摘要:

Mendeleev Communications Electronic Version, Issue 2, 2000 (pp. 83–124) Resolution of racemates with achiral reagents Remir G. Kostyanovsky,*a Vasilii R. Kostyanovsky,a Gul’nara K. Kadorkinaa and Vladimir Yu. Torbeevb 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 Higher Chemical College, Russian Academy of Sciences, 125047 Moscow, Russian Federation 10.1070/MC2000v010n03ABEH001330 The conditions required for resolving racemates into enantiomers by crystallization involve a deficiency of either the conglomerator or the achiral solubiliser to promote the solubility of the compound to be resolved; in both cases the resolution can also be achieved by internal entrainment, adding a single crystal of the resolvable conglomerate.L. Pasteur was the first (May 1848) to separate1 the enantiomorphic levorotatory and dextrorotatory crystals of an entirely racemic conglomerate such as sodium ammonium tartrate 1·4H2O. The individual crystals taken from this conglomerate were shown to be optically active2 as were also the first crystals precipitated from its supersaturated solution.3–5 The most important consequence of these classical experiments is the possibility to effect homochiral crystallization and, thereby, spontaneous resolution at the level of single crystals.Independently, theoretical investigation of the meanings of this fundamental phenomena — the possibility of homochiral crystal formation — was begun by O.Bravais in France in the same year 1848 (Bravais crystal lattices) and completed by E. S. Fedorov in Russia and A. Schönflies in Germany (1890– 1891). Sixty five of the 230 Fedorov space groups are chiral,† and hence all those chemical compounds which crystallise in these space groups produce crops of homochiral crystals (conglomerates) and are capable of spontaneous resolution into enantiomers by crystallization.However, all the methods developed by Pasteur (manual sorting of conglomerate crystals by a ‘chiral’ experimenter, separation via diastereomers and utilization of enzymes) and his followers (using an optically active seed by Gernez,6 see also refs. 7–9) implicate the use of either chiral chemical reagents or asymmetric physical fields.9–12 In particular, a modification of the second Pasteurian method included the use of a chiral resolving reagent in nonstoichiometric quantity.7 W.Marckwald (1896) applied a half-mole quantity of the alkaloid cinchonine to resolve tartaric acid; W. J. Pope and S. J. Peachey (1899) introduced a neutralization technique for the second moiety of the compound to be resolved. In the present work, we propose and realise a simple idea for the resolution of conglomerates without the involvement of any chiral reagent (preliminary results13 were obtained at the 150th anniversary of Pasteur’s discovery).(±)-Tartaric acid itself forms heterochiral crystals (space group P ) whereas the Pasteur salt (±)-1·4H2O containing achiral addends (NaOH, NH3, H2O) crystallises as a conglomerate7 (space group P21212).Such addends are called conglomerators (C). On the onset of crystallization from a supersaturated solution of a conglomerate, the first formed ‘left’ (L) or ‘right’ (D) crystal initiates the formation of further crystals of the same type. If the supersaturation is sufficiently great, crystallization of the opposite enantiomer occurs. Such an alternating process (circulatory crystallization14) leads to a racemic precipitate.If the same procedure is performed with a deficiency of the conglomerator, after the first enantiomer has crystallised, the second enantiomer in the form of its adduct with the conglomerator is no longer supersaturated (Scheme 1). † Chiral space groups compatible with a single enantiomer in a crystal: triclinic P1; monoclinic P2, P21, C2; orthorhombic P222, P2221, P21212, P212121, C222, C2221, I222, I 212121, F222; tetragonal P4, (P41, P43), P42, I4, I41, P422, P4212, (P4122, P4322), (P41212, P43212), P4222, P42212, I422, I4122; trigonal P3, (P31, P32), R3, P312, P321, (P3112, P3212), (P3121, P3221), R32; hexagonal P6, (P61, P65), (P62, P64), P63, P622, (P6122, P6522), P6322, (P6422, P6222); cubic P23, P213, I23, I213, F23, P432, (P4132, P4332), P4232, I432, I4132, F432, F4132 (in brackets, 11 enantiomorphous couples of space groups).We found that crystallization of the Pasteur salt from H2O at 18 °C (1 day) leads to crystals of (±)-1·4H2O in 36% yield. Crystallization under the same conditions, except that a halfmole quantity of Na2CO3 was added, gives (–)-1 (yield 30%, optical purity 55%).The experiment with Na2CO3 was then repeated but with the addition of a random crystal of conglomerate (±)-1·4H2O as a seed to the supersaturated solution; this resulted in the formation of crystalline (+)-1 (yield 35%, optical purity 65%). Such a procedure is essentially a modification of the Gernez method. It does not use any external chiral seed and can be called the internal entrainment procedure.Finally, the optical purity of (+)-1·4H2O was increased to 80% by its crystallization at 30 °C (1 day), separation of the precipitated racemate (±)-1·4H2O and evaporation of the mother liquor. Thus, the temperature conditions for conglomerate formation found by van't Hoff15 can be used for increasing the optical purity by successive conglomerate–racemate crystallizations.Racemic coordination complexes were originally resolved by A. Werner (1811) using chiral reagents.16 Many of these complexes were later shown to form conglomerates,17,18 e.g. cis- [Co(NO2)2(en)2]Cl 2 (space group P21). We have found that its analogue cis-[Co(NO2)2(en)2]Br 3 is also a conglomerate, as follows from the optical activity of its individual crystals.This is confirmed by a single crystal X-ray diffraction study of (±)-3 (space group P21). Like 1, compounds 2 and 3 can be resolved by crystallization from supersaturated solutions upon adding their individual conglomerate crystals as seeds. Bromide 3 is less soluble19 than chloride 2 by a factor of 3.6, and we succeeded in obtaining optically active bromides (–)-3 and (+)-3 (yields 15–30%, optical purity 20–50%) by crystallization from aqueous solutions of chloride (±)-2 containing a half-mole quantity of NaBr as a conglomerator.Chiral ammonium salt Me(Et)N+(Allyl)PhI–·CHCl3 4 forms a conglomerate as shown by the observed hemihedrism20 and optical activity of its single crystals,21 and confirmed by us by X-ray study (space group P212121).Obviously, the molecule of CHCl3 in compound 4 is a conglomerator. After evacuation of CHCl3 by a vacuum treatment of rubbed salt 4 (under NMR monitoring), crystallization from EtOH–Et2O with a half-mole quantity of CHCl3 gave enantiomer (+)-4 (yield 35%, optical purity 61%). A similar resolution was experimentally observed earlier with a new conglomerate.22 The high efficiency of the above method of racemate resolution was illustrated by the salts of a-phenylethylamine, one of the simplest, most easily available and widely used chiral amines.23 The sulfate,7 cinnamate24(a)–(d) 5, and succinate24(e) 6 of a-phenylethylamine are known conglomerates.Using this fact, D. Kozma et al.25 increased the optical purity of the partly enriched amine, and K.Saigo et al.24(a) described a classical resolution of (±)-5 by entrainment. The mixture of cinnamate 5 and hydrochloride (mole ratio ~1:2) containing (+)-5 as a seed crystallised from H2O–MeOH (2:3) to give first crystals of (+)-5 (yield 22%, optical purity 75%). Next, (±)-5 in a quantity ap- 1 Cryst. (LC)n nL + nD + nC (DC)n cryst. [C] ®0 [C]®0 Scheme 1Mendeleev Communications Electronic Version, Issue 3, 2000 (pp. 83–124) proximately equivalent to that of the separated crystals and a seed of (–)-5 were added to the mother liquor; crystals of (–)-5 were isolated (yield 35%, optical purity 85%). In our work, no optically active salt 5 was added as a seed which is an essential novelty of our procedure. We repeated Saigo’s experiment without the seed and obtained three portions of crystals of alternating sign of optical rotation in total yields of 35% and optical purity up to 95%.We call such a resolution as a single conglomerate method. Our second approach, called the double conglomerate method, was applied to the crystallization of an equimolar mixture of 5 and 6. This gave four portions of crystals of (+)-5 (yield 45%, optical purity 30–54%); (–)-6 was isolated from the mother liquor (yield 35%, optical purity 40%).Salts 5 and 6 differ in solubility and other properties, and they can be easily identified by NMR spectra. Thus, we can call them quasi-diastereomers. Significantly, the single conglomerate method and the double conglomerate method both provide products of optical purity up to 95.6% after a single crystallization.These methods can be applied to the resolution of amino acids as salts with aromatic sulfonic acids, which are known to be conglomerates.26 The idea of using a deficiency of the conglomerator (Scheme 1) was extended to crystallization with a deficiency of the achiral solubiliser promoting the solubility of the compound to be resolved. As mentioned above, Werner complexes 2 and 3 highly differ in solubility, therefore the Cl– anion can be considered as a solubiliser, and a spontaneous resolution can be expected when crystallising with its deficiency.Indeed, a single crystallization of an equimolar mixture of racemates 2 and 3 leads to the formation of (–)-3 or (+)-3 in 15–20% yields with an optical purity of 50–60%. Another substrate suitable for a similar resolution was found in a structural study of N-succinopyridine,25 C5H5N+CH(CO2 –)- CH2CO2H 7, an adduct of pyridine with maleic acid (space group P212121), synthesis and the properties of which have not been published.We found that prompt evaporation of the solution of an equimolar mixture of pyridine with maleic acid leads to only pyridinium maleate, whereas the prolonged keeping of the same solution (1–2 weeks) results in formation of bright transparent crystals of adduct 7 (mp 214 °C), and each crystal of which shows optical activity.By crystallization of (±)-7 from H2O with a half-mole quantity of pyridine or CF3CO2H, enantiomers (–)-7 and (+)-7 were obtained (yield 12–15%, optical purity 40-70%). We have initiated a systematic search22,28 and design29 of conglomerates.Proposed methods can be widely used for spontaneous optical enrichment of new conglomerates, which could be found, under scrutinity, among various chemicals in any laboratory. Significantly, spontaneous resolution of conglomerates in the presence of nonstoichiometric quantities of either a conglomerator or a solubiliser could be the most natural explanation for the origin of biohomochirality.This work was supported by the Russian Foundation for Basic Research (grant nos. 00-03-32738 and 00-03-81187) and INTAS (grant no. 99-0157). References 1 (a) L. Pasteur, Ann. Phys., 1848, 24, 442; (b) L. Pasteur, Compt. Rend. Acad. Sci., 1848, 26, 535; (c) L. Pasteur, Lecons de chimie professees en 1860. Soc.Chim. de Paris, Paris, 1861. 2 L.Pasteur, Ann. Chim. Phys., 1850 (3), 28, 56. 3 L.Pasteur, Ann. Chim. Phys., 1852, 34, 30. 4 G. B. Kauffman and R. D. Myers, J. Chem. Ed., 1975, 52, 777. 5 G. B. Kauffman, I. Bernal and H.-W. Schütt, Enantiomer, 1999, 4, 33. 6 D. Gernez, Compt. Rend. Acad. Sci., 1866, 63, 843. 7 J. Jacques, A. Collet and S. H. Wilen, Enantiomers, Racemates, and Resolution, Krieger Publ.Comp., Malabar, Florida, 1994. 8 (a) A. Collet, in Comprehensive Supramolecular Chemistry, ed. D. N. Reinhoudt, Pergamon, 1996, vol. 10, ch. 5, p. 13; (b) A. Collet, Enantiomer, 1999, 4, 157. 9 W. A. Bonner, Origin Life Evol. Biosphere, 1994, 24, 63; (b) W. A. Bonner, Origin Life Evol. Biosphere, 1995, 25, 175; (c) W. A. Bonner, Origin Life Evol. Biosphere, 1996, 26, 27. 10 (a) V. I. Goldanskii, in The Role of Radiation in the Origin and Evolution of Life, eds. M. Akaboshi, N. Fujii and R. Gonzales, Kioto University Press, Kioto, Japan, 2000, p. 291; (b) V. I. Goldanskii, Khim. Fiz., 1999, 18, 3 (Chem. Phys. Rep., 1999, 18, 627). 11 M. Avalos, R. Babiano, P. Cintas, J. L. Jimenez, J. C. Palacios and L. D. Barron, Chem. Rev., 1998, 98, 2391. 12 Y. Inoue, H. Tsuneishi, T. Hakushi, K. Yagi, K. Awazu and H. Onuki, J. Chem. Soc., Chem. Commun., 1996, 2627. 13 (a) R. G. Kostyanovsky, invited lecture at Paul Walden Symposium, Riga, 1998 [P. Trapencieris, Khim. Geterotsikl. Soedin., 1998, 712 (in Russian)]; (b) V. R. Kostyanovsky, Diploma, Moscow, 1998; (c) R. G. Kostyanovsky, Abstracts of the Interdisciplinary Symposium on Biological Homochirality, 1998, Italy, OP20, p. 41; (d) R.G. Kostyanovsky, G. K. Kadorkina and V. R. Kostyanovsky, Abstracts of the V Scientific Conference of N. N. Semenov Institute of Chemical Physics, RAS, 1999, Moscow, p. 60 (in Russian); (e) V. R. Kostyanovsky, Abstracts of the First Workshop on Organic Chemistry and Catalysis between Russian Academy of Sciences and Bayer AG, Moscow, 2000, p. 24. 14 G. A. Potter, Ch. Garcia, R. McCague, B. Adger and A. Collet, Angew. Chem., Int. Ed. Engl., 1996, 35, 1666. 15 J. H. van’t Hoff and C. M. van Deventer, Z. Phys. Chem., 1887, 1, 165. 16 A. Werner, Ber., 1911, 44, 1887. 17 I. Bernal and G. B. Kauffman, J. Chem. Ed., 1987, 64, 604. 18 G. B. Kauffman and I. Bernal, J. Chem. Ed., 1989, 66, 293. 19 K. Yamanari, J. Hidaka and Y.Shimura, Bull. Chem. Soc. Jpn., 1973, 46, 3724. 20 (a) E. Wedekind, Zur Stereochemie des fünfwertigen Stickstoffs, Leipzig, 1899, p. 56; (b) A. Fock, Z. Kristallogr., 1902, 35, 399; (c) E. Wedekind, Ber., 1903, 36, 3793. 21 E. Havinga, Biochim. Biophys. Acta, 1954, 13, 171. 22 R. G. Kostyanovsky, A. P. Avdeenko, S. A. Konovalova, G. K. Kadorkina and A. V. Prosyanik, Mendeleev Commun., 2000, 16. 23 E. Juaristi, J. Escalante, J. L. Leon-Romo and A. Reyes, Tetrahedron: Asymmetry, 1998, 9, 715. 24 (a) H. Nohira, M. Kai, M. Nohira, J. Noshikawa, T. Hoshiko and K. Saigo, Chem. Lett., 1981, 951; (b) K. Saigo, H. Kimoto, H. Nohira, K. Yanagi and M. Hasegawa, Bull. Chem. Soc. Jpn., 1987, 60, 3655; (c) K. Saigo, Y. Hashimoto, K. Kinbara and A. Sudo, Proc. Indian Acad.Sci. (Chem. Sci.), 1996, 108, 555; (d) K. Kinbara, Y. Hashimoto, M. Sukegawa, H. Nohira and K. Saigo, J. Am. Chem. Soc., 1996, 118, 3441; (e) D. Kozma, Z. Bocskei, K. Simon and E. Fogassy, J. Chem. Soc., Perkin Trans. 2, 1994, 1883. 25 (a) D. Kozma, Z. Madarasz, M. Acs and E. Fogassy, Chirality, 1995, 7, 381; (b) D. Kozma and E. Fogassy, Enantiomer, 1997, 2, 51. 26 H. Kimoto, K. Saigo, Y. Ohashi and M. Hasegawa, Bull. Chem. Soc. Jpn., 1989, 62, 2189. 27 M. N. G. James and M. Matsushima, Acta Crystallogr., 1976, B32, 999. 28 (a) R. G. Kostyanovsky, P. E. Dormov, P. Trapenzieris, B. Strumfs, G. K. Kadorkina, I. I. Chervin and I. Ya. Kalvin’s, Mendeleev Commun., 1999, 26; (b) D. A. Lenev, K. A. Lyssenko and R. G. Kostyanovsky, Izv. Akad. Nauk, Ser. Khim., 2000, in press. 29 (a) 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; (b) R. G. Kostyanovsky, K. A. Lyssenko, Yu. I. El’natanov, O. N. Krutius, I. A. Bronzova, Yu. I. Strelenko and V. R. Kostyanovsky, Mendeleev Commun., 1999, 106; (c) R. G. Kostyanovsky, K. A. Lyssenko, D. A. Lenev, Yu. I. El’natanov, O. N. Krutius and I. A. Bronzova, Mendeleev Commun., 1999, 151; (d) R. G. Kostyanovsky, K. A. Lyssenko and D. A. Lenev, Mendeleev Commun., 1999, 154; (e) R. G. Kostyanovsky, K. A. Lyssenko and V. R. Kostyanovsky, Mendeleev Commun., 2000, 44. Received: 23rd May 2000; Com. 00/1656

