摘要:

L e t t e r Conversion of aldehydes to amides via dimethyl sulfoxide oxidation of the corresponding a-aminonitriles Dieter Enders,* Andreç s S. Amaya and Fabrice Pierre Institut Organische Chemie, T echnische Hochschule, f ué r Rheinisch-W estf aé lische Professor-Pirlet-str. 1, 52074 Aachen, Germany. E-mail : Enders=RWTH-Aachen.de Received (in Montpellier, France) 11th December 1998, Accepted 17th December 1998 Unsaturated, aryl and heteroaryl N-dialkyl-a-aminonitriles are easily oxidized at room temperature to their corresponding tertiary N-dialkylamides by using dimethyl sulfoxide and bases like ButOK or KOH.The reaction oÜers an efficient two-step conversion of aldehydes to N-dialkyamides. Owing to their ease of operation and eÜectiveness, oxidations with activated dimethyl sulfoxide (DMSO) are regularly employed in organic synthesis.A variety of diÜerent reactions such as the Kornblum,1 P–tzner-MoÜat,2 Swern3 and Corey- Kim4 reactions, amongst others5 have become classical procedures, especially those utilised for the conversion of alcohols to carbonyl compounds. Only a few methods for the direct conversion of aldehydes to their corresponding amides have been reported so far, through electrophilic amination of O-(trimethylsilyl) aldehyde cyanohydrins,6 and through radical-mediated oxidation of aldehydes to acid bromides.7 Two papers have also reported a base-catalysed of a-aminonitriles.8 O2-autoxidation We now wish to report a new DMSO-based reaction, the oxidation of unsaturated, aryl and heteroaryl N-dialkyl-aaminonitriles to their corresponding N-dialkylamides.Since the a-aminonitriles are easily accessible from the parent aldehydes, the reaction allows an efficient two-step conversion of aldehydes to tertiary amides. The oxidation was performed by simply adding the aaminonitriles 2 (obtained from the aldehydes 1 by Strecker synthesis9) to a DMSO solution containing one equivalent of potassium tert-butoxide or potassium hydroxide, and subsequently stirring at room temperature for 12 h»3 days (Scheme 1).§ After work-up and puri–cation by distillation or chromatography, the N-dialkylamides 3a»h were isolated in satisfactory yields and were fully characterized (Table 1).A relatively broad range of dialkylaminonitriles bearing a heterocycle (furyl, pyridyl), an aromatic or an unsaturated chain could be Scheme 1 Table 1 Reaction conditions and yields for the oxidation of the a-aminonitriles 2 to the amides 3 Entry Amide Reaction timea Yieldb(%) Physical characteristicsc/ °C Torr~1 3a 3 days 64 (50) m.p.: 95»97 (Lit.10 95»96) 3b 3 days 79 (66) m.p. : 47»49 (Lit.11 48»49) 3c 12 h 89 (70) m.p. : 40 (Lit.10 41»43) 3d 3 daysd 43 (30) Oil12 3e 3 days 56 (50) Oil13 3f 3 days 66 (44) See typical procedure 3g 3 days 91 (76) b.p.: 55»60/0.05 (Lit.6») 3h 3 days 68 (65) b.p. : 83»84 (Lit.14») a 2 (5 mmol), ButOK (5 mmol), DMSO (15 ml), room temperature. b Value in parentheses is calculated from 1. c Uncorrected. The spectroscopic data are in accordance with the literature data. d 2 (5 mmol), KOH (5 mmol), DMSO (15 ml), room temperature.New J. Chem., 1999, 261»262 261Scheme 2 transformed, but attempts to carry out the oxidation on aminonitriles bearing alkyl chains Et, But) were (R\C5H11, unsuccessful. DMSO was unambiguously implicated in the oxidation process since no transformation occurred in its absence. In addition, dimethyl sul–de resulting from the reduction of DMSO was clearly detected by its characteristic odour.We also ensured that oxygen was not involved in the process, by carrying out all the reactions in dry and degassed DMSO, under an argon atmosphere. Bearing this in mind, a reasonable mechanism for the oxidation might involve the formation of an alkoxysulfonium intermediate A (Scheme 2) after substitution of the cyanide by DMSO in the a-aminonitriles 2.This intermediate A could then form the –nal products through the corresponding sulfonium ylide, intramolecular proton transfer and loss of dimethyl sul–de, as is usually described for related reactions.5 In summary, the transformations of Scheme 1 are experimentally simple, mild and efficient and the second step constitutes a new application of DMSO in oxidation reactions.Acknowledgements work was supported by the Fonds der Chemischen This Industrie and the Deutsche Forschungsgemeinschaft (Leibniz Prize). We wish to thank Degussa AG, BASF AG, Bayer AG, Hoechst AG and Knoll AG for their donation of chemicals. One of the authors (F.P.) thanks the Alexander von Humboldt foundation for support. Notes and references § Typical procedure for the preparation of 3f : Under an argon atmosphere, ButOK (0.56 g, 5 mmol) was dissolved in 10 ml of degassed DMSO and stirred at room temperature for 10 min.A solution of the aminonitrile 2f (1.02 g, 5 mmol) in 5 ml of DMSO was added dropwise and the resulting mixture stirred at room temperature for 3 days. After adding 30 ml of ether, the organic phase was washed with 5% aqueous solution of KOH to eliminate the residual DMSO.The organic layer was washed with a saturated solution of NaHCO3 , dried over and the solvent removed in vacuo. Puri–cation by MgSO4 column chromatography (neutral alumina, activity grade III, ether) and distillation (73 °C/0.01 Torr) aÜorded the amide 3f, isolated as a yellow oil (0.64 g, 66%). IR (neat) : 3090, 3000»2840, m(CO) 1630, 1500, 1400, 1190, 1070, 890 cmv1; 1H NMR d 1.51 (s, 3H), 1.51»2.4 (CDCl3) : (m, 7H), 2.95 (s, 6H), 4.68 (m, 2H), 5.77 (m, 1H); MS : m/z (%) 193 (M`, 96), 192 (16), 178 (17), 165 (50), 164 (36), 152 (34), 150 (27), 149 (53), 148 (24), 147 (27), 126 (18), 121 (42), 113 (24), 111 (11), 105 (30), 93 (96), 91 (34), 82 (45), 81 (46), 80 (11), 79 (86), 77 (31), 72 (100), 67 (18), 55 (19), 53 (58), 45 (11), 44 (26), 41 (24) ; anal.calcd for C12H19NO (193.29) : C, 74.57 ; H, 9.9 ; N, 7.25. Found: C, 74.10 ; H, 10.2 ; N, 7.38%. 1 N. Kornblum, J. W. Powers, G. J. Anderson, W. J. Jones, H. O. Larson, O. Levand and W. M. Weaver, J. Am. Chem. Soc., 1957, 79, 6562. 2 (a) K. E. P–tzner and J. G. MoÜat, J. Am. Chem. Soc., 1963, 85, 3027; (b) K. E. P–tzner and J. G. MoÜat, J. Am. Chem.Soc., 1965, 87, 5661. 3 A. J. Mancuso and D. Swern, Synthesis, 1981, 165. 4 (a) E. J. Corey and C. U. Kim, J. Am. Chem. Soc., 1972, 94, 7586 ; (b) E. J. Corey, C. U. Kim, J. Org. Chem., 1973. 38, 1233. 5 For a review see : T. T. Tidwell, Synthesis, 1990, 857. 6 G. Boche, F. Bossold and M. T etrahedron L ett., 1982, 23, Nieêner, 3255. 7 I. E. Markoç and A. Mekhal–a, T etrahedron L ett., 1990, 31, 7237. 8 (a) F. Yuste, A. E. Origel and L. J. Bren8 a, Synthesis, 1983, 109; (b) T.-H. Chuang, C.-C. Yang, C.-J. Chang and J.-M. Fang, Synlett, 1990, 733. 9 (a) H. Ahlbrecht and C. Vonderheid, Synthesis, 1975, 512; (b) W. L. Matier, D. A. Owens and W. T. Comer, J. Med. Chem., 1973, 16, 901; (c) S. F. Dyke, E. P. Tiley, A. W. C. White and D. P. Gale, T etrahedron, 1975, 31, 1219. 10 J. Kopecky and J. Smejkal, Chem. Ind. (L ondon), 1966, 1529. 11 H. Schindlbauer, Monatsh. Chem., 1968, 99, 1799. 12 R. P. Woodbury and M. W. Rathke, J. Org. Chem., 1978, 43, 1947. 13 R. Da Dacosta, M. Gillard, J. B. Falmague and L. Ghosez, J. Am. Chem. Soc., 1979, 101, 4381. 14 A. Brunno and G. Purello, Gazz. Chim. Ital., 1966, 96, 986. L etter 8/09770D 262 New J. Chem., 1999, 261»262

ISSN:1144-0546    

DOI:10.1039/a809770d

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

L e t t e r Light-emitting diode device from a luminescent organocopper(I) compound Yu-Guang Ma,a,b Wing-Han Chan,a Xue-Mei Zhoub and Chi-Ming Che*a a Department of Chemistry, T he University of Hong Kong, Pokfulam Road, Hong Kong. E-mail : cmche=hkucc.hku.hk b Key L ab for Supramolecular Structure and Spectra, Jilin University, Changchun 130023, P. R. China Received (in Montpellier, France) 9th November 1998, Accepted 8th December 1998 A highly luminescent tetranuclear copper(I) compound, [L= 1,8-bis(diphenylphosphino)- [Cu4(CyCPh)4L2 ] (Cu4) 3,6-dioxaoctane ] , was used as an emitting material for fabrication of light-emitting diode (LED) devices.The doped Cu4 poly(vinyl carbazole) device displays electroluminescence nm) that is similar in energy to the photolumine- (kmax = 520 scence recorded in dichloromethane.The device characteristics and mechanism of electroluminescence have been studied. The use of luminescent inorganic binary compounds such as GaN and metal chalcogenides as light-emitting diode (LED) materials is well-documented. In contrast, related studies with luminescent metal-organic compounds such as [tris(8- Alq3 hydroxyquinoline)aluminum] and its derviatives are still in their infancy.1h3 In contrast to semiconductor nanocrystallites, polynuclear d10 metal compounds have well-de–ned molecular structures and recent studies have established their rich and diverse photoluminescent properties.Importantly, their excited state and physiochemical properties are tunable through variation of the metal atom and auxiliary ligands.4h6 This, in principle, would allow the study of structure-property relationships in LED devices and provide a —exible entry to the fabrication process. Herein is described the luminescent complex [L\1,8-bis(diphenylphosphino)- [Cu4(CyCPh)4L2] 3,6-dioxaoctane] as an advanced material for LEDs.This Cu4 compound is air-stable, can be readily prepared from inexpensive starting materials,7 and has an emission quantum yield comparable to that of the classic compound.8 Moreover, Alq3 the intramolecular CuI»CuI interaction may lead to a high carrier mobility and provide a new pathway for the carriertransfer process. These advantages allow and its related Cu4 complexes6,9 to have useful optoelectronic applications.Fig. 1 Diagram showing the core structure of and the con–guration of the LED device.Cu4 New J. Chem., 1999, 263»265 263Fig. 2 (a) Photoluminescence spectrum of the –lm and Cu4 : PVK (b) electroluminescence spectrum of the ITO/Cu4 : PVK/TAZ/Al device excited at 320 nm. The core structure of the complex and the con–gu- Cu4 ration of the LED device are depicted in Fig. 1. The complex has a zigzag core wrapped by two bridging phosphine Cu4 ligands.It displays an intense low energy absorption at 408 nm (e\7.8]103 M~1 cm~1) that is attributable to an admixture of the metal-centered 3d]4s/4p and CuI]p*(CyCPh) transitions. In room temperature dichloromethane solution, the complex exhibits an intense photoluminescence at 520 nm with a lifetime of 9.8 ls and quantum yield of 0.42.In solid state, the complex emits at 533 nm. As the complex is non-volatile, a –lming process using vacuum deposition cannot be used. We therefore immobilize it into a PVK polymer matrix that has a high hole drift mobility. In this work, two LEDs with the structures ITO/Cu4 : PVK (1000 (single layer) and (1000 ”)/Al ITO/Cu4 : PVK derivative (TAZ 45 (double layer) were ”)/triazole ”)/Al fabricated.§ The TAZ layer, which transports electrons and blocks holes eÜectively, is generally used together with the electron-injection layer.However, the latter is not Alq3 employed in our study since shows photoluminescence at Alq3 a similar energy as that of the compound. Cu4 Fig. 2 shows the photoluminescence (PL) and electroluminescence (EL) spectra of the –lm measured at Cu4 : PVK room temperature.î The PL spectrum of the (10 Cu4 : PVK wt.% –lm shows two peaks at 516 and 420 nm (intensity Cu4) ratio The low energy emission (lifetime\4.1 I516/I420\3).ls) originates from the triplet excited state of the Cu4 complex, whereas the high energy emission (lifetime \10 ns) is assigned to an exciplex emission of the carbazole groups in PVK.10 In the EL spectrum, the emission intensity at kmax\ 516 nm is much enhanced while the high energy band becomes a weak shoulder (intensity ratio This I516/I420\10). can be explained by a higher formation efficiency of the triplet than that of the singlet state.10 The EL spectrum of a single layer device exhibits a Cu4 weak emission with at 520 nm (driving voltage P12 V), kmax which is nearly identical with the PL spectrum of the complex measured in dichloromethane solution.This indicates that both the PL and EL originate from the same excited state. Fig. 3 Room temperature electroluminescence spectra of an device at variable applied voltage. Insert : ITO/Cu4 : PVK/TAZ/Al current density and EL intensity versus driving voltage. (L) (=) For the double layer device, the turn-on voltage for EL (B12 V) is similar to that of the single layer device but the emission efficiency is ten times higher for the same injecting current density.Presumably, the TAZ layer leads to a lower eÜective barrier for electron injection and/or a better con–nement of holes to the emissive layer.11 The EL spectra of the double layer device at various applied voltages are depicted in Fig. 3. The insert graph shows plots of the current density and emission intensity versus the applied voltage. A non-linear relationship between the EL intensity and the current density is observed, which indicates a non-balanced charge injection in the device. When the applied voltage is less than 15 V, the injection barrier for the electrons in the TAZ layer is high, thus resulting in low EL efficiency.At high driving voltage, the electrons have sufficient energy to overcome the barrier and subsequently transport to the TAZ»PVK interface, where they combine with the holes to give the observed emission. The brightness at a current density of 20 mA cm~2 is ca. 50 cd m~2 and the estimated EL yield is 0.1%. We envisage that both the EL yield and the turn-on voltage can be further improved by using a low workfunction cathode to enhance electron injection or materials with a higher electron affinity as the injection layer. In this work, we have demonstrated the successful application of a polynuclear metal alkynyl compound for LED device fabrication.When compared to the d8 poly-yne complexes of PtII,9 the d10 metal alkynyl complexes are advantageous because of their high luminescence quantum yields.5 Thus, we envisage that polynuclear d10 metal alkynyl compounds, with their rich and diverse photoluminescent properties, have important applications in future development of LED technology.Acknowledgements acknowledge support from the University of Hong Kong, We the Croucher Foundation and the Hong Kong Research Grants Council.Notes and references § LEDs were fabricated on indium»tin»oxide (ITO) coated glass substrates (sheet resistance 20 which were cleaned successively in )/K), ultrasonic baths of detergent, water, acetone, water and ethanol, and then dried under vacuum at 200 °C for 2 h. A thin emitting layer (typically 1000 of (10 wt.% was formed by spin- ”) Cu4 : PVK Cu4) coating a chloroform solution onto the ITO-coated glass substrate.TAZ (45 was deposited on ITO at B2]10~6 Torr (B2.67]10~8 ”) Pa) at a rate of B1 s~1. An Al cathode (2000 was vacuum ” ”) deposited on top of these multilayer –lms at an evaporation rate of 3»5 s~1. The substrate was kept at room temperature during depo- ” sition. The active area of the LEDs was 2]5 mm2. 264 New J.Chem., 1999, 263»265î Time-resolved luminescence decay was recorded on a boxcar averager using time-correlated counting methods. Excitation was monitored at 355 nm by a mode-locked Q-switched Nd : YAG laser with pulse width of 8 ns, frequency of 50 Hz and an integration power of 50 mW. All photophysical measurements were performed at room temperature. 1 (a) C. W. Tang and S.A. VanSlyke, Appl. Phys. L ett., 1987, 51, 913; (b) C. W. Tang, S. A. VanSlyke and C. H. Chen, J. Appl. Phys., 1989, 65, 3610; (c) M. Strukelj, R. H. Jordan and A. Dodabalapur, J. Am. Chem. Soc., 1996, 118, 1213. 2 (a) C. C. Wu, J. K. M. Chun, P. E. Burrows, J. C. Sturm, M. E. Thompson, S. R. Forrest and R. A. Register, Appl. Phys. L ett., 1995, 66, 653; (b) M. Era, S. Morimoto, T.Tsutsui and S. Saito, Synth. Metals, 1995, 71, 2013; (c) L. S. Sapochak, P. E. Burrows, D. Garbuzov, D. M. Ho, S. R. Forrest and M. E. Thompson, J. Phys. Chem., 1996, 100, 17766; (d) J. Kido and J. Endo, Chem. L ett., 1997, 593; (e) J. Kido and J. Endo, Chem. L ett., 1997, 633; ( f ) J. Kido and Y. Iizumi, Chem. L ett., 1997, 963; (g) C. H. Chen, Macromol. Symp., 1997, 125, 1. 3 Y. Kunugi, K. R. Mann, L. L. Miller and C. L. Exstrom, J. Am. Chem. Soc., 1998, 120, 589. 4 (a) C. W. Chan, T. F. Lai, C. M. Che and S. M. Peng, J. Am. Chem. Soc., 1993, 115, 11245; (b) D. Li, X. Hong, C. M. Che, W. C. Lo and S. M. Peng, J. Chem. Soc., Dalton T rans., 1993, 2929; (c) X. Hong, K. K. Cheung, S. M. Peng and C. M. Che, J. Chem. Soc., Dalton T rans., 1994, 3607; (d) C. W.Chan, L. K. Cheng and C. M. Che, Coord. Chem. Rev., 1994, 132, 87; (e) B. C. Tzeng, W. C. Lo, C. M. Che and S. M. Peng, Chem. Commun., 1996, 181. 5 (a) V. W. W. Yam, W. K. Lee and T. F. Lai, Organometallics, 1993, 12, 2383; (b) V. W. W. Yam, W. K. Lee and T. F. Lai, J. Chem. Soc., Chem. Commun., 1993, 1571; (c) V. W. W. Yam, W. K. M. Fung and K. K. Cheung, Angew. Chem. Intl. Ed.Eng., 1996, 35, 3459; (d) V. W. W. Yam and K. K. W. Lo, Comm. Inorg. Chem., 1997, 19, 209. 6 (a) A. Vogler and H. Kunkely, J. Am. Chem. Soc., 1986, 108, 7211; (b) A. Vogler and P. C. Ford, Acc. Chem. Res., 1993, 26, 220; (c) A. Juris, V. Balzani, F. Barigelletti, S. Campagna, P. Belser and A. von Zelewsky, Coord. Chem. Rev., 1988, 84, 85; (d) M. Maestri, C. Deuschel-Cornioley and A. von Zelewsky, Coord. Chem. Rev, 1991, 111, 117. 7 W. H. Chan, Z. Z. Zhang, T. C. W. Mak and C. M. Che, J. Organomet. Chem., 1998, 556, 169. 8 D. Z. Garbuzov, V. Bulovic, P. E. Burrows and S. R. Forrest, Chem. Phys. L ett., 1996, 249, 433. 9 (a) A. E. Dray, F. Wittmann, R. H. Friend, A. M. Donald, M. S. Khan, J. Lewis and B. F. G. Johnson, Synth. Met., 1991, 41»43, 871; (b) A. Kohler, H. F. Wittmann, R. H. Friend, M. S. Khan and J. Lewis, Synth. Met., 1994, 67, 245. 10 Y. G. Ma, H. Y. Zhang, J. C. Shen and C. M. Che, Synth. Met., 1998, 94, 245. 11 J. Kido, M. Kimura and K. Nagai, Science, 1995, 267, 1332. L etter 8/08763F New J. Chem., 1999, 263»265 265

