Radical alkylation of N-alkyl 1,2,4-triazoles

Article abstract of DOI:10.1016/S0040-4039(01)01518-0

Reaction of 1-N-alkyl triazoles with an alkyl radical generated from the corresponding carboxylic acid, affords the triazole ring alkylated selectively in the 5 position. Secondary radicals perform best in the intermolecular reaction, while primary radica

Full text of DOI:10.1016/S0040-4039(01)01518-0

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 7353–7355
Radical alkylation of N-alkyl 1,2,4-triazoles
Karl B. Hansen,* Shawn A. Springfield, Richard Desmond, Paul N. Devine,
Edward J. J. Grabowski and Paul J. Reider
Department of Process Research, Merck and Co., Rahway, NJ, USA
Received 1 June 2001; accepted 19 July 2001
Abstract—Reaction of 1-N-alkyl triazoles with an alkyl radical generated from the corresponding carboxylic acid, affords the
triazole ring alkylated selectively in the 5 position. Secondary radicals perform best in the intermolecular reaction, while primary
radicals alkylate the triazole in an intramolecular fashion. © 2001 Elsevier Science Ltd. All rights reserved.
We recently wished to prepare alkylated derivatives of
1-N-methyl triazole 1 as part of our research efforts
(Scheme 1). N-Alkyl triazoles are typically alkylated by
deprotonation of the triazole ring, followed by reaction
with electrophiles such as ketones, amides or primary
iodides.1 A complementary approach to formation of
the carbonꢀtriazole bond would be the addition of alkyl
radicals directly to the triazole ring. Carbon radicals
generated from carboxylic acids via silver catalysis have
been shown to alkylate many six-membered nitrogen-
containing ring systems under aqueous acidic condi-
tions,2 however, there are fewer reports of the
alkylation of five-membered nitrogen heterocycles using
this method.3 Herein we report the extension of this
methodology to the alkylation of 1-N-alkyl 1,2,4-
triazoles.
in the isolation of a mixture of 5-cyclobutyl triazole 2,
unreacted starting material and a small amount of
dicyclobutyl triazole 3 (Scheme 1). The HPLC assay
yield of 2 was 33% and no triazole monoalkylated in
the 3 position was isolated from the reaction. Increas-
ing the amount of oxidant and carboxylic acid in the
reaction did not significantly enhance the conversion of
the methyl triazole substrate. However, by controlling
the addition rate of the carboxylic acid by dissolving it
as its sodium salt in the oxidant solution and slowly
adding this mixture to the reaction, the yield of 2 was
raised to 62% with 10–15% of 3 as the major by-
product along with unreacted 1.
The scope of the reaction was explored by examining
other alkyl carboxylic acids under the conditions out-
lined above. The results are summarized in Table 1.6
The desired products could be separated from dialkyl-
ated material by vacuum distillation of the crude reac-
tion mixture. However, they were generally isolated as
a mixture with dialkylated triazole (>90:10 ratio of
monoalkylated to dialkylated by 1H NMR) after
removing unreacted starting material by elution
through a plug of silica gel. Secondary carboxylic acids,
In order to assess the feasibility of this strategy, we
subjected 14 to conditions we had developed for the
alkylation of 3,6-dichloropyridazine, which were
modified from those originally reported by Samartoni.5
Treatment of a trifluoroacetic acid (TFA)/water solu-
tion of methyltriazole, 1.1 equiv. cyclobutanecarboxylic
acid and 10 mol% AgNO3 at 70°C with an aqueous
solution of 1.5 equiv. (NH4)2S2O8 resulted after workup
CO2H
Me
Me
Me
N
N
N
N
N
N
N
N
+
AgNO3, (NH4)2S2O8
TFA/H2O
N
1
2
3
Scheme 1.
* Corresponding author.
0040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)01518-0