ISSN:0959-9436     

出版商:RSC    

年代:2000

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 2, 2000 (pp. 83–124) One-pot method for the generation of the trication [1,2-(CH2)2C5Me3RuC5Me4CH2]3+ from decamethylruthenocene Margarita I. Rybinskaya,* Alla A. Kamyshova, Arkadii Z. Kreindlin and Pavel V. Petrovskii A. N. Nesmeyanov Institute of Organoelement Compounds, Russian Academy of Sciences, 117813 Moscow, Russian Federation. Fax: + 7 095 135 5085; e-mail: krekis@ineos.ac.ru 10.1070/MC2000v010n03ABEH001286 The title trication was generated for the first time by the interaction of decamethylruthenocene with oxygen in CF3SO3H.Previously,1–3 multistage methods were developed for the synthesis of metalonium monocations [CH2C5Me4MC5Me5]+ 1 and homoannular and heteroannular dications [1,2-(CH2)2C5Me4- MC5Me5]2+ 2 and [1,1'-(CH2C5Me4)2M]2+ 3 (where M = Fe, Ru or Os), respectively.Single-stage methods based on permethylmetallocenes4 –7 were also proposed for the synthesis of cations 1–3. Simple and convenient methods with the use of strong protic acids exhibited the most promise.6,7 The generation of a mixture of dications 2 and 3 on dissolving (C5Me5)2Ru in oleum at 20 °C was of prime interest.7 The dications are readily formed and exhibit reasonably high thermodynamic stability due to the donor–acceptor interactions between the carbocationic sites CH2 + and two lone electron pairs of the metal atom.This suggested that structurally similar trications can also be generated because a transition metal atom in metallocenes has three lone electron pairs (dxy, dx2 – y2, dz2).It seemed reasonable to generate trications under similar conditions with the use of strong protic acids. We examined the behaviour of (C5Me5)2Ru in a CF3SO3H solution both in an inert atmosphere and in air. According to the 1H NMR spectra, the protonation product [(C5Me5)2RuH]+ and monocation 1 were formed in argon, whereas dications 2 and 3 along with a small amount of new cationic species 4 were detected in the reaction mixture in air in addition to the above products. Thus, we discovered that oxygen of the air in the presence of a strong protic acid can oxidise one, two or even three methyl groups in (C5Me5)2Ru.With the use of oxygen in this reaction, we found that the concentration of [(C5Me5)2RuH]+ decreased with a simultaneous increase in the amounts of cations 1–3 and new species 4.The 1H and 13C NMR data allowed us to describe this species as trication [1,2-(CH2)2C5Me3Ru- C5Me4CH2]3+ 4. To monitor the reaction, we performed it in a 0.7 ml NMR tube, which was filled with new portions of oxygen at regular intervals. This 1H NMR monitoring demonstrated that monocation 1 completely disappeared after twice flushing the tube with oxygen.After the flushing with oxygen was repeated three times, the distribution of reaction products was as follows: 2, 21%; 3, 22%; 4, 57%. In this case, the content of either of cations 2 and 3 decreased by ~6%. These data indicate that cations 1–3 are precursors of trication 4, and this fact together with the NMR data substantiates the structure of 4. Note that the use of monocation 1 in place of (C5Me5)2Ru in this reaction also resulted in the generation of a mixture of dications 2 and 3 and trication 4.The determination of the structure of trication 4 presented no serious problems because its 1H NMR spectrum was similar to that of CH2C5Me4 and 1,2-(CH2)2C5Me3 units in cations 1–3 with downfield shifted signals.† Trication 4 has a plane of symmetry and does not contain the C5Me5 ring.Thus, isomers with the arrangement of CH2 groups at the 1,2,3- or 1,2,4-positions of the same ring cannot exist. To compare the chemical shifts in the 1H NMR spectra with those in the spectra2 of dications 2 and 3, CD2Cl2 was added. The signals of protons of † The NMR spectra were measured on a Bruker AMX-400 spectrometer (400.13 and 100.61 MHz for 1H and 13C, respectively).C6D6 was used as an external standard for acid solutions (d C6D5H 7.25 and 127.96 ppm for 1H and 13C, respectively). 13C NMR data for 4, d (1JCH/Hz): 9.22 (133) (2Me), 9.28 (131) (2Me), 10.09 (132) (3Me), 65.66 (157) (CH2), 88.57 (172) (2CH2), 99.51, 107.16, 110.35, 114.01, 128.44, 138.50 (CCp). dication 3 in CF3SO3H without solvent are broadened and shifted downfield.The difference between the signals of protons of cations 2 and 4 is insignificant. The structure of trication 4 was also supported by 13C NMR data. In the 13C {1H} NMR spectrum, three carbon atoms of CH2 groups appear as triplets, and both carbon atoms of 1,2-(CH2)2 groups with the chemical shift 88.57 ppm exhibit the constant 1JCH 172 Hz, which is close to that of dication 2 (1JCH 171 Hz).2,7 The signal of the carbon atom of the third CH2 group is upfield (65.66 ppm, 1JCH 157 Hz).In this case, the constant is close to the constant obtained for monocation 1 (1JCH 167 Hz).‡,2 Thus, we found that a metal atom in Group VIII element metallocenes can stabilise three carbocationic centres on the formation of trication 4, which is the first example of organometallic onium complexes (note that in contrast to the salt of monocation 1,6 the salts of cations 2–4 are sensitive to trace water and were not isolated as individual compounds).It is unusual that three lone electron pairs of the metal atom participate in the formation of Ru–CH2 bonds. The procedure developed for the oxidation of methyl groups can also be helpful in the case of other transition metal complexes.The only example similar to the reaction considered is the recently reported8 oxidation of methane with oxygen in concentrated H2SO4 in the presence of PtII catalysts to form methanol derivatives. This work was supported by the Russian Foundation for Basic Research (grant nos. 97-03-32917 and 00-03-32894). ‡ 1H NMR spectra of the ruthenium cations complexes (d/ppm).The chemical shifts in CF3SO3H are given in parentheses, the other values were measured upon addition of CD2Cl2 to a CF3SO3H solution. 1: (1.96) (C5Me5), (1.71) (a-Me), (2.04) (b-Me), (4.56) (s, CH2). 2: 2.35 (2.31) (C5Me5), 2.22 (2.18) (a-Me), 2.55 (2.51) (b-Me), 4.86 (4.82) and 5.26 (5.27) (2d, CH2 AB, 2JHH gem 2.0 Hz). 3: 2.28 (2.01) and 2.33 (2.23) (a,a'-Me), 2.47 (2.32) and 2.59 (2.55) (b,b'-Me), 4.99 (5.26 s) and 5.41 (5.86 s) (2d, CH2 AB, 2JHH gem 2.0 Hz). 4: 2.29 (2.28) and 2.31 (2.33) (2×6H, a,a'-Me), 2.48 (2.47) and 2.62 (2.59) (6H and 3H, b,b'-Me), 5.51 (5.50) (s, 2H, CH2), 5.07 (5.03) and 5.49 (5.51) (2d, 2×2H, CH2 AB, 2JHH gem 1.9 Hz). Ru CF3SO3H, O2 20 °C CH2 Ru CH2 CH2 Ru + 2+ CH2 CH2 Ru 2+ H2C CH2 CH2 Ru 3+ 1 2 3 4 Scheme 1Mendeleev Communications Electronic Version, Issue 3, 2000 (pp. 83–124) References 1 A. Z. Kreindlin, P. V. Petrovskii, M. I. Rybinskaya, A. I. Yanovsky and Yu. T. Struchkov, J. Organomet. Chem., 1987, 319, 229. 2 A. Z. Kreindlin, E. I. Fedin, P. V. Petrovskii, M. I. Rybinskaya, R. M. Minyaev and R. Hoffmann, Organometallics, 1991, 1206. 3 M. I. Rybinskaya, A. Z. Kreindlin, R. Hoffmann and R. M. Minyaev, Izv. Akad. Nauk, Ser. Khim., 1994, 1701 (Russ. Chem. Bull., 1994, 43, 1605). 4 U. Kolle and J. Grub, J. Organomet. Chem., 1985, 289, 133. 5 A. Z. Kreindlin, P. V. Petrovskii and M. I. Rybinskaya, Izv. Akad. Nauk, Ser. Khim., 1987, 1620 (Russ. Chem. Bull., 1987, 36, 1489). 6 A. A. Kamyshova, A. Z. Kreindlin, M. I. Rybinskaya and P. V. Petrovskii, Izv. Akad. Nauk, Ser. Khim., 1999, 587 (Russ. Chem. Bull., 1999, 48, 581). 7 A. A. Kamyshova, A. Z. Kreindlin, M. I. Rybinskaya and P. V. Petrovskii, Izv. Akad. Nauk, Ser. Khim., 2000, 517 (Russ. Chem. Bull., 2000, 49, 520). 8 R. A. Periana, D. J. Taube, S. Gamble, H. Taube, T. Satoh and H. Fujii, Science, 1998, 280, 560. Received: 21st February 2000; Com. 00/1612

ISSN:0959-9436     

出版商:RSC    

年代:2000

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 2, 2000 (pp. 83–124) Outer-sphere anion–anion charge transfer in a solid hexacyanoferrate Sergei I. Gorel’skii,a Tatyana G. Kim,b Tamara P. Klimova,c Vitalii Yu. Kotov,*b Boris V. Lokshin,c Yurii D. Perfil’ev,d Taisia I. Sherbakd and Galina A. Tsirlinad a Department of Chemistry, York University, Toronto, Ontario, M3J1P3 Canada. Fax: +1 416 736 5936; e-mail: serge@yorku.ca b Higher Chemical College, Russian Academy of Sciences, 125047 Moscow, Russian Federation.Fax: +7 095 200 4204; e-mail: tsir@elch.chem.msu.ru c A. N. Nesmeyanov Institute of Organoelement Compounds, Russian Academy of Sciences, 117813 Moscow, Russian Federation. Fax: +7 095 135 5085; e-mail: bloksh@ineos.ac.ru d Department of Chemistry, M. V.Lomonosov Moscow State University, 119899 Moscow, Russian Federation. Fax: +7 095 939 0171; e-mail: tsir@elch.chem.msu.ru 10.1070/MC2000v010n03ABEH001280 A solid solution based on Cs2Mg[Fe(CN)6] containing both [Fe(CN)6]4– and [Fe(CN)6]3– ions exhibits a green colour associated with electronic absorption spectra bands due to the outer-sphere charge transfer between complex anions.Aqueous solutions simultaneously containing the hexacyanoferrate ions [Fe(CN)6]3– and [Fe(CN)6]4– can exhibit a green colour, which was never observed in systems with only one of the ions. The electronic absorption spectra demonstrate1 outer-sphere charge-transfer bands in the region 12000–12500 cm–1. Inasmuch as the position and intensity of these bands depend on the type and concentration of alkali metal cations,2 we can suggest that association of the anions results from cooperative interactions.Similar interactions were also observed in other anion–anion systems.3–6 A comparative analysis of spectral and kinetic parameters of electron transfer for ionic associates in solutions containing K+, [Fe(CN)6]4– and [Fe(CN)6]3– ions revealed7 that the best results can be obtained at a contact distance between ions taken to be 6.9 Å.This value is close to the distance between iron atoms in the crystalline hexacyanoferrates K3[Fe(CN)6] and K4[Fe(CN)6]·3H2O.8,9 This fact allowed us to expect that spectra of solid solutions containing hexacyanoferrate ions will also exhibit outer-sphere charge-transfer bands. We studied solid solutions based on caesium and magnesium hexacyanoferrates and related systems in order to find outer-sphere charge-transfer bands.We decided on these test materials because isostructural analogues of these compounds are known, for which the charges of complex ions differ by unity, for example, Cs2Mg[Fe(CN)6]– Cs2Li[Fe(CN)6] and KLa[Fe(CN)6]·4H2O–La[Fe(CN)6]·5H2O.10–13 We obtained three compounds in the Cs2Mg[Fe(CN)6]– Cs2Li[Fe(CN)6] system.By adding caesium chloride to a concentrated aqueous solution simultaneously containing lithium chloride and K3[Fe(CN)6], a well crystallised orange precipitate was obtained. X-Ray diffraction data (Guignet monochromator; CuKa radiation, Ge standard) indicate that this compound is crystallised in the space group Fm3m with the lattice parameter a = 10.561(6) Å, which is consistent with the published data11 for Cs2Li[Fe(CN)6].Upon the addition of a caesium chloride solution (0.1 mol dm–3) to a solution containing K4[Fe(CN)6] and MgSO4 (0.1 mol dm–3 each), a poorly crystalline white precipitate was formed. After washing with water and acetone and drying in air, the precipitate became pale green. The compound crystallised in the space group Fm3m.The lattice parameter a = 10.396(3) Å was substantially lower than the published12 value a = 10.446 Å for Cs2Mg[Fe- (CN)6]. The IR spectrum (Nicolet-Magna 750, Nujol) of the compound in the region of 400–4000 cm–1 exhibited bands due to water molecules and sulfate anions in addition to the valence bands and bands of deformation vibrations for Cs2Mg[Fe(CN)6]14 [Figure 1(a)].According to the thermogravimetry data (TGD- 7000, Ulvac Sinku-Riko), approximately 1.4 water molecules accounts for the formula unit Cs2Mg[Fe(CN)6]. The presence of water and sulfate ions in the isolated compound suggests that this compound is a solid solution in which a part of [Fe(CN)6]4– ions in the Cs2Mg[Fe(CN)6] matrix is displaced by sulfate ions.Moreover, the charge is compensated by the displacement of caesium ions in the tetrahedral lattice interstitials by water molecules. Additional water molecules enter the lattice ligand positions and saturate the coordination positions of magnesium ions. Upon the addition of a caesium chloride (0.1 mol dm–3) solution to a solution simultaneously containing lithium chloride (0.1 mol dm–3), magnesium sulfate (0.1 mol dm–3), K3[Fe(CN)6] (0.05 mol dm–3) and K4[Fe(CN)6] (0.05 mol dm–3), a lettuce-green precipitate was isolated. The colour of the precipitate remained unchanged after the addition of an alkali solution to the initial solution (alkalies usually destroy coloured bridge cyanide complexes).The compound crystallises in the space group Fm3m. The cubic lattice parameter a = 10.370(5) Å is substantially lower than that in two samples isolated earlier.Such a deviation from the Vegard law allowed us to assume that, during the formation of this compound, the substitution of [Fe(CN)6]3– for [Fe(CN)6]4– 2.5 2.0 1.5 1.0 0.5 0.0 4000 3000 2000 1500 1000 500 Absorbance n/cm–1 (a) (b) 2.5 2.0 1.5 1.0 0.5 0.0 Figure 1 IR spectra of solid solutions based on Cs2Mg[Fe(CN)6]: (a) without [Fe(CN)6]3– and (b) with [Fe(CN)6]3–. 3405 2087 1635 1127 994 592 475 3382 2090 1645 1135 994 593 479Mendeleev Communications Electronic Version, Issue 3, 2000 (pp. 83–124) ions was accompanied by the displacement of caesium ions with water molecules rather than by the displacement of magnesium ions with lithium ions. Indeed, the removal of lithium ions from the initial solution had no effect on the colour and lattice parameters of the precipitate.The peak positions and intensities in the IR spectra of the isolated substance [Figure 1(b)] were consistent with the parameters of similar peaks for samples obtained in the absence of [Fe(CN)6]3– ions. In accordance with the thermogravimetry data, approximately 1.5 water molecules account for the formula unit Cs2Mg[Fe(CN)6].The diffuse reflection (Hitachi M 340) and electronic absorption spectra (Hitachi EPS 3T, Nujol, quartz substrate) of the solid solution exhibited additional bands in comparison with the spectrum of a FeIII-free sample (Figure 2). The absorption band at 23000 cm–1 is associated with the [Fe- (CN)6]3– ion. The absorption at 11000 (absorption spectrum) or 10000 cm–1 (reflection spectrum) can be attributed to the outersphere charge transfer between the ions [Fe(CN)6]4– and [Fe- (CN)6]3–.Note that the intensity ratio between the above bands in the difference spectrum is about 4, whereas the ratio between the molar absorption coefficients of [Fe(CN)6]3– and the associate [Fe(CN)6]4–, nK+, [Fe(CN)6]3– is about 37 in aqueous solutions.4 Probably, this is due to specific effects of the cationic composition and different FeII:FeIII ratios in an aqueous solution (1:1)4 and in a solid. The relative [Fe(CN)6]3– content of the solid solution was estimated as 4–6% by comparing the absorbance of [Fe(CN)6]3– under similar conditions.Hence, a [Fe(CN)6]3– ion can be surrounded by up to eight [Fe(CN)6]4– ions (the closest distance 7.33 Å).The conclusion concerning the predominance of FeII is consistent with the Mössbauer data for isolated compounds at 293 K (a ‘Perseus’ spectrometer with laser stabilization, control and calibration of the velocity; 57Co source in chromium). The isomeric shift with reference to sodium nitroprusside and the width at half-height of the singlet signal for green samples are 0.175(2) and 0.285(3) mm s–1, respectively.The analogous values for Cs2Li[Fe(CN)6] are 0.120(2) and 0.257(3) mm s–1 and for a solid solution based on Cs2Mg[Fe- (CN)6] containing no [Fe(CN)6]3–, 0.179(2) and 0.321(7) mm s–1, respectively. The low concentration of [Fe(CN)6]3– ions in the green solid solution is explained by a much lower solubility of FeII compounds as compared with FeIII compounds.Indeed, an aqueous solution containing potassium hexacyanoferrate(III), magnesium sulfate and sodium chloride (0.1 mol dm–3 each) is stable in storage. A cyclic voltammogram [glassy carbon electrode, PI-50-1.1 potentiostat, cycling in the range 0 < E < 1.2 V (SCE), scan rate 50 mV s–1] exhibits a peak that corresponds to the reduction of complex ions in solution at E = 0.2 V.The peak corresponding to the oxidation of complex ions is shifted to E = = 0.7 V (in the absence of magnesium and caesium salts, the corresponding current peak is observed at E = 0.34 V). In the first approximation, the voltammograms are reproducible in the first 10–12 cycles; this fact is probably associated with quasi-reversibility of the formation of insoluble films on the electrode during the reduction of the complex and the oxidative dissolution.However, the latter process is incomplete because, in the course of further cycling, the nonconductive film becomes thicker and can cause a substantial potential drop, which leads to a gradual smoothing and, ultimately, the complete disapperance of peaks.The lower the potential of the anodic cycling limit, the quicker the increase in the potential difference between the peaks and the greater their distortion. The results obtained allowed us to expect the appearance of outer-sphere charge-transfer bands in the spectra of other systems similar in composition, for instance, in the KLa[Fe(CN)6]· ·4H2O–La[Fe(CN)6]·5H2O system. However, upon the addition of lanthanum nitrate to solutions with the K3[Fe(CN)6]: K4[Fe- (CN)6] ratios 1:1 and 10:1, colourless precipitates were formed. According to the X-ray diffraction data, the unit-cell parameters of these compounds coincide with those of KLa[Fe(CN)6]·4H2O.The replacement of potassium by caesium in this system or of magnesium by calcium in the system studied earlier did not result in the formation of coloured precipitates.Taking into account similar contact distances between hexacyanoferrate ions in all of the isolated systems, we can conclude that two factors are responsible for the formation of coloured precipitates. The first factor is associated with different mutual solubilities of hexacyanoferrates. The second is determined by the arrangement of ions in solid solutions.Hexacyanoferrate ions are localised on the second-order axes with respect to one another only for the magnesium system in solid solutions with relatively high contents of different-valence ions. In such a mutual arrangement of complex ions, the overlap between the t2g orbitals responsible for the electron transfer between ions was found to be maximum, and this is favourable for an increase in the intensity of charge-transfer bands. This work was supported by the Russian Foundation for Basic Research (grant no. 99-03-32367a). References 1 D. E. Khostariya, R. Meusinger and R. Billing, J. Phys. Chem., 1995, 99, 3592. 2 D. E. Khostariya, A. M. Kjaer, T. A. Marsagishvili and J. Ulstrup, J. Phys. Chem., 1991, 95, 8797. 3 F. Pina and M. Maestri, Inorg. Chim. Acta, 1988, 142, 223. 4 R. Billing and D. E. Khostariya, Inorg. Chem., 1994, 33, 4038. 5 A. B. Nikol’skii and V. Yu. Kotov, Mendeleev Commun., 1995, 139. 6 V. Yu. Kotov and S. I. Gorelsky, Izv. Akad. Nauk, Ser. Khim., 1999, 833 (Russ. Chem. Bull., 1999, 48, 823). 7 G. A. Tsirlina and V. Yu. Kotov, Mendeleev Commun., 1999, 181. 8 N. G. Vannerberg, Acta Chem. Scand., 1972, 26, 2863. 9 R. Kiriyama, H. Kiriyama, T. Wada, N. Niizeri and H. Hirabayashi, J. Phys. Soc. Jpn., 1964, 19, 540. 10 B. I. Swanson, S. I. Hamburg and R. R. Ryan, Inorg. Chem., 1974, 13, 1685. 11 G. W. Beall, W. O. Milligan, J. Korp, I. Bernal and R. K. McMullan, Inorg. Chem., 1977, 16, 207. 12 G. W. Beall, D. F. Mullica and W. O. Milligan, Acta Crystallogr. B., 1978, 34, 1446. 13 W. E. Bailey, R. J. Williams and W. O. Milligan, Acta Crystallogr. B., 1973, 29, 1365. 14 B. I. Swanson and J. J. Rafalko, Inorg. Chem., 1976, 15, 249. 0.18 0.16 0.14 0.12 0.10 0.08 0.06 0.04 0.02 0.00 6000 12000 18000 24000 Absorbance n/cm–1 Figure 2 Difference electronic absorption spectrum of a solid solution based on Cs2Mg[Fe(CN)6] with [Fe(CN)6]3–, as measured with reference to a sample without [Fe(CN)6]3–. Received: 10th February 2000; Com. 00/1606