ISSN:1144-0546    

DOI:10.1039/a808763f

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

L e t t e r A new 1D bimetallic thiocyanate-bridged copper(II)ñcobalt(II) compound§ Giancarlo Francese,a Sylvie Ferlay,a Helmut W. Schmalleb and Silvio Decurtinsa* a Departement Chemie und Biochemie, Bern, Freiestrasse 3, CH-3012 Bern, f ué r Universitaé t Switzerland. E-mail : Silvio.Decurtins=iac.unibe.ch b Anorganisch-chemisches Institut, W interthurerstrasse 190, CH-8057 Universitaé t Zué rich, Zué rich, Switzerland Received (in Strasbourg, France) 5th October 1998, Revised m/s received 21st December 1998, Accepted 21st December 1998 The crystal structure of the –rst one-dimensional heterometallic compound containing thiocyanate as bridging ligands, has been determined, together { [Cu(cyclam)] [Co(NCS)4 ] }n , with a preliminary study of the magnetic properties.In recent years, the –eld of molecular magnetism has attracted considerable attention, both for theoretical reasons and with a view to developing new materials for magneto-optic devices.1 In particular, special attention has been paid to the design and construction of one-dimensional, metal ion-containing molecular compounds (both homo and heterometallic) in order to undertake theoretical studies of their magnetic properties.2 The design of such materials requires appropriate components, such as suitable bridging ligands between the magnetic centres.The thiocyanate ligand, with the ambidentate ability of its ììend-to-endœœ coordination mode, seems to play an important role in the design of extended one-,3h5 two-6 and three-dimensional3,7 copper(II) compounds with speci–c magnetic properties.Other extended homometallic compounds, in which the metals are bridged by the thiocyanate ligand, include nickel(II),8 manganese(II)9 and cobalt(II)10 derivatives. However, there is no information in the literature about in–nite heterometallic copper(II)-containing networks with thiocyanate as a bridging ligand. Only recently the crystal structures and magnetic properties of tetranuclear heterometallic compounds have been reported,11 but no compounds of higher dimensionality are known.In this preliminary communication, we report the synthesis, crystal structure and magnetic properties of the –rst bimetallic cobalt(II)»copper(II) one-dimensional complex 1. Compound 1 was prepared by carefully adding 40 ml of a 1 : 1 solution containing cyclam (1.53 mmol) and CH3CN»EtOH (1.53 mmol) to 10 ml of an alcoholic solu- Cu(ClO4)2 … 6H2O tion (1.53]10~2 M) of (1.53 mmol).12 The K2[Co(NCS)4] solution was left in air to evaporate and deep-blue single crystals appeared after several days.î The infrared spectrum° of 1 exhibits two bands in the region near 2000 cm~1 (2107 and 2073 cm~1), which correspond to the stretching mas(NCS) mode, resulting from a combination of four eÜects : (i) the presence of bridging thiocyanate ligands (CowNCSwCu); (ii) the presence of non-bridging thiocyanate ligands (CowNCS); (iii) the distorted octahedral copper(II) environment and (iv) the tetrahedral environment of cobalt(II).It is therefore difficult to assign unambiguously both bands, in accordance with earlier discussions in the literature.13 The molecular structure“ of 1 is presented in Fig. 1 as an ORTEP view of the molecule, together with the atomic labeling scheme, while Fig. 2 shows the one-dimensional character of the compound, and especially the coordination geometry around the metallic ions. Each copper(II) atom is bound to two sulfur atoms from thiocyanate ligands, which are also coordinated to cobalt(II) ions.These sulfur atoms lie in the axial positions of a deformed octahedron around the copper(II), the equatorial plane being –lled with the four nitrogen atoms of the cyclam ligand. The cobalt(II) atom possesses Fig. 1 ORTEP drawing of the molecules. [Cu(cyclam)][Co(NCS)4] Thermal ellipsoids of the non-hydrogen atoms are drawn at the 50% probability level.H atoms are drawn as small circles of arbitrary size. Selected bond lengths and angles [°] : CuwN8 2.004(3), CuwN1 [”] 2.014(3), CuwN4 2.020(3), CuwN11 2.025(3), CuwS1 2.891(3), CuwS2 3.160(3), CowN21 1.951(4), CowN41 1.951(4), CowN31 1.955(3), CowN22 1.976(3), S1wC21 1.625(4), C21wN21 1.152(5), S2wC22 1.634(4), C22wN22 1.141(5), S3wC31 1.625(4), C31wN31 1.155(5), S4wC41 1.612(5), N1wC2 1.479(5), N1wC14 1.488(5), C2wC3 1.491(7), C3wN4 1.480(5), C41wN41 1.138(6), N4wC5 1.486(5), C5wC6 1.498(7), C6wC7 1.514(7), C7wN8 1.476(5), N8wC9 1.491(5), C9wC10 1.482(6), C10wN11 1.478(5), N11wC12 1.480(5), C12wC13 1.509(7), C13wC14 1.504(7).N8wCuwN1 177.18(13), N1wC2wC3 107.9(3), N8wCuwN4 94.31(13), N4wC3wC2 108.8(4), N1wCuwN4 85.57(13), N41wC41wS4 178.5(5), N8wCuwN11 85.88(12), C31wN31wCo 172.5(3), N1wCuwN11 94.10(13), C3w N4wC5 113.2(3), N4wCuwN11 177.11(14), C3wN4wCu 107.2(3), S1wCuwS2 169.96(12), C5wN4wCu 117.6(3), N21wCowN41 114.1(2), N4wC5wC6 111.7(4), N21wCowN31 107.9(2), C5w C6wC7 115.2(4), N41wCowN31 105.8(2), N8wC7wC6 111.3(3), N21wCowN22 109.8(2), C7wN8wC9 113.0(3), N41wCowN22 103.8(2), C7wN8wCu 117.9(2), N31wCowN22 115.44(14), C9w N8wCu 107.6(2), N21wC21wS1 179.3(3), C10wC9wN8 108.6(3), C21wN21wCo 171.8(3), N11wC10wC9 108.7(3), N22wC22wS2 178.7(4), C12wN11wC10 113.2(3), C22wN22wCo 169.7(3), C12w N11wCu 117.7(3), N31wC31wS3 178.8(4), C10wN11wCu 105.9(2), C31wN31wCo 172.5(3), N11wC12wC13 111.9(4), C2wN1wC14 113.2(3), C12wC13wC14 114.7(4), C2wN1wCu 106.9(3), N1w C14wC13 111.9(3), C14wN1wCu 117.3(3). New J.Chem., 1999, 267»269 267Fig. 2 A perspective view of the chain of M[Cu(cyclam)]- running along the b axis. Black circles indicate [Co(NCS)4]Nn , cobalt(II) atoms and grey circles indicate copper(II) atoms. two bridging and two non-bridging thiocyanate ligands, each attached to the metal by the nitrogen atom, giving rise to a deformed tetrahedral environment around the cobalt(II) centre.Both molecular building blocks possess two points of connectivity ; as such, they assemble as a 1 : 1 complex consisting of an in–nite helical bimetallic chain running along the b axis, in which the copper(II) and cobalt(II) atoms alternate, as shown in Fig. 2. Bond lengths in the planar cyclam ligand and CuwN distances (see Table 3 in supplementary material) are in accordance with the ones found in the literature for similar copper(II) complexes.14 The CuwS axial distances are equal to 2.891(3) and 3.160(3) respectively, for CuwS1 and CuwS2.These ”, distances are close to the ones found in other complexes in which the sulfur atoms adopt the axial positions of a distorted octahedron around copper(II).15 The S1wCuwS2 angle is equal to 169.96(12)°.The CowN distances are in agreement with those found in the literature for other cobalt(II) thiocyanato complexes.16 The thiocyanato groups are almost linear with a mean value of the NwCwS angles of 178.8(4)°. The CowNwC linkages are bent with angles varying from 169.7(3)° to 172.5(3)°. These structural features have already been observed in other thiocyanato-containing metal complexes. 17 The SwC average distance of 1.625(13) and CwN ” average distance of 1.145(9) are in accordance with the ” values given in the literature.18 The Co … … … Co, Cu … … … Cu and Co … … … Cu distances within the chain are equal to 7.619(3), 10.254(3) and 6.909(3) ”, respectively. The Co … … … Co, Cu … … … Cu and Co … … … Cu distances between the chains are equal to 7.701(3), 7.472(3) and 6.361(2) respectively. The chains are aligned parallel with ”, respect to each other in the crystal structure.Magnetic susceptibility data on the polycrystalline sample were collected with a MPMS Quantum Design SQUID magnetometer in the temperature range of 260»1.7 K at a –eld of 1000 Oe. The sample holder was a quartz tube. The data were corrected for the experimentally determined diamagnetism of the sample holder, whilst the diamagnetic contribution from the sample was calculated from Pascalœs constants.The magnetic behaviour is presented in Fig. 3. The room temperature value (1.95 cm~3 mol~1 K), is slightly lower than the vMT value expected for isolated copper(II) and cobalt(II) ions [2.25 cm~3 mol~1 K, assuming g\2 for copper(II) and cobalt(II)].Upon cooling from room temperature, the values vMT decrease slowly down to 30 K and sharply below 15 K. This feature could be explained by a weak antiferromagnetic interaction between the metallic centres, probably together with an antiferromagnetic coupling between the chains. This must be con–rmed by the –t of the experimental data to an adapted law. Because of the non-compensation of spins of copper(II) and cobalt(II) and from a physical point of (S\12) (S\32) view, the compound is expected to behave as a ferrimagnetic chain at temperatures below 1 K.Further magnetic measurements at very low temperatures (T \1 K) should allow us to con–rm this behaviour. The experimental susceptibility data were analysed by an expression derived by Drillon and co-workers19 that is valid for bimetallic alternating ferrimagnetic chains with quantum classical spins.This expression is derived for quantum S\12 Fig. 3 Thermal dependence of the molar susceptibility for vM in a 1000 Oe applied –eld. The solid line M[Cu(cyclam)][Co(NCS)4]Nn shows the best –t using eqn. (1). Insert shows the thermal dependence of the product of the molar susceptibility and temperature under vM the same conditions.spins like the spin of copper(II), alternating with a classical spin, the spin of cobalt(II) in our case : vM\ NA gCu 2 b2 4kB T cr2 3a(a2]2)sinh a[6a(a]sinh a) cosh a]3(a2]1)sinh2 a]a2(a2]3) 3a3(a]sinh a) d [2r a cosh a[sinh a a2 ] sinh a a (1) with r\ 2gCo SCo gCu and a\[ JSCo kB T . is the Avogadro constant, b is the Bohr magneton, is NA kB the Boltzmann constant, and J is the exchange coupling constant.The best –t gave a g value of 2.42(1) for cobalt(II) and 2.12(1) for copper(II), and a J value of [1.48(3) cm~1 (coefficient of determination r2\0.9996), as indicated by the solid curve in Fig. 3. Attempts to introduce mean –eld theory in this equation, in order to determine the interaction between the chains, failed.This very weak character of the antiferromagnetic interaction is due to intrinsic structural features and especially to the large distance between the copper(II) and cobalt(II) centres, which leads to a poor overlap of the magnetic orbitals. In conclusion, this work presents the synthesis, crystal structure and preliminary magnetic characterization of the –rst bimetallic chain containing thiocyanate as bridging ligands.The weak antiferromagnetic interaction observed between the magnetic centres could –nally reveal a heterometallic ferrimagnetic chain at low temperatures. Acknowledgements We express our gratitude to Dr. Marc Drillon for helpful discussions concerning the magnetic data and to the Swiss National Science Foundation for –nancial support through Project No. 20-45750.95. Notes and references § Non-SI units employed: 104 G\1 T; 1 Oe\79.6 A m~1. î Anal. calcd for C, 30.29 ; H, 4.36 ; N, 20.18. C14H24N8S4CoCu: Found: C, 29.60 ; H, 4.40 ; N, 19.75. 268 New J. Chem., 1999, 267»269° The infrared spectrum was recorded on a Bio-Rad IR.FT spectrophotometer, in the 4000»250 cm~1 range, using pressed KBr pellets containing 1% by mass of the sample.“ Crystal data for 1: M\555.12 g mol~1, mono- C14H24N8S4CoCu, clinic, space group a\8.568(2), b\14.463(3), c\18.562(4) P21/c, ”, b\96.62(3)°, U\2284.8(9) Z\4, Mg m~3, ”3, Dc\1.614 l\2.040 mm~1, T \296(2) K. Of 5845 re—ections collected [Enraf- Nonius CAD-4, radiation (k\0.71073 5488 unique re—ec- MoK a ”)], tions were used for the Patterson interpretation routine of SHELXS- 97 (G.M. Sheldrick, Acta Crystallogr., Sect. A, 1990, 46, 467). Re–nement was carried out using full-matrix least-squares techniques, with SHELXL-97 (G. M. Sheldrick, SHEL XL -97, Program for Crystal Structure Re–nement, University of Goé ttingen, Goé ttingen, Germany, 1997). Hydrogen atomic positions were calculated at distances of 0.93 and were –xed by using a riding model in the re–nement.”R \0.0465 and Rœ(all data)\0.1436 with 346 parameters. CCDC reference number 440/089. See http ://www.rsc.org/suppdata/nj/1999/ 267/ for crystallographic –les in .cif format. 1 See for example: P. Day, Science, 1993, 261, 431; J. S. Miller and A. J. Epstein, Angew. Chem., Int. Ed. Engl., 1994, 33, 385; D. Gatteschi, Adv. Mater., 1994, 6, 635; O.Kahn, Adv. Inorg. Chem, 1996, 43, 179. 2 O. Kahn, Molecular Magnetism, VCH Press, Weinheim, Germany, 1993. 3 G. De Munno, G. Bruno, F. Nicolo` , M. Julve and J. A. Real, Acta Crystallogr., Sect. C, 1993, 49, 457; M. Julve, M. Verdaguer, G. De Munno, J. A. Real and G. Bruno, Inorg. Chem., 1993, 32, 795; P. C. Healy, C. Pakawatchai, R. I. Papasergio, V. A. Patrick and A.H. White, Inorg. Chem., 1984, 23, 3769. 4 J. A. R. Navarro, M. A. Romero, J. M. Salas, M. Quiros and E. R. T. Tiekink, Inorg. Chem., 1997, 36, 4988. 5 J. Ribas, C. Diaz, X. Solans and M. Font-Bardia, Inorg. Chim. Acta, 1995, 231, 229. 6 S. Ferlay, G. Francese, H. W. Schmalle and S. Decurtins, Inorg. Chim. Acta, in press. 7 M. B. Hursthouse, K. J. Izod, M. A. Mazid and P. Thornton, Polyhedron, 1990, 9, 535. 8 R. Vincente, A. Escuer, J. Ribas and X. Solans, J. Chem. Soc., Dalton T rans., 1994, 259; M. Montfort, C. Bastos, C. Diaz, J. Ribas and X. Solans, Inorg. Chim. Acta, 1994, 218, 185; A. Escuer, S. Kumar, F. Mautner and R. Vicente, Inorg. Chim. Acta, 1998, 269, 313. 9 J. N. McElearny, L. L. Balagot, J. A. Muir and R. D. Spence, Phys. Rev. B, Condens. Mater., 1979, 19, 306. 10 J. Lu, T. Paliwala, S. Lim, C. Yu, T. Niu and A. J. Jacobson, Inorg. Chem., 1997, 36, 923; F. Lloret, G. De Munno, M. Julve, J. Cano, R. Ruiz and A. Caneschi, Angew. Chem, Int. Ed. Engl., 1998, 37, 135. 11 J. Ribas, C. Diaz, R. Costa, J. Tercero, X. Solans, M. Font-Bardia and H. Stoeckli-Evans, Inorg. Chem., 1998, 37, 233. 12 M.-C. Chow and T. C. W. Mak, Aust. J. Chem., 1992, 43, 1307. 13 R. A. Bailey, S. L. Kosak, T. W. Michelson and W. N. Mills, Coord. Chem. Rev., 1971, 6, 407. 14 P. A. Tasker and L. Sklar, J. Crystallogr. Mol. Struct., 1975, 5, 329; X. Chen, G. Lay, R. D. Willet, T. Howks, S. Molnar and K. Brewer, Acta Crystallogr., Sect. C, 1996, 52, 1924. 15 M. Kabesova, R. Boca, M. Melnik, D. Valigura and M. Dunaj- Jurco, Coord. Chem. Rev., 1995, 140, 115. 16 J. S. Wood and R. K. McMullan, Acta Crystallogr., Sect. C, 1984, 40, 1803. 17 A. Sedov, J. Kozisek, M. Kabesova, M. Dunaj-Jurco, J. Kazo and J. Garaj, Inorg. Chim. Acta, 1983, 75, 73; F. H. Cano, S. Garcia- Blanco and A. Guerrero-Laverat, Acta Crystallogr., Sect. B, 1976, 32, 1526; M. Nardelli, A. Gasparri, A. Musati and A. Manfedotti, Acta Crystallogr, 1966, 21, 910. 18 H. van Rooyen and J. C. A. Boeyens, Acta Crystallogr., Sect. B, 1975, 31, 2933. 19 J. Cureç ly, R. Georges and M. Drillon, Phys. Rev. B, 1986, 33, 6243. L etter 9/00125E New J. Chem., 1999, 267»269 269

ISSN:1144-0546    

DOI:10.1039/a900125e

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

L e t t e r Titanium imido complexes with 1,3,5-triazacyclohexane ligands : syntheses, solution dynamics and solid state structures Paul J. Wilson, Paul A. Cooke, Alexander J. Blake, Philip Mountford*§ and Martin Schroé der* School of Chemistry, University of Nottingham, Nottingham, UK NG7 2RD. E-mail : or martin.schroder=nottingham.ac.uk philip.mountford=chemistry.oxford.ac.uk Received (in Cambridge, UK) 14th October 1998, Accepted 18th November 1998 The multi-gram scale syntheses of the –rst 1,3,5-triazacyclohexane imido complexes (R = But, [Ti(NR)(R3 º tach)Cl2 ] Rº = Me, But) are described together with the C6H3Pr2 i -2,6 ; X-ray structures of and [Ti(NBut)(Me3tach)Cl2 ] the complexes of exhibit [Ti(NBut)(Bu3 t tach)Cl2 ]; Me3 tach dynamic NMR behaviour an unusual trigonal twist of the øia facially coordinated ligand and Me3tach (Me3tach Bu3 t tach= and tri-tert-butyl-1,3,5-triazacyclohexane, 1,3,5-trimethylrespectively). There is a great deal of interest in the development of new early transition metal complexes for synthetic and catalytic purposes.1 Of particular importance are compounds of the type (M\d0 metal centre ; ligand [M(Ln)X2] Ln\supporting set ; X\halide or hydrocarbyl) in which the frag- cis-MX2 ment is a well recognised structural pre-requisite for the stoichiometric or catalytic transformation of organic substrates, including alkene polymerisation.A wide range of support- Ln ing ligand sets have been employed, including bis(g5-cyclopentadienyl), cyclopentadienyl-amido and -amino, bis(amido), bis(alkoxide), SchiÜ-base and tetraazamacrocyclic moieties.2,3 Of relevance to this contribution are recently reported group 5 and 6 compounds that feature one or two imido cis-MX2 (NR where R\alkyl or aryl) groups in the supporting Ln ligand set.4,5 Although now well established, the reaction chemistry of Group 4 imido complexes has focused almost exclusively on transformations involving the M\NR linkage, rather than employing the imido group simply as an ancilliary ligand.6,7 Very recently, however, we reported two classes of monochloride complexes (L\amidinato or [Ti(NR)(L)Cl(py)n] cyclopentadienyl) in which the imido group readily supports halide metathesis reactions giving amido, aryloxide, cyclopentadienyl, borohydride, alkyl and other derivatives.8,9 Since Group 4 imides with units are likely to be signi–- cis-MX2 cantly more valuable targets than the mono-chloride complexes, we recently reported the 1,4,7-triazacyclononane complexes (R\H or Me).10 These isolobal [Ti(NBut)(R3[9]aneN3)Cl2] analogues of [Ti(g5- allow ready substitution of C5H5)2Cl2] the halide ligands and already suggest an extensive reaction chemistry.We report here new imido-supported cis-dichloride complexes using 1,3,5-triazacyclohexane ligands R3tach (R\Me or But) which are more conveniently prepared in large quantities than the 1,4,7-triazacyclononanes. Complexes of these ligands have recently been reported for middle to late transition metals and some p-block metals,11h16 but complexes of Group 4 or any with metal imido fragments have not been described previously.The synthesis and proposed structures of the new compounds are shown in Scheme 1. The starting titanium imido compounds (n\2, R\But 1a; n\3, [Ti(NR)Cl2(py)n] R\ 1b)17 and the 1,3,5-triazacyclohexane ligands C6H3Pr2 i -2,6 (R\Me or But)18,19 can all be readily prepared in R3tach ìone-potœ reactions on large scales. Addition of to a Me3tach solution of 1a or 1b in gives quantitative formation of CH2Cl2 spectroscopically pure (R\But 2a or [Ti(NR)(Me3tach)Cl2] 2b) within hours at room temperature.î Reac- C6H3Pr2 i -2,6 tions of the bulkier ligand with 1a and 1b are some- Bu3 t tach what slower as might be expected, and give the analogous products (R\But 3a or [Ti(NR)(Bu3 t tach)Cl2] C6H3Pr2 i -2,6 3b) in high yields (ca. 86%). Analytically pure samples were obtained by recrystallisation from mixtures.CH2Cl2»hexane The new compounds 2a»3b can all be prepared on a multigram scale. In a typical preparation, reaction of 2.55 g of and 3.48 g of 1a aÜords 4.58 g of recrystallised 3a Bu3 t tach (86%). DiÜraction-quality crystals of the tert-butylimido complexes 2a and [Ti(NBut)(Me3tach)Cl2] [Ti(NBut)(Bu3 t tach)Cl2] 3a were obtained by layering a solution of 2a in dichloromethane with hexane or by cooling a saturated solution of 3a in toluene.° The molecular structure of 3a is shown in Fig. 1; that of 2a is broadly analogous. Selected bond distances and angles for both complexes are given in the caption of Fig. 1. The structures of 2a and 3a comprise pseudo-octahedral titanium centres with (R\Me or But) rings and fac-R3tach mutually cis tert-butylimido and chloride ligands.The respective and Ti»Cl distances for the two compounds are Ti»Nimide identical within error, whereas the distances are sig- Ti»Nring ni–cantly longer (by ca. 0.05»0.09 for 3a, presumably re—ec- Aé ) ting the increased steric demands of The Bu3 t tach. Ti»Nring distances in are intermediate [Ti(NBut)(Me3[9]aneN3)Cl2]10 between those of 2a and 3a.The Ti»Cl distances for the derivative (av. 2.393 are somewhat longer Me3[9]aneN3 Aé ) than for 2a and 3a whereas the distance [1.694(2) Ti»Nimide Aé ] is comparable. The Cl»Ti»Cl angle of 95.75(4)° in is smaller than for either [Ti(NBut)(Me3[9]aneN3)Cl2] R3tach Scheme 1 Reagents and conditions : i, r.t., 14 h, Me3tach, CH2Cl2 , 94% for 2a, r.t., 7 d, 86% for 3a; ii, Bu3 t tach, CH2Cl2, Me3 tach, r.t., 14 h, 89% for 2b; r.t., 7 d, 86% for 3b.CH2Cl2, Bu3 t tach, CH2Cl2 , New J. Chem., 1999, 271»273 271Fig. 1 Displacement ellipsoid (25% probability) plot for 3a with hydrogen atoms omitted. Selected [Ti(NBut)(Bu3 t tach)Cl2] bond lengths and angles (°) for 3a with the corresponding values (Aé ) for 2a in single brackets : Ti(1)»N(1) 1.692(4) [Ti(NBut)(Me3tach)Cl2] [1.699(4)], Ti(1)»N(2) 2.292(4) [2.241(5)], Ti(1)»N(4) 2.290(4) [2.247(5)], Ti(1)»N(6) 2.513(4) [2.424(4)], Ti(1)»Cl(1) 2.351(2) [2.356(2)], Ti(1)»Cl(2) 2.351(1) [2.363(2)] ; N(2)»Ti(1)»N(4) 61.2(1) [61.4(2)], N(2)»Ti(1)»N(6) 58.5(1) [59.0(2)], N(4)»Ti(1)»N(6) 58.4(1) [58.9(2)], Cl(1)»Ti(1)»Cl(2) 99.44(7) [102.72(6)], Ti(1)»N(1)»C(10) 164.0(4) [176.3(4)], N(1)»Ti(1)»Cl(1) 101.6(2) [106.3(2)], N(1)»Ti(1)» Cl(2) 100.7(1) [106.2(2)].derivative, with that in 3a being slightly smaller than in 2a. The geometries of the ligands are similar in each R3tach complex and comparable to those found in previous examples. 11h15 The solution 1H and 13C NMR spectroscopic data for the complexes 3a and 3b are fully consistent with the Bu3 t tach solid state structures and show sharp resonances for the imido N-substituents along with two types of ring NBut substituents and diastereotopic methylene protons on the triazacyclohexane ring.In contrast, the resonances for the Me3tach ligand in (R\But 2a or [Ti(NR)(Me3tach)Cl2] C6H3Pr2 i -2,6 2b) are very broad at room temperature, whereas those for the imido N-substituents are sharp and temperature independent.Detailed variable temperature NMR spectroscopic experiments [including spin saturation transfer (SST) and 1H lineshape (rate constant) analysis] were used to characterise the dynamic process. The NMR spectra for 2a and 2b are consistent with an inplace trigonal twist of the ligand. In the slow Me3tach exchange limit (268 K) the 1H NMR spectra for both complexes show two sharp resonances in a ratio 3H : 6H for the methyl groups trans- and cis- to the imido ligand, respectively.On increasing the probe temperature these groups undergo site exchange (con–rmed by SST) with the rate constants for the cis-Me]trans-Me transformations being half of those for trans-Me]cis-Me, as expected.20 In addition, SST 1H NMR experiments con–rm exchange between the two types of ìdownœ methylene hydrogens of and also between Me3tach, the two types of ìupœ hydrogens.Signi–cantly, there was no exchange between the ìdownœ and ìupœ hydrogens indicating that the dynamic process does not involve dissociation of the triazacyclohexane ring. Analysis of the methyl group exchange rate constants (using standard Eyring plots21) for 2a aÜords: *Hî\65.7^0.5 kJ mol~1; *Sî\9.6^2 J mol~1 K~1; *Gî\62.9^0.5 kJ mol~1 at 292 K. The eÜectively zero value for *Sî lends further support to a non-dissociative mechanism for the exchange process.Previous studies of —uxional processes for a range of bis(acetylacetonato)titanium complexes proposed to undergo trigonal twists found negative *Sî values in the range ca.[55 to [90 J mol~1 K~1.22 The more positive *Sî for 2a indicates a less-ordered transition state compared to those for the previously studied titanium systems. The trigonal twist mechanism has long been established for bis(bidentate ligand) chelate complexes of titanium(IV),22,23 but to our knowledge the —uxional processes for 2a and 2b are the –rst such examples for any triazacycloalkane ligand.That the complexes 2a and 2b are —uxional whereas 3a and 3b and those of the larger rings (R\H or Me)10 are R3[9]aneN3 not, is consistent with previous reports concerning the importance of steric factors and ligand bite angle on the energies of activation for such processes.22,24 The proposed mechanism presumably proceeds via a trigonal prismatic transition state (or intermediate).Ground state trigonal prismatic six-coordinate transition metal complexes are well established and their structures have been ML6 rationalised theoretically.25 It is now generally accepted that a trigonal prismatic geometry can be favoured over the octahedral alternative for d0 complexes where L is a r-only (or is only weakly p-donating) ligand.For complexes where ML6 the ligand set contains eÜective p-donors the octahedral geometry is preferred.26 This consideration of electronic eÜects appears to account for the ground-state pseudooctahedral geometries found for 2 and 3 which feature strongly p-donating organoimido ligands. However, it appears that the trigonal prismatic alternatives are accessible on the NMR timescale for the less sterically crowded homologues Me3tach 2.In summary we have –rmly established the –rst 1,3,5-triazacyclohexane complexes both for Group 4 metal and for any metal imido fragment. The complexes 2a»3b are new isolobal analogues of metallocene dihalides and can be readily prepared in synthetically useful quantities. Both ring and imido substituents can be varied, and the new compounds therefore promise a rich and diverse reaction chemistry.The —uxional complexes 2a and 2b represent the –rst examples of a trigonal twist mechanism for any pseudo-octahedral triazacycloalkane complex. Synthetic and catalytic investigations of all the new compounds and their homologues are currently underway in our laboratories. Acknowledgements We thank the EPSRC, Leverhulme Trust, Royal Society and University of Nottingham for support, and Professors O.Eisenstein and B. E. Mann for helpful comments. Philip Mountford is the Royal Society of Chemistry Sir Edward Frankland Fellow for 1998»1999. Notes and references § Present address : Inorganic Chemistry Laboratory, South Parks Road, Oxford, UK OX1 3QR. î Satisfactory spectroscopic and analytical data have been obtained for the new compounds.° Crystal data for 2a and 3a: crystals were mounted in a –lm of RS3000 per—uoropolyether oil (Hoechst) on a glass –bre and transferred to a Stoeé Stadi-4 four-circle diÜractometer equipped with an Oxford Cryosystems low-temperature device.27 Data were collected at 150(2) K using graphite-monochromated Mo-Ka radiation (j\0.710 73 and u»h scans in the range 2.5OhO25°, and semi- Aé ) empirical absorption corrections based on t-scans were applied.Crystallographic calculations were performed using SIR9228 and CRYSTALS-PC;29 R\&pFo o[oFcp/&oFo o, Rw\M&w(Fo [Fc)2/&w(Fo)2N1@2. 272 New J. Chem., 1999, 271»2732a: M\319.14, orthorhombic, space group C10H24Cl2N4Ti, a\7.053(2), b\13.818(5), c\16.452(5) U\1603.4(7) P212121, Aé , Z\4, k\0.85 mm~1, pale yellow rod of dimensions Aé 3, 0.40]0.16]0.13 mm, 1435 independent observed [I[2p(I)] re—ections used in re–nement, no.of parameters re–ned 154, (Rmerge\0.02) full-matrix least squares on F with Chebychev polynomial weighting scheme, R\0.0455, GOF\1.119, absolute structure Rw\0.0475, parameter 0.09(9), –nal largest residual peaks 0.36 (*/p)max\0.003, and [0.63 e Aé ~3. 3a: M\445.27, monoclinic, space group C19H42Cl2N4Ti, P21/n, a\9.871(5), b\16.677(11), c\15.020(11) b\103.51(5)°, Aé , U\2404(2) Z\4, k\0.59 mm~1, yellow block of dimensions Aé 3, 0.35]0.27]0.26 mm, 3079 independent observed [I[2p(I)] re—ections used in re–nement, no. of parameters re–ned 262, full-matrix least squares on F with unit weights, R \0.068, Rw\0.067, GOF\0.938, –nal 0.003, largest residual peaks 0.41 and (*/p)max [0.63 e Aé ~3.CCDC reference number 440/083. See http ://www.rsc.org/ suppdata/nj/1999/271/for crystallographic –les in cif format. 1 Applied Homogeneous Catalysis with Organmetallic Compounds; ed. B. Cornils and W. A. Herrmann, VCH, Weinheim, 1996. 2 M. Bochmann, J. Chem. Soc., Dalton T rans., 1996, 255. 3 H. H. Brintzinger, D. Fischer, R. Mué lhaupt, B. Rieger and R. M. Waymouth, Angew. Chem., Int. Ed. Engl., 1995, 34, 1143. 4 M. P. Coles, C. I. Dalby, V. C. Gibson, W. Clegg and M. R. J. Elsegood, J. Chem. Soc., Chem. Commun., 1995, 1709. 5 S. Scheuer, J. Fischer and J. Kress, Organometallics, 1995, 14, 2627. 6 P. Mountford, Chem. Commun., 1997, 2127. 7 D. E. Wigley, Prog.Inorg. Chem., 1994, 42, 239. 8 P. J. Stewart, A. J. Blake and P. Mountford, Organometallics, 1998, 17, 3271. 9 S. C. Dunn, P. Mountford and D. A. Robson, J. Chem. Soc., Dalton T rans., 1997, 293. 10 P. J. Wilson, A. J. Blake, P. Mountford and M. Schroé der, Chem. Commun., 1998, 1007. 11 R. D. Koé hn, M. Haufe, G. Kociok-Koé hn and A. C. Filippou, Inorg. Chem., 1997, 36, 6064. 12 N. L. Armanasco, M. V. Baker, M. R. North, B. W. Skelton and A. H. White, J. Chem. Soc., Dalton T rans., 1998, 1145. 13 R. D. Koé hn, G. Seifert and G. Kociok-Koé hn, Chem. Ber., 1996, 129, 1327. 14 R. D. Koé hn, G. Kociok-Koé hn and M. Haufe, Chem. Ber., 1996, 129, 25. 15 M. Haufe, R. D. Koé hn, R. Weimann, G. Seifert and D. Zeigan, J. Organomet. Chem., 1996, 520, 121. 16 G. Willey, T.J. Woodman, U. Somasundaram, D. R. Aris and W. Errington, J. Chem. Soc., Dalton T rans., 1998, 2573. 17 A. J. Blake, P. E. Collier, S. C. Dunn, W.-S. Li, P. Mountford and O. V. Shishkin., J. Chem. Soc., Dalton T rans., 1997, 1549. 18 J. M. Lehn, F. G. Riddell, B. J. Proce and I. O. Sutherland, J. Chem. Soc. A, 1967, 387. 19 J. Graymore, J. Chem. Soc., 1924, 125, 2283. 20 M. L. H. Green, L.-L. Wong and A. Sella, Organometallics, 1992, 11, 2660. 21 J. Sandstroé m, Dynamic NMR Spectroscopy, Academic Press, London, 1992. 22 D. C. Bradley and C. E. Holloway, J. Chem. Soc. A, 1969, 282. 23 N. Serpone and R. C. Fay, Inorg. Chem., 1967, 6, 1835. 24 E. L. Muetterties and L. J. Guggenberger, J. Am. Chem. Soc., 1972, 94, 8046. 25 For leading references see : M. Kaupp, Chem. Eur. J., 1998, 4, 1678; S. Kleinhenz, V. Pfennig and K. Seppelt, Chem. Eur. J., 1998, 4, 1687. 26 M. H. Chisholm, I. P. Parkin, W. E. Streib and O. Eisenstein, Inorg. Chem., 1994, 33, 812. 27 J. Cosier and A. M. Glazer, J. Appl. Crystallogr., 1986, 19, 105. 28 A. Altomare, G. Cascarano, G. Giacovazzo, A. Guagliardi, M. C. Burla, G. Polidori and M. Camalli, J. Appl. Crystallogr., 1994, 27, 435. 29 D. J. Watkin, C. K. Prout, J. R. Carruthers and P. W. Betteridge, CRYSTALS Issue 10, Chemical Crystallography Laboratory, University of Oxford, 1996. L etter 8/07985D New J. Chem., 1999, 271»273 273

