摘要:

Mendeleev Communications Electronic Version, Issue 1, 2000 (pp. 1–42) First in situ 1H NMR spectroscopic monitoring of manganese species in the MnIII(salen) + PhIO catalytic system Konstantin P. Bryliakov,a Dmitrii E. Babushkinb and Evgenii P. Talsi*b a Natural Science Department, Novosibirsk State University, 630090 Novosibirsk, Russian Federation b G. K. Boreskov Institute of Catalysis, Siberian Branch of the Russian Academy of Sciences, 630090 Novosibirsk, Russian Federation.Fax: +7 3832 34 3766; e-mail: talsi@catalysis.nsk.su DOI: 10.1070/MC2000v010n01ABEH001183 High-valence manganese species formed in the MnIII(salen) + PhIO catalytic system were characterised using 1H NMR spectroscopy. The Jacobsen catalyst (R,R)-(–)-N,N'-bis(3,5-di-tert-butylsalicylidene)- 1,2-cyclohexane diaminomanganese(III) chloride, [MnIII- (salen)] 1 and related manganese complexes are practically attractive catalysts for enantioselective epoxidation of unfunctionalised alkenes.1–5 The observed enantioselectivities were explained by models based on a reactive oxomanganese(V) species.However, in none of these cases the reactive species was isolated or characterised.Only recently, direct evidence for its existence was obtained by electrospray tandem mass spectrometry.6 Nevertheless, it is still unclear whether the detected [(salen)MnV=O]+ intermediate really exist in a detectable amount in the catalytic system (1 + PhIO) or it is formed in the course of a MS experiment via fragmentation of the m-oxo dimeric complex [(salen)- MnIV–O–MnIV(salen)]2+.Thus, the in situ spectroscopic characterization of the [(salen)MnV=O]+ species remains an intriguing problem. Jin and Groves7 observed an unstable diamagnetic (porphyrin)MnV=O species by 1H NMR spectroscopy.7 Here we describe the first 1H NMR spectroscopic monitoring of manganese species formed in the MnIII(salen) + PhIO catalytic system. Based on the 1H NMR spectrum and the reactivity pattern some of these species can be identified as the oxomanganese(V) intermediate [(salen)MnV=O]+.To assign 1H NMR resonances of complex 1, the signals were compared with those of N,N'-bis(salicylidene)ethylenediaminomanganese( III) chloride 2 and N,N'-bis(3,4,5,6-tetradeuterosalicylidene)- 1,2-cyclohexanediaminomanganese(III) chloride 3.† The 1H NMR spectrum of complex 2 in [2H6]DMSO at 20 °C is shown in Figure 1(a).The resonances at –22.2 ppm (Dw1/2 = = 450 Hz) and –26.0 ppm (Dw1/2 = 500 Hz) were previously unambiguously assigned to protons at the 5- and 4-positions of aromatic rings of complex 2, respectively.10 We have additionally observed the resonance at –125 ppm (Dw1/2 = 4 kHz) assigned to two protons of the ethylene bridge of complex 2 and a very † Complex 1 was purchased from Aldrich.Complexes 2 and 3 were prepared as described in ref. 8. Deuterated salicylic aldehyde for the synthesis of 3 was prepared according to ref. 9. The 1H NMR spectra were recorded on a Bruker MSL 400 MHz spectrometer. The EPR spectra were recorded in 5 mm quartz tubes on a Bruker ER 200 D spectrometer. broad resonance at –405 ppm (Dw1/2 = 10 kHz) assigned to the imine protons.The latter signal can be detected in complexes 1, 2 and 3 at approximately the same field position. The resonances of protons at the 3- and 6-positions of aromatic rings of complex 2 are masked by those of residual undeuterated water and DMSO [Figure 1(a)]. The 2D NMR spectrum of complex 3 [Figure 1(b)] shows that the deuterons at the 3- and 6-positions (and thus protons) of complex 3 display resonances at –1.9 and 2.0 ppm, respectively.Thus, the protons at the 3- and 6-positions of complex 2 and protons at the 6-position of complex 1 would exhibit signals in the same region. Figure 1(c) demonstrates the 1H NMR spectrum of complex 1 in [2H6]DMSO at 20 °C. A comparison of the 1H NMR spectra of 1, 3 and 2 [Figure 1(a), (c) and (d)] allowed us to assign the signals denoted in Figure 1(c) by B to the diaminocyclohexane bridge of 1.We cannot assign these signals to particular protons of the bridge. Their total intensity corresponds to four protons. The signals of the remaining six protons of the bridge may be too broad or can be masked by the intense signals of residual H2O and DMSO.The resonance at –27 ppm (Dw1/2 = 700 Hz) belongs to protons at the 4-position of aromatic rings in 1. The 1H and 2D NMR spectra of 2, 3 and 1 [Figure 1(a)–(d)] were recorded in [2H6]DMSO and DMSO, respectively. These solvents are unsuitable for the epoxidation of alkenes by the 1 + PhIO catalytic system; thus, the reaction of 1 with PhIO O H N Mn O H N H H Cl O H N Mn O H N Cl 1 2 O H N Mn O H N H H Cl 3 D D D D D D D D 3 4 5 6 4 6 30 0 –20 d/ppm (a) (b) (c) (d) (e) H2 O +DMSO 5-H 4-H 6-D, 3-D 5-D 4-D B 4-H CHCl3 ×4 But 4-H 20 10 –30 –10 B B B ×4 B B B Figure 1 1H NMR spectra ([2H6 ]DMSO, 20 °C) of (a) 2, (c) 1 and (d) 3; (b) 2D NMR spectrum (DMSO, 20 °C) of 3; (e) 1H NMR spectrum (CDCl3, –20 °C) of 1 (0.025M solutions).Mendeleev Communications Electronic Version, Issue 1, 2000 (pp. 1–42) was studied in CDCl3 as a solvent. The 1H NMR spectrum of 1 in CDCl3 at –20 °C [Figure 1(e)] displays an intense resonance of four But groups at 2.6 ppm (Dw1/2 = 200 Hz), two resonances at –29.0 ppm (Dw1/2 = 800 Hz) and –32.7 ppm (Dw1/2 = 800 Hz) from nonequivalent protons at the 4-positions of 1 and signals, denoted by B, from the protons of the diaminocyclohexane bridge in 1.The signals of protons at the 6-positions of 1 are masked by that of the But groups. The nonequivalence of two aromatic rings in 1 is consistent with the X-ray diffraction data.11 The paramagnetic line broadening prevents the detection of this nonequivalence by 1H NMR. Thus, four nonequivalent But groups of 1 display a peak at 2.6 ppm. The nonequivalence of protons at the 4-positions of 1 is revealed only at low temperature.We cannot say what particular effect masks this nonequivalence at room temperature. Stable oxomanganese(V) complexes are few in number; tetradentate ligands are used to stabilise the high-valence manganese centre.12,13 By analogy to isoelectronic nitridomanganese(V) salen complexes14 and known stable oxo MnV complexes, the reactive [(salen)MnV=O]+ species is expected to be a low-spin d2 complex. The 1H NMR spectrum of (salen)MnVºN prepared according to the known procedure14 is as follows (CDCl3, 400 MHz, –20 °C) d: 1.28 (s, 18H, CMe3), 1.45 (s, 9H, CMe3), 1.49 (s, 9H, CMe3), 2.03–3.46 (m, 10H, cyclohexane H), 6.97 (s, 1H, aromatic H), 7.02 (s, 1H, aromatic H), 7.44 (s, 1H, aromatic H), 7.46 (s, 1H, aromatic H), 7.95 (s, 1H, CH=N), 8.00 (s, 1H, CH=N).Note that But groups of (salen)MnVºN display three peaks at 1.28, 1.45 and 1.49 ppm in the 1H NMR spectrum, while those of uncoordinated H2salen exhibit two peaks at 1.24 and 1.42 ppm. In order to detect the oxomanganese(V) species in the catalytic system, a standard NMR tube (d = 5 mm) containing 0.6 ml of a cooled (–40 °C) 0.02 M solution of 1 in CDCl3 was placed in an NMR spectrometer immediately after shaking the solution with PhIO powder (2 mg) at –40 °C for 30 s.Only a small portion of PhIO was dissolved as a result of this procedure. The 1H NMR spectrum was recorded at –20 °C 5 min after the onset of the reaction. Several new signals are observed in the region 1.3–1.8 ppm [cf. Figure 2(a) and (b)].They can be assigned to But groups of three manganese complexes 4–6. Note that H2salen is not liberated at the initial stage of the reaction. Complex 4 is very unstable. Its concentration diminished with a characteristic time of about 20 min at –20 °C and shorter than 3 min at 0 °C, while concentrations of complexes 5 and 6 increased. The attempt to increase the concentration of complex 4 by additional shaking of the sample [Figure 2(b)] with an initially added portion of PhIO at –40 °C gave rise to a predominant increase in the concentration of complexes 5 and 6 [Figure 2(c) and (d)].The achieved concentration of complex 4 was no higher than 3% of the initial concentration of 1 and those of complexes 5 and 6 can be higher than 50% of the initial concentration of 1 [Figure 2(e) and (f)].The increase in the concentrations of complexes 5 and 6 was accompanied by dissolution of PhIO. Complexes 5 and 6 are stable at –20 °C and very slowly react with styrene at this temperature (the characteristic time was longer than 2 h, [styrene] = 0.1 mol dm–3). In contrast, the addition of styrene (to a concentration of 0.1 mol dm–3) to the sample presented in Figure 2(d) at –20 °C leads to an immediate drop (by a factor of about two) of the concentration of complex 4 [Figure 2(e)].This drop was accompanied by the appearance of styrene oxide resonances in the 1H NMR spectrum. In the absence of styrene, the concentrations of complexes 4–6 remained almost unchanged in 5 min at –20 °C. These data indicate that complex 4 can be reactive towards styrene.When styrene was added to the sample containing 1 prior to the shaking with PhIO at –20 °C, the immediate growth of the styrene oxide concentration was observed by 1H NMR, while formation of complexes 4–6 was almost entirely suppressed. Complex 4 displays three resonances of But groups at 1.68, 1,64 and 1.42 ppm. These peaks were assigned to one complex because of a strictly parallel change in their intensities.The overall intensity of the signals at 1.68 and 1.64 ppm equals to that of the signal at 1.42 ppm. The observed pattern for But groups of complex 4 resembles that for the nitridomanganese complex (salen)MnVºN (at 1.49, 1.45 and 1.28 ppm), when one compares differences in the chemical shifts between the signals and their relative intensities.Unfortunately, we have not detected signals of the aromatic and imine protons of complex 4 using CDCl3 and CD2Cl2 as solvents. Most probably, they are obscured by the intense resonances of PhI formed in the reaction of 1 with PhIO. It is important that the widths of the resonances of But groups in complex 4 (20 Hz) are close to those of the signals of diamagnetic species (e.g., CHCl3 or PhI) in our particular sample (the line broadening is caused by the presence of paramagnetic MnIII species). This fact evidences in favour of complex 4 to be also diamagnetic.Corresponding signals of complexes 5 and 6 are broader than those of complex 4 (Figure 2) and can belong to paramagnetic species. Complex 4 displays a characteristic pattern of But groups closely resembling that for diamagnetic nitridomanganese(V) salen species.It is very unstable and predominates only at the early stage of the reaction of 1 with PhIO at low temperature. The effect of styrene on the concentration of 4 evidences in favour of its reactivity towards this substrate. Based on these data, complex 4 can be identified as the oxomanganese intermediate [(salen)MnV=O]+.Let us discuss the structure of complexes 5 and 6. The concentration of complex 5 grows after warming the sample [Figure 2(d)] for 2 min at room temperature [Figure 2(f )]. The sample displays two resonances of But groups at 1.72 and 1.60 ppm, two resonances at 10.9 and 11.3 ppm, several signals in the range 4–5 ppm (not shown) and two signals at –4.1 and –4.2 ppm.The field positions and widths (30–80 Hz) of the ob- 3.0 2.4 1.8 1.5 1.2 11.0 9.0 –2.0 –4.0 (a) (b) (c) (d) (e) (f) (g) d/ppm 1 4 5 6 * 11.0 9.0 –2.0 –4.0 1 4 4 4 4 5 5 5 5 6 6 6 Figure 2 1H NMR spectra of a 0.02 M solution of 1 in CDCl3 (0.6 ml) before and after shaking with a suspension of PhIO (2 mg) at –40 °C: (a) before shaking; (b) shaking for 30 s; (c) shaking for 1.5 min; (d) shaking for 2.5 min; (e) 1 min after the addition of styrene to a concentration of 0.1 mol dm–3 to the sample shown in Figure 1(d); (f) sample (d) after 2 min warming at room temperature; (g) sample (f) after shaking with an additional portion of PhIO (4 mg) at 0 °C.The 1H NMR spectra were recorded at (a)–(f) –20 °C and (g) 0 °C. ×2 ×2 4Mendeleev Communications Electronic Version, Issue 1, 2000 (pp. 1–42) served resonances of complex 5 are typical of antiferromagnetically coupled m-oxo dinuclear manganese(IV) species.15,16 For these species, the 1H NMR signals are placed much closer to the positions in diamagnetic complexes than for corresponding mononuclear MnIV and MnIII complexes. Thus, the resonances at 10.9 and 11.3 ppm probably belong to aromatic protons of complex 5.The concentration of complex 5 decreased and that of complex 6 increased as the sample [Figure 2(f)] was treated with an additional portion of PhIO (4 mg) at 0 °C [Figure 2(g)]. The field positions and widths of the signals of complex 6 are also typical of MnIV–O–MnIV dimers. Probably, complexes 5 and 6 are binuclear complexes [L(salen)MnIV–O–MnIV(salen)L' ]2+ with different axial ligands L, L' (iodosylbenzene and chloride anion).The dimeric cation [(salen)MnIV–O–MnIV(salen)]2+ with PhIO molecules at axial sites was detected in the 1 + PhIO catalytic system by electrospray tandem mass spectrometry.6 Thus, at least three types of manganese species (4–6) are formed upon the interaction of complex 1 with PhIO at low temperature.Complexes 5–6 are relatively inert dimers [L(salen)- MnIV–O–MnIV(salen)L' ]2+ with different axial ligands. Complex 4 can be identified as the oxomanganese(V) intermediate [(salen)- MnV=O]+ based on its 1H NMR spectrum and the reactivity pattern. This work was supported by the Russian Foundation for Basic Research (grant no. 97-03-32495a). References 1 E. N.Jacobsen, W. Zhang, A. R. Muci, J. R. Ecker and L. Deng, J. Am. Chem. Soc., 1991, 113, 7063. 2 T. Katsuki, Coord. Chem. Rev., 1995, 140, 189. 3 T. Palucki, N. S. Finney, P. J. Pospisil, M. L. Guler, T. Ishida and E. N. Jacobsen, J. Am. Chem. Soc., 1998, 120, 948. 4 C. T. Dalton, K. M. Ryan, V. M. Wall, C. Bousquet and D. G. Gilheany, Topics in Catalysis, 1998, 5, 75. 5 W. Adam, R. T. Fell, V. R. Stegmann and Ch. R. Saha-Moller, J. Am. Chem. Soc., 1998, 120, 708. 6 D. Feichtinger and D. A. Plattner, Angew. Chem., Int. Ed. Engl., 1997, 36, 1718. 7 N. Jin and J. T. Groves, J. Am. Chem. Soc., 1999, 121, 2923. 8 K. Srinivasan, P. Michaud and J. K. Kochi, J. Am. Chem. Soc., 1986, 108, 2309. 9 D. S. Kemp, J. Org. Chem., 1971, 36, 202. 10 J. A. Bonadies, M. J. Maroney and V. L. Pecoraro, Inorg. Chem., 1989, 28, 2044. 11 P. J. Pospisil, D. H. Carsten and E. N. Jacobsen, Chem. Eur. J., 1996, 2, 974. 12 T. J. Collins and S. W. Gordon-Wylie, J. Am. Chem. Soc., 1989, 111, 4511. 13 F. M. MacDonnell, P. Fackler, C. Stern and T. V. O’Halloran, J. Am. Chem. Soc., 1994, 116, 7431. 14 J. Du Bois, J. Hong, E. M. Carreira and M. W. Day, J. Am. Chem. Soc., 1996, 118, 915. 15 J. A. Smegal, B. C. Schardt and C. L. Hill, J. Am. Chem. Soc., 1983, 105, 3510. 16 J. T. Groves and M. K. Stern, J. Am. Chem. Soc., 1988, 110, 8628. Received: 5th July 1999; Com. 99/1511