ISSN:0959-9436     

出版商:RSC    

年代:2000

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 2, 2000 (pp. 83–124) The nature of the intramolecular transannular Si···N interaction in crystalline 1-methylsilatrane, as found from X-ray diffraction data Konstantin A. Lyssenko,*a Alexander A. Korlyukov,b Mikhail Yu. Antipin,a Sergey P. Knyazev,b Valerii N. Kirin,b Nicolay V. Alexeevb and Eugenii A. Chernyshevb a A. N. Nesmeyanov Institute of Organoelement Compounds, Russian Academy of Sciences, 117813 Moscow, Russian Federation.Fax: +7 095 135 5085; e-mail: kostya@xrlab.ineos.ac.ru b M. V. Lomonosov Moscow State Academy of Fine Chemical Technology, 117571 Moscow, Russian Federation. Fax: +7 095 430 7983; e-mail: xteoc2@mitht.unesco.rssi.ru 10.1070/MC2000v010n03ABEH001270 The primarily electrostatic nature of the intramolecular transannular Si···N interaction was found from the electron-density distribution in crystalline 1-methylsilatrane at 100 K.Silatranes are of great interest in theoretical chemistry because of the dominance of the endo-form in their structure, which is related to the intramolecular transannular Si···N interaction (see, for example, ref. 1). Numerous structural studies have shown that the Si···N distance in silatranes varies in a wide range (2.0– 2.3 Å) and depends on the nature of the substitutent at the silicon atom.2 The character of the Si···N interaction, as well as the electron influence of a substituent at the silicon atom on the Si···N distance, is usually described in terms of a hypervalent model as a three-centre–four electron bond (3c–4e bond).2–3 To test this model, we carried out a topological analysis of the electron-density distribution function r(r) in crystalline 1-methylsilatrane 1 on the basis of high-resolution X-ray diffraction data at 100 K (Figure 1).† Experimental studies of the r(r) in silatranes are limited only by a qualitative analysis of the deformation electron density (DED) distribution in 1-fluorosilatrane4(a) and 1-chloromethylsilatrane.4(b) The molecular geometry parameters of 1 at 100 K (Figure 1) are close to those at 298 K.5 Molecules of 1 in a crystal are characterised by the approximate C3 symmetry.The Si and N atoms that participate in the Si···N interaction have distorted bipyramidal and trigonal configurations with the deviations in opposite directions from the planes of their neighbouring atoms by 0.202 and 0.371 Å, respectively.The Si···N distance [2.1604(3) Å] is typical of silatranes with electron-donor substitutients.2 The Si(1)–C(7) bond is significantly elongated [1.8801(4) Å] in comparison with the corresponding values for alkylsiloxanes (1.858 Å).6 Because of the weak C–H···O contacts in crystalline 1 [C(3)– H(3B)···O(3') (2 – x, –y, 1 – z), H(3B)···O(3') 2.48 Å, C(3)– H(3B)–O(3') 156°, C(3)···O(3') 3.4864(5) Å], the molecules are arranged in centrosymmetrical dimers.An analysis of the electron-density distribution‡ was performed by the DED maping and topological analysis of the r(r) function in terms of the Bader theory ‘Atoms in molecules’ (AIM).7,§ According to the DED map in the plane of Si(1), N(1), O(1) † Crystallographic data for 1: crystals of C7H15NO3Si are monoclinic at 100 K, space group P21/n, a = 7.5663(1) Å, b = 12.0408(2) Å, c = = 9.6210(1) Å, b = 91.740(1)°, V = 876.11(2) Å3, Z = 4, M= 189.29, dcalc = 1.435 g cm–3, m(MoKa) = 0.236 mm–1, F(000) = 408. Intensities of 20349 reflections were measured with a Smart 1000 CCD diffractometer at 100 K [l(MoKa) = 0.71072 Å, w-scans with a 0.3° step in w and 10 s per frame exposure, 2q < 90°], and 7397 independent reflections (Rint = = 0.0135) were used in the further refinement.The structure was solved by a direct method and refined by the full-matrix least-squares technique against F2 in the anisotropic–isotropic approximation. Hydrogen atoms were located from the Fourier synthesis and refined in the isotropic approximation.The refinement converged to wR2 = 0.0911 and GOF = = 1.031 for all independent reflections [R1 = 0.0307 was calculated against F for 6305 observed reflections with I > 2s(I)]. All calculations were performed using SHELXTL PLUS 5.0 on IBM PC AT. 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 1135/64. and C(7) atoms, a concentration of the electron density is observed in the covalent Si–C, Si–O and C–H bonds and along the line of the hypervalent Si(1)···N(1) interaction.The corresponding DED maximum (0.7 e Å–3) at the Si(1)···N(1) line is significally shifted towards the nitrogen atom and is located at an approximate distance of 0.45 Å. This fact allowed us to suggest that the nitrogen lone pair which participates in the formation of the 3c–4e bond (according to the hypervalent model) has a predominantly atomic character.‡ The analytical form of the electron density was obtained by a multipole refinement based on the Hansen–Coppens8 formalism using the XD program package.9 The level of the multipole expansion was hexadecapole for Si(1) and N(1), octadecapole for all oxygen and carbons atoms and dipole for hydrogens. The scattering factor of the hydrogen atoms was calculated from the contracted radial density functions (k = 1.2).The refinement was carried out against F without any symmetry restraints with the exception of hydrogens for which a cylindrical symmetry was assumed. The refinement converged to R = 0.0211, wR = 0.0238, GOF = = 1.69. The ratio of the number of reflections to the number of refined parameters was more than 20. § According to the AIM theory, a chemical bond is described by means of analysis of the electron density r(r) and its Laplacian Ñ2r(r) values in the so-called critical points (CPs), where gradient of the electron density vanishes, Ñr(r) = 0.The type of the CP is determined by the number of non-zero eigenvalues of the Hessian matrix, as well as by the sum of their signs. A set of the nondegenerated CP constitutes a molecular graph in which CP of the (3, –3) type correspond to a nuclear position, (3, +3), to a cage, (3, +1) to a cycle, and (3, –1), to a chemical bond.Thus, the existence of the CP (3, –1) is equivalent to the existence of a bond path (line of the maximum electron-density gradient) and is a necessary condition for the formation of a chemical bond. C(7) Si(1) O(2) C(3) C(4) N(1) C(2) C(6) C(5) O(3) O(1) C(1) Figure 1 The molecular structure of 1 (85% probability level).The hydrogen atoms are omitted for clarity. Selected bond lengths (Å): Si(1)– O(2) 1.6776(3), Si(1)–O(3) 1.6795(3), Si(1)–O(1) 1.6803(3), Si(1)–C(7) 1.8801(4), Si(1)–N(1) 2.1604(3), O(1)–C(1) 1.4199(5), O(2)–C(3) 1.4235(4), O(3)–C(5) 1.4196(5), N(1)–C(2) 1.4722(5), N(1)–C(6) 1.4735(5), N(1)– C(4) 1.4765(4); selected bond angles (°): O(2)–Si(1)–O(3) 119.246(16), O(2)–Si(1)–O(1) 118.925(16), O(3)–Si(1)–O(1) 117.554(16), O(2)–Si(1)– C(7) 96.80(2), O(3)–Si(1)–C(7) 97.00(2), O(1)–Si(1)–C(7) 96.92(2), O(2)– Si(1)–N(1) 83.19(1), O(3)–Si(1)–N(1) 83.01(1), O(1)–Si(1)–N(1) 83.08(1), C(7)–Si(1)–N(1) 179.99(2), C(2)–N(1)–C(6) 113.77(3), C(2)–N(1)–C(4) 114.25(3), C(6)–N(1)–C(4) 113.67(3), C(2)–N(1)–Si(1) 104.56(2), C(6)– N(1)–Si(1) 104.67(2), C(4)–N(1)–Si(1) 104.48(2).Mendeleev Communications Electronic Version, Issue 3, 2000 (pp. 83–124) Similar conclusions can be made from the results of a topological analysis of the electron density distribution, in particular, from the values of r(r) and Ñ2r(r) at the critical points (CPs) of the type (3, –1) bonds.§ A topological analysis of the r(r) demonstrated the coincidence between the molecular graph obtained by the CP analysis and the structural formula.Note that CP (3, –1) were found not only at the Si–O, C–C, C–H and N–C bonds, but also in the regions of the transannular Si···N interaction and the intermolecular C–H···O contact. The Si···N bond resulted in the appearance of three CPs of the type (3, +1), which correspond to the cycle formation.Thus, the characteristic set of CPs in 1-methylsilatrane satisfies the Poincare–Hopf equation. 10 The difference between the bond path length7,§ and the interatomic distances in 1 does not exceed 0.002 Å. This fact is indicative of the absence of steric strains in the five-membered rings formed by the Si···N transannular interaction.An analysis of the atomic charges calculated from the monopole occupancies showed that the positive charge (0.44 e) is mainly localised at the Si atom, while the negative charge (–0.34 and –0.42 e), at the N and O atoms. The dipole moment was calculated to be 6.9 D, which is close to the experimental value of 5.3 D.11 A similar increase in the molecular dipole moment in a crystal is usually related to the molecular polarization in a crystal field.12 An analysis of the topological characteristics of r(r) in the CP (3, –1) for the Si···N and Si–O bonds and C–H···O contacts revealed that Ñ2r(r) values are positive for these bonds (0.83, 7.17 and 0.78 e A–5, respectively). Thus, the formation of chemical bonds results in an electron density depletion rather than accumulation in the interatomic region.This is typical of the ‘closed shell’ type interaction.7,¶ The character of the interatomic interaction in the area of the C–H···O contact is clear, while positive Ñ2r(r) values at the covalent Si–O and transannular Si···N bonds makes it impossible to describe their nature unambiguously. It was found previously by quantum-chemical calculations for Si–O bonds that despite of an electron-density depletion in the Si–O bond, r(r) values in the corresponding CP (3, –1) are relatively high, and the local energy density E(r) is negative, which is a typical characteristic of the intermediate type interaction. 13,16,17 Probably, such a character of the Si–O bond is caused by its high polarity, as well as by a repulsion of the electron density of the Si–O bond and nonbonded oxygen lone pairs (lone pair weakening effect18).An analysis of the electron density and its topology demonstrated that the value of r(r) at the Si–O bond (0.99 e A–3) is significantly higher than that at the Si···N and C–H···O bonds (0.45 and 0.04 e Å–3, respectively), and the Ñ2r(r) for the last two bonds are almost equal [for comparison, r(r) and Ñ2r(r) in the LiF molecule are 0.51 e Å–3 and 9.23 e Å–5, respectively, and E(r) is positive (0.0198 a.u.)15].Thus, we can conclude that the Si···N bond has most probably an electrostatic nature. Such a character of the Si···N bond is consistent with its high polarizability and explains an essential decrease of the Si···N distance in the transition from a gas phase (2.45 Å)19 to a crystal [2.1604(3) Å].On the contrary, the Ñ2r(r) values for C–C, C–O, N–C and C–H covalent bonds are negative and equal to –9.78, 10.83, –11.29 and –14.77 e Å–5, respectively. The r(r) values for these bonds (1.66, 1.79, 1.73 and 1.75 e Å–3, respectively) are much higher in comparison with the corresponding values for the Si···N and C–H···O bonds. The electron-density topology characteristics at the CP (3, –1) for C–C, C–H, C–N and C–O bonds in 1 are in good agreement with experimental and theoretical data on the r(r) in organic compounds (see, for example, refs. 20 and 21). Thus, the X-ray diffraction study of the electron density in 1-methylsilatrane demonstrated that the distribution in the transannular Si···N interaction belongs to the so-called closed-shell interaction with a very small covalent contribution.A further investigation into silatrane derivatives with acceptor substituents will make it possible to evaluate the influence of substituents at the Si atom on the character of the transannular interaction and to check the correctness of the theoretical procedures used for the description of hypervalent bonds.¶ ‘Shared interactions’ and ‘closed shell’ interactions differ mainly by the sign of the Ñ2r(r) in the CP (3, –1).7 Shared interactions are characterised by negative Ñ2r(r) values and high r(r) values, while in the closed shell interactions the value of Ñ2r(r) is positive, and the total r(r) is small.13–15 However, the positive Ñ2r(r) value is not a unique criterion of the closed shell interaction, the necessary condition is the positive value of the local energy density, which is related to Ñ2r(r) by the equation: E(r) = V(r) + G(r) =G(r) – ( 2/4m)Ñ2r(r), where V(r) and G(r) are the local potential and kinetic energy densities, respectively.It can be seen that if the Ñ2r(r) is positive, the value of E(r) may remain negative if the potential energy density (a priori negative) exceeds the kinetic energy in the absolute value.The bonds characterised by a positive value of Ñ2r(r) and a negative value of E(r) correspond to an intermediate type of the interatomic interaction. H(2B) N(1) C(1) O(1) Si(1) C(7) H(7C) Figure 2 Static deformation electron density in the plane of Si(1), O(1), N(1) and C(7) atoms [atoms C(1) and C(2) are deviated from this plane by 0.15 and –0.44 Å, respectively].Interval between isolines is 0.05 e Å–3, negative contours are dashed. H(2B) N(1) C(1) O(1) Si(1) H(7C) C(7) Figure 3 Laplacian of the electron density [–Ñ2r(r)] in the plane shown in Figure 2. Negative lines are dashed, contours are drawn using logarithmic scale. hMendeleev Communications Electronic Version, Issue 2, 2000 (pp. 83–124) Note that the recent theoretical calculation [MP2/6-31G(d), B3LYP/6-31G(d)]22 of 1-methylsilatrane supported the idea that Si···N is the closed-shell interaction and hence is characterised by the electrostatic nature. For comparison, the r(r) and Ñ2r(r) values in the CP (3, –1) for the Si···N bond, according to the MP2/6-31G(d) calculations, are 0.26 e Å–3 and 0.82 e Å–5, respectively.These values are close to the experimental data (0.45 e Å–3 and 0.83 e Å–5) taking into account the elongation of this bond by 0.28 Å in the theoretical calculation. This work was supported by the Russian Foundation for Basic Research (grant nos. 00-15-97359 and 00-03-32807a). References 1 M. W. Schmidt, T. L. Windus and M. S. Gordon, J. Am. Chem. Soc., 1995, 117, 7480. 2 V. E. Shklover, Yu. T. Struchkov and M. G. Voronkov, Usp. Khim., 1989, 58, 353 (Russ. Chem. Rev., 1989, 58, 211). 3 V. F. Sidorkin, V. A. Pestunovich and M. G. Voronkov, Dokl. Akad. Nauk SSSR, 1977, 235 [Dokl. Chem. (Engl. Transl.), 1977, 160]. 4 (a) L. Parkanyi, P. Hencsei, L. Bitatsi and T. Muller, J. Organomet. Chem., 1984, 269, 1; (b) Z. Najue and L. Yumin, Proceedings of the Symposium ‘Molecular Structure: Chemical Reactivity and Biological Activity’, China, 1986, p. 272. 5 Lai Wu-Jiang, Hang Man-Shui, Huang Ming-Sheng and Hu Sheng-Zhi, Jiegou Huaxue (J. Struct. Chem.), 1991, 10, 258. 6 Structure Correlation, eds. H. B. Burgi and J. D. Dunitz, VCH Publishers, New York, 1994, vols. 1, 2. 7 R. F. W. Bader, Atoms in Molecules. A Quantum Theory, Clarendron Press, Oxford, 1990. 8 N. K. Hansen and P. Coppens, Acta Crystallogr., 1978, A34, 909. 9 T. Koritsansky, S. T. Howar, T. Richter, P. R. Mallinson, Z. Su and N. K. Hansen, XD. A Computer Program Package for Multipole Refinement and Analysis of Charge Densities from X-Ray Diffraction Data, 1995. 10 K. Collard and G. G. Hall, Int. J. Quantum Chem., 1977, 12, 623. 11 M. G. Voronkov, I. B. Mazheika and I. G. Zelchan, Khim. Geterotsikl. Soedin., 1965, 58 [Chem. Heterocycl. Compd. (Engl. Transl.), 1965, 45]. 12 G. Gatti, V. R. Saunders and C. Roetti, J. Chem. Phys., 1994, 101, 10686. 13 R. W. F. Bader and H. Essen, J. Chem. Phys., 1984, 80, 1943. 14 D. Cremer and E. Kraka, Croat. Chem. Acta, 1984, 57, 1259. 15 R. W. F. Bader, J. Chem. Phys., 1998, A102, 7314. 16 R. J. Gillespie and S. A. Johnson, Inorg. Chem., 1997, 36, 3031. 17 G. V. Gibbs, M. B. Boisen, F. C. Hill, O. Tamada and R. T. Downs, Phys. Chem. Minerals, 1998, 25, 574. 18 S. Snaik, P. Maitre, G. Sini and P. C. Hiberty, J. Am. Chem. Soc., 1992, 114, 7861. 19 Q. Shen and R. L. Hildebrandt, J. Mol. Struct., 1980, 64, 257. 20 P. Roversi, M. Barzaghi, F. Merati and R. Destro, Can. J. Chem., 1996, 74, 1145. 21 T. Koritsanszky, J. Buschmann and P. Luger, J. Phys. Chem., 1996, 100, 10547. 22 J. M. Anglada, C. Bo, J. M. Bofill, R. Crehuet and J. M. Poblet, Organometallics, 1999, 18, 5584. Received: 28th January 2000; Com. 00/1596

