An investigation into the alkylation of 1,2,4-triazole

Article abstract of DOI:10.1016/S0040-4039(99)02272-8

The alkylation of 1,2,4-triazole with 4-nitrobenzyl halides and a variety of bases afforded the 1- and 4-alkylated isomers with a consistent regioselectivity of 90:10. Previously reported regiospecific alkylations of 1,2,4-triazole were re-examined and the quoted isomer ratios were shown to depend on the isolation procedure. The use of DBU as base in the alkylation of 1,2,4-triazole allows for a convenient and high yielding synthesis of 1- substituted-1,2,4-triazoles. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(99)02272-8

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 1297–1301
An investigation into the alkylation of 1,2,4-triazole
Paul G. Bulger, Ian F. Cottrell, Cameron J. Cowden,∗ Antony J. Davies ∗ and Ulf-H. Dolling
Department of Process Research, Merck Sharp & Dohme Research Laboratories, Hertford Road, Hoddesdon, Hertfordshire,
EN11 9BU, UK
Received 10 November 1999; accepted 7 December 1999
Abstract
The alkylation of 1,2,4-triazole with 4-nitrobenzyl halides and a variety of bases afforded the 1- and 4-alkylated
isomers with a consistent regioselectivity of 90:10. Previously reported regiospecific alkylations of 1,2,4-triazole
were re-examined and the quoted isomer ratios were shown to depend on the isolation procedure. The use of DBU
as base in the alkylation of 1,2,4-triazole allows for a convenient and high yielding synthesis of 1-substituted-1,2,4-
triazoles. © 2000 Elsevier Science Ltd. All rights reserved.
Keywords: 1,2,4-triazole; DBU; regioselective alkylation; rizatriptan.
Fused and pendant 1,2,4-triazoles are a ubiquitous feature of many pharmaceutical and agrochemical
products.1 The 1-substituted-1,2,4-triazole nucleus is particularly common and examples can be found in
marketed drugs such as fluconazole,2 terconazole3 and rizatriptan (1).4 An alkylation reaction would be
an obvious method of introducing this triazole unit. However, the controlled, regioselective alkylation of
1,2,4-triazole can be problematic, with mixtures of isomers produced and over-alkylation leading to salt
formation possible. While such problems have been noted,5 there are also several contrasting reports of
regiospecific alkylation reactions.6 We recently required the efficient synthesis of 1-alkyl-1,2,4-triazole
derivatives and now report our findings on this apparently trivial reaction. We also disclose a mild
and convenient method for the alkylation of 1,2,4-triazole, which uses the weakly nucleophilic base
1,8-diazabicyclo[5.4.0]undec-7-ene (DBU).
Scheme 1.
Rizatriptan (1) is a recently marketed 5-HT1B/1D receptor agonist which is prescribed for the acute
treatment of migraine.4 Two syntheses of this compound have been reported, both of which proceed
∗
Corresponding authors. Fax: 01992 470437; e-mail: cameron_cowden@merck.com (C. J. Cowden),
tony_davies2@merck.com (A. J. Davies)
0040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(99)02272-8
tetl 16200

