Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.chem.msu.ru/eng/misc/babaev/papers/005e.pdf Дата изменения: Fri Jan 21 12:17:02 2011 Дата индексирования: Sat Feb 12 03:03:24 2011 Кодировка:

STRUCTURE AND AMBIPHILIC REACTIVITY OF INDOLIZINES. 2.* 8(6)-ACETYL- AND CYANOINDOLIZINES S. I. Bobrovskii, E. V. Babaev, S. P. Gromov, K. A. Paseshnichenko, and Yu. G. Bundel'

UDC 547.754.04:543.422

Quaternization of 2-methyl-3(5)-acetyl and cyanopyridines with =-bromoketones followed by cyclocondensation has given a number of 8(6)-acetyl- and cyanoindolizines.

The effects of substituents on the reactivity of indolizines have received little attention. This is particularly true of indolizines substituted in the pyridine ring, for the synthesis of which by the most widely-used Chichibabin reaction [2] difficultlyaccessible substituted alkylpyridines are required. At the present time, the Chichibabin reaction has given 6- and 7-cyano-, 6-ethoxycarbonyl and 6-carbamoylindolizines [3], 7- and 8-ethoxycarbonyl [4, 5], and some 6- and 8-nitroindolizines [6], the latter having been subjected to isomerization recyclization to give 5- and 7-nitroindoles respectively [7]. In the synthesis of 8-cyano- [8] and 1,8-dicyanoindolizine [9] and 6-cyano-l-nitro-2-methylindolizine [i0], other methods of construction of the indolizine ring have been employed. However, with the exception of the work of Loseva et al. [4, II, 19] on 6-, 7-, and 8- ethoxycarbonylindolizines, the effects of substituents in different positions of the pyridine ring on the structure and reactivity of indolizines have not been examined. Likewise, no studies have been made on the possible isomerizational recyclization of indolizines with other electronacceptor substituents (other than the NO 2 group) in positions 6 or 8. It is noteworthy that 6(8)-acylindolizines have not been described. 6- and 8-Acetyl and cyanoindolizines were obtained by the Chichibabin method from ~methyl- ~(~')-acetyl and cyanopyridines and ~-bromoketones.
X

i

Br

la-c,g

CH2COR II.a~d-g :.

ma-g

a-c

R=CH3, d--g R=C6Hs; a f. X=CN, b,e ,g.~ X=H,c,d Y=H,b,,e Y=COCH3, g Y=CN

X=COCHa; a,c,d.f

Treatment of the salts (IIa, d-g), isolated in the pure state, with bases has given good yields of the indolizines (Tables 1 and 2). The bromoacetonylates (IIb, c), which could not be purified, were subjected to cyclocondensation without isolation. (The acetonylates of 3-nitro- [6], and of 3- and 5-ethoxy-2-picolines [24, 25], likewise could not be obtained pure, unlike the corresponding bromophenacylates). The compounds were purified by column chromatography. The indolizines (IIIa-g) were orange or yellow in color. In the IR spectra of indolizines (IIlb-e), the absorption for C-O stretching (1672-1675) corresponded to that in ~, ~-unsaturated ('%1675) rather than aromatic ketones (%1690 cm'1), perhaps in consequence of considerable alternation in bond lengths in the pyridine moiety of indolizine (see, e.g., [12]). It is also noteworthy that in I- and 3-acetylindolizines, in which greater conjugation of the acetyl group with the ~-excessive pyrrole ring must occur, the values of the stretching vibrations of the C~O group are smaller, at 1595-1625 cm -! [5]. In the UV spectra (Table 3) of 6- and 8-indolizines, three groups of absorption bands may be discerned (characteristic of indolizines in general), namely at short (221-285), *For Part i, see [i]. M. V. Lomonosov Moscow State University. Translated from Khimiya Geterotsiklicheskikh Soedinenii, No. 2, pp. 209-213, February, 1987. Original article submitted July 25, 1985.