ISSN:0959-9436     

出版商:RSC    

年代:2000

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 1, 2000 (pp. 1–42) Regioselective monophosphorylation of glycols containing primary and secondary hydroxyl groups Edward E. Nifantiev,*a Mikhail K. Gratcheva and Stephen F. Martinb a Department of Chemistry, Moscow State Pedagogical University, 119882 Moscow, Russian Federation. Fax: +7 095 248 0162; e-mail: chemdept@mtu-net.ru b Department of Chemistry and Biochemistry, University of Texas at Austin, Austin, Texas 78712-1167, USA.Fax: +1 512 471 4180; e-mail: sfmartin@mail.utexas.edu DOI: 10.1070/MC2000v010n01ABEH001212 A new approach to the synthesis of monophosphorylated glycols with the use of PIII-benzimidazolide has been suggested. Regioselective phosphorylation of polyhydroxyl systems is an urgent problem of fine organic synthesis.Definite progress towards this direction was achieved using the preferable interaction of primary hydroxyl groups of polyols with sterically hindered amines, which are activators of the nucleophilicity of hydroxyl groups.1 For solving this problem, we suggest another way based upon the use of commonly available and highly reactive amides of trivalent phosphorus acids (ATPA) as phosphorylating agents.2 Note that in contrast to chlorophosphites and other widespread phosphorylating agents they can be dramatically different from each other in the spatial organization of a leaving (amine) group.This difference can be important in selecting reagents for the regioselective phosphorylation of nucleophiles having several reactive centres.Glycols with primary and secondary hydroxyl groups can serve as an example. ATPA with small amido groups do not exhibit selectivity with respect to such glycols. However, ATPA with bulky amido groups, which are responsible for steric hindrances during an attack on secondary hydroxyl groups, can provide high selectivity for the phosphorylation of only primary hydroxyl groups. In this connection, we decided on phospho(III)benzimidazolides3 in order to examine the regioselective monophosphorylation of glycols with primary and secondary hydroxyl groups. 1,3-Butylene glycol 1 was taken as a model substrate for the phosphorylation; diethylamide 2 and benzimidazolide 3 of neopentylenephosphorous acid served as phosphorylating agents. We found that the interaction of equimolecular quantities of 1 and 2 resulted in a mixture of two regioisomers 4 and 5, which exhibit dramatically different 31P NMR spectra (122 and 128 ppm, respectively, the integral ratio 1:1).† At the same time, according to 31P NMR data, the interaction of equimolecular quantities of 1 and 3 leads to only one product 4 (Scheme 1).In both cases, the complete disappearance of signals of phosphorylating reagents 2 and 3 (147 and 114 ppm, respectively) in the 31P NMR spectra of reaction mixtures was observed.Monophosphorylation product 4 was treated with sulfur to form thiophosphate 6, which was isolated in the individual form.‡ Note that diphosphite 7 was formed by the phosphorylation of glycol 1 by two moles of ATPA. The 31P NMR spectrum of the reaction mixture exhibited a signal at 128 ppm, which was previously observed in the phosphorylation of glycol 1 by amide † Dioxane, c 1 mol dm–3, 100 °C, 3 h.‡ An equimolar quantity of sulfur was added to the reaction mixture, and the contents were stirred at 20 °C for 12 h. Thiophosphate 6 was isolated by flash chromatography on silica gel with the eluent CHCl3– MeOH (7:1), Rf 0.5. Yield 70%, pale yellow viscous oil. 1H NMR (250 MHz, CDCl3, TMS as an internal standard) d: 0.88 (s, 3H, Mee cyclo), 1.13 (s, 3H, Mea cyclo), 1.17 (d, 3H, MeCH, 3JHCCH 6.4 Hz), 1.62–1.88 (m, 2H, CCH2C), 2.34 (br. s, 1H, OH), 3.80–4.03 (m, 6H, CH2Oacyclo, 3JHCOP 9.0 Hz; CH2Ocyclo, 3JHCOP 8.5 Hz), 4.08–4.35 (m, 1H, CH). 13C NMR (50.32 MHz, CDCl3, TMS as an internal standard) d: 20.97 (MeCH), 21.5 (Mee cyclo), 23.44 (Mea cyclo), 32.17 (MeCMe, 3JCCOP 6.6 Hz), 39.00 (CH2CH2CH, 3JCCOP 6.9 Hz), 63.97 (CH), 65.35 (CH2CCH2, 2JCOP 5.5 Hz), 77.61 (CH2CH2O, 2JCOP 5.6 Hz). 31P NMR (32.4 MHz, CDCl3, 85% H3PO4 as an external standard) d: 64. The elemental analysis data are consistent with the theoretical values. 2. Thus, we found that reagent 3 is capable of the regioselective phosphorylation of a primary hydroxyl group in the presence of secondary hydroxyl groups.An additional confirmation for the location of a phosphoruscontaining residue in compound 6 was obtained by an analysis of 1H and 13C NMR spectra measured under conditions of homoand heteronuclear double NMR. Thus, for example, spin–spin coupling constants (SSCC) of oxymethylene protons of 1,3-butylene glycol residue with phosphorus were present in the 1HNMR spectra, and SSCC of methine proton with phosphorus were absent.Analogously, SSCC of oxymethylene carbon with phosphorus were present in 13C NMR spectra and SSCC of methine carbon with phosphorus were absent.‡ We used the found regioselectivity for the directed phosphorylation of natural ceramide 8,§ which is of importance for lipidology.We found that ceramide 8 was regioselectively mono- § We used commercial ceramide 8 (Sigma), which had the following 1H NMR spectrum (250 MHz, CDCl3, TMS as an internal standard) d: 0.81 (t, 6H, Me1, 3JHCCH 7.2 Hz), 1.09–1.27 (m, 58H, CH2 2 ), 1.42–1.60 (m, 4H, CH2 3 ), 1.91–2.05 (m, 2H, CH2 4), 2.16 (t, 2H, CH2 8, 3JHCCH 7.1 Hz), 2.56 (br. s, 1H, CH2OH), 3.58–3.68 (m, 2H, CH2OH), 3.76–3.90 (m, 3H, CH2 5, NCH), 4.25 (br.s, 1H, HCOH), 5.25–5.30 (m, 1H, CH6), 5.39–5.51 (m, 1H, HCOH), 5.64–5.78 (m, 1H, CH7), 6.17 (d, 1H, NH, 3JHNCH 6.8 Hz). OH OH O O Et2NP 1:1 O O OP OH O O OP OH N N P O O 1 + 4 1:1 S O O OP OH S 1 + 2 1 2 4 5 O O P P O O O O 1:2 3 6 7 Scheme 1Mendeleev Communications Electronic Version, Issue 1, 2000 (pp. 1–42) phosphorylated by an equimolar quantity of phosphoamide 3 by analogy with the above reaction (Scheme 2).Phosphite 9 was treated in the reaction mixture by sulfur with the formation of thiophosphate 10, which was isolated in the individual form.¶ Its identity and structure were determined with using the above methodology. The synthesis of cyclothiophosphate 10 is not only principally important, but also synthetically useful since it opens the way for preparing various analogues of phosphosphingolipids (sphingomyelins).4 Thus, the traditional use of ATPA can find new applications, for example, to the regioselective phosphorylation of complex natural compounds containing different proton-donor groups. This work was supported by the International Science Foundation (ISF) and the U.S.Civilian Research and Development Foundation (CRDF) (grant no. RC1-177). References 1 K. Ishihara, H. Kurihara and H. Yamamoto, J. Org. Chem., 1993, 58, 3791. 2 E. E. Nifantiev and M. K. Gratchev, Usp. Khim., 1994, 63, 602 (Russ. Chem. Rev., 1994, 63, 575). 3 V. Yu. Iorish, M. K. Gratchev, A. R. Bekker and E. E. Nifantiev, Zh. Obshch. Khim., 1993, 63, 783 (Russ.J. Gen. Chem., 1993, 63, 551). 4 D. A. Predvoditelev and E. E. Nifantiev, Zh. Org. Khim., 1995, 31, 1761 (Russ. J. Org. Chem., 1995, 31, 1559). ¶ Phosphorylation conditions: dioxane–CH2Cl2 (4:1), c 0.1 mol dm–3, 45 °C, 12 h. 31P NMR of compound 9 (in the reaction mixture) d: 122. An equimolar quantity of sulfur was added to the reaction mixture at 45 °C for 12 h. Cyclothiophosphate 10 was isolated and purified by double flash chromatography on silica gel with the eluent CHCl3–MeOH (7:1), Rf 0.5; next, with the eluent C6H6–dioxane (3:1), Rf 0.6.Yield of compound 10 50%, pale yellow amorphous solid, mp 95–96 °C. 1H NMR (250 MHz, CDCl3, TMS as an internal standard) d: 0.84–0.92 (m, 9H, Me1, Mee cyclo), 1.16–1.36 (m, 61H, CH2 2, Mea cyclo), 1.56–1.70 (m, 4H, CH2 3), 1.95–2.09 (m, 2H, CH2 4), 2.22 (t, 2H, CH2 8, 3JHCCH 7.2 Hz), 3.67– 3.73 (m, 2H, CH2OP), 3.74–4.10 (m, 7H, CH2 5, NCH, CH2Ocyclo), 4.31 (br. s, 1H, HCOH), 5.33–5.37 (m, 1H, CH6), 5.46–5.58 (m, 1H, HCOH), 5.72–5.86 (m, 1H, CH7), 6.23 (d, 1H, NH, 3JHNCH 6.8 Hz). 31P NMR (32.4 MHz, CDCl3, 85% H3PO4 as an external standard) d: 63. The elemental analysis data are consistent with the theoretical values. CHOH CH CH CH2OH Me(CH2)9CH2CH2CH2C H NH Me(CH2)20CH2CH2C O 1 2 3 4 5 6 7 8 1 2 3 8 + 3 9 S CHOH CH CH H2C Me(CH2)9CH2CH2CH2C H NH Me(CH2)20CH2CH2C O 1 2 3 4 5 6 7 8 1 2 3 10 OP O O S Scheme 2 CHOH CH CH H2C Me(CH2)9CH2CH2CH2C H NH Me(CH2)20CH2CH2C O OP O O Received: 13th October 1999; Com. 99/1540

ISSN:0959-9436     

出版商:RSC    

年代:2000

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 1, 2000 (pp. 1–42) Formation of stable 1,2,3-benzodithiazolyl radicals by thermolysis of 1,3,2,4-benzodithiadiazines Victor A. Bagryansky,*a Ivan V. Vlasyuk,b Yuri V. Gatilov,c Alexander Yu. Makarov,c Yuri N. Molin,a Vladimir V. Shcherbukhind and Andrey V. Zibarev*c a Institute of Chemical Kinetics and Combustion, Siberian Branch of the Russian Academy of Sciences, 630090 Novosibirsk, Russian Federation.Fax: +7 3832 34 2350; e-mail: vbag@kinetics.nsc.ru b Department of Natural Sciences, Novosibirsk State University, 630090 Novosibirsk, Russian Federation c N. N. Vorozhtsov Novosibirsk Institute of Organic Chemistry, Siberian Branch of the Russian Academy of Sciences, 630090 Novosibirsk, Russian Federation. Fax: +7 3832 34 4752; e-mail: zibarev@nioch.nsc.ru d N.D. Zelinsky Institute of Organic Chemistry, Russian Academy of Sciences, 117913 Moscow, Russian Federation. Fax: +7 095 135 5328 DOI: 10.1070/MC2000v010n01ABEH001198 Mild thermolysis (at 110–150 °C) of 1,3,2,4-benzodithiadiazine 1 and its derivatives 2–6 in hydrocarbon solvents quantitatively yielded stable 1,2,3-benzodithiazolyl p-radicals via a first-order reaction (Ea = 76.2 kJ mol–1, k0 = 4.34×105 s–1 for 1).The heteroatom reactivity of 1,3,2,4-benzodithiadiazine1 1 and its derivatives2,3 (Scheme 1) which exhibit formal features of antiaromaticity4 (such as a planar or nearly planar geometry, a united molecular 12p-electron system, and low-energy excited states)1–3,5,6 is poorly known.1,3,7 Reasonably, the first step in studying the chemistry of these compounds is the investigation of their thermal stability and the identification of decomposition products.We found that mild (~110–150 °C) thermolysis of 1–6 in hydrocarbon solvents (squalane, trans-decalin, cyclohexane or hexane) resulted with nearly quantitative yields in stable radicals 1·–6·, which were identified by EPR spectroscopy (Figures 1 and 2, Table 1).† The EPR spectra of these radicals generated from 1–3 corresponded to those published earlier for 1,2,3- benzodithiazolyl p-radicals 1·–3· prepared by other methods,8–11 mainly by reduction of corresponding Herz salts (1,2,3-arenodithiazolium chlorides12).‡ Radicals 1·–3· were initially assigned the 1,2-benzothiazetyl structure8,9 (Scheme 1), which was further corrected to 1,2,3-benzodithiazolyl on the basis of EPR experiments with 33S-enriched species. These experiments indicated the presence of two nonequivalent sulfur atoms.11 The 1,2,3-benzodithiazolyl structure is also consistent with the fact that the radicals can be oxidised into Herz salts by molecular chlorine.10 Radicals 4·–6· have been synthesised for the first time; thus, this approach is superior to methods reported previously.8–11 In particular, the corresponding Herz salts12 were not yet described.The assignment of HFI constants in 1·–6· (Table 1) was based on earlier data,8–11 substitutional effects and the results of the ab initio B3LYP/CC-pVDZ calculations of spin density distribution. The calculated constants were consistent with the experimental data (Table 1). 10 G 1· 2· 3· Experiment Simulation Figure 1 Experimental (in squalane) and simulated EPR spectra of 1·–3·. Differences in the HF line widths were ignored in simulating the spectra. R4 R1 R3 R2 S N S N R4 R1 R3 R2 S S N R4 R1 R3 R2 S N or D 1–6 1·–6· 1, 1· R1 = R2 = R3 = R4 = H 2, 2· R1 = R2 = R4 = H, R3 = Me 3, 3· R1 = R2 = R4 = H, R3 = OMe 4, 4· R1 = R2 = R3 = R4 = F 5, 5· R1 = R2 = R4 = H, R3 = F 6, 6· R1 = R3 = R4 = H, R2 = FMendeleev Communications Electronic Version, Issue 1, 2000 (pp. 1–42) The kinetic build-up curve of radical 1· in the thermolysis of compound 1 in squalane (Figure 3) was adequately approximated by the first-order equation A = B(1 – e–kt) (where B and k are † Compounds 1–6 were synthesised and purified as described earlier.1–3 The EPR spectra were recorded on a Bruker EMX spectrometer (MW power, 0.64 mW; modulation frequency, 100 KHz; modulation amplitude, 0.1 G).The spectra simulation was performed with the Simfonia-Bruker program. The spin density distribution and HFI constants were calculated at the B3LYP/CC-pVDZ level of theory using the Gaussian 94™ program.13 In a typical experiment, 10–3 M solutions of 1–6 in a hydrocarbon (squalane, trans-decalin, cyclohexane or hexane), outgassed by three freeze–pump–thaw cycles, was gradually heated in an EPR valve-equipped quartz capillary up to 150 °C (in squalane, the detectable amounts of radicals appeared at 110 °C, whereas in cyclohexane even at 90 °C).After holding for 1 h at this temperature, the sample was cooled to 20 °C, and the EPR spectrum was measured. The g-factors of 1·–6· were measured using a DPPH standard.Nearly quantitative conversion of 1–6 into 1·–6· was determined by a CuCl2·2H2O standard with an accuracy of ±15%. At 20 °C, the concentrations of 1·–6· in air-protected solutions decreased only by 30% for 3 weeks. In the presence of oxygen, the radicals are less stable, especially under heating.Thus, in an air-saturated solution at 150 °C, the EPR signal completely decayed in 5 min. ‡ 1,2,3-Benzodithiazolium chloride12 is readily reduced to 1· with Ph3Sb in toluene at 20 °C. The EPR spectrum is identical to that of the radical arising from the thermolysis of 1. the optimised parameters: B is the concentration of radicals at t = •, k is the reaction rate constant).The results are summarized in Table 2. The activation energy Ea = 76.2 kJ mol–1 and the pre-exponential factor k0 = 4.34×105 s–1 were calculated from the equation ln k = ln k0 – Ea/RT (Figure 4). Thus, a novel promising approach to the synthesis of thermally stable 1,2,3-benzodithiazolyl radicals was developed. This method provides the basis for further in-depth studies of these interesting species including their individual isolation and structural characterization.a3H (Me). b3H (OMe). cInterchangeable values. Table 1 HFI constants (in parentheses, theoretical values), G, and g-factors of radicals 1·–6·. N R1 R2 R3 R4 g 1· 8.22 (8.1) 2.93 (–2.5) 0.97 (1.2) 3.73 (–2.9) 0.81 (1.2) 2.0080 2· 8.38 (8.2) 2.94 (–2.4) 1.03 (1.1) 3.97a (3.2) 0.75 (1.3) 2.0076 3· 8.67 (8.4) 2.93 (–2.3) 1.08 (0.9) 0.44b (0.7) 0.47 (1.2) 2.0076 4· 8.15 (8.1) 5.67 (7.3) 3.52 (–3.6) 9.96 (10.2) 2.56 (–4.1) 2.0078 5· 8.55 (8.3) 3.15 (–2.6) 1.06 (1.0) 8.55 (9.6) 0.71 (1.3) 2.0079 6· 7.88 (7.7) 2.44c (–2.3) 2.65c (–3.3) 3.87 (–3.0) 0.92 (1.2) 2.0081 aB is the concentration of 1 at t = •, and N is the initial concentration of 1.b±15%.Table 2 Rate constant k of thermolysis of 1 at different temperatures.a T/°C k/s–1 B/Nb 110 1.55×10–5 0.95 125 5.80×10–5 0.92 145 1.42×10–4 0.95 150 1.83×10–4 0.80 10 G 4· 5· 6· Experiment Simulation Figure 2 Experimental (in squalane) and simulated EPR spectra of 4·–6·. Note minor extra lines of an unidentified radical in the spectrum of 5·. Differences in the HF line widths were ignored in simulating the spectra.Conversion t/min 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0 100 200 300 400 500 600 700 Figure 3 The kinetic build-up curve of 1· by thermolysis of 1 in squalane at 125 °C (initial concentration of 1 was equal to 10–3 mol dm–3).Mendeleev Communications Electronic Version, Issue 1, 2000 (pp. 1–42) This work was supported in part by the Russian Foundation for Basic Research (grant no. 99-03-33115) and by the Leading Scientific Schools Programme (grant no. 96-15-97564). References 1 A. W. Cordes, M. Hojo, H. Koenig, M. C. Noble, R. T. Oakley and W. T. Pennington, I norg. Chem., 1986, 25, 1137. 2 A. V. Zibarev, Yu. V. Gatilov and A. O.Miller, Polyhedron, 1992, 11, 1137. 3 I. Yu. Bagryanskaya, Yu. V. Gatilov, A. Yu. Makarov, A.M. Maksimov, A. O. Miller, M. M. Shakirov and A. V. Zibarev, Heteroatom Chem., 1999, 10, 113. 4 V. I. Minkin, B. Ya. Simkin and M. N. Glukhovzev, Aromaticity and Antiaromaticity. Electronic and Structural Aspects, Wiley–Interscience, New York, 1994. 5 A. P. Hitchcock, R. S. DeWitte, J.M. van Esbroek, P. Aebi, C. L. French, R. T. Oakley and N. P. C. Westwood, J. Electron Spectrosc.Relat. Phenom., 1991, 57, 165. 6 N. E. Petrachenko, Yu. V. Gatilov and A. V. Zibarev, J. Electron Spectrosc. Relat. Phenom., 1994, 67, 489. 7 A. V. Zibarev, Yu. V. Gatilov, I. Yu. Bagryanskaya, A. M. Maksimov and A. O. Miller, J. Chem. Soc., Chem. Commun., 1993, 298. 8 R. Mayer, S. Bleisch, G. Domschke, A. Tkac, A. Stasko and A. Bartl, Org. Magn. Reson., 1979, 12, 532. 9 S.Bleisch, G. Domschke and R. Mayer, GDR Patent 138 062DI, 1979 (Chem. Abstr., 1980, 92, 128892). 10 R. Mayer, S. Bleisch, G. Domschke and A. Bartl, Z. Chem., 1981, 21, 146. 11 R. Mayer, G. Domschke, S. Bleisch, J. Fabian, A. Bartl and A. Stasko, Collect. Czech. Chem. Commun., 1984, 49, 684. 12 G. Kirsch, in Methoden der Organische Chemie (Houben-Weyl), ed. E. Schaumann, Thieme, Stuttgart, 1994, vol. E8d, pp. 3–12. 13 M. J. Frisch, G.W. Trucks, H. B. Schlegel, P. M.W. Gill, B. G. Johnson, M. A. Robb, J. R. Cheeseman, T. Keith, G. A. Petersson, J. A. Montgomery, K. Raghavachari, M. A. Al-Latam, V. G. Zakrzewski, J. V. Ortiz, J. B. Foresman, J. Cioslowski, B. B. Stefanov, A. Nanayakkara, M. Challacombe, C. Y. Peng, P. Y. Ayala, W. Chen, M.W.Wong, J. L. Andres, E. S. Replogle, R. Gomperts, R. L. Martin, D. J. Fox, J. S. Binkley, D. J. Defrees, J. Baker, J. P. Stewart, M. Head-Gordon, C. Gonzales and J. A. Pople, Gaussian 94™. Revision C.3, Gaussian Inc., Pittsburg PA, 1995. ln k T–1/10–3 K–1 –8.5 –9.0 –9.5 –10.0 –10.5 –11.0 –11.5 2.30 2.35 2.40 2.45 2.50 2.55 2.60 2.65 Figure 4 Arrhenius plot of the formation rate constant of 1· in squalane. Received: 17th August 1999; Com. 99/1526

