HETEROCYCLES

Paper | Regular issue | Vol. 78, No. 3, 2009, pp. 623-633
Received, 28th August, 2008, Accepted, 14th October, 2008, Published online, 20th October, 2008.
DOI: 10.3987/COM-08-11535

■ A Convenient Method of Preparation of 3,3’-Dichloro-5,5’-bi-1,2,4-triazine and Its Synthetic Applications

Ewa Wolińska*

Department of Chemistry, University of Podlasie, ul. 3-go Maja 54, 08-110 Siedlce, Poland

Abstract

A convenient method for preparing of 3,3'-dichloro-5,5'-1,2,4-triazine (4) and its application to the synthesis of 3,3'-diamino-5,5'-bi-1,2,4-triazines 6a i by the nucleophilic aromatic substitution are described. An attempt to synthesis of cyclophanes containing 5,5'-bi-1,2,4-triazine subunit by ring-closing metathesis of the alkenyl ethers 7a,b have been unsuccessful. A crossover experiment clearly shows that nitrogen atoms of the 1,2,4-triazine ring coordinate to the ruthenium catalyst and deactivate it.

Substituted 1,2,4-triazines are important class of compounds due to their biological activity and importance in organic synthesis.2 The presence of three nitrogen atoms makes the 1,2,4-triazine ring one of the most π-deficient nitrogen-containing heterocycles. Every position of 3, 5 and 6 in 1,2,4-triazine is susceptible to nucleophilic attack, however their reactivity depends on the kind of substituents and nature of the leaving group in the ring.3 Furthermore, nucleophilic addition at 1,2,4-triazine ring carbon carrying hydrogen, SNH reaction, produces stable σ-adducts, which can be converted into aromatic compounds on several ways.4
In contrast to well known 1,2,4-triazine chemistry only a limited number of reports have appeared regarding synthesis and chemical properties of its dimeric analogue, namely 5,5'-bi-1,2,4-triazine and its derivatives. These compounds have been easily prepared by homocoupling reactions of monocyclic 1,2,4-trazine derivatives unsubstituted at C-5 in the presence of cyanide.5 The direct functionalization of such obtained 5,5'-bi-1,2,4-triazines by nucleophilic displacement of substituents present in the 1,2,4-triazine ring is still undeveloped area and have not been studied in details. According to literature there are only two reports concerning nucleophilic displacement of methylsulfinate from readily available 3,3'-bis(methylsulfanyl)-5,5'-bi-1,2,4-triazine (2) with aqueous dimethylamine6 and ethoxide.5a However, in some cases a serious limitation for the above approach is rather difficult access to suitable substrates.7a Also oxidation of methylsulfanyl group in 2 to methylsulfonyl one being more reactive toward nucleophilic displacements, cannot be completed due to the instability of 3,3'-bis(methylsulfonyl)-5,5'-bi-1,2,4-triazine (2a) (Scheme 1).7b
In searching for more effective nucleophugal group for nucleophilic displacements we undertook study on the preparation of 3,3'-dichloro-5,5'-bi-1,2,4-triazine (4) and its application in organic synthesis. Since some amino and diamino derivatives of 1,2,4-triazine are biologically active and have medicinal value8 reaction of 4 with various amines leading to new diamino 5,5'-bi-1,2,4-triazines have been investigated.9 We also describe in this paper an attempt to synthesis of 5,5'-bi-1,2,4-triazine - based macrocycles by ring closing metathesis (RCM).


RESULTS AND DISCUSSION
Synthetic approach to 3,3'-dichloro-5,5'-bi-1,2,4-triazine (4) starts with 3-(methylsulfanyl)-1,2,4-triazine (1) easily prepared using literature procedure10 (Scheme 1). Compound 1 was smoothly converted into disodium salt of 3,3'-dihydroksy-5,5'-bi-1,2,4-triazine (3a), via a two-step one-pot procedure which involved the homocoupling of 1 in the presence of 1.5 equivalents of potassium cyanide to give 3,3'-bis(methylsulfanyl)-5,5'-bi-1,2,4-triazine11 (2), followed by nucleophilic displacement of methylsulfinate from the latter with sodium hydroxide. This route is practical for a large scale operation and avoids the use of time and solvent consuming isolation of bitriazine 2 from the reaction mixture. After 24 hours of stirring at rt disodium salt 3a was isolated as precipitated solid in 86 % yield. When instead of sodium hydroxide, potassium hydroxide is applied under the identical reaction conditions a better soluble dipotassium salt 3b is obtained in 80 % yield, after partial evaporation of solvent. Subsequent chlorination of salts 3a or 3b with an excess of phosphoryl chloride afforded 3,3'-dichloro-5,5'-bi-1,2,4-triazine (4) in 65 and 70 % yield respectively (Scheme 1). The resulted dichloro compound 4 exhibits good air and moisture stability. Evidence for its structure is obtained from 1H, 13C NMR and elemental analysis (see EXPERIMENTAL).

