Mendeleev Communications Electronic Version, Issue 5, 1998 (pp. 169–205) A porphyrin chlorination reaction Andrei F. Mironov,* Valentina D. Rumyantseva and Olga N. Ponamoreva M. V. Lomonosov Moscow State Academy of Fine Chemical Technology, 117571 Moscow, Russian Federation. Fax: +7 095 434 8711 Treatment of nickel and palladium porphyrin complexes with thionyl chloride readily affords products of meso- and b-chlorination; further reaction leads to chlorination of macrocyclic methyl groups.A reaction for the chlorination of porphyrin metal complexes has been found. In this study, instead of traditional chlorinating agents (e.g. HCl and hydrogen peroxide,1,2 sulfuryl chloride3 and chlorosulfonic acid4), we tried thionyl chloride, as used in the case of tetraazaporphyrin.5 This reagent is known to readily replace the hydroxyl group in alcohols and carboxylic acids but has scarcely been used in reactions involving C–H linkages.6 Thionyl chloride has been widely used in porphyrin chemistry for transforming carboxylic acids into the corresponding chlorides.However, on attempting the activation of the carboxylic acids in palladium coproporphyrin III with thionyl chloride, we noted an abrupt colour change from orange–red to deep green. The electron absorption spectrum of the product showed a significant bathochromic shift of both the Soret band and the a- and b-bands, with the intensity ratio of the a- and b-bands being considerably lower, evidencing disturbance of the porphyrin macrocycle.Unfortunately, full characterisation of the compound thus obtained was unsuccessful, presumably due to the formation of a highly reactive chloride.Further study of this reaction was performed using the palladium coproporphyrin III tetramethyl ester 1a. This was dissolved in SOCl2, kept for 2 h at 20 °C, poured into ice and the resulting precipitate was filtered to give almost pure products (TLC assay) in 92% yield.Prior to elemental analysis, the porphyrin obtained was passed through alumina and recrystallised from chloroform–methanol. Elemental analysis data showed that the product contained four additional chlorine atoms. The mass spectrum also provided evidence in favour of four chlorine atoms (m/z 952).† The 1H NMR spectrum showed the disappearance of four meso-protons. Therefore, the structure 2a was assigned to the new compound.‡ In the case of palladium deuterioporphyrin 1b, not only the meso-protons, but also both b-positions were replaced to give the hexachloro-substituted porphyrin 2c.§ This showed an even greater colour change and bathochromic shift of absorption bands.Similar behaviour was observed for nickel porphyrins. In the case of coproporphyrin 3a, the meso-tetrachloro-derivative 4a¶ was obtained at 4 °C for 15 min in 93% yield.Deuterioporphyrin 3b was transformed into the hexachloro-derivative 4c.†† However, nickel porphyrins demonstrated higher reactivity. Prolonged treatment of 3a led to substitution of not only the meso-protons, but also the methyl groups. Heating for 1 h † Mass spectra were measured on a MSBKh instrument (SELMI, Sumy, Ukraine).Ionisation was effected by 252Cf fission products and a timeof- flight monitoring ion analyser was employed. ‡ Data for 2a, methyl ester: mp 199–202 °C. 1H NMR (CDCl3) d: 4.08 (t, 8H, CH2CH2CO2Me), 3.81 (s, 6H, COOMe), 3.78 (s, 6H, COOMe), 3.26 (s, 6H, Me), 3.24 (s, 6H, Me), 2.95 (m, 8H, CH2CH2CO2Me). UV [CHCl3, lmax/nm (e×10–3)]: 440 (146), 561 (9.6), 606 (6.4).MS, m/z: 952 (M+). Found (%): C 50.23, H 4.18, Cl 15.41, N 5.81. Calc. for C40H40Cl4N4O8Pd (%): C 50.41, H 4.23, Cl 14.88, N 5.88. § Data for 2c, ethyl ester: mp >300 °C. 1H NMR (CDCl3) d: 4.22 (m, 8H, CH2CH2CO2Me and CH2Me), 3.30 (s, 12H, Me), 3.00 (t, 4H, CH2CH2CO2Me), 1.30 (t, 6H, CH2Me). UV (CHCl3, lmax/nm): 452, 561 (b), 618 (a) (a/b = 0.70). MS, m/z: 878 (M+).