ISSN:0959-9436     

出版商:RSC    

年代:2000

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 1, 2000 (pp. 1–42) Selective photolysis of the azido groups in 2,4-diazidopyridines Sergei V. Chapyshev,*a Richard Waltonb and Paul M. Lahtib a Institute for Problems of Chemical Physics, Russian Academy of Sciences, 142432 Chernogolovka, Moscow Region, Russian Federation. Fax: +7 096 515 3588; e-mail: chap@icp.ac.ru b Department of Chemistry, University of Massachusetts, Amherst, MA 01003–4510, USA.Fax: +1 413 545 4490; e-mail: lahti@grond.chem.umass.edu DOI: 10.1070/MC2000v010n01ABEH001210 Selective photolysis of the a-azido group in 2,4-diazidopyridines has been determined by EPR spectroscopy and theoretically treated in terms of non-adiabatic photodissociation processes. The selective photochemical cleavage of chemical bonds in organic compounds is a very important problem in chemistry.1 In continuation of our research on selective derivatization of the azide groups in aromatic polyazides,2 an EPR study of the primary products formed in the photolysis of 2,4- and 2,6-diazidopyridines has been performed, the results of which are described here.Azides 1a–e, 3a,b, 5a–e and 8a,b were irradiated in degassed, frozen solutions of 2-methyltetrahydrofuran at l > 300 nm for 5 min at 77 K.† Upon irradiation all samples became blue and displayed X-band EPR spectra typical3–6 of triplet nitrene signals in the region of 6870–7100 G (Table 1).In order to differentiate the isomeric triplet pyridylnitrenes on the basis of their EPR characteristics, published data3–5 on the related EPR spectra were considered.In addition, PM3 computations‡ for a series of heteroarylnitrenes were carried out. Contrary to the case of parasubstituted phenylnitrenes,3,6 the zero-field splitting (zfs) D-values of heteroarylnitrenes correlate fairly well, although non-linearly, with electron-withdrawing properties of the heteroaryl substituents and with computed C–N bond lengths (Table 2).The adequacy of the PM3 computations is supported by an excellent fit of the geometrical parameters for triplet phenylnitrene calculated by this and CASSCF(8,8)/6-31G*8 methods (Figure 1). Using the correlation between the D-values and the PM3 C–N bond lengths for triplet heteroarylnitrenes, the EPR spectra of the primary products formed by photolysis of azides 1a–e, 3a,b, 5a–e and 8a,b were analysed.As can be seen in Table 1, among nitrenes 2a–e (generated from 1a–e), nitrenes 2b–d have the same D-values and C–N bond lengths, whereas 2a and 2e have lower D-values and shorter C–N bonds. An explanation of these effects comes from an analysis of the structures of 2a–e. The structures of 2b–d are better described by the resonance form I, those of 2a,e, by form II.The former has shorter C(5)–Cl and longer C(6)– NHAr bonds. The low D-value for 4a, formed by photolysis of 3a, is presumably associated with the presence of a strong electron- donating aziridine substituent at the pyridine ring of this nitrene. In comparison with 4a, dicyano derivative 4b has a higher D-value and longer C–N bond, due to the stronger electron-withdrawing character of the pyridine ring and the inability of a nitrene unit to conjugate with the para-chlorine atom. † The synthesis of azides 1a–e, 2a,b, 3a,b and 4a–e was described elsewhere. 13 The photolysis and EPR measurements were described previously. 14 ‡ The structures of azides, triplet nitrenes and excited states were calculated with the full optimization of geometrical parameters using the PM3 method (UHF, SCF level).15 The photolysis of 5a–d led to EPR spectra with two triplet nitrene signals.By analogy with 4a,b, the low-field signals were assigned to nitrenes 6a–e bearing the chlorine atom at the paraposition to the nitrene unit, whereas the high-field signals were assigned to nitrenes 7a–e. This assignment is supported by shorter C–N bond lengths in 6a–e, comparing with those for 7a–e (Table 1).The effect is quite small, but reasonably consistent. The nitrene peaks from photolysis of 5b,e showed shoulders, but were insufficiently resolved to distinguish two different nitrene signals. The small effect of substituents in the ArNH moieties on the D-value of 6a–e and 7a–e is likely explained by a weak conjugation of these moieties with the pyridine ring.The C(5)– C(4)–N–C(1' ) torsion angles in 6a–e and 7a–e are about 122°, while the related angles in 2a–e are about 20°. In contrast to 5a–e, diazides 8a,b upon irradiation displayed in the EPR spectra only one triplet nitrene, the line position of which was almost the same as that for nitrenes 2b–d. An answer to the question of which nitrenes, 2-nitreno 9a,b or/and 4-nitreno 10a,b, give the EPR signals in the experiment, was suggested by an analysis of the C–N bond lengths in 9a,b and 10a,b.The substantial difference in these bonds for 9a,b and 10a,b (Table 1) suggests that signals of these isomers should not coincide in the EPR spectra. Nitrenes 9a,b match better to the spectral line positions observed. N N N SCF PM3 CASSCF(8,8)/6-31G* MCSCF(8,8)/3-21G 3A2 1.326 1.434 1.387 1.402 1.338 1.425 1.386 1.404 1.402 1.404 1.391 1.396 3A2 3A2 Figure 1 The molecular geometry parameters of triplet phenylnitrene, computed by the SCF PM3, CASSCF(8,8)/6-31G*8 and MCSCF(8,8)/3-21G9 methods.aShoulder. Table 1 The line positions, D-values and computed C–N bond lengths of nitrenes 2a–e, 4a,b, 6a–e, 7a–e, 9a,b and 10a,b.Nitrene Field/G |D hc–1|/cm–1 C–N/Å 2a 7008 1.026 1.3291 2b 7045 1.040 1.3297 2c 7045 1.040 1.3297 2d 7105 1.063 — 2e 6911 0.989 1.3282 4a 6877 0.976 1.3282 4b 7020 1.031 1.3299 6a 6877 0.976 — 6b 6877 0.976 1.3260 6c 6877 0.976 1.3260 6d 6877 0.976 — 6e 6899 0.983 — 7a 7059 1.046 — 7b —a — 1.3264 7c 6989 1.019 1.3264 7d 6990 1.018 — 7e —a — — 9a 7040 1.038 1.3290 9b 7105 1.062 1.3293 10a — — 1.3277 10b — — 1.3278 Table 2 The D-values and computed C–N bond lengths of nitrenes.Nitrene |D hc–1|/ cm–1 C–N/Å Reference Py-3-N 1.005 1.3248 3 Py-2-N 1.051 1.3308 4 5-Trifluoromethylpyridyl-2-nitrene 1.087 1.3315 5 3,5-Bis(trifluoromethyl)pyridyl-2-nitrene 1.108 1.3333 5 Diazidocyanurylnitrene 1.44 1.3571 7Mendeleev Communications Electronic Version, Issue 1, 2000 (pp. 1–42) Both the C–N bond lengths and the D-values for 9a,b are almost equal to those for 2-nitreno species 2b–d, which have similar nitrene peak positions and D-values. The selective photolysis of the a-azido group in 8a,b can be rationalised from an analysis of the energies for two different excited T0 states of 8a,b (a-T0 and g-T0 states for dissociation of the a- and g-azido group, respectively).The photodissociation of chemical bonds in the molecules occurs as a result of energy transfer from the potentials of the excited states (curves 2 and 2' in Figure 2) onto the potential of the repulsive term (curve 3).10 The higher the energy of the excited states, the higher the probability that surfaces 2 or 2' will be crossed by surface 3.As has been recently shown,11 the most prominent feature of the azides in their excited T0 state is the considerable bending of the azido group (from about 170° in the ground state to about 128° in the T0 state) and a dramatic fall in the activation energy of N–N2 bond dissociation (from about 35 to 2 kcal mol–1, respectively). The computational distortion of the a- or g-azido group in 8a,b to 135° followed by the PM3 geometry optimization of the molecules in the T0 state allowed us to determine the energies of the a-T0 and g-T0 states of 8a,b.The energy of the former state was found to be higher by 4 kcal mol–1. This finding supports the idea that the a-excited states (T0, S1, T1, etc.)11 of 8a,b, indeed, lie more closely to the reaction coordinate, so that photodissociation of the a-azido groups in these compounds should be1,10 the preferable process.Note that the PM3 computation of the T0 state for 8a without preliminary bending of one of the azido groups was found to yield the a-T0 state by preference. Another interesting finding is the substantial difference in the orbital density distribution between the a- and g-azido groups in the HOMO of 8a,b (Figure 3).Assuming that the loss of stronger bonding interactions in the molecules leads to the formation of the less stable intermediates,12 the presence of a higher bonding orbital density at the Na and Nb atoms11 of the a-azido group in the HOMO of 8a,b explains fairly well the higher E G 1 2 3 2' 2'' Ed hn kd Figure 2 Schematic diagram of the electronic terms for non-adiabatic dissociation (1, 2, 2' and 2'' are electronic terms for the ground and excited states, respectively; 3 is the repulsive term; Ed is the energy of dissociation; kd is the dissociation rate constant).N HN N3 CN Cl Cl N HN N CN Cl Cl N HN N CN Cl Cl N HN N CN Cl Cl N N3 N3 R N R N N3 N R N R N N3 N CN NHC6H4R Cl N N3 N3 CN NHC6H4R Cl N N N3 CN NHC6H4R Cl N RC6H4HN N3 CN N Cl N RC6H4HN N3 CN N3 Cl N RC6H4HN N CN N3 Cl a R = p-OMe b R = p-Me c R = H d R = p-Br e R = m-NO2 1a–e 2a–e I II 1, 2, 5–7: 3a,b 4a,b R R R R a R = Cl b R = CN a R = p-Me b R = p-Br 3, 4: 8–10: hn hn hn hn hn hn 6a–e 5a–e 7a–e 8a,b 9a,b 10a,b d+ d+ d– d–Mendeleev Communications Electronic Version, Issue 1, 2000 (pp. 1–42) energies for the a-T0 state of these molecules.Figure 3 also demonstrates that the azido groups in the HOMO of 5b are equal (have almost no orbital density at the Na atoms). Thus, it is no surprise that the energies of the a- and a'-T0 states for 5b were also computed to be equal, and the photolysis of this diazide led to formation of both isomers 6b and 7b. It seems reasonable to assume that selective photolysis of the azido groups in polyazides can be possible only for those compounds in which the azido groups significantly differ in the HOMO orbital density distribution.Conversely, the computational modelling of the bonding orbital density will be helpful to predict the relative photolability of azido groups in aromatic polyazides. This work was supported by the U.S. National Science Foundation (grant nos.CHE-951595 and CHE-9740401). References 1 F. F. Crim, J. Phys. Chem., 1996, 100, 12725. 2 (a) S. V. Chapyshev, Mendeleev Commun., 1999, 164; (b) S. V. Chapyshev, Mendeleev Commun., 1999, 166; (c) S. V. Chapyshev, R. Walton and P. M. Lahti, Proceedings of International Workshop on Reactive Inter-mediates, IWRI’99, Szczyrk, Poland, 1999, p. 8. 3 E. Wasserman, Prog. Phys. Org.Chem., 1971, 8, 319. 4 M. Kuzaj, H. Lüerssen and C. Wentrup, Angew. Chem., Int. Ed. Engl., 1986, 25, 480. 5 R. A. Evans, M. Wong and C. Wentrup, J. Am. Chem. Soc., 1996, 118, 4009. 6 J. H. Hall, J. M. Fargher and M. R. Gisler, J. Am. Chem. Soc., 1978, 100, 2029. 7 R. M. Moriarty, M. Rahman and G. J. King, J. Am. Chem. Soc., 1966, 88, 842. 8 W. L. Karney and W. T. Borden, J.Am. Chem. Soc., 1997, 119, 1378. 9 D. A. Hrovat, E. E. Waali and W. T. Borden, J. Am. Chem. Soc., 1992, 114, 8698. 10 (a) M. Ya. Mel’nikov and V. A. Smirnov, Fotokhimiya organicheskikh radikalov (Photochemistry of Organic Radicals), Izd. Mosk. Univ., Moscow, 1994, p. 93 (in Russian); (b) R. Schinke, Photodissociation Dynamics, Cambridge University Press, Cambridge, 1993, p. 1. 11 M. F. Budyka and T. S. Zyubina, J. Mol. Struct. (Theochem), 1997, 419, 191. 12 H. Fujimoto and K. Fukui, in Chemical Reactivity and Reaction Paths, ed. G. Klopman, Wiley–Interscience, New York, 1974, ch. 3. 13 (a) S. V. Chapyshev and T. Ibata, Heterocycles, 1993, 36, 2185; (b) S. V. Chapyshev and N. V. Chapysheva, Khim. Geterotsikl. Soedin., 1994, 666 [Chem. Heterocycl. Compd. (Engl. Transl.), 1994, 30, 585]; (c) S. V. Chapyshev and V. M. Anisimov, Khim. Geterotsikl. Soedin., 1997, 1521 [Chem. Heterocycl. Compd. (Engl. Transl.), 1997, 33, 1315]. 14 (a) C. Ling, M. Minato, P. M. Lahti and H. van Willigen, J. Am. Chem. Soc., 1992, 114, 9959; (b) C. Ling and P. M. Lahti, J. Am. Chem. Soc., 1994, 116, 8784; (c) R. S. Kalgutkar and P. M. Lahti, J. Am. Chem. Soc., 1997, 119, 4771; (d) S. V. Chapyshev, R. Walton, J. A. Sanborn and P. M. Lahti, J. Am. Chem. Soc., in press. 15 (a) J. J. P. Stewart, J. Comput. Chem., 1989, 10, 221; (b) Spartan, version 4.0, Wavefunction, Inc., USA, 1995. Figure 3 The orbital density distribution in the HOMO of 5b and 8a. The higher bonding orbital density at the Na atom of the a-azido group in the HOMO of 8a indicates that a-excited states (T0, S1, T1, etc.) of 8a,b should lie more closely to the reaction coordinate. 5b 8a Received: 4th October 1999; Com. 99/1538