ISSN:0959-9436     

出版商:RSC    

年代:2000

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 2, 2000 (pp. 83–124) Synthesis of 5,7-diacetamido-3,5,7,9-tetradeoxy-L-glycero-D-galacto- and -L-glycero-D-talo-nonulosonic acids, putative components of bacterial lipopolysaccharides Yury E. Tsvetkov,*a Alexander S. Shashkov,a Yuriy A. Knirel,a,b Leon V. Backinowskya and Ulrich Zähringerb a N. D. Zelinsky Institute of Organic Chemistry, Russian Academy of Sciences, 117913 Moscow, Russian Federation.Fax: +7 095 135 5328 b Forschungszentrum Borstel, Zentrum für Medizin und Biowissenschaften, 23845 Borstel, Germany 10.1070/MC2000v010n03ABEH001287 The title acids were synthesised by condensation of 2,4-diacetamido-2,4,6-trideoxy-L-gulose with oxalacetic acid and characterised by 1H and 13C NMR spectroscopy. N-Acyl and O-acetyl derivatives of 5,7-diamino-3,5,7,9-tetradeoxynonulosonic acids are the components of lipopolysaccharides of Gram-negative bacteria, in which they play a role in serological specificity and endow the bacterial surface with peculiar physico-chemical properties.Pseudaminic and legionaminic acids were discovered in the mid-1980s, and the L-glycero-L-manno and D-glycero-L-galacto configurations, respectively, were ascribed to them.1 More recently, the configuration of legionaminic acid was revised to L-glycero-D-galacto,2,3 and its C4-epimer was isolated from a lipopolysaccharide from Legionella pneumophila.4 Reliable identification of these sugars, including the determination of absolute configurations, requires authentic samples.Here, we report on the first synthesis of compounds of this class.By analogy with the preparation of N-acetylneuraminic acid,5 the condensation of 2,4-diacetamido-2,4,6-trideoxy-L-gulose 14, in which the functionalisation and configuration of C2–C6 correspond to those of C5–C9 in the target nonulosonic acids, with oxalacetic acid was used at the key step of the synthesis. L-Rhamnose, a readily available 6-deoxyhexose, served as the progenitor of 14.Since the introduction of an azido group (as a precursor of the acetamido function) is accompanied by inversion of configuration, azidation at the 2- and 4-positions in L-rhamnose would lead directly to the desirable configurations of C2 and C4. Hence, an additional inversion at C3 has to be performed to achieve the target L-gulo configuration.The prerequisite for successful nucleophilic substitution of axial O2 sulfonates is that the substituent at C1 is equatorial.6 Therefore, benzyl b-L-rhamnopyranoside 5 was thought to be the precursor of choice. It was prepared from 1,2-diol 1 by Bu2SnOmediated benzylation.7 As expected,8 the reaction occurred in a regio- and stereoselective manner, though it was accompanied by migration of benzoyl protecting groups (Scheme 1).As a result, a mixture of benzyl b-L-rhamnopyranoside dibenzoates 2–4 was obtained in a total yield of 90–95% with the 2:3:4 ratio approximately equal to 4:3:1. Debenzoylation of the mixture of 2–4 with NaOMe in methanol gave 5. The treatment of 5 with trimethyl orthoacetate in the presence of TsOH followed by acetylation of OH4 and hydrolytic opening of the orthoester ring in the resultant 2,3-orthoester yielded 2,4-diacetate 6.The reaction of 6 with triflic anhydride in the presence of pyridine led to triflate 7, which readily gave 3,4- anhydro-6-deoxyaltroside 8† on treatment with NaOMe in methanol. Compound 8 was the key intermediate having (i) a necessary configuration of C3, (ii) a free OH2 group required for the subsequent introduction of an azido group and (iii) a 3,4-epoxy function suitable for the introduction of the second azido group at the 4-position.As anticipated, the conversion of 8 to triflate 9 and subsequent reaction with NaN3 in DMF resulted in azide 10 in a high yield. A large J1,2 coupling constant of 7.6 Hz in the 1H NMR spectrum of 10 showed that the azido group was pseudo-equa- † Compound 8: [a]D +87° (c 2.6, CHCl3). 1H NMR (CDCl3) d: 1.47 (d, 3H, H6, J5,6 7.0 Hz), 3.04 (d, H4, J3,4 3.9 Hz), 3.44 (dd, H3, J2,3 1.7 Hz), 4.02 (t, H2, J1,2 1.6 Hz), 4.07 (q, H5), 4.54 (d, H1), 4.58, 4.89 (2d, 2H, CH2Ph, Jgem 11.9 Hz), 7.27–7.42 (m, 5H, Ph). torial. The opening of the epoxide ring in 10 with NaN3 in the presence of NH4Cl in boiling aqueous ethanol9 afforded diazide 11.Large J1,2 and small J2,3, J3,4 and J4,5 coupling constants in the 1H NMR spectrum of derived acetate 12‡ proved unambiguously that 11 had the b-gulo configuration. The chemical shifts ‡ Compound 12: 1H NMR (CDCl3) d: 1.38 (d, 3H, H6, J6,5 6.5 Hz), 2.05 (s, 3H, AcO), 3.48 (dd, H4, J4,5 1.7 Hz), 3.69 (dd, H2, J2,3 3.4 Hz), 4.03 (dq, H5), 4.69, 4.96 (2d, 2H, CH2Ph, Jgem 11.8 Hz), 4.79 (d, H1, J1,2 8.1 Hz), 5.33 (t, H3, J3,4 3.5 Hz), 7.30–7.42 (m, 5H, Ph).Scheme 1 Reagents and conditions: i, Bu2SnO, PhH, reflux; ii, BnBr, Bu4NBr, PhH, reflux; iii, MeONa, MeOH; iv, MeC(OMe)3, TsOH, MeCN; v, Ac2O, pyridine; vi, aqueous 80% AcOH; vii, Tf2O, pyridine, CH2Cl2, 0 °C; viii, NaN3, DMF, 20 °C; ix, NaN3, NH4Cl, EtOH–H2O, reflux; x, H2, Pd(OH)2 /C, MeOH; xi, Ac2O, MeOH; xii, oxalacetic acid, Na2B4O7, pH 10.5.O OH Me BzO BzO OH 1 O Me R3O R2O OR1 2 R1 = H, R2 = R3 = Bz i, ii 95% OBn 3 R1 = R3 = Bz, R2 = H 4 R1 = R2 = Bz, R3 = H iii 85% O Me HO HO OH 5 OBn iv, v, vi 74% O Me AcO RO OAc 6 R = H 7 R = Tf OBn vii iii 79% O Me OR 8 R = H 9 R = Tf OBn vii viii 83% O O Me 10 OBn O N3 ix 86% O Me R1 OR2 11 R1 = N3, R2 = H OBn 12 R1 = N3, R2 = Ac 13 R1 = NHAc, R2 = H x 94% R1 v x, xi 85% O Me NHAc OH 14 NHAc xii O AcHN R2 15 R1 = OH, R2 = H CO2H 16 R1 = H, R2 = OH R1 OH Me OH AcHN OHMendeleev Communications Electronic Version, Issue 3, 2000 (pp. 83–124) of H2–H4 in the 1H NMR spectrum of 12 demonstrated the location of the azido groups at C2 and C4. The consecutive reduction of the azido groups in 11 by hydrogenation over Pd(OH)2, N-acetylation, and removal of the protective benzyl group from 13 led to target sugar 14§ in a high yield.The condensation of 14 with oxalacetic acid in the presence of Na2B4O7 at pH 10.55 resulted in a mixture of nonulosonic acids 15 and 16. This mixture was isolated by anion-exchange chromatography (Dowex 1×8, 0.3 M formic acid) and separated by reversed-phase HPLC (C18, 0.05% TFA) to afford 15, [a]D 22 +15.4° (c 1.6, H2O), and 16, [a]D 22 –19.2° (c 1.7, H2O), in 18 and 20% yields, respectively. According to the 1H and 13C NMR spectral data,¶ both of the compounds were isolated as mixtures of anomers (a:b ª 1:19 and 1:8 for 15 and 16, respectively). However, the concentration § Compound 14: mp 144–146 °C (MeOH–diethyl ether), [a]D +6.3 ® +76° (c 1.6, MeOH). 1H NMR (D2O) d: 14a, 1.16 (d, 3H, H6, J5,6 6.6 Hz), 2.06, 2.07 (2s, 6H, 2MeCON), 3.91 (t, 1H, H3, J3,4 3.8 Hz), 3.95 (dd, 1H, H4, J4,5 1.7 Hz), 4.11 (t, 1H, H2, J2,3 3.5 Hz), 4.62 (dq, 1H, H5), 5.15 (d, 1H, H1, J1,2 4.0 Hz); 14b, 1.18 (d, 3H, H6, J5,6 6.5 Hz), 2.04, 2.08 (2s, 6H, 2MeCON), 3.85 (dd, 1H, H2, J2,3 3.2 Hz), 3.87 (dd, 1H, H4, J4,5 1.6 Hz), 3.94 (t, 1H, H3, J3,4 3.4 Hz), 4.29 (dq, 1H, H5), 4.94 (d, 1H, H1, J1,2 8.9 Hz). The ratio 14a:14b ª 1:5.¶ Spectral data (D2O, 303 K, pD 1.7, acetone as an internal standard: dH 2.225, dC 31.45). 15b: 1H NMR, d: 1.16 (d, 3H, H9, J8,9 6.2 Hz), 1.85 (dd, H3a, J3a,4 12.2 Hz, J3a,3e 13.1 Hz), 1.96, 2.00 (2s, 6H, 2MeCON), 2.30 (dd, H3e, J3e,4 4.8 Hz), 3.70 (t, H5, J4,5 10.2 Hz), 3.89 (quintet, H8, J7,8 6.4 Hz), 3.92 (ddd, H4), 3.93 (dd, H7, J6,7 2.0 Hz), 4.14 (dd, H6, J5,6 10.3 Hz). 13C NMR, d: 20.0 (C9), 23.3, 23.6 (2MeCON), 40.6 (C3), 54.3 (C5), 54.7 (C7), 68.5 (C4), 69.5 (C8), 73.1 (C6), 96.7 (C2), 174.3 (C1), 175.4, 175.5 (2MeCON). 16b: 1H NMR, d: 1.21 (d, 3H, H9, J8,9 5.7 Hz), 1.97, 2.01 (2s, 6H, 2MeCON), 2.13 (dd, H3a, J3a,4 3.3 Hz, J3a,3e 14.9 Hz), 2.18 (dd, H3e, J3e,4 2.9 Hz), 3.89 (dd, H5, J4,5 2.8 Hz, J5,6 10.6 Hz), 3.95 (m, H7), 3.96 (m, H8), 4.11 (m, H4), 4.48 (dd, H6, J6,7 1.3 Hz). 13C NMR, d: 19.9 (C9), 23.0, 23.1 (2MeCON), 37.6 (C3), 49.9 (C5), 54.8 (C7), 66.9 (C4), 68.6 (C6), 69.3 (C8), 96.2 (C2), 174.4 (C1), 174.6, 175.2 (2MeCON). 16a: 1H NMR, d: 1.28 (d, 3H, H9, J8,9 6.3 Hz), 1.94 (dd, H3a, J3a,4 2.7 Hz, J3a,3e 14.4 Hz), 1.96, 2.04 (2s, 6H, 2MeCON), 2.65 (dd, H3e, J3e,4 3.6 Hz), 3.84 (dd, H5, J4,5 2.7 Hz, J5,6 10.5 Hz), 3.89 (H7), 4.08 (m, 2H, H4, H8), 4.47 (dd, H6). 13C NMR, d: 19.9 (C9), 40.0 (C3), 50.1 (C5), 54.8 (C7), 66.6 (C4), 69.8 (C8), 72.7 (C6). of 15a was too small for the reliable assignment of the NMR spectra. The J3a,4, J4,5 and J5,6 coupling constant values in the 1H NMR spectra of 15 and 16 clearly indicated that the substituents at C4, C5, and C6 in 15 and at C5 and C6 in 16 were equatorial, whereas the hydroxy group at C4 in 16 was axial.The 13C NMR spectrum showed upfield shifts by 1.6–4.5 ppm for the C3–C6 signals in 16b, as compared to the corresponding signals in 15b (cf. similar data for N-acetylneuraminic acid and its C4-epimer10). The most marked difference between 16a and 16b was observed for the chemical shifts of C6 (d 72.7 and 68.6, respectively).This provides a basis for the determination of the anomeric configuration of nonulosonic acids from this class. A comparison of the synthetic and natural 5,7-diacetamido- 3,5,7,9-tetradeoxynonulosonic acids will be published elsewhere. This work was supported by the Russian Foundation for Basic Research (grant no. 99-03-32955) and the Deutsche Forschungsgemeinschaft (grant no. 436 RUS 113/314/0). References 1 Yu. A. Knirel and N. K. Kochetkov, FEMS Microbiol. Rev., 1987, 46, 381. 2 P. Edebrink, P.-E. Jansson, J. Bogwald and J. Hoffman, Carbohydr. Res., 1996, 287, 225. 3 Yu. A. Knirel, J. H. Helbig and U. Zähringer, Carbohydr. Res., 1996, 283, 129. 4 Yu. A. Knirel, H. Moll, J. H. Helbig and U. Zähringer, Carbohydr. Res., 1997, 304, 77. 5 M. J. How, M. D. A. Halford and M. Stacey, Carbohydr. Res., 1969, 11, 313. 6 W. Karpiesiuk, A. Banaszek and A. Zamojski, Carbohydr. Res., 1989, 186, 156. 7 N. K. Kochetkov, N. E. Byramova, Yu. E. Tsvetkov and L. V. Backinowsky, Tetrahedron, 1986, 41, 3363. 8 V. K. Srivastava and C. Schuerch, Tetrahedron Lett., 1979, 35, 3269. 9 Y. Ali and A. C. Richardson, Carbohydr. Res., 1967, 5, 441. 10 F. Baumberger and A. Vasella, Helv. Chim. Acta, 1986, 69, 1205. Received: 24th February 2000; Com. 00/1613