ISSN:1144-0546    

DOI:10.1039/a807985d

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Liquid-crystalline, polycatenar complexes of silver(I) : dependence of the mesomorphism on the ligand and the anion Bertrand Donnio§ and Duncan W. Bruce* School of Chemistry, University of Exeter, Stocker Road, Exeter, UK EX4 4QD. E-mail : d.bruce=exeter.ac.uk Received 30th November 1998, Accepted 7th January 1999 We have synthesised several series of poly(alkoxy)stilbazole ligands, and investigated the thermal behaviour of their complexes with some silver salts. Most of the complexes, with the general formula where X\dodecyl [AgL2][X], sulfate (DOS) or tri—uoromethanesulfonate (OTf), and L\poly(alkoxy)stilbazole, were found to be liquid-crystalline over a wide-temperature range, the vast majority of them being low-melting.The mesomorphic properties were mainly dominated by the formation of hexagonal columnar phases and to a lesser extent, of (Colh) bicontinuous cubic phases; monotropic nematic and smectic C phases were also seen for one structural type of Ia3 6 d compound only.The mesophases were assigned on the basis of their optical textures under polarised optical microscope (POM) prior to being identi–ed by X-ray diÜraction (XRD).The temperatures and enthalpies of the transitions were determined by diÜerential scanning calorimetry (DSC). The mesomorphism was found to be strongly dependent on the substitution pattern of the ligand, that is on the number, position and length of alkoxy chains, as well as on the type of the counter anion. We explain the mesomorphism in terms of ionic interactions and in terms of mean interfacial curvature between the hydrophobic, aliphatic chains and the rigid, aromatic, central core, by analogy to lyotropic systems.Introduction For many years, thermotropic cubic phases were scarcely studied, essentially due to their rare appearance in conventional liquid-crystalline materials,1 and thus, very little information was known about their structure. More recently, they have been more extensively studied.The two general types of cubic phases2,3 commonly found in lyotropic systems are now being observed in many thermotropic systems. For instance, micellar thermotropic cubic phases (analogous to the lyotropic and phases) have been undoubtedly identi–ed in I1 I2 some dendrimeric macromolecules,4 in several mesogens containing a carbohydrate or a polyhydroxy moiety,5 and also, most probably, in some polar, globular mesogens.6 Bicontinuous thermotropic cubic phases (analogous to the lyotropic phase) have been observed in some more conventional V2 systems, for instance in calamitic systems such as the 4- alkoxy-3-nitro- (or -cyano) biphenyl-4@-carboxylic acids and 1,2-bis(4-alkoxybenzoyl)hydrazine,7 disc-like compounds such as the hexa-O-pentanoylscyllo inositol and a few derivatives of the ellagic acids,8 as well as in polyol and carbohydrate liquid crystals,5a,9 silver stilbazole complexes,10 polycatenar and biswallow- tailed mesogens,11 dialkylphosphate salts,12 oligoalkylene amides,13 rod-coil calamitic mesogens14 and, probably, in some silver thiolate complexes.15 Other mesomorphic systems have been found to form cubic phases in the melt, such as diblock co-polymers16 in which the mesophases occur as a function of the volume fraction of the two blocks, and polyelectrolyte»surfactant systems.17 With such a variety of molecular structures, it is not straightforward to predict the occurrence of these three-dimensional mesophases and thus, the possibility for molecular design remains low.However, one can notice that most of the compounds listed above § Present address : Institut fué r Makromolekulare Chemie, Albert- Ludwigs Universitaé t, Hermann-Staudinger-Haus, Stefan Meier Strasse 31, Freiburg i. Br., 79104, Germany. possess a common denominator, in that they are amphiphilic in nature. Indeed, all the molecules listed above are made of two (or more) diÜerent constituent blocks which are structurally and chemically diÜerent.The concept of amphiphilicity and liquid-crystallinity was, for a long time, used exclusively to describe the behaviour of lyotropic liquid crystal systems. More recently, the amphiphilic character of thermotropic mesogens has been taken into account and is felt to be a very important driving force for thermotropic mesophase formation.18 Consequently, it is now common to remark on the similarities in the mesomorphism of thermotropic and lyotropic mesogens, rather than look to the diÜerences.This approach, developed in two major papers18 which emphasise the amphipatic character of the molecules, should progressively set up the foundation of a new concept unifying the description of mesophases on the one hand, and leading to enhanced understanding of phase formations and transitions based on the evolution of the shape of the interfaces on the other.This concept has been the subject of an updated comprehensive review recently published, which highlights the importance of micro-segregation in nonconventional mesogens, and further extends the essential role of amphipathy in liquid crystals.19 Indeed, an important property of amphiphilic molecules, due to the diÜerent nature of their block units, is the ability to organise to form structures which divide the space into subspaces with diÜerent properties, the diÜerent sub-spaces thus being separated by interfaces. The degree of amphiphilic character, in addition to the degree of molecular anisometry, will in—uence the micro-segregation and aÜect the shape of the surface (interface) separating the diÜerent parts of the molecules.Bearing this in mind, a lamellar phase can, therefore, be described as a succession of planar surfaces, and a columnar phase as a two-dimensional arrangement of cylindrical surfaces. However, since this is purely a geometrical approach, care must be taken as these interfaces are not so precise and well-de–ned, and in addition to thermal agitation, diÜusion New J.Chem., 1999, 275»286 275occurs so that the interfaces can be modulated (smectic) or undulating (columnar), i.e. the interfaces are not planar, rather curved with a wave-like periodicity (see e.g. refs 11a and 12c). Furthermore, transitions from smectic or columnar organisation to nematic phases can be explained by molecular diÜusion from adjacent lamellar planes (S-to-N), or adjacent columns (Col-to-N) until the interfaces have totally vanished.In the last 15 years, there has been a growing attraction to the study of metallomesogens, as evidenced by an abundant literature.20 This interest arose because of the new properties that may be expected on the introduction of metals into liquid-crystalline materials.In addition, metals oÜer wider possibilities for structural variations (coordination, geometry) than simple organic molecules, and a large number of metal» ligand combinations can be envisaged to generate mesomorphic materials.20 Another interesting aspect is the search for the establishment of well de–ned structure»property relationships in order to be able to predict liquid-crystalline properties in metallomesogens, as well as to enrich our understanding of phase formation and molecular organisation, since additional parameters than those considered in organic liquid crystals may have to be taken into account.Many such studies have already been undertaken,20 and, in particular, an important part of our work has concentrated on the silver(I) complexes of alkoxystilbazoles.11,20b,21 In a previous study, we described the mesomorphism of some bis(3,4-dialkoxystilbazole) silver(I) alkyl sulfate complexes, and were particularly interested in determining the factors governing the formation of the cubic mesophases present in these systems.11a,21 A part of this work consisted of varying the length of the alkyl sulfate anion only, and we remarked that the mesophase range and type were dependent on whether or not the (alkyl sulfate) alkyl chain crossed the aliphatic/ aromatic interface.21 This was explained in terms of contribution (or not) of the anion alkyl chain to the molecular chain area, resulting in the modi–cation (increase) of the aliphatic/ aromatic ratio, and hence, of the degree of curvature at the core»chain interface.Thus, a possible way to obtain mesomorphic materials with ììcurved mesophasesœœ (cubic and columnar phases) is to vary the aliphatic/aromatic ratio of amphiphilic molecules in order to modify the associated interface, by analogy with the lyotropic mesophases. As was previously shown,11a this can be achieved, for example, by grafting more than one peripheral chain on the mesogen, which can be approximated, in some ways, to the addition of an apolar solvent to amphiphilic molecules. We now report a further development of this work in which we focus primarily on the structure of the ligand, and thus, we have synthesised –ve more series of polysubstituted stilbazoles in order to observe and try to rationalise some eÜects on the thermal behaviour of the resulting silver complexes.Furthermore, two diÜerent silver salts were used: silver(I) dodecyl sulfate (AgDOS) and silver(I) tri—uoromethanesulfonate (AgOTf). Results Synthesis and characterisation of the ligands and their complexes The synthetic pathway used to prepare the ligands and the silver complexes is shown in the scheme below. 2,4-Di-, 2,3,4- tri- and 2,4,5-tri-alkoxybenzaldehydes were obtained directly by etheri–cation of the corresponding di- or tri-hydroxybenzaldehydes (Scheme 1, route 2) with the appropriate 1- bromoalkane, under Williamsom ether conditions, in butanone or pentan-2-one (the latter being found considerably to ameliorate the yield of the O-alkylation reaction).However, 3,5-di- and 3,4,5-tri-alkoxybenzaldehydes were prepared diÜerently in three steps due to the high cost of the parent hydroxybenzaldehyde (Scheme 1, route 1). The –rst step consisted of the etheri–cation of methyl 3,5-di- and 3,4,5-tri-hydroxybenzoates with 1-bromoalkane. The resulting methyl poly(alkoxy)benzoates were then reduced to the corresponding benzyl alcohol using lithium tetrahydroborate (LiBH4) in a 5 : 1 EtOH»THF mixture, and the alcohols were –nally oxidised by pyridinium chlorochromate (PCC) in dichloromethane to furnish the desired poly(alkoxy)- benzaldehydes.22 The poly(alkoxy)stilbazoles were synthesised, as described previously, by condensation of the poly- (alkoxy)benzaldehydes with 4-methylpyridine using lithium diisopropylamide (LDA23) in THF, and followed by the dehydration of the intermediate alcohols by pyridinium toluenep- sulfonate salt (PPTS) in toluene at re—ux, the water formed being continuously removed from the reaction mixture by the use of a Dean»Stark apparatus.The complexes were subsequently prepared by stirring two equivalents of stilbazole with one equivalent of the silver salt in dichloromethane (for AgDOS), or in acetone (for AgOTf), at ambient temperature and with the vessel protected from light.The NMR spectroscopic data, 1H, 13C and 19F, were in agreement with the proposed structures of the intermediates and –nal substances, and the purity was con–rmed by elemental analysis (C,H,N,S). All the stilbazoles were trans as evidenced by 1H NMR, in which the coupling constant of the AB system was ca. 16»16.5 Hz. The complexes generate almost identical spectra to the free ligands (for which the trans con- –guration of the double bond was retained), although some appreciable diÜerences in chemical shifts (1H and 13C) indicated that the complexation was achieved. 13C and 19F NMR clearly showed that the tri—ate complexes were formed, as in the carbon spectra, a quartet was observed at d 121 corresponding to the C»F coupling, with a Hz, as well 1JCF\320 as a singlet in the —uorine spectra at d ca. [80.(CF3) Thermal behaviour of the stilbazoles The ligands will be abbreviated hereafter as St(n-x,y,z), where n represents the number of carbon atoms in the —exible, alkoxy chains, and x, y and z the position of the alkoxy chains grafted on the terminal benzene ring [e.g.St(6-3,4) represents 3,4-dihexyloxystilbazole]. None of the stilbazoles was mesomorphic. 24 The loss of the liquid-crystalline properties was nevertheless expected owing to the considerable reduction in the molecular anisotropy resulting from the grafting of more than one alkoxy chains in the 2- and/or 3-positions of the terminal benzene ring. However, they were noticeable for their low-melting temperatures (Table 6), some of them did even not crystallise but remained as oils instead, a desired condition since when complexed to silver, the metal complexes should equally show a substantial decrease in the transition temperatures.Thermal behaviour of the complexes In general, the melting points of the complexes described here were signi–cantly lower than that of the ìparentœ complexes, bis(4@-alkoxy-4-stilbazole)silver(I) dodecyl sulfate11b and tri- —ate,25 showing that the grafting of several alkoxy chains in the periphery of the complex can be a productive approach to obtain low-melting materials.A consequence of this important depression in the melting temperatures of the mesomorphic complexes is re—ected by the observation of larger mesomorphic domains than in the corresponding parent systems, since the clearing points remained almost unchanged.At –rst glance, it seems that the type of mesophase was more dependent on the substitution pattern of the ligand than the type of anion (vide infra). However, appreciable diÜerences between the DOS series and the OTf series of polycatenar mesogens were found for the melting points of the complexes, which were higher in the tri—ate series.This we largely attribute to a greater degree of ionic character in the silver»anion inter- 276 New J. Chem., 1999, 275»286Scheme 1 Synthesis of the stilbazoles and their complexes. i, butanone or pentan-2-one, re—ux, 24 h; ii, THF» CnH2n`1 Br, K2CO3 , LiBH4 , MeOH (5 : 1), re—ux, 12 h; iii, PCC, r.t., 4 h; iv, LDA, THF, [78 °C, 12 h; r.t., 2 h; PPTS, toluene, re—ux, 12 h; v, N2, CH2 Cl2, N2, H2O»HCl, AgX, or acetone, r.t., 4 h.CH2Cl2 action, thus requiring more energy to break down the lattice. This is supported by the stronger tendency of the tri—ate complexes to crystallise on cooling, when compared to their dodecyl sulfate analogues.11b,c,25 For all the mesomorphic complexes described here, the mesophases were characterised initially by optical microscopy, the nematic (N), smectic C cubic (Cub) and columnar (Col) phases giving character- (SC), istic textures1 and the transition temperatures detected by DSC analysis.Thermal behaviour of the silver(I) dodecyl sulfate complexes The mesomorphism of the [AgSt(n-3,4) has already 2][DOS] be described,11a but will be recalled here in order to act as a reference (Fig. 1). Thus, a bicontinuous cubic phase (Cub) Ia3 6 d was seen for 4OnO10 and a hexagonal columnar phase for 6OnO12. The domains of mesophase stability (Colh) were fairly large, especially those of the columnar phases, with melting points in the range 55 to 70 °C, and clearing points varying between 100 and 170 °C (Table 1).Complexation of St(n-3,5) to silver(I) dodecyl sulfate also induced strongly mesomorphic materials, which are unique examples of mesomorphic tetracatenar mesogens having four chains in the meta positions and only four aromatic rings.12 Thus, complexes with n\6»14 were mesomorphic showing one type of mesophase from almost room temperature up to 90»120 °C (Fig. 2). The low-melting behaviour is likely connected to the meta position of all the chains, and the absence of chains in the para positions. On the basis of the optical texture, the phase was assigned as columnar, and the presence of large homeotropic areas further suggested the symmetry of the mesophase to be hexagonal (uniaxial).Some of the complexes were obtained as amorphous solids, and on cooling from the isotropic liquid, they did not crystallise, although the mesophase was not —uid below the temperature where the compounds were expected to crystallise. In the following heat»cool cycles, no sign of crystal-to-columnar phase transition could be detected by DSC, and the optical texture of the mesophase was preserved as seen in the optical microscope.This peculiar thermal behaviour is consistent with the formation of anisotropic liquid-crystalline glasses (in the present case columnar glasses),26 although we were not able to obtain and from what turned out to be a poorly resolved Tg *Cp DSC trace. The tetradecyloxy homologue presents two meso- New J. Chem., 1999, 275»286 277Table 1 Transition temperatures and thermal data for [AgSt(n-3,5) and [AgSt(n-2,4) 2][DOS] 2][DOS] Complex n Transition T /°C *H/kJ mol~1 *Sm/J mol~1 K~1 [AgSt(n-3,5)2][DOS] 1 Crys»I 160 » » 4 Crys»Crys@ 98 8.4 22.6 Crys»I 119 39.2 99.0 6 Crys»Colh 32 4.8 15.6 Colh»I 95 1.5 4.0 7 ga»Colh 34 » » Colh»I 93 0.3 0.8 9 ga»Colh 29 » » Colh»I 113 1.9 5.0 12 Crys»Colh 32 14.3 46.9 Colh»I 107 1.8 4.8 14 Crys»Colh1 41 94.9 302.6 Colh1»Colh2 76 14.8 42.5 Colh2»I 123 1.9 4.8 [AgSt(n-2,4)2][DOS] 7 Crys»Crys@ 61 28.1 83.9 Crys@»I 71 18.9 54.9 (I»N) (65) » 14 Crys»I 57 90.3 273.4 (I»N) (53) ([0.8) ([2.6) (N»SC) (50) ([2.5) ([7.6) a g stands for frozen mesophase or anisotropic glass.phases (a reversible transition was clearly observed by DSC), although no textural change was observed at the transition.XRD studies corroborated the optical observations in the sense that no structural change was detected. This may suggest a transition between two hexagonal columnar phases (the –rst order DSC transition is consistent with a transition between two phases of the same symmetry), with a diÜerent type of order (or disorder) in the molecular organisation within the columns, these columns being nevertheless associated according to a 2-D hexagonal array.In the second set of complexes, namely [AgSt(n-2,4)2]- [DOS], only monotropic mesophases were observed (Table 1) ; for n\7 this was a nematic phase, while for n\14, both a nematic and a smectic C phase were observed. Both phases were recognised by their characteristic optical texture, and the N-to- transition was readily detected both by DSC and SC optical microscopy (e.g.the formation of transition bars in the Fig. 1 Phase diagram for the complexes [AgSt(n-3,4)2][DOS]. texture just before the transition). In both cases, crystallisation occured on reaching room temperature. Amongst the hexa-substituted substances, only the [AgSt(n- 3,4,5) complexes showed mesomorphic properties 2][DOS] (Table 2).All the complexes from n\4 to 14 showed the expected columnar mesophase (Fig. 3), over an extended temperature range (*T \100»120 °C). Most of them did not crystallise on cooling, retaining the texture of the mesophase, suggesting once more, the formation of anisotropic hexagonal columnar glasses, although the glass transition could not be observed by DSC.The complexes of the other two series of compounds, [AgSt(n-2,3,4) and [AgSt(n-2,4,5) melted 2][DOS] 2][DOS], directly to the isotropic liquid without showing any mesophase on heating, nor any metastable phase on cooling. These complexes showed strong supercooling eÜects, but they eventually crystallised after a few hours. Fig. 2 Phase diagram for the complexes [AgSt(n-3,5)2][DOS]. 278 New J. Chem., 1999, 275»286Table 2 Transition temperatures and thermal data for [AgSt(n-3,4,5)2]- [DOS] n Transition T /°C *H/kJ mol~1 *Sm/J mol~1 K~1 1 Crys»I 180 » » 4 Crys»Crys@ 42 2.3 7.3 Crys@»Colh 53 9.9 30.3 Colh»I 127 1.2 2.9 5 Crys»Colh 38 8.5 27.3 Colh»I 133 2.1 5.1 6 Crys»Colh 34 7.0 22.8 Colh»I 125 2.3 5.8 7 ga»Colh 34 » » Colh»I 141 3.0 7.2 8 ga»Colh 30 » » Colh»I 141 3.4 8.1 9 ga»Colh 28 » » Colh»I 141 3.2 7.9 10 ga»Colh 32 » » Colh»I 137 3.4 8.3 11 ga»Colh 37 » » Colh»I 133 3.9 9.5 14 Crys»Colh 41 87.6 278.9 Colh»I 122 4.5 11.4 a g stands for frozen mesophase or anisotropic glass.Thermal behaviour of the silver(I) tri—ate complexes The –rst series of tri—ate complexes, [AgSt(n-3,4) dis- 2][OTf], played a very rich mesomorphism (Table 3), in a way very similar to the corresponding dodecyl sulfate complexes,11a including both cubic and columnar mesophases (Fig. 4). The cubic phase was seen for almost all homologues (7OnO14), while the columnar phase appeared only for nP13. XRD studies con–rmed these observations ; the space group of the cubic phase was assigned as and the columnar phase has Ia3 6 d, a 2-D hexagonal lattice.The temperature range of the cubic phase was found to increase until n\12 (mainly on account of a rather pronounced reduction in melting point with increasing chain length) before it started to decrease, completely vanishing at the expense of the columnar phase at n\14. For the compounds showing only the cubic phase, the transition to the isotropic liquid was easily seen as the distorted, imprisoned air-bubbles relaxed to a spherical shape, as already seen in some lyotropic systems.27 The columnar phase Fig. 3 Phase diagram for the complexes [AgSt(n-3,4,5)2][DOS]. Table 3 Transition temperatures and thermal data for [AgSt(n-3,4)2]- [OTf] n Transition T /°C *H/kJ mol~1 *Sm/J mol~1 K~1 4 Crys»Crys@ 138 12.8 31.3 Crys@»I 188 37.6 81.7 6 Crys»Crys@ 144 16.2 38.8 Crys@»I 148 » » 7 Crys»Crys@ 132 18.7 46.2 Crys@»Cub 138 3.9 9.5 Cub»I 153 3.3 7.8 8 Crys»Crys@ 72 1.9 5.6 Crys@»CrysA 76 2.6 7.4 CrysA»Crys” 91 10.0 27.4 CrysA»Cub 125 4.3 10.7 Cub»I 154 3.1 7.2 10 Crys»Crys@ 63 10.7 31.8 Crys@»CrysA 88 55.2 152.6 CrysA»Cub 99 4.8 13.0 Cub»I 161 3.6 8.2 12 Crys»Cub 97 84.0 227.1 Cub»I 164 3.7 8.6 13 Crys»Cub 96 89.7 242.9 Cub»Colh 152 1.6 3.8 Colh»I 166 » » 14 Crys»Crys@ 84 1.5 4.2 Crys@»Cub 100 101.1 271.0 Cub»Colh 149 2.3 5.6 Colh»I 172 1.3 2.8 18 Crys»Crys@ 81 21.1 59.6 Crys@»Colh 101 125.6 335.6 Colh»I 151 1.3 3.0 also exists over a wide temperature range, as it does in the corresponding dodecyl sulfate series.In the second series of complexes, [AgSt(n-3,5)2][OTf] (Table 4), only the tetradecyloxy derivative was found to be mesomorphic, the optical texture suggesting a hexagonal columnar phase.The phase exists over a wide temperature range (*T \70 °C), with a melting point slightly above ambient (42 °C). On cooling, no crystallisation peak was seen by DSC, but it did appear in the following heat (cold crystallisation), followed by melting of the complex at 42 °C. In the next series, [AgSt(n-2,4) (Table 4), the two 2][OTf] complexes synthesised (n\7 and 14) exhibited a monotropic (metastable) nematic phase, in common with the analogous dodecyl sulfate salts, though the smectic C found in the latter Fig. 4 Phase diagram for the complexes [AgSt(n-3,4)2][OTf]. New J. Chem., 1999, 275»286 279Table 4 Transition temperatures and thermal data for [AgSt(n-3,5) and [AgSt(n-2,4) 2][OTf] 2][OTf] Complex n Transition T /°C *H/kJ mol~1 *Sm/J mol~1 K~1 [AgSt(n-3,5)2][OTf] 6 Crys»I 124 18.7 47.1 9 Crys»Crys@ 45 8.0 