ISSN:0959-9436     

出版商:RSC    

年代:2000

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 1, 2000 (pp. 1–42) Characteristic luminescence of mercury and mercury-like ions at electrodes covered with insulating films Vladimir V. Yagov* and Andrei S. Korotkov V. I. Vernadsky Institute of Geochemistry and Analytical Chemistry, Russian Academy of Sciences, 117975 Moscow, Russian Federation. Fax: +7 095 938 2054; e-mail: zubor@geokhi.ru DOI: 10.1070/MC2000v010n01ABEH001196 Intense electrogenerated luminescence of tin, indium and mercury at Al/Al2O3, Be/BeO, Mg/MgO and Mg/MgF2 electrodes was observed, and a mechanism of direct excitation of mercury-like centres by reduction with hot electrons was proposed.Cathodic polarisation of insulating film-coated electrodes proceeds with the participation of hot electrons, and these processes can result in light emission.1 Because of the barrier properties of Al2O3, ac electrolysis with an aluminium electrode is a versatile technique for generating electroluminescence (EL).2–7 Various heavy metals can exhibit EL at Al/Al2O3.Indium,7 tin8 and mercury3,5,9 show the most intense emission. An attempt to induce EL of these activators at different insulator-covered electrodes is undertaken here.The dependence of EL on the electrode material has two basic aspects. First, the energetic structure of a barrier layer determines the type of the energy distribution function of cathodic electrons. In some sense the alternation of electrode materials in the case of EL is similar to a shift of the wavelength of excitation light in the case of photoluminescence (PL).Second for, In, Sn and Hg, the barrier layer probably plays a role of the base of a surface phosphor.7 It is known that hydrated In, Sn and Hg ions do not exhibit PL at room temperature. Therefore, it seems reasonable that the EL of these metals is a solid-state electroluminescence and strongly depends on the nature of an insulating film. Preliminary experiments with Al/Al2O3, Mg/MgO, Mg/MgF2, Be/BeO, Ca/CaF2, Y/Y2O3 and Zr/ZrO2 revealed that only aluminium, magnesium and beryllium electrodes are suitable for inducing the characteristic EL of tin, indium and mercury.Thus, only the results concerning Al/Al2O3, Mg/MgO, Mg/MgF2 and Be/BeO will be discussed below. Note that Mg/MgF2 and Be/BeO electrodes were not used for EL generation before.A Mg/MgO electrode was utilised to induce electrochemiluminescence.10 The three-electrode electrochemical cell (Figure 1) in a quartz cuvette 1 was positioned in front of a MDR-3 monochromator. Working electrode 2 (99.99% Al, 99.985% Mg or 99.98% Be) was fixed against a 8 mm diameter window in the wall of Teflon sample holder 3 by Teflon screw 4. Graphite rod 5 was used as an auxiliary electrode.Working and auxiliary electrode compartments were separated by sample holder 3. All potentials were measured against a saturated Ag/AgCl reference electrode. Rectangular potential pulses were applied to the EL cell using a PI-50-1.1 potentiostat. In some cases continuous ac electrolysis was used, or 0.5 s periods of electrolysis were alternated with 5 s pauses.The response of a PM tube was recorded with an L154 PC data acquisition board. The integrated signal from a fixed number (several hundreds) of cathodic pulses was used as the EL intensity in spectra. The spectra were measured in a range from 300 to 700 nm and corrected for the spectral sensitivity of the recording system. A preliminary deposition procedure9 was used to study the EL of mercury at Mg/MgO and Al/Al2O3 electrodes. After cementation of mercury from a 0.001 M Hg(NO3)2 solution in 0.5 M NaCl, the electrode was washed with distilled water and transferred to a cell filled with 0.02 M NaBrO3 for EL generation.In all other cases an activator was added directly to the solution in which the EL generation was carried out. The conditions of EL generation (Table 1) were optimised to provide the most intense and stable luminescence.The most intense emission (systems 1, 2 and 4) was visible in daylight. The cathodic current densities varied from less than 0.03 A cm–2 for bromate solutions (systems 7 and 9) or 0.2 A cm–2 for ammonium fluoride solutions (systems 3, 5 and 8) to 1 A cm–2 for H3PO4 solutions (systems 1, 2 and 4).It is impossible to induce the EL of Sn and In at Mg/MgO because, on the one hand, the hydrolysis of SnII, SnIV and InIII inhibits the generation, and, on the other hand, MgO is readily soluble in acids. Thus, only a Mg/MgF2 electrode can be used for the generation of Sn and In electroluminescence. As can be seen in Figure 2 and Table 1, the spectra and intensities of the EL of Al2O3–Sn (curve 1) and BeO–Sn (curve 2) practically coincide.Note that the EL spectra are very similar to the PL spectra12 of solid phosphates doped with Sn2+. It seems reasonable that Sn2+, which is extensively used as an activator in phosphors, also acts as an emitter in the case of EL. The maximum of the MgF2–Sn spectrum (curve 3) is shifted by about 30 nm towards the short-wave region.This fact is probably associated with a fluoride environment of Sn2+ in this system. The EL spectrum of indium is complex and can vary depending on experimental conditions.7 As Figure 2 indicates, the spectra of Al2O3–In (curve 4) and MgF2–In (curve 5) contain strong blue components and weak yellow components that manifest themselves as shoulders. The EL spectrum of BeO–In (curve 6) has a maximum at 580 nm, which corresponds to the shoulders in curves 4 and 5, whereas there are no peaks in the blue–green region in curve 6.As follows from a comparison with the PL data,13,14 the blue component is associated with the 5sp ® 5s2 transition in In+. The yellow component appears to be associated with the luminescence of In2+. This assumption is in agreement with the results of PL measurements carried out in glasses doped with In+ under conditions of powerful laser irradiation. 14 It was found that an increase of the yellow component in PL spectra corresponds to the production of In2+, which results from the photoionization of In+ centres.14 The EL of mercury can be observed at Al/Al2O3, Mg/MgF2 and Mg/MgO electrodes rather than at a Be/BeO electrode. As Table 1 indicates the intensity of EL is maximum for Mg/MgF2 and minimum for Mg/MgO.The spectra of Al2O3–Hg (Figure 2, curve 7) and MgF2–Hg (Figure 2, curve 8) are rather similar, whereas in the spectrum of Mg/MgO–Hg (Figure 2, curve 9) the peak is strongly broadened and shifted to the long-wave region. The similarity of curves 7 and 8 and the dissimilarity of curves 7 and 9 are surprising when it is considered that the initial forms in which mercury was introduced are different for the first couple of curves and coincide for the second one (Table 1).Meulenkamp et al. suggested5 that the Hg2+ cation is responsible for the EL at an Al/Al2O3 electrode. However, as distinct from other d10 ions, there is no information on the PL of Hg2+.In our opinion,9 the mercury atom with the d10s2 structure produces EL. The fact that the EL intensity of mercury is at least 1 2 3 4 5 6 7 Figure 1 Schematic diagram of the EL cell: (1) quartz cuvette, (2) working electrode, (3) sample holder, (4) screw, (5) auxiliary electrode, (6) Luggin capillary, (7) cover.Mendeleev Communications Electronic Version, Issue 1, 2000 (pp. 1–42) by two orders of magnitude higher than that for other d10 emitters5 and is close to that of d10s2 ions is evidence in favour of this assumption. The production of excited mercury atoms and excited mercury-like ions (Sn2+ and In+), apparently, proceeds as the reduction of oxidised species with the d10 structure (HgII, SnIV and InIII) by hot electrons according to the scheme: When mercury is introduced as Hg0, an oxidant is required for EL generation.The active intermediate products of bromate reduction convert Hg0 into Hg2+; next, the above scheme is valid. Thus, two ways can lead to light emission. Way I appears to be of secondary importance with respect to way II, because of very short lifetimes of intermediate species (Sn3+, In2+ or Hg+). Nevertheless, the contribution of In2+ to the EL spectrum is significant.As there is no information on the Sn3+ or Hg+ luminescence, we assume that Sn2+ or Hg0 are the main EL emitters. Because luminescence centres are localised in a solid insulating film, it is reasonable to expect that EL is sensitive to its properties. This sensitivity is evident in comparison of the EL at a Be/BeO electrode with other cases.Mercury does not exhibit EL at a Be/BeO electrode, and the spectrum of BeO–In (Figure 2, curve 6) dramatically differs from the spectra of Al2O3–In (curve 4) and MgF2–In (curve 5). This is likely due to great differences in size between a very small Be2+ cation (0.034 nm) and rather large Hg2+ (0.11 nm) and In3+ (0.092 nm) cations, which makes their incorporation into a BeO lattice difficult.The Sn4+ cation appears to exhibit a rather small size (0.067 nm) to enter into a BeO film. The surprising thing is that the structural selectivity is observed only in the case of a drastic difference in size between host and guest cations. Probably, an easy incorporation of guest cations into a barrier layer is facilitated by defects produced by impacts of hot electrons.Moreover, during a cathodic pulse, the entry of guest cations into the film is promoted by a high electric field. Note that a high electric field strength (several V nm–1) can result in a significant Stark effect on EL spectra. Thus, not only an Al/Al2O3 electrode is appropriate to induce the EL of mercury-like species. In our opinion, the EL of mercurylike species at insulating film-coated electrodes is a special type of recombination luminescence that occurred in thin solid layers when electrons injected from a metal directly interact with d10 cations sorbed from the solution. By now, the elements exhibiting EL upon electrochemical reduction from d10 to d10s2 species are Hg, Sn, Pb, Ga, In and Tl,3,5,6 and it is likely that this range will be extended.This work was supported by the Russian Foundation for Basic Research (grant no. 98-03-32668). References 1 S. Kulmala, T. Ala-Kleme, A. Hakanen and K. Haapakka, J. Chem. Soc., Faraday Trans., 1997, 93, 165. 2 M. I. Eidelberg and T. Z. Tseitina, Zavod. Lab., 1978, 2, 197 (in Russian). 3 K. Haapakka, J. Kankare and S. Kulmala, Anal. Chim. Acta, 1985, 171, 259. 4 S. P. Eremenko and M. I. Eidelberg, Zh. Prikl. Spektrosk., 1975, 22, 842 (in Russian). 5 E. A. Meulenkamp, J. J. Kelly and G. Blasse, J. Electrochem. Soc., 1993, 140, 84. 6 V. V. Yagov and B. K. Zuev, Zh. Anal. Khim., 1996, 51, 287 [J. Anal. Chem. (Engl. Transl.), 1996, 51, 266]. 7 V.V.Yagov, Dokl. Ross. Akad. Nauk, 1996, 350, 226 [Dokl. Chem. (Engl. Transl.), 1996, 350, 219]. 8 V.V.Yagov, Zh.Anal. Khim., 1997, 52, 536 [J. Anal. Chem. (Engl. Transl.), 1997, 52, 477]. 9 V.V.Yagov, Zh. Anal. Khim., 1996, 51, 502 [J. Anal. Chem. (Engl. Transl.), 1996, 51, 462]. 10 S. Kulmala, A. Kulmala, T. Ala-Kleme and J. Pihlaja, Anal. Chim. Acta, 1998, 367, 17. 11 N. D. Tomashov and G. P. Chernova, Teoriya korrozii i korrozionnostoikie konstruktsionnye splavy (Theory of Corrosion and Corrosion- Resistant Constructional Alloys), Metallurgiya, Moscow, 1993, p. 416 (in Russian). 12 H. Donker, W. H. A. Smit and G. Blasse, J. Electrochem. Soc., 1989, 136, 3130. 13 M. U. Belyi, B. A. Ohrimenko and S. M. Yablochkov, Ukr. Fiz. Zh., 1981, 26, 1790 (in Russian). 14 M. U. Belyi, S. E. Zelenskyi, B. A. Ohrimenko and S. M. Yablochkov, Izv. Akad. Nauk SSSR, Ser. Fiz., 1985, 49, 2010 (in Russian). Table 1 Conditions of EL generation. no. Barrier film– activator Cathodic pulse Anodic pulse Solution Intensity (arbitrary units) Length/ms Potential/V Length/ms Potential/V 1 Al2 O3 –Sn 5 –6 5 3 0.01 M SnIV in 1 M H3PO4 12 2 BeO–Sn 1 –7 0.5 4.5 0.005 M SnIV in 0.2 M H3PO4 12 3 MgF2 –Sn 1 –8 1 1 0.002 M InIII in 0.35 M NH4F·HF 3 4 Al2 O3 –In 5 –6 5 3 0.02 M InIII in 1 M H3PO4 15 5 MgF2 –In 1 –8 0.5 2.5 0.002 M InIII in 0.35 M NH4F·HF 2 6 BeO–In 2 –8 1 5 0.02 M InIII in 1 M H3PO4 3 7 Al2 O3 –Hg 5 –8 5 4 Hg film at Al in 0.02 M NaBrO3 4 8 MgF2 –Hg 1 –8 0.5 0.5 0.001 M Hg(NO3)2 in 0.35 M NH4F·HF 5 9 MgO–Hg 5 –8 1 8 Hg film at Mg in 0.02 M NaBrO3 1 d10p d10sp d10 d10s d10s2 e e* e e e* I II 1 0 Intensity (arbitrary units) l/nm 350 400 450 500 550 600 650 700 1 2 3 4 5 6 7 8 9 2 6 3 Figure 2 EL spectra (curve numbers correspond to those in Table 1). Received: 5th August 1999; Com. 99/1524