ISSN:0959-9436     

出版商:RSC    

年代:2000

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 2, 2000 (pp. 83–124) Dolichyl phosphate derivatives with a fluorescent label at an internal isoprene unit Natalia Ya. Grigorieva,*a Olga A. Pinsker,a Sergei D. Maltsev,a Leonid L. Danilov,a Vladimir N. Shibaeva and Mark J. Jedrzejasb a N. D. Zelinsky Institute of Organic Chemistry, Russian Academy of Sciences, 117913 Moscow, Russian Federation. Fax: + 7 095 135 5328; e-mail: ves@cacr.ioc.ac.ru b Department of Microbiology, The University of Alabama at Birmingham, Birmingham, Alabama 35294, USA 10.1070/MC2000v010n03ABEH001257 A general approach based on directed aldol condensation followed by reductive amination with fluorescent amines and phosphorylation was developed and illustrated by the synthesis of two dolichyl phosphate derivatives with the 1-aminonaphthalene fluorophore at the g-isoprene unit of the chain.The carbohydrate chains of glycoproteins are involved in many cellular activities1 and are often structurally changed in diseases.2 The biosynthesis of Asn-linked oligosaccharides on eukaryotic glycoproteins, the O-glycosylation of fungal glycoproteins and the formation of glycosylphosphatidylinositol anchors present in numerous membrane proteins require the participation of dolichyl phosphate (Dol-P), a derivative of a linear long-chain polyisoprene alcohol, and its glycosylated forms as key intermediates.3 The fluorescence methodology seems to be promising for studies of molecular mechanisms of the interaction of Dol-P and its glycosylated derivatives with biological membrane components.The synthesis of Dol-P derivatives with fluorescent labels at the w-end of the chain was reported recently.4 Here, we describe a procedure for the incorporation of fluorophores in an internal isoprene unit of Dol-P. The methodology is based on the preferential formation of thermodynamically more stable (E)-isomers of a,b-disubstituted acroleins in directed aldol condensation, which was found previously5 and successfully used in the syntheses of dolichols, polyprenols and their analogues.6,7 The key step in the synthetic strategy was the construction of dolichyl-like (E)-acroleins through the cross-condensation of two appropriate aldehydes, one of which was used as an aldimine. The resulting (E)-acrolein was subjected to reductive amination with a fluorescent amine, and the amino alcohol formed was phosphorylated.This approach is illustrated by Scheme 1 where the synthesis of two Dol-P analogues (5a,b) with a 1-aminonaphthalene label at the g-isoprene unit is shown. The aldehydoacetate 14 was used as an aldehyde component to build the key intermediates, (E)-acroleins 3a,b. Condensation of 1 with aldimine 2a8 deprotonated by treatment with LDA at –70 °C leads to acetoxyacrolein 3a (40%, after flash chromatography on SiO2) which contains ~2% of the (2Z)-isomer (1H NMR data).In a similar manner, (E)-acetoxyacrolein 3b [~3% of the (2Z)-isomer, according to 1H NMR data] was obtained in ~21% yield through the condensation of 1 with aldimine 2b prepared in five synthetic steps (cf.ref. 9) from a mixture of polyprenols WT2Cn-OH isolated from birch tree10 with C35- and C40-prenols (n = 4, 5) as main components.† The reductive amination of 3a,b with 1-aminonaphthalene and NaBH3CN (cf. ref. 11) led to amino alcohols 4a,b in 37 and ~20% yields (after flash chromatography on SiO2), respectively.‡ To prepare fluorescent Dol-P derivatives, 4a,b were subjected to phosphorylation with Bu4N·H2PO4/CCl3CN12 under conditions essentially similar to those described for the synthesis of w-labeled analogues.4 The resulting phosphates 5a,b§ were isolated as ammonium salts after purification by ion-exchange chromatography (DE-52, AcO–) in 18 and 28% yields, respectively.As expected, they exhibited intense fluorescence with an excita- † 3a: 1H NMR (CDCl3) d: 0.92 (d, 3H, MeC-10, J 6.5 Hz), 1.05–1.50 (m, 5H, H2C-9,11, HC-10), 1.62 (s, 12H, cis-Me), 1.71 (s, 6H, trans- MeC-6,16'), 2.0 (m, 16H, CH2C=C), 2.01 (s, 3H, MeCO), 2.25 (m, 4H, H2C-5,1'), 2.45 (td, 2H, H2C-4, J1 = J2 = 7 Hz), 4.09 (td, 2H, H2C-12, J1 7 Hz, J2 2.5 Hz), 5.12 (m, 5H, HC=C), 6.43 (t, 1H, HC-3, J 7 Hz), 9.34 (s, 1H, HC-1). 13C NMR, d: 15.9 (cis-Me), 17.5 (cis-MeC-16'), 19.3 (MeC-10), 20.9 (MeCO), 23.1 (MeC-6), 24.2 (C-1'), 25.5 (trans-MeC- 16'), 25.2, 25.6, 26.0, 26.7, 27.0, 27.3 (CH2CH=C), 29.5 (C-10), 30.6 (C-5), 35.4 (C-9), 37.1 (C-11), 39.6 [H2CC(Me)=C of (E)-units], 62.8 (C-12), 123.2, 124.2, 124.35, 126.6 (HC=C), 131.1, (C-16'), 133.2, 134.8, 134.9, 136.0 (MeC=C), 143.3 (C-2), 154.3 (C-3), 171.0 (COMe), 194.9 (C-1). 3b: 1H NMR (CDCl3) d: 0.93 (d, 3H, MeC-10, J 6.5 Hz), 1.10–1.60 (m, 5H, H2C-9,11, HC-10), 1.59 (s, 9H, cis-Me), 1.68 (s, 19.5H, trans- MeC), 2.0 (m, 33H, CH2C=C, MeCO), 2.20 (m, 4H, H2C-5,1'), 2.45 (td, 2H, H2C-4, J1 = J2 = 7 Hz), 4.11 (td, 2H, H2C-12, J1 7 Hz, J2 2.5 Hz), 5.15 (m, 8.5H, HC=C), 6.45 (t, 1H, HC-3, J 7 Hz), 9.35 [s, 0.97H, HC-1 of (E)-isomer], 10.10 [s, 0.03H, HC-1 of (Z)-isomer]. 13C NMR, d: 15.9 (cis-Me), 17.6 (cis-Me of w-terminal unit), 19.3 (MeC-10), 20.9 (MeCO), 23.1 (MeC-6), 23.4 (trans-Me), 24.4 (C-1'), 25.6 (trans-Me of w-terminal unit), 25.2, 26.3, 26.6, 26.7, 26.8, 27.0, 27.3 (CH2CH=C), 29.4 (C-10), 30.5, 31.9, 32.1 [H2CC(Me)=C of (Z)-units], 35.3 (C-9), 37.1 (C-11), 39.7 [H2CC(Me)=C of (E)-units], 62.8 (C-12), 124.1, 124.15, 124.3, 124.4, 124.95, 126.6 (HC=C), 131.15, (MeC=C of w-terminal unit), 133.2, 134.8, 135.1, 135.3, 135.95 (MeC=C), 143.6 (C-2), 154.45 (C-3), 171.0 (COMe), 194.9 (C-1).Scheme 1 Reagents and conditions: i, LDA, Et2O–hexane (1:1), –10 °C ® ® 0 °C, 30 min; ii, 1, Et2O, –70 °C (2 h) ® –20 °C (2.5 h); iii, 3.5% aq. HCl, 20 °C, 3 h, then Ac2O/Py–DMAP, 20 °C, 4 h; iv, 1-aminonaphthalene, NaBH3CN/AcOH–MeOH (for 4a) or MeOH–Et2O (for 4b), 20 °C, 92 h, then HCl to pH 2, 20 °C, 10 min, then NaOH/MeOH, 20 °C, 15 min; v, Bu4N·H2PO4, CCl3CN/CH2Cl2, 20 °C, 48 h, chromatography on DE-52 (AcO–).O OAc H NBut m n 1 H O m n OAc H NH m n OR 2a,b i, ii, iii iv 3a,b 4a,b R = H 5a,b R = PO3(NH4)2 v a m = 4, n = 0 b m = 3, n = 4, 5 1 2 3 4 5 6 7 8 9 10 11 12 1' 2' 3' 4' 5' 1 2 3 4 5 6 7 8 9 10 11 12Mendeleev Communications Electronic Version, Issue 3, 2000 (pp. 83–124) tion/emission maximum at 340/410 nm (in n-heptane–2-propanol, 4:1) and are quite similar in this respect to the derivatives of Dol-P with a 1-naphthylamine residue at the w-end of the chain4 (see ref. 13 for the excitation and emission spectra). In preliminary biochemical tests, phosphate 5a was found to serve as a substrate for recombinant yeast Dol-P-mannose synthase (EC 2.4.1.83) isolated from Escherichia coli.14 It is likely that the described approach to the incorporation of a fluorescent label at an internal isoprene unit of Dol-P can be employed for the preparation of a wide range of related derivatives starting from the corresponding aldehydes and aldimines ‡ 4a: 1H NMR (CDCl3) d: 0.91 (d, 3H, MeC-3, J 6.6 Hz), 1.10–1.60 (m, 5H, H2C-2,4, HC-3), 1.65 (s, 12H, cis-Me), 1.72 (s, 6H, trans-Me), 2.05 (m, 18H, CH2C=C), 2.27 (m, 4H, H2C-9,12), 3.20 (s, 1H, NH), 3.63 (dt, 2H, H2C-1, J1 1.5 Hz, J2 6.6 Hz), 3.87 (s, 2H, CH2N), 5.14 (m, 5H, HC=C), 5.45 (t, 1H, HC-10, J 6.7 Hz). 13C NMR, d: 16.0, 16.1 (cis-Me), 17.6 (cis-MeC-27), 19.5 (MeC-3), 23.4 (MeC-7), 25.1, 25.3, 26.2, 26.6, 26.7, 27.2 (CH2CH=C), 25.7 (trans-MeC-27), 29.2, 29.4 (C-3,12), 31.8 (C-8), 37.4 (C-4), 39.7 [CH2C(Me)=C of (E)-units], 39.8 (C-2), 50.0 (CN), 61.1 (C-1), 123.8, 124.2, 124.4, 125.5, 125.55, 127.1 (HC=C), 131.2 (C-27), 134.2, 134.6, 135.0, 135.4, 136.2 (MeC=C).Signals for 1-aminonaphthalene fragment are not shown. 4b: 1H NMR (CDCl3) d: 0.90 (d, 3H, MeC-3, J 6.5 Hz), 0.95–1.55 (m, 5H, H2C-2,4, HC-3), 1.57 (s, 9H, cis-Me), 1.65 (s, 19.5H, trans-Me), 2.01 (m, 32H, CH2C=C), 2.20 (m, 4H, H2C-9,12), 3.20 (s, 1H, NH), 3.70 (dt, 2H, H2C-1, J1 1.5 Hz, J2 6.6 Hz), 3.87 (s, 2H, H2CN), 5.13 (m, 8.5H, HC=C), 5.53 (t, 1H, HC-10, J 6.7 Hz). 13C NMR, d: 16.0 (cis- Me), 17.7 (cis-Me of w-terminal unit), 19.5 (MeC-3), 23.4 (trans-Me), 25.7 (trans-Me of w-terminal unit), 25.1, 26.2, 26.4, 26.6, 26.7, 27.0 (CH2CH=C), 29.1 (C-3), 29.6 (C-12), 31.8, 31.9, 32.2 [CH2C(Me)=C of (Z)-units], 37.4 (C-4), 39.7 [CH2C(Me)=C of (E)-units], 39.8 (C-2), 50.0 (CN), 61.0 (C-1), 123.8, 124.1, 124.2, 124.3, 124.7, 124.9, 125.6, 125.7, 127.1, 128.8 (HC=C), 131.2 (MeC=C of w-terminal unit), 134.2, 134.6, 135.15, 135.2, 135.3, 135.7, 136.0 (MeC=C).Signals for 1-aminonaphthalene fragment are not shown.§ 5a: UV, lmax/nm: 252, 335; emax: 18000, 5500. ESI–MS, m/z: 716 [M (free acid) – H]–. 5b: UV, lmax/nm: 252, 335; emax: 18500, 5000; 31P NMR (CD3OD– CDCl3, 1:2) d: 2.25. 1H and 13C NMR spectra of 5a,b are similar to the spectra of 4a,b except for signals of HC-1 (3.90 and 3.85 for 5a and 5b, respectively) and C-1 (64.0 and 63.8 for 5a and 5b, respectively). available from different terpenols.7 The use of these Dol-P analogues is of paramount importance for studies of the interaction of Dol-P with biological membrane components.This work was supported by the U.S. Civilian Research and Development Foundation (grant no. RN1-404). References 1 A. Varki, Glycobiology, 1993, 3, 97. 2 I. Brockhausen, J. S. Schutzbach and W. Kuhns, Acta Anatom., 1998, 161, 36. 3 F. W. Hemming, in New Comprehensive Biochemistry, eds. J. Montreuil, H. Schachter and J. F. G. Vliegenthart, Elsevier, Amsterdam, 1995, vol. 29a, p. 127. 4 V. N. Shibaev, V. V. Veselovsky, A. V. Lozanova, S. D. Maltsev, L. L. Danilov, W. T. Forsee, J. Xing, H. C. Cheung and M. J. Jedrzejas, Bioorg. Med. Chem. Lett., 2000, 10, 189. 5 N. Ya. Grigorieva, E. P. Prokof’ev and A.V. Semenovsky, Dokl. Akad. Nauk SSSR, 1979, 245, 366 [Dokl. Chem. (Engl. Transl.), 1979, 245, 112]. 6 N. Ya. Grigorieva, V. V. Veselovsky and A. M. Moiseenkov, Khim.- Farm. Zh., 1987, 21, 845 (in Russian). 7 N. Ya. Grigorieva and O. A. Pinsker, Usp. Khim., 1994, 63, 177 (Russ. Chem. Rev., 1994, 63, 169). 8 N. Ya. Grigorieva, I. M. Avrutov and A. V. Semenovsky, Tetrahedron Lett., 1983, 24, 5531. 9 N. Ya. Grigorieva, O. A. Pinsker, E. D. Daeva and A. M. Moiseenkov, Izv. Akad. Nauk SSSR, Ser. Khim., 1991, 2325 (Bull. Acad. Sci. USSR, Div. Chem. Sci., 1991, 40, 2037). 10 A. R. Wellburn and F. W. Hemming, Nature, 1966, 212, 1364. 11 A. Schepartz and R. Breslow, J. Am. Chem. Soc., 1987, 109, 1814. 12 L. L. Danilov, T. N. Druzhinina, N. A. Kalinchuk, S. D. Maltsev and V. N. Shibaev, Chem. Phys. Lipids, 1989, 51, 191. 13 J. Xing, W. T. Forsee, E. Lamani, S. D. Maltsev, L. L. Danilov, V. N. Shibaev, J. S. Schutzbach, H. C. Cheung and M. J. Jedrzejas, Biochemistry, submitted. 14 J. S. Schutzbach, J. W. Zimmerman and W. T. Forsee, J. Biol. Chem., 1993, 268, 24190. Received: 28th December 1999; Com. 99/1583

