Decarboxylative Bromination of Heteroarenes: Initial Mechanistic Insights

Article abstract of DOI:10.1055/s-0040-1707901

After an initial report from our laboratory describing metal-free decarboxylative halogenation of various azaheteroarenes, we set out to investigate the possible mechanism by which this chemistry occurs. Evidence from this mechanistic investigation sugges

Full text of DOI:10.1055/s-0040-1707901

SYNLETT
0
9
3
6
-
5
2
1
4
1
4
3
7
-
2
0
9
6
©
Georg Thieme Verlag Stuttgart · New York
2020, 31, A–E
letter
A
en
Synlett
P. R. Patel et al.
Letter
Decarboxylative Bromination of Heteroarenes:
Initial Mechanistic Insights
Pritesh R. Patela
Scott H. Hendersonb
O
OH
Br
Mark S. Roec
Mark A. Honey*d
NBS, hυ
N
+ CO2 + HBr
N
0
0
0
0-
0
0
0
1-7
2
7
2-4
7
6
X
DMF, rt, 16 h
N
N
H
H
a
School of Pharmacy, Faculty of Science and Engineering, Uni-
versity of Wolverhampton, Wolverhampton, WV1 1LY, UK
Sussex Drug Discovery Centre, University of Sussex,
Brighton, BN1 9RH, UK
School of Life Sciences, University of Sussex,
Brighton, BN1 9RH, UK
School of Science, Faculty of Engineering and Science,
University of Greenwich, Chatham, Kent, ME4 4TB, UK
m.a.honey@gre.ac.uk
- substrate activation of NBS
- light sensitive
- potential radical pathway
b
c
d
Received: 11.06.2020
Accepted after revision: 30.06.2020
Published online: 17.07.2020
high reaction temperatures.12,13 We recently reported a
metal-free selective decarboxylative halogenation of het-
eroaromatic compounds under mild conditions (Scheme
DOI: 10.1055/s-0040-1707901; Art ID: St-2020-d0343-l
14
1). We found that indole-, indazole-, azaindole-, and
Abstract After an initial report from our laboratory describing metal-
free decarboxylative halogenation of various azaheteroarenes, we set
out to investigate the possible mechanism by which this chemistry oc-
curs. Evidence from this mechanistic investigation suggests that this
chemistry occurs via a radical pathway, with H NMR studies suggesting
that the acidic substrates activate NBS.
azaindazolecarboxylic acids underwent spontaneous de-
carboxylative bromination in the absence of any catalyst or
initiator. To exploit this chemistry further, we needed to
understand the mechanism by which this reaction occurs.
Previous reports by the Larrosa group indicated that, under
their basic conditions with aromatic substrates, decarbox-
1
Key words decarboxylation, halogenation, reaction mechanism, bro-
mosuccinimide, azahetarenes
ylative halogenation probably proceeds by a concerted
mechanism.1
2,13
As we believed it was possible for the
mechanism of our chemistry to proceed by a radical path-
way or by a concerted pathway, we set out to determine ex-
perimentally which pathway was the most likely.
The manipulation of heterocyclic compounds to modu-
late or change their reactivity is of fundamental importance
1,2
to a vast range of chemical industries. To be able to intro-
duce new functionality selectively under mild conditions is
something that numerous academic and industrial research
groups strive to achieve. The ability to halogenate aromat-
ic/heteroaromatic compounds selectively can be an essen-
tial element in the synthesis of pharmaceutically active
O
OH
Br
R
R
NBS (1 equiv)
DMF, rt, 16 h
Y
Y
N
N
X
X
H
H
Scheme 1 Decarboxylative halogenation of azaheterocycles14
3–5
molecules within drug-discovery programmes
or it can
create handles for easy manipulation of functionality when
investigating structure–activity relationships during lead
optimization.
Having previously established that the decarboxylative
bromination of indazolecarboxylic acids occurs sponta-
neously in DMF in the presence of NBS,14 we next sought to
investigate whether this reaction was able to proceed in a
range of solvents (Table 1). We found that the reaction is
tolerant of polar protic and polar aprotic solvents, although
lower yields are obtained for solvents in which the solubili-
ty of the starting material appears to be poor. Mixed solvent
systems were also investigated, but did not provide any im-
provement on the standard conditions employing DMF. Be-
cause DMF still appeared to be the best solvent for this
chemistry, we used it in all our initial mechanistic studies.
