Potassium Dichloroiodate(I) as a reagent for synthesis of iodinated oxaphospholenes

I. V. Alabugin, G. A. Sereda, V. K. Brel', N. V. Zyk, and N. S. Zefirov

Chemistry Department of the Moscow State University, Moscow, Russia

This research has been carried out at the Moscow State University, Russian Federation

Recently it was shown [1] that potassium dichloroiodate(I) (KIC1₂) is a very interesting iodating reagent, which allows, in particular, smooth and effective iodinating of olefins. This fact lets to regard this reagent as a donor of iodine(I). However the problem of its real nature and the mechanism of its action remains yet unsolved. Since as the synthetic processes performed with use of KICI₂, are very valuable, this problem requires further investigation.

For the more detailed study of properties of KICl₂ we presumed that it would be reasonable to use the specificy in structure and reactivity of phosphorylated allenes.

In addition, we tried to apply the synthetic performance of this reagent for development of methods of electrophilic iodocyclization of allenylphosphonates, which do not exist yet. The studied earlier reactions with I₂, IBr and ICI required heating, and the yields of the final products were no higher than 50% [5]. Therefore the search of new iodinating agents seemed to be an actual synthetic task.

We found that dimethyl 3-methyl-1,2-butadienephosphonate (I) reacts with KIC1₂ in chloroform at room temperature in l5-20 min. After washing the reaction mixture with sodium thiosulfate we obtained 4-iodo-1,2-oxaphosphol-3-ene (II) in 80-85% yield. Its formation corresponds to the ordinary scheme of reactions of electrophilic reagents with such substrate, assuming that initial electrophilic attack occurs with iodine chloride. This fact well agrees with the common view on KICI₂ as a source of ICl [1]. However, in the reaction with allene I is realised an earlier unknown type of reactivity of KICI₂. As a minor product (yield about 10%), in the reaction mixture we detected dimethyl 3-methyl-2-chloro-2-butenephosphonate (III).

All structures on this scheme have been optimized using molecular mechanics calculations.

To the oxaphospholene II in the ¹H NMR spectrum correspond dublets at 6.5 (²J_HP 26 Hz) and 3.7 ppm ( ³J_HP 12 Hz) and two singlets of nonequivalent methyl groups at 1.50 and 1.45 ppm. Formation of adduct III is indicated by the signals characteristic of methylene group, at 2.95 ppm (2H, J_HP 22 Hz), and of two different methyl groups at the double bond, at 1.85 ppm (J_HP 4.5 Hz) and 1.80 ppm (J_HP 7.0 Hz). The spectral characteristics completely coinside to the published ones [2].

The regioselectivity of chlorine addition shows that compound III is formed as a result of nucleophilic attack of chloride anion at the a,b-double bond followed by protonation of the intermediate. Since the iodine chloride displays the properties of a Lewis acid [3], it seems possible that coordination at phosphoryl oxygen of the iodine chloride, formed intermediately from the original reagent, promotes addition of chloride anion. The alternative route with one-electron oxidation of the substrate with iodine chloride also may not be excluded.

The possibility that the process proceeds by SET mechanism agrees with the recent report on probable reaction of ICl by this mechanism with aromatic substrates, also leading to chloro-substituted compounds [4].

In this step we did not attempted to elucidate unambiguously which mechanism acts predominantly in our case. Nethertheless, the fact that alternative reactivity of potassium dichloroiodate(I) actually exists illustrates spectacularly the possibilities to use the phosphorylated allenes as model substrates for the study of new reagents.

To obtain an additional information on both the reaction mechanism and its synthetic performance we studied reaction of KICl₂, with other allenephos phonates, differing by substituents at the C¹ atom of allene system and (or) at phosphorus atom. In this aspect, 1-alkoxy-2,3-pentadiene-2-phosphonates are particularly interesting due to their possible to allene- diene isomerization with formation of 3-halo-1,3- butadiene-2-phosphonates in the presence of hydrogen halides acids [5] or nucleophilic agents [6]. They are also capable to form cyclization products involv ing oxygen atom of alkoxymetylene fragment in the reactions with iodonium salts [7].

The reaction of diethyl 1-methoxy-2,3-penta-diene-2-phosphonates (IV) proceeds smoothly and leads to formation of a single product, 4-iodo-1,2-oxaphosphol-3-ene (V) only in almost quantitative yield.