25.1 Crys@»I 81 6.5 18.3 12 Crys»Crys@ 35 8.2 26.4 Crys@»I 59 3.4 10.4 14 Crys»Colh 42 92.9 294.8 Colh»I 112 2.7 6.9 [AgSt(n-2,4)2][OTf] 7 Crys»Crys@ 92 41.2 112.8 Crys@»I 113 44.8 115.9 (I»N) (88) ([1.2) ([3.3) 14 Crys»Crys@ 50 10.8 33.4 Crys@»I 81 94.5 267.1 (I»N) (78) ([0.9) ([2.7) was absent here.The two compounds recrystallised a few degrees below the isotropic»nematic phase transition and showed this behaviour reproducibly in the following heat» cool cycles. Of the hexacatenar complexes, [AgSt(n-2,3,4) and 2][OTf] [AgSt(n-2,4,5) melted to the isotropic liquid at 100» 2][OTf] 120 °C for the octyloxy derivatives, and at 70»80 °C for the tetradecyloxy derivatives, with no metastable phase being detected on cooling from the isotropic liquid.However, the complexes [AgSt(n-3,4,5) (Table 5) 2][OTf] exhibited a rather complicated thermal behaviour, and optical microscopy suggested the following sequence (Fig. 5). The butyloxy derivative melted at high temperature to the isotropic liquid without showing any mesophase.The hexyloxy and heptyloxy derivatives displayed a cubic phase, over a narrow temperature range, the phase being assigned by the growth of black, distorted domains until the whole area became completely optically extinct. The phase this produced was also rather viscous. The transitions were con–rmed by DSC, showing a broad peak at the crystal-to-cubic phase transition, and another –rst order peak for the clearing point.The octyloxy homologue showed a rather unusual mesomorphism. Indeed, the optical observations, in addition to the DSC traces, suggested that the silver complex crystallised as two diÜerent, thermodynamically stable polymorphic crystals. Table 5 Transition temperatures and thermal data for [AgSt(n-3,4,5)2][OTf] T *H *Sm n Transition /°C /kJ mol~1 /J mol~1 K~1 4 Crys»I 205 » » 6 Crys»Cub 99 10.8 29.0 Cub»I 113 0.9 2.4 7 Crys»Cub 107 7.1 18.8 Cub»I 121 3.8 9.6 8a Crys»Cub]Crys@ 53 7.6 23.3 Cub]Crys@»Cub]Colh 65 3.1 9.1 Cub]Colh»Colh{]Colh 100 2.3 6.1 Colh{]Colh»I]Colh 108 2.5 6.5 I]Colh»I 127 0.6 1.5 9 Crys»Colh 82 2.6 7.4 Colh»I 162 1.5 3.4 10 Crys»Crys@ 40 6.3 20.1 Crys@»Colh 73 3.1 8.9 Colh»I 174 1.8 4.1 11 Crys»Colh 38 80.8 259.8 Colh»I 174 1.8 4.0 12 Crys»Colh 37 71.4 230.3 Colh»I 170 2.2 5.0 14 Crys»Crys@ 50 40.9 126.4 Crys@»Colh 62 82.6 246.4 Colh»I 165 2.4 5.5 a Crys and Crys@: two crystalline phases of diÜerent thermodynamic stability ; and two immiscible hexagonal columnar phases Colh Colh{ : with diÜerent degree of ordering.On heating, a –rst transition at 53 °C gave rise to a biphasic region, comprising a cubic phase and a crystalline phase, Crys@.The preparation was maintained at 60»62 °C for two hours to see whether this phenomenom was due to the slow kinetics of the crystal-to-cubic phase transition, thus, explaining the delay in the melt of the crystal phase, or whether Crys@ was another thermodynamically stable crystalline structure.No change was observed, suggesting that the initial bulk compound contained two diÜerent crystalline phases. On further increasing the temperature, the crystalline phase, Crys@, melted to give a columnar mesophase, at 65 °C, the phase being Colh , recognisable by its optical texture. In this temperature range, two mesophases were thus co-existing, a cubic and a columnar phase. Here again, the preparation was left for two hours at 95 °C, but no change was observed.At 100 °C, the cubic phase melted to a columnar phase, Surprisingly, the two Colh @ . columnar phases were not miscible as evidenced by large boundaries between the two textures, although they displayed identical defects. The mesophases did not mix when pressure was applied on the cover-slip, nor when the temperature was kept at 105 °C for two hours.Anyway, the non-miscibility of the two columnar mesophases does not necessarily contradict their possible identical nature. At 108 °C, the columnar phase melted to the isotropic liquid, and the other even- Colh Colh{ tually melted at 127 °C. On cooling, the same mesophases and biphasic regions were again observed, and this behaviour was reproducible on subsequent heat»cool cycles.The presence of a cubic phase in this series was rather unexpected, and there is Fig. 5 Phase diagram for the complexes [AgSt(n-3,4,5)2][OTf]. 280 New J. Chem., 1999, 275»286still no rational explanation so far for its occurence in the phase diagram. However, the most remarkable feature of the mesomorphism of this homologue is that the biphasic behaviour described was observed on cooling from the isotropic liquid and on subsequent reheating cycles.This means that the species giving rise to the two sets of mesomorphism are somehow diÜerent, and that this diÜerence is not ìannealedœ out in the isotropic phase. What we know is that the ì aspreparedœ and ì as-crystallised œ complex gives an elemental analysis consistent with the formulation [AgSt(8-3,4,5)2]- [OTf], and so the initial supposition is that the species can exist in more than one form, and these forms are not simply related as diÜerent crystal polymorphs.One possibility is that the complex could exist as a mixture of monomer and dimer. Recall that in the single crystal X-ray structure of [AgSt(1-4) the complex was found to exist as 2][C8H17OSO3], a dimer with two silver complex cations being held together by bridging sulfate groups from the octyl sulfate chains.28 It is conceivable that the tri—ate salts have a similar structure, and that a monomeric form can also exist. Thus, it is possible that the behaviour is explained by the presence of noninterconverting monomer and dimer.The higher homologues (nP9) behaved as typical hexacatenar mesogens do, in that they all showed an enantiotropic columnar phase, with large anisotropic domains, increasing from n\9 to 11, and then decreasing for nP12. Compounds with n\9 to 12 did not recrystallise on cooling, and did not show a melting behaviour on subsequent heating. Similarly to the previous compounds showing anisotropic glasses, it was not possible to measure the and the of the glass tran- *Cp Tg sition.Discussion Systems containing four —exible alkoxy chains Four alkoxy chains in the positions 3 and 4. The complexes having this substitution pattern behave typically as tetracatenar mesogens, showing cubic and columnar phases, although N and phases are not seen.12a,b Slight diÜerences SC in the mesomorphism of the tri—ate and dodecyl sulfate salts were observed and while both showed enantiotropic cubic and columnar mesophases with a wide temperature range, the tri- —ate salts had slightly higher transition temperatures.However, the most important contrast is that in the tri—ate salts, the cubic phase appeared to longer chain length (n\7 to 14) than in the related DOS salts (n\4 to 10).This is explained by recalling that in the DOS salts, the alkyl sulfate chain adds to the curvature at the paraffinic/aromatic interface whereas in the tri—ate systems, this is not the case. Thus, for a given chain length on the stilbazole, the interfacial curvature will always be greater in the DOS salts, resulting in the onset of columar mesophases at shorter chain lengths.A related factor is likely to be due to the fact that mesophase formation is goverened by the interfacial curvature which, in turn, depends on the relative volumes of the core and the chains. In the tri—ate salts, not only does the anion not contribute to the aliphatic chain density, but in fact it can act to the opposite eÜect, increasing the volume of the core relative to the chains.Thus, phases with more strongly curved interfaces will require longer chain lengths. That the tri—ate complexes are likely to be more ionic is supported by their higher melting points, but the fact that the clearing points of the two salts series are essentially the same points to the fact that the global breakdown in ionic interactions is not responsible for clearing, not inconsistent with a lattice melting model.28 By XRD and dilatometry,11a a linear variation of the lattice parameters and of the speci–c volume of both phases of the DOS series as a function of n was observed, even during the cross-over from one phase to the other.After deduction of the epitaxial relationships between the two mesophases, it was concluded that the transition Col-to-Cub occured with rather small structural changes, and a model based on undulated columns was proposed.In the case of the tri—ate compounds, preliminary X-ray results suggest a completely diÜerent mechanism, since an important jump in the variation of the lattice parameters of the two phases was recorded. The results of this study will be reported in due course.29 Interestingly, the cubic phase-to-isotropic liquid transition shows by DSC a fairly broad diÜuse endotherm which may suggest a clearing process involving several steps. This peculiar phenomenom has been attributed by others to the breakdown of the molecular order of the cubic lattice.30 This may further explain the supercooling observed on cooling of the isotropic liquid.Four alkoxy chains in the positions 3 and 5. The diÜerence in thermal behaviour as a function of the counter anion is more evident here, and clearly shows the crucial role of the dodecyl sulfate chain in stabilising mesophases. While only the tetradecyloxy homologue of the OTf series is mesomorphic showing a columnar phase, all members of the DOS series are mesomorphic for n[4.As mentioned above, these complexes are virtually unique examples of mesomorphic 3,5-disubstituted tetracatenar materials with only four rings in the structure. 12a,b For the DOS salts, the minimum requirement for mesomorphism (six methylene groups per chain) is probably linked to the absence of the alkoxy chains in the 4-position, that is a minimum volume of aliphatic chains is required to be able to generate an aliphatic matrix of sufficient density and in an appropriate position to stabilise the columns and to compensate the void created by the absence of chains in the 4- position. This reinforces the important role of the DOS chain in the promotion of mesomorphism in these systems.Thus, on crossing the aromatic/paraffinic boundary, the DOS chain interacts sterically with the meta chains causing them to adopt a spatial disposition diÜerent to that in the tri—ate salts where no such steric interactions take place, as well as contributing positively to the interfacial curvature.Thus, we propose that in addition to enhancing the alkyl chain density, the DOS chain modi–es the spatial distribution and allows the void of the 4-position chain to be more readily addressed.A test of this idea will be available when the X-ray data are considered as there should be no appreciable interdigitation of columns if this model is correct. Four alkoxy chains in the positions 2 and 4. The complexes having this substitution pattern showed only monotropic behaviour, consisting of nematic and, in one case only, smectic C phases.It is likely that the destabilisation of the liquidcrystalline properties results from the alkoxy chain in the 2- position which, it is assumed, acts as an additional lateral group. This behaviour is consistent with that shown by some organic liquid crystals having one or two lateral chain(s), although in the latter, the nematic phases showed a higher thermodynamic stability than in the present situation.Systems containing six —exible alkoxy chains Six alkoxy chains in the positions 3, 4 and 5. Almost all the complexes having this substitution pattern showed a columnar phase typical for such systems, although it appeared at smaller values of n in the DOS salts (n\4) than in the OTf salts (n\8), indicating once more the positive eÜect of the dodecyl sulfate chains on the chain density and interfacial curvature.The melting points are also much lower in the former (28»53 °C) than in the latter (40»60 °C), although the phase exists over a much wider temperature in the tri—ate series because of higher clearing temperatures. One particularly striking feature is the rather unsual thermal behaviour of the short homologues of the tri—ate salts New J.Chem., 1999, 275»286 281(n\6, 7, 8). Indeed, the cubic phase displayed by these homologues was not at all expected primarily owing the strong curvature at the interface generated by the three periphal alkoxy chains. However, recall the argument above which related the enhanced volume of the core of these complexes to the presence of the tri—ate anion.This argument is also consistent with the observation of cubic phases in these hexacatenar materials as the increased core volume will reduce the interfacial curvature increasing the likelihood of a cubic phase being seen. Six alkoxy chains in the positions 2, 3, 4 and 2, 4, 5. In these systems, liquid-crystal properties were not expected because of the location of two chains in the positions 2 and 3/5.Lateral interactions were not favoured, as evidenced by the low transition temperatures, and the non-efficient packing of the chains, both reasons excluding mesomorphism. Conclusions New low-melting metallomesogens have been obtained by simple reaction of a silver fragment with various polysubstituted stilbazole ligands. Furthermore, the wide collection of substances herein prepared allowed us to be able to predict the liquid-crystalline properties, and thus to propose some structure»property relationships.Interestingly, the complexation of the non-mesomorphic ligands to some silver salts induced mesomorphism in many ligand/metal combinations, proving in this case, that the mesomorphism is not necessarily the result of molecular anisotropy itself, but depends on the molecular aggregation and on the shape of these aggregates.This highlights a very interesting analogy to lyotropic liquid crystals, and can allow speculation about a general description of liquid crystalline mesophases. It seems indeed clearer that similar principles of phase descriptions used in lyotropic systems can be used here. In order to probe this analogy to lyotropic systems, we will study the eÜects of aliphatic apolar solvents on the mesomorphism of these silver(I) complexes, as several other groups have already done.31 Preliminary results from simple contact preparations between some silver(I) compounds and dodecane look very promising, con–rming the amphiphilic character of these complexes. In this case, they behave typically as the socalled amphotropic systems.6a,19,32 Experimental procedures All the solvents were distilled prior to use according to standard procedures:33 was distilled over calcium CH2Cl2 hydride, methanol from magnesium and iodine, tetrahydrofuran (THF) from sodium and benzophenone, toluene from sodium, acetone from potassium permanganate.Diethyl ether was dried and stored over sodium wire. 4- Methylpyridine was distilled from sodium hydroxide and stored over sodium hydroxide pellets. Potassium carbonate was kept in the oven. All other chemicals were used as supplied. 1H, 13C and 19F NMR spectra were recorded on a Bruker ACL250 or AM400 spectrometer and referenced to external tetramethylsilane. The coupling constants were measured by the use of a program Window NMR (1D-Win NMR, Bruker, MS Window).One-bond and multiple-bond (1JCH) (2JCH , and C»H correlation spectra were recorded to 3JCH , 4JCH) allow a correct and full assignment of the proton and carbon peaks of all the substances. Microanalysis was performed by the University of Sheffield micro-analytical service. Infrared spectra were recorded on a Perkin»Elmer 684 spectrophotometer.The study of the thermal behaviour was achieved by DSC analysis, carried out using a Perkin»Elmer DSC7 instrument using various heating rates (2, 5 and 10 K min~1) and the mesomorphism was studied by hot-stage, polarising microscopy using a Zeiss Labpol microscope equipped with a Linkam TH600 hot-stage and PR600 temperature controller. All of the mesophases were characterised by their optical textures, and then con–rmed by XRD.The spectroscopic and analytical data of all the intermediate substances (aldehydes, esters and alcohols) were good and in agreement with the proposed structures. Preparation of the poly(alkoxy)benzaldehydes 2,4-Di-, 2,3,4-tri- and 2,4,5-tri-alkoxybenzaldehydes were synthesised as described previously, although for the trisubstituted benzaldehydes, pentan-2-one was prefered due to its higher boiling point, and the reaction was carried out under an inert atmosphere Two equivalents of were (N2).K2CO3 systematically used per hydroxy group. They were obtained in good yields, ranging from 66 to 89%, and of analytical purity after puri–cation by —ash column chromatography (silica gel, IR(CsI/Nujol, cm~1) : 2,4-didecyloxybenzaldehyde: CH2Cl2). 1680s, 2715w; 2,3,4-trioctyloxybenzaldehyde : l(C/O) l(HhC/0) 1688s, 2730w; 2,4,5-trioctyloxybenzaldehyde : l(C/O) l(HhC/0) 1660s, 2730w. l(C/O) l(HhC/0) 3,5-Di and 3,4,5-tri-alkoxybenzaldehydes were obtained differently, and the preparation of one example is given. Methyl 3,5-dihydroxybenzoate (3.5 g, 20.2 mmol), (11.2 g, 81.0 K2CO3 mmol) and 1-bromooctane (8.7 g, 44.4 mmol) were placed in a —ask –lled with pentan-2-one (100 cm3), and the reaction mixture was heated at re—ux for 24 h.When cooled to room temperature, the mixture was –ltered through a pad of Celite and the solvent evaporated under reduced pressure. The product was puri–ed by column chromatography (SiO2 , 4 : 1). The yield was 84% (6.7 g). All of the CH2Cl2»hexane esters were puri–ed similarly, and obtained in good yields ranging from 75 to 94%. IR(CsI/Nujol, cm~1) : 1730s, l(C/O) 1300m.l(ChOMe) The reduction was carried out under an atmosphere of nitrogen. Methyl 3,5-dioctyloxybenzoate (6 g, 15.3 mmol) was dissolved in THF (50 cm3), and (0.83 g, 38.2 mmol) was LiBH4 added by portions to the solution ; the mixture was heated at re—ux for 1 h.Then, dry methanol (10 cm3) was added dropwise to the re—uxing mixture over a period of 1 h. The mixture was left 6 h at re—ux and under vigorous stirring, and was then allowed to cool to room temperature. A solution of dilute HCl (20 cm3, 3% v/v) was added until the pH was neutral. The alcohol was extracted in diethyl ether (2]250 cm3), washed with water (2]200 cm3), rinsed with brine (100 cm3), and the organic layer was dried over The sol- MgSO4 .vents were removed under reduced pressure and the alcohol puri–ed by —ash-column chromatography (SiO2 , 99:1 to 90:10). The yield was eÜectively CH2Cl2»MeOH quantitative (97%, 5.4 g). All of the benzyl alcohols were puri- –ed similarly, and obtained in good yields ranging from 85 to 97%. IR(CsI/Nujol, cm~1) 3340br, s.l(OhH) The oxidation of the alcohols was carried out in the air and at room temperature. 3,5-Dioctyloxybenzyl alcohol (4 g, 11.0 mmol) was dissolved in (10 cm3), and added to a sus- CH2Cl2 pension of pyridinium chlorochromate (3.6 g, 16.5 mmol) in (40 cm3). The mixture immediately turned dark as the CH2Cl2 solution of alcohol was added, and was left stirred for 4 h.The product was directly recovered following –ltration over a pad of silica gel and after evaporation of the solvent to give the aldehyde in good yield (85%, 3.4 g). When necessary, the aldehydes were once more puri–ed by —ash-column chromatography All of the benzaldehydes were puri–ed (SiO2, CH2Cl2). similarly, and obtained in good yields ranging from 70 to 95%. IR(CsI/Nujol, cm~1) : 1705s, 2710w.l(C/O) l(HhC/0) Preparation of the poly(alkoxy)stilbazoles The –ve new series of ligands were prepared and puri–ed in the identical manner as the already given detailed procedure 282 New J. Chem., 1999, 275»286for the preparation of 3,4-dialkoxy-4-stilbazole.11a All the ligands were found to be analytically pure (Table 6), except for some homologues of the 3,5-disubstituted series which were found difficult to purify, and were used as such.Nevertheless, the structure of all the ligands were con–rmed by NMR spectroscopy, for which no extra signals or impurities were observed. The NMR data of one derivative of each series are given below, and the C»H correlation spectra were made available to the referees. The lettering of the hydrogen and carbon atoms follows the same system as already used.12c The protons of the double bond (k, l) and those of the pyridine ring (n, o) were studied as a AB system and as a AA@XX@ system respectively.Spectroscopic and analytical data for the ligands [ St(n-x,y,z) ] St(6-3,5). MHz, 0.90 (a@, t, 6.6 Hz, 6 dH(250.13 CDCl3) : 3JHH H), 1.41 (b@, m, 12 H), 1.78 (c@, m, 4 H), 3.96 (d@, t, 6.6 Hz, 3JHH 4 H), 6.43 (e, t, 2.3 Hz, 1 H), 6.67 (g, d, 2.3 Hz, 2 H), 4Jeg 4Jge 6.97 (l, AB, 16.2 Hz, 1 H), 7.27 (k, AB, 16.2 Hz, 1 transJlk transJkl H), 7.33 (n, AA@XX@, 6.4 Hz, 2 H), 8.56 (o, AA@XX@, o Jno]Jno{ o 6.7 Hz, 2 H).MHz, 14.0 (a@), 22.6, o Jon]Jon{ o dC(62.9 CDCl3) : 25.7, 29.2, 31.6 (b@, c@), 67.8 (d@), 101.9 (e), 105.6 (g), 120.9 (n), 126.3 (l), 133.3 (k), 137.9 (h), 144.5 (m), 150.1 (o), 160.6 (f ).St(7-2,4). MHz, 0.89, 0.90 (a, aA, 2 t, dH(250.13 CDCl3) : 3JHH 6.4 Hz, 6 H), 1.40 (b, bA, m, 16 H), 1.82 (c, cA, m, 4 H), 3.96, 3.99 (d, dA, 2 t, 6.4 Hz, 4 H), 6.45 (f, d, 2.0 Hz, 1 H), 3JHH 4Jfj 6.49 (j, dd, 8.5 Hz, 2.0 Hz, 1 H), 6.95 (l, AB, 16.5 3Jji 4Jjg transJlk Hz, 1 H), 7.31 (n, AA@XX@, 6.1 Hz, 2 H), 7.47 (i, d, o Jno]Jno{ o 8.5 Hz, 1 H), 7.58 (k, AB, 16.5 Hz, 1 H), 8.51 (o, 3Jij transJkl AA@XX@, 6.1 Hz, 2 H).MHz, 14.1 o Jon]Jon{ o dC(62.9 CDCl3) : (a, aA), 22.6, 26.0, 26.2, 29.1, 29.2, 29.3, 31.8 (b, bA, c, cA), 68.2, 68.5 (d, dA), 99.8 (f), 105.8 (j), 118.1 (h), 120.6 (n), 123.7 (l), 128.2 (i), 128.3 (k), 145.9 (m), 150.0 (o), 158.2 (g), 160.9 (e). St(8-3,4,5). MHz, 0.96 (a, t, 6.5 Hz, dH(250.13 CDCl3) : 3JHH 3 H), 0.98 (a@, t, 6.5 Hz, 6 H), 1.37 (b, b@, m, 30 H), 1.74 (c, 3JHH c@, m, 6 H), 3.98 (d, t, 6.5 Hz, 2 H), 4.02 (d@, t, 6.5 Hz, 3JHH 3JHH 4 H), 6.72 (g, s, 2 H), 6.86 (l, AB, 16.2 Hz, 1 H), 7.19 (k, transJlk AB, 16.2 Hz, 1 H), 7.32 (n, AA@XX@, 6.4 Hz, transJkl o Jno]Jno{ o 2 H), 8.54 (o, AA@XX@, 6.4 Hz, 2 H).MHz, o Jon]Jon{ o dC(62.9 14.0 (a@), 14.1 (a), 22.6, 22.7, 25.8, 29.4, 29.7, 30.3, 31.6, CDCl3) : 31.8 (b, b@, c, c@), 68.9 (d@), 73.5 (d), 105.7 (g), 120.8 (n), 124.8 (l), 131.3 (e), 133.4 (k), 139.2 (h), 144.8 (m), 150.1 (o), 153.4 (f ).St(8-2,3,4). MHz, 0.88, 0.90 (a, a@, aA, 2 t, dH(250.13 CDCl3) : 6.7 Hz, 9 H), 1.39 (b, b@, bA, m, 30 H), 1.80 (c, c@, cA, m, 6 3JHH H), 3.97, 3.98, 4.04 (d, d@, dA, 3 t, 6.7 Hz, 6 H), 6.67 (j, d, 3JHH 9.0 Hz, 1 H), 6.90 (l, AB, 16.6 Hz, 1 H), 7.28 (i, d, 3Jji transJlk 9.0 Hz, 1 H), 7.32 (n, AA@XX@, 6.4 Hz, 2 H), 3Jij o Jno]Jno{ o 7.58 (k, AB, 16.6 Hz, 1 H), 8.52 (o, AA@XX@, transJkl o Jon]Jon{ o 6.4 Hz, 2 H).MHz, 14.0, 14.1 (a, a@, aA), 22.6, dC(62.9 CDCl3) : 22.7, 25.8, 26.0, 29.3, 30.3, 30.4, 31.6, 31.7 (b, b@, bA, c, c@, cA), 68.8, 73.7, 74.3 (d, d@, dA), 108.6 (j), 120.6 (n), 120.9 (i), 123.2 (h), 124.4 (l), 128.1 (k), 141.9 (f), 145.5 (m), 150.0 (o), 151.8 (g), 154.1 (e).St(8-2,4,5). MHz, 0.86, 0.88 (a, a@, aA, 2 t, dH(250.13 CDCl3) : 6.7 Hz, 9 H), 1.38 (b, b@, bA, m, 30 H), 1.80 (c, c@, cA, m, 6 3JHH H), 3.97, 3.98, 4.00 (d, d@, dA, 3 t, 6.7 Hz, 6 H), 6.49 (f, s, 1 3JHH H), 6.87 (l, AB, 16.5 Hz, 1 H), 7.12 (i, s, 1 H), 7.32 (n, transJlk AA@XX@, 6.1 Hz, 2 H), 7.61 (k, AB, 16.5 Hz, o Jno]Jno{ o transJkl 1 H), 8.52 (o, AA@XX@, 6.1 Hz, 2 H).MHz, o Jon]Jon{ o dC(62.9 14.1 (a, a@, aA), 22.7, 26.1, 26.3, 29.3, 29.4, 29.6, 31.8 (b, CDCl3) : b@, bA, c, c@, cA), 69.3, 69.7, 70.7 (d, d@, dA), 100.4 (f), 113.5 (i), 117.5 (h), 120.6 (n), 123.6 (l), 127.9 (k), 143.3 (j), 145.7 (m), 150.0 (o), 151.3 (e), 152.4 (g). Preparation of the silver complexes The preparation of the silver(I) dodecyl sulfate was achieved as previously described.11a In the case of the silver(I) tri—ate com- Table 6 Microanalytical data for the stilbazolesa St(n-x,y,z) Yield (%) MP/°C C H N St(13-3,4) 89 89 80.7 (81.0) 11.1 (11.0) 2.3 (2.4) St(14-3,4) 70 91 80.5 (81.3) 11.3 (11.1) 2.7 (2.3) St(18-3,4) 78 94 81.2 (81.9) 12.0 (11.6) 1.8 (1.9) St(1-3,5) 58 92 74.5 (74.7) 6.4 (6.3) 5.8 (5.8) St(4-3,5)b 89 » 75.4 (77.5) 8.1 (8.4) 3.9 (4.3) St(6-3,5) 80 36 78.1 (78.7) 9.2 (9.2) 4.0 (3.7) St(7-3,5) 75 49 78.5 (79.2) 9.8 (9.6) 3.7 (3.4) St(9-3,5) 62 55 79.5 (80.0) 10.3 (10.2) 2.8 (3.0) St(11-3,5) 53 51 79.6 (80.6) 11.0 (10.6) 2.4 (2.7) St(12-3,5) 86 55 79.8 (80.8) 10.9 (10.8) 2.7 (2.5) St(14-3,5) 85 64 80.6 (81.3) 11.2 (11.1) 2.5 (2.3) St(7-2,4) 40 64 78.9 (79.2) 9.8 (9.6) 3.3 (3.4) St(10-2,4) 77 44 80.0 (80.3) 10.4 (10.4) 3.0 (2.8) St(14-2,4) 82 55 81.3 (81.3) 11.4 (11.1) 2.3 (2.3) St(1-3,4,5) 58 97 70.4 (70.8) 6.2 (6.3) 5.1 (5.2) St(4-3,4,5) 67 64 75.7 (75.5) 9.0 (8.9) 3.5 (3.5) St(5-3,4,5)b 85 » 76.5 (76.5) 9.5 (9.4) 3.2 (3.2) St(6-3,4,5)b 90 » 76.6 (77.3) 9.8 (9.8) 2.7 (2.9) St(7-3,4,5) 85 49 77.9 (78.0) 10.1 (10.2) 2.6 (2.7) St(8-3,4,5) 86 48 78.6 (78.5) 10.7 (10.5) 2.5 (2.5) St(9-3,4,5) 78 41 78.9 (79.0) 11.1 (10.8) 2.0 (2.3) St(10-3,4,5) 82 44 79.6 (79.4) 11.1 (11.0) 2.1 (2.1) St(11-3,4,5) 76 48 79.8 (79.8) 11.3 (11.2) 2.1 (2.0) St(12-3,4,5) 66 51 79.9 (80.2) 11.3 (11.4) 1.7 (1.9) St(14-3,4,5) 90 57 80.5 (80.7) 11.8 (11.7) 1.6 (1.7) St(6-2,3,4)b 74 » 76.8 (77.3) 9.4 (9.8) 2.9 (2.9) St(8-2,3,4)b 90 » 78.3 (78.5) 10.8 (10.5) 2.4 (2.5) St(10-2,3,4) 76 41 79.6 (79.4) 11.0 (11.0) 2.4 (2.1) St(12-2,3,4) 61 51 79.9 (80.2) 11.4 (11.4) 1.8 (1.9) St(14-2,3,4) 89 63 80.8 (80.7) 11.8 (11.7) 1.4 (1.7) St(8-2,4,5) 84 66 78.5 (78.5) 10.6 (10.5) 2.6 (2.5) St(14-2,4,5) 86 65 80.3 (80.7) 12.0 (11.7) 1.4 (1.7) a Calculated values in parentheses (%).b Obtained as oils.New J. Chem., 1999, 275»286 283plexes, the reactions were carried out in dry acetone. When cooled, the resulting yellow precipitate was –ltered, crystallised from hot acetone (once or twice), recovered by –ltration and dried under high vacuum. The complexes were obtained in satisfying yields, with good analytical purity (Tables 7 and 8). The NMR data (1H, 13C and 19F) were in agreement with the expected structures, and are given below for one homolog of each series.Spectroscopic and analytical data for the complexes [AgSt(nx, y,z)2 ] [DOS] MHz, 0.84 (s, t, [AgSt(6-3,5)2 ] [DOS] . dH(250.13 CDCl3) : 6.7 Hz, 3 H), 0.91 (a@, t, 6.7 Hz, 12 H), 1.34 (b@, r, m, 3JHH 3JHH 42 H), 1.67 (q, m, 2 H), 1.76 (c@, m, 8 H), 3.90 (d@, t, 6.7 Hz, 3JHH 8 H), 4.13 (p, t, 6.7 Hz, 2 H), 6.41 (e, t, 2.1 Hz, 2 H), 3JHH 4Jeg 6.57 (g, d, 2.1 Hz, 4 H), 6.87 (l, AB, 16.2 Hz, 2 H), 4Jge transJlk 7.17 (k, AB, 16.2 Hz, 2 H), 7.39 (n, AA@XX@, transJkl o Jno]Jno{ o 6.4 Hz, 4 H), 8.69 (o, AA@XX@, 6.4 Hz, 4 H).o Jon]Jon{ o dC(62.9 MHz, 14.1 (a@, s), 22.7, 25.8, 26.1, 29.3, 29.5, 29.6, 29.7, CDCl3) : 31.9 (b@, c@, r, q), 68.1 (p), 69.2 (d@), 102.6 (e), 105.7 (g), 121.7 (n), 124.6 (l), 135.8 (k), 137.1 (h), 146.9 (m), 151.8 (o), 160.5 (f).MHz, 0.84 (s, t, [AgSt(7-2,4)2 ] [DOS] . dH(250.13 CDCl3) : 6.7 Hz, 3 H), 0.89 (a, aA, t, 6.7 Hz, 12 H), 1.35 (b, bA, 3JHH 3JHH r, m, 50 H), 1.67 (q, m, 2 H), 1.81 (c, cA, m, 8 H), 3.94 (d, dA, t, 6.7 Hz, 8 H), 4.11 (p, t, 6.7 Hz, 2 H), 6.39 (f, d, 3JHH 3JHH 4Jfj Table 7 Microanalytical data for the AgDOS complexesa Complex n Yield (%) C H N S [AgSt(n-3,5)2][DOS] 1 74 58.5 (58.9) 6.5 (6.5) 3.2 (3.3) 2.7 (3.7) 4 28 63.1 (63.3) 7.8 (7.8) 2.8 (2.7) 3.2 (3.1) 6 58 66.1 (65.5) 8.6 (8.4) 2.4 (2.5) 2.8 (2.8) 7 38 66.7 (66.5) 8.5 (8.7) 2.4 (2.3) 2.8 (2.7) 9 69 68.1 (68.1) 9.4 (9.2) 2.2 (2.1) 2.6 (2.5) 12 51 70.2 (70.1) 9.8 (9.8) 1.9 (1.9) 2.4 (2.2) 14 67 71.5 (71.2) 10.2 (10.1) 1.6 (1.8) 2.2 (2.0) [AgSt(n-2,4)2][DOS] 7 67 65.8 (66.5) 8.8 (8.7) 2.2 (2.3) 2.8 (2.7) 14 74 71.1 (71.2) 10.2 (10.1) 1.7 (1.8) 2.0 (2.0) [AgSt(n-3,4,5)2][DOS] 1 76 57.6 (57.7) 6.5 (6.5) 3.0 (3.1) 3.5 (3.5) 4 56 63.5 (63.7) 8.2 (8.2) 2.3 (2.4) 2.6 (2.7) 5 15 65.6 (65.2) 8.9 (8.6) 2.4 (2.2) 2.6 (2.5) 6 32 66.6 (66.5) 9.0 (9.0) 2.1 (2.1) 2.7 (2.4) 7 68 67.4 (67.6) 9.3 (9.3) 2.2 (2.0) 2.4 (2.3) 8 64 68.8 (68.6) 9.7 (9.6) 1.8 (1.9) 2.1 (2.1) 9 51 69.4 (69.5) 10.0 (9.8) 1.9 (1.8) 2.1 (2.0) 10 58 70.5 (70.3) 10.2 (10.1) 1.6 (1.7) 2.0 (1.9) 11 55 71.3 (71.1) 9.5 (10.3) 2.0 (1.6) 1.8 (1.8) 14 64 72.7 (72.9) 10.8 (10.8) 1.7 (1.4) 1.5 (1.6) [AgSt(n-2,3,4)2][DOS] 8 53 68.6 (68.6) 9.5 (9.6) 1.8 (1.9) 2.1 (2.1) 14 79 73.4 (72.9) 10.9 (10.8) 1.4 (1.4) 1.4 (1.6) [AgSt(n-2,4,5)2][DOS] 8 52 68.6 (68.6) 9.7 (9.6) 1.8 (1.9) 2.2 (2.1) 14 45 72.8 (72.9) 10.9 (10.8) 1.6 (1.4) 1.6 (1.6) a Calculated values in parentheses (%).Table 8 Microanalytical data for the AgOTf complexesa Complex n Yield (%) C H N S [AgSt(n-3,4)2][OTf] 4 82 56.5 (59.6) 5.8 (6.0) 3.2 (3.1) 4.0 (3.5) 6 88 59.6 (60.0) 6.9 (6.9) 2.8 (2.7) 3.7 (3.1) 7 75 60.9 (61.4) 7.2 (7.3) 2.6 (2.6) 3.7 (3.0) 8 80 61.8 (62.6) 7.7 (7.7) 2.8 (2.5) 2.7 (2.8) 10 64 64.9 (64.7) 8.4 (8.3) 2.1 (2.2) 3.0 (2.6) 12 94 66.4 (66.4) 8.6 (8.8) 1.9 (2.1) 2.7 (2.4) 13 80 67.0 (67.2) 8.9 (9.0) 2.0 (2.0) 2.2 (2.3) 14 89 67.4 (67.9) 9.5 (9.2) 1.9 (1.9) 2.4 (2.2) 18 88 71.1 (70.2) 10.2 (9.9) 1.9 (1.6) (1.9) [AgSt(n-3,5)2][OTf] 6 56 59.1 (60.0) 6.9 (6.9) 2.6 (2.7) 3.7 (3.1) 9 66 63.4 (63.7) 8.1 (8.0) 2.6 (2.3) 2.9 (2.7) 12 40 66.4 (66.4) 8.7 (8.8) 2.0 (2.1) 2.4 (2.4) 14 79 67.6 (67.9) 9.3 (9.2) 1.7 (1.9) 2.8 (2.2) [AgSt(n-2,4)2][OTf] 7 61 61.3 (61.4) 7.4 (7.3) 2.8 (2.6) 3.3 (3.0) 14 87 67.3 (67.9) 9.2 (9.2) 1.9 (1.9) 1.8 (2.2) [AgSt(n-3,4,5)2][OTf] 4 54 57.8 (58.2) 6.7 (6.7) 2.7 (2.7) 3.1 (3.0) 6 50 61.7 (62.0) 7.6 (7.8) 2.3 (2.3) 2.9 (2.6) 7 64 63.3 (63.5) 8.2 (8.2) 2.2 (2.1) 2.6 (2.5) 8 80 64.7 (64.9) 8.6 (8.6) 2.0 (2.0) 2.7 (2.3) 9 53 66.1 (66.0) 8.9 (8.9) 1.9 (1.9) 2.5 (2.2) 10 51 66.8 (67.1) 9.1 (9.2) 1.7 (1.8) 2.3 (2.1) 11 60 67.7 (68.1) 9.6 (9.5) 1.4 (1.7) 1.8 (1.9) 12 61 68.3 (68.9) 9.5 (9.7) 1.6 (1.6) 1.9 (1.9) 14 78 70.3 (70.4) 10.3 (10.1) 1.3 (1.5) 1.8 (1.7) [AgSt(n-2,3,4)2][OTf] 8 85 65.3 (64.9) 8.6 (8.6) 2.0 (2.0) 2.4 (2.3) 14 92 69.6 (70.4) 10.4 (10.1) 1.6 (1.5) 1.8 (1.7) [AgSt(n-2,4,5)2][OTf] 8 75 64.4 (64.9) 8.7 (8.6) 2.0 (2.0) 2.5 (2.3) 14 76 70.6 (70.4) 10.6 (10.1) 1.4 (1.5) 2.0 (1.7) a Calculated values in parentheses (%). 284 New J. Chem., 1999, 275»2862.1 Hz, 2 H), 6.41 (j, dd, 8.3 Hz, 2.1 Hz, 2 H), 6.87 (l, 3Jji 4Jjg AB, 16.5 Hz, 2 H), 7.33 (n, AA@XX@, 6.4 Hz, transJlk o Jno]Jno{ o 4 H), 7.38 (i, d, 8.3 Hz, 2 H), 7.58 (k, AB, 16.5 Hz, 2 3Jij transJkl H), 8.64 (o, AA@XX@, 6.4 Hz, 4 H).MHz, o Jon]Jon{ o dC(62.9 14.1 (a, aA, s), 22.6, 26.0, 26.1, 29.0, 29.1, 29.3, 29.5, CDCl3) : 29.7, 31.8, 31.9 (b, bA, c, cA, r, q), 68.0 (p) 68.1, 68.4 (d, dA), 99.5 (f), 105.7 (j), 117.5 (h), 121.3 (n), 122.3 (l), 128.6 (i), 130.6 (k), 148.1 (m), 152.1 (o), 158.5 (g), 161.4 (e). MHz, 0.84 (s, t, [AgSt(8-3,4,5)2 ] [DOS] .dH(250.13 CDCl3) : 6.7 Hz, 3 H), 0.85 (a, t, 6.4 Hz, 6 H), 0.88 (a@, t, 3JHH 3JHH 3JHH 6.4 Hz, 12 H), 1.35 (b, b@, r, m, 78 H), 1.67 (q, m, 2 H), 1.78 (c, c@, m, 12 H), 3.97 (d, t, 6.4 Hz, 4 H), 3.98 (d@, t, 6.4 3JHH 3JHH Hz, 8 H), 4.09 (p, t, 6.7 Hz, 2 H), 6.70 (g, s, 4 H), 6.81 (l, 3JHH AB, 16.2 Hz, 2 H), 7.22 (k, AB, 16.2 Hz, 2 H), transJlk transJkl 7.42 (n, AA@XX@, 6.4 Hz, 4 H), 8.67 (o, AA@XX@, o Jno]Jno{ o 6.4 Hz, 4 H).MHz, 14.1 (a, a@, s), o Jon]Jon{ o dC(62.9 CDCl3) : 22.6, 22.7, 25.8, 26.1, 29.1, 29.3, 29.4, 30.4, 31.8, 31.9 (b, b@, c, c@, r, q), 68.1 (p), 69.2 (d@), 73.6 (d), 106.0 (g), 121.7 (n), 123.4 (l), 130.6 (e), 135.9 (k), 139.8 (h), 147.2 (m), 151.8 (o), 153.4 (f ). MHz, 0.85 (s, t, [AgSt(8-2,3,4)2 ] [DOS] . dH(250.13 CDCl3) : 6.7 Hz, 3 H), 0.86, 0.87 (a, a@, aA, 2 t, 6.7 Hz, 18 H), 3JHH 3JHH 1.35 (b, b@, bA, r, m, 78 H), 1.66 (q, m, 2 H), 1.80 (c, c@, cA, m, 12 H), 3.96, 3.97, 4.03 (d, d@, dA, 3 t, 6.7 Hz, 12 H), 4.10 (p, t, 3JHH 6.7 Hz, 2 H), 6.62 (j, d, 8.7 Hz, 2 H), 6.86 (l, AB, 3JHH 3Jji 16.5 Hz, 2 H), 7.22 (i, d, 8.7 Hz, 2 H), 7.39 (n, transJlk 3Jij AA@XX@, 6.4 Hz, 4 H), 7.61 (k, AB, 16.5 Hz, o Jno]Jno{ o transJkl 2 H), 8.67 (o, AA@XX@, 6.4 Hz, 4 H).MHz, o Jon]Jon{ o dC(62.9 14.1 (a, a@, aA, s), 22.6, 26.0, 26.1, 26.2, 29.3, 29.4, 29.6, CDCl3) : 29.7, 30.3, 31.8, 31.9 (b, b@, bA, c, c@, cA, r, q), 68.0 (p), 68.7, 73.7, 74.4 (d, d@, dA), 108.4 (j), 121.3 (i), 121.5 (n), 122.6 (h), 123.0 (l), 130.5 (k), 141.8 (f), 147.8 (m), 152.0 (o), 152.1 (g), 154.6 (e). Thermal data [T /°C (*H/kJ mol~1)] : n\8, Crys Æ 98 (23) Æ I ; n\14, Crys Æ 41 (77) Æ Crys@ Æ52 (37) Æ I. MHz, 0.84 (s, t, [AgSt(8-2,4,5)2 ] [DOS] .dH(250.13 CDCl3) : 6.7 Hz, 3 H), 0.87, 0.88 (a, a@, aA, 2 t, 6.7 Hz, 18 H), 3JHH 3JHH 1.35 (b, b@, bA, r, m, 78 H), 1.66 (q, m, 2 H), 1.82 (c, c@, cA, m, 12 H), 3.93, 3.94, 3.99 (d, d@, dA, 3 t, 6.7 Hz, 12 H), 4.10 (p, t, 3JHH 6.7 Hz, 2 H), 6.43 (f, s, 2 H), 6.83 (l, AB, 16.5 Hz, 2 3JHH transJlk H), 7.06 (i, s, 2 H), 7.39 (n, AA@XX@, 6.6 Hz, 4 H), o Jno]Jno{ o 7.64 (k, AB, 16.5 Hz, 2 H), 8.64 (o, AA@XX@, transJkl o Jon]Jon{ o 6.6 Hz, 4 H).MHz, 14.1 (a, a@, aA, s), 22.7, 26.0, dC(62.9 CDCl3) : 26.1, 26.2, 29.3, 29.4, 29.6, 29.7, 31.8 (b, b@, bA, c, c@, cA, r, q), 68.0 (p), 69.2, 69.5, 70.6 (d, d@, dA), 99.8 (f), 113.5 (i), 116.6 (h), 121.5 (n), 122.0 (l), 130.5 (k), 143.2 (j), 148.2 (m), 151.8 (o), 151.9 (e), 152.9 (g).Thermal data [T /°C (*H/kJ mol~1)] : n\8, Crys Æ 88 (78) Æ I ; n\14, Crys Æ 64 (55) Æ Crys@ Æ78 (89) Æ I. Spectroscopic and analytical data for the complexes [AgSt(nx, y,z)2 ] [OTf] MHz, 0.90 (a, a@, t, [AgSt(7-3,4)2 ] [OTf] . dH(250.13 CDCl3) : 6.7 Hz, 12 H), 1.40 (b, b@, m, 32 H), 1.81 (c, c@, m, 8 H), 3JHH 3.94, 3.98 (d, d@, 2 t, 6.7 Hz, 8 H), 6.69 (l, AB, 16.5 3JHH transJlk Hz, 2 H), 6.71 (j, d, 8.5 Hz, 2 H), 6.88 (i, dd, 8.5 Hz, 3Jji 3Jji 1.8 Hz, 2 H), 6.96 (g, d, 1.8 Hz, 2 H), 7.18 (k, AB, 4Jig 4Jig 16.5 Hz, 2 H), 7.29 (n, AA@XX@, 6.4 Hz, 4 H), transJkl o Jno]Jno{ o 8.56 (o, AA@XX@, 6.4 Hz, 4 H).MHz, o Jon]Jon{ o dC(62.9 14.1 (a, a@), 22.7, 26.0, 29.2, 29.3, 31.9 (b, b@, c, c@), 69.0, CDCl3) : 69.2 (d, d@), 111.5 (g), 113.0 (j), 120.8 q, 320 Hz), (CF3 , JCF 121.4 (n), 121.5 (i), 121.8 (l), 128.1 (h), 135.9 (k), 147.5 (m), 149.1 (e), 150.6 (f ), 152.0 (o).MHz, [78 s). dF(235.36 CDCl3) : (CF3 , MHz, 0.89 (a@, t, [AgSt(6-3,5)2 ] [OTf] . dH(250.13 CDCl3) : 6.7 Hz, 12 H), 1.34 (b@, m, 24 H), 1.75 (c@, m, 8 H), 3.86 (d@, 3JHH t, 6.7 Hz, 8 H), 6.38 (e, t, 2.1 Hz, 2 H), 6.52 (g, d, 3JHH 4Jeg 4Jge 2.1 Hz, 4 H), 6.81 (l, AB, 16.2 Hz, 2 H), 7.15 (k, AB, transJlk 16.2 Hz, 2 H), 7.33 (n, AA@XX@, 6.4 Hz, 4 H), transJkl o Jno]Jno{ o 8.59 (o, AA@XX@, 6.4 Hz, 4 H).MHz, o Jon]Jon{ o dC(62.9 14.1 (a@), 22.7, 26.1, 29.3, 29.5, 29.6, 31.9 (b@, c@), 68.1 CDCl3) : (d@), 102.8 (e), 105.6 (g), 120.8 q, 320 Hz), 121.9 (n), (CF3 , JCF 124.3 (l), 136.2 (k), 136.9 (h), 147.3 (m), 152.0 (o), 160.5 (f).MHz, [78 s). dF(235.36 CDCl3) : (CF3 , MHz, 0.90 (a, aA, [AgSt(7-2,4)2 ] [OTf] . dH(250.13 CDCl3) : t, 6.7 Hz, 12 H), 1.40 (b, bA, m, 32 H), 1.81 (c, cA, m, 8 H), 3JHH 3.95, 3.96 (d, dA, 2 t, 6.7 Hz, 8 H), 6.39 (f, d, 2.1 Hz, 2 3JHH 4Jfj H), 6.41 (j, dd, 8.3 Hz, 2.1 Hz, 2 H), 6.88 (l, AB, 3Jji 4Jjg transJlk 16.5 Hz, 2 H), 7.35 (n, AA@XX@, 6.7 Hz, 4 H), 7.37 o Jno]Jno{ o (i, d, 8.3 Hz, 2 H), 7.60 (k, AB, 16.5 Hz, 2 H), 8.56 (o, 3Jij transJkl AA@XX@, 6.4 Hz, 4 H).MHz, 14.1 o Jon]Jon{ o dC(62.9 CDCl3) : (a, aA), 22.6, 26.0, 26.1, 29.1, 29.3, 31.8, 31.9 (b, bA, c, cA), 68.2, 68.5 (d, dA), 99.6 (f), 105.8 (j), 117.4 (h), 120.8 q, 320 (CF3 , JCF Hz), 121.5 (n), 122.0 (l), 128.9 (i), 131.4 (k), 148.9 (m), 151.0 (o), 158.7 (g), 161.6 (e).MHz, [78 s). dF(235.36 CDCl3) : (CF3 , MHz, 0.88 (a, t, [AgSt(7-3,4,5)2 ] [OTf] . dH(250.13 CDCl3) : 6.7 Hz, 6 H), 0.89 (a@, t, 6.7 Hz, 12 H), 1.39 (b, b@, m, 3JHH 3JHH 48 H), 1.78 (c, c@, m, 12 H), 3.95 (d@, t, 6.7 Hz, 8 H), 3.97 3JHH (d, t, 6.7 Hz, 4 H), 6.67 (g, s, 4 H), 6.79 (l, AB, 16.2 3JHH transJlk Hz, 2 H), 7.21 (k, AB, 16.2 Hz, 2 H), 7.40 (n, AA@XX@, transJkl 6.4 Hz, 4 H), 8.58 (o, AA@XX@, 6.4 Hz, o Jno]Jno{o oJon]Jon{ o 4 H).MHz, 14.1 (a, a@), 22.6, 26.1, 29.1, 29.3, dC(62.9 CDCl3) : 29.4, 30.4, 31.9 (b, b@, c, c@), 69.2 (d@), 73.6 (d), 105.9 (g), 120.8 q, 320 Hz), 121.8 (n), 123.0 (l), 130.4 (e), 136.3 (k), (CF3 , JCF 139.9 (h), 147.5 (m), 152.0 (o), 153.4 (f). MHz, dF(235.36 [78 s). CDCl3) : (CF3 , MHz, 0.85, 0.89 [AgSt(8-2,3,4)2 ] [OTf] .dH(250.13 CDCl3) : (a, a@, aA, 2 t, 6.7 Hz, 18 H), 1.39 (b, b@, bA, m, 60 H), 1.78 3JHH (c, c@, cA, m, 12 H), 3.95, 3.96, 4.02 (d, d@, dA, 3 t, 6.7 Hz, 12 3JHH H), 6.56 (j, d, 8.8 Hz, 2 H), 6.81 (l, AB, 16.5 Hz, 2 H), 3Jji transJlk 7.15 (i, d, 8.8 Hz, 2 H), 7.33 (n, AA@XX@, 6.7 3Jij o Jno]Jno{ o Hz, 4 H), 7.59 (k, AB, 16.5 Hz, 2 H), 8.59 (o, AA@XX@, transJkl 6.7 Hz, 4 H).MHz, 14.1 (a, a@, aA), o Jon]Jon{ o dC(62.9 CDCl3) : 22.7, 26.2, 29.3, 29.4, 29.5, 29.6, 30.4, 31.8, 31.9 (b, b@, bA, c, c@, cA), 68.7, 73.7, 74.4 (d, d@, dA), 108.4 (j), 120.8 q, 320 (CF3 , JCF Hz), 121.3 (i), 121.5 (n), 122.5 (h), 122.9 (l), 130.7 (k), 141.8 (f), 147.9 (m), 152.0 (o), 152.1 (g), 154.6 (e). MHz, dF(235.36 [78 s). Thermal data [T /°C (*H/kJ mol~1)] : CDCl3) : (CF3 , n\8, Crys Æ 119 (24) Æ I ; n\14, Crys Æ 65 (44) Æ Crys@ Æ72 (27) Æ I.MHz, 0.88, 0.89 [AgSt(8-2,4,5)2 ] [OTf] . dH(250.13 CDCl3) : (a, a@, aA, t, 6.7 Hz, 18 H), 1.39 (b, b@, bA, m, 60 H), 1.81 (c, 3JHH c@, cA, m, 12 H), 3.92, 3.93, 3.99 (d, d@, dA, 3 t, 6.7 Hz, 12 3JHH H), 6.42 (f, s, 2 H), 6.83 (l, AB, 16.5 Hz, 2 H), 7.04 (i, s, 2 transJlk H), 7.39 (n, AA@XX@, 6.4 Hz, 4 H), 7.64 (k, AB, o Jno]Jno{ o 16.5 Hz, 2 H), 8.57 (o, AA@XX@, 6.4 Hz, 4 H).transJkl o Jon]Jon{ o MHz, 14.1 (a, a@, aA), 22.7, 26.1, 26.2, 29.3, dC(62.9 CDCl3) : 29.4, 29.6, 31.8 (b, b@, bA, c, c@, cA), 69.2, 69.4, 70.6 (d, d@, dA), 99.7 (f), 113.5 (i), 116.4 (h), 120.8 q, 320 Hz), 121.5 (n), (CF3 , JCF 121.6 (l), 130.8 (k), 143.2 (j), 148.6 (m), 151.9 (o), 152.0 (e), 152.9 (g). MHz, [78 s).Thermal data dF(235.36 CDCl3) : (CF3 , [T /°C (*H/kJ mol~1)] : n\8, Crys Æ 58 (10) Æ Crys@ Æ110 (60) Æ I ; n\14, Crys Æ 67 (41) Æ Crys@ Æ80 (40) Æ I. References 1 G. W. Gray and J. W. Goodby, in Smectic L iquid Crystals : T extures and Structures, Leonard Hill, Glasgow, 1984; S. Diele and P. Goé ring, in Handbook of L iquid Crystals, ed.D. Demus, J. W. Goodby, G. W. Gray, H.-W. Speiss and V. Vill, Wiley-VCH, Weinheim, 1998 vol. 2B, ch. XIII, p. 887; A.-M. Levelut and M. Clerc, L iq. Cryst., 1998, 24, 105. 2 P. Sakya, J. M. Seddon, R. H. Templer, R. J. Mirkin and G. J. T. Tiddy, L angmuir, 1997, 13, 3706; A. Gulik, H. Delacroix, G. Kirschner and V. Luzzati, J. Phys. Fr. II, 1995, 5, 445; R. Vargas, New J. Chem., 1999, 275»286 285P.Mariani, A. Gulik and V. Luzzati, J. Mol. Biol., 1992, 225, 137; H. Delacroix, T. Gulik-Krzywicki, P. Mariani and V. Luzzati, J. Mol. Biol., 1993, 229, 526; T. Gulik-Krzywicki and H. Delacroix, Bio. Cell., 1996, 80, 193; H. Delacroix, T. Gulik-Krzywicki and J. M. Seddon, J. Mol. Biol., 1996, 258, 88. 3 J. M. Seddon and R. H. Templer, Philos. T rans. R.Soc. L ondon, Ser. A, 1993, 344, 377; P. Mariani, V. Luzzati and H. Delacroix, J. Mol. Biol., 1988, 204, 165; M. Clerc and A.-M. Levelut, J. Phys. Fr. II, 1991, 1, 1263. 4 M. J. Baena, P. Espinet, M. C. Lequerica and A.-M. Levelut, J. Am. Chem. Soc., 1992, 114, 4182. 5 V. S. K. Balagurusamy, G. Ungar, V. Percec and G. Johansson, J. Am. Chem. Soc., 1997, 119, 1539; S. D. Hudson, H.-T.Jung, V. Percec, W.-D. Cho, G. Johansson, G. Ungar and V. K. S. Balagurusamy, Science, 1997, 278, 449. 6 (a) C. Tschierske, Prog. Polym. Sci., 1996, 21, 775; (b) K. Borisch, S. Diele, P. Goé ring, H. Kresse and C. Tschierske, J. Mater. Chem., 1998, 8, 529; (c) K. Borisch, S. Diele, P. Goé ring and C. Tschierske, L iq. Cryst., 1997, 22, 427; (d) K. Borisch, S. Diele, P. Goé ring, H.Kresse and C. Tschierske, Angew. Chem., Int. Ed. Engl., 1997, 36, 2087. 7 C. Soulieç , P. Bassoul and J. Simon, J. Chem. Soc., Chem. Commun., 1993, 114. 8 G. W. Gray, B. Jones and F. Marson, J. Chem. Soc., 1957, 393; S. Diele, P. Brand and H. Sackmann, Mol. Cryst. L iq. Cryst., 1972, 17, 163; D. Demus, A. Gloza, H. Hartung, A. Hauser, I. Rapthel and A. Wiegeleben, Cryst. Res.T echnol., 1982, 16, 1445; P. Goé ring, S. Diele, S. Fischer, A. Wiegeleben, G. Pelzl, H. Stegemeyer and W. Tyen, L iq. Cryst., 1998, 25, 467; S. Kutsumizu, R. Kato, M. Yamada and S. Yano, J. Phys. Chem. B, 1997, 101, 10666 and references therein. 9 B. Kohne, K. Praefcke and J. Billard, Z. Naturforsch., T eil B, 1986, 41, 1036; J. Billard, H. Zimmermann, R. Poupko and Z. Luz, J. Phys. Fr., 1989, 50, 539; H.Zimmermann, J. Billard, H. Gutmann, E. J. Watchtel, R. Poupko and Z. Luz, L iq. Cryst., 1992, 12, 245. 10 S. Fischer, H. Fischer, S. Diele, G. Pelzl, K. Jankowski, R. R. Schmidt and V. Vill, L iq. Cryst., 1994, 17, 855; G. Lattermann and G. StauÜer, Mol. Cryst. L iq. Cryst., 1990, 191, 199; M. Schellhorn and G. Lattermann, L iq. Cryst., 1994, 17, 529; G.StauÜer, M. Schellhorn and G. Lattermann, L iq. Cryst., 1995, 18, 519; C. M. Paleos and D. Tsiourvas, Angew. Chem., Int. Ed. Engl., 1995, 34, 1696; K. Praefcke, B. Kohne, A. Eckert and J. Hempel, Z. Naturforsch., T eil B, 1990, 45, 1084; S. Yano, Y. Mori and S. Kutsumizi, L iq. Cryst., 1991, 9, 907. 11 (a) B. Donnio, B. Heinrich, T. Gulik-Krzywicki, H. Delacroix, D. Guillon and D.W. Bruce, Chem. Mater., 1997, 9, 2951; (b) D. W. Bruce, D. A. Dunmur, S. A. Hudson, E. Lalinde, P. M. Maitlis, M. P. McDonald, R. Orr, P. Styring, A. S. Cherodian, R. M. Richardson, J. L. Feijoo and G. Ungar, Mol. Cryst. L iq. Cryst., 1991, 206, 79; (c) D. W. Bruce and S. A. Hudson, J. Mater. Chem., 1994, 4, 479; (d) D. W. Bruce, B. Donnio, S. A. Hudson, A.-M. Levelut, S. Megtert, D.Petermann and M. Veber, J. Phys. II Fr., 1995, 5, 289. 12 (a) J. Malthe� te, H. T. Nguyen and C. Destrade, L iq. Cryst., 1993, 13, 171; (b) H. T. Nguyen, C. Destrade and J. Malthe� te, Adv. Mater., 1997, 9, 375; (c) B. Donnio and D. W. Bruce, J. Chem. Soc., Dalton T rans., 1997, 2745; (d) K. E. Rowe and D. W. Bruce, J. Mater. Chem., 1998, 8, 331; (e) W. Weiss—og, G. Pelzl, I.Letko and S. Diele, Mol. Cryst. L iq. Cryst., 1995, 260, 157; ( f ) H. T. Nguyen, C. Destrade and J. Malthe� te, in Handbook of L iquid Crystals, ed. D. Demus, J. W. Goodby, G. W. Gray, H.-W. Speiss and V. Vill, Wiley-VCH, Weinheim, 1998, vol. 2B, ch. XII, p. 865; (g) R. P. Tuffin, K. J. Toyne, J. W. Goodby and G. J. Cross, J. Mater. Chem., 1996, 6, 1271. 13 D. Tsiourvas, D. Kardassi, C.M. Paleos, S. Gehant and A. Skoulios, L iq. Cryst., 1997, 23, 269; C. M. Paleos, D. Kardassi, D. Tsiourvas and A. Skoulios, L iq. Cryst., 1998, 25, 267. 14 U. Stebani, G. Lattermann, R. Festtag, M. Wittenberg and J. H. Wendorf, J. Mater. Chem., 1995, 5, 2247; U. Stebani, G. Lattermann, M. Wittenberg, R. Festtag and J. H. Wendorf, Adv. Mater., 1994, 6, 572. 15 M. Lee, B.K. Cho, H. Kim and W. C. Zin, Angew. Chem., Int. Ed., 1998, 37, 638; M. Lee, B. K. Cho, H. Kim, J. Y. Yoon and W. C. Zin, J. Am. Chem. Soc., 1998, 120, 9168. 16 D. A. Hajduk, P. E. Harper, S. M. Gruner, C. C. Honeker, G. Kim, E. L. Thomas and L. J. Fetters, Macromolecules, 1994, 27, 4063; S. Foé rster, A. K. Khandpur, J. Zhao, F. S. Bates, I. W. Hamley, A. J. Ryan and W. Bras, Macromolecules, 1994, 27, 6922; M. Antonietti and C. Goé ltner, Angew. Chem., Int. Ed. Engl., 1997, 36, 910. 17 C. K. Ober and G. Wegner, Adv. Mater., 1997, 9, 17. 18 A. Skoulios and D. Guillon, Mol. Cryst. L iq. Cryst., 1988, 165, 317; Y. Hendrikx and A.-M. Levelut, Mol. Cryst. L iq. Cryst., 1988, 165, 233. 19 C. Tschierske, J. Mater. Chem., 1998, 8, 1485. 20 (a) J. L. Serrano, in Metallomesogens: Synthesis, Properties, and Applications, VCH, Weinheim, 1996; (b) D. W. Bruce, in Inorganic Materials, ed. D. W. Bruce and D. OœHare, Wiley, Chichester, 2nd edn., 1996; (c) A.-M. Giroud-Godquin, in Handbook of L iquid Crystals, ed. D. Demus, J. W. Goodby, G. W. Gray, H.-W. Speiss and V. Vill, Wiley-VCH, Weinheim, 1998, vol. 2B, ch. XIV, p. 901. 21 B. Donnio and D. W. Bruce, J. Mater. Chem., 1998, 8, 1993. 22 References for the reduction and oxidation reactions were taken from J. March, in Advanced Organic Chemistry: Reactions, Mechanisms, and Structure, Wiley, New York, 4th edn., 1992. 23 For a scope of the use of this reagent in organic synthesis, see : L. Brandsma, H. Verkruijsse, in Preparative Polar Organometallic Chemistry, Springer-Verlag, Berlin, 1987, vol. 1; L. Brandsma, in Preparative Polar Organometallic Chemistry, Springer-Verlag, Berlin, 1990, vol. 2. 24 D. W. Bruce, D. A. Dunmur, E. Lalinde, P. M. Maitlis and P. Styring, L iq. Cryst., 1988, 3, 385. 25 D. W. Bruce, D. A. Dunmur, S. A. Hudson, P. M. Maitlis and P. S. Styring, Adv. Mater. Opt. Electron., 1992, 1, 37. 26 W. Wedler, D. Demus, H. Zaschke, K. Mohr, W. Schaé fer and W. Weiss—og, J. Mater. Chem., 1991, 1, 347; W. Wedler, P. Hartmann, U. Bakowsky, S. Diele and D. Demus, J. Mater. Chem., 1992, 2, 1195. 27 P. Sotta, J. Phys. II Fr., 1991, 1, 763. 28 H. A. Adams, N. A. Bailey, D. W. Bruce, S. C. Davis, D. A. Dunmur, P. D. Hempstead, S. A. Hudson and S. Thorpe, J. Mater. Chem., 1992, 2, 395. 29 B. Heinrich and D. Guillon, work in progress. 30 J. W. Goodby, D. A. Dunmur and P. J. Collings, L iq. Cryst., 1995, 19, 703. 31 M. Ibn-Elhaj, D. Guillon, A. Skoulios, A.-M. Giroud-Godquin and J. C. Marchon, J. Phys. II Fr., 1992, 2, 2197; R. Seghrouchni and A. Skoulios, J. Phys. II Fr., 1995, 5, 1385; N. Usoltœseva, G. Hauck, H. D. Koswig, K. Praefcke, B. Heinrich and D. Guillon, L iq. Cryst., 1996, 20, 731; B. Heinrich, K. Praefcke and D. Guillon, J. Mater. Chem., 1997, 7, 1363; N. Usol@tseva, P. Espinet, J. Buey and J. L. Serrano, J. Mater. Chem., 1997, 7, 215; K. Saito, A. Sato and M. Sorai, L iq. Cryst., 1998, 25, 525. 32 H. Ringsdorf, B. Schlarb and J. Venzmer, Angew. Chem., Int. Ed. Engl., 1988, 27, 113; D. Blunk, K. Praefcke and V. Vill, in Handbook of L iquid Crystals, ed. D. Demus, J. W. Goodby, G. W. Gray, H.-W. Speiss and V. Vill, Wiley-VCH, Weinheim, 1998, vol. 3, ch. VI, p. 305. 33 B. S. Furniss, A. J. Hannaford, P. W. G. Smith and A. R. Tatchell, in V ogelœs T extbook of Practical Organic Chemistry, Longman Scienti–c and Technical, Wiley, Glasgow, 5th edn., 1989. Paper 8/09320B 286 New J. Chem., 1999, 275&raq