6,7
A major method employed to introduce halogens selec-
tively onto aromatic and aliphatic substrates employs de-
carboxylative halogenation. This area of research has a rich
history, but the greatest advances have mainly been
8
achieved by using alkanoic acids as substrates. There are
several reported methods for performing decarboxylative
halogenation on aromatic substrates, but they often rely on
metal catalysts, require high reaction temperatures, or are
9–11
unselective.
Recent reports by Larrosa and co-workers
have advanced this area of research, but their method re-
quires stoichiometric quantities of a base and relatively
©
2020. Thieme. All rights reserved. Synlett 2020, 31, A–E
Georg Thieme Verlag KG, Rüdigerstraße 14, 70469 Stuttgart, Germany

B
Synlett
P. R. Patel et al.
Letter
Table 1 Investigation of Solvent15
nism further, we investigated whether free radicals are in-
volved in this chemistry by adding the radical quencher
O
2,2,6,6-tetramethyl-1-piperidinyloxyl (TEMPO) to our reac-
OH
Br
N
tion mixture, both in the presence of ambient light and in
darkness. Notably, even in the presence of the radical
quencher TEMPO, the initial instant colour change of the re-
action mixture was observed, and therefore the suspected
initial bromine formation still occurred. When TEMPO was
present in reactions conducted in ambient light and in
darkness (entries 2 and 4), the reaction failed to proceed,
with only the starting material being recovered. The sup-
pression of the reaction in the presence of the radical
quencher suggests a fundamental involvement of free radi-
cals in the progress of this reaction. Having established that
the presence of TEMPO in the reaction mixture for the
chemistry with NBS stopped the reaction from proceeding,
the effect of TEMPO on the progress of the reaction in the
presence of NCS was investigated. Although the isolated
yield of 2-chloro-1H-indazole (entry 5) was low, the reac-
tion was able to proceed under the standard conditions.
However, as with the case for the chemistry with NBS, the
addition of TEMPO appeared to prevent any reaction from
occurring (entry 6).
NBS (1 equiv)
solvent, rt, 16 h
N
N
N
H
H
Entry
Solvent
Yield (%)
116
2
DMF
67
EtOAc
56
3a
4
CH
Cl
17
2
2
MeOH
37
5a
6a
7a
8a
9
H
O
0
2
MeCN
37
toluene
23
Et
CHCl
MeOH–H
CH Cl –EtOAc
2
O
trace
trace
trace
trace
3
10
2
O
11
2
2
a
Solubility issues were observed.
Having established that 1H-indazole-3-carboxylic acids
Table 2 Effect of TEMPO on the Progress of the Reaction
undergo spontaneous decarboxylative bromination in the
presence of NBS at room temperature, we sought to investi-
gate the mechanism by which this transformation occurs.
Our general procedure for this chemistry started with the
addition of the acidic substrate to the reaction solvent.
Once the substrate had dissolved, often giving a colourless
solution, NBS was added. It was at this point that we ob-
served an instant change in the colour of the reaction mix-
ture: the mixture typically turned yellow, brown, or red.
O
OH
X
NXS (1 equiv)
N
N
N
DMF, rt, 16 h
conditions
N
H
H
a
Entry Halogen Conditions
TEMPO (equiv)
Yield (%)
1
2
3
4
5
Br
ambient light
0
2
0
2
0
3
65
trace
38
0
1
Our belief, supported by H NMR spectroscopic evidence
Br
ambient light
detailed later, is that the acidic substrate activates NBS to-
wards dissociation, and so quickly forms a low concentra-
tion of bromine. As bromine can react by an electrophilic or
a radical pathway, our next thought was to investigate
whether the bromine in the reaction mixture reacts with
the substrate by either of these pathways. Our initial efforts
focused on the effect of light on our reaction system, as bro-
mine is known to be activated by the presence of light. En-
tries 1 and 3 in Table 2 show the isolated yields of 3-bromo-
Br
foil-covered vessel
foil-covered vessel
ambient light
Br
Cl
Cl
12
0
6
ambient light
a
Isolated yield.