It is noteworthy that neither allene-diene isomerization products nor the products of cyclization involving oxygen of methoxymethylene group are detected, unlike the reaction with iodonium salts [7], which obviously is caused by lower polarity of the reacting moiety.

lodinating of 1,2-alkadienephosphonic dichlorides has not yet been described earlier. Therefore it seems that study of reaction of one of them with KICl₂ is of particular interest.

Iodinating of 3-methyl-1,2-butadienephosphonic dichloride (Vl) with potassium dichloroiodate leads to 2-iodo-3-methyl- l,3-butadiehephosphonic dichloride (Vll) as the main product, what is typical for the re actions of allenephosphonic dichlorides with electro- philes [7].

Diene structure of compounds VII and VIll is confirmed by IR and NMR spectroscopy. In the ¹H NMR spectrum of ester VIII there is doublet of the proton attached to C¹ (d 6.4 ppm), and signals of terminal olefinic protons and the protons of methyl groups in an isopropenyl fragment. In the IR spectra occur absorption bands of two different double bonds (n 1590, 1640 cm^-1).

According to ¹H NMR spectrum, the reaction mixture also contains several unstable minor products formed probably as a result of addition at the 1-2 double bond of the allene system and by further transformation of the 1,3-alkadiene VII.

Thus, we developed a new approach to the synthesis of 4-iodo-1,2-oxaphosphol-3-enes and derivatives of 2-iodo-1,3-butadienephosphonic acids on the ground of reaction of KICl₂ with allenephosphonic acid derivatives.

The unique combination in the allenephosphonate molecule of two double bonds essentially different in reactivity allowed us to discover an unusual reaction pathway involving KICl₂.

EXPERIMENTAL

The ¹H NMR spectra were registered on a Varian, T-60 and Bruker WM-250 instruments with internal reference HMDS. Chromatography was carried out using Silufol UV-254 plates. The IR spectra were taken on a Specord 75IR spectrophotometer.

Reaction of potassium dichloroiodate(l) with dimethyl 3-methyl-1,2-butadienephosphonate (I). (a) A mixture of phosphonate (I) (0.176 g. I mmol) and KICl₂, (0.237 g, 1 mmol) in 3 ml of anhydrous methylene chloride was stirred for 15 min at room temperature. After washing the reaction mixture with a solution of potassium thiosulfate it was treated with chloroform, the organic layer was dried over sodium sulfate, and the solvent was removed in vacuo. A mixture of 0.22 g (81%) 4-iodo-5,5-dimethyl-2-oxo-1,2-oxaphosphol-3-ene (II) and 0.02 g (10%) dimethyl 3-metbyl-2-chloro-2-butene-1-phosphonate (III) formed. After crystallization the mixture from heptane O.18 g (67%.) of pure phospholene II, mp 101 ^oC (7).

(b) The reaction was performed like in (a), but with ethanol as a solvent. We obtained O.170 g (63%) of iodide II, identical with that obtained in (a). Phospholene II. IR spectrum (film, n, cm^-1): 1550 (C=C), l260 (P=O). ¹H NMR spectrum (s, ppm): 6.5 d (1H, HC=C, J_HP 26 Hz), 3.7 d (6H, OMe J_HP, 12 Hz), 1.5 s (3H, Me), 1.45 s (3H, Me). ³¹P NMR spectrum, s_P 34.7ppm.

Phosphonate III. IR spectrum: 1558 (C=C), 1260 (P=O). ¹H NMR spectrum (s, ppm) 3 7 d (6H, OMe, JHP 12 Hz), 2.95 d (2H, CH₂, J_HP 22 Hz), 1.85 d (3H, Me, J_{_HP} 4.5 Hz). 1.80 d (3H, Me, J_HP 7.0 Hz). ³¹P NMR spectrum. s_P 24.5 ppm