ISSN:0959-9436     

出版商:RSC    

年代:2000

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 2, 2000 (pp. 83–124) Kinetics of diffusionally induced gelation and ordered nanostructure formation in surfactant–polyelectrolyte complexes formed at water/water emulsion type interfaces Valery G. Babak,*a Elena A. Merkovich,a,b Leonid S. Galbraikh,b Eleonora V. Shtykovac and Margueritte Rinaudod a A. N. Nesmeyanov Institute of Organoelement Compounds, Russian Academy of Sciences, 117813 Moscow, Russian Federation. E-mail: babak@ineos.ac.ru b A.N. Kosygin Moscow State Textile Academy, 117918 Moscow, Russian Federation c A. V. Shubnikov Institute of Crystallography, Russian Academy of Sciences, 117333 Moscow, Russian Federation d CERMAV–CNRS, 38041 Grenoble, France 10.1070/MC2000v010n03ABEH001227 A kinetic approach to study the physical gel formation from insoluble surfactant–polyelectrolyte complexes has been developed in connection with the problem of microencapsulation. Ionic surfactants in mixed aqueous solutions with oppositely charged polyelectrolytes may form insoluble surfactant–polyelectrolyte complexes (SPECs) and self-organise in micelle-like clusters bound to polyelectrolyte chains at a concentration referred to as a critical aggregation concentration (c.a.c.).This c.a.c. is about two orders of magnitude smaller than the critical micelle concentration (c.m.c.) of the surfactants in their individual solutions. The binding of ionic surfactants to oppositely charged polyelectrolytes has been reported as cooperative,1 whereas the SPECs are believed to be stoichiometric relative to ionic functional groups of polyelectrolyte macroions.At a relatively high surfactant concentration Cs and ratio j = Cs /CPE in a mixed solution, the hydrophobic interactions between surfactant molecules may produce the over-stoichiometric binding of surfactants to the complexes and the inversion of the electric charge of these complexes.2 The SPECs condense as colloidal or microscopic disperse-phase particles which precipitate in more or less dense sediments depending on the parameters Cs, CPE, j, electric charge density and rigidity of polyelectrolyte chains, as well as on the mechanical power applied to stir or sonicate the mixed solution.Small angle X-ray scattering (SAXS) measurements have evidenced an ordered nanostructure in the SPECs sediments, which was attributed to the side-by-side hydrophobic interactions between bound surfactants.2–5 The more pronounced ordering of clusters and even the formation of regular and crystalline nanostructures with cubic or hexagonal symmetries have been found by SAXS in the complexes of covalently cross-linked polyelectrolyte gels and oppositely charged surfactants.6–14 The formation of SPECs by mixing two aqueous solutions seems to be improbable.The diffusion of surfactant molecules from the outside solution into the macroscopic samples of these gels makes possible and enhances the formation of these super-regular nanostructures on the macroscopic level. On the other hand, a high cross-linker density of the polyelectrolyte network chains could hinder the formation of a highly ordered structure of surfactant molecules in the complexes.10 The degree of order in the complexes also decreases with increasing electrolyte concentration in the solution.14 The aims of this work were (i) to study the kinetics of the physical gel formation in the production of microcapsules from SPECs by the frontal diffusion of components [oppositely charged cationic polyelectrolyte chitosan and anionic surfactant sodium dodecyl sulfate (SDS)] through water/water emulsion type interfaces and (ii) to estimate the structure-mechanical properties of these physical gels.The physical gelation of polyelectrolyte solutions by the action of surfactants can find practical applications in the production of pH-, electrolyte- and thermosensitive hydrogel beads and microbeads for medicine, cosmetics, food industry, etc.15 Naturally occurring polyelectrolytes (such as polysaccharides and proteins) and surfactants (fatty acids and phospholipids) may be used for this purpose.Avoiding cross-linking reagents in the formulation reduces the toxicity of carriers made on the basis of these physical gels. Chitosan (Ch), MW = 550.000, DA=0.12;16,17 and the anionic surfactant SDS (Fluka) were used.All solutions were prepared in an acetate buffer with pH 3.6 and ionic strength 0.05 M. Capsules with a diameter of ~1 mm were formed by the dropwise addition of a chitosan solution to an SDS solution. The capsule membranes were formed via the formation of insoluble SPECs as a result of the frontal diffusion of surfactant molecules through the water/water emulsion type interface inside the chitosan solution in the drop.Optical studies (the measurement of capsule volume, contraction and deformation) were performed using a video camera attached to a microscope and recorded with a video printer (Figure 1). Plane hydrogel layers with the thickness d = 10–300 mm and the area A = 71 cm2 for mechanical measurements in an ambient environment were formed by carefully placing a surfactant solution over the chitosan solution (to avoid mixing). Variable parameters were the concentrations C and volumes V of SDS and chitosan solutions.Scattering measurements were performed on an AMUR-K small-angle X-ray scattering diffractometer (Institute of Crystallography, Russian Academy of Sciences) with a linear positionsensitive detector and a single-crystal monochromator at the wavelength l = 1.542 Å.The Kratky-type geometry was used with a sample-to-detector distance of 673 mm and a sample slit width of 0.2 mm to cover the range of momentum transfer 0.012 < q < 0.55 Å–1 (here, q = 4psinq/l, where 2q is the scattering angle). The windows of the sample holder were made from poly(ethylene terephthalate) with a thickness of 0.01 mm; the sample thickness was about 1 mm.The latter varied along the length of the holder window (10 mm) so that the irradiated volume cannot be estimated and no absolute calibration was possible. The data were normalised to the intensity of the incident beam and corrected for the detector response according to standard procedures.18 To analyse the structure of the gel–surfactant complexes, mean long-range order dimensions L in the systems were estimated from the Bragg peaks in the SAXS patterns by the Scherrer formula19 L = l/bscosq = 19.2 nm, where bs = 0.33 nm is the full 5 min 15 min 30 min 60 min d Figure 1 The thickness d of the gel bead wall visibly grows in time.CCh = 0.1 base-mol dm–3; CSDS = 0.05 mol dm–3.Mendeleev Communications Electronic Version, Issue 3, 2000 (pp. 83–124) width at a half-maximum intensity of the Bragg peak observed at the mean scattering angle 2q. From SAXS curves, the radii of interaction rm = 0.16p2l/bs = 30.4 nm and the degrees of disorder in the system D/ =p–1(bs /l)0.5 = 0.14 were also calculated (here, = 2p/qmax is the characteristic size of the ordered elements of the gel–surfactant complex, D is the mean-square deviation of distances between the neighbouring molecules).The elasticity modulus E and the thickness d of gel-like layers with the thickness 10–300 mm were estimated in the Hertz approximation (a hard sphere of radius 1 mm in contact with an elastic body) using an original device that allowed us to record the force (f)–deformation (Dh) curves with an accuracy of ~1 mN or ~1 mm, respectively.15 From the fp–Dh curves, using the Hertz formula fp ER1/2Dh3/2,20 we estimated the elasticity modulus as E = 104–105 Pa.21 We found that the kinetics of formation of a gel-like wall in the beginning of the process (the first 30–60 min) is limited by the diffusion of surfactant molecules into the chitosan solution in the drop, and we have d ~ t1/2 (Figure 2).Later on, the limiting factor is the mass of chitosan (if the surfactant is in an excess). From the expression for the frontal diffusion of surfactant molecules d (2Dst)1/2, we obtained Ds 10–12 m2 s–1 for curves 2 and 3 in Figure 2. Obviously, the final thickness d of the gellike layer is proportional to the chitosan mass in the solution.A twofold increase in the bulk chitosan concentration (curve 1) gave Ds 10–13 m2 s–1, which was consistent with a ~10-fold increase in the viscosity of the chitosan solution (h 0.5 or 7 Pa s for CCh = 0.05 or 0.1 base-mol dm–1, respectively).21 Elemental analysis showed that the ratio Z = CSDS/CCh [mol/ base-mol] in the SPECs is close to unity,21 i.e., the complexes are stoichiometric in terms of the electric charge of the polycations.From Figure 2, one can estimate the volume fraction of the solid phase in the gel as where MCh = 165 g base-mol–1, MSDS = 249 g mol–1, rgel 1.1×103 kg m–3, Vgel = Ad. We found that j 0.3, which corresponds to a ~30% solution. The SAXS measurements (Figure 3) show the existence of an ordered nanostructure in these gels, which is manifested by the appearance of characteristic peaks with the d-spacing at 1.84 and 3.80 nm–1 (curves 1 and 2 in Figure 3 and Table 1).The relative positions of the main and secondary peaks are indicative of the formation of fragments with a lamellar structure. The SPEC sediment (curve 3) also exhibited the formation of an ordered structure; however, the maximum is rather weekly pronounced. Chitosan physical gel beads obtained by dropping the polysaccharide solution into the NaOH solutions without a surfactant (Figure 3, curve 3) display practically no central scattering (which would indicate the presence of clusters, particles or micelles) and no characteristic peaks (which would appear because of ordering the gel structure).Thus, the formation of an ordered nanostructure is typical of SPEC gels formed by the frontal diffusion of surfactant molecules in the polyelectrolyte solution. This study demonstrates the possibility to control the preparation of capsules and microcapsules from the physical gel of chitosan–surfactant complexes. References 1 Y. V. Khandurina, A.T. Dembo, V. B. Rogacheva, A. B. Zezin and V. A. Kabanov, Polym. Sci., 1994, 36, 189. 2 L. Chen, S. Yu, Y. Kagami, J. Gong and Y. Osada, Macromolecules, 1998, 31, 787. 3 E. A. Ponomarenko, A. Waddon, D. Tirrell and W. J. MacKnight, Langmuir, 1996, 12, 2169. 4 M. Gawronski, G. Aguirre, H. Conrad, T. Springer and K.-P. Stahmann, Macromolecules, 1996, 29, 1516. 5 A. F. Thünemann and K.Lochhaas, Langmuir, 1999, 15, 4867. 6 Y. Khandurina, V. Alexeev, G. A. Evmenenko, A. Dembo, V. B. Rogacheva and A. B. Zezin, J. Phys. II Fr., 1995, 36, 337. 7 H. Okuzaki and Y. Osada, Macromolecules, 1995, 28, 380. 8 E. L. Sokolov, F. Yeh, A. Khokhlov and B. Chu, Langmuir, 1996, 12, 6229. 9 E. Sokolov, F. Yeh, A. Khokhlov, V. Y. Grinberg and B. Chu, J. Phys. Chem. B, 1998, 102, 7091. 10 S. Zhou, C. Burger, F. Yeh and B. Chu, Macromolecules, 1998, 31, 8157. 11 P. Hansson, Langmuir, 1998, 14, 4059. 12 P. Hansson, Langmuir, 1998, 14, 2269. 13 L. M. Bronstein, O. A. Platonova, A. N. Yakunin, I. M. Yanovskaya, P. M. Valetsky, A. T. Dembo, E. E. Makhaeva, A. V. Mironov and A. R. Khokhlov, Langmuir, 1998, 14, 252. 14 A. V. Mironov, S. G. Starodoubtsev, A. R.Khokhlov, A. T. Dembo and A. N. Yakunin, Macromolecules, 1998, 31, 7698. 15 V. G. Babak and E. A. Skotnikova, in Proceedings of International Conference ‘Lipid and Surfactant Dispersed Systems. Fundamentals, Design, Formulation, Production’, Moscow, 1999. 16 V. Babak, I. Lukina, G. Vikhoreva, J. Desbrière and M. Rinaudo, Colloids Surf., A, 1999, 147, 139. 17 J. Desbrières, M. Rinaudo, V.Babak and G. Vikhoreva, Polym. Bull., 1997, 39, 209. 18 L. A. Feigin and D. I. Svergun, Structure Analysis by Small-Angle X-ray and Neutron Scattering, Plenum Press, New York, 1987. 19 B. K. Vainshtein, Diffraction of X-rays by Chain Molecules, Elsevier, Amsterdam, 1966, p. 203. 20 D. C. Andrei, B. J. Briscoe, P. F. Luckham and D. R. Williams, J. Chem. Phys., 1996, 93, 960. 21 V. G. Babak, E. A. Merkovich, L. S. Galbraikh and M. Rinaudo, Journal of Microencapsulation, in preparation. d d d 200 150 100 50 0 50 100 150 t1/2 d/mm 3 2 1 Figure 2 The thickness d of gel-like layers as a function of time (t) for CSDS = 0.1mol dm–3 and VSDS = 50 cm3: (1) CCh = 0.1 base-mole dm–3, VCh = = 10 cm3; (2) CCh = 0.05 base-mole dm–3, VCh = 10 cm3; (3) CCh = = 0.05 base-mole dm–3, VCh = 20 cm3 (two sets of experimental points illustrate the reproducibility of measurements). = ~ 100 10 1 0 1 2 3 4 5 6 I(q) q/nm–1 1 2 3 Figure 3 X-ray intensity profiles for chitosan–SDS complexes: (1) chitosan gel beads in SDS solutions; (2) sedimented chitosan–SDS complex; (3) chitosan gel beads in NaOH solutions. CCh = 0.1 base-mol dm–3, CSDS = = 0.1 mol dm–3 (curve 1) and CSDS = 10–3 mol dm–3 (curve 2). = ~ = ~ = ~ = ~ j = CChMCh DSrgel 1 + ZDS VCh Vgel MSDS MCh , = ~ = ~ = ~ Received: 11th November 1999; Com. 99/1555