ISSN:1144-0546    

DOI:10.1039/a809320b

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Ultrasounds in voltammetry of irreversible processes of silicon organic compounds In search of thermodynamic consequences§ Viatcheslav Jouikov,* Claude Bernard and Chantal Degrand SEESIB Equipe Electroanalyse et Organique, (CNRS UMR 6504), Electrosynthe` se Universiteç Blaise Pascal, 63177 France Aubie` re ceç dex, Received (in Montpellier, France) 10th November 1998, Accepted 28th December 1998 The electron transfer in the course of the electrochemical oxidation of hexamethyldisilane and the electroreduction of chlorotrimethylsilane in anhydrous acetonitrile was shown to follow a dissociative mechanism.The same processes carried out upon sonication with ultrasounds of low (20 kHz) and high (500 kHz) frequency did not reveal any decrease of the eÜective potential.The eÜect of ultrasounds was only to intensify the mass transfer and to accelerate the Cl»F exchange of chlorotrimethylsilane with the —uorine anion provided by the supporting electrolyte. Les ultrasons en voltameç trie des processus irreç versibles des composeç s organique du silicium. A la recherche de conseç quences thermodynamique. Il a eç teç deç montreç que les reç actions eç lectrochimiques dœoxydation de lœhexameç thyldisilane et de reç duction du chlorotrimeç thylsilane dans lœaceç tonitrile anhydre avaient lieu suivant un meç canisme de transfert dœeç lectron dissociatif.Lorsque ces me� mes reç actions sont conduites en preç sence dœultrasons de basse ou de haute freç quence (20 ou 500kHz), aucune diminution du potentiel eÜectif nœest observeç .Lœunique eÜet des ultrasons a eç teç dœaccroï� tre le transfert de masse et dœacceç leç rer lœeç change Cl»F du chlorotrimeç thylsilane avec lœanion —uorure de lœeç lectrolyte support. The in—uence of ultrasounds on electrochemical processes and, above all, on the mass-transfer intensi–cation has been studied in a series of papers.1h5 The lowering the potential of the irreversible oxidation of aromatic disul–des under the action of ultrasounds6 has also been reported.Since the ultrasounds do not aÜect characteristics of stepwise processes,7h9 one can suppose that it acts by disturbing the diÜusion layer or by depassivating the electrode surface of adsorbed particles. Besides accelerating mass transfer and clearing the electrode surface, one can suppose another possible action of ultrasounds on voltammetric characteristics of irreversible processes.Sonication of an electrolyte can aÜect the double layer structure and, hence, the eÜective potential at the reaction site, so this potential might oscillate with the applied frequency. Higher harmonics of ultrasounds with a main frequency of several hundreds of kHz might overlap with the beginning of a zone of non-linearity in dielectric properties with fre- (es) quency of common organic solvents.As a result, the solvent reorganization energy might change, resulting in a variation in the electron transfer activation barrier. On the other hand, the electrosynthesis of silicon organic compounds is a very promising –eld of chemistry and an eventual possibility to modify the reaction mechanism by ultrasounds could be a good way to expand the synthetic potentialities of this method.The purpose of this work was to determine the nature of the electron transfer in electrochemical oxidation and reduction of some silicon-containing compounds and to clarify possible consequences of the action of ultrasounds on electrochemically irreversible processes.* Present address : Department of Chemistry, University of Arizona, Tucson AZ 85721, USA. § Non-SI units employed: 1 eVB9.65 J mol~1; 1 calB4.18 J. Results and discussion Electrooxidation of hexamethyldisilane Upon oxidation at a stationary Pt oxidized electrode, hexamethyldisilane shows an oxidation peak V vs. (Ep\1.667 SCE) whose current is linear with concentration and with the square root of the potential sweep rate.The ratio ip m~1@2 remains constant within the sweep rate interval 0.05»1000 V s~1. This indicates that the process is diÜusion-controlled under the given experimental conditions. Peak width (at m\1 V s~1) corresponds to the transfer coefficient a\0.38. The number of electrons involved in oxidation was shown to be 2 per molecule.The slope of the dependence of on the sweep Ep rate is 76 mV (Fig. 1). [*Ep/*log(m)] In the process of cation elimination, the cation is a Me3Si` better leaving group than the proton.10 For this reason the former does not exist in its free form in solution.11 This particle is immediately intercepted by a nucleophile (e.g., the anion of the supporting salt in the double or reaction layer) at the stage of charge induction in the transition state.Being nucleophile-assisted, electron transfer is therefore dissociative. The dissociative character of the process can also be deduced from the following consideration. Three main parameters, which determine the reactivity of the disilane upon oxidation [energy of the highest occupied molecular orbital, eHOMOD standard potential of the leaving E(Me3Si’`Me3Si`)@Me3SiSiMe3 0 ; group, and the energy of homolytic dissociation EMe3Si`@Me3Si’ 0 ; of a bond being broken in the process D(SiwSi)], are related by the equation obtained from the analysis of the thermochemical cycle of oxidative cleavage of the SiwSi bond: E(Me3Si’`Me3Si`)@Me3SiSiMe3 0 \EMe3Si`@Me3Si’ 0 ]D(SiwSi)[T *S.(1) New J. Chem., 1999, 287»290 287Fig. 1 Oxidation potentials of hexamethyldisilane plotted against Ep log(m) ; oxidized Pt stationary electrode of 1 mm diameter, concentration of is 10~3 mol l~1; T \20 °C. (Me3Si)2 Potentials are taken with a positive sign for oxidation processes ; *S is the change of entropy for the SiwSi bond dissociation, for *S\SMe3Si’]SMe3Si’[SMe3SiSiMe3 ; Me3 SiSiMe3 both radical fragments are similar, so *S\2SMe3Si’[ SMe3SiSiMe3 .Taking into account the relation between the eÜective and the standard potentials and assuming the linear approach of the quadratic Marcus equation, one obtains a ratio between D(SiwSi), and similar to that derived for the Ep EMe3Si`@Me3Si’ 0 , reduction of organic halides :12 D(SiwSi)+2/3(Ep[EMe3Si`@Me3Si’ 0 ) ][2/3(2*Gº[k0/2]T *S[z0)] (2) or D(SiwSi)+2/3(Ep[EMe3Si`@Me3Si’ 0 )]C, (3) where and are the solvent reorganization energy and the k0 z0 outer Helmholtz plane potential, respectively.The value of C was shown to vary slightly within the interval 0.25»0.3 eV for a large number of organic halogen compounds (AlkX, etc.).13,14 Now, taking for C an ArCH2X, average value 0.27 and using the known value D298 0 (SiwSi)\ eV,15 one obtains the value of the standard oxidation 3.49 potential of the leaving group cation) (Me3Si` EMe3Si`@Me3Si’ 0 \ V.This value is very negative and for this reason the [3.19 oxidation of silyl radicals to silicenium cations has a large driving force throughout most of the range of accessible potentials and the process keeps its two-electron character for all sweep rates.Using eqn. (1) and taking T *S equal to [0.25 eV (the maximal value for such processes13), one obtains the standard potential of dissociative oxidation of hexamethyldisilane V. This value is 1 V less posi- E(Me3Si’`Me3Si`)@Me3SiSiMe3 0 \0.525 tive as compared to the eÜective potential and this fact is in Ep good agreement with the dissociative character of the process.The theoretical value of the transfer coefficient a, calculated for the sweep rate m\0.1 V s~1,16 is 0.32, whereas the experimental value, derived from the peak width, was found to be 0.38. Such a reduced value is related to the overestimation of the reorganization factor because of an uncertainty in the k0 size of the fragment taken for the calculation. Sinhe Me3Si trimethylsilicenium cation does not exist in a ìì free œœ form in solution, its hard sphere eÜective radius is substantially larger than that corresponding to its own size.This fact provokes a decrease of and, therefore, the observed increase in a. k0 The oxidation of hexamethyldisilane has been studied under sonication with high (500 kHz) and low (20 kHz) frequencies. Low frequency ultrasounds caused a total increase of the oxidation current but the shape of the signal was not very informative (Fig. 2). When oxidizing the disilane under sonication by high frequency ultrasounds, the process was characterized by practically the same value of the oxidation potential as in the absence of ultrasounds at a freshly depassivated electrode (Fig. 2). It was shown that several consecutive sweeps do not cause a shift of the oxidation peak towards more positive potentials as in the case of non-sonochemical oxidation.Besides this, there were no changes in the electrochemical behaviour, so this phenomenon can probably be attributed to the cleaning of the electrode surface. We did not succeed in evaluating the parameters for the characteristic dependencies of the oxidation of hexamethyldisilane under sonication with high frequency (500 kHz) ultrasounds because of the low precision of the measurements under these conditions.According to the Levich equation, the growth of the limiting current upon sonication corresponds to an increase of the rotation speed of a conventional rotating disk electrode of the same surface by approximately 1600 times (Fig. 2). The oxidation potential (at least remains practically unchanged. Thus, Ep@2) no evidence for in—uence of ultrasounds on the characteristics of dissociative oxidation of hexamethyldisilane was obtained. Electroreduction of chlorotrimethylsilane The cathodic process»conventional and sonoelectrochemical reduction of chlorotrimethylsilane»has been considered as well.Usually, in order to obtain a well-shaped reduction peak of chlorotrimethylsilane, one must operate under inert atmosphere using a freshly distilled solvent, vacuum-dried supporting salt and activated Even doing so, the reduction Al2O3 . signal is observable during several minutes only. After 5»10 min it transforms into another signal, presumably attributed to the reduction of a hydroxy or a —uoro derivative.When a —uoro-containing supporting electrolyte is used in an anhy- Fig. 2 Voltammograms of the oxidation of (C\10~3 mol (Me3Si)2 l~1) at a Pt oxidized stationary electrode ; M CH3CN»0.01 m\1 V s~1, T \20 °C. (a) without ultrasound; (b) and (c) Bu4NBF4 , upon sonication at 20 and 500 kHz, respectively. 288 New J. Chem., 1999, 287»290Fig. 3 Voltammograms of the reduction of (C\8]10~3 Me3SiCl mol l~1) at a Pt oxidized stationary electrode ; M CH3CN»0.1 m\1 V s~1, T \20 °C.(a) without ultrasound; (b) upon Me4NBF4 , sonication at 500 kHz. drous solvent, it is probably the Cl»F exchange that provides the —uorosilane.17,18 This exchange is promoted by nucleophilic solvents or by Lewis acids. Fluoride ion arises from the supporting salt anion as the Cl»F exchange reaction consuming F~ progressively displaces the equilibrium of its formation from BF4~.Upon reduction in M media CH3CN»0.1 Me4NBF4 without ultrasounds, chlorotrimethylsilane shows a quite distinguishable signal. The peak current, is linear with concen- ip , tration when C[5]10~3 mol l~1 and the ratio m~1@2 ip remains constant within the range of sweep rates 0.05»20 V s~1, indicating a diÜusional or quasi-diÜusional character of the process.The high negative reduction potential (Ep\ V vs. SCE), as well as the high dissociation energy of [2.68 the SiwCl bond, the number of electrons n\2 and a\0.29 suggest the dissociative character of the electron transfer in the reduction of this compound. The estimation of the standard potential EMe3SiCl@Me3Si’`Cl~ 0 using eqn.(1) and the known gas phase bond dissociation energy for the SiwCl bond [D(SiwCl)\5.49 eV19] provide the standard potential for the dissociative reduction of this compound, as about [3.3 V. Since the EMe3SiCl@Me3Si’`Cl~ 0 , reduction occurs at a less negative potential than the calculated standard potential, the reductive cleavage of the SiwCl bond most probably includes a substantial polarization of this bond in a polar solvent and/or a preceding (or simultaneous) interaction with the cation of the supporting electrolyte.By weakening and lengthening the SiwCl bond, both these factors lower the eÜective bond strength D(SiwCl) with respect to its value in the gas phase. The eÜective SiwCl bond energy, estimated according to eqn.(3) (taking V, C\0.27 eV13 and ECl’@Cl~ 0 \1.89 Ep\ V), should be as large as 3.3 eV or about 76 kcal [2.68 mol~1. Evidently, some interactions, like pre-dissociation or ion-pair formation, and the increasing electrophilicity of silicon in the chlorosilane in solution play an important role during the electroreduction of In the absence of such Me3SiCl. interactions this chlorosilane would not be reducible at all because it would have had a peak potential Ep+[(5.49 V.Given that the SiwF bond is ](3/2)[0.27[1.89)\[6 stronger than the SiwCl bond, similar factors probably take place in the reduction of —uorosilanes too. Indeed, at the observed reduction potential V) and the stan- (Ep\[3.15 dard potential of the leaving group V13), the (EF’@F~ 0 \2.62 eÜective SiwF bond energy should have the value D(SiwF)\(2.62]3.15)(2/3)]0.27\3.58 eV or about 82 kcal mol~1.When tracing reduction curves of chlorotrimethylsilane upon sonication ( f\500 kHz), the signal of the chlorosilane does not shift, but the signal of a —uoro derivative appears a few seconds after ultrasounds are applied (Fig. 3). The chlorosilane peak disappears and no peak at less negative potentials is observed. Apparently, the action of the ultrasounds is purely kinetic and consists in acceleration of the Cl»F exchange, while the concentration of residual water is not large enough to compete with the F~anion, which is, in addition, known to be a weak nucleophile.It is to be noted that ultrasonic intensi–cation of electrochemical processes involving chlorosilanes (even the processes of ììanodic reductionœœ20) are widely used for the electrosynthesis of carbosilanes21,22 but no phenomena related to the thermodynamics of the process were reported. Experimental The oxidation of hexamethyldisilane (Aldrich) was carried out in a M solution at an oxidized plati- CH3CN»0.01 Bu4NBF4 num electrode using a PAR-273 potentiostat.The auxiliary electrode was a Pt wire and the reference electrode was a SCE. For the reduction of chlorotrimethylsilane, the reference electrode was separated from the analyte by an electrolytic bridge –lled with a 0.1 M solution of in in order Me4NBF4 CH3CN to prevent AgCl formation at the tip of the reference electrode. Acetonitrile (Aldrich) was distilled over immediately CaH2 prior to use.Ultrasonic experiments were carried out in a previously described23 Te—on cell of 60 ml using a PAR-273 potentiostat and an ultrasonic generator (Sonic and Materials, Inc., Connecticut). The ultrasonic power at 20 kHz, as measured by calorimetry, was 2.4 W cm~2. At 500 kHz, the output power of the generator was 20 W, which, given the bottom area of the cell equal to 19.63 cm2, approximately corresponds to an apparent total power of 1 W cm~2.Notes and references 1 V. Yegnarman and S. Bharathi, Bull. Electrochem., 1992, 8, 84. 2 F. Marken, D. L. Goldfarb and R. G. Compton, Electroanalysis, 1998, 10, 562. 3 R. P. Akkermans, M. Wu, C. D. Bain, M. Fidel-Suarez and R. G. Compton, Electroanalysis, 1998, 10, 613. 4 F. Marken, R. P. Akkerman and R. G.Compton, J. Electroanal. Chem., 1996, 415, 55. 5 R. G. Compton, J. C. Eklund, F. Marken, T. O. Rebbitt, R. P. Akkerman and D. N. Walder Electrochim. Acta, 1997, 42, 19. 6 C. Bernard, C. Degrand, F. Moutin and J. Klima, Journees dœElectrochimie de Strasbourg, 1995, pp. CO 1»3. 7 J. Klima, C. Bernard and C. Degrand, J. Electroanal. Chem., 1994, 367, 297. 8 A. Durant, H.Francois, J. Reisse and A. Kirsch-Demesmaeker, Electrochim. Acta, 1996, 41, 2. 9 F. Marken, S. Kumbhat, G. H. W. Sanders and R. G. Compton, J. Electroanal. Chem., 1996, 414, 95. 10 S. Fornarini, J. Org. Chem., 1988, 53, 1314. 11 G. Olah, G. Rasul, L. Heiliger, J. Baush and G. K. Prakash, J. Am. Chem. Soc., 1992, 114, 7734. 12 R. Popielarz and D. R. Arnold, J. Am. Chem. Soc., 1990, 112, 3068. 13 C. P. Andrieux, A. Le Gorande and J.-M. Saveant, J. Am. Chem. Soc., 1992, 114, 6892. 14 J.-M. Saveant, Acc. Chem. Res., 1993, 26, 455. 15 CRC Handbook of Chemistry and Physics, 75nd edn., ed. D. R. Lide, The Chemical Rubber Co., Cleveland, 1994, p. 9. 16 V. Zhuikov, Zh. Obslch. Khim., 1997, 67, 1037. 17 A. Kunai, E. Toyoda, T. Kawakami and M. Ishikawa, Electrochim. Acta, 1994, 39, 2089. 18 A. Kunai, E. Toyoda, T. Kawakami and M. Ishikawa, Organometallics, 1992, 11, 2899. New J. Chem., 1999, 287»290 28919 R. Walsh, in T he Chemistry of Organic Silicon Compounds, eds. S. Patai and Z. Rappoport, John Wiley & Sons Ltd, New York, 1989, p. 371. 20 A. J. Fry, J. Touster, U. N. Sirasoma and B. Raimundo, in Electroorganic synthesis, eds. R. D. Little and N. L. Weinberg, Marcel Dekker, Inc., New York, 1991, p. 99. 21 A. J. Fry, J. Touster and U. N. Sirasoma, J. Electrochem. Soc., 1990, 137, 521. 22 M. Bordeau, C. Biran, M.-P. Leger-Lambert and J. Dunogues, J. Chem. Soc., Chem. Commun., 1991, 1476. 23 J. Klima, C. Bernard and C. Degrand, J. Electroanal. Chem., 1995, 399, 147. Paper 8/08808J 290 New J. Chem., 1999, 287»290