7354
K. B. Hansen et al. / Tetrahedron Letters 42 (2001) 7353–7355
Table 1.
R2
R2CO2H
R1
R1
N
N
N
N
N
N
AgNO3, (NH4)2S2O8
TFA/H2O
1 R1 = Me
4 R1 = Et
5 R1 = Me
6 R1 = Et
Entry
Substrate
R2
Productd
HPLC yield (%)
Isolated yield (%)
1
2
3
4
5
6
7
8
1
1
1
1
1
1
4
1
Octyl
B5
62
62
36
53
69
55
B5
–
Cyclopropylc
Cyclobutyl
Cyclopentyl
Cyclohexyl
Isopropyl
Isopropyl
tert-Butyl
5a
2
5b
5c
5d
6
60
54a
35
54b
53b
–
–
a Purified by vacuum distillation.
b Isolated as a mixture of mono and dialkylated triazole after removal of starting material via elution of crude reaction mixture through a plug
of SiO2 with 1:1 EtOAc:hexanes.
c 3 equiv. of carboxylic acid were used in this reaction.
d For 1H NMR data, see Ref. 7.
experiments. In the case of the butane tether 7b, the 6–5
fused system 8b was observed as the only product of the
reaction after workup. When the chain was extended
one methylene, the 7–5 system 8c was isolated, albeit in
lower yield. Shortening the chain by one methylene
resulted in none of the 5–5 cyclized product 8a being
formed, most likely due to strain of the ring system.
Thus, primary radicals can alkylate the triazole ring
system when held in close proximity to the triazole ring.
both cyclic and acyclic, performed best in the reaction
(entries 2–7) with yields in the 50–60% range. Pivalic
acid and octanoic acid (entries 1 and 8) were poor
substrates for the alkylation and gave little or no
desired products. Replacement of the methyl group on
the nitrogen with an ethyl group afforded the corre-
sponding N-ethyl derivative 6 (entry 7).
Once generated, alkyl radicals can either attack the
electrophilic, protonated triazole in a productive fash-
ion to afford 52a or react with other species in solution
to afford alkane by-products.8 This paradigm explains
why highly electron-deficient heterocycles such as
dichloropyridizine are alkylated in high yields using this
procedure.2b The lower yields observed for the alkyla-
tion of N-alkyl triazoles 1 and 4 may be due to it being
less electrophilic than the dichloropyridizine ring sys-
tem. The poor performance of octanoic and pivalic acid
in the reaction can also be explained using this ratio-
nale. Highly reactive primary radicals may react with
other species in solution rather than the desired hetero-
cycle, while a tertiary radical may not be nucleophilic
enough to their react with protonated triazole.
We have demonstrated the alkylation of a triazole
heterocycle with alkyl radicals generated from car-
boxylic acids. While triazoles are not as reactive under
these conditions as other heterocyclic substrates
reported in the literature, 1,5-dialkyl triazoles with sec-
ondary alkyl substituents in the 5 position can be
isolated in synthetically useful yields using this method.
While secondary radicals work best in the intermolecu-
lar sense, primary radicals alkylate triazoles in an
intramolecular fashion to form bicyclic ring systems.
Table 2.
1. TFA, H2O, 80 ºC
2. AgNO3, (NH4)2S2O8(aq)
n
In the case of primary radicals, covalently attaching the
carboxylic acid precursor to the heterocycle should
promote reactivity between the two species. To test this
hypothesis, we synthesized triazoles 7a–c with a pri-
mary ester moiety tethered to the triazole nitrogen
through an alkyl chain. The esters groups of 7a–c were
hydrolyzed by heating in TFA/water at 70°C. Upon
complete hydrolysis of the ethyl ester, the reaction was
then charged with 20 mol% AgNO3 followed by slow
addition of an aqueous solution of ammonium persul-
fate (2 equiv.). Table 2 summarizes the results of these
N
N
N
CO2Et
N
N
n
N
7
8
Substrate
n
Producta
Yield%
7a
7b
7c
1
2
3
8a
8b
8c
0
84
63
a For 1H NMR data, see Ref. 9.