In our previous report,14 we described the utility of this
reaction for several azaheterocycles, including the three ex-
amples shown in Scheme 2. We found that, in general,
more-electron-deficient heterocycles showed greater reac-
tivity, and good yields of halogenated azaindoles and azain-
dazoles were obtained. Although the relatively electron-de-
ficient indazole-3-carboxylic acid readily underwent halo-
decarboxylation, the reaction failed to proceed when the
indazole was N-methylated. This raised the question of
whether a free N–H is needed for this reaction to proceed or
the failure to react is the result of an electronic effect.
16
1
H-indazole from reactions performed in DMF at room
temperature in the presence of ambient light and in dark-
ness, respectively. All experiments were conducted with
new reaction vials and stirrer bars to avoid the introduction
of any trace metals from previous chemistry. It is clear that
the presence of light significantly increases the overall yield
of the brominated product compared with that of the reac-
tion conducted in the dark. It is known that in the presence
of light bromine typically undergoes homolytic cleavage to
17
form bromine radicals. Therefore, to probe the mecha-
©
2020. Thieme. All rights reserved. Synlett 2020, 31, A–E

C
Synlett
P. R. Patel et al.
Letter
ity. Decarboxylative bromination of substituted thiophenes
also proved successful (Scheme 4), although the products
were not isolated as they were obtained only transiently
and reacted in situ as part of a drug-discovery programme.
Unfortunately, the use of unsubstituted thiophene-2-car-
boxylic acid proved unsuccessful, with no product being
observed. Although disappointing, this result also adds
weight to the argument that electron deficiency of the aro-
matic ring, and the ability to stabilize a free radical facili-
tates this reaction.
Br
N
Br
N
N
H
NBS (1 equiv)
DMF, rt, 16 h
(90%)
Heterocyclic acid
N
Br
(0%)
N
N
N
H
(74%)
O
Scheme 2 Current scope of method14
NBS (1.8 equiv)
Br
S
S
DMF, rt, 16 h
OH
N
N
As part of an ongoing project, we subjected pyrazo-
lo[1,5-b]pyridazine-3-carboxylic acid to our standard con-
ditions (Scheme 3), and we found that the reaction pro-
ceeded smoothly to give the desired brominated product in
excellent yield, with the structure of the product being con-
firmed by single-crystal X-ray crystallography (Figure 1).18
Although the heterocyclic core was slightly different, this
reaction proceeded in the absence of a free N–H group, sug-
gesting that its presence might not be important for certain
substrates in the mechanism of this reaction.
5
0% conversion (LC/MS)
O
NBS (1.8 equiv)
DMF, rt, 16 h
Br
S
S
OH
25% conversion (LC/MS)
O
NBS (1.8 equiv)
DMF, rt, 16 h
Br
S
S
OH
Scheme 4 Decarboxylative bromination of thiophenes
O
OH
Br
To investigate the possible mechanism further, the stan-
dard reaction was studied by H NMR spectroscopy in deu-
1
NBS (2 equiv)
terated DMF. NBS has been known to react with numerous
solvents, so we first investigated the possibility of activa-
N
DMF, rt, 16 h
N
N
N
19
N
N
(92%)
tion of NBS by DMF-d . NBS was dissolved in DMF-d and
7
7
left at room temperature for 30 minutes. Had NBS reacted
with DMF, the resultant H NMR of the reaction mixture
Scheme 3 Decarboxylative bromination of pyrazolo[1,5-b]pyridazine-
1
3-carboxylic acid
would have shown a mixture of compounds. However, the
1
H NMR showed only one peak for the four equivalent ali-
phatic protons of NBS, and no evidence could be found for
the existence of the N–H corresponding to the formation of
succinimide. This suggests that, although DMF acts as a sol-
vent and interacts with NBS in this regard, it does not ap-
pear to alter its reactivity significantly. It has been reported
20
that N-halosuccinimides can be activated by both Brønsted
and Lewis acids,21 the lone pair of electrons on the car-
bonyl oxygen acting as a Lewis base. Protonation of the car-
bonyl oxygen results in a weaker N–Br bond, and so in-
creases the rate of dissociation. As NBS appeared to be
relatively stable in DMF, we explored potential activation of
1
NBS by the substrate itself through protonation. The
H
NMR spectrum for 1H-indazole-3-carboxylic acid in DMF
clearly showed each proton on the aromatic ring, as well as
a broad singlet corresponding to a mixture of the carboxylic
acid, indazole N–H, and traces of water. When the acid was
combined with NBS, there was a clear chemical shift of the
aliphatic protons of the NBS, as well as the appearance of a
new aliphatic peak associated with the byproduct, succin-
imide. The observed chemical shift for NBS suggested that
Figure 1 X-ray crystal structure of 3-bromopyrazolo[1,5-b]pyri-
dazine18
As part of a continuing medicinal chemistry project, de-
carboxylative bromination has formed an integral step in
the synthesis of compounds with potential biological activ-
©
2020. Thieme. All rights reserved. Synlett 2020, 31, A–E