ISSN:1144-0546    

DOI:10.1039/a808808j

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Dissolution of cholesterolñcalcium bilirubinate compressed discs by microemulsions Amina Ben Mouaz,a Marc Lindheimer,*a Anne-Marie Montet,b Serge Lagergea and Jean-Claude Montetb a L aboratoire des et Inorganiques (CNRS ESA 5072), Agreç gats Moleç culaires Mateç riaux Montpellier II, CC015, Place Bataillon, 34095 Montpellier, France Universiteç Euge` ne b INSERM, L aboratoire de Physiopathologie 46 Boulevard de la Gaye, Heç patique, 13009 Marseille, France Received (in Montpellier, France) 24th November 1998, Accepted 23rd December 1998 The purpose of the present work was to characterize the composition and the structure of a monophasic solution able to dissolve cholesterol»calcium bilirubinate compressed discs simulating cholesterol and pigment biliary stones.Thermodynamically stable microemulsions composed of methyl tert-butyl ether, sodium dodecyl sulfate, n-butanol, and of an aqueous ethylenediaminetetraacetate solution were investigated.Their structure was de–ned by viscosity and conductivity measurements. The kinetics of compressed disc dissolution served to determine the optimal composition of the mixtures for rapid dissolution of the two types of discs.All the experimental results show that cholesterol and calcium bilirubinate dissolutions occur simultaneously and that the more efficient microemulsions are those exhibiting a bicontinuous structure. Dissolution de pastilles de cholesteç rolñbilirubinate de calcium par des microeç mulsions. Cette eç tude a pour objet de deç –nir la composition et la structure dœune solution monophasique capable de dissoudre des pastilles de cholesteç rol et de bilirubinate de calcium simulant les calculs biliaires cholesteç roliques et pigmentaires humains.Des microeç mulsions thermodynamiquement stables composeç es de meç thyl tertio-butyl eç ther, de dodeç cylsulfate de sodium, de n-butanol et dœune solution aqueuse dœeç thyle` ne diamine teç traaceç tate ont eç teç eç tudieç es.Leur structure a eç teç caracteç riseç e par des mesures de viscositeç et de conductiviteç . Lœeç tude cineç tique de la dissolution des pastilles placeç es au contact de ces microeç mulsions montre que les dissolutions du cholesteç rol et du bilirubinate se font simultaneç ment et que les meilleures performances sont obtenues avec des microeç mulsions a` structure bicontinue. The treatment of biliary lithiasis is essentially surgical to date.However, for high operative risk patients, a nonsurgical alternative is required. Recent advances in endoscopic catheterization now permit the direct instillation of stone solvents into the biliary system. The –rst clinical studies consisted in injecting methyl tert-butyl ether (MTBE) in the gallbladder by percutaneous transhepatic cannulation, allowing at least 95% dissolution of cholesterol stones in 72 of 75 patients.1,2 On the other hand, we have shown, in vitro and in animals, that it was possible to dissolve mixed stones composed of cholesterol and pigments with multicomponent contact solvents.3,4 These solvents were MTBE»dimethyl sulfoxide mixtures, and a complex aqueous solution intended to dissolve the noncholesterol part of stones.When applied alternately these solvents led to complete and rapid disappearance of stones. Furthermore, we reported that in patients with large duct stones, contact litholysis therapy contributed to bile duct clearance in 40 of 44 patients.5 However, alternating infusion with the solvents constitutes a heavy procedure in clinical practice, and it would be useful to dispose of one monophasic solvent efficiently and simultaneously to dissolve the main biliary stone components.In this work, we have studied the eÜectiveness of diÜerent types of microemulsions to dissolve cholesterol»calcium bilirubinate compressed discs. Viscosimetric and conductometric analysis of these microemulsions allow us to relate their solubilizing power to their structure.Thus, we show that the microemulsions exhibiting a bicontinuous structure are the most efficient for dissolving model biliary stones. This work is a step in the development of new topical solvents for the efficient dissolution of the two major stone types in the human biliary tract. Experimental Materials Sodium dodecyl sulfate (SDS) and MTBE were supplied by Sigma Chemical Co (St Louis, Mo).Ethylenediaminetetraacetic acid (EDTA), bilirubin and anhydrous cholesterol were provided by Fluka (Buchs, Switzerland). Tritiated cholesterol. M[1a, 2a (n) 3H] cholesterolN with a speci–c activity of 50 lCi mmol~1 was a CEA product (France). n-Butanol and the kit for cholesterol analysis were purchased from Merck (Darmstadt, Germany) and Boehringer-Mannheim (Mannheim, Germany), respectively.All other products used in this investigation were of A grade. Methods Preparation of the disc components. The anhydrous cholesterol was recrystallized three times from an ethyl alcohol solution at 333 K. Monohydrate cholesterol was prepared as described by Igimi and Carey.6 Calcium bilirubinate (6.4% Ca) was synthesized from diacidic bilirubin according to Leuschner et al.7 New J.Chem., 1999, 291»296 291Fig. 1 Diagrammatic representation of rotating disc apparatus for dissolution rate studies. Microemulsion preparation. Microemulsions consisted of MTBE, of a 1% (w/w) aqueous solution of EDTA buÜered at pH 9.4 with 50 mM NaOH»glycine, of an anionic surfactant (SDS) and of a cosurfactant (n-butanol).The SDS]n-butanol mixture was prepared to achieve a mass ratio of 1 to 2. The pseudoternary diagram of this system has been plotted in order to delimit the monophasic region. This diagram was determined by adding increasing amounts of the SDS»nbutanol mixture to the EDTA aqueous solution. DiÜerent quantities of MTBE were then introduced into each of these solutions.The resulting solution was equilibrated by gentle rotation in a thermostat at a constant temperature of 310^0.5 K and the occurrence of a possible phase separation was observed by nephelometry. The monophasic region being de–ned, diÜerent microemulsions were then prepared to investigate their viscosity and conductivity. Viscosity study. Viscosity measurements were performed using a Couette viscosimeter to ensure the Newtonian —ow of the microemulsions. The kinematic viscosities (m) were evaluated by means of an Ubbelohde tube (Viscosimetric MS, Fica) placed in a thermostated bath whose temperature was regulated at 310^0.5 K.Densities (q) of the solutions were measured with a PAAR DMA 602 density meter, at 310^0.5 K. Conductivity study.All conductometric analyses were performed with a Knick 702 conductometer at 310^0.5 K and in a closed-circuit device to avoid evaporation of the organic phase. The microemulsions were obtained as follows : mixtures of (SDS]n-butanol)»EDTA aqueous solution were prepared to achieve mass ratios of 0.33, 0.54, 0.82, 1.00, 1.22, 1.86 and 3.00. Known volumes of MTBE were then progressively added to each of the previous mixtures using a dilution device in order to cover all the monophasic range.The medium was continuously homogenized by means of a peristaltic pump and the conductivity measurements were accomplished after each addition of MTBE. Solubility of cholesterol and calcium bilirubinate. The solubility of cholesterol monohydrate in the above described microemulsions has been evaluated as follows : each microemulsion containing a monohydrate cholesterol excess was maintained under agitation in a thermostat at 310^0.5 K.The supernatants collected from each microemulsion were analysed for the solubilized cholesterol content using the enzymatic kit : the concentration of the lutidine dye (3,5-diacetyl- 1,4-dihydrolutidine) formed is stoichiometric to the amount of cholesterol and is measured by the increase of light absorbance at 405 nm.The solubility of calcium bilirubinate was measured in the dark because of the high photosensitivity of bilirubin, which easily oxidizes to biliverdin. The microemulsions were placed in the presence of calcium bilirubinate powder, under gentle agitation, in a thermostat at 310^0.5 K for 24 h.Supernatants collected from each microemulsion were diluted with methyl alcohol, and the calcium bilirubinate concentrations assayed spectrophotometrically at 453 nm. Preparation of the discs. Compressed discs composed of 80% cholesterol and 20% calcium bilirubinate w/w or 80% calcium bilirubinate and 20% cholesterol w/w, simulating mixed cholesterol stones and pigment stones, respectively, were prepared.Calcium bilirubinate-rich discs contained additional tritiated cholesterol. The mixtures were homogenized by using an agate mortar, then an aliquot of about 350 mg was compressed at 2300 Kg cm~2 for 1 min to form a 1.30 cm diameter disc. Dissolution kinetics. Dissolution kinetics were studied by using the rotating disc procedure.6,8 This method consists in immersing a disc, initially –xed at the end of a rod, in 25 ml of microemulsion (Fig. 1). Te—on was applied to the upper disc face to ensure good waterproo–ng and to avoid wetting. The rod was rotated at 100 rpm. During the dissolution process (lasting 2»8 h) 25 ll of microemulsion was withdrawn every 5 min and assayed for bilirubin and cholesterol. All dissolution kinetics measurements were performed at 310^0.5 K in the dark.Chemical analyses of bilirubin and cholesterol. The dissolved bilirubin was diluted with methyl alcohol prior to analysis by spectrophotometry at 453 nm. Cholesterol was assayed with the enzymatic method and by radioactivity determination, for high- and low-cholesterol discs, respectively. Results and discussion Phase diagram The pseudoternary diagram SDS»aqueous phase»MTBE exhibited a small monophasic region (Fig. 2). This region increased signi–cantly when a mixture of SDS]n-butanol was substituted for SDS. Consequently, n-butanol as cosurfactant is necessary to obtain a large monophasic zone. This zone includes (SDS]n-butanol)»aqueous phase mixtures with mass ratios between 0.33 and 3.00.Microemulsion viscosity Fig. 3 shows the viscosity of the microemulsions as a function of the amount of MTBE for diÜerent (SDS]n-butanol)» aqueous phase mass ratios. Without MTBE, increasing the (SDS]n-butanol)»aqueous phase mass ratio by ten led to a twofold higher viscosity. With progressive addition of MTBE, the viscosity of mixtures with the highest (SDS]n-butanol) contents regularly decreased.Moreover, for MTBE proportions higher than 30%, the microemulsions exhibited similar viscosities for all (SDS]n- 292 New J. Chem., 1999, 291»296Fig. 2 Pseudoternary phase diagrams: The hatched area corresponds to the monophasic region of the SDS»aqueous phase»MTBE diagram. The grey area corresponds to the monophasic region of the (SDS]n-butanol)»aqueous phase»MTBE diagram.The horizontal line represents microemulsions containing 30% (SDS]n-butanol). butanol)»aqueous phase mass ratios. For the lowest (SDS]n-butanol) contents (mass ratios 0.33 and 0.54), the dynamic viscosity –rst increased and then reached a maximum with a 15% and 10% MTBE fraction, respectively. These mixtures correspond to points A and B in Fig. 5 (shown later).This viscosity pattern suggests the occurrence of micellar aggregates whose size increases with the addition of MTBE. Moreover, bringing the aggregates closer together favours their mutual interactions, which might consequently increase the viscosity. The decrease in the viscosity pattern with the subsequent addition of MTBE occurs from a particular MTBE weight fraction, which may indicate a structure transition from the aqueous to the bicontinuous medium.For the mixtures with large surfactant»cosurfactant contents in which the viscosity continuously decreased as the amount of MTBE increased, it seems that the n-butanol concentration was high enough to play the role of an organic phase. Thus, the continuously declining dynamic viscosity may already indicate the occurrence of bicontinuous structure systems.This supports the results of Kathopoulis9 and Lindman et al.10 on the octylbenzenesulfonate»n-pentanol»ndecane »water system. Conductivity of microemulsions Fig. 4 represents the conductivity of (SDS]n-butanol)» aqueous phase mixtures to which were added increasing Fig. 3 Dynamic viscosity of microemulsions as a function of the amount of MTBE for various (SDS]n-butanol)»buÜer mass ratios : 0.33, 0.54, 0.82, (]) 1.22, (») 1.86, 3.()) (L) (|) (K) Fig. 4 Conductivity of microemulsions as a function of MTBE percentage for various (SDS]n-butanol)»buÜer mass ratios : 0.33, ()) 0.54, 0.82, (]) 1, (]) 1.22, (») 1.86, 3. (L) (|) (K) amounts of MTBE. Microemulsions containing large proportions of water had the highest conductivity.The conductivity decreased continuously when the microemulsions were enriched with MTBE and tended to very low values. Considering the microemulsions with (SDS]n-butanol)»aqueous phase mass ratios of 1.22, 1.86 and 3.00, one observes that the conductivity becomes extremely low for microemulsions containing more than 45% MTBE. The portion of the diagram corresponding to the transition between the bicontinuous and the reverse medium was determined by the composition of mixtures for which the slope of the curves v vs.% MTBE became nearly constant. This transition area is reported in Fig. 5 and passes through points C (60% MTBE), D (52% MTBE) and E (45% MTBE). Moreover, concerning the conductivity values versus the weight fraction of MTBE with microemulsions containing 30% (SDS]n-butanol) (Fig. 6), one notes that the decrease in conductivity is more marked for MTBE proportions higher than about 15% (Fig. 5, point F). Such a percentage was also found in the viscosity pattern. For weight fractions higher than 55% MTBE (Fig. 5, point G), the slope of the curve tends to stabilize. This limiting point is located around the area separating bicontinuous and reverse media.Concerning the microemulsions without MTBE, maximum conductivity was reached with about 35»40% (SDS]n-butanol) (Fig. 5, point H) ; for larger percentages, the micellar structure no longer occurs. These data are in accordance with the transitions previously described on the basis of viscosity measurements between the aqueous and the bicontinuous medium.Fig. 5 Delineation of boundaries between aqueous (I), bicontinuous (II) and reverse (III) media. Changes in structure were in fact gradual. New J. Chem., 1999, 291»296 293Fig. 6 Conductivity of microemulsions containing 30% (SDS]nbutanol) as a function of MTBE. BuÜer conductivity (1.071 S m~1) is indicated by the horizontal line. The boundaries between the various media are reported in Fig. 5. Solubility of cholesterol monohydrate and calcium bilirubinate in microemulsions The solubility of cholesterol monohydrate was evaluated in microemulsions containing (SDS]n-butanol)»aqueous phase mixtures with mass ratios between 0.33 and 3.00 (Fig. 7). As expected, the cholesterol solubility increased with concentration of MTBE in the microemulsions. In some cases, this solubility was even higher than the solubility of cholesterol monohydrate in pure MTBE (0.418 mol l~1).This clearly indicates that (SDS]n-butanol) is actually involved in the dissolution of cholesterol. Likewise, given the experimental results with increasing quantities of surfactants, it is evident that SDS micelles act as a solubilizing agent. The most efficient microemulsions, which exhibited a cholesterol solubility higher than that obtained in pure MTBE, correspond to media with a bicontinuous structure.The solubility data of calcium bilirubinate (Fig. 8) show that the higher the aqueous phase in the microemulsion, the higher the solubility. The presence of EDTA, a powerful chelating agent at alkaline pH, accounts for this result.Sodium bilirubinate, formed as a result of the cationic exchange, is soluble in water at pH 9.4. Its solubility in (SDS]n-butanol)» aqueous phase mixtures reached a limiting value at about 8]10~3 mol l~1 a concentration higher than that obtained with an alkaline solution of EDTA (6.41]10~3 mol l~1). Fig. 7 Cholesterol solubility in (SDS]n-butanol)»aqueous phase» MTBE microemulsions as a function of MTBE percentage for various (SDS]n-butanol)»buÜer mass ratios : 0.33, 0.54, 0.82, ()) (L) (|) (]) 1.22, (]) 1.86, 3.Cholesterol monohydrate solubility in (K) MTBE (0.418 mol l~1) is indicated by the horizontal line. Fig. 8 Calcium bilirubinate solubility in (SDS]n-butanol)»aqueous phase»MTBE microemulsions as a function of MTBE percentage for various (SDS]n-butanol)»buÜer mass ratios : 0.33, 0.54, ()) (L) (|) 0.82.Calcium bilirubinate solubility (6.41]10~3 mol l~1) in EDTA aqueous solution (pH 9.4) is indicated by the horizontal line. Therefore, the surfactant improves the EDTA solubilizing action. Dissolution of cholesterol-rich and calcium bilirubinate-rich discs. Investigations of the initial —ow To dissolve the two types of compressed discs, microemulsions with (SDS]n-butanol)»aqueous phase ratios of 0.33, 0.54 and 0.82 were selected. From a medical point of view, these microemulsions have two advantages: (1) their low viscosity allows for better homogenization with bile during their in vivo instillation and (2) they have a low (SDS]n-butanol) content and, in consequence, they are potentially less toxic.The dissolution kinetics of the discs in microemulsions were evaluated at 310^0.5 K in the dark using the rotating disc method.Initially, the surface of the disc is macroscopically homogeneous. During a kinetic experiment, the most soluble compound is preferentially dissolved leading to a surface essentially composed of the least soluble compound. Therefore, the relative surfaces of both solids in contact with the solvent are modi–ed as the solubilization process is occurring.As a result, the permeabilities towards both constituents, related to both the surface composition and the surface area, also vary. Moreover, the solvent was not renewed during the experiments. Consequently, its structure, composition and viscosity are continuously changing as the concentration gradient in the diÜuse layer and thereby the mutual diÜusion coefficient vary.Hence, the solubility of each compound of the disc, in the presence of the other, can be subsequently modi- –ed during the kinetic process. To be free from the simultaneous variations of all these parameters, the experimental results were considered only at very short times. Then, for each dissolution study11 we report the initial dissolution rates (Fig. 9). This procedure mimics the clinical case, in which the solvent instilled is continuously changing.12 When the disc is introduced into solvents, a very thin –lm consisting of a saturated layer appears at the solid»liquid interface where the interfacial reaction is occurring. A diÜerence of concentration takes place on both sides of (Cs1[Cs2 ) the –lm.Then, solute will —ow into the bulk solution across a diÜusion layer of thickness h, which is not stirred and directly in contact with the previous saturated –lm. The diÜerence of concentration within the diÜusion layer is (Cs2[Cb). From Berthoudœs law the —ow of matter going through a surface of section A as a function of the diÜerence of concentration between the solid surface and the bulk phase of volume V can be expressed13 as : G\dn/Adt\(Cs1[Cb)/R (1) 294 New J.Chem., 1999, 291»296Fig. 9 Initial —ows (mol cm~2 min~1) of cholesterol and of calcium bilirubinate from cholesterol-rich discs (panels A and B) and calcium bilirubinate-rich discs (panels C and D). The (SDS]n-butanol)»aqueous phase ratios were: 0.33, 0.54, 0.82.()) (L) (|) where R is the sum of the diÜusional resistance (h/D) and the interfacial resistance (1/P). D is the diÜusion coefficient of the diÜusing species and P is the permeability coefficient related to the solid surface. The following diÜerential equation is obtained: [(dCb/dt)](A/V R)]Cb[(A/V R)Cs1\0 (2) where and R are assumed to be constant. The solution of Cs1 eqn. (2) is written as : Cb\Cs1 [1[exp[([A/V R)t]] (3) Eqn.(3) represents the law describing the variation of solute concentration as a function of time at the beginning of the kinetics. During the dissolution process and R may vary Cs1 with the composition of the medium. R may also change with the surface composition. For the previously set out reasons, we have reported the initial dissolution rates versus the weight fraction of MTBE (Fig. 9). This plot allows us to de–ne the microemulsion composition that is optimal for dissolving the disc constituents. For very short dissolution times, one observes a linear variation of with time: Cb Cb\Cs1 [(A/V R)t] (4) which gives the same expression of the initial —ow. The variation of these —ows as a function of the (SDS]nbutanol) »aqueous phase ratio and the percentage of MTBE is displayed in Fig. 9. It appears that the initial —ows of cholesterol and calcium bilirubinate from cholesterol-rich discs (Fig. 9, A and B) increased to a large extent above 15»25% MTBE. For calcium bilirubinate-rich discs (Fig. 9, C and D), the increase starts abruptly around 20% MTBE. In both cases, these MTBE percentages correspond to the location of the bicontinuous medium in the pseudoternary diagram.Therefore, solvents that may be used to dissolve both the cholesterol-rich and the pigment-rich discs require a weight fraction of MTBE larger than 25%. Conclusions and physiological implications We have established in this study that the solubilizing power of (SDS]n-butanol)»aqueous phase»MTBE microemulsions depends on their composition and their structure, the most efficient exhibiting a bicontinuous structure.These microemulsions are able to dissolve cholesterol and calcium bilirubinate conglomerates such as those found in human biliary lithiasis. This –rst approach provides experimental results that do not take into account the potential toxicity of the solvents. Although animal and human studies with MTBE and alkaline EDTA14,15 have shown resistance of the gallbladder to these solvents, investigations have to be carried out to evaluate the toxic eÜect of the described microemulsions.Acknowledgements would like to thank Dr. B. Blaive for helpful We discussions. References 1 M. J. Allen, T. J. Borody, T. F. Bugliosy, G. R. May, N. F. LaRusso and J.L. Thistle, New Engl. J. Med., 1985, 312, 217. 2 J. L. Thistle, G. R. May, C. E. Bender, H. J. Williams, A. J. LeRoy, P. E. Nelson and C. J. Peine, New Engl. J. Med., 1989, 320, 633. 3 K. Y. Dai, J. C. Montet, X. M. Zhao, J. Amic and R. Choux, Gastroenterol. Clin. Biol., 1988, 12, 312. 4 K. Y. Dai, J. C. Montet, X. M. Zhao, J. Amic and A. M. Montet, J. Hepatol., 1989, 9, 301. 5 T. Takacs, J. Lonovics, F. X. Caroli-Bosc, A. M. Montet and J. C. Montet, Gastroenterol. Clin. Biol., 1997, 21, 655. 6 H. Igimi and M. C. Carey, J. L ipid Res., 1981, 22, 254. New J. Chem., 1999, 291»296 2957 U. Leuschner, H. Baumgaé rtel, M. Leuschner, H. Frenk and X. Klicic, Digestion, 1986, 34, 36. 8 P. Roé schlau, E. Bernt and W. Gruber, Z. Klin. Chem. Klin. Biochem., 1974, 12, 403. 9 T. M. Kathopoulis, PhD Thesis, Universiteç de Montpellier, France, 1979. 10 B. Lindman, N. Kamenka, T. M. Kathopoulis, B. Brun and P. G. Nilsson, J. Phys. Chem., 1980, 84 , 2485. 11 A. Ben Mouaz, PhD Thesis, Universiteç de Montpellier, France, 1996. 12 S. F. Zakko and A. F. Hofmann, Gastroenterology, 1990, 99, 1807. 13 K. H. Kwan, W. I. Higuchi, A. M. Molokhia and A. F. Hofmann, J. Pharm. Sci., 1977, 66, 1094. 14 E. Van Sonnenberg, S. Zakko, A. F. Hofmann, H. DœAgostino, H. Jinich, D. B. Hoyt, K. Miyai, G. Ramsby and A. Moossa, Gastroenterology, 1991, 100, 1718. 15 C. Y. Chen, K. K. Chang, N. H. Chow, T. C. Leow, T. C. Chou and X. Z. Lin, Digest. Dis. Sci., 1995, 40, 419. Paper 8/09228A 296 New J. Chem., 1999, 291»296