0009-3122/87/2302-0175512.50

~ 1987 Plenum Publishing Corporation

175


TABLE I.
:om,ound

Properties of Salts (II) Found, % Empirical formula C
46,3 57,5 57,3 56,8 56,5

Reacticr Mp, deg time, h C
10 20 15 20 25 200--201 154--I56 199--20[ 183--185 220--222

Calc., %
c H
4,3 4,8 4,8 4,1 4,1

Yield %

H
4,3 5,3 5,2 4,1 4,3 C,oH,IBrN20 Ct6H,6BrNO2 CI~H~BrNO~ C,sHI3BrN20

Ila ITd lie llf llg

CIsH,3BrN20

47,1 57,5 57,5 56,8 56,8

40 75 72 41 40

TABLE 2.
COm -

Properties of Indolizines (III)
C
M+

pound
Ilia IIIb llIc IIId IIIe IIIf IIIg

Mp, deg

Vc=O,Cm"

II

VC___ N ,cm-

I Yield,
I00 16" 98 57 52 65 70

35--37 32--33 15--16 163--165 188--190 163--165 194---196

156
173 173 235 235 218 218 1672 1675 1675 1675

2213

2235 2226

*Yield based on starting acetylpyridine (Ib). medium (300-353), and long wavelengths (Imax > 373 nm). As compared with the unsubstituted 2methyl- and 2-phenylindolizines, all the bands undergo a bathochromic shift. This shift is greatest for the long-wavelength band, and it increases in the order CN < COCH 3 < NO= both for the 6- and the 8-substituted indelizines (6-substituted 2-phenylindolizines are exceptiona this band appearing as a shoulder in the region 340-450 nm). In the case of the 8-isomers, the long-wavelength region of the spectrum (the first and second absorption bands) undergoes a bathochromic shift as compared with those of the 6-isomers, whereas the short wavelength band is usually shifted towards shorter wavelengths. In the case of 2-methyl-8-substituted indolizines (Ilia, c), the UV spectra are characterized by the appearance of the short-wavelength band appearing as a doublet, which does not occur in the 6-isomers, nor in the 2-phenyl derivatives. A similar double maximum has been observed previously in the methyl derivatives of 8-, but not 6-nitroindolizines [i]. Some of the features of the UV spectra of substituted indolizines may be rationalized in terms of quantum chemical calculations of the indolizine molecule. Using the MO LCAO method, the following values of the ~-charges in particular have been obtained for the ground and firs excited singlet states of indolizine: for C(6 ) 1.00 and 1.08, and for C(8) 0.79 and 1.13 respectively [13]. It has been shown [14] that the long-wavelength band is almost exclusively due (to the extent of 89%) to the HOMO-LUMO transition (of the ~ + ~*-type). It is further suggested that electron-acceptor substituents (COCH3, CN, NO 2) in positions 6 and 8 of indolizinc influence only the energies of the molecular orbitals, and do not introduce their own absorption bands (in other words, they are auxochromes rather than chromophores). Since the H-charges at C(6 ) and C(a ) are increased on passing from the ground to the first excited state but to a greater extent for C(,) than for C(s), it may be concluded (cf. [6]) that electronacceptor substituents in positions 6 and 8 of the indolizine nucleus should deepen the color as compared with pyridine ring-unsubstituted 2-methyl- or 2-phenylindolizines, the 8-isomers being more deeply colored than the 6-isomers, and the Imax values should increase in the sequence COCH 3 < CN < NO2, in accordance with the electronegativity sequence as measured, for example, by their a constants [15]. These conclusions have in general been confirmed experimentally. In the cyano- and acetylindolizines the long-wavelength bands are close, although the bathochromlc shift is somewhat greater in the latter. It is notable that this behavior is not in conflict with the UV spectra of 2-methyl-6- and -8-ethoxycarbonylindolizines (dr. Fig. 1 in [ii]). The PMR spectra of the indolizines prepared here (Table 4) confirm their structures. The 8-acetylindolizine (lllc) shows a low-field shift of the signal for I-H under the influenc, of the magnetically anisotropic substituent in the peri-position. A similar perishift is

176


TABLE 3. Groups of Bands in the UV Spectra of Substituted Indolizines (in Ethanol).
Compound* ~'max' rim Ig 8

IIih[23]** lIlf IIld III i lIIg IIIe IIlj lllk Ill a IIIb Ill s Ilia film

254,5, 325. 350, 366, 384.5 260, 336, shoulder. 370--420 256, 350, shoulder. 415--460 248, 378, shoulder. 450--500 271, 300--320, shoulder 340--435 285, 308--322, shoulder 360--435 242, 313, shoulder. 350--450 240, 287, 291, 299, 348 248, 265, 315, 321, 394 254, 262, 334, 403 258, 265, 353, 480 221, 274, shoulder. 310, 322, 373 226, 303, 436