¶ Data for 4a: mp 111–113 °C. 1H NMR (CDCl3) d: 4.41 (m, 8H, CH2CH2CO2Me), 3.78 (s, 12H, COOMe), 3.24 (s, 12H, Me), 2.92 (m, 8H, CH2CH2CO2Me). UV [CHCl3, lmax/nm (e×10–3)]: 442 (121), 577 (9.7), 621 (5.0). MS, m/z: 905 (M+). Found (%): C 53.44, H 4.24, Cl 15.01, N 5.81. Calc. for C40H40Cl4N4O8Ni (%): C 53.07, H 4.45, Cl 15.67, N 6.19. resulted in chlorination of all four methyl group to produce the octachloroporphyrin 5.‡‡ Shortening the reaction time makes it †† Data for 4c: mp 164–166 °C. 1H NMR (CDCl3) d: 4.14 (m, 4H, CH2CH2CO2Me), 3.77 (s, 6H, COOMe), 3.23 (s, 3H, Me), 3.18 (s, 3H, Me), 3.14 (s, 3H, Me), 3.12 (s, 3H, Me), 2.83 (m, 4H, CH2CH2CO2Me). UV [CHCl3, lmax/nm (e×10–3)]: 442 (156), 585 (7.0), 632 (4.9). MS, m/z: 802 (M+). Found (%): C 47.53, H 3.75, N 6.63.Calc. for C32H26Cl6N4O4Ni (%): C 47.92, H 3.27, N 6.99. ‡‡ Data for 5: mp 120–122 °C. 1H NMR (CDCl3) d: 5.75 (s, 4H, CH2Cl), 5.71 (s, 4H, CH2Cl), 4.12 (m, 8H, CH2CH2CO2Me), 3.76 (s, 12H, COOMe), 3.05 (m, 8H, CH2CH2CO2Me). UV [CHCl3, lmax/nm (e×10–3)]: 459 (130), 593 (10.4), 641 (6.6). MS, m/z: 1042 (M+). Found (%): C 45.64, H 3.39, Cl 26.91, N 5.12. Calc. for C40H36Cl8N4O8Ni (%): C 46.06, H 3.48, Cl 27.19, N 5.37.N N N N CO2R' CO2R' Me Me Me R Me R Pd N N N N CO2R' CO2R' Me Me Me R Me R Pd Cl Cl Cl Cl SOCl2 1a,b 2a,c N N N N CO2Me CO2Me Me Me Me R Me R Ni 3a,b SOCl2 N N N N CO2Me CO2Me Me Me Me R Me R Ni 4a,c Cl Cl Cl Cl N N N N CO2Me CO2Me Ni 5 Cl Cl Cl Cl Cl Cl CO2Me Cl Cl CO2Me R' = Me, Et a R = CH2CH2CO2Me b R = H c R = Cl Scheme 1Mendeleev Communications Electronic Version, Issue 5, 1998 (pp. 169-206) possible to obtain products with partially chlorinated methyl groups: we succeeded in isolating heptachloro-substituted nickel coproporphyrin III tetramethyl ester 6.§§ The number of chlorine atoms in 4a, 4c, 5 and 6 was proved by mass spectrometry, 1H NMR spectroscopy and elemental analysis data. The presence of a transition metal ion in the porphyrin molecule presumably plays the determining role in this reaction.On treatment of deuterioporphyrin dimethyl ester and coproporphyrin tetramethyl ester with thionyl chloride under similar conditions, no chlorination was observed. It seems likely that coordination of thionyl chloride with a central metal atom followed by a chain of redox reactions leads to the reduction of sulfur and to the formation of a highly reactive chlorine species (possibly, the chlorine radical), which attacks the macrocycle.Finding crystalline sulfur in the reaction mixture obtained during the synthesis of 2c provides evidence in favour of this assumption. §§ Data for 6: mp 190–192 °C. 1H NMR (CDCl3) d: 5.76 (s, 6H, CH2Cl), 4.18 (m, 8H, CH2CH2CO2Me), 3.77 (s, 12H, COOMe), 3.25 (s, 3H, Me), 3.00 (m, 8H, CH2CH2CO2Me).UV [CHCl3, lmax/nm (e×10–3)]: 448 (157), 582 (13.1), 627 (8.5). MS, m/z: 1007 (M+). Found (%): C 47.31, H 3.93, Cl 24.31, N 5.93. Calc. for C40H37Cl7N4O8Ni (%): C 47.63, H 3.70, Cl 24.60, N 5.56. This work was supported by the Russian Foundation for Basic Research (grant no. 96-15-97-709). References 1 H. Fischer and W. Klendauer, Ann. Chem., 1941, 547, 123. 2 R. Bonnett, P. Brewer, K. Noro and T. Noro, Tetrahedron, 1978, 34, 379. 3 D. Dolphin, The Porphyrins, Academic Press, New York, 1978, vol. 2, p. 153. 4 E. Samuels, R. Shuttleworth and T. S. Stevens, J. Chem. Soc. C, 1968, 145. 5 O. G. Khelevina, S. V. Timofeeva and B. D. Berezin, Zh. Org. Khim., 1994, 30, 295 (Russ. J. Org. Chem., 1994, 30, 312). 6 L. Fieser and M. Fieser, Reagents for Organic Synthesis, Wiley, New York, 1968. Received: Moscow, 23rd March 1998 Cambridge, 1st June 1998; Com. 8/02397B