ISSN:1144-0546    

DOI:10.1039/a809228a

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Electrochemical oxidative polymerization of binuclear ìanilœ and ìsalenœ-type complexes and tetrahydro derivatives Pierre-Henri Aubert,a Pierre Audebert,*a Patrice Capdevielle,b Michel Maumy,*b and Maxime Rochea a L aboratoire de Chimie et dœElectrochimie de Moleç culaire, Universiteç Franche-Comteç , 25030 France Besanc” on, b L aboratoire de Chimie Organique (CNRS ESA 7084), ESPCI, 10 rue V auquelin, 75231 Paris cedex 05, France Received (in Montpellier, France) 17th November 1998, Accepted 11th January 1999 New kinds of modi–ed electrodes have been prepared on the basis of electroactive polymers containing bimetallic ì anil œ and ìsalenœ-type complexes.New tetrahydro ì anil œ ligands and complexes have been synthesized, giving birth to a new family of electroactive polymers.The polymers have been prepared potentiostatically and exhibit electrochemical properties analogous to simple polysalens. In the case of binuclear complexes both imino- and amino-linked monomers polymerize, but the polymers exhibit quite diÜerent charge transfer dynamics. Electropolymeç risation anodique de complexes binucleç aires de type ìanilœ et ìsalenœ et de leur deç riveç s tetrahydrogeç neç s.Un nouveau type dœeç lectrodes modi–eç es a eç teç eç laboreç sur la base de polyme` res eç lectroactifs issus de complexes bimeç talliques de type ì anil œ et ìsalenœ. Nous avons syntheç tiseç des ligands et complexes tetrahydro ì anil œ originaux donnant acce` s a` une nouvelle famille de polyme` res eç lectroactifs. Les complexes binucleç aires posseç dant des liaisons imine et amine polymeç risent eç galement, mais les polyme` res preç sentent des comportements tre` s diÜeç rents sur le plan du transfert eç lectronique. The electrochemical behaviour of metal complexes is arousing more and more interest, especially when new polymers can be built and used for various applications,1 including electrode modi–ers,2 electrocatalysis,3h10 sensors11,12 and nonlinear» and near-–eld»optics.13 The behaviour of metal-salens and analogous square planar complexes is attractive, given the ease with which these compounds electropolymerize into conducting polymers,14h20 exhibiting in some cases electrocatalytic properties.21 Also thiophene-salen compounds have been recently electropolymerized and their electrochemical properties have been reported.22 We describe here the electrochemical behaviour of bimetallic copper(II) and nickel(II) complexes derived from pentadentate binucleating ì anil œ and ìsalenœ SchiÜ base ligands.More precisely, these complexes are Robsonœs23h25 ì anil œ structures (Scheme 1, 1a and 1b) and Mazurekœs26,27 ìsalenœ series (structure 3). These ligands are derived from reaction of 2- aminophenol with 3-formyl-5-methyl-salicylaldehyde (ì anil œ) Scheme 1 New J.Chem., 1999, 297»301 297and from reaction of salicylaldehyde with 1,3-diaminopropan- 2-ol (ìsalenœ). Furthermore, we have synthesized and studied new binuclear complexes (Scheme 1, structures 2) derived from saturated diamino ligands prepared by sodium borohydride reduction of the anil SchiÜ bases.In these complexes, the electronic environment of the metals is square planar and mimics the classical salens ; however, the metal ions are close enough so that they can interact and have their electron-exchange properties modi–ed in comparison with the mononuclear complexes. Moreover, the para positions of the phenolates are free in each of the complexes, thus allowing oxidative polymerization (e.g., electrochemical) when this is a feasible process (e.g., when the oxidation potential is low enough).The monomeric substrates have been studied both in oxidation and reduction. Electropolymerization of the substrates has been performed for complexes 1a, 1b, 2a, 2b and 3. The electrochemical behaviour of the polymers has been investigated both in oxidation and reduction, and is reminiscent of that of salen complexes.Experimental Syntheses of the monomer complexes Precursor ligand 5-methylisophthalaldehyde di-2@-hydroxyanil has been prepared by reaction of ortho-aminophenol (H3L1) with 5-methylisophthalaldehyde in methanol and its derived metallic complexes 1a and 1b by (Cu2L1OMe) (Ni2L1OMe) subsequent reaction with a CuII or a NiII salt in basic medium according to Robson.23,24 The previously unknown tetrahydro ligand is prepared from the SchiÜ base H3[H4]L1 H3L1 by reduction of both CxN double bonds with sodium borohydride in methanol. The copper and nickel complexes 2a and 2b are obtained by reaction of with copper and H3[H4]L1 nickel salts in basic medium. 1,3-Bis(salicylideneamino)propan-2-ol and copper (H3L2)26 complex 3 have been prepared according to (Cu2L2OAc)27 Mazurekœs procedures from commercially available salicylaldehyde and 1,3-diaminopropan-2-ol.ligand. The red ligand (1.038 g, 3 mmol) is H3 [H4 ]L1 H3L1 dissolved in 20 ml methanol and excess sodium borohydride (0.114 g, 3 mmol) is slowly added (10 min) with stirring at ambient temperature. The solution rapidly decolourizes ; 30 min later, MeOH is distilled oÜ under reduced pressure.The residue is –rst dissolved in 10% phosphoric acid (15 ml) and then 10% sodium hydroxide is added to establish pH 7. The colourless crystalline product, which begins to separate, is extracted with ethyl acetate (150 ml), dried over magnesium sulfate and –ltered. It crystallizes after solvent evaporation, and is dried in vacuo (25 °C).Yield : 0.885 g (84%), mp 160 °C (decomp.) (diethyl oxide). Anal. calcd. % for C, C21H22N2O3: 71.98 ; H, 6.33 ; N, 7.99. Found C, 71.83 ; H, 6.36 ; N, 7.87. 1H NMR d) : 2.13 (s, 3H, 4.27 (s, 4H, 2 (d6-DMSO, CH3), CH2), 6.5»7.5 (m, 10H, arom.), 9.32 (s, 2H). 13C NMR (d6-DMSO, d) : 20.51, 43.11, 110.68, 113.54, 116.39, 119.71, 126.38, 127.57, 137.29, 144.49, 150.89.Copper complex 2a. ligand (Cu2 [H4 ]L1OMe) H3[H4]L1 (0.350 g, 1 mmol) is dissolved in hot methanol (15 ml) and treated with (0.400 g, 2 mmol) in 20 ml hot Cu(OAc)2 …H2O methanol. The solution is stirred for 2 h. A green solid separates, which is dried to a constant weight in vacuo (25 °C). Yield : 0.394 g (78%). Anal. calcd. % for C, C22H22N2O4Cu2: 52.27 ; H, 4.38 ; N, 5.54 ; Cu, 25.14.Found C, 52.15 ; H, 4.42 ; N, 5.52 ; Cu, 25.12. Nickel complex 2b. ligand (Ni2 [H4 ]L1OMe) H3[H4]L1 (0.350 g, 1 mmol) is dissolved in methanol (20 ml) and 0.8 ml of 5 M sodium methoxide in methanol (4 mmol) is added. Then, 0.498 g (2 mmol) dissolved in 10 ml Ni(OAc)2 … 4H2O methanol is added under stirring at room temperature. Methanol is distilled under reduced pressure.The residue is dissolved in chloroform (150 ml), –ltered and the solvent distilled under reduced pressure. An orange-brown complex is isolated, which is dried to a constant weight in vacuo (25 °C). Yield : 0.396 g (80%). Anal. calcd. % for C, 53.29 ; C22H22N2O4Ni2: H, 4.47 ; N, 5.65 ; Ni, 23.67. Found C, 53.21 ; H, 4.50 ; N, 5.61 ; Ni, 23.60. Electrochemical setup Polymer electrosyntheses and analytical experiments were performed in a three-compartment cell –tted with a saturated calomel reference (SCE), a glassy carbon electrode (diameter 1.2 mm) or a platinum electrode (diameter 1 mm) and a platinum counter electrode. The electrochemical apparatus was a home-made potentiostat28 (equipped with an Ohmic drop compensation system) –tted with a PAR 173 Universal programmer, a Nicolet digital oscilloscope and a Sefram 164 plotter.The solvent was spectroscopic grade acetonitrile (distilled over and stored on 3 molecular sieves) with CaH2 Aé 0.1 M tetraethylammonium perchlorate (Fluka puriss. recrystallized once in acetonitrile»diethyl oxide) as the supporting electrolyte. Concentration of the monomeric substrates was usually 2]10~3 M and the cells were —ushed with argon throughout the experiments.Ohmic compensation was used when necessary (i.e. when scan rates were over 1 V s~1). For –lm studies, two cells were used in series, one containing a saturated monomer solution for the –lm synthesis, another containing a clean electrolyte solution, into which the –lms were transferred for electrochemical studies.The electrode was carefully cleaned between each experiment. The –lms were prepared potentiostatically at various potentials ranging between 1.2 and 1.6 V as stated in the discussion. When a precise potential determination was desired, ferrocene was used as an internal standard; its potential was assumed to be ]405^5 mV vs. SCE; the ferrocene potential was checked after each set of experiments, but values in tables are given against SCE.For the diÜusion coefficient determination, the classical Cottrell law has been used [i\nFAc°D1@2/(pt)1@2], assuming n\1 per monomer unit in the –lm and a volumic mass of l\1.5 g cm~3 as usually assumed for most conducting polymers; therefore the concentration c° of electroactive moieties inside the –lms is expressed as c°\l/M, where M represents the molar mass of the monomer unit (taking into account one counter ion per moiety in the oxidized polymer). Results and discussion Behaviour of monomer complexes All complexes were studied by means of cyclic voltammetry prior to polymer synthesis, both in oxidation and reduction.The behaviour in reduction of these complexes is much less complicated than that of classical salens.Nickel-containing compounds exhibit a large, multielectronic, irreversible reduction peak around [1.8 V, whereas copper-containing compounds exhibit a monoelectronic reversible wave around [1.1 V with a slow charge transfer as attested by the peak separation of about 200 mV, attributable to the CuII/CuI reduction for one ion per cage (The number of electrons has been estimated by comparison to a ferrocene standard). The electrochemical reduction has been performed both in acetonitrile, a solvent able to complex the metals, and in dichloromethane, a non-complexing solvent, but the behaviours are quite similar.The voltammograms in Fig. 1 display the electrochemical responses of some complexes. Upon electrochemical oxidation at a potential in the 1.1»1.4 V range, all complexes (1, 2 and 3) exhibit an irreversible two- 298 New J.Chem., 1999, 297»301Fig. 1 Electrochemical reduction (left) and oxidation (right) of complexes 1 (v\1.2 V s~1), 2 (v\2 V s~1) and 3 (v\1.2 V s~1) in TEAP (0.1 M) performed on a 1.2 mm diameter carbon elec- CH3CN, trode. electron oxidation that leads to polymer formation on the electrodes (Fig. 2). As in several cases examined previously,17 the coupling between two entities occurs probably via a CR»CR (cation-radical»cation-radical) coupling mechanism at the para position of the phenolate as shown in Scheme 2. The synthesis of the polymer –lms may occur either at a controlled potential, or upon repetitive cycling in the solution. However, although the oxidation begins slightly before 1.0 V, a potential between 1.3 and 1.6 V is required to obtain clean –lms.Usually, optimum polymerization is obtained at potential values quite above that required for the formation of classical polysalens. The potential range where efficient polymerization takes place is also wider than in the case of the polymerization of simple salens or functionalized conducting polymers in general (e.g., polypyrrole).We have observed that mononuclear complexes 4a,b (Scheme 3) prepared according to the literature29 and dinuclear 5a,b tetrahydrosalen complexes prepared from the Fig. 2 Accumulative synthesis via electrochemical oxidation of complexes 2a (v\700 mV s~1), 2b (v\250 mV s~1) and 3 (v\280 mV s~1) in TEAP (0.1 M) performed on a 2 mm diameter Pt CH3CN, electrode.ligand30 cannot be transformed into electroactive H3[H4]L2 polymers under the same experimental conditions as their unsaturated analogous complexes. The peculiar behaviour of the ì tetrahydroanilœ series in particular should be outlined since the binuclear amine complexes 2a and 2b readily polymerize, like their SchiÜ-base analogues 1a and 1b.To our knowledge this is the –rst example of the polymerization of a copper or nickel amine complex. This made us wonder about the possibility –rst to oxidize (dehydrogenate) the amino complexes 2a and 2b into the imino compounds 1a and 1b (via intermediate CuIII and NiIII oxidation states), which would then be easily polymerized. Such a metal-catalyzed dehydrogenation of the HCwNH bond has been previously observed in some copper, nickel and cobalt para-substituted tetrahydrosalen complexes31 that are transformed into half-salen (dihydro) derivatives.But this hypothesis is ruled out because poly-2a and poly-2b exhibit very diÜerent electrochemical properties compared with their poly-1a and poly-1b analogues (see Discussion below), sug- Scheme 2 New J.Chem., 1999, 297»301 299Scheme 3 gesting that the monomer structure (imine or amine bonds) is indeed retained in the subsequent –nal polymer. Behaviour of polymers All polymers derived from compounds 1, 2 and 3 behave similarly to the polysalens upon electrochemical oxidation, with in many cases a better de–nition of the oxidation process, as evidenced by the voltammograms shown in Fig. 3. The potentials and the doping levels are listed in Table 1. As with the monomers, the peak potentials are more positive than in the case of simple salens. The peaks are better de–ned and the capacitive tailing is less important. The Fig. 3 Cyclic voltammograms of several polymers derived from complexes 1a (v\1.0 V s~1), 2a (v\1.0 V s~1) and 3 (v\1.5 V s~1). Polymerization charge mC cm~2.Qpolymerization\6.4 Table 1 Electrochemical data of complexes 1, 2 and 3 and their associated polymers Potential/mV vs. SCE Complex Monomer/(Epic)a Polymer/(E°) db D/cm2 s~1 1a [1105/955 805 0.88 8.1]10~10 1b [1690/890 775 0.67 1.3]10~9 2a [1150/1020 1035 0.87 4.5]10~11 2b [1730/990 990 0.92 1.93]10~12 3 [1140/1090 950 0.56 1.52]10~10 a Reduction/oxidation. b Calculated from eqn.(1). doping levels d have been calculated from the following classical equation:32 d\ 2Qredox Qtotal[Qredox (1) where is the total charge during electropolymerization : Qtotal is given by coulometric integration Qpolymerization]Qredox . Qtotal during electrosyntheses. is calculated by integration of Qredox the cyclic voltammograms as to avoid (Qanodic]Qcathodic)/2 problems linked to the basis line in the voltammograms.The polymerization charge the redox charge Qpolymerization , Qredox and therefore the total charge are de–ned by the Faradaic law as : nmonomer\ mpolymer Mmonomer \ Qpolymerization 2F \ Qredox dF \ Qtotal (2]d)F (2) Except for poly-3, d values are close to 1, which strongly suggests the exchange of one electron per monomer unit in the –lm; the diÜerence between the supposed theoretical doping level of 1 and the calculated experimental one is attributable to synthesis yields slightly lower than 100% for the polymer –lms.Even in the remaining case of poly-3, this explication is probably still valid.33 Therefore, even in polymerized binuclear complexes, it does not appear possible to oxidize each site more than once, despite the presence of two metal ions and the larger size of the cage.There is no signi–cant change in the doping levels when passing from copper to nickel central ions, in contrast with classical polysalens. The kinetics of electron transport in the polymers have been estimated on the basis of chronoamperometry experiments. As expected, there is no in—uence of the –lm thickness on the measurements.In fact, the electrochemical response of the –lm Fig. 4 Cottrell plot of poly-1b (top) and poly-2b (bottom) coated on a Pt electrode. The polymerization charge is 2.5 mC cm~2. 300 New J. Chem., 1999, 297»301exhibits a less important capacitive current and better de–ned peaks, the Cottrell plots (Fig. 4) display a behaviour closer to ideality, and the kinetics of charge transport are easily determinable from the experimental data. The diÜusion coefficients for the charge transfer are slightly below that measured in the polysalen case.It is striking that, with the exception of poly-3, which has a diÜerent structure, the polymers with amine links 2a and 2b exhibit a much slower (see Table 1) charge transport than the polymers with imine links (1a and 1b).Therefore, it is clear that some delocalization occurs through the imine bond in poly-1a and poly-1b, whereas poly-2a and poly-2b behave more like classical pendant-group polymers with a non-conducting backbone. In addition, while polysalens easily undergo degradation, the binuclear compounds appear resistant to high potential cycling (up to 1.7 V), without appreciable loss of electroactivity even after several hundreds of cycles.The polymers are reducible at relatively low potentials. However, only in the case of poly-2a is the reduction reversible, featuring the reduction of one copper ion per unit in the –lm according to the integration of the voltammograms. Since the]1 oxidation state of nickel is not stable, the Ni-containing –lms are reduced with two electrons per monomer unit.Poly-2b is not reducible, probably because the polymer is not conducting enough in this potential range. Conclusion In this paper we have describe the –rst oxidative electrochemical polymerization of binuclear copper and nickel complexes. Polymerization of amine-ligated copper and nickel (2a , 2b) complexes is also an unprecedented transformation.The results for these new binuclear monomers and polymers show that, although the general trends are comparable to polysalens, noticeable diÜerences are clearly discernible, especially an electrochemical behaviour closer to ideality and an improved stability towards high potentials. Notes and references 1 (a) G. Bidan, in Initiation la Chimie et Physico-chimie a` Groupe des Polyme` re, Strasbourg, Macromoleç culaire, Franc” ais 1993, vol. 9, ch. 5, pp. 137»254. (b) J. J. Andreç , in Initiation la a` Chimie et Physico-chimie Group des Macromoleç culaire, Franc” ais Polyme` res, Strasbourg, 1993, vol. 9, ch. 2, pp. 63»86. 2 (a) A. Deronzier and J.-C. Moutet, Coord. Chem. Rev., 1996, 147, 339. (b) F. Bedioui, J. Devynck and C.Bied-Charreton, Acc. Chem. Res., 1995, 28, 30. (c) P. Audebert, T rends Electrochem., 1994, 3, 459. (d) F. C. Anson, D. D. Montgomery, K. Shigehara and E. Tsuchida, J. Am. Chem. Soc., 1984, 106, 7991. 3 F. Bedioui, C. Bied-Charreton, J. Devynck, L. Gaillon and S. Guttierez-Granados, Stud. Surf. Sci. Catal., 1991, 66, 221. 4 F. Bedioui, C. Bied-Charreton, J. Devynck and S. Guttierez- Granados, New J.Chem., 1991, 15, 939. 5 J. P. Colin and J. P. Sauvage, J. Chem. Soc., Chem. Commun., 1987, 1075. 6 K. Rajeshwar, Y. Son and N. R. de Tacconi, J. Phys. Chem., 1993, 97, 1042. 7 G. Bidan, B. Divisia-Blohorn, M. Lapkowski, J. M. Kern and J. P. Sauvage, J. Am. Chem. Soc., 1992, 114, 5986. 8 R. J. Forster and J. G. Vos, Macromolecules, 1990, 23, 4372. 9 R. Cervini, R.J. Fleming and K. S. Murray, J. Mater. Chem., 1992, 2, 1115. 10 L. Coche, B. Ehui, D. Limosin and J. C. Moutet, J. Org. Chem., 1990, 55, 5905. 11 D. Bouaziz, B. Keita and L. Nadjo, J. Electroanal. Chem., 1988, 255, 303. 12 B. Fabre, G. Bidan and M. Lapkowski, J. Chem. Soc., Chem. Commun., 1994, 1509. 13 D. J. Williams, Non L inear Properties of Organic and Polymeric Materials, ACS Symposium Series 233, American Chemical Society, Washington, DC, 1983. 14 J.K. Blaho, K. A. Goldsby and L. A. Hoferkamp, Polyhedron, 1988, 8, 113. 15 L. A. Hoferkamp and K. A. Goldsby, Chem. Mater., 1989, 1, 348. 16 F. Bedioui, E. Labbe, S. Guttierez-Granados and J. Devynck, J. Electroanal. Chem., 1991, 301, 267. 17 P. Audebert, P. Capdevielle and M. Maumy, New J. Chem., 1991, 15, 235. 18 P. Audebert, P. Capdevielle and M. Maumy, New J. Chem., 1992, 16, 697. 19 P. Audebert, P. Hapiot, P. Capdevielle and M. Maumy, J. Electroanal. Chem., 1992, 338, 269. 20 P. Capdevielle, M. Maumy, P. Audebert and B. Plaza, New J. Chem., 1994, 18, 519. 21 C. E. Dahm and D. G. Peters, Anal. Chem., 1994, 66, 3117. 22 J. L. Reddinger and J. R. Reynolds, Chem. Mater., 1998, 10, 1236 and references therein. 23 (a) R. Robson, Aust. J. Chem., 1970, 23, 2217. (b) R. Robson, Inorg. Nucl. Chem. L ett., 1970, 6, 125. 24 B. F. Hoskins, R. Robson and G. A. Williams, Inorg. Chim. Acta, 1976, 6, 121 and references therein. 25 (a) H. Okawa, Bull. Chem. Soc. Jpn., 1970, 43, 3019. (b) H. Okawa, I. Ando and S. Kida, Bull. Chem. Soc. Jpn., 1974, 47, 3041. 26 W. Mazurek, K. J. Berry, K. S. Murray, M. J. OœConnor, M. R. Snow and A. G. Wedd, Inorg. Chem., 1982, 21, 3071. 27 W. Mazurek, B. J. Kennedy, K. S. Murray, M. J. OœConnor, J. R. Rodgers, M. R. Snow, A. G. Webb and P. R. Zwack, Inorg. Chem., 1985, 24, 3258. 28 D. Garreau and J. M. Saveç ant, J. Electroanal. Chem., 1972, 35, 309. 29 J. Csaszar, Acta Phys. Chem., 1984, 30, 61. 30 M. Maumy and P. Capdevielle, unpublished results. 31 A. Boé ttcher, H. Elias, E.-G. Jaé ger, H. Langfelderova, M. Mazur, L. Mué ller, H. Paulus, P. Pelikan, M. Rudolph and M. Valko, Inorg. Chem., 1993, 32, 4131. 32 For details concerning the calculation of the doping level, see ref. 19 and C. P. Andrieux and P. Audebert, J. Electroanal. Chem., 1990, 285, 163. 33 For poly-3, the d values were commonly measured to be between 0.5 and 0.6 and were averaged over several experiments, but one experiment displayed a d value equal to 0.76. Paper 8/08995G New J. Chem., 1999, 297»301 301

ISSN:1144-0546    

DOI:10.1039/a808995g

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Three-dimensional open-framework zinc phosphates with the structure-directing organic amines acting as ligands§ S. Neeraj, Srinivasan Natarajan and C. N. R. Rao* Chemistry and Physics of Materials Unit, Jawaharlal Nehru Centre for Advanced Scienti–c Research, Jakkur P.O., Bangalore 560 064, India. E-mail : cnrrao=jncasr.ac.in Received (in Montpellier, France) 18th November 1998, Accepted 8th January 1999 Two new three-dimensional zinc phosphates I, and II, [NH(CH2)2NH2(CH2)2NH3]2`[Zn5(PO4)4]2~, where the structure-directing organic amine acts as a ligand, have been [CN5H6]`[Zn2(PO4)(HPO4)]~, synthesized hydrothermally. Crystal data for I : a\27.071(2), b\5.215(1), c\17.920(1) b\130.3(1)° ; ”, U\1930.9(3) space group\Cc (no. 9) ; Z\4; M\811.01 ; g cm~3; MoKa and for II : ”3 ; Dcalc\2.789 a\8.089(1), b\12.771(1), c\10.067(1) b\105.3(1)° ; U\1000.3(2) space (no. 14) ; Z\4; ”, ”3 ; group\P21/n M\409.8 ; g cm~3; MoKa. Compound I is novel in the sense that it is dominated by the presence of Dcalc\2.713 a large number of three-coordinated oxygen atoms (25%), leading to the formation of in–nite ZnwOwZn chains. The presence of a distorted bipyramidal unit as well as a 2-membered ring in II is noteworthy.These ZnO3N2 structures are formed by the networking of and moieties, leading to the formation of ZnO4 , ZnO3N, PO4 ZnO3N2 three-dimensional structures possessing channels with I forming a 10-membered one-dimensional channel system and II forming two 8-membered channels. A variety of open-framework metal phosphates of diÜerent structures have been synthesized in the last few years by employing hydrothermal or sovothermal conditions in the presence of structure-directing organic amines.Amongst these, the zinc phosphates possessing 3-, 4-, 6-, 8-, 12-membered and other size rings and in–nite wZnwOwZnw chains constitute a relative large family.1h5 Surprisingly, 10-membered rings have not so far been reported in the zinc phosphates, although they are commonly found in other open-framework systems.Our continued research for metal phosphate open-framework structures with novel features has enabled us to isolate two new zinc phosphates having three-dimensional connectivity and possessing channels. The zinc phosphate I forms with a 10-membered one-dimensional channel, akin to that in aluminosilicates.6 More importantly, the structure-directing amine, in the middle of the channel, also satis–es the Zn coordination by acting as a ligand.Such a bifunctional role of the amine is also seen by us in the second zinc phosphate, II, where the amine directs the formation of two 8-membered channels (8- or 10-member refers to the number of tetrahedral atoms [Zn, P] forming the rings).In this paper, we describe the synthesis, structural and other characterizations of the two open-framework zinc phosphates where the structuredirecting amines act as ligands. In the context of such organicinorganic hybrid open-framework structures, the recent report of Halasyamani et al.7 of a zinc —uorophosphate where the nitrogen atoms of the 4,4@-bipyridyl bound to Zn form a pillared-layer network is noteworthy. Experimental Synthesis and initial characterization Compound I was synthesized by the following procedure: 0.407 g of ZnO was dispersed in 9 ml of water, and 0.365 g of § Supplementary material available : Atomic cordinates and isotropic displacement parameters for I and II.For direct electronic access see http ://www.rsc.org/suppdata/nj/1999/303/, otherwise available from BLDSC (No.SUP 57483, 3 pp.) or the RSC Library. See Instructions for Authors, 1999, Issue 1 (http :/www.rsc.org/njc). HCl and 0.98 g of 85 wt% of added to the mixture, H3PO4 which was then stirred for 10 min. To this mixture 0.516 g of diethylenetriamine (DETA) was added and the mixture was homogenized, transferred into a Parr pressure bomb and heated initially at 150 °C for 5 days.At the end of the initial heating, half the mixture was removed from the container and 2 ml of were added to the remaining mixture, which was H2O heated for a further period of 4 days at 150 °C. This resulted in the formation of needle- and plate-shaped crystals. The –nal composition of the mixture was ZnO : 2 H3PO4 : 2 HCl : DETA : 100 The plate-shaped crystals were separ- H2O.ated easily from the mixture as they are the predominant product of the synthesis and all further characterizations and studies were carried out on them. Compound II was synthesized as follows : 0.407 g of ZnO was dispersed in 5 ml of water, and 1.145 g of 85 wt% of added to the mixture under continuous stirring. 1,3- H3PO4 Diaminoguanidinemonohydrochloride(DAG,2.98 g)wasadded to the above and stirring continued until the mixture became homogeneous.The composition of the mixture was ZnO : 2.3 DAG : 55 The mixture was transferred into a H3PO4 : 4.8 H2O. te—on-lined stainless steel autoclave (Parr, USA) and heated at 110 °Cfor4days.Theproductcontainedlargequantitiesofrod-like single crystals. The initial characterizations of the samples were carried out by powder X-ray diÜraction (XRD) and thermogravimetric analysis (TGA).The XRD pattern of the crushed single crystals of I (plates) and II indicate the products to be new materials ; the individual patterns are entirely consistent with the structures determined by single crystal X-ray diÜraction (simulated pattern from the atomic coordinates derived from the single crystal study).Single crystal structure determination A suitable single crystal of each compound (plate for I and rod for II) was carefully selected under a polarizing microscope and glued to a thin glass –ber with cyanoacrylate (superglue) adhesive. Crystal structure determination by X-ray diÜraction was performed on a Siemens SMART-CCD diffractometer equipped with a normal focus, 2.4 kW sealed tube New J.Chem., 1999, 303»308 303Table 1 Crystal data and structure re–nement parameters for I, and II, [NH(CH2)2NH2(CH2)2NH3]2`[Zn5(PO4)4]2~ [CN5H6]`- [Zn2(PO4)(HPO4)]~ I II Empirical formula Zn5P4O16C4N3H14 Zn2P2O8C1N5H7 Crystal system Monoclinic Monoclinic Space group Cc (no. 9) P21/n (no. 14) Crystal size/mm 0.04]0.1]0.16 0.06]0.08]0.14 a/” 27.071(2) 8.089(1) b/” 5.215(1) 12.771(1) c/” 17.920(1) 10.067(1) a/° 90.0 90.0 b/° 130.3(1) 105.3(1) c/° 90.0 90.0 Volume/”3 1930.8(3) 1003.2(2) Z 4 4 Formula mass 811.01 409.80 qcalc/g cm~3 2.789 2.713 k (MoKa)/” 0.71073 0.71073 l/mm~1 6.543 5.151 h range/° 1.97»23.28 2.63»23.28 Total data collected 3775 4109 Index ranges [29OhO29, [5OkO5, [19OlO17 [8OhO5, [14OkO13, [11OlO10 Unique data 2089 1436 Observed data [r[2r(I)] 1994 1193 Re–nement method Full-matrix least-squares on F2 Full-matrix least-squares on F2 R indices [I[2r(I)] R1\0.0389 ; wR2\0.0901 R1\0.0422 ; wR2\0.993 R indices (all data) R1\0.0433 ; wR2\0.1168 R1\0.0554 ; wR2\0.1095 Goodness of –t (S) 1.14 1.08 No.of variables 291 163 Largest diÜerence map peak and hole/e ”~3 1.159 and [1.002 0.769 and [0.813 X-ray source (MoKa radiation, k\0.71073 operating at ”) 40 kV and 40 mA.A hemisphere of intensity data was collected at room temperature in 1321 frames with x scans (width of 0.30° and exposure time of 20 s per frame). The –nal unit-cell constants were determined by a least squares –t of 3568 re—ections for I and 2148 re—ections for II in the range 3°\2h\46.5°.Pertinent experimental details for the structure determination are presented in Table 1. The structure was solved by direct methods using SHELXS- 868 and diÜerence Fourier syntheses. All the hydrogen positions, in I and II, were initially located in the diÜerence Fourier maps for both the compounds. For the –nal re–nement, the hydrogen atoms were placed geometrically and held in the riding mode.No absorption correction was applied for either I or II. The last cycles of re–nement included atomic positions for all the atoms, anisotropic thermal parameters for all non-hydrogen atoms and isotropic thermal parameters for all the hydrogen atoms. Full-matrix least-squares structure re–nement against oF2 o was carried out using the SHEXLXTL-PLUS9 suite of programs.Details of the –nal re–nements are given in Table 1. The selected bond distances and bond angles are given in Table 2 for I and