With 3,3'-dichloro-5,5'-bi-1,2,4-triazine (4) in hand we next evaluated its reactivity toward nucleophilic substitution of chlorine atoms with a variety of amines 5a-j. To optimize an amination conditions the reaction of 4 with n-butylamine 5a was first investigated. Compound 4 was treated with n-butylamine 5a in dioxane at ambient temperature. To avoid monosubstitution process five equivalents of amine were necessary to use. The same conditions were applied for reactions of 4 with other amines (Scheme 1). The reactions were completed within hours and appropriate diamino derivatives of 5,5'-bi-1,2,4-triazine 6a-f were precipitated from the reaction mixture as yellow solids or oils in the case of 6g-i. The latter, after evaporation of solvent were turned to solids by treatment with methanol. The yields of the prepared diamino compounds were good or excellent (Table 1). Only diisopropylamine 5j did not undergo reaction with 4 due to steric effect of two diisopropyl groups.
Continuing our research on diamino bitriazines we decided to explore their derivatives 6h and 6i as potential intermediates for the synthesis of 5,5'-bi-1,2,4-triazine containing cyclophanes. The sulfur analogues of such systems and their application in Diels-Alder/retro Diels-Alder reactions were recently reported.12 The approach presented in Scheme 2 is focused on the construction of the alkenyl ethers 7a,b which may be converted into the target molecules by ring-closing metathesis.

Reactions of 6h and 6i with allyl bromide in the presence of sodium hydride afforded 7a,b in good yield. However, treatment of the olefin substrates 7a,b with rutenium benzylidene complex (Grubbs’ catalyst I) (10 mol%) in 0.01 M solution of dichloromethane at reflux for 5 hours did not result in the formation of desired products; only starting 7a,b were recovered unchanged. It is well documented that amino group being present in the olefin containing substrate can coordinate to ruthenium catalyst and deactivated it. To overcome these difficulties conversion of amine to the corresponding amide or the addition of Lewis acids into the reaction mixture is recommended.13 The latter, binding nitrogen prior to its reaction with ruthenium catalyst may improve olefin metathesis. Therefore, RCM of compounds 7a,b were repeated in the presence of equivalent amounts of Ti(Oi-Pr)4 and B(Et2O)4, however, no expected products 8a,b could be isolated (Scheme 2).

Also acetylation of the exocyclic amine group in 7a into amide 7c and RCM reaction of the latter did not result in the formation of 8c (Scheme 2). On the other hand, 1,2,4-triazines are known to form stable complexes with transition metals including ruthenium.14 It seems likely that ring nitrogen atoms of compounds 7a,b are able to coordinate to ruthenium catalyst and prevent its RCM reaction. The colour of the RCM reaction mixture is deeply green what indicates that Grubbs' catalyst can be destroyed under these reaction conditions. In order to explain which nitrogen atoms in 7a and 7b are responsible for coordination to the ruthenium catalyst, pyridotriazine derivative 10 without the exocyclic amino group was prepared (Scheme 3) and subjected to RCM reaction under conditions described above. When the compound 10 containing 1,2,4-triazine ring was treated with first generation Grubbs’ catalyst, no reaction occurred (Scheme 3). This result clearly shows that nitrogen atoms of the 1,2,4-triazine part of 10 must deactivate ruthenium benzylidene complex, since its 2,2'-bipyridine analogue 12 easily undergoes RCM to give cyclophane 13 in high yield15 (Scheme 4).

This conclusion is supported by a crossover experiment. When compound 12 was treated with Grubbs’ catalyst in the presence of equivalent amount of 7a, 7b or 4 cyclophane 13 was not formed (Scheme 4). The deep green colour of the reactions mixtures again indicates that the catalyst is destroyed by the compounds 7a, 7b or 4.

Thus, in marked contrast to the 2,2'-bipyridine containing alkenyl ethers, their bi-1,2,4-triazine 7a-c or 5-(pyridyl-2-yl)-1,2,4-triazine 10 analogues do not undergo ring – closing metathesis by treatment with ruthenium benzylidene complex (Grubbs catalyst I). We suppose that the 1,2,4-triazine ring nitrogen atoms are responsible for deactivation of the catalyst. However, an influence of the nitrogen atoms of the exocyclic amine groups in compounds 7a,b on the catalyst is also possible.