ISSN:0959-9436     

出版商:RSC    

年代:2000

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 1, 2000 (pp. 1–42) Tunnelling effects in the oxidative addition of a dihydrogen molecule to the palladium ethylenediphosphine complexes [H2P(CH2)2PH2]Pd and [H2P(CH2)2PH2]Pd2 Viktor M. Mamaev,*a Igor P. Gloriozov,a Vahan V. Simonyan,a Elena V. Zernova,a Eugene M. Myshakina and Yuri V. Babinb a Department of Chemistry, M. V. Lomonosov Moscow State University, 119899 Moscow, Russian Federation.Fax: +7 095 938 8846; e-mail: vmam@qnmr.chem.msu.su b Far-Eastern State Academy of Economics and Management, 690091 Vladivostok, Russian Federation. Fax: +7 4232 40 6634; e-mail: yub@mail.primorye.ru DOI: 10.1070/MC2000v010n01ABEH001205 In terms of the reaction-path Hamiltonian formalism, the ligand environment has been found to have almost no effect on the activation barrier in the oxidative addition of H2 to the title complexes (the under-barrier tunnelling in the reaction was predominant at T < 230 K).The currently available technologies of oil and gas processing are based on high-temperature catalytic processes and result in high energy consumption. Therefore, it is important to develop new ways of activation of small molecules by transition metal complexes.Theoretical studies devoted to the activation of unsaturated bonds are few in number. As a rule, a static quantumchemical model is used, which describes the reaction kinetics and mechanism only in terms of energy and geometry of the structures corresponding to the stationary points of potential energy surfaces (PES).Nevertheless, the oxidative addition of dihydrogen implies a substantial contribution of tunnelling to the overall kinetics of the reaction. This contribution can be estimated only by a dynamic model developed using the reactionpath Hamiltonian (RPH)1 formalism, which includes an analysis of dynamic structures of stationary points on PES. We report here a quantum-chemical study of the oxidative addition of the hydrogen molecule to [H2P(CH2)2PH2]Pd complex 1 taking into account a contribution of tunnelling to the overall reaction rate constant: Complex 1 can be used as a model of an unsaturated 14- electron palladium complex with bis(diphenylphosphino)ethane, whose active participation in the activation of small molecules was found experimentally.2 The oxidative addition of hydrogen and methane molecules to transition metal atoms, clusters and complexes was not examined by ab initio methods in the RPH approximation taking into account a tunnelling contribution to the thermal rate constant.With the use of high-level ab initio methods, the consideration was usually restricted by calculations of stationary points or a unidimensional potential depending on the interior coordinate.3 Note that, even in the calculation of PES stationary points of the activation of H–H and C–H bonds by transition metals and their complexes, the results of the calculations significantly depended on the ab initio method used. The reaction energies for the activation of a C–H bond by Pd and Ni atoms (DE) were compared in ref. 4. The energies were calculated by ab initio methods of two levels, which were different in the size of the basis and in the consideration for the electron correlation (Large: IC-ACPF, large basis, CH-correlation. Standard: CCI + Q, standard basis, no CH-correlation).In the case of Pd, DE = 9.1 kcalmol–1 (Large) and 17.6 kcal mol–1 (Standard); in the case of Ni, DE = = –3.3 kcal mol–1 (Large) and 4.3 kcal mol–1 (Standard).Hence it follows that, with the use of ab initio methods, it is also difficult to calculate the most accurate values of relative energies for stationary points. As for the applicability of the density functional theory (DFT), which significantly shortens the calculation time and is a good alternative to the ab initio method with correlation corrections, it gives a minimum in the Pd + H2 system only for the pre-reaction complex with the H–Pd–H angle 28° and the distance R(H–H) = 0.85 Å.This is in contradiction with the most accurate calculations.5 We obtained a similar result by calculating the molecular system under discussion using the DFT method. The calculations of PES were performed by the CNDO/S2 method.6 This semiempirical method was specially developed for calculating PES of molecular systems containing transition metals.The method was parametrised on the basis of both experimental data and high-level ab initio calculations. It was reliable in a study of the activation of H–H and C–H bonds by the Pd atom and Pd2 cluster.7 Relative energies and geometry parameters of the stationary points were also calculated by the ab initio Hartree– Fock (HF) and MP2 methods from the GAMESS program package8 in the 6-31G(d) basis set with the SBK pseudo-potential9 on the palladium atom (Table 1).For comparison, Table 1 also includes data10 on the high-level calculations (MP4) of methane PH2 + H2 Pd H2P PH2 Pd H2P H H (I) 1 E/kcal mol–1 w/cm–1 s/Å m1/2 –1.5 –1.0 –0.5 0.0 0.5 1.0 8 6 4 2 0 –2 –4 –6 4500 4000 3500 3000 2500 2000 1500 1000 500 0 –500 –1000 PC TS RP q(H–H) c(HH–Pd) q(HH–Pd) q(Pd–H), (A) q(Pd–H), (B) a(H–Pd–H), (A) Figure 1 Potential V0(s) of the [H2P(CH2)2PH2]Pd + H2 reaction (thick solid line) and frequencies of the most important normals against the reaction coordinate s (m is the proton mass).Table 1 Relative energies and geometry parameters (bond lengths in Å) of stationary points on PES in reaction (I).Method E/kcal mol–1 r(H–H) r(Pd–H) �H–Pd–H r(Pd–P) CNDO/S2 –5.3 0.82 2.19 21° 2.72 HF –1.5 0.75 2.07 21° 2.47 MP2 –1.5 0.82 1.85 25° 2.29 MP410 –5.1 — — — 2.32 CNDO/S2 6.2 1.31 1.60 48° 2.66 HF 13.3 1.32 1.59 49° 2.52 MP2 5.0 1.39 1.65 50° 2.34 MP410 — — 1.56 — 2.38 2.46 CNDO/S2 3.0 1.95 1.55 78° 2.54 HF 10.1 1.93 1.57 76° 2.50 MP2 4.0 1.89 1.62 71° 2.33 MP410 — — 1.55 — 2.39 2.45 PC TS RPMendeleev Communications Electronic Version, Issue 1, 2000 (pp. 1–42) C–H bond activation by complex 1. It can be seen in Table 1 that the stationary points found by different methods are rather close in geometry (except for Pd–P bond lengths, which were overestimated by the CNDO/S2 method as well as by ab initio calculations without considering the correlation energy) and comparable in energy; this fact confirms the reliability of our results.Note that, for stationary points of the Pd + H2 reaction, the results of the CNDO/S2 calculations11 are consistent with the ab initio calculations by Low and Goddard.5(a) We plotted the RPH for the oxidative addition of H2 to complex 1 and calculated the V0(s) potential along the reaction path (RP) and the vibration frequencies wi(s) of modes orthogonal to RP, where s is the reaction coordinate expressed in weighted Cartesian coordinates.We found that the reaction path passes through three stationary points on PES: pre-reaction complex (PC), transition state (TS), and reaction product (RP). The molecular system retained the axis of symmetry C2. The dependence of the energy and vibration frequencies orthogonal to RP on the reaction coordinate s for reaction (I) is presented in Figure 1 (vibration frequencies that change insignificantly during the reaction are not presented; the region in which the frequency exhibits imaginary values is represented by the negative semi-axis).In the PC region, the reaction coordinate is the vibration with the frequency wq(Pd–HH) = 206 cm–1 corresponding to the motion of H2 to the Pd complex.The vibration frequency of H–H is close to that in the dihydrogen molecule [q(H–H) = 4390 cm–1]. The third vibration with a low frequency [wc(HH–Pd) = 200 cm–1] corresponds to the rotation of hydrogen about the line connecting the palladium atom with the centre of the H–H bond.The geometry of the complex in the PC region remains almost unchanged asompared to the separated structure. The Pd–H bond length indicates a weak interaction of the dihydrogen molecule and the Pd atom: the Pd–H bond order is less than 0.1. An analysis of the RP vector shows that, in the TS region at s > –0.5, the change in the system geometry is mainly determined by a change in one internal coordinate, namely, a(H–Pd–H) [in TS, wa(H–Pd–H) = 944i cm–1).During the motion along RP, the q(H–H) vibration of the H2 molecule is transformed into the stretching vibration q+(Pd–H) (symmetry A), and the free rotation of the H2 molecule about the axis perpendicular to the C•(H–H) axis (whose frequency is equal to zero in the separated reactants) is transformed into the stretching vibration q–(Pd–H) (symmetry B).In the region from PC to TS, a considerable decrease in the vibration frequency of the H–H bond q(H–H) exceeds an increase in the rotation frequency of the dihydrogen molecule c(H–H). As the result, taking account of the frequencies [see equation (2)] leads to a decrease in the reaction barrier (unlike that in the Pd + H2 reaction11).In the TS and RP regions, the Pd–H bond length in the complex structure changes especially strongly. In the PC region, the bond has the distinct s-character, and d-orbitals do almost not participate in the bond formation, while the Pd–H distance decreases due to sd-hybridization in the TS and RP regions. The overall thermal rate constants of the bimolecular reactions were calculated in terms of the transition state theory with tunnelling correction:12,13 where m is the effective mass of motion of the molecular system (MS) along RP, kB is the Boltzmann constant, E is the energy of collisions of particles and Pn(E) is the probability of the reaction in the case when reactants considered as MS exist in the nth vibrational state (n = {n1, n2, ..., n3N – 7} is the vector of quantum numbers, and N is the number of atoms of MS).The probability of the reaction Pn(E) was calculated in the quasi-classic approximation12 with vibration-adiabatic functions V(n,s): Plotting RPH makes it possible to calculate the dependence of the frequencies of transversal vibrations on the reaction coordinate. Taking account of these frequencies can change substantially both the height and the shape of the energy barrier.Moreover, the effect of selective vibrational excitation on the shape of the potential curve and, hence, on the kinetics of reactions can be analysed. Note that, in the reaction at temperatures below 500 K, the population of vibrational levels is insignificant (< 1%), except for the ground state.Therefore, all data presented below are related to the ground vibrational level (n = 0). Figure 2 illustrates the results of calculations of the probabilities P(E) for reaction (I). As in the case of the reaction of the palladium atom with the dihydrogen molecule, tunnelling and repulsion from the barrier make a great contribution to the reaction probability. Therefore, these quantum effects should be taken into account in the calculation of the thermal rate constant.The account of them results in the deviation from the linear dependence of the logarithm of the overall thermal rate constant on the inverse temperature (Figure 3). 1.0 0.8 0.6 0.4 0.2 0.0 4 5 6 7 8 9 10 P E/kcal mol–1 Figure 2 Reaction probability P(E) for the reactions of [H2P(CH2)2PH2]Pd with dihydrogen. The Heaviside function is designated by a dotted line.–2 –4 –6 –8 –10 –12 –14 –16 –18 0.002 0.004 0.006 0.008 0.010 lg k T–1/K–1 Figure 3 Temperature dependence of the logarithms of the overall quantum (solid) and classic (dotted) rate constants for the [H2P(CH2)2PH2]Pd + H2 reaction. k(n,T) = P(n,E)exp(– )dE 1 2pmkBT E kBT ò E • (1) V(n,s) = V0(n,s) + h (n + 1/2)wj(s) (2) Sj = 2 3N – 6 k 14 12 10 8 6 4 2 0 100 200 300 400 T/K ~230 K Figure 4 Temperature dependence of the transmission coefficient for the [H2P(CH2)2PH2]Pd + H2 reaction.Mendeleev Communications Electronic Version, Issue 1, 2000 (pp. 1–42) The integral in the expression for the rate constant [equation (1)] can be presented as the sum: where Vmax is the potential barrier.The first term corresponds to the under-barrier particle flow, and the latter two terms correspond to the over-barrier particle flow through the barrier. This makes it possible to estimate separately the under-barrier (tunnelling) and over-barrier (activation) contributions to the rate constant. It was accepted that P(E) = 1 at E > 2Vmax. The temperature dependence of the transmission coefficient was calculated for the quantitative estimation of the tunnelling contribution to the total rate constant (Figure 4).It is evident that at k > 2 the tunnelling contribution to the rate constant is higher than the activation contribution. Thus, PES in the RPH approximation and the kinetics of reaction (I) with account of tunnelling have been completely analysed for the first time.An analysis of the mechanism of activation of the H–H bond shows that the [H2P(CH2)2PH2]Pd + H2 reaction is similar to the Pd + H2 reaction in both the geometry of stationary points and the activation barriers, except for the fact that the account of vibration frequencies results in a decrease in the barrier of reaction (I), unlike that of the Pd + H2 reaction.11 The kinetic behaviours of these reactions are also similar.For example, tunnelling also predominates in reaction (I) at temperatures below 230 K. At room temperature, the contribution of tunnelling to the reaction rate constant is about 25%. It is well known that the ligand environment at the Pd atom increases the activation barrier of oxidative addition of small molecules as, for example, in the Pd(PH3)2 and PdCl2(PH3)2 complexes.14 In this case, the use of H2P(CH2)2PH2 as the ligand did almost not affect the potential barrier as compared to that in the activation of H2 by the Pd atom.Fayet et al.15 experimentally examined the activation of small molecules by Pdn clusters (n = = 1–25). Neutral palladium clusters were found to be much more active than atomic palladium.Maximum activities were observed for Pd2 and Pdn > 5. Earlier,7 we also theoretically studied the activation of a dihydrogen molecule by the Pd2 cluster and found that its activity is higher than that of the Pd atom. The next step was a study on the oxidative addition of a dihydrogen molecule to the hypothetical complex [H2P(CH2)2PH2]Pd2 2. The RPH was constructed, and the stationary points were found.A rather deep potential well (EPC = –8.6 kcal mol–1) corresponds to the optimised PC structure (C2 symmetry). The dihydrogen molecule is perpendicular to the plane of the complex, and r(H–H) = 0.85 Å. The TS and RP exhibit the same symmetry (Figure 5). The energies of all optimised structures on the PES are lower than the energy of the separated reactants (SR); that is, the reaction is formally barrierless and exothermic.However, we suggested that the formation of RP from TS is the rate-limiting step and performed corresponding calculations for this step. Taking into account the frequencies of vibrations transverse with respect to the reaction path results in a significant decrease in the barrier height (from 5.1 to 3.8 kcal mol–1).Tunnelling was found to be predominant below 100 K, whereas at high temperatures the reaction mainly proceeds by an activation mechanism. In conclusion, activation of the H–H bond by the binuclear palladium complex with ethylenediphosphine significantly lowers the barrier for oxidative addition of a dihydrogen molecule as compared with the mononuclear complex. In this case, the energy of the TS becomes lower than that of the SR.Thus, catalytic systems designed on the basis of binuclear cyclic complexes of transition metals (palladium) will exhibit higher activation ability, as compared with the corresponding mononuclear complexes. This work was supported by the Russian Foundation for Basic Research (RFBR) (grant no. 96-03-32536a) and by INTAS–RFBR (grant no. 95-0163). References 1 W. H. Miller, J. Phys. Chem., 1983, 87, 3811. 2 (a) M. Portnoy and D. Milstein, Organometallics, 1994, 13, 600; (b) Y. Pan, J. Mague and M. J. Fink, J. Am. Chem. Soc., 1993, 115, 3842. 3 (a) D. Day, D. W. Liao and K. Balasubramanian, J. Chem. Phys., 1995, 102, 7530; (b) E. Broclavic, R. Yamauchi, A. Endou, M. Kubo and A. Miyamoto, Int.J. Quantum Chem., 1997, 61, 673; (c) S. Castillo, A. Cruz, V. Bertin, E. Poulain, J. S. Arellan and G. del Angel, Int. J. Quantum Chem., 1997, 62, 29. 4 M. R. A. Blomberg, P. E. M. Siegbahn, U. Nagashima and J.Wennerberg, J. Am. Chem. Soc., 1991, 113, 424. 5 (a) J. J. Low and W. A. Goddard, III, Organometallics, 1986, 5, 609; (b) K. Balasubramanian, B. Y. Fend and M. Z. Liao, J.Chem. Phys., 1988, 88, 6955. 6 M. J. Filatov, O. V. Gritsenko and G. M. Zhidomirov, Theor. Chim. Acta, 1987, 72, 211. 7 (a) V. M. Mamaev, I. P. Gloriozov, V. V. Simonyan, E. V. Zernova, A. V. Prisyajnyuk and Yu. A. Ustynyuk, Mendeleev Commun., 1997, 246; (b) V. M. Mamaev, I. P. Gloriozov, V. A. Khmara, V. V. Orlov and Yu. A. Ustynyuk, Dokl. Ross. Akad. Nauk, 1994, 338, 65 [Dokl.Chem. (Engl. Transl.), 1994, 338, 171]; (c) V. M. Mamaev, I. P. Gloriozov, S. Ya. Ishchsenko, V. V. Simonyan, E. M. Myshakin, A. V. Prisyajnyuk and Yu. A. Ustynyuk, J. Chem. Soc., Faraday Trans., 1995, 91, 3779. 8 M. V. Schmidt, K. K. Baldridge, J. A. Boatz, S. T. Elbert, M. S. Gordon, J. H. Jensen, S. Koseki, N. Matsunada, K. A. Nguyen, S. J. Su, T. L. Windus, M. Dupui and J. A. Montgomery, J. Comput. Chem., 1993, 14, 1347. 9 M. Krauss and W. J. Stevens, Ann. Rev. Phys. Chem., 1985, 35, 357. 10 S. Sakaki, B. Biswas and M. Sugimoto, Organometallics, 1998, 17, 1278. 11 V. M. Mamaev, I. P. Gloriozov, V. V. Simonyan, A. V. Prisyajnyuk and E. V. Zernova, Zh. Fiz. Khim., 1999, 73, 143 (in Russian). 12 R. P. Bell, The Tunnel Effect in Chemistry, Chapman and Hall, New York, 1980. 13 R. T. Skodje and D. G. Truhlar, J. Chem. Phys., 1982, 77, 5955. 14 N. Koga and K. Morokuma, Chem. Rev., 1991, 91, 823. 15 P. Fayet, A. Kaldor and D. M. Cox, J. Chem. Phys., 1990, 92, 254. P(0,E)exp(– )dE = P(0,E)exp(– )dE + E kBT ò 0 • (3) ò 0 Vmax E kBT + P(0,E)exp(– )dE + exp(– )dE E kBT ò ò E kB T Vmax 2Vmax 2Vmax • k(T) = ktotal(T)/kactivation(T) (4) Pd Pd P P C C Pd Pd P P C C Pd Pd P P C C 0.85 1.93 2.69 2.51 1.88 1.51 1.28 1.77 2.66 2.70 2.72 1.70 2.67 2.76 1.88 1.88 1.51 1.51 PC TS RP Figure 5 Structures of the stationary points of the reaction [H2P(CH2)2– PH2]Pd2 + H2 (bond lengths in Å). Received: 16th September 1999; Com. 99/1533

