Broadening the scope of 1,2,4-triazine synthesis by the application of microwave technology

Article abstract of DOI:10.1016/S0040-4039(02)02845-9

By the application of microwave technology, a general protocol has been developed for the rapid synthesis of diverse 3,5,6-trisubstituted 1,2,4-triazines in excellent yield and purity, including many previously unknown 3-heterocyclic-1,2,4-triazines.

Full text of DOI:10.1016/S0040-4039(02)02845-9

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 1123–1127
Broadening the scope of 1,2,4-triazine synthesis by the
application of microwave technology
Zhijian Zhao,* William H. Leister, Kimberly A. Strauss, David D. Wisnoski and Craig W. Lindsley
Department of Medicinal Chemistry, Technology Enabled Synthesis Group, Merck Research Laboratories, PO Box 4,
West Point, PA 19486, USA
Received 3 December 2002; accepted 17 December 2002
Abstract—By the application of microwave technology, a general protocol has been developed for the rapid synthesis of diverse
3,5,6-trisubstituted 1,2,4-triazines in excellent yield and purity, including many previously unknown 3-heterocyclic-1,2,4-triazines.
© 2003 Elsevier Science Ltd. All rights reserved.
Substituted triazines represent an important class of
nitrogen-containing heterocycles. The 1,2,4-triazine
core is a versatile synthetic platform to access a wide-
range of condensed heterocyclic ring systems via
intramolecular Diels–Alder reactions with a vast array
of dienophiles.1 In addition, the triazine ring system is a
key component of commercial dyes, herbicides, insecti-
cides and more recently, pharmaceutical compositions.2
As our efforts are directed at an iterative analog library
approach to support nascent medicinal chemistry pro-
grams, this latter application for the triazine scaffold
attracted our attention.
Scheme 1.
While developing structure activity relationships (SAR)
for a small heterocyclic lead compound, the need arose
for a general protocol to synthesize 3-heterocyclic-1,2,4-
triazines, preferably in a manner amenable to analog
library synthesis (Fig. 1). Examination of the literature
provided numerous methods for the synthetic prepara-
tion of triazines.3 However, the examples highlighted by
these protocols typically focus on only simple aliphatic,
phenyl and ester substituents at R1–R3.1–4 This was
Scheme 2.
especially true for the synthesis of 1,2,4-triazines via the
condensation of 1,2-diketones with acyl hydrazides and
ammonium acetate under traditional thermal and ‘dry
Figure 1. Generic and target 3-heterocyclic-1,2,4-triazine.
media’ microwave-assisted reaction conditions (Scheme
1).4 Due to our in-house reagent library of diverse acyl
hydrazides, our efforts centered on exploring this syn-
thetic route.
* Corresponding author. Tel.: 215-652-5892; fax: 215-652-6345;
e-mail: zhijian zhao@merck.com
–
0040-4039/03/$ - see front matter © 2003 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)02845-9

1124
Z. Zhao et al. / Tetrahedron Letters 44 (2003) 1123–1127
The traditional thermal conditions involve heating a
1,2-diketone and an acyl hydrazide, in a 1:1 ratio, with
excess ammonium acetate in refluxing acetic acid for
6–24 h.
to consume starting materials (10–24 h) and resulted in
numerous side products. Applying these same reaction
conditions with non-heterocyclic starting materials
delivered the desired triazines in under 8 h and in >65%
isolated yields, in accord with literature precedent.1,4a
In our hands, these conditions with either a heterocyclic
acyl hydrazide or heterocylic-containing 1,2-diketone
afforded low yields (<30%), required extended heating
Recently, ‘dry media’ microwave-assisted protocols
have emerged wherein an inorganic support, such as
Table 1. Representative 3-substituted-1,2,4-triazines