D
Synlett
P. R. Patel et al.
Letter
Br
O
H
O
H+ catalyst
O
(from substrate)
hυ
2
x
N
Br
Br Br
N
O
O
N
H
NH
O
CO2 + HBr
Br Br
Br
N
N
N
N
H
H
Scheme 5 Possible mechanism through a radical pathway
the indazole interacts with NBS, most likely through the
carboxylic acid O–H and the carbonyl groups of NBS.
Acknowledgment
We would like to thank Surila Darbar, David Townrow, and Keith
Holding for their help with LC/MS, NMR, and managing the laborato-
ry, respectively, and Dr. Ryan West for helpful discussions.
1
With the H NMR evidence suggesting that the substrate
is involved in the activation of NBS towards dissociation,
and the use of a radical quencher stopping the reaction,
even in the apparent presence of bromine, we believe it is
highly likely that the reaction proceeds by a radical path-
way. Although not definitive, a possible mechanism is
shown in Scheme 5. The reaction is initiated by protonation
of NBS by the indazolecarboxylic acid, which helps to acti-
vate the N–Br bond towards homolytic cleavage to form
bromine. The bromine then undergoes light-mediated
and/or thermal homolytic cleavage to produce bromine
radicals. A bromine radical abstracts the acidic proton of
the indazolecarboxylic acid, and subsequent decarboxyl-
ation produces an indazole free radical that reacts with an-
other equivalent of bromine to form the brominated product.
Having previously established that azaheteroarene car-
boxylic acids undergo metal- and catalyst-free decarboxyl-
ative halogenation, we sought to investigate the mechanism
by which the reaction occurs. Although we cannot com-
pletely rule out alternative mechanisms, our studies sug-
gest that the reaction proceeds by a radical pathway. The
yield of the reaction is higher when the chemistry is con-
ducted in the presence of light, and the reaction fails to pro-
ceed in the presence of the radical quencher TEMPO. We
believe that the acidic reaction environment influences the
reactivity of NBS, and so catalyses the formation of bromine
that subsequently reacts through a radical pathway to form
the decarboxylative brominated product.
Supporting Information
Supporting information for this article is available online at
https://doi.org/10.1055/s-0040-1707901.
S
u
p
p
orti
n
g Inform ati
o
n
S
u
p
p
orti
n
g Inform ati
o
n
References and Notes
(1) Pozharskii, A. F.; Soldatenkov, A. T.; Katritzky, A. R. Heterocycles
in Life and Society: An Introduction to Heterocyclic Chemistry,
Biochemistry and Applications, 2nd ed; Wiley: Chichester, 2011.
(2) Heterocyclic Chemistry in Drug Discovery; Li, J. J., Ed.; Wiley:
Hoboken, 2013.
(
(
3) Pathak, T. P.; Miller, S. J. J. Am. Chem. Soc. 2012, 134, 6120.
4) Chung, W.-J.; Vanderwal, C. D. Angew. Chem. Int. Ed. 2016, 55,
4396.
(
5) Bacsa, I.; Herman, B. E.; Jójárt, R.; Herman, K. S.; Wölfling, J.;
Schneider, G.; Varga, M.; Tömböly, C.; Rižne, T. L.; Szécsi, M.;
Mernyák, E. J. Enzyme Inhib. Med. Chem. 2018, 33, 1271.
(6) Mallinger, A.; Schiemann, L.; Rink, C.; Sejberg, J.; Honey, M. A.;
Czodrowski, P.; Stubbs, M.; Poeschke, O.; Busch, M.; Schneider,
R.; Schwarz, D.; Musil, D.; Burke, R.; Urbahns, K.; Workman, P.;
Wienke, D.; Clarke, P. A.; Raynaud, F. I.; Eccles, S. A.; Esdar, C.;
Rohdich, F.; Blagg, J. ACS Med. Chem. Lett. 2016, 7, 573.