ISSN:0959-9436     

出版商:RSC    

年代:2000

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 1, 2000 (pp. 1–42) Synthesis, molecular and crystal structure of a tricarbonylchromium complex of 7-(2-phenyl-o-carboran-1-yl)cyclohepta-1,3,5-triene Valery N. Kalinin,*a Igor V. Shishkov,a Sergey K. Moiseev,a Pavel V. Petrovskii,a Zoya A. Starikovaa and Dae Dong Sungb a A. N. Nesmeyanov Institute of Organoelement Compounds, Russian Academy of Sciences, 117813 Moscow, Russian Federation.Fax: +7 095 135 6549; e-mail: vkalin@ineos.ac.ru b Department of Chemistry, Dong-A University, Saha-Gu, Pusan, 604-714, Korea. E-mail: ddsung@seunghak.donga.ac.kr DOI: 10.1070/MC2000v010n01ABEH001184 The reaction of substituted 1-lithio-o-carboranes with tropylium tetrafluoroborate produces 7-(o-carboran-1-yl)cyclohepta-1,3,5- trienes, which can also be obtained by decomplexation of the corresponding tricarbonylchromium complexes prepared from 1-lithio-o-carboranes and [(C7H7)Cr(CO)3]+BF4 – .Higher-order cycloaddition reactions are a promising approach to the preparation of otherwise hardly available carbocyclic compounds. 1 Cyclohepta-1,3,5-triene derivatives are very effective 6p-participants in [6p + 4p] and [6p +2p] cycloaddition reactions. 2,3 Tricarbonylchromium complexes of substituted cycloheptatrienes can also be involved in the reaction.4,5 The complexes can be prepared by reactions of a tropylium tricarbonylchromium complex with the corresponding nucleophiles.6 Their tricarbonylchromium complexes can also enter the reactions, and the cycloaddition products obtained differ in structure from those prepared from the uncomplexed starting cyclohepta-1,3,5-triene derivatives.1,7 The only cycloheptatriene derivative described to date is 7-(2-methyl-o-carboran-1-yl)cyclohepta-1,3,5-triene,† which was obtained from 1-lithio-2-methyl-o-carborane and 7-methoxycyclohepta- 1,3,5-triene.8 Here, we describe a common method for preparation of 7-(o- or m-carboran-1-yl)cyclohepta-1,3,5- triene derivatives along with their h6-tricarbonylchromium complexes.† The terms ‘o-carborane’ and ‘m-carborane’ denote 1,2- and 1,7-dicarba- closo-dodecaborane(12), respectively. We found that 1-lithio-2-R-o-carboranes 1a,b readily react with tropylium tetrafluoroborate 2 in a diethyl ether–hexane solution to form corresponding 7-(2-R-o-carboran-1-yl)cyclohepta- 1,3,5-trienes 5a,b in good yields (Scheme 1).‡ Only one regioisomer of product 5 was isolated in each case.The corresponding 1H NMR spectra (in CDCl3) show three signals attributed to three pairs of the alkenyl hydrogens and a peak of one nonalkenyl hydrogen of the cycloheptatrienyl ring. These data unambiguously indicate that the carboranyl moiety in pro-duct 5 is attached to the 7-position of cyclohepta-1,3,5- triene.That is, the carboranyl substituent occupies the allyl position regarding the carbon–carbon double bond system of the ring. This is important for cycloaddition reactions because a rather strong electron-accepting effect and the presence of a bulk carboranyl group lead to a decrease of the reactivity of the carborane-connected carbon–carbon double bonds in the cycloaddition processes.9 A reaction of lithiated carboranes 1a–d with [(C7H7)Cr- (CO)3]+BF4 – was used to prepare the tricarbonylchromium complexes of o- and m-carboranyl derivatives of cyclohepta-1,3,5- triene 4a–d.§ The chromium complexes obtained in 65–75% yields are red air-stable compounds slowly decomposing in solution. ‡ General procedure for preparation of 5a,b.To a solution of 1a or 1b (0.01 mmol) in dry diethyl ether (30 ml) a hexane solution of BunLi (0.01 mmol) was added. The mixture was stirred at room temperature for 30 min. Next, compound 2 (0.01 mmol) was added. After additional vigorous stirring for 2.5–3 h, the reaction mixture was quenched with water (20 ml). The organic layer was separated and dried with anhydrous sodium sulfate.The solvent was removed under reduced pressure, and column chromatography (silica gel, diethyl ether–petroleum ether) of the residue followed by recrystallization from hexane gave 5a (85%) or 5b (80%). § Complex 4c was isolated in 2–3% yield. C(16) C(15) C(17) C(14) C(18) C(13) C(1) H(3') H(6') C(2) B(11) B(12) B(10) B(9) B(8) B(6) B(5) B(4) B(3) C(19) H(19) C(25) C(24) C(23) C(22) C(21) C(20) Cr(1) C(27) O(27) C(26) O(26) C(28) O(28) Figure 1 General view of a molecule of 4b.Selected bond lengths (Å): Cr(1)–C(20) 2.323(8), Cr(1)–C(21) 2.215(7), Cr(1)–C(22) 2.198(8), Cr(1)– C(23) 2.212(7), Cr(1)–C(24) 2.218(7), Cr(1)–C(25) 2.295(6), C(2)–C(19) 1.550(9), C(1)–C(2) 1.700(9), C(1)–C(13) 1.522(9), C(1)–B(3) 1.680(11), C(1)–B(4) 1.679(9), C(1)–B(5) 1.695(11), C(1)–B(6) 1.737(8), C(2)–B(3) 1.723(9), C(2)–B(6) 1.711(11), C(2)–B(7) 1.741(10), C(2)–B(11) 1.708(10), Cr(1)···C(19) 2.825(8); selected bond angles (°): C(13)–C(1)–C(2) 118.8(5), C(1)–C(2)–C(19) 117.6(5), C(19)–C(20)–C(21) 127.3(7), C(19)–C(25)– C(24) 127.4(7), C(20)–C(19)–C(25) 108.8(6), Cr(1)–C(20)–C(19) 92.7(5), Cr(1)–C(25)–C19 93.9(4).H(20) BF4 RCB10H10CLi i RCB10H10CLi Cr(CO)3 BF4 RCB10H10C RCB10H10C ii Cr(CO)3 H 1 2 3 4 5 6 7 1 2 3 4 5 6 7 a R = o-Me b R = o-Ph c R = o-H d R = m-Me 1a,b 2 5a,b,d 4a–d iii, iv 1a–d 3 Scheme 1 Reagents and conditions: i, hexane–Et2O, 20 °C, 3 h; ii, hexane– Et2O, 20 °C, 10–20 min; iii, MeCN–THF (35:15); iv, (NH4)2Ce(NO3)6, H2O–MeCN (35:20).Mendeleev Communications Electronic Version, Issue 1, 2000 (pp. 1–42) An X-ray study of complex 4b¶ showed that the carboranyl moiety is attached to the cycloheptatrienyl ligand at the 7-position and has an exo orientation relative to the Cr(CO)3 group (Figure 1). 1H NMR spectra (in CDCl3) of compounds 4a–d unambiguously indicate that the carboranyl moiety is attached to the cyclohepta-1,3,5-triene ring at the 7-position. The tricarbonylchromium group can be easily removed from complexes 4a,b,d by the action of (NH4)2Ce(NO3)6 in a THF solution releasing free ligands 5a,b,d.Satisfactory analyses as well as IR and 1HNMR†† spectra were obtained for all 7-(carboran-1-yl)cyclohepta-1,3,5-triene derivatives and their h6-tricarbonylchromium complexes. ¶ Crystallographic data for 4b: C18H22B10CrO3, M = 446.46, orthorhombic crystals, spase group P212121, a = 8.292(2), b = 8.674(2), c = = 30.658(11) Å, V = 2205(1) Å3, z = 4, dcalc = 1.345 g cm–3, m(MoKa) = = 5.38 cm–1, F(000) = 912.The intensities of 3426 reflections were measured on a Siemens P3/PC diffractometer at –120 °C (lMoKa radiation, q/2q scan technique, 2q < 50°), and 3048 independent reflections were used in further calculations and refinement.The absolute conformation for the molecule of 4b was determined by calculation of the Flack parameter [k = 0.05(5)]. The structure was solved by a direct method and refined by a full-matrix least-squares technique against F2 in an anisotropic –isotropic approximation. The positions of hydrogen atoms were located from the difference Fourier syntheses. The refinement was converged to wR2 = 0.1633 and GOF = 0.974 for all 3023 independent reflections [R1 = 0.0565 was calculated against F for the 2143 independent reflections with I > 2s(I)].The number of the refined parameters is 377. All the calculations were performed using SHELXTL PLUS 5.0 on an IBM computer. Atomic coordinates, bond lengths, bond anlges and thermal parameters have been deposited at the Cambridge Crystallographic Data Centre (CCDC).For details, see ‘Notice to Authors’, Mendeleev Commun., Issue 1, 2000. Any request to the CCDC for data should quote the full literature citation and the reference number 1135/58. †† 1H NMR spectra (400 MHz, CDCl3, d/ppm). 4a: 0.80–2.95 (m, 10H, B10H10), 2.05 (s, 3H, Me), 3.83 (m, 1H, 7-H), 3.64 (m, 2H, 1-H, 6-H), 5.04 (m, 2H, 2-H, 5-H), 6.03 (m, 2H, 3-H, 4-H). 4b: 0.9–3.30 (m, 10H, B10H10), 3.10–3.50 (m, 3H, 1-H, 6-H, 7-H), 4.94 (m, 2H, 2-H, 5-H), 5.96 (m, 2H, 3-H, 4-H), 7.23–7.95 (m, 5H, Ph). 4c: 0.90–2.85 (m, 10H, B10H10), 3.16 (br. s, 1H, HCCB10H10), 3.66 (m, 2H, 1-H, 6-H), 4.02 (t, 1H, 7-H, 3J7,1 = 3J6,7 = 8.8 Hz), 5.00 (m, 2H, 2-H, 5-H), 5.95–6.05 (m, 2H, 3-H, 4-H). 4d: 1.15–3.18 (m, 10H, B10H10), 1.59 (s, 3H, Me), 3.59–3.70 (m, 3H, 1-H, 6-H, 7-H), 4.91–5.00 (m, 2H, 2-H, 5-H), 5.95 (m, 2H, 3-H, 4-H). 5a: 1.00–3.75 (m, 10H, B10H10), 1.65 (t, 1H, 7-H, 3J6,7 = 3J7,1 = 6.0 Hz), 1.80 (s, 3H, Me), 5.38 (dd, 2H, 1-H, 6-H, 3J1,7 = 3J6,7 = 6.0 Hz, 3J1,2 = = 3J5,6 = 8.8 Hz), 6.26 (m, 2H, 2-H, 5-H), 6.77 (m, 2H, 3-H, 4-H). 5b: 1.13 (t, 1H, 7-H, 3J1,7 = 3J6,7 = 6.0 Hz), 1.80–3.70 (m, 10H, B10H10), 5.32 (dd, 2H, 1-H, 6-H, 3J1,2 = 3J5,6 = 8.8Hz, 3J6,7 = 3J1,7 = 6.0 Hz), 6.04 (m, 2H, 2-H, 5-H), 6.42 (m, 2H, 3-H, 4-H, 3J3,2 = 3J4,5 = 3.2 Hz), 7.23 (t, 2H, m-HPh, J 7.6 Hz), 7.34 (t, 1H, p-HPh, J 7.6 Hz), 7.44 (t, 2H, o-HPh, J 7.6 Hz). 5d: 1.05–3.65 (m, 10H, B10H10), 1.72 (s, 3H, Me), 1.68–1.78 (m, 1H, 7-H), 5.26 (dd, 2H, 1-H, 6-H, 3J1,2 = 3J5,6 = 8.8 Hz, 3J1,7 = 3J6,7 = 6.0 Hz), 6.16 (m, 2H, 2-H, 5-H), 6.70 (m, 2H, 3-H, 4-H).This work was supported by the Russian Foundation for Basic Research (grant nos. 97-03-33783a and 99-03-32899) and by the Scientific Training Centres on the Chemistry of Organometallic Compounds and Biomedical Chemistry (grant nos. 234 and K0599, the ‘Integratsiya’ Special Federal Program). References 1 J. H. Rigby, S. D. Rege, V. P. Sandanayaka and M. Kirova, J. Org. Chem., 1996, 61, 842. 2 K. N. Houk and R. B. Woodward, J. Am. Chem. Soc., 1970, 92, 4143. 3 J.H.Rigby, Org. React., 1997, 49, 331. 4 J. H. Rigby, N. M. Niyaz, K. Short and M. Heeg, J. Org. Chem., 1995, 60, 7720. 5 K. Chaffee, J. B. Sheridan and A. Aistars, Organometallics, 1992, 11, 18. 6 W. R. Roth and W. Grimme, Tetrahedron Lett., 1966, 2347. 7 J. H. Rigby, in Advances in Metal-Organic Chemistry, ed. L. S. Liebeskind, JAI Press Inc., Greenwich, 1995, vol. 4, p. 89. 8 K. M. Harmon, A. B. Harmon and B. C. Thompson, J. Am. Chem. Soc., 1967, 89, 5309. 9 L. I. Zakharkin and V. N. Kalinin, Izv. Akad. Nauk SSSR, Ser. Khim., 1967, 937 (Bull. Acad. Sci. USSR, Div. Chem. Sci., 1967, 16, 908). Received: 5th July 1999; Com. 99/1512