4,62; 4,61; 4,60; 4,62; 4,81; 4,72; 4,35; 4,50; 4,62; 4,49; 4,17; 4,41; 4,53;

3,27; 3,45; 3,43; 3,15 3,95; 3,3--3,0 2,81; 3,3--3,1 3,85; 3,4 3,99--3,90; 3,78--3,70 4,18--4,04; 3,20--3,03 4,26; 3,90--3,40 3,71; 3,73; 3,75; 3,63 4,66; 3,62; 3,72; 3,28 3,48; 2,81; 2,51 4,18; 3,55; 3,33 4,85; 4,15; 3,97; 2,92 4,51; 3,36

*(lllh-k), R = CsHs; (lllk-m), R = CHs; (llli, l), X = NO 2, Y = H; (IIIJ, m), X = H, Y = NO2; (IIlh, k), X = Y = H. **In cyclohexane. TABLE 4. Cations
Compound

PMR Spectral Data for Indolizines (III) and Their Chemical shifts,6 , ppm

Soivent I-H
CC14 CF3COOH CC14 CF3COOH CC14 CF~COOH CF3COOH CF~COOH CF~COOH CF~COOH 6,48 7,03 6,09 7,66 7,05 7,60 8,36
2-R

3-H

5-H

6-H

7-H

8-H

IIIa IIIb IIIc IIId IIIe llIf Illg

2,30 2,36 2,30 3,41

2,31

7,8--8,1 7,6--7,8
7,6--8,0

2,83

7,5--7,9

7,17 5,50 6,90 6,29 7,05 5,30 6,06 6,15 6,18 6,13

7,95 9,05 8,34 10,23 7,85 9,00 9,16 9,70 9,25 9,48

6,36

7,73

2,16* 3, i4' 6,28 7,80 7,8--8,1 2,95, 7,85

6,97 8,60 &90 9,53

8,72
2,31" 2,43* 2,98* 8,20 8,19

7,22

--

9,00 9,09 9,04 8,78 8,65

*For the acetyl CH 3 group. seen in 8-ethoxycarbonyl-[5] and 8-nitroindolizines [i, 6], but not in 8-cyano-derivatives. The spectra of 8-cyano [8] and 2-methyl-8-cyanoindolizine are similar*. We have attempted to carry ont the base-catalyzed rearrangement of acetylindolizines to the indoles, by analogy with nitroindolizines [7], but it was found that under isomerizational recyclization conditions the indolizines (IIIc-e) were recovered unchanged. Nor were anionic o-complexes, such as those observed previously with nitroindolizines [7], formed. It is likel5 that the addition of the HO- ion followed by recyclization requires the presence in the 6- or 8-position of the indolizine of a nitro-group, or at least a substituent whose electronegativit is no smaller than this. The protonation of the acetyl- and cyanoindolizines (III) in CFsCO2H has been examined by PMR. The appearance in the spectrum of a singlet of double intensity at 5.3-6.3 ppm (Table 4) unambiguously demonstrates [18] that protonation takes place in the 3-position:

X m

R

X W

R

*Note that Jutz et al., the quadruplet at T 2.56 smaller value of 3j (2.8 However, in indolizines proton is seen at 6.72,

[8] ppm Hz) 3Jl2 and

inaccurately assigned the (6 7.44 ppm) to the I-H [for the quadruplet at 9 > 3J23 (cf. e.g., [17]), that for 3-H at 7.44 ppm.

signals for I-H and 3-H, attributing proton, probably on the grounds of the 3.28 ppm (8 6.72 ppm) 3j = 4.0 Hz]. and therefore the signal for the I-H

177


In the case of the 8-substituted cations (IV), a low-field the influence of the peri-oriented CH3CO group is also seen, but CN group. Protonation at C(,) has been observed previously in [19], and nitroindolizines [I], and hence the introduction of the 6- or 8-position does not affect the mode of protonation of EXPERIMENTAL

shift of the I-H proton under this is not the case with the carbamoyl- [3], ethoxycarbonylelectron-acceptor groups into indolizines.

