed in an oil bath under argon for 24 h. The
reaction mixture was cooled to rt and concentrated under reduced
pressure. The residue was purified by column chromatography (SiO2,
R = 0.27 in CH Cl /hexanes 1:1) to give N-chlorophthalimide (8a) as a
4
5, 193. (c) Manh, G. T.; Hazard, R.; Pradère, J. P.; Tallec, A.;
Raoult, E.; Dubreuil, D. Tetrahedron Lett. 2000, 41, 647.
d) Manh, G. T.; Hazard, R.; Tallec, A.; Pradère, J. P.; Dubreuil, D.;
(
Thiam, M.; Toupet, L. Electrochim. Acta 2002, 47, 2833.
4) Photochemical ring contraction: (a) Hatch, C. E.; Johnson, P. Y.
Tetrahedron Lett. 1974, 15, 2719. (b) Johnson, P. Y.; Hatch, C. E.
III. J. Org. Chem. 1975, 40, 909. (c) Johnson, P. Y.; Hatch, C. E. III.
J. Chem. Soc., Chem. Commun. 1975, 725. (d) Johnson, P. Y.;
Schmuff, N. R.; Hatch, C. E. Tetrahedron Lett. 1975, 4089.
f
2
2
white solid.
K S O
8
2
2
Following GP IV, 7 (105 mg, 0.50 mmol) and K S O (270 mg, 1.00
mmol) were reacted. Ring contraction product 8a was not detected.
2
2
8
(5) Thermal ring contraction: Galanin, N. E.; Yakubov, L. A.;
Shaposhnikov, G. P. Macroheterocycles 2010, 3, 41.
(
6) Electrophile-promoted, oxidative ring contraction is known for
triazine derivatives: Gazieva, G. A.; Karpova, T. B.; Nechaeva, T.
V.; Kravchenko, A. N. Russ. Chem. Bull. Int. Ed. 2016, 65, 2172.
7) (a) Parrick, J.; Ragunathan, R. J. Chem. Soc., Perkin Trans. 1 1993,
TCICA and K S O
8
2
2
Following GP IV, 7 (105 mg, 0.50 mmol), TCICA (232 mg, 1.00 mmol),
and K S O (270 mg, 1.00 mmol) were reacted to give 8a; yield: 13 mg
(
2
2
8
2
1
11. (b) Omote, Y.; Yamamoto, H.; Sugiyama, N. J. Chem. Soc. D
970, 914.
(14%).
(
8) Im, J. K.; Yang, B.; Jeong, I.; Choi, J.-H.; Chung, W.-j. Tetrahedron
Lett. 2020, 61, 152048.
TCICA and n-Bu NCl
4
Following GP IV, 7 (105 mg, 0.50 mmol), TCICA (232 mg, 1.00 mmol),
(9) Shiri, A.; Khoramabadizad, A. Synthesis 2009, 2797.
and n-Bu NCl (278 mg, 1.00 mmol) were reacted to give 8a; yield: 41
4
(
(
(
10) Clement, R. A. J. Org. Chem. 1960, 25, 1724.
11) Kealy, T. J. J. Am. Chem. Soc. 1962, 84, 966.
12) Ineffective reagents include N-chlorosuccinimide (NCS), 1,3-
mg (45%).
TCICA
dichloro-5,5-dimethylhydantion
(DCDMH),
Palau’Chlor,
Following GP IV, 7 (105 mg, 0.50 mmol) and TCICA (232 mg, 1.00
mmol) were reacted to give 8a ; yield: 24 mg (26%).
2,3,4,5,6,6-hexachlorocyclohexa-2,4-dien-1-one (HCCO), N-bro-
mosuccinimide (NBS), N-fluorobenzenesulfonimide (NFSI), and
TM
Selectfluor
.
(
13) Other soluble chloride such as Me NCl exhibited comparable
4
promoting reactivity. In contrast, insoluble chloride such as LiCl
was completely ineffective.
Funding Information
This research was supported by Basic Science Research Program
through the National Research Foundation of Korea (NRF) funded by
the Ministry of Science & ICT (NRF-2020R1F1A1076028) and by ‘GIST
(14) Other common reductive workup procedures with aq Na S O
2 2 3
or aq NaHSO
resulted in less quantitative mass recovery.