ISSN:0959-9436     

出版商:RSC    

年代:2000

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 1, 2000 (pp. 1–42) Spontaneous resolution of new conglomerates in the series of 4-arenesulfonyliminocyclohex-2-en-1-ones Remir G. Kostyanovsky,*a Anatolii P. Avdeenko,b Svetlana A. Konovalova,b Gulnara K. Kadorkinaa and Alexander V. Prosyanikc a N. N. Semenov Institute of Chemical Physics, Russian Academy of Sciences, 117977 Moscow, Russian Federation.Fax: +7 095 137 3227; e-mail: kost@center.chph.ras.ru b Donbass State Machinery Academy, 343913 Kramatorsk, Ukraine. Fax: +38 0626 416 676; e-mail: dgma@dgma.donetsk.ua c Ukrainian State Chemical Technology University, 320005 Dnepropetrovsk, Ukraine. Fax: +38 0562 477 478; e-mail: ugxtu@dicht.dnepropetrovsk.ua DOI: 10.1070/MC2000v010n01ABEH001208 Racemic mixtures 1a–f, 2a–e crystallise as conglomerates at room temperature and lead to spontaneous resolution; tosylimines 1e and 2c give homochiral crystals (space group P1), whereas similar benzenesulfonyloximes 3a,b give heterochiral packings (space groups P21/n and P21/c, respectively).Conglomerate formation is a necessary condition both for spontaneous resolution of enantiomers and for resolution by crystallization from optically active solvents or by an entrainment procedure.1 Conglomerate formation is of poor occurrence; up to 1979 only 250 conglomerates were reported, according to nowaday estimations the frequency of organic conglomerates does not exceed 10%,1 so that a search for conglomerates comprises an essential challenge.Some previous studies2 have shown that this proportion can fluctuate to a large extent in some particular series of organic compounds. Earlier we have found conglomerates among various classes of organic compounds using X-ray data,3,4 resolution by crystallization from optically active solvents,5,6 and engineering homochiral crystals.7,8 In this work, another intriguing instance of conglomerate formation has been found in the rather wide series of derivatives of 4-arenesulfonyliminocyclohex-2-en-1-ones 1, 2.Compounds of this class have been synthesised recently by the halogenation of corresponding N-arenesulfonyl-p-quinone imines or 4-arenesulfonylaminophenols. 9 Similar 4-arenesulfonyloximinocyclohex- 2-en-1-ones such as 3 have been obtained by the chlorination of corresponding O-arenesulfonyl-p-quinone oximes.10–12 We optimised the above methodology and thus rised the yields by 11–25% up to 75–86% (cf.ref. 12), improved the purity of 1b,c,f, 2b, 3a (the analytically pure products were obtained after single crystallizations) and synthesised 1a,c,d, 2a,d and 3b for the first time. All the products were characterised by spectroscopic data† and elemental analysis; the structures of 1e, 2c and 3a,b were also confirmed by X-ray diffraction analysis (the C(17a) C(13a) C(14a) C(15a) C(12a) C(16a) C(11a) O(3a) S(1a) O(2a) N(1a) C(4a) C(3a) Cl(1a) C(2a) Cl(2a) C(10a) C(5a) C(6a) C(7a) C(8a) C(9a) C(1a) O(1a) Figure 1 Crystal structure of 2c. Above: two independent molecules of 2c (a and b) and a molecule of the entrained CCl4, in the molecule b a statistical disorder on two positions of the C(2b)C(3b) fragment is observed; below: the molecular structure of an independent molecule a of 2c.Selected bond lengths (Å): S(1a)–N(1a) 1.672(4), Cl(1a)–C(2a) 1.800(6), Cl(2a)–C(3a) 1.843(7), O(1a)–C(1a) 1.205(7), N(1a)–C(4a) 1.287(7), C(1a)–C(9a) 1.478(8), C(1a)–C(2a) 1.517(8), C(2a)–C(3a) 1.462(9), C(3a)–C(4a) 1.513(8), C(4a)– C(10a) 1.482(7), C(9a)–C(10a) 1.390(8); selected dihedral angles (°): S(1a)–N(1a)–C(4a)–C(10a) 179.9(4), S(1a)–N(1a)–C(4a)–C(3a) 3.0(8), C(3a)–C(4a)–C(10a)–C(9a) 17.7(7), C(2a)–C(3a)–C(4a)–C(10a) 45.2(7), Cl(1a)–C(2a)–C(3a)–Cl(2a) 177.9(3).(a) (b) (a) aOptical rotation was measured on a Polamat A polarimeter. bIn MeOH. cIn Me2CO. dOptical rotation remained unchanged after 1 week at 20 °C.eIn EtOH. f2c (100 mg) was powdered, treated under a vacuum (1 Torr, 2 h) in order to remove the entrained CCl4, and then crystallised from a mixture of the solvents with the addition of deficiency of CCl4 (the ratio 2c:CCl4 = = 4:1). Table 1 Conditions of crystallization and optical activity of the obtained compounds. Compound Solvent for the crystal growth Number of crystals Weight of each crystal/ mg Optical activity of each crystala [a]20 546/° (c, CHCl3) 1a CHCl3– n-heptane (1:1) 43 3.6–13.2 9.1–29.1 +2.7–9.3 (0.6–1.1) –4.0–7.0 (0.8–2.4) 1b CHCl3 21 10.7–30.4 11.1 +2.4–5.6 (0.9–2.5) –3.2 (0.9) 1c CHCl3– n-heptane (3:2) 11 26.6 6.3 +0.8 2.2) –3.8 (0.5) 1d CHCl3– n-heptane (1:1) 12 12.4 11.5–29.0 +4.8 (1.0) –2.9–3.1 (1.0–2.4) 1e CHCl3 313 2.7–12.2 12.2 7.5–63.4 +2.1–7.8 (0.2–0.9) +2.1 (0.9)b –2.8–6.4 (0.5–5.3) 1f CHCl3– n-heptane (3:2) 111 28.2 15.2 56.1 +12.3 (2.4) +2.4 (1.3)c –7.2 (4.7) 2a CHCl3– n-heptane (1:2) 21 20.5–26.4 14.0 +2.3–2.7 (1.7–2.2) –2.6 (1.2) 2b CHCl3– CCl4 (1:1) 13 15.9 11.1–14.2 +8.7 (1.3) –7.2–9.6 (0.9–1.2)d 2c CCl4 2211 11.5–12.6 1.3–2.5 29.0 53.5 +1.9–2.1 (1.0) +8.8–21.2 (0.1–0.2)c +0.8 (2.4)e –2.2 (4.4) 2d CHCl3– n-heptane (1:1) 11 12.6 11.5 +1.0 (1.1) –1.1 (1.0) 2e CHCl3– n-heptane (2:1) 21 3.3–13.1 7.5 +3.2–5.5 (0.3–1.1) –5.1 (1.0) 2cf Me2CO– n-hexane (1:5) the entire precipitate 45.2 –1.3 (3.8)Mendeleev Communications Electronic Version, Issue 1, 2000 (pp. 1–42) data for 1e and 3a,b will be published later). Compounds 1 and 2 both in solution and in crystals exist solely in the form of E-isomers relative to the double cyclohexene bond.For compound 3 in solution Z- and E-isomers are observed, whereas only the E-isomer is detected in a crystal (cf. refs. 10–12). All the above compounds are stable and give well-formed, rather large-sized, transparent crystals. By testing the optical activity of individual crystals, (+)- and (–)-enantiomers of compounds 1 and 2 were isolated.This results in the identification of 11 new conglomerates (Table 1). Consistently, X-ray diffraction analysis performed on 1e and 2c demonstrated that single crystals contain homochiral molecules (space groups P1) (Figure 1).‡ By contrast, structurally related O-benzenesulfonyloximes 3a,b form centrosymmetric crystals and thus do not lead to any spontaneous resolution.† Characteristics and spectroscopic data. 1H and 13C NMR spectra were measured at 300 and 75 MHz, respectively, in CDCl3. 1a: yield 72%, mp 124–125 °C (AcOH). 1H NMR, d: 2.25 (s, 3H, 3-Me), 6.64 (s, 1H, 5-H), 7.62, 7.72 and 8.05 (m, 5H, Ph). IR, n/cm–1: 1725 (C=O), 1610, 1584 (C=N, C=C), 1338, 1170 (SO2). 1b:9 yield 78%, mp 137–138 °C (AcOH). 1H NMR, d: 6.69 (s, 1H, 5-H), 7.63, 7.74 and 8.07 (m, 5H, Ph). 13C NMR {1H}, d: 58.2 [C(5)], 81.2 [C(6)], 127.8, 129.4, 134.5 and 138.2 (Ph), 138.6, 143.8 [C(2), C(3)], 160.3 [C(4)], 173.4 [C(1)]. 1c: yield 86%, mp 155–156 °C (AcOH). 1H NMR, d: 2.19 (s, 3H, 6-Me), 2.28 (s, 3H, 3-Me), 6.48 (s, 1H, 5-H), 7.60, 7.70 and 8.05 (m, 5H, Ph). 13C NMR {1H}, d: 20.12 (3-Me), 25.48 (6-Me), 46.59 [C(5)], 55.58 [C(6)], 127.55, 129.18, 133.91 and 139.32 (Ph), 133.95 [C(2)], 148.22 [C(3)], 167.10 [C(4)], 181.50 [C(1)].IR, n/cm–1: 1710 (C=O), 1600 (C=N, C=C), 1320, 1169 (SO2). 1d: yield 62%, mp 142–143 °C (AcOH). 1H NMR, d: 2.24 (s, 3H, 3-Me), 2.50 (s, 3H, Me), 6.66 (s, 1H, 5-H), 7.40 and 7.92 (dd, 4H, C6H4, 3J 8.4 Hz). 13C NMR {1H}, d: 17.02 (3-Me), 21.71 (Me), 58.64 [C(5)], 81.63 [C(6)], 127.82, 129.93, 145.52, and 146.46 (C6H4), 135.83 [C(2)], 145.52 [C(3)], 164.67 [C(4)], 174.83 [C(1)].IR, n/cm–1: 1723 (C=O), 1595, 1570 (C=N, C=C), 1339, 1161 (SO2). 1e:9 yield 77%, mp 141–142 °C (AcOH). 1HNMR, d: 2.49 (s, 3H, Me), 6.71 (s, 1H, 5-H), 7.40 and 7.97 (dd, 4H, C6H4, 3J 8.3 Hz). 13C NMR {1H}, d: 21.77 (Me), 52.8 [C(5)], 81.40 [C(6)], 128.03, 130.05, 135.46, and 145.89 (C6H4), 138.70 and 143.96 [C(2), C(3)], 160.02 [C(4)], 173.52 [C(1)].IR, n/cm–1: 1724 (C=O), 1610, 1555 (C=C, C=N), 1346, 1168 (SO2). 1f:9 yield 75%, mp 141–142 °C (AcOH). 1HNMR, d: 6.63 (s, 1H, 5-H), 7.60 and 8.00 (dd, C6H4, 3J 8.7 Hz). IR, n/cm–1: 1723 (C=O), 1600, 1553 (C=C, C=N), 1350, 1170 (SO2). 2a: yield 81%, mp 130–131 °C (CCl4). 1H NMR, d: 4.76 (d, 1H, 2-H, 3J 3.9 Hz), 6.55 (d, 1H, 3-H, 3J 3.9 Hz), 7.61, 7.68 and 8.13 (m, 5H, Ph), 7.61 and 8.12 (m, 4H, 5,6,7,8-H). 2b:9 yield 82%, mp 141–142 °C (AcOH). 1HNMR, d: 6.83 (s, 1H, 3-H), 7.61, 7.70 and 8.10 (m, 5H, Ph), 7.78 and 8.14 (m, 4H, 5,6,7,8-H). IR, n/cm–1: 1719 (C=O), 1612, 1583 (C=C, C=N), 1330, 1161 (SO2). 2c:9 yield 84%, mp 136–137 °C (CCl4). 1H NMR, d: 2.48 (s, 3H, Me), 4.75 (d, 1H, 2-H, 3J 3.6 Hz), 6.57 (d, 1H, 3-H, 3J 3.6 Hz), 7.40 and 7.99 (dd, 4H, C6H4, 3J 8.1 Hz), 7.75 and 8.13 (mm, 4H, 5,6,7,8-H).IR, n/cm–1: 1705 (C=O), 1612, 1587 (C=N, C=C), 1330, 1165 (SO2). 2d: yield 86%, mp 138–139 °C (CCl4). 1H NMR, d: 4.77 (d, 1H, 2-H, 3J 3.3 Hz), 6.51 (d, 1H, 3-H, 3J 3.6 Hz), 7.58 and 8.04 (m, 4H, C6H4, 3J 8.7 Hz), 7.77 and 8.12 (m, 4H, 5,6,7,8-H). 2e:9 yield 80%, mp 167–168 °C (AcOH). 1HNMR, d: 6.78 (s, 1H, 3-H), 7.60 and 8.05 (dd, 4H, C6H4, 3J 8.7 Hz), 7.80 and 8.17 (m, 4H, 5,6,7,8-H).IR, n/cm–1: 1725 (C=O), 1620, 1589 (C=C, C=N), 1343, 1170 (SO2). 3a:12 yield 78%, mp 110 °C (AcOH); the ratio of isomers E/Z = 2.1. E-isomer: 1HNMR, d: 1.85 (s, 3H, 6-Me), 2.02 (d, 3H, 2-Me, 4J 1.5 Hz), 5.47 (d, 1H, 3-H, 4J 1.5 Hz), 6.76 (s, 1H, 5-H), 7.58, 7.71 and 8.01 (m, 5H, Ph). 13C NMR {1H}, d: 16.74 (6-Me), 22.33 (2-Me), 53.18 [C(5)], 63.35 [C(6)], 122.67, 129.05, 129.25 and 130.03 (Ph), 134.56 and 140.54 [C(2), C(3)], 157.24 [C(4)], 188.80 [C(1)].Z-isomer: 1.86 (s, 3H, 6-Me), 2.07 (d, 3H, 2-Me, 4J 1.5 Hz), 4.89 (d, 1H, 3-H, 4J 1.5 Hz), 7.27 (s, 1H, 5-H), 7.58, 7.71 and 8.01 (m, 5H, Ph). 13C NMR {1H}, d: 16.98 (6-Me), 22.75 (2-Me), 61.71 [C(5)], 65.10 [C(6)], 122.67, 129.00, 129.24 and 130.03 (Ph), 134.44 and 141.74 [C(2), C(3)], 155.75 [C(4)], 188.60 [C(1)].IR, n/cm–1: 1703 (C=O), 1630, 1590 (C=N, C=C), 1394, 1209 (SO2). 3b: yield 73%, mp 143 °C (AcOH), the ratio of isomers in CDCl3 E/Z = 4.0. E-isomer: 1H NMR, d: 1.90 (s, 3H, 6-Me), 5.49 (s, 1H, 3-H), 7.15 (s, 1H, 5-H), 7.59, 7.72 and 8.00 (m, 5H, Ph). Z-isomer: 1.90 (s, 3H, 6-Me), 5.48 (s, 1H, 3-H), 7.14 (s, 1H, 5-H), 7.59, 7.72 and 8.00 (m, 5H, Ph).Compound 2c crystallises as a CCl4 solvate with the 2:1 stoichiometry. This achiral solvent molecule could favour the crystallization in non-centrosymmetric space group and thus acts as a conglomerator.13 Therefore, a predetermined optical enrichment of compound 2c can be performed by crystallization in the presence of a half-mole quantity of CCl4.In this case, the entire precipitate possessed optical activity whereas upon crystallization from CCl4 only individual crystals were optically active, but the entire precipitate was a racemic conglomerate. This work was supported by the Russian Foundation for Basic Research (grant no. 97-03-33021) and INTAS (grant no. 157). References 1 J.Jacques, A. Collet and S. H. Wilen, Enantiomers, Racemates, and Resolutions, Krieger Publ. Co, Malabar, Florida, 1994. 2 (a) A. Collet, Enantiomer, 1999, 4, 157; (b) A. Collet, in Comprehensive Supramolecular Chemistry, ed. D. N. Reinhoudt, Pergamon Press, Oxford, 1996, vol. 10, ch. 5, p. 113; (c) G. Coquerel, M. N. Petit and F. Robert, Acta Crystallogr., 1993, C49, 824. 3 A. B. Zolotoi, Yu.I. El’natanov, I. I. Chervin, S. V. Konovalikhin, O. A. Dyachenko, L. O. Atovmyan and R. G. Kostyanovsky, Khim. Geterotsikl. Soedin., 1988, 909 [Chem. Heterocycl. Compd. (Engl. Transl.), 1988, 746]. 4 R. G. Kostyanovsky, Yu. I. El’natanov, I. I. Chervin, S. V. Konovalikhin, A. B. Zolotoi and L. O. Atovmyan, Izv. Akad. Nauk, Ser. Khim., 1996, 1796 (Russ. Chem. Bull., 1996, 45, 1707). 5 R.G. Kostyanovsky, V. F. Rudchenko and G. V. Shustov, Izv. Akad. Nauk SSSR, Ser. Khim., 1977, 1687 (Bull. Acad. Sci. USSR, Div. Chem. Sci., 1977, 26, 1560). 6 V. F. Rudchenko, O. A. Dyachenko, A. B. Zolotoi, L. O. Atovmyan, I. I. Chervin and R. G. Kostyanovsky, Tetrahedron, 1982, 38, 961. 7 R. G. Kostyanovsky, K. A. Lyssenko, G. K. Kadorkina, O. V. Lebedev, A. N. Kravchenko, I. I.Chervin and V. R. Kostyanovsky, Mendeleev Commun., 1998, 231. 8 R. G. Kostyanovsky, K. A. Lyssenko, Yu. I. El’natanov, O. N. Krutius, I. A. Bronzova, Yu. A. Strelenko and V. R. Kostyanovsky, Mendeleev Commun., 1999, 106. ‡ Crystallographic data for 2c at 20 °C: (C17H13NO3SCl2)2·CCl4, triclinic, crystal size 0.36×0.39×0.48 mm, space group P1, a = 12.959(5) Å, b = = 13.271(6) Å, c = 13.951(7) Å, a = 63.03(4)°, b = 77.56(4)°, g = 71.69(5)°, V = 2023 Å3, Z = 2, dcalc = 1.508 g cm–3, m(MoKa) = 0.706 mm–1, F(000) = = 932.The intensities of 7581 reflections were measured on an Enraf- Nonius CAD-4 diffractometer at 20 °C (lMoKa radiation, q/2q scan, 12° < q < 23°), and 6075 independent reflections were used in further calculations and refinement.The structure was solved by a direct method and refined using the full-matrix least-squares method against F2 in the anisotropic–isotropic approximation. The refinement is converged to wR2 = = 0.2148 and GOF = 0.988 for all independent reflections [R1 = 0.0789 is calculated against F for 6044 observed reflections with I > 2s(I)]. The number of refined parameters is 514.All the calculations were performed using the SHELXS and SHELXL 93 programs. Atomic coordinates, bond lengths, bond angles and thermal parameters have been deposited at the Cambridge Crystallographic Data Centre (CCDC). For details, see ‘Notice to Authors’, Mendeleev Commun., 2000, Issue 1. Any request to the CCDC for data should quote the full literature citation and the reference number 1135/60.N O H Y Y R'' R' Y O2S X N O H Cl Cl R O2S X N O H Cl Cl Me H R O2SO a X = H, R' = Me, R'' = Y = Cl b X = H, R' = R'' = Y = Cl c X = H, R' = R'' = Me, Y = Br d X = R' = Me, R'' = Y = Cl e X =Me, R' = R'' = Y = Cl f X = Cl, R' = R'' = Y = Cl 1a–f a X = H, R = H b X = H, R = Cl c X =Me, R = H d X = Cl, R = H e X = Cl, R = Cl 2a–e a R = Me b R = Cl 3a,bMendeleev Communications Electronic Version, Issue 1, 2000 (pp. 1–42) 9 A. P. Avdeenko and S. A. Zhukova, Zh. Org. Khim., 1999, 35, 412 (Russ. J. Org. Chem., 1999, 35, 388). 10 (a) A. P. Avdeenko, N. M. Glinanaya and V. V. Pirozhenko, Zh. Org. Khim., 1993, 29, 1402 (Russ. J. Org. Chem., 1993, 29, 1164); (b) A. P. Avdeenko, N. M. Glinanaya and V. V. Pirozhenko, Zh. Org. Khim., 1995, 31, 1523 (Russ. J. Org. Chem., 1995, 31, 1380). 11 A. P. Avdeenko and N. M. Glinanaya, Zh. Org. Khim., 1995, 31, 1679 (Russ. J. Org. Chem., 1995, 31, 1507). 12 A. P. Avdeenko, S. A. Zhukova and N. M. Glinanaya, Zh. Org. Khim., 1999, 35, 586 (Russ. J. Org. Chem., 1999, 35, 560). 13 R. G. Kostyanovsky, Proceedings of the Interdisciplinary Symposium on Biological Homochirality, Serramazzoni (Modena), Italy, 1998, OP 20, p. 41. Received: 22nd September 1999; Com. 99/1536

ISSN:0959-9436     

出版商:RSC    

年代:2000

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 1, 2000 (pp. 1–42) A new ring transformation in the series of 1,2,3-thiadiazoles. Synthesis of 5H-[1,2,3]triazolo[5,1-b][1,3,4]thiadiazines Yury Yu. Morzherin,a Tatiana V. Glukhareva,a Irina N. Slepukhina,a Vladimir S. Mokrushin,a Alexey V. Tkachevb and Vasiliy A. Bakulev*a a Department of Technology of Organic Synthesis, Urals State Technical University, 620002 Ekaterinburg, Russian Federation.Fax: +7 3432 745483; e-mail: morjerine@htf.ustu.ru b N. N. Vorozhtsov Novosibirsk Institute of Organic Chemistry, Siberian Branch of the Russian Academy of Sciences, 630090 Novosibirsk, Russian Federation. Fax: +7 3832 34 4752; e-mail: atkachev@nioch.nsc.ru DOI: 10.1070/MC2000v010n01ABEH001218 The first example of the ring transformation of 1,2,3-thiadiazoles involving four atoms of the side chain to form 5H-[1,2,3]triazolo- [5,1-b][1,3,4]thiadiazines is presented.Several types of ring transformation reactions and rearrangements of 1,2,3-thiadiazoles leading to various heterocyclic compounds have been discovered.1 These processes are governed by the following factors: (i) the facile cleavage of the weak N–S bond, (ii) the existence of an equilibrium between 1,2,3-thiadiazoles and a-diazo thiocarbonyl compounds and (iii) the capacity of both thiocarbonyl and diazo groups to cyclise onto electrophilic and nucleophilic functionalities.It was shown that 1,2,3-thiadiazoles could be transformed with involvement of one (Dimroth type rearrangement),2 two (Conforth type)3 or three (L’abbé type)4 atoms of the side chain.This paper presents the first example of the ring transformation of 1,2,3-thiadiazoles with the participation of four atoms of the side chain. Starting compounds 2a–d for this novel ring transformation were obtained by one pot synthesis from 5-N-nitrosylamino- 1,2,3-thiadiazole5 1 (Scheme 1).† We found that compounds 2 are transformed to ethyl 6-aryl-5-chloro-5H-[1,2,3]triazolo[5,1-b]- [1,3,4]thiadiazin-3-carboxylates 4a–d in moderate yields by a treatment with thionyl chloride at room temperature for 1 h.The structures of products 4a–d were assigned on the basis of elemental analyses, IR, mass and NMR spectra.‡ We have also found that this reaction being carried out at –80 °C for 30 min leads to intermediate compounds 3a–d that could be transformed further into 4a–d under similar conditions.The structures of compounds 3a–d obtained as triazolothiazines were confirmed by 1H NMR and mass spectrometry.§ The fact that the melting points and NMR spectra of 3a,b were found to be identical to those of compounds obtained earlier by the reaction of ethyl 1-amino-5-mercapto-1,2,3-thiadiazol-4-carboxylate with bromoacetophenones6 also confirmed the structures of 3a–d.The rearrangement probably involves a Dimroth rearrangement and chlorination by thionyl chloride. The order in which these processes occur is not clear and will be the subject of a further study. † The 1H and 13C NMR spectra were recorded in [2H6]DMSO solutions with a Bruker DRX-500 instrument (500 MHz for 1H and 125 MHz for 13C), and the IR spectra were recorded in KBr using a UR-20 spectrometer.Synthesis of 2. A suspension of N-nitrosoamine 1 (4 g, 0.02 mol) in 200 ml of 1 M HCl was treated with SnCl2 (4.7 g, 0.025 mol) at 10–15 °C. After stirring for 3 h, the reaction mixture was filtered. To the filtrate 0.02 mol of a ketone and 0.1 g of Et4NCl were added, and the mixture was stirred at room temperature for 12 h.The precipitate of 2 was filtered off and recrystallised from ethanol. For 2a: yield 62%, mp 120–122 °C. 1H NMR, d: 10.45 (s, 1H, NH), 7.79–7.83 (m, 2H, ArH), 7.44–7.48 (m, 3H, ArH), 4.44 (q, 2H, OCH2, J 7.3 Hz), 2.42 (s, 3H, Me), 1.40 (t, 3H, Me, J 7.3 Hz). For 2b: yield 57%, mp 190–192 °C. 1H NMR, d: 10.47 (s, 1H, NH), 7.81 (d, 2H, ArH, J 8.85 Hz), 7.52 (d, 2H, ArH, J 8.85 Hz), 4.43 (q, 2H, OCH2, J 7.0 Hz), 2.40 (s, 3H, Me), 1.40 (t, 3H, Me, J 7.3 Hz).For 2c: yield 65%, mp 144–146 °C. 1H NMR, d: 10.41 (s, 1H, NH), 7.70 (d, 2H, ArH, J 8.24 Hz), 7.27 (d, 2H, ArH, J 8.24 Hz), 4.43 (q, 2H, OCH2, J 7.0 Hz), 2.38 (s, 3H, Me), 1.39 (t, 3H, Me, J 7.0 Hz). For 2d: yield 49%, mp 205–206 °C. 1H NMR, d: 10.6 (s, 1H, NH), 8.54–8.55 (m, 1H, ArH), 8.20–8.28 (m, 2H, ArH), 7.69–7.76 (m, 1H, ArH), 4.46 (q, 2H, OCH2, J 7.3 Hz), 2.50 (s, 3H, Me), 1.44 (t, 3H, Me, J 7.3 Hz).‡ Synthesis of 4. A suspension of hydrazone 1 (0.02 mol) in 50 ml of SOCl2 was stirred for 2 h at room temperature, and the excess of SOCl2 was removed at a reduced pressure. The product was recrystallised from ethanol. For 4a: yield 45%, mp 150–152 °C. 1H NMR, d: 8.05–8.15 (m, 2H, ArH), 7.62–7.74 (m, 3H, ArH), 7.57 (1H, s, CHCl), 4.39 (q, 2H, OCH2, J 7.3 Hz), 1.36 (t, 3H, Me, J 7.3 Hz). MS, m/z: 324 (9%, M + 2), 322 (20%, M+). For 4b: yield 55%, mp 196–198 °C. 1H NMR, d: 8.14 (d, 2H, ArH, J 10.0 Hz), 7.65 (d, 2H, ArH, J 10 Hz), 7.50 (1H, s, CHCl), 4.41 (q, 2H, OCH2, J 7.5 Hz), 1.42 (t, 3H, Me, J 7.5 Hz). 13C NMR, d: 159.9 (CO), 149.0 (C3a), 139.7 (CArCl), 134.8 (C3), 129.8 (CArH), 129.0 (CArH), 128.9 (CAr), 122.2 (C6), 61.8 (OCH2), 46.2 (C5), 14.1 (Me).MS, m/z: 359 (3.5%, M + 2), 357 (8.3%, M+). For 4c: yield 65%, mp 162–164 °C. 1H NMR, d: 8.03 (d, 2H, ArH, J 8.24 Hz), 7.44 (d, 2H, ArH, J 8.24 Hz), 7.46 (1H, s, CHCl), 4.40 (q, 2H, OCH2, J 7.3 Hz), 1.42 (t, 3H, Me, J 7.3 Hz). MS, m/z: 338 (4.5%, M + 2), 336 (8.3%, M+).For 4d: yield 48%, mp 215–216 °C. 1H NMR, d: 8.91–8.93 (m, 1H, ArH), 8.44–8.58 (m, 2H, ArH), 7.91–7.98 (m, 1H, ArH), 7.74 (s, 1H, CHCl), 4.44 (q, 2H, OCH2, J 10.0 Hz), 1.43 (t, 3H, Me, J 10.0 Hz). MS, m/z: 369 (1.5%, M + 2), 367 (2.3%, M+). § Synthesis of 3. A suspension of hydrazone 1 (0.02 mol) in 50 ml of SOCl2 was stirred at –80 °C for 30 min, and the excess of SOCl2 was removed at a reduced pressure.The product was recrystallised from ethanol. For 3a: yield 35%, mp 183–185 °C (lit.,6 185 °C). 1H NMR (CDCl3) d: 7.95–8.15 (m, 2H, ArH), 7.46–7.54 (m, 3H, ArH), 4.40 (q, 2H, OCH2, J 7.0 Hz), 3.95 (s, 2H, CH2), 1.45 (t, 3H, Me, J 7.0 Hz). MS, m/z: 288 (8%, M). For 3b: yield 23%, mp 215–216 °C (lit.,6 216 °C). 1H NMR (CDCl3) d: 8.05 (d, 2H, ArH), 7.52 (d, 2H, ArH), 4.45 (q, 2H, OCH2, J 7.3 Hz), 3.95 (s, 2H, CH2), 1.40 (t, 3H, Me, J 7.3 Hz).MS, m/z: 324 (9%, M + 2), 322 (20%, M+). For 3c: (mixture with 4c) 1H NMR, d: 7.98 (d, 2H, ArH, J 7.9 Hz), 7.36 (d, 2H, ArH, J 7.9 Hz), 4.36 (q, 2H, OCH2, J 7.3 Hz), 4.26 (s, 2H, CH2), 2.26 (s, 3H, Me), 1.31 (t, 3H, Me, J 7.3 Hz). MS, m/z: 302 (19%, M). For 3d: (mixture with 4d) 1H NMR, d: 8.87 (dd, 1H, ArH), 8.55 (dd, 1H, ArH), 8.47 (dd, 1H, ArH), 7.87 (dd, 1H, ArH), 4.37 (q, 2H, OCH2, J 7.0 Hz), 4.41 (s, 2H, CH2), 1.40 (t, 3H, Me, J 7.0 Hz).MS, m/z: 333 (10%, M). N N S COOEt NH NO 1 N N S COOEt NH NH2 i ii N N S COOEt NH N 2a–d Ar Me N S N N N COOEt Ar H H N S N N N COOEt Ar Cl H iii iv 3a–d 4a–d a Ar = Ph b Ar = 4-ClC6H4 c Ar = 4-MeC6H4 d Ar = 3-NO2C6H4 Scheme 1 Reagents and condition: i, SnCl2, 1 M HCl, 3 h, room temperature; ii, ArCOMe, 1 M HCl, Et4NCl, 10 h, room temperature; iii, SOCl2, –80 °C, 30 min; iv, SOCl2, room temperature, 2 h.Mendeleev Communications Electronic Version, Issue 1, 2000 (pp. 1–42) Thus, we have found the first example of the ring transformation of 1,2,3-thiadiazole where four atoms of the side chain take part in the reaction to afford 5H-[1,2,3]triazolo[5,1-b]- [1,3,4]thiadiazine.This work was supported by the Russian Foundation for Basic Research (grant no. 98-03-33045a). References 1 G. L’abbé, B. D’hooge and W. Dehaen, Molecules, 1996, 64, 190. 2 (a) G. L’abbé, J. Heterocycl. Chem., 1984, 21, 627; (b) Yu. M. Shafran, V. A. Bakulev, V. A. Shevyrin and M. Yu. Kolobov, Khim. Geterotsikl. Soedin., 1993, 840 [Chem. Heterocycl. Compd. (Engl. Transl.), 1993, 724]; (c) T. Kindt-Larsen and C. Pedersen, Acta Chem. Scand., 1962, 16, 1800. 3 (a) Yu. Yu. Morzherin, V. A. Bakulev, E. F. Dankova and V. S. Mokrushin, Khim. Geterotsikl. Soedin., 1994, 4, 548 [Chem. Heterocycl. Compd. (Engl. Transl.), 1994, 438]; (b) G. L’abbé, E. Vanderstede, W. Dehaen and S. J. Toppet, J. Chem. Soc., Perkin Trans. 1, 1993, 1719. 4 (a) G. L’abbé, Bull. Soc. Chim. Belg., 1990, 99, 281; (b) V. A. Bakulev, E. V. Tarasov, Yu. Yu. Morzherin, I. Luyten, S. Toppet and W. Dehaen, Tetrahedron, 1998, 54, 8501. 5 J. Goerdeler and G. Gnad, Chem. Ber., 1966, 99, 1618. 6 G. L’abbé and E. Vanderstede, J. Heterocycl. Chem., 1989, 1811. Received: 1st November 1999; Com. 99/1546