(7) Li, F.; Hu, Y.; Wang, Y.; Ma, C.; Wang, J. J. Med. Chem. 2017, 60,
1580.
(8) Li J. J.; In Name Reactions; Springer, Berlin, 2009; 298; DOI:
10.1007/978-3-642-01053-8_131
(
9) Fu, Z.; Li, Z.; Song, Y.; Yang, R.; Liu, Y.; Cai, H. J. Org. Chem. 2016,
1, 2794.
(10) Jiang, Q.; Li, H.; Zhang, X.; Xu, B.; Su, W. Org. Lett. 2018, 20,
424.
8
Funding Information
2
Biotechnology
BB/L017105/1).
and
Biological
Sciences
Research
Council
(
B
i
ote
c
h
n
o
l
o
g
y
a
n
d
B
i
o
l
o
g
i
c
a
lS
c
i
e
n
c
es
R
esearc
h
C
o
u
n
c
i
l (BB/L
0
1
7
1
0
5/1)
©
2020. Thieme. All rights reserved. Synlett 2020, 31, A–E

E
Synlett
P. R. Patel et al.
Letter
(
11) Hammamoto, H.; Umemoto, H.; Umemoto, M.; Ohta, C.; Fujita,
E.; Nakamura, A.; Maegawa, T.; Miki, Y. Heterocycles 2015, 91,
61.
(16) 3-Bromo-1H-indazole (Table 1, Entry 1)
14
Colourless solid; yield: 36 mg (67%); mp 145–150 °C (Lit. 142–
1
5
143 °C). H NMR (400 MHz, CDCl ):  = 1.57 (br s, 1 H, NH), 7.58
3
(
(
(
(
12) Perry, G. J. P.; Quibell, J. M.; Panigrahi, A.; Larrosa, I. J. Am. Chem.
(m, 2 H, ArH), 7.39 (m, 1 H, ArH), 7.17 (m, 1 H, ArH). 13C NMR
Soc. 2017, 139, 11527.
(100 MHz, CDCl ):  = 141.3, 128.3, 123.1, 122.6, 121.9, 120.2,
3
+
13) Quibell, J. M.; Perry, G. J. P.; Cannas, D. M.; Larrosa, I. Chem. Sci.
110.7. HRMS (ESI): m/z [M + H] calcd for C H BrN : 196.9709;
7
6
2
2018, 9, 3860.
found: 196.9717.
14) Henderson, S. H.; West, R. A.; Ward, S. E.; Honey, M. A. R. Soc.
Open Sci. 2019, 5, 180333.
15) Bromination of Azahetarenecarboxylic Acids; General Proce-
dure
(17) Tanner, D. D.; Ruo, T. C. S.; Takiguchi, H.; Guillaume, A.; Reed, D.
W.; Setiloane, B. P.; Tan, S. L.; Meintzer, C. P. J. Org. Chem. 1983,
48, 2743.
(18) CCDC 1913779 contains the supplementary crystallographic
data for 3-bromopyrazolo[1,5-b]pyridazine (Figure 1). The data
can be obtained free of charge from The Cambridge Crystallo-
graphic Data Centre via www.ccdc.cam.ac.uk/getstructures.
(19) Shimizu, S.; Imamura, Y.; Ueki, T. Org. Process Res. Dev. 2014, 18,
354.
(20) Prakash, G. K. S.; Mathew, T.; Hoole, D.; Esteves, P. M.; Wang, Q.;
Rasul, G.; Olah, G. A. J. Am. Chem. Soc. 2004, 126, 15770.
(21) Xue, H.; Tan, H.; Wei, D.; Wei, Y.; Lin, S.; Liang, F.; Zhao, B. RSC
Adv. 2013, 3, 5382.
The appropriate carboxylic acid (0.280 mmol) was dissolved in
the solvent (1.5 mL) and NBS (0.280 mmol) was added. The
vessel was sealed and the mixture was stirred at rt for 16 h. H O
2
(
10 mL) was added and the organic material was extracted with
EtOAc (2 × 4 mL). The organic layers were combined, dried
MgSO ), filtered, and concentrated under reduced pressure.
(
4
The crude residue was purified by chromatography (silica gel,
EtOAc–PE or CH Cl –MeOH).
2
2
©
2020. Thieme. All rights reserved. Synlett 2020, 31, A–E