Z. Zhao et al. / Tetrahedron Letters 44 (2003) 1123–1127
1125
Table 2. 1,2-Diketone diversity in 1,2,4-triazine synthesis
conventional thermal methods.7 In addition, the Smith-
synthesizer™ is ideal for library synthesis as the entire
library can be run unattended under carefully con-
trolled and reproducible reaction conditions.8
silica gel, is employed as the energy transfer medium in
lieu of solvent.5 This approach has been applied to the
preparation of 1,2,4-triazines to successfully deliver
products in good yields using
a
conventional
microwave oven.4b Unfortunately, attempts to repro-
duce this work utilizing Personal Chemistry’s Smithsyn-
thesizer™ afforded poor yields with non-heterocyclic
reactants and no desired product when heterocyclic
reactants were employed.6
Conventional thermal conditions were quickly adapted
and optimized on the Smithsynthesizer™ (Scheme 2) to
deliver the previously unknown 3-imidazoly-1,2,4-tri-
azine 3 in 85% isolated yield.9 The optimized conditions
involved reacting benzil 1 and an imidazoyl acyl hydra-
zide 2, in a 1:1 ratio, with 10 equiv. of ammonium
acetate in 1 mL of acetic acid for 5 min at 180°C, 60°C
above the boiling point of HOAc. During the course of
the reaction, the pressure never exceeded 5 psi. Upon
rapid cooling of the reaction vessel to 40°C (a standard
feature of the Smithsynthesizer™), a yellow precipitate
formed which proved to be 3. With this result in hand,
a 48-membered library was then synthesized employing
However, developing a microwave-assisted protocol
was still of paramount interest. Abbreviated reaction
times held promise for less product loss from decompo-
sition due to the prolonged heating required for a
traditional thermal reaction with heterocyclic reactants.
Indeed, microwave protocols have been developed to
rapidly access a number of heterocylic frameworks with
abbreviated reactions times and yields far exceeding