K. B. Hansen et al. / Tetrahedron Letters 42 (2001) 7353–7355
7355
Acknowledgements
of addition. Once the addition was complete, the reaction
was aged for 1 h as it cooled to room temperature. The
temperature was then lowered to 0°C and concentrated
NH4OH was slowly added, keeping the temperature below
20°C, until the pH was adjusted to approximately 12. The
mixture was then transferred to a separatory funnel and
extracted with ethyl acetate (2×400 mL). The organic
layers were combined, washed with brine (200 mL), dried
with MgSO4 and filtered. The solution was then concen-
trated under reduced pressure to afford a yellow oil.
Purification of the product was performed by either vac-
uum distillation of the crude concentrate or elution
through a plug of silica gel with 1:1 ethyl acetate/hexanes.
We thank Dr. Peter Dormer for aid in structure deter-
mination.
References
1. Katritzky, A. R.; Lue, P.; Yannakopoulou, K. Tetra-
hedron 1990, 46, 641.
2. (a) Minisci, F.; Benardi, R.; Bertini, F.; Galli, R.; Perch-
inummo, M. Tetrahedron 1971, 27, 3575; (b) Samaritoni, J.
G.; Babbitt, G. J. Heterocyclic Chem. 1991, 28, 583.
3. (a) Jain, R.; Cohen, L. A.; Nahia, A. E.-K.; King, M. M.
Tetrahedron 1997, 53, 2365; (b) Begg, C. G.; Grimmett, M.
R.; Yu-Man, L. Aus. J. Chem. 1973, 26, 415.
1
7. Compound 5a: H NMR (CDCl3) l 7.61 (s, 1H), 3.82, (s,
3H), 1.78 (quint., J=6.7 Hz, 1H), 1.00 (d, J=6.7 Hz, 4H).
Compound 2: 1H NMR (CDCl3) l 7.72 (s, 1H), 3.69 (s,
3H), 3.54 (m, 1H), 2.39 (m, 4H), 2.07 (m, 1H), 1.97 (m,
1H). Compound 5b: 1H NMR (CDCl3) l 7.70 (s, 1H), 3.78
(s, 3H), 3.09 (m, 1H), 2.00 (m, 2H), 1.82 (m, 4H), 1.64 (m,
2H). Compound 5c: 1H NMR (CDCl3) l 7.77 (s, 1H), 3.83
(s, 3H), 2.72 (m, 1H), 1.89 (m, 4H), 1.70 (m, 3H), 1.35 (m,
3H). Compound 5d: 1H NMR (CDCl3) l 7.69 (s, 1H), 3.76
(s, 3H), 3.02 (sept., J=6.7 Hz, 1H), 1.26 (d, J=6.7 Hz,
4. Bulger, P. C.; Cottrell, I. F.; Cowden, C. J.; Davies, A. J.;
Dolling, U.-H. Tetrahedron Lett. 2000, 41, 1297.
5. Samartoni, J. G. Org. Prep. Proc. Int. 1988, 20, 117.
6. General procedure: A flask equipped with an addition
funnel and mechanical stirrer was charged with 1 (10.0 g,
0.12 mol), water (10 mL) and trifluoroacetic acid (27.8
mL, 0.36 mol). The solution was heated to 70°C. A
homogeneous solution of NH4S2O8 (68.66 g, 0.30 mol),
water (69 mL), carboxylic acid (0.24 mol) and 5N NaOH
(48 mL, 0.24 mol) was prepared and charged to the
addition funnel. A solution of AgNO3, prepared by dis-
solving AgNO3 (2.0 g, 0.012 mol) in water (6 mL), was
charged to the reaction flask. The drop-wise addition of
the NH4S2O8/carboxylic acid solution was initiated and
within 5–10% of the addition the temperature began to rise
rapidly and gas evolution was observed. The temperature
was maintained between 80–90°C by controlling the rate
1
6H). Compound 6: H NMR (CDCl3) l 7.77 (s, 1H), 4.10
(q, J=7.3 Hz, 2H), 3.05 (sept., J=6.8 Hz 1H), 1.44 (t,
J=7.3 Hz, 3H), 1.33 (d, J=6.8 Hz, 6H).
8. Anderson, J. M.; Kochi, J. K. J. Am. Chem. Soc. 1970, 92,
1651.
1
9. Compound 8b: H NMR (CDCl3) l 7.76 (s, 1H), 4.09 (m,
2H), 2.85 (m, 2H), 2.02 (m, 2H), 1.93 (m, 2H). Compound
1
8c: H NMR (CDCl3) l 7.69 (s, 1H), 4.29 (m, 2H), 2.96
(m, 2H), 1.95 (m, 2H), 1.81 (m, 2H), 1.71 (m, 2H).