Reaction of potassium dichloroiodate(l) wilh diethyl 1-methoxy-2,3-pentadiene-2-phosphonale (IV). Phosphonate IV (0.234 g, 1 mmol) and KICI₂, (0.237 g, 1 mmol) were stirred for 15 min at room temperature, treated with aqueous solution of sodium thiosulfate (3 ml), then with chloroform. The organic layer was dried over sodium sulfate, and the solvent was removed m vacuo. 0.321 g (97%) of 5,5-dimethyl-4-iodo-3-methoxy and methyl-2-oxo-1,2-oxa-phosphol-1-ene (V) was obtained, mp 77劍 (sublimed at 110 ^oC/2 mm). ³¹P NMR spectrum (CHCI₃): d_P 30.10 ppm. IR spectrum (mineral oil, n, cm^-1): 1260 (P=O). ¹H NMR spectrum, (s, ppm): 4.2 d (2H, CH₂OMe, J_HP14 Hz), 4.2- 3.8 m (2H, OCH₂CH₃), 3.4 s (3H, OMe), 1.5 s (3H. Me), 1.45 s (3H, Me), 1.2 t (3H, CH₂CH₃, J 7.0 Hz). Found. % C 31.93; H 5.23 C₉H₁₆IO₄P. Calculated, % C 31.23, H 4.66.

Reaction of potassium dichloroiodate(l) with 3-methyl-1,2-buladienephosphonate dichloride (VI). To a solution of acyl dichloride Vl (2.55 g, 13.78 mmol) in 10 ml of anhydrous ethylene chloride while stirring at room temperature was added portionwise in 5 min 3.27 g (13.78 mmol) of KICI₂. Stirrng was continued for 30 min (strong evolution of hydrogen chloride was observed), and the eluent was removed in vacuo. 3-Methyl-2-iodo-1,3. buudiene-1-phosphonic dichloride (VII) [¹H NMR spectrum, d, ppm: 6.5 d (1H, HC=C), J_HP, 14.0 Hz), 5.3 br.s (1H HC=C), 5.15 br.s (1H, HC=C), 2.1 s (3H, Me); ³¹P NMR spectrum s_P 22.9 ppm] was treated without isolation with 2.2 ml (27.56 mmol) of dried pyridine and 3 ml of anliydrous methanol. After 15 min the mature was dissolved in 10 ml of chloroform, washed with aqueous solution of sodium thiosulfate (10 ml) an then with water (2 X 10 ml), dried over sodium sulfate. The solvent was removed, and after distilling the residue 2.1 g (50%) of dimethyl 2-iodo-3-methyl-1,3-butadiene-1-phosphonate (VII) was obtained, bp 133 ^oC (3 mm). IR spectrum (mineral oil, n, cm^-1): 1640, 1590 (C=C-C=C), 1260 (P=O). ¹H NMR spectrum, s, ppm: 6.4 d (¹H, HC=C, J_HP 14.0 Hz), 5.22 d.d. (1H, HC=C, J₁, J₂less than 1 Hz); 5.0 d.q (1H HC=C, J₁ 1.4 Hz, J₂ less than 1 Hz), 3.7 d (6H. OMe. J_HP 12.0 Hz), 2.0 m (3H, Me). ³¹P spectrum, s_P 14.0 ppm. Found, % C 27.87, H 4.59; P 10.31. C₇H₁₂IO₃P. Calculated, %. C27.84; H 4.00; P 10.26.

ACKNOWLEDGMENTS

This work was performed under financial support of Russian Foundation for Basic Research (grant 96-03-33250).

REFERENCES

1. Zefirov, N.S., Sereda, G.A., Sosonuk, S.S., Zyk, N.Y,, and Likhomanova, T.l., Synthesis, 1995, no. 11, pp. 1359-1361.
2. Angelov, Kh.M., Enchev, D.D., and Kirilov, M., Zh. Obshch. Khim., l 983, vol. 53, no. 9, pp. 1958-1960.
3. Andrews, L.J. and Keefer, R.M. J. Org. Chem., 1987, vol. 52, no. 13, pp. 2690-2694.
4. Turner, P.E., O'Malley, R.F., Sardella, D.J., Barinelli, L.S., and Kaul, P.J., J. Org. Chem., 1994, vol. 59, no. 24, pp. 7335-7340.
5. Brel', V.K., Chunin, E.D., Dogadina, A.D., Skvortsov, N.K., Ignat'ev, V.M., and lonin, B.l. Zh. Ohshch. Khim., 1980, vol. 50, no. 4, pp. 770-773.
6. Brel', V.K. and Ahramkin, E.V., Zh. Ohshch. Khim., 1991, vol. 6l, no. 4, pp. 770-773.
7. Zefirov, N.S., Koz'min, A.S., Kasumov, T., Polekhin, K.A.. Sorokin, V.D., Brel, V.K., Ahramkin. E.V., Struchkov, Yu.T., Zhdankin, V.V,, and Stang, P.J., J. Org. Chem., 1992, vol. 57, no. 8, pp. 2433-2437.