1126
Z. Zhao et al. / Tetrahedron Letters 44 (2003) 1123–1127
a diverse set of acyl hydrazides using 1 as the 1,2-dike-
tone component. The desired product was obtained in
every instance, with crude LCMS purity in excess of
75% and isolated yields in excess of 79%. For ꢀ60% of
this library, the desired product precipitated out of
solution upon rapid cooling, and clean material could
be obtained by filtration and washing. The remaining
40% of the library was purified by mass-triggered
preparative LCMS on a custom Agilent 1100 instru-
ment.10 Representative library members are depicted in
Table 1 and include such 3-position heterocycles as
oxazole (entry 1), pyrimidine (entry 2), pyrrole (entry
3), triazole (entry 5), thiazole (entry 6) and pyridine
(entry 7). This new protocol also allowed for the syn-
thesis of saturated heterocyclic congeners (entry 4) as
well as other aminoalkyl derivatives (entry 8) demon-
strating the generality of this methodology for analog
library synthesis.11 The majority of the triazine analogs
from this library have not been previously described in
the primary or patent literature and represent novel
heterobicyclic structures.12
3. (a) Robinson, L.; Vanderwala, P. H.; Laakso, P. V.
Tetrahedron 1957, 1, 103; (b) Metze, R. Chem Ber. 1955,
88, 772; (c) Metze, R. Chem Ber. 1954, 87, 1540; (d)
Saraswathi, T. V.; Srinivasan, V. R. Tetrahedron Lett.
1977, 33, 1043; For excellent reviews, see: (e) Neunhoef-
fer, H. Chemistry of 1,2,3-Triazines and 1,2,4-Triazines,
Tetrazines and Pentazines; Wiley-Interscience: New York,
1978; pp. 194–200; (f) Neunhoffer, H. In Comprehensive
Heterocyclic Chemistry; Katrizky, A. R.; Rees, C. W.;
Boulton, A. J.; McKilop, A., Eds.; Pergamon: Oxford,
1984; Vol. 3, pp. 385–456.
4. (a) Taylor, E. C.; French, L. G. J. Org. Chem. 1989, 54,
1245 and references cited therein; (b) Rostamizadeh, S.;
Sadeghi, K. Synth. Commun. 2002, 32, 1899; (c)
Mazaahir, K.; Pooja, S.; Bhushan, K.; Pretti, M. Synth.
Commun. 2001, 31, 1639.
5. Kidwai, M. Pure Appl. Chem. 2001, 73, 147.
6. Due to vague experimental procedures in Ref. 4b, repeat-
ing the published work proved difficult. However,
attempts to optimize reaction variables in the Smithsyn-
thesizer™, in place of a conventional microwave oven,
afforded very poor results with ‘dry media’.
This methodology is not only general with respect to
the acyl hydrazide component, but also appears to be
general for the 1,2-diketone component as well. Repre-
sentative examples from another library aimed at this
diversity element are illustrated in Table 2. Again,
excellent crude LCMS purities (>70%) and isolated
yields were attained under standard reaction conditions
for heterocyclic (entries 1 and 2), and 1-alkyl-2-phenyl-
3,4-diketones (entries 3 and 4).13 As entries 3 and 4
involved unsymmetrical 1,2-diketones, a 1:1 ratio of
regioisomers was obtained; moreover, extending the
reaction time from 5 to 10 min increased the yields by
ꢀ15% for these entries.
7. For a review with accounts of heterocycles synthesized by
microwave heating, see: Lidstrom, P.; Tierney, J.;
Wathey, B.; Westman, J. Tetrahedron 2001, 57, 925.
8. For information on Personal Chemistry’s micro-
wave
technology
for
organic
synthesis,
see:
http:¯¯www.personalchemistry.com.
9. Experimental for 3: To a 5 mL Smithsynthesizer™ reac-
tion vial (Part c 351521) with a stir bar was placed
benzil, 1, (42 mg, 0.2 mol) the imidazoly acyl hydrazide,
2, (26 mg, 0.2 mmol), ammonium acetate (154 mg, 2.0
mmol) and 1 mL of glacial HOAc. The reaction vessel
was heated in the Smithsynthesizer™ reactor cavity for 5
min at 180°C. After 5 min, the vessel was rapidly cooled
to 40°C by the unit. Upon removal from the reactor
cavity, a bright yellow precipitate was collected by filtra-
tion from the reaction vessel. The solid was washed with
water and dried in a vacuum oven overnight at 50°C to
afford 51 mg (85%) of 3 as a bright yellow solid. Analyt-
ical LCMS indicated a single peak (2.190 min, CH3CN/
H2O/0.1%TFA, 4 min gradient) >98% pure by UV (214
nm) and 100% pure by ELSD. 1H NMR (300 MHz,
DMSO-d6): l 13.3 (bs, 1H), 7.83 (d, J=3 Hz, 2H), 7.76
(d, J=9 Hz, 2H), 7.54 (m, 2H), 7.48 (m, 6H); HRMS
calcd for C18H13N5(M+H), 300.1244; found 300.1248
(M+H).
In summary, a microwave-assisted protocol for the
general synthesis of functionalized 1,2,4-triazines has
been developed on a Smithsynthesizer™. In addition to
providing high yielding access to a number of previ-
ously unknown 3-heterocyclic-1,2,4-triazines, overall
reaction times have been reduced 60–300-fold over con-
ventional thermal conditions. Additional applications
of microwave technology for analog library synthesis
are in progress and will be reported in due course.
Acknowledgements
10. Leister, W. H.; Strauss, K. A.; Wisnoski, D. D.; Zhao,
Z.; Lindsley, C. W. J. Comb. Chem., submitted.
11. Representative analytical data for library members in
Table 1: (entry 3): Analytical LCMS indicated a single
peak (3.403 min, CH3CN/H2O/0.1%TFA, 4 min gradient)
We would like to thank Dr. Charles W. Ross III for
obtaining HRMS data (accurate mass measurements).
1
>98% pure by UV (214 nm) and 100% pure by ELSD. H
NMR (300 MHz, CDCl3): l 7.65 (m, 1H), 7.62 (m, 2H),
7.60 (m, 2H), 7.39 (m, 7H), 4.47 (s, 3H); HRMS calcd for
C19H15N5(M+H), 314.1400; found 314.1403 (M+H);
(entry 4): Analytical LCMS indicated a single peak (3.230
min, CH3CN/H2O/0.1%TFA, 4 min gradient) >98% pure
by UV (214 nm) and 100% pure by ELSD. 1H NMR (300
MHz, CDCl3): l 7.56 (m, 4H), 7.36 (m, 6H), 5.44 (dd,
J=1.5, 5.7 Hz, 1H), 4.27 (m, 1H), 4.1 (m, 1H), 2.51 (m,
1H), 2.39 (m, 1H), 2.26 (m, 1H), 2.11 (m, 1H); HRMS
calcd for C19H17N3O (M+H), 304.1445; found 304.1445
References
1. For an excellent review, see: (a) Boger, D. L. Chem Rev.
1986, 86, 781; (b) Benson, S. C.; Li, J.-H.; Snyder, J. K.
J. Org. Chem. 1992, 57, 5285.
2. (a) Hurst, D. T. Prog. Heterocyclic Chem. 1995, 7, 244;
(b) Groger, H.; Sans, J.; Gunther, T. Chim. Oggi 2000,
18, 12; (c) Bondinell, W. E. et. al. WO 0,276,984, 2002;
(d) Bettati, M. et. al. WO 0,238,568, 2002.