References

1. Part 45 in "1,2,4-triazines in organic synthesis". For Part 44, see D. Branowska, J. Ławecka, W. Wysocki, and A. Rykowski, Arkivoc, 2008, (in print).
2. a) C. W. Lindley and M. E. Layton, 'Science of Synthesis: 1,2,4-Triazines,' Vol. 17, ed. by S. M. Weinreb, Thieme Verlag, Stuttgard, New York, 2004, pp. 357-447; b) H. Neunhoeffer, 'Comprehensive Heterocyclic Chemistry: 1,2,4-Triazines and their benzo derivatives,' Vol. 6, ed by A. R. Katritzky, Pergamon Press, Oxford, 1996, pp. 507-573; c) H. Neunhoeffer, 'Houben –Weyl: 1,2,4-Triazines,' 4th edition, Vol. E9c, ed. by E. Schaumann, Thieme Verlag, Stuttgard, New York, 1988, pp. 582-666.
3. a) V. N. Charushin, S. G. Alexeev, O. N. Chupahkin, and H. C. van der Plas, in 'Advances in Heterocyclic Chemistry: Behavior of Monocyclic 1,2,4-triazines in reactions with C-, N-, O-, and S- nucleophiles,' Vol. 46, ed. by A. R. Katritzky, Academic Press, Inc., London, 1986, pp. 73-142.
4. a) A. Rykowski and M. Mąkosza, Liebigs Ann. Chem., 1988, 36, 627; CrossRef b) D. N. Kozhevnikov, V. L. Rusinov, O. N. Chupakhin, M. Mąkosza, A. Rykowski, and E. Wolińska, Eur. J. Org. Chem., 2002, 1412; CrossRef c) A. Rykowski and E. Wolińska, Tetrahedron Lett., 1996, 37, 5795; CrossRef d) A. Rykowski, E. Wolińska, and H. C. Van der Plas, J. Heterocycl. Chem., 2000, 37, 879; CrossRef e) A. Rykowski, E. Wolińska, and D. Branowska, and H. C. Van der Plas, Arkivoc, 2004, , MM-942G, 74; f) A. Rykowski, E. Guzik, M. Mąkosza, and W. Holzer, J. Heterocyclic Chem., 1993, 30, 413; CrossRef g) A. Rykowski, D. Branowska, M. Mąkosza, and P. Van Ly, J. Heterocycl. Chem., 1996, 33, 1567. CrossRef
5. a) D. K. Krass and W. W. Paudler, J. Heterocycl. Chem., 1974, 11, 43; CrossRef b) D. K. Krass and W. W. Paudler, J. Heterocycl. Chem., 1973, 10, 343; CrossRef c) A. Rykowski and H. C. Van der Plas, Recl. Trav. Chim. Pays-Bas, 1975, 94, 204; d) D. Shamroc, US patent 4105434, 1978; e) D. Branowska and J. Kielak, Polish J. Chem., 2003, 77, 1149, and references cited therein.
6. D. Branowska and B. Iwańska, Molbank, 2004, M372. CrossRef
7. a) J. Daunis and C. Pigiere, Bull., Soc., Chim., Fr., 1973, 2493; b) A. Rykowski, to be published.
8. For monoamino 1,2,4-triazine drugs: 3-amino-1,2,4-benzotriazine 1,4-dioxide (Tirapazamine), 3-amino-6-[(5-nitro-2-furyl)vinyl]-1,2,4-triazine (Panfuran), and diamino 1,2,4-triazine drug 3,5-diamino-6-(2,3-dichlorophenyl)-1,2,4-triazine (Lomatrigine) see ref. 1b.
9. Biological evaluation of the compounds will be published elsewhere .
10. W. W. Paudler and T.-K. Chen, J. Heterocycl. Chem., 1970, 7, 767. CrossRef
11. D. Branowska, Molecules, 2005, 10, 265. CrossRef
12. J. Ławecka, E. Olender, P. Piszcz, and A. Rykowski, Tetrahedron Lett., 2008, 49, 723. CrossRef
13. Q. Yang, W.-J. Xiao, and Z. Yu, Org. Lett., 2005, 7, 871, and references cited therein. CrossRef
14. X.-L. Hong, H. Chao, L.-J. Lin, K.-Ch. Zeng, H. Li, X.-L. Wang, F.-C. Yun, and L.-N. Ji, Helv. Chim. Acta, 2004, 87, 1180. CrossRef
15. D. Branowska and A. Rykowski, Tetrahedron, 2005, 61, 10713. CrossRef
16. D. Branowska, Tetrahedron, 2004, 60, 6021. CrossRef