PMR spectra were obtained on Tesla BS-467 (60 MHz) and XL-100 instruments, in CCI 4 or CF3CO2H, internal standard TMS. UV spectra were recorded on a Specord M-40 in 96% ethanol, and IR spectra on a UR-20 in Vaseline oil. Mass spectra were obtained on an MX-1303 (E = 70 eV) with direct sample introduction. The purities of the compounds obtained were checked by chromatography on Silufol plates. The starting acetyl' and cyanopyridines (Ia-c, g) were obtained as described in [20-22]. l-Acetonyl-2-methyl-3-cyanopyridinium Bromide (iIa). A mixture of 0.8 g (6.8 mmole) of 2-methyl-3-cyanopyridine (Ia) and bromoacetone 1.7 g (l.lml, 13.0m mole) was boiled in i0 ml of ethyl methyl ketone for 10 h, until no more solid separated. The solution was cooled, and the rose-colored crystals which separated were filtered off, washed with ethyl methyl ketone and ether, and dried to give 0.7 g of the salt (IIa). Similarly obtained were the salts (IId-g) (Table I). The ~-acetylpyridinium salts were difficult to purify, and it was difficult to obtain a satisfactorily pure sample of the salt (IIc) (mp 145-147~ from methanol; 42%). It was not possible to isolate the bromo-acetonylate of (Ib). 2-Methyl-8-cyanoindolizine (IIIa). A mixture of 0.52 g (2.0 mmole) of the salt (IIa) and a saturated solution of sodium bicarbonate in 75 ml of 20% ethanol was heated for 30 min. After cooling, the mixture was extracted with ether, the extract evaporated, the residue chromatographed on a plate (silica gel L40/100, chloroform),the fractlonwith Rf 0.7 being collected. There was obtained 0.32 g (2.0 mmole, 100%) of low-melting yellow crystals of (IIIa). Found: C 77.0; H 5.7; H 17.9%. CIoHsN 2. Calculated: C 76.9; H 5.1; N 17~9%. The indolizines (IIIc-g) were obtained in a similar way (Table 2). In the preparation of (IIIb), the mixture obtained by reacting the pyridine (Ib) with bromoacetone was evaporated under reduced pressure and worked up as for (IIa). Compounds (IIId-f) crystallized from ethyl methyl ketone, and (IIIa-c, g) were purified by plate chromatography (silica gel L40/100, chloroform). Reaction of 2-Methyl-8-acetylindolizine with Alkali. A solution of 0.145 of the indolizine (IIIc) in 20 ml of 50% ethanol and 2 g (36 mmole) of KOH was h. After cooling, the mixture was neutralized with HCI to pH 7, and extracted form. The extract was evaporated and subjected to column chromatography (SiO 2, to give 0.069 g (48%) of the starting indolizine (IIIc), Rf 0.51. g (0.84 mmole) boiled for i0 with chlorochloroform)

On heating with alkali under the same conditions, the indolizines (IiId, e) were recovered in yields of 80 and 74% respectively. LITERATURE CITED I. 2. 3. 4. 5. 6. 7. 8. 9. i0. Ii. 12. S. I. Bobrovskii, E. V. Babaev, and Yu. G. Bundel', Khim. Geterotsikl. Soedin., No. 2, 203 (1987). k. E. Tschitschibabin, Ber. Chem. Ges., 60, 1607 (1927). G. Jones and J. Stanyer, J. Chem. Soc., C, No. 11, 901 (1969). T. S. Loseva, k. D. Yanina, E. E. Mikhlina, and L. N. Yakhontov, Khim. C~terotsikl. Soedin., No. 3, 348 (1976). I. Dainis, kust. J. Chem., 25, 1003 (1972). k. N. Kost, R. S. Sagitullin, and S. P. Gromov, Khim. Geterotsikl. Soedin., No. 7, 922

(1976).
A. N. Kost, R. S. Sagitullin, and S. P. Gromov, Dokl. Akad. Nauk C. Jutz, R. M. Wagner, and H.-G. Lobering, Angew. Chem. Int. Ed., L. Frohlich and F. Krohnke, Chem. Ber., 104, 1621 (1971). W. Kiel and F. Krohnke, Chem. Ber., 105, 3709 (1972). T. S. Loseva, M. S. Goizman, G. G. Dvoryantseva, E. E. Mikhlina, Khim. Geterotstikl. Soedin., No. 5, 627 (1979). V. A. Tafeenko, L. A. Aslanov, A. N. Kost, R. S. Sagitullin, and Akad. Nauk SSSR, 252, 126 (1980). SSSR, 230, 1106 (1976). 13, 737 (1974).

and L. N. Yakhontov, S. P. Gromov, Dokl.