ISSN:0959-9436     

出版商:RSC    

年代:2000

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 2, 2000 (pp. 83–124) Synthesis and electrochemical properties of N-isocyanurate-substituted aziridino[1,6][60]fullerene, an unusual product of cycloaddition to the 5,6-junction of fullerene Oleg G. Sinyashin,* Irina P. Romanova, Gulshat G. Yusupova, Valery I. Kovalenko, Vitaly V. Yanilkin and Nail M. Azancheev A. E. Arbuzov Institute of Organic and Physical Chemistry, Kazan Scientific Centre of the Russian Academy of Sciences, 420083 Kazan, Russian Federation.Fax: +7 8432 75 2253; e-mail: oleg@iopc.kcn.ru 10.1070/MC2000v010n03ABEH001202 The main product of the cycloaddition of 1,3-diallyl-5-(5'-azidopentyl)-1,3,5-triazine-2,4,6-(1H,3H,5H)-trione to C60 at 100 °C is an N-isocyanurate-substituted aziridino[1,6][60]fullerene. The reactions of fullerenes with organic azides result in various products. It is well known1,2 that N-substituted triazolino[1,2]- [60]fullerenes are formed at the first stage of the reactions of azides with C60.The subsequent thermal elimination of N2 from triazolinofullerenes led to azahomo[60]fullerenes (the structures with an open 5,6-junction of fullerene) as major products and to aziridino[1,2][60]fullerenes (the product of addition to a closed 6,6-junction of fullerene) as minor products.Moreover, it was suggested1 that aziridino[1,6][60]fullerenes at the 5,6-junction of C60 are intermediate products of dissociation without opening the fullerene framework. This formation of monoadducts was proved by the isolation of triazolino[1,2]- and aziridino[ 1,2][60]fullerenes, as well as azahomo[60]fullerenes, from reaction mixtures.The formation of aziridino[1,6][60]fullerenes is still an open question. Theoretically, these structures can exist and even be stable. Banks et al.3 have reported on the isolation of aziridino[1,6][60]fullerene. However, Smith et al.4 and Schick et al.5 have cast doubt on these results. In this study, we examined the reaction of an N-isocyanurate-substituted azide with [60]fullerene and isolated a product, which was characterized as the 5,6-closed product of cycloaddition of the azide to C60 using spectral methods.Earlier,6 we found that the main product of the cycloaddition reactions of 1,3-diallyl-5-(5'-azidopentyl)-1,3,5-triazine-2,4,6- (1H,3H,5H)-trione 1 to C60 in o-dichlorobenzene at 180 °C was 1,3-diallyl-5-[5'-(azahomo[60]fullereno)pentyl]-1,3,5-triazine- 2,4,6-(1H,3H,5H)-trione 3, which exhibited Rf = 0.59 (eluent: toluene–diethyl ether, 10:1). In addition to 3, products 2 and 4 were also isolated from the reaction mixture by column chromatography in low yields.We supposed6 that compound 4 (Rf = 0.1) was the diadduct of azide 1 and [60]fullerene.The spectral characteristics of product 2 (Rf = 0.75) were not reported previously6 because of a very low yield. According to the value of Rf , we suppose this product to be a low-polarity monoadduct of [60]fullerene and azide 1. To obtain compound 2 in amounts sufficient for a structural study by spectral methods, the conditions of the reaction of azide 1 with C60 were changed (o-dichlorobenzene as a solvent, 100 °C).The reaction mixture was heated for 4 h and chromatographed on silica gel. The unreacted fullerene (13% of the initial amount) and products 2 (23% yield) and 3 (1–2% yield) were isolated. After the removal of the eluent in a vacuum, product 2 was stirred in diethyl ether and dried in a vacuum at 50 °C for 2 h. The elemental analysis of compound 2† showed that it is a monoadduct of C60 and azide 1.The 13C NMR spectrum† of this monoadduct exhibited 31 signals in the region typical of sp2 carbon atoms of fullerene derivatives (d between 130 and 150 ppm). The intensities of 4 signals corresponded to one carbon atom, and the intensities of 27 signals corresponded to two carbon atoms. An additional signal at d 120.21 ppm was observed in the spectrum, the intensity of which corresponded to two carbon atoms.Earlier,6 the signal at d 120 ppm was not observed in the 13C NMR spectra of compound 3 and azahomo[60]fullerenes. 4 If this signal is due to the sp3 carbon of fullerene, such a set of signals corresponds to a monoadduct at the 6,6-junction or closed 5,6-junction of fullerene.7 Note that for known aziridino[ 1,2][60]fullerenes the sp3 carbon signals were usually detected at d 75 and 85 ppm.4 Banks et al.3 observed a signal at 104.2 ppm in the 13C NMR spectrum of aziridino[1,6][60]fullerene.This signal was ascribed to sp3 carbons of fullerene. † Compound 2: 1H NMR (400 MHz, CDCl3) d: 4.50 (d, 4H, CH2 7,10, 3JHH 6.5 Hz), 5.88 (ddt, 2H, CH2 8,11, 3JHH 9cis 10.9 Hz, 3JHH 9trans 18.0 Hz), 5.24 (d, 2H, CHtrans 9,12 ), 5.32 (d, 2H, CHcis 9,12), 4.00 (m, AA'XX', 2H, CH2 13, 3JHH 7.4 Hz), 1.84 (m, 2H, CH2 14), 1.79 (m, 2H, CH2 15), 2.07 (m, 2H, CH2 16), 3.81 (m, 2H, CH2 17, 3JHH 7.0 Hz). 13C NMR (100.62 MHz, CDCl3) d: 148.69 (s, 2C, C2,4), 148.42 (s, 1C, C6), 44.9 (C7,10), 130.89 (dm, C8,11, 1JCH 156.6 Hz), 118.99 (tm, C9,12, 1JCH 158.2 Hz), 42.9 (C13), 27.6 (C14), 24.4 (C15), 28.9 (C16), 51.3 (C17); C60N: 120.21 (2C), 133.67 (2C), 135.77 (2C), 136.16 (2C), 137.10 (2C), 137.27 (1C), 137.78 (1C), 137.99 (2C), 139.15 (2C), 139.45 (2C), 140.68 (2C), 140.70 (2C), 141.36 (2C), 142.57 (2C), 142.67 (2C), 142.73 (2C), 142.84 (2C), 143.04 (2C), 143.32 (2C), 143.47 (1C), 143.57 (2C), 143.77 (2C), 144.04 (2C), 144.08 (2C), 144.25 (2C), 144.27 (2C), 144.36 (2C), 144.49 (2C), 144.65 (2C), 144.96 (2C), 146.68 (1C), 147.73 (2C).UV–VIS [CH2Cl2, lmax/nm (lg e)]: 261 (5.15), 330 (4.54), 403 (3.71), 430 (3.48), 555 (3.15). IR (KBr, n/cm–1): 1690 (C=O), 1645 (C=C), 928.991 (=C–H), 2921, 2851, 1455 (CH), 526 (C60). Found (%): C, 87.30; H, 2.10; N. 5.48. Calc. for C74H20N4O3· ·(C6H5CH3)0.2 (%): C, 87.81; H, 2.09; N, 5.43.N N N N O (CH2)5 O O All All N3 C60 + N N N O O O All All 180 °C solvent 1 2 + 3 + 4 All = CH2CH=CH2 N N N N O (CH2)5 O O All All N3 C60 + N N N O O O CH2CH=CH 2 CH2CH=CH 2 100 °C solvent 1 2 + 3 All = CH2CH=CH2 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17Mendeleev Communications Electronic Version, Issue 3, 2000 (pp. 83–124) The UV–VIS spectrum of compound 2† (Figure 1) is similar to the spectrum of azahomo[60]fullerene 3; the difference is in the position of a weak broad band at 500–600 nm (lmax = 555 or 546 nm in the spectrum of compound 2 or 3, respectively).Moreover, a weak and sharp peak at lmax = 420–430 nm, which is typical of monoadducts at the closed 6,6-junction of fullerene, 4 is absent from the spectrum of compound 2. Thus, according to the UV–VIS and 13C NMR spectra, compound 2 is a monoadduct of azide 1 at the 5,6-junction of fullerene (aziridino[1,6][60]fullerene).In the 13C NMR spectrum, the signal at d 120.21 ppm should be assigned to sp3 carbon of the fullerene fragment. This position of the signal of sp3 carbons of the fullerene fragment can be explained by either the electron-withdrawing influence of the isocyanurate heterocycle or the structure of aziridino[1,6][60]fullerene, in which the fullerene sphere was retained.The sp3 carbon signals at d 120.62 and 94.19 ppm were observed in the spectrum of the product of reaction between C60 and diazomethane.8 The sp3 carbon signals of the fullerene fragment of a phosphorylated isoxazoline derivative of fullerene were also detected in the same field (d 124.7 and 104.1 ppm).9 The assignment of the signal at d 120.21 ppm to the sp3 carbon of the fullerene fragment in the 13C NMR spectrum of compound 2 excludes the azahomofullerene structure of this compound, which might be an isomer of 3 with a different orientation of the substituent at nitrogen of the fullerene fragment.Note that the probability of separating such isomers is low because of the fast inversion at N at ambient temperature.10 The 1H NMR spectrum of compound 2† shows the signals of two equivalent allyl groups in the isocyanurate heterocycle and of five methylene groups, which bind the isocyanurate heterocycle to fullerene. Moreover, the signals of the methylene groups are 0.2–0.4 ppm downfield shifted in comparison with the position of these signals in the 1H NMR spectrum of azide 1.6 The protons of the methylene group closest to the nitrogen atom of aziridinofullerene are more strongly shifted.Note that a downfield shift of the signals of protons is characteristic of fullerene derivatives at the 5,6-junction.11 The IR spectrum of compound 2 exhibits bands of both isocyanurate and fullerene fragments, and no absorption is detected in the azide region at 2100 cm–1.† Thus, the spectral data show that the main product of the cycloaddition of azide 1 to C60 at 100 °C is an aziridinofullerene at the 5,6-junction of C60 without opening the fullerene framework.In connection with the isolation of aziridino[1,6][60]fullerene 2 and azahomo[60]fullerene 3 from the reaction mixture of 1 and C60, the question arises of whether an N-isocyanuratecontaining aziridino[1,2][60]fullerene (an aziridinofullerene at the closed 6,6-junction of C60) can be formed.The latter was obtained by the thermal isomerization of azahomo[60]fullerene 3.6 As a result of 20 h heating of azahomo[60]fullerene 3 (10 mg) in boiling o-dichlorobenzene (5 ml), the partial decomposition of azahomo[60]fullerene 3 to parent C60 (4 mg, 57% yield) and the formation of compound 5 (3 mg, 30% yield) were observed.The fullerene and compound 5 were separated by column chromatography on silica gel using toluene–diethyl ether (10:1) as an eluent (Rf 0.62 for 5). The UV–VIS spectrum‡ of 5 shows an absorption band at lmax = 424 nm (Figure 1), which is typical of closed 6,6-bridged fullerene monoadducts.4 The IR spectrum of compound 5 indicates the retention of fullerene and isocyanurate fragments.Note that compound 5 cannot be synthesised by the reaction of azide 1 and C60 in boiling o-dichlorobenzene because of a subsequent reaction of obtained azahomo[60]fullerene 3 with parent azide 1. A viscous dark-brown liquid was formed by heating a mixture of azide 1 and C60 at 100 °C for 10 h.The isolation of individual products from this mass was unsuccessful. Earlier,6 we showed by cyclic voltammetry that azahomo[60]- fullerene 3 is electrochemically reduced at less negative potentials than the reduction potentials of C60. Here, we examined the electrochemical reduction of compound 2. Similarly to azahomo[ 60]fullerene 3,6 compound 2 shows three reversible peaks of reduction (E1 red = –0.97, E2 red = –1.40 and E3 red = –1.86 V),§ each of them corresponds to the transfer of one electron to the aziridinofullerene molecule.However, these peaks are in the region of more negative potentials relative to the peaks of the reduction potential of C60 (E1 red = –0.92, E2 red = –1.34 and E3 red = = –1.80 V).§ Such an electrochemical behaviour is typical of aziridino[1,2][60]fullerenes.12 This work was supported by the Russian Foundation for Basic Research (grant no. 99-03-33001) and the Russian Ministry of Science (grant no. 99031). References 1 C. Bellavia-Lund, M. Kashavarz-K., R. Gonzalez, J.-C. Hummelen, R. Hicks and F. Wudl, Phosphorus Sulfur Silicon Relat. Elem., 1997, 120, 107. 2 T. Grosser, M.Prato, V. Lucchini, A. Hirsch and F. Wudl, Angew. Chem., Int. Ed. Engl., 1995, 34, 1343. ‡ Compound 5: UV–VIS [CH2Cl2, lmax/nm (lg e)]: 261 (5.15), 331 (4.48), 411 (4.38), 424 (4.37), 498 (4.30), 598 (4.22), 681 (4.16). IR (KBr, n/cm–1): 1696 (C=O), 1645 (C=C), 932.990 (=C–H), 2924, 2852, 1457 (CH), 526 (C60). § The conditions of the experiment: solution, a mixture of o-dichlorobenzene and MeCN (2:1); temperature, 25 °C; solution concentration, 1×10–3 mol dm–3; working electrode, Pt; reference electrode, Ag/AgNO3 0.01 mol dm–3 in MeCN; supporting electrolyte, 0.05 mol dm–3 Et4N·BF4; scan rate, 50 mV s–1.Absorbance 350 400 500 600 800 l/nm 1 2 3 Figure 1 UV–VIS spectra of (1) N-isocyanurate-substituted aziridino[1,6]- [60]fullerene 2, (2) azahomo[60]fullerene 3 and (3) aziridino[1,2][60]fullerene 5 in CH2Cl2.N C60 + N N N O O O All All 180 °C solvent 3 All = CH2CH=CH2 N N N O O O All All 5 N (CH2)5Mendeleev Communications Electronic Version, Issue 2, 2000 (pp. 83–124) 3 R. Banks, J. I. G. Cadogan, I. Gosney, P. K. G. Hodgson, P. R. R. Langridge-Smith, J. R. A. Millar, J. A. Parkinson, D. W. Rankin and A. T. Taylor, J. Chem.Soc., Chem. Commun., 1995, 887. 4 A. B. Smith and H. Tokuyama, Tetrahedron, 1996, 52, 5257. 5 G. Schick, T. Grösser and A. Hirsch, J. Chem. Soc., Chem. Commun., 1995, 2289. 6 O. G. Sinyashin, I. P. Romanova, G. G. Yusupova, A. A. Nafikova, V. I. Kovalenko, N. M. Azancheev, V. V. Yanilkin and Yu. G. Budnikova, Mendeleev Commun., 2000, 61. 7 V. I. Kovalenko, in Struktura i dinamika molekulyarnykh sistem (Structure and Dynamics of Molecular Systems), ed. V. D. Skirda, Unipress, Kazan, 1997, p. 91 (in Russian). 8 T. Suzuki, O. Li, K. C. Khemani and F. Wudl, J. Am. Chem. Soc., 1992, 114, 7301. 9 O. G. Sinyashin, I. P. Romanova, F. R. Sagitova, V. A. Pavlov, V. I. Kovalenko, Yu. V. Badeev, N. M. Azancheev, A. V. Ilyasov, A. V. Chernova and I. I. Vandyukova, Mendeleev Commun., 1998, 79. 10 M. Prato, Q. C. Li, F. Wudl and V. Lucchini, J. Am. Chem. Soc., 1993, 115, 1148. 11 A. Hirsch, The Chemistry of the Fullerenes, Thieme, Stuttgart, 1994, p. 89. 12 J. Zhou, A. Rieker, T. Grosser, A. Skiebe and A. Hirsch, J. Chem. Soc., Perkin Trans. 2, 1997, 1. Received: 31st August 1999; Com. 99/1530