1298
via 4-nitrobenzyl triazole (2) with either a Fisher indole reaction4 or a Pd-mediated cyclisation7 used to
construct the indole heterocycle (Scheme 1).
The Process Chemistry route to compound 2 utilised a two-step alkylation of 4-aminotriazole with
4-nitrobenzyl bromide followed by sodium nitrite deamination with evolution of nitrous oxide.8 This was
an efficient sequence affording an overall yield of >90% from 4-aminotriazole. An attractive alternative
is the alkylation of 1,2,4-triazole with a 4-nitrobenzyl halide, as this is one synthetic step shorter and
would allow the use of cheaper reagents (triazole and 4-nitrobenzyl chloride versus 4-aminotriazole and
4-nitrobenzyl bromide). However, the reported yield of alkylated product by this method was moderate
(52%) but, of relevance to the present study, the authors observed that the alkylation gave “as expected,
a single isomer”.4
As part of our ongoing research, the alkylation route to compound 2 was re-examined. Following
the described procedure (NaH, 4-nitrobenzyl bromide, DMF, rt), we obtained a 90:10 mixture of the
1- and 4-isomers, 2 and 3, respectively (Scheme 2). When this reaction was repeated with a range of
conditions, on each occasion, a surprisingly consistent ratio of alkylated 1,2,4-triazoles was obtained. At
no stage could we significantly alter this ratio, either by changing the base used (NaH, K2CO3, Cs2CO3,
DBU, LiOH, NaOH, tetramethylguanidine); changing the electrophile (X=Cl, Br) or changing the solvent
(MeOH, THF, toluene, CH3CN, DMF). This could imply alkyl group isomerisation, however, literature
precedent for such an equilibration is only reported at high temperatures.9 The isomer ratios were also
constant (±2%) from the first analysis point of 5 min up to 3 days, apparently indicating that the ratios
do not result from an equilibration.
Scheme 2.
The optimum synthetic conditions we found were the use of DBU (1.2 equiv.), 4-nitrobenzyl chloride
(1.0 equiv.), triazole (1.1 equiv.) and THF whereby a 93% yield of adducts in a 90:10 ratio was obtained.10
Interestingly, successive aqueous washes increased the isomer ratio of 2:3 to 98:2. This preferential
partition of the minor isomer to the aqueous layer may account for the earlier report of the alkylation
reaction proceeding to give a single isomer.4
We were intrigued to see whether this simple partitioning of 4-alkyl-1,2,4-triazoles to an aqueous
layer on work-up might explain some previously reported regiospecific alkylation reactions. Katritzky et
al. have disclosed an improved method for the alkylation of 1,2,4-triazole and they give five examples
of isomer ratios of 100:0 using a DMF/NaOH system.6a The experimental procedure for these reactions
included an aqueous work-up and it was thought that the isomer ratio might not have been determined on
the crude reaction mixtures. Table 1 illustrates our results following this procedure. Indeed, excellent
isomer ratios were observed after work-up, however, the isomer ratios of the crude mixtures before
work-up were only ca. 90:10.
In a related project, we required the synthesis of 1-ethyl-1,2,4-triazole. The use of NaOH/DMF
was not practical, however, due to the difficulties in separation of the water soluble and volatile
1-ethyl-1,2,4-triazole from DMF. Instead, using DBU as base and THF as solvent, we obtained an
88% yield of products as a 90:10 mixture of isomers, again favouring the 1-isomer. An advantage of
this method was that DBU·HX precipitated from the reaction and was easily removed by filtration.
Concentration of the resulting liquors then gave the crude product which was easily purified by distillation
to afford pure 1-ethyl-1,2,4-triazole (78% yield).

1299
Table 1
(a) Isomer ratio based on 1H NMR analysis of the product mixture. (b) All
reactions were conducted at 20°C under an atmosphere of nitrogen. (c) We were
unable to observe the expected product from this reaction, even at 0°C, due to
competing hydrolysis of the ester.
Compared to previously reported alkylation methods,1a–c the use of DBU as base is a synthetically
simple and mild method for the alkylation of 1,2,4-triazole.11 Hence, we applied these conditions to a
range of alkylating agents and obtained good yields for all substrates with the regioselectivity of the
reactions varying from 86:14 up to 94:6 (Table 2).
Table 2
1
(a) Isomer ratio based on H NMR analysis of the product mixture. (b) Yield
refers to adduct 6 isolated after distillation. (c) Ethyl acrylate was formed in this
reaction and may be the alkylating species. (d) Yield after silica gel chromato-
graphy. (e) This compound was isolated as a 92:8 mixture of isomers.