3
1
(15) An approximate H NMR yield of 8a was measured with 1,1,2,2-
Research Institute (GRI)’ grant funded by the GIST in 2020.Nati
o
nal ResearchFo
u
n
datio
n
o
f
K
orea(N
R
F-2
0
2
0
R
1F1
A
1
0
7
6
0
2
8)G
w
a
n
g
juInsitute
o
f
Scie
n
cean
d
T
e
ch
n
o
lo
g
y
()
tetrachloroethane as an internal standard. From the reaction of
6
a with TCICA (4 equiv) and n-Bu NCl (4 equiv), ca. 42% yield of
4
©
2020. Thieme. All rights reserved. Synthesis 2021, 53, 1760–1770

1770
Synthesis
J. K. Im et al.
Paper
8
a was observed in the crude mixture, and a small amount of 7
(24) (a) Schmidt, M. W.; Baldridge, K. K.; Boatz, J. A.; Elbert, S. T.;
Gordon, M. S.; Jensen, J. H.; Koseki, S.; Matsunaga, N.; Nguyen, K.
A.; Su, S.; Windus, T. L.; Dupuis, M.; Montgomery, J. A. Jr.
J. Comput. Chem. 1993, 14, 1347. (b) Gordon, M. S.; Schmidt, M.
W. Advances in Electronic Structure Theory: GAMESS a Decade
Later, In Theory and Applications of Computational Chemistry:
The First Forty Years; Dykstra, C. E.; Frenking, G.; Kim, K. S.;
Scuseria, G. E., Ed.; Elsevier: Amsterdam, 2005, 1167–1189.
(c) The General Atomic and Molecular Electronic Structure
System. https://www.msg.chem.iastate.edu/gamess/
(25) The initial input geometry was obtained by Molecular Orbital
PACkage 2016 Stewart, J. J. P. Stewart Computational Chemistry.
Colorado Springs (USA) 2016, http://openmopac.net/
(26) 3-D structures were obtained using CYLview 1.0b Legault, C. Y.
Sherbrooke (Canada) Université de Sherbrooke 2009,
http://www.cylview.org.
was detected as the only major side product. This result indi-
cates the moderate ring contraction efficiency as well as the
intractable decomposition of 6a during the reaction. See the SI
for details.
(
16) Prabhakar, V.; Babu, K. S.; Ravindranath, L. K.; Lakshmireddy, I.;
Latha, J. Heterocycl. Lett. 2016, 6, 643.
17) Yanilkin, V. V.; Buzykin, B. I.; Morozov, V. I.; Nastapova, N. V.;
Maksimyuk, N. I.; Eliseenkova, R. M. Russ. J. Gen. Chem. 2001, 71,
(
1636.
(
18) (a) Kende, A. S. Org. React. 1960, 11, 261. (b) Akhrem, A. A.;
Ustynyuk, T. K.; Titov, Y. A. Russ. Chem. Rev. 1970, 39, 732.
(c) Chenier, P. J. J. Chem. Educ. 1978, 55, 286. (d) Guijarro, D.;
Yus, M. Curr. Org. Chem. 2005, 9, 1713.
(
19) (a) Rees, C. W.; Sale, A. A. J. Chem. Soc., Perkin Trans. 1 1973, 545.
(b) Laakso, P. V.; Robinson, R.; Vandrewala, H. P. Tetrahedron
1
957, 1, 103.
(27) (a) Gobbi, A.; Frenking, G. J. Chem. Soc., Chem. Commun. 1993,
1162. (b) Bronzolino, N.; Grandinetti, F. J. Mol. Struct.: THEO-
CHEM 2003, 635, 221.
(28) The addition of NaBH4 as an EtOH solution led to vigorous
quenching and lower yields.
(29) Najda-Bernatowicz, A.; Łebska, M.; Orzeszko, A.; Kopańska, K.;
Krzywińska, E.; Muszyńska, G.; Bretner, M. Bioorg. Med. Chem.
2009, 17, 1573.
(
(
(
20) The activation free energy for the N–N bond cleavage in i was
estimated to be >70 kcal/mol by DFT calculation (see the SI).
21) (a) Zhao, Y.; Truhlar, D. G. Theor. Chem. Acc. 2008, 120, 215.
(
b) Zhao, Y.; Truhlar, D. G. Acc. Chem. Res. 2008, 41, 157.
22) (a) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. J. Chem. Phys.
010, 132, 154104. (b) Peverati, R.; Baldridge, K. K. J. Chem.
2
Theory Comput. 2008, 4, 2030.
(
23) Marenich, A. V.; Cramer, C. J.; Truhlar, D. G. J. Phys. Chem. B
(30) Clapham, K. M.; Batsanov, A. S.; Greenwood, R. D. R.; Bryce, M.
R.; Smith, A. E.; Tarbit, B. J. Org. Chem. 2008, 73, 2176.
2
009, 113, 6378.
(31) Chuang, K. V.; Xu, C.; Reisman, S. E. Science 2016, 353, 912.
©
2020. Thieme. All rights reserved. Synthesis 2021, 53, 1760–1770