ISSN:0959-9436     

出版商:RSC    

年代:2000

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 2, 2000 (pp. 83–124) Mixed-ligand complexes of lanthanide dialkyldithiocarbamates with 1,10-phenanthroline as precursors of lanthanide sulfides Roman A. Ivanov,* Igor E. Korsakov, Natalya P. Kuzmina and Andrey R. Kaul Department of Chemistry, M. V. Lomonosov Moscow State University, 119899 Moscow, Russian Federation. E-mail: rivanov@inorg.chem.msu.ru 10.1070/MC2000v010n03ABEH001263 A simple method for the synthesis of the title complexes, which can be used as precursors of lanthanide sulfides, in aqueous solutions has been developed.Lanthanide sulfides are of great interest in modern science and technology because of wide application of these compounds as ceramic and film materials.1 The synthesis of lanthanide sulfides is a rather difficult task because pure metal lanthanides and hydrogen sulfide should be used.2 Coordination compounds containing lanthanide–sulfur bonds can be promising precursors of lanthanide sulfides.The mixed-ligand complexes of lanthanide dialkyldithiocarbamates with 1,10-phenanthroline [Ln(dalkdtc)3- (phen)] were used as volatile precursors to obtain thin films of lanthanide sulfides.3,4 However, the wide application of [Ln- (dalkdtc)3(phen)] is restricted by a complex synthetic procedure. 4–6 The well-known method of the synthesis involves the preparation of [Ln(dalkdtc)3] complexes in anhydrous solvents followed by the formation of mixed-ligand complexes by mixing solutions containing 1,10-phenanthroline and [Ln(dalkdtc)3]. The main difficulty in this method is due to the fact that [Ln- (dalkdtc)3] solutions exhibit extremely high sensitivity to hydrolysis, and the synthesis should be performed under strongly anhydrous conditions.The prospects of [Ln(dalkdtc)3(phen)] as precursors of lanthanide sulfides stimulated our interest in the development of a more suitable synthetic route to these coordination compounds. Here, we describe a new method for the synthesis of [Ln- (dalkdtc)3(phen)] in aqueous solutions and the use of the [Ln- (dalkdtc)3(phen)] complexes for the preparation of rare-earth metal sulfides.As examples, we used Eu and Er complexes with the following three dialkyldithiocarbamate ligands: (ddtc is diethyldithiocarbamate; pmdtc is pentamethylendithiocarbamate; and mechdtc is cyclohexylmethyldithiocarbamate). This one-pot synthetic approach involves two consecutive reactions:† the interaction between LnCl3·6H2O and phen·H2O † Synthesis of Ln(dalkdtc)3(phen): a solution of 1,10-phenanthroline (1 mmol) in boiling water (20 ml) was added to an aqueous solution of LnCl3 (1 mmol, 10 ml) with intense stirring; next, aqueous solution of Na(dalkdtc) (3 mmol, 20 ml) was added dropwise to the [Ln(phen)aq]Cl3 solution, and coloured crystals were collected and dried in a vacuum at room temperature.Yield ~90%. IR (Nujol and hexachlorobutadiene mulls, n/cm–1): 1000–1020 (nC–S); 1480–1490 (nC–N); 2980–2990 (nC–H); 1600, 1620, 1630, 850, 750 (phen vibration frequencies). Eu(ddtc)3(phen). Found (%): C, 42.0; H, 5.2; N, 9.2; Eu, 19.7. Calc. for C27H38N5S6Eu (%): C, 41.7; H, 4.9; N, 9.0; Eu, 19.6.Er(ddtc)3(phen). Found (%): C, 41.1; H, 4.3; N, 9.0; Er, 21.9. Calc. for C27H38N5S6Er (%): C, 41.1; H, 4.8; N, 8.9; Er, 21.1. Eu(pmdtc)3(phen). Found (%): C, 44.8; H, 5.1; N, 8.9; Eu, 20.5. Calc. for C30H38N5S6Eu (%): C, 44.3; H, 4.7; N, 8.6; Eu, 21.1. Er(pmdtc)3(phen). Found (%): C, 43.6; H, 4.4; N, 8.5; Er, 17.8. Calc. for C30H38N5S6Er (%): C, 43.5; H, 4.6; N, 8.5; Er, 18.3.Eu(mechtc)3(phen). Found (%): C, 48.8; H, 5.0; N, 7.7; Eu, 19.3. Calc. for C36H50N5S6Eu (%): C, 48.2; H, 5.6; N, 7.8; Eu, 19.5. Er(mechtc)3(phen). Found (%): C, 47.2; H, 5.7; N, 8.1; Er, 16.9. Calc. for C36H50N5S6Er (%): C, 47.4; H, 5.5; N, 7.7; Er, 17.0. in an aqueous solution with the formation of [Ln(phen)aq]Cl3 followed by the addition of an aqueous Na(dalkdtc)·H2O solution to form coloured precipitates of [Ln(dalkdtc)3(phen)].The products obtained were dried in a vacuum at room temperature and characterised by elemental analysis and IR spectroscopy.† The new synthetic method was found to be successful for all dithiocarbamate ligands, and the product compositions were adequately described by the formulae [Ln(dalkdtc)3(phen)].Thus, the formation of [Ln(phen)aq]Cl3 at the first stage of the synthesis allowed us, at the second stage, to obtain stable rare-earth metal dialkyldithiocarbamates in an aqueous solution. The behaviour of the [Ln(dalkdtc)3(phen)] complexes on heating in a dry nitrogen flow was examined. The thermal stability of the complexes depends on the nature of ligands and decreases in the order ddtc > pmdtc > mechdtc (Table 1).‡ According to the TGA data, the thermal decomposition of [Ln(ddtc)3- (phen)] leads to the formation of lanthanide sulfides.Previously,4 it was found that the [Ln(ddtc)3(phen)] prepared by the known method can be evaporated at 250–300 °C and 0.01 Torr. The [Ln(ddtc)3(phen)] complexes prepared by the developed method demonstrated the same ability to evaporation.§ For [Ln(pmdtc)3(phen)], only 80–85% of the starting sample was evaporated at 240–280 °C because of considerable thermal degradation at this temperature. In the case of the [Ln- (mechdtc)3(phen)] complex, the evaporation was not observed. The powders of lanthanide sulfides were found in the evaporation boat after heating to 450 °C. The data obtained at a low pressure correlate with the difference in the thermal stability of complexes found by thermal analysis in nitrogen.The low thermal stability of [Ln(dalkdtc)3(phen)] forms the basis of the preparation of lanthanide sulfides by thermal decomposition of these complexes. This method can be implemented either at an atmospheric pressure in nitrogen or at a reduced pressure depending on the nature of dithiocarbamate ligands.These modes were applied to the most available diethyldithiocarbamate complex with 1,10-phenanthroline and the thermally unstable [Ln(mechdtc)3(phen)] complex, respectively. The samples of [Ln(ddtc)3(phen)] were slowly decomposed in a flow of dry oxygen-free nitrogen in the temperature range 200–400 °C for 3 h and then annealed at a higher temperature.Fine crystals of lanthanide sulfides were obtained at 1000– ‡ Thermogravimetric analysis was performed on an OD-102 derivatograph in a nitrogen atmosphere at a heating rate of 5 °C min–1. § Isothermal vacuum sublimation experiments were performed in glass tubes at a pressure of 0.01 Torr using 100 mg samples. N S S Et Et ddtc N S S pmdtc N S S Me mechdtc Table 1 Thermal analysis data for Ln(dalkdtc)3(phen).Compound Ts/°C (±5) Dm (%) (±5) SDm (%) calc. for lanthanide sulfide formation Eu(ddtc)3(phen) 290 75 65 Er(ddtc)3(phen) 280 70 62 Eu(pmdtc)3(phen) 270 76 77 Er(pmdtc)3(phen) 275 70 74 Eu(mechtc)3(phen) 250 74 79 Er(mechtc)3(phen) 255 71 76Mendeleev Communications Electronic Version, Issue 3, 2000 (pp. 83–124) 1100 °C.In the case of [Ln(mechdtc)3(phen)], lanthanide sulfides were obtained on heating at 500 °C for 2 h at a low pressure. The formation of crystalline EuS and Er2S3 was confirmed by X-ray diffraction analysis¶ (Figure 1). Note that the crystallinity of sulfides prepared from [Ln- (mechdtc)3(phen)] at a low pressure was slightly worse than that of the samples obtained at a higher temperature in a nitrogen flow.Thus, the temperature conditions of lanthanide sulfide formation at a low pressure should be optimised. ¶ X-ray diffraction data for EuS obtained using [Eu(ddtc)3(phen)] as a precursor: a = 17.0 (0.1), b = 4.00 (0.04) and c = 10.11 (0.03) Å, b = = 99.4 (0.4)°; CCDC no. 21-324: a = 17.404, b = 3.978 and c = 10.092 Å, b = 98.67°. In summary, we developed a new very simple method for the synthesis of mixed-ligand dialkyldithiocarbamate complexes of lanthanides with 1,10-phenanthroline and demonstrated that these complexes can be used as precursors of lanthanide sulfides.This work was supported by the programmes ‘Universities of Russia’ and ‘Basic Research in Chemical Technologies’ and by INTAS (grant no. 96-2359). References 1 P.Maestro and D. Huguenin, J. Alloys Compd., 1995, 225, 520. 2 Gmelin Handbook of Inorgan. Chem., 8th Edn., Springer, Berlin, 1984, 61B67, p. 105. 3 V. I. Bessergenev, E. N. Ivanova, Ya. A. Kovalevskaya and S. V. Larionov, Spring Meeting of Electrochem. Soc., Los Angeles, 1996, p. 105. 4 N. P. Kuzmina, R. A. Ivanov, S. E. Paramonov and L. I. Martynenko, Electrochem. Soc. Proceed. 14th Intern. Conf. on CVD and EUROCVD, ed. M. Allendorf, Electrochemical Society, Inc., Pennington, 1997, p. 872. 5 Comprehensive Coordination Chemistry, ed. V. C. Willkinson, Pergamon Press, Oxford, 1986, p. 92. 6 N. I. Gorshkov, G. V. Sidorenko and D. N. Suglobov, Radiokhimiya, 1994, 36, 154 [Radiochem. (Engl. Transl.), 1994, 36, 167]. 20 25 30 35 40 45 50 55 60 65 2q I(a.u.) (1) (2) Figure 1 X-ray diffraction patterns of (1) EuS obtained by the thermolysis of [Eu(ddtc)3(phen)] at 1100 °C in a nitrogen flow and (2) a reference sample of EuS powder. Received: 18th January 2000; Com. 00/1589

ISSN:0959-9436     

出版商:RSC    

年代:2000

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 2, 2000 (pp. 83–124) Surface chemistry of noble metal complexes anchored from cationic complexes on a graphitised carbon support Aleksandr Yu. Stakheev,*a Galina N. Baeva,a Natal’ya S. Telegina,a Anatoly B. Volynsky,b Leonid M. Kustova and Khabib M. Minacheva a N. D. Zelinsky Institute of Organic Chemistry, Russian Academy of Sciences, 117913 Moscow, Russian Federation.Fax: +7 095 135 5328 b V. I. Vernadsky Institute of Geochemistry and Analytical Chemistry, Russian Academy of Sciences, 117975 Moscow, Russian Federation. Fax: +7 095 938 2054 10.1070/MC2000v010n03ABEH001247 The adsorption and chemical transformations of ammonia complexes of Pd, Pt, Rh and Ir on the surface of a modified carbon support were studied by X-ray photoelectron spectroscopy.Porous carbon materials are widely used for the preparation of metal-supported catalysts, particularly for liquid-phase processes. A new generation of graphite-like materials expands further the application of carbon supports. Among them, a ‘Sibunit’ mesoporous graphite-like carbon attracts particular attention because of its chemical inertness, mechanical strength, and high surface area (200–600 m2 g–1).1 Several studies have been devoted to researching noble metal deposition on the carbon surface by ion exchange, since this technique allows the preparation of highly dispersed supported metal catalysts.2–4 However, most studies performed so far have focused only on the preparation of catalysts by ion exchange with anionic metal complexes like H2PdCl4, HAuCl4, and H2PtCl6.The main disadvantage of this approach is the reduction of the anionic complexes in the course of ion exchange resulting in an uncontrolled deposition of reduced metal species. As a result, the metal dispersion becomes poor and the amount and distribution of the deposited metal cannot be controlled.3,4 Our preliminary experiments with cationic complexes demonstrated that the ion-exchange capacity of the parent Sibunit is negligible (Table 1).Therefore, the aim of this study was an attempt to modify the Sibunit surface in order to increase its exchange capacity towards cationic complexes and to study their deposition with the final goal of preparing highly dispersed metal supported catalysts. The presence of active oxygen-containing groups on the surface of the carbon material is a prerequisite for the efficiency of ion exchange of cationic complexes.5 Therefore, before ion exchange, Sibunit (5 g) was treated with 150 ml of an aqueous KMnO4 solution (0.2 mol dm–3) for 3 h at 80 °C.The resulting material was rinsed with distilled water to pH 7. After that, the Sibunit was treated with 4 M HCl overnight at room temperature. Next, the Sibunit was exhaustively washed with distilled water until the sample became free of nitrate and manganese ions, as evidenced by XPS.The pretreated Sibunit was immersed in a 0.01 or 0.0004 M aqueous solution of [Pd(NH3)4](NO3)2, [Pt(NH3)4]Cl2, [Rh(NH3)5- Cl]Cl2 or [Ir(NH3)5Cl]Cl2 and kept under stirring for 6 h at room temperature.Preliminary experiments indicated that an equilibrium concentration of the adsorbed complexes is achieved in less than 3 h. The excess of the complex was washed out thoroughly, and the samples were dried at 120 °C overnight. X-ray photoelectron spectra were measured on an XSAM-800 spectrometer (Kratos) using Al Ka1,2 radiation for spectra excitation. Atomic ratios were calculated from the integral intensities of XPS peaks using Scofield’s photoionization cross-sections for Al Ka1,2 excitation.6 The accuracy of measurements was checked using the initial complexes and was estimated as ±15%.The binding energies of peaks were corrected to account of sample charging by referencing to the C 1s peak at 285.0 eV. Figure 1 shows the XPS spectra of the samples containing Pt and Pd species.Obviously, the technique applied to the adsorp- (a) (b) 4 3 2 1 Pt2+ Pt0 Pd2+ Pd0 85 80 75 70 65 60 350 345 340 335 330 Binding energy/eV Figure 1 XPS spectra of (a) Pt 4f and (b) Pd 4f in complexes supported on the modified Sibunit surface: (1) initial complex; (2) surface complex deposited from a 0.0004 M solution; (3) surface complex deposited from a 0.01 M solution; (4) metal foil. 4 3 2 1 aCalculated on the basis of XPS data using the Kerkof model.12 Table 1 Ion-exchange capacity of Sibunit towards noble metal complexes. Sibunit Complex Concentration of the complex in solution/mol dm–3 Exchange capacitya/ mmol g–1 parent [Pd(NH3)4](NO3)2 0.01 < 0.005 parent [Pt(NH3)4]Cl2 0.01 < 0.005 modified [Pd(NH3)4](NO3)2 0.01 0.34 modified [Pd(NH3)4](NO3)2 0.0004 0.18 modified [Pt(NH3)4]Cl2 0.01 0.036 modified [Pt(NH3)4]Cl2 0.0004 0.051 modified [Rh(NH3)5Cl]Cl2 0.01 0.15 modified [Rh(NH3)5Cl]Cl2 0.0004 0.22 modified [Ir(NH3)5Cl]Cl2 0.01 0.039 modified [Ir(NH3)5Cl]Cl2 0.0004 0.055 7 6 5 4 3 2 1 0 Pt Pd Ir Rh M/C×10–3 0.0004 M 0.01 M Figure 2 Effect of the nature of the complexed metal and its concentration in solution on the metal concentration at the Sibunit surface.Mendeleev Communications Electronic Version, Issue 3, 2000 (pp. 83–124) tion of cationic complexes allows us to avoid the formation of metal particles during ion exchange. Similar results were obtained for the Ir- and Rh-containing samples. The dependence of the surface metal concentration on the nature and concentration of the metal in solution is depicted in Figure 2.The ion-exchange capacity of the Sibunit calculated on the basis of these data is given in Table 1. The exchange capacity of the modified Sibunit surface appears to be a function of the nature of the complex rather than its solution concentration. Modified Sibunit shows a maximum capacity towards Pd and Rh, while the capacity towards Pt and Ir is markedly lower.As the XPS data indicate, the nitrogen-to-metal (N/M) atomic ratio decreases with increasing surface concentration of the metal complex (Figure 3). For the samples containing low concentrations of Pt and Ir, the N/M ratio remains practically the same as in the initial complexes. The positions of the Pt 4f and Ir 4f peaks also remain almost unchanged. However, a further increase in the concentration of the adsorbed complex is accompanied by a significant decrease in the N/M ratio.This effect is well pronounced for Pd- and Rh-containing complexes. A shift of the corresponding XPS peaks is also observed [Figure 1(b)], which can indicate a change in the structure of the complex ion on the Sibunit surface (e.g., due to a change in the coordination number of the metal ion7).The following tentative interpretation may be proposed in explanation of the behaviour of the noble metal complexes on the surface of the carbon support. According to the mechanism proposed by Morikawa,8 oxidation of the carbon surface results in the formation of carboxyl groups. Protons of the carboxyl groups can be exchanged with metal ions.In accordance with the Morikawa mechanism, the inner sphere of the complex remains intact and the complex is bound to the surface groups only by electrostatic forces. However, our data indicate that, in part, the ligands can be substituted, presumably with oxygen atoms of carboxyl groups. This leads probably to the formation of M–O–C bonds. The final structure of the adsorbed complex seems to be a function of the stability constants of the corresponding complexes.According to Grinberg et al.,9,10 the overall stability constant for [Pt(NH3)4]2+ is about five orders of magnitude higher than that for [Pd(NH3)4]2+. This effect is evidently caused by the so-called ‘lanthanide contraction’.11 Presumably, the stability constants for [Ir(NH3)5Cl]2+ and [Rh(NH3)5Cl]2+ change in the same direction (the corresponding data could not be found in the literature): Since Pt and Ir complexes are significantly more stable than Rh and Pd complexes, the elimination of NH3 ligands and the formation of –O–M bonds in the course of the ion exchange occurs more easily for the Pd and Rh complexes.This results in their stronger bonding to the support surface and, in turn, in a higher concentration of the supported metal.Thus, the oxidative modification of the Sibunit surface followed by ion exchange with noble metal complexes makes it possible to avoid reduction of the adsorbed metal ions and facilitates the control of the adsorption process. The surface concentration of adsorbed metal ions and the structure of the metal complexes appear to be a function of the stability constants of the corresponding inner-sphere complexes.This work was supported by the Russian Foundation for Basic Research (grant nos. 99-03-32222 and 99-03-32750). References 1 N. S. Polyakov, G. A. Petukhova and V. F. Surovikin, Izv. Akad. Nauk, Ser. Khim., 1993, 1377 (Russ. Chem. Bull, 1993, 42, 1308). 2 V. B. Fenelonov, L. B. Avdeeva and O.V. Gonacharova, Stud. Surf. Sci. Catal., 1995, 91, 825. 3 P. A. Simonov, E. M. Moroz, A. L. Chuvilin, V. N. Kolomiichuk, A. I. Boronin and V. A. Likholobov, Stud. Surf. Sci. Catal., 1995, 91, 977. 4 P. A. Simonov, E. M. Moroz, V. A. Likholobov and G. V. Plaksin, Izv. Akad. Nauk SSSR, Ser. Khim., 1990, 1478 (Bull. Acad. Sci. USSR, Div. Chem. Sci., 1990, 39, 1335). 5 K.Amine, M. Mizuhata, K. Oguro and H. Takenaka, J. Chem. Soc., Faraday Trans., 1995, 91, 4451. 6 J.H.Scofield, J. Electron Spectrosc., 1976, 9, 29. 7 D. P. Woodruff and T. A. Delchar, Modern Techniques of Surface Science, Cambridge University Press, Cambridge, 1989. 8 K. Morikawa, T. Shirasaki and M. Okada, Adv. Catal., 1969, 60, 98. 9 A. A. Grinberg and M. I. Gel’sman, Dokl. Akad.Nauk SSSR, 1961, 137, 87 [Dokl. Chem. (Engl. Transl.), 1961, 137, 257]. 10 A. A. Grinberg, N. V. Kiseleva and M. I. Gel’sman, Dokl. Akad. Nauk SSSR, 1967, 172, 856 [Dokl. Chem. (Engl. Transl.), 1967, 172, 114]. 11 J. E. Heheey, Inorganic Chemistry. Principles of Structure and Reactivity, 3rd edn., Harper and Row, New York, 1983. 12 F. P. J. M. Kerkof and J. A. Moulijn, J. Phys. Chem., 1979, 83, 1612. C O C O H O C O H Pd H3N NH3 NH3 NH3 2+ – 2H+ Pd H3N NH3 NH3 NH3 2+ C O C O O C O – 2NH3 C O C O O C O Pd H3N NH3 [Pt(NH3)4]2+ ª [Ir(NH3)5Cl]2+ >> [Pd(NH3)4]2+ ª [Rh(NH3)5Cl]2+ 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 0 2 4 6 7 1 3 5 N/M M/C×10–3 Figure 3 Dependence of the N/M ratio in the supported complexes on the surface metal concentration. Empty and filled symbols correspond to the samples prepared by ion exchange from 0.0004 M and 0.01 M solutions, respectively. Pt(NH3)4 2+ Ir(NH3)5Cl2+ Rh(NH3)5Cl2+ Pd(NH3)4 2+ Received: 7th December 1999; Com. 99/1573

ISSN:0959-9436     

出版商:RSC    

年代:2000

数据来源: RSC