ISSN:0959-9436     

出版商:RSC    

年代:2000

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 1, 2000 (pp. 1–42) Photoinduced activation of arene C–H bonds with ( 5-cyclopentadienyl)- trimethylplatinum(IV): a possible role of CpPtR2H Andrei N. Vedernikov,* Sergei V. Borisoglebski, Alexei B. Solomonov and Boris N. Solomonov Department of Chemistry, Kazan State University, 420008 Kazan, Russian Federation. Fax: +7 8432 38 0994; e-mail: ave@ksu.ru DOI: 10.1070/MC2000v010n01ABEH001168 (h5-Cyclopentadienyl)trimethylplatinum(IV) reacts with ArMe arenes rather than alkanes or benzene under heating to 150 °C in the dark or under irradiation with light at room temperature by a radical mechanism to produce (h5-cyclopentadienyl)dimethyl- (arylmethyl)platinum(IV) complexes, and the expected complex CpPtR2(H) is unstable under the reaction conditions, as found by the reduction of CpPtMe2I with complex metal hydrides.According to recent quantum-chemical calculations,1,2 the 16- electron cyclopentadienyl complexes of platinum(II) [CpPt(CO)]+ and [CpPtCl] can undergo exothermic oxidative addition with methane to produce methyl hydrido complexes of cyclopentadienylplatinum( IV), which can be considered to be similar to platinum(IV) derivatives with fac-chelating ligands.3 It has been expected that the CpPtX species are highly reactive short-lived intermediates, which can be generated by decomposition of corresponding more stable 18-electron precursors.The well-known methods for generating coordinatively unsaturated 16-electron complexes CpMX (M = Rh or Ir) involve photolysis of the dihydrides CpMH2(L) (M = Ir) or dicarbonyls CpM(CO)2 or thermolysis of the hydrido complexes CpM(L)(R)H (refs. 4, 5 and 6, respectively). In this work, we attempted to obtain the platinumcontaining species via elimination of ligands from cyclopentadienyltrimethylplatinum( IV), CpPtMe3. Complexes of the general formula CpPtR3 (R3 = Alk3, Alk2R' etc.) are well known for R' = alkyl or acyl7 rather than for R' = H.Irradiation of solutions containing 3–5 mg of CpPtMe3 and 0.5 ml of a hydrocarbon (benzene, toluene, p-xylene, mesitylene, n-pentane or cyclohexane) was performed with a 10 W highpressure mercury lamp or a 250 W halogen lamp for 0.5–3 h using a water filter to absorb the IR component of light. The reactants were placed in sealed tubes under argon or in a vacuum.A longer time of irradiation leaded to decomposition of the starting complex and to deposition of platinum metal and platinum containing white polymer of the formula PtC8H14. The heating of the same solutions at 150 °C in the dark also leaded to decomposition of the complex. According to 1H NMR spectroscopy data, both photoinduced and thermally induced decomposition of CpPtMe3 in alkylarene solutions resulted in new cyclopentadienylplatinum( IV) derivatives, which were isolated from the reaction mixture by filtering off platinum and evaporating the solvent.Photolysis was found to be a more effective tool: the product yield was as high as 10% on a platinum basis (cf. 1–3% obtained by thermolysis). In the course of the reaction, new sets of typical peaks appeared in the NMR spectra at 5.0 (CpPt group, 2JPt–H ª 6 Hz), 3.0 (Pt–CH2Ar, 2JPt–H ª 99 Hz) and 0.8–1.3 ppm (Pt–Me, 2JPt–H ª 83 Hz).New absorption bands also appeared in the low-field region (7 ppm).† No signals were detected in the high-field region from 0 to –30 ppm; this fact indicates that no platinum hydrides are formed. New peaks were not detected in the absence of CpPtMe3.The experimental data correspond to a reaction mechanism involving the formation of dimethyl arylmethyl derivatives of † CpPtMe2(3,5-Me2C6H3CH2): 1H NMR (250 MHz, 20 °C, CDCl3) d: 1.01 (s, 6H, Pt–Me, 2JPt–H 83.4 Hz), 2.30 (s, 6H, 3,5-Me), 2.89 (s, 2H, Pt–CH2, 2JPt–H 98.8 Hz), 5.38 (s, 5H, C5H5, 2JPt–H 6.6 Hz), 6.82 (s, 1H, p-C–H), 6.98 (s, 2H, o-C–H). CpPtMe2(4-MeC6H4CH2): 1H NMR (250 MHz, 20 °C, CDCl3) d: 1.00 (s, 6H, Pt–Me, 2JPt–H 83.4 Hz), 2.25 (s, 3H, 4-Me), 2.93 (s, 2H, Pt–CH2, 2JPt–H 98.8 Hz), 5.38 (s, 5H, C5H5, 2JPt–H 6.2 Hz), 6.9–7.1 (m, C6H4).CpPtMe2(CH2Ph): 1H NMR (250 MHz, 20 °C, CDCl3) d: 1.02 (s, 6H, Pt–Me, 2JPt–H 83.2 Hz), 2.96 (s, 2H, Pt–CH2, 2JPt–H 98.4 Hz), 5.37 (s, 5H, C5H5, 2JPt–H 6.4 Hz), 7.10–7.24 (m, 5H, Ph). cyclopentadienylplatinum(IV).Indeed, according to mass-spectrometric data, the lipophilic residue obtained after photolysis of a CpPtMe3–toluene mixture and evaporation of the solvent contained CpPtMe2(CH2Ph) (M+, m/z 380, 381, 382 and 384). Benzene was found to be inert under the reaction conditions, and the examined alkanes produced only traces of new cyclopentadienyls. The main reaction for the above substrates is decomposition of CpPtMe3 to form the platinum containing polymer.The reproducibility of product yields for methylarenes was in a range of 20–30%. Traces of other unidentified compounds containing the CpPt unit were detected in toluene solutions after irradiation for a long time. We performed the simplest tests to reveal the reaction mechanism. The addition of 5 mg of hydroquinone to the reaction mixtures completely inhibited the photoinduced reactions.The average yields of products obtained under photolysis of CpPtMe3 in [2H8]toluene were lower than those in toluene by 1–1.5 orders of magnitude. Photolysis of CpPtMe3 in toluene mixtures with cyclohexane, p-xylene or mesitylene gave following results. Alkanes remained inert, and polymethylbenzenes competed for the metal complex.Photolysis of CpPtMe3 in the presence of iodine leaded to [PtMe3I]4 and, in a moderate yield, cyclopentadienyldimethyliodoplatinum( IV), CpPtMe2I, which was identified in the reaction mixture by spectroscopy.‡ Photolysis of CpPtMe3 in a toluene mixture with p-xylene (50 vol%) gave benzyl and p-methylbenzyl derivatives in the 1:1.5 ratio. The ratio between benzyl and 2,3-dimethylbenzyl complexes obtained in a mixture of toluene and mesitylene (1:1, v/v) was 1:2.5.Taking into account the difference in molar concentrations of active C–H bonds in the mixtures of substrates (1:1.7 and 1:2.3, respectively), we can conclude that there is no considerable difference in the reactivity of benzylic C–H bonds of these substrates. At the same time, the hydrogen kinetic isotope effect of the reaction is sufficiently high; this fact allowed us to conclude that the cleavage of benzylic C–H bonds is the rate-determining step.Taking into account the low reactivity of methylarenes towards CpPtMe3, the inertness of alkanes containing stronger C–H bonds seems to be in agreement with the above explanation.An analysis of the substrate selectivity and regioselectivity of the reaction and an analogy with the published data8 concerning the photoinduced activation of arene C–H bonds with organic platinum(II) complexes allowed us to suggest a radical mechanism of the metathesis of hydrocarbon C–H bonds and metal complex Pt–C bonds as follows: ‡ CpPtMe2I: 1H NMR (250 MHz, 20 °C, C6D6) d: 2.07 (s, 6H, Pt–Me, 2JPt–H 84.4 Hz), 5.04 (s, 5H, C5H5, 2JPt–H 12.4 Hz).h Pt Me Me Me Pt Me CH2Ar Me hn Ar–Me – CH4Mendeleev Communications Electronic Version, Issue 1, 2000 (pp. 1–42) Iodine is able to scavenge the [CpPtMe2]· radical to form CpPtMe2I. The last-mentioned reaction is responsible for the deposition of platinum-containing polymer, the main product of CpPtMe3 transformation.The absence of platinum(IV) alkyl hydrides in the reaction mixtures suggests that either the CpPtMe species are not formed or the expected organic hydridoplatinum(IV) cyclopentadienyls CpPtMe(R)(H) are unstable under the reaction conditions. To test the latter hypothesis, we attempted to synthesise CpPtMe2H. A dimethyl halogeno derivative of cyclopentadienylplatinum( IV), e.g., CpPtMe2I, can serve as a precursor in this synthesis.Cyclopentadienyldimethyliodoplatinum(IV), CpPtMe2I, was obtained in 25% yield by the treatment of polymeric dimethyldiiodoplatinum(IV) with cyclopentadienylthallium in absolute THF in an evacuated and sealed tube. The other reaction products are (h5-Cp)PtMe2(h1-Cp)§ and CpPtMe3. Cyclopentadienyldimethyliodoplatinum(IV) was treated with various hydride-ion donors (lithium aluminium hydride, lithium § (h5-Cp)PtMe2(h1-Cp): 1H NMR (250 MHz, 20 °C, C6D6) d: 1.41 (s, 6H, Pt–Me, 2JPt–H 82.1 Hz), 4.01 (s, 1H, CH–Pt, 2JPt–H 148.9 Hz), 4.93 (s, 5H, C5H5, 2JPt–H 7.0 Hz), 6.40–6.90 (m, 4H, CH=CH). dibutylaluminohydride and sodium borohydride) in THF purified with the dipotassium–anthracene adduct in evacuated and sealed NMR tubes.In all cases, the reaction mixtures became colourless at room temperature, and platinum metal was deposited. Thus, it is our opinion that the CpPtMe2H complex is insufficiently stable under the reaction conditions, and it cannot be detected in the room-temperature photolysis of CpPtMe3 solutions in hydrocarbons. A possible way of CpPtMe2H decomposition may include, e.g., its isomerization to the unstable planar complex (h1-Cp)PtMe2H; this way of decomposition was not examined theoretically in refs. 1 and 2. Nevertheless, the corresponding cationic analogues [CpPtL(Me)H]+ can exhibit higher stability. Low-temperature experiments on detecting CpPtMe2H and its cationic analogues are now in progress in our laboratory. This work was supported by the Russian Foundation for Basic Research (grant no. 97-03-33120). References 1 M. D. Su and S. Y. Chu, Organometallics, 1997, 16, 1621. 2 A. N. Vedernikov, G. A. Shamov and B. N. Solomonov, Zh. Obshch. Khim., 1998, 68, 709 (Russ. J. Gen. Chem., 1998, 68, 667). 3 (a) S. A. O’Reilly, P. S. White and J. L. Templeton, J. Am. Chem. Soc., 1996, 118, 5684; (b) A. J. Canty, A. Dedieu, H. Jin, A. Milet and M. K. Richmond, Organometallics, 1996, 15, 2845; (c) D. D. Wick and K. I. Goldberg, J. Am. Chem. Soc., 1997, 119, 10235; (d) H. A. Jenkins, G. P. A. Yap and R. J. Puddenphatt, Organometallics, 1997, 16, 1946. 4 A. H. Janowicz and R. G. Bergman, J. Am. Chem. Soc., 1982, 104, 352. 5 J. K. Hoyano and W. A. G. Graham, J. Am. Chem. Soc., 1982, 104, 3723. 6 W. D. Jones and F. J. Feher, J. Am. Chem. Soc., 1982, 104, 4240. 7 A. Shaver, Can. J. Chem., 1978, 56, 228. 8 A. Miyashita, M. Hotta and Y. Saida, J. Organometal. Chem., 1994, 473, 353. CpPtMe3 [CpPtMe2]· + Me·; [CpPtMe2]·Me· + Me–Ar [CpPtMe2]·(CH2Ar)· + CH4; [CpPtMe2]·(CH2Ar)· [CpPtMe2(CH2Ar)] or [CpPtMe2]·R· [CpPtMe] + MeR; [CpPtMe] + RH CpPtMe(R)H and nCpPtMe3 [PtC8H14]n. hn [PtMe2I2]n + nCpTl nCpPtMe2I + TlI Received: 31st May 1999; Com. 99/1493

ISSN:0959-9436     

出版商:RSC    

年代:2000

数据来源: RSC