Letter pubs.acs.org/OrgLett
Synthesis of N,N‑Dioxopyridazines Shlomo Rozen* and Avshalom Shaffer School of Chemistry, Tel Aviv University, Tel Aviv 69978, Israel S Supporting Information *
ABSTRACT: Despite many efforts, one of the smallest heterocycles containing two nitrogen atoms, pyridazine, could not be converted to its N,N-dioxide (see, however, Nakadate et al. Chem. Pharm. Bull. 1970, 18, 1211−1218). HOF· CH3CN, made easily from diluted fluorine, was able to accomplish this task in a fast reaction with good yields.
T
CPBA.14 It was not surprising that HOF·CH3CN could also quantitatively form 7 from either 6 or 6a in a matter of seconds. Even pyrazines with two EWG could be readily oxidized despite the fact that the nitrogen atoms are somewhat less basic. Thus, 2,5-dibromopyrazine (8) was converted into 2,5-dibromo-N,Ndioxopyrazine (9) at room temperature in a matter of seconds in quantitative yield (Scheme 2).
ertiary amine oxides are a class of compounds that are gaining increasing importance. They are used in a variety of processes and final products, such as fiber preparation,1 hair tonics,2 topical pharmaceuticals,3 cellulose solutions,4 and more.5 In addition, certain poly-N-oxides are considered highly energetic materials.6 In many cases, however, it is not easy to prepare heterocyclic N-oxides, which contain more than one nitrogen atom, as is evident from 1,10-phenanthroline N,Ndioxide (1), 4,5-diazafluorenone N,N-dioxides (2), 1,8-diazafluoren-9-one (DFO) N,N-dioxide (3),7 or tetrazole 3-N-oxide (4).8 Pyridazine and its derivatives are yet another family of those “stubborn” heterocycles which were practically impossible to fully oxidize to form pyridazine N,N-dioxide (5) and its derivatives (Scheme 1).
Scheme 2
Scheme 1
The picture is completely different within the pyridazine series. Applying m-CPBA, AcOOH, or H2O2 (30%) yielded pyridazine mono-oxide only.15 A plausible explanation for this behavior is the difference in the electron densities around the nitrogen atoms in the two series. By using Gaussian programs, with a cc-pVQZ basis set and the RB3LYP version of the DFT suite, we have calculated the classical Mulliken electron densities on the nitrogen atoms of pyrazine and its mono-Noxide to be −0.28. The basicity of the nitrogen atoms in pyridazine, on the other hand, is considerably lower (−0.169 for pyridazine itself) and goes further down for the mono-oxide (−0.159), making oxygen transfer to both nitrogen atoms a quite difficult task indeed. There is only one paper, published almost 50 years ago,16 describing the reaction of this family derivatives with boiling 50% and 90%(!) H2O2 resulting in the corresponding bis-dioxide in yields of 0.7−1.1% (Scheme 3). Such poor results were explained later in a theoretical paper which predicted lengthening of the N−N bond would result in a ring opening.17
Years ago, we initiated the introduction of elemental fluorine to the realm of organic synthesis. It should be stressed that fluorine is less toxic than chlorine or bromine9 and it cannot spread into large areas. No fatal accidents were reported with it in the last 50 years. This halogen has proved to be a very unique and selective reagent in its own right,10 but in addition, it was somewhat surprising to find that it could be a tool for creating many secondary reagents, which were able to perform unprecedented chemistry. Among those reagents, one can find AcOF,11 BrF3,12 and especially HOF·CH3CN, which has become an indispensable tool to many chemists.13 The last reagent possesses a strong electrophilic oxygen and proved to be one of the best oxygen transfer agents organic chemistry has to offer. In this work, we employ this reagent to address a standing problem of more than half a century of transferring oxygen atoms to some members of the diazine family. Attaching oxygen atoms to pyrazine (6) or pyrazine monooxide (6a) to form pyrazine N,N-dioxide (7) is an easy task achievable in good yields by reactions with 30% H2O2 or m© 2017 American Chemical Society
Received: July 27, 2017 Published: August 24, 2017 4707
DOI: 10.1021/acs.orglett.7b02310 Org. Lett. 2017, 19, 4707−4709
Letter
Organic Letters Scheme 3
Scheme 5
The HOF·CH3CN complex is easily made by passing nitrogen-diluted fluorine through aqueous acetonitrile. While it is quite unusual for an oxygen to act as an electrophile, such a mode is not surprising when it is bonded to the most electronegative element of the periodical table, the fluorine. What is more, the reactions are especially encouraged by the thermodynamically favored HF formation. It usually reacts with substrates under very mild conditions such as 0 °C and very short reaction times,13 suppressing the chances for decompositions. When this reagent is reacted with pyridazine (10), the practically elusive N,N-dioxopyridazine (5) was obtained in 80% yield almost instantaneously. The balance of the reaction was the mono-oxide 11. In light of the fact that this compound is practically unknown despite the efforts to prepare it with various oxidants, we have submitted it to X-ray diffraction proving that the two oxygen atoms were indeed attached to the two nitrogen ones (Scheme 4). Scheme 4. X-ray of 5a
a
electron density around the nitrogen atoms, and again only a mixture of 1:1 N-mono-oxides 36 and 37 was obtained. There are two points that must be addressed. The first is based on the fact that unlike pyrazine it was not possible to convert any mono-oxide of the pyridazine series to its dioxide counterpart, even when a large excess of the HOF·CH3CN complex was used, and the second focuses on our inability to connect two oxygen atoms to the two nitrogen ones in the pyrimidine series, although mono-oxygenation was accomplished in seconds and in practically quantitative yields (Scheme 6). We believe these two issues are related.