Z. Zhao et al. / Tetrahedron Letters 44 (2003) 1123–1127
1127
1
(M+H); (entry 6): Analytical LCMS indicated a single
peak (3.393 min, CH3CN/H2O/0.1%TFA, 4 min gradient)
>98% pure by UV (214 nm) and 100% pure by ELSD. H
ELSD. H NMR (300 MHz, CDCl3): l 7.98 (s, 1H), 7.72
(m, 1H), 7.67 (m, 1H), 7.51 (s, 1H), 7.25 (m, 1H), 7.10 (d,
J=3.8 Hz, 1H), 6.69 (m, 1H), 6.61 (m, 1H); HRMS calcd
for C14H8N4O3 (M+H), 281.0669; found 281.0670 (M+
H); (entry 4): Analytical LCMS indicated a single peak
(2.909 min, CH3CN/H2O/0.1%TFA, 4 min gradient)
>98% pure by UV (214 nm) and 100% pure by ELSD. By
NMR, a 1:1 ratio of regioisomers: Isomer 1: 1H NMR
(300 MHz, CDCl3): l 7.99 (s, 1H), 7.66 (m, 2H), 7.57 (m,
3H), 7.52 (s, 1H), 2.97 (t, J=7.8 Hz, 2H), 1.84 (m, 2H),
0.93 (t, J=8 Hz, 3H); Isomer 2: 1H NMR (300 MHz,
CDCl3): l 7.96 (s, 1H), 7.74 (m, 2H), 7.57 (m, 3H), 7.49
(s, 1H), 3.19 (t, J=7.8 Hz, 2H), 1.84 (m, 2H), 0.93 (t,
J=8 Hz, 3H); HRMS calcd for C15H14N4O (M+H),
267.1241; found 267.1242 (M+H).
1
NMR (300 MHz, CDCl3): l 8.99 (s, 1H), 8.94 (s, 1H),
7.60 (m, 4H), 7.41 (m, 6H); HRMS calcd for C18H12N4S
(M+H), 317.0857; found 317.0856 (M+H).
12. Searches of CAS, SciFinder and Bielstein indicated that
these compounds are new chemical entities, unknown in
the primary and/or patent literature.
13. Note, as an unsymmetrical 1,2-diketone was employed in
entries 3 and 4, a 1:1 mixture of regioisomers was
obtained. Representative analytical data for library mem-
bers in Table 2: (entry 2): Analytical LCMS indicated a
single peak (2.519 min, CH3CN/H2O/0.1%TFA, 4 min
gradient) >98% pure by UV (214 nm) and 100% pure by