178


13. 14. 15. 16. 17. 18. 19.

20. 21. 22. 23. 24. 25.

E. M. EVleth, Theor. Chim. Acta, 16, 22 (1970). V. Galasso, G. de Alti, and A. Bitotto, Theor. Chim. Acta, 2, 222 (1968). A. Gordon and R. Ford, Chemist's Compansion, Wiley (1973). S. F. Melson, in: Physical Methods in the Chemistry of Heterocyclic Compounds [in Russian], Khimiya, Moscow-Leningrad (1966), p. 360. P. J. Black, M. L. Hefferman, L. M. Jackm~n, Q. N. Porter, and G. R. Underwood, Aust. J. Chem., 17, 1128 (1964). M. Fraser, S. McKenzie, and D. H. Reid, J. CHem. Soc., B, No. i, 44 (1969). L. M. ALekseeva, G. G. Dvoryantseva, M. S. Goizman, T. 5. Loseva, A. D. Yanina, E. E. Mikhlina, Yu. N. Sheinker, and L. N. Yakhontov, Khim. Geterotsikl. Soedin., No. i, 154 (1976). P. Baumgarten and A. Dornow, Ber. Chem. Ges., 72, 563 (1939). P. Knudsen, Ber. Chem. Ges., 25, 2985 (1892). M. J. Relder and R. C. Elderfield, J. Org. Chem., !, 286 (1942). J. D. Bower, J. Chem. Soc., No. ii, 4510 (1957). T. S. Loseva, M. S. Goizman, O. R. Shvarts, L. M. Alekseeva, E. E. Mikhlina, and L. N. Yakhontov, Khim. Geterotsikl. Soedin., No. 6, 802 (1978). T. S. Loseva, A. D. Yanina, E. E. Mikhlina, and L. N. Yakhontov, Khim. Geterotsikl. Soedin., No. 2, 209 (1976).

PORPHYRINS. 21.* SYNTHESIS AND REACTIVITIES OF

13,17-DISUBSTITUTED DERIVATIVES OF 13,17-DESETHYLETIOPORPHYRIN III

G. B. Maravin, G. V. Ponomarev, and A. M. Shul'ga

UDC 547.979.733.04

A method was developed for the preparation, from hemin, of a porphyrin with hydroxypropyl substituents in the 13 and 17 positions of the macroring and a number of its derivatives.

In the present research we solved the problem of the preparation, from hemin (I), of a porphyrin (lla) with hydroxypropyl substituents in the 13 and 17 positions of the macroring and anumber of its derivatives. In the synthesis of the substituted porphyrins special attention was directed to obtaining substances that contain tertiary amino groups, since it is known [2-4] that porphyrins that contain amino groups display diverse biological activity. The preparation of porphyrin lla from mesoporphyrin IX (Ilia) by reduction of its dimethyl ester (lllb) with lithium aluminum hydride has been described in the literature [5]; the preparation of an aminoethylporphyrin from porphyrin IIIa by Curtius cleavage of the corresponding azide IIlc has also been described [6]. Porphyrin IIIa can be obtained in turn by treatment of hemin (I) with concentrated HI [7] or, more efficiently, by hydrogenation of it in formic acid on a palladium catalyst [8]. The process that we developed to obtain porphyrin IIa directly from hemin (I) [9] proceeds in up to 90% overall yield without isolation of intermediates; this made it possible to use this porphyrin for further chemical modification. *See [i] for Communication 20. Institute of Biophysics, Ministry of Public Health of the USSR, Moscow 123182. Translated from Khmiya Geterotsiklicheskikh Soedinenii, No. 2, pp. 214-220, February, 1987. Original article submitted August 21, 1985. 0009-3122/87/2302-0179512.50 ~ 1987 Plenum Publishing Corporation 179