1300
Notable examples include the reaction of alkyl esters (entries 9–12), which proceeded without compet-
ing hydrolysis, and secondary alkyl halides (entries 5 and 7), where competing dehydrohalogenation of
the alkylating agent by DBU was not a serious competing problem.
Purification of the isomeric mixtures of triazoles was achieved by distillation (except entries 14
and 15). In this manner, pure 1-alkyl-1,2,4-triazoles were obtained (6:7≥99:1) without any attempt at
fractionation. In each case, the 1-isomer had the lower boiling point as indicated by an examination of
the distillation residue, which showed the presence of the 4-alkyl isomer. This highlights the difference in
physical properties12 of the two triazole isomers and the care needed in reporting regiospecific alkylation
reactions, as distillation of a mixture invariably leads to the isolation of a single adduct.
In conclusion, DBU is a mild and convenient base for the alkylation of 1,2,4-triazole. Some literature
reports of regiospecific alkylations of 1,2,4-triazole have been re-examined and in each case, the 1-isomer
is accompanied by formation of the 4-isomer. However, such are the differing volatilities and polarities
of these isomers, they are readily separable by distillation, recrystallisation or silica gel chromatography.
Hence, in alkylation reactions of 1,2,4-triazoles which produce isomeric mixtures, the 4-isomer often
goes undetected and thus unreported.
References
1. For reviews of 1,2,4-triazole chemistry, see: (a) Garratt, P. J. In Comprehensive Heterocyclic Chemistry II; Katritzky, A.
R.; Rees, C. W.; Scriven, E. F. V., Eds.; Pergamon Press: Oxford, 1996; Vol. 4, 127. (b) Temple, C. In The Chemistry of
Heterocyclic Compounds: Triazoles 1,2,4; Montgomery, J. A., Ed.; John Wiley & Sons: Chichester, 1981. (c) Turnbull, K.
Prog. Heterocycl. Chem. 1998, 10, 153. For reviews of the therapeutic use of triazoles, see: (d) Terrell, C. L. Mayo Clin.
Proc. 1999, 74, 78. (e) Koltin, Y.; Hitchcock, C. A. Curr. Opin. Chem. Biol. 1997, 1, 176.
2. Fluconazole is marketed as Diflucan™, see: Richardson, K.; Brammer, K. W.; Marriott, M. S.; Troke, P. F. Antimicrob. Agents
Chemother. 1985, 27, 832.
3. Terconazole is marketed as Terazole™, see: Heeres, J.; Hendrickx, R.; Van Cutsem, J. J. Med. Chem. 1983, 26, 611.
4. Rizatriptan is marketed as Maxalt™, see: Street, L. J.; Baker, R.; Davey, W. B.; Guiblin, A. R.; Jelley, R. A.; Reeve, A.
J.; Routledge, H.; Sternfeld, F.; Watt, A. P.; Beer, M. S.; Middlemiss, D. N.; Noble, A. J.; Stanton, J. A.; Scholey, K.;
Hargreaves, R. J.; Sohal, B.; Graham, M. I.; Matassa, V. G. J. Med. Chem. 1995, 38, 1799.
5. (a) Polya, J. B. In Comprehensive Heterocyclic Chemistry; Katritzky, A. R.; Rees, C. W.; Potts, K. T., Eds.; Pergamon Press:
Oxford, 1981; Vol. 5, 746. (b) Abenhaïm, D.; Díez-Barra, E.; de la Hoz, A.; Loupy, A.; Sánchez-Migallón, A. Heterocycles
1994, 38, 793. (c) Olofson, R. A.; Kendall, R. V. J. Org. Chem. 1970, 35, 2246. (d) Begtrup, M.; Larsen, P. Acta Chem.
Scand. 1990, 44, 1050. (e) Balasubramanian, M.; Keay, J. G.; Scriven, E. F. V. Heterocycles 1994, 37, 1951.
6. (a) Katritzky, A. R.; Kuzmierkiewicz, W.; Greenhill, J. V. Recl. Trav. Chim. Pays-Bas 1991, 110, 369. (b) Gasparini, J. P.;
Gassend, R.; Maire, J. C.; Elguero, J. J. Organomet. Chem. 1980, 188, 141. (c) Díez-Barra, E.; de la Hoz, A.; Rodríguez-
Curíel, R. I.; Tejeda, J. Tetrahedron 1997, 53, 2253. (d) Smith, K.; Small, A.; Hutchings, M. G. Synlett 1991, 485. (e) Urban,
F. J.; Breitenbach, R. Synth. Commun. 1999, 29, 645.
7. Chen, C.-Y.; Lieberman, D. R.; Larsen, R. D.; Reamer, R. A.; Verhoeven, T. R.; Reider, P. J.; Cottrell, I. F.; Houghton, P. G.
Tetrahedron Lett. 1994, 35, 6981.
8. Astleford, B. A.; Goe, G. L.; Keay, J. G.; Scriven, E. F. V. J. Org. Chem. 1989, 54, 731.
9. (a) Smith, K.; Small, A.; Hutchings, M. G. Chem. Lett. 1990, 347. (b) Carlsen, P.; Gautun, O. R.; Jørgensen, K. B. Molecules
1996, 1, 242. (c) A general anionic mechanism for the thermodynamic control of regioselectivity in the N-alkylation of
heterocycles has been postulated, see: Bentley, W. T.; Jones, R. V. H.; Wareham, P. J. Tetrahedron Lett. 1998, 30, 4013.
10. Compounds 2 and 3: To a mechanically stirred suspension of 1,2,4-triazole (10.0 g, 0.145 mol) and 4-nitrobenzyl chloride
(22.4 g, 0.130 mol) in THF (100 mL) was added DBU (24.2 g, 0.159 mol) in THF (20 mL) via syringe pump over 1 h.
The solution was stirred at room temperature for 16 h, filtered and the cake washed with THF (100 mL). The filtrate was
concentrated to residue and partitioned between water (250 mL) and EtOAc (500 mL and 100 mL). The combined organic
layers were concentrated to afford 2 and 3 (24.7 g, 93%) as a 94:6 mixture. Purification by silica gel chromatography
(eluent EtOAc→EtOAc:MeOH, 85:15), yielded in elution order: 1-[(4-nitrophenyl)methyl]-1,2,4-triazole, 2 (19.0 g, 73%):
1
mp 100–101°C (ref.13: mp 95–100°C); Rf (EtOAc:MeOH, 9:1) 0.71; H NMR (250 MHz, d6-DMSO): δ 8.61 (1H, s),
8.12–8.09 (2H, m), 7.92 (1H, s), 7.39–7.36 (2H, m), 5.49 (2H, s); 13C NMR (62.5 MHz, d6-DMSO): δ 153.0, 148.0, 145.6,
144.7, 129.8, 124.7, 52.1. Anal. calcd for C9H8N4O2: C, 52.94; H, 3.95; N, 27.44; O, 15.67. Found: C, 53.01; H, 3.84; N,
27.30; O, 15.68. 4-[(4-Nitrophenyl)methyl]-1,2,4-triazole, 3 (1.1 g, 4%): mp 143–145°C; Rf (EtOAc:MeOH, 9:1) 0.41; 1H

1301
NMR (250 MHz, d6-DMSO): δ 8.54 (2H, s), 8.14–8.10 (2H, m), 7.42–7.38 (2H, m), 5.35 (2H, s); 13C NMR (62.5 MHz,
d6-DMSO): δ 148.0, 145.0, 144.3, 129.6, 124.9, 47.6. Anal. calcd for C9H8N4O2: C, 52.94; H, 3.95; N, 27.44; O, 15.67.
Found: C, 53.03; H, 3.85; N, 27.42; O, 15.70.
11. To the best of our knowledge, the alkylation of 1,2,4-triazole using DBU as base has only been reported once, see: Norcross,
R. D.; von Matt, P.; Kolb, H. C.; Belluš, D. Tetrahedron 1997, 53, 10289.
12. El Ghomari, M. J.; Mokhlisse, R.; Laurence, C.; Le Questel, J.-Y.; Berthelot, M. J. Phys. Org. Chem. 1997, 10, 669.
13. Andreini, F.; Biagi, G.; Giorgi, I.; Livi, O.; Scartoni, V. Farmaco 1989, 44, 831.