CCDC deposition no.:1537826. The cif file can be found in the SI.
Scheme 5 presents the scope and limitation of this reaction. Methyl-substituted compounds such as 3-methyl- and 3,6dimethylpyridazine (12 and 13) were successfully converted to the corresponding N,N-dioxide derivatives 14 and 15 with the balance again being the mono-oxides 16 and 17. Even when the methyl was not located near any nitrogen atom as in 4methylpyridazine 18 the dioxide 19 was obtained in 80% yield with the rest being again the mono-oxides 20 and 21. When one of the methyl groups in 13 was replaced with an electronwithdrawing chlorine atom, 6-chloro-3-methylpyridazine (22) did not change the basicity of the nitrogen atoms was not changed, and the corresponding N,N-dioxide 23 was obtained in good yield, the mass balance being the mono-oxide 24. Removing the electron-donating methyl caused some reduction of the yields, and 3-bromopyridazine (25) formed the N,Ndioxo 26 in 75% yield, while 3-cyanopyridazine (27) with its stronger electron-withdrawing group produced the N,N-dioxo compound 28 in 20% yield only. In both cases, the balance of the reactions was the N-mono-oxides 29 and 30. Placing two EWG on the pyridazine skeleton proved to be too much even for the HOF·CH3CN. The low nitrogen basicity of 3,6dichloropyridazine (31) or even of 3,6-dibromopyridazine (32) was the reason that only the mono-oxides 33 and 34 were obtained in quantitative yields regardless of how much excess HOF·CH3CN was used. Adding an electron-donating group as in 3,6-dichloro-4-methylpyridazine (35) did not change the
Scheme 6
Obviously, after a mono-N-oxygenation takes place within the pyrazine series it does not affect the basicity of the second nitrogen atom because of the two resonance structures of the NO group and the dioxides 7 and 9 were obtained almost quantitatively. This is not the case with pyrimidines, and after a mono-oxygenation the second nitrogen is no longer available for any electrophilic attack. From an electronic point of view, the situation within the pyridazine family is very similar, and indeed, any mono-N-oxide of this group could not be further oxidized to the N,N-dioxide. This situation brings to mind the reaction of thiophenes or sulfides,18 even very deactivated ones, with HOF·CH3CN to form the SO2 moiety rather than the 4708
DOI: 10.1021/acs.orglett.7b02310 Org. Lett. 2017, 19, 4707−4709
Organic Letters
■
sulfoxide one. The reason for such behavior is the hydrogen bond existing between a couple of HOF molecules and the weak complexation between them and acetonitrile.19 Thus, two HOF molecules approach the sulfur atom of a sulfide, or as in our case to a nitrogen atom, but only in pyridazines is the second reagent molecule near the second nitrogen atom, and a simultaneous attack takes place on both heteroatoms to form the corresponding N,N-dioxide (option A in Scheme 7). This
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. ORCID
Shlomo Rozen: 0000-0002-8034-9530 Notes
The authors declare no competing financial interest.
■ ■
Scheme 7
ACKNOWLEDGMENTS This work was supported by the Israel Science Foundation REFERENCES
(1) Blaschke, G.; Kleber, R. Ger. Offen. DE 3,545,152, 1987; Chem. Abstr. 1987,107, 156320t. (2) Uchikawa, K.; Miyazawa, K.; Nakanishi, J.; Ishino, A. US Patent US 5,156,836, 1992; Chem. Abstr. 1993, 118, 260685d. (3) Cerecetto, H.; Gonzalez, M.; Lavaggi, M.; Azqueta, M.; Lopez de Cerain, A.; Monge, A. J. Med. Chem. 2005, 48, 21. (4) Rabideau, R. D.; Ismail, A. E. J. Phys. Chem. B 2015, 119, 15014− 15022. (5) Babu, N. J.; Reddy, L. S.; Nangia, A. Mol. Pharmaceutics 2007, 4, 417−434. (6) He, P.; Zhang, J. G.; Yin, X.; Wu, J. T.; Wu, L.; Zhou, Z. N.; Zhang, T. L. Chem. - Eur. J. 2016, 22, 7670−7685. (7) Harel, T.; Shefer, N.; Hagooly, Y.; Rozen, S. Tetrahedron 2010, 66, 3297−3300. (8) Harel, T.; Rozen, S. J. Org. Chem. 2010, 75, 3141−3143. (9) American Environmental Group, Ltd. AEGL (Acute Exposure Guideline Level) Oct 2, 2009, 50. (10) (a) Rozen, S.; Gal, C. J. Org. Chem. 1987, 52, 4928−4933. (b) Gal, C.; Rozen, S. Tetrahedron Lett. 1984, 25, 449−452. (11) Hebel, D.; Rozen, S. J. Org. Chem. 1988, 53, 1123−1125. (12) Rozen, S.; Mishani, E.; Bar-Haim, A. J. Org. Chem. 1994, 59, 2918. (13) (a) Kol, M.; Rozen, S. J. Org. Chem. 1993, 58, 1593−1595. (b) Chavez, D. E.; Parrish, D. A.; Mitchell, L. Angew. Chem., Int. Ed. 2016, 55, 8666. (c) Wei, S.; Xia, J.; Dell, E. J.; Jiang, y.; Song, R.; Lee, H.; Rodenbough, P.; Briseno, A. L.; Campos, L. M. Angew. Chem., Int. Ed. 2014, 53, 1832−1836. (d) Cochran, J. E.; Amir, E.; Sivanandan, K.; Ku, S. Y.; Seo, J. H.; Collins, B. A.; Tumbleston, J. R.; Toney, M. F.; Ade, H.; Hawker, C. J.; Chabinyc, M. L. J. Polym. Sci., Part B: Polym. Phys. 2013, 51, 48−56. (e) McPake, C. B.; Murray, C. B.; Sandford, G. Tetrahedron Lett. 2009, 50, 1674−1676. (f) Rozen, S. Eur. J. Org. Chem. 2005, 2005, 2433−2447. (14) Cowden, W. B.; Jacobson, N. W. Aust. J. Chem. 1980, 33, 131− 135. (15) Kanchupalli, V.; Joseph, D.; Katukojvala, S. Org. Lett. 2015, 17, 5878−5881. (16) Nakadate, M.; Sueyoshi, S.; Suzuki, I. Chem. Pharm. Bull. 1970, 18, 1211−1218. (17) Aitken, R. A.; Fodi, B.; Palmer, M. H.; Slawin, A. M. Z.; Yang, J. Tetrahedron 2012, 68, 5845−5851. (18) Beckerbauer, R.; Smart, B. E.; Bareket, Y.; Rozen, S. J. Org. Chem. 1995, 60, 6186−6187. (19) (a) Rozen, S.; Bareket, Y. J. Org. Chem. 1997, 62, 1457−1462. (b) Srnec, M.; Oncak, M.; Zahradnik, R. J. Phys. Chem. A 2008, 112, 3631−3637. (20) Igeta, H. Chem. Pharm. Bull. 1960, 8, 559−560. (21) (a) Henkelis, J. J.; Barnett, S. A.; Harding, L. P.; Hardie, M. Inorg. Chem. 2012, 51, 10657−10674. (b) Sarma, R.; Boudalis, A. K.; Baruah, J. B. Inorg. Chim. Acta 2010, 363, 2279−2286.
mechanism was supported by an experiment where MeOH was deliberately added to the reaction mixture, thus breaking the hydrogen bonds between the reagent molecules and creating new ones with the methanol. The result was a quantitative formation of pyridazine mono-oxide 11 independent of the concentration of the reagent and the reaction times (option B in Scheme 7). The distance between the heteroatoms in the pyrimidine series does not allow a complexation with a pair of the reagent molecules, resulting in mono-oxidation only (Scheme 6). This simultaneous attack of both nucleophilic nitrogens on the receptive nearby oxygens is responsible for the fact that 3chloropyridazine (38) forms only the mono-oxide 3920 (Scheme 5, last entry). Although chlorine is a less powerful electron-withdrawing group than the cyan in 27, the yield of the dioxide is practically 0%. The chlorine atom is bulkier than the linear CN, interfering with the second molecule of the HOF dimer to attack simultaneously the additional nitrogen situated near the chlorine atom. In conclusion, for many decades, it was practically impossible to oxidize the two nitrogen atoms of one of the simplest heterocycle systems, the pyridazine family. Elemental fluorine, through one of its offspring reagents, HOF·CH3CN, succeeded where all other oxygen-transfer agents have failed. We believe that this dioxide may find many uses in the near future including serving as a new ligand in many studies.21
■
Letter
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.orglett.7b02310. X-ray data for compound 5 (CIF) General experimental procedures, X-ray structure, and spectroscopic data for all new compounds (PDF) 4709
DOI: 10.1021/acs.orglett.7b02310 Org. Lett. 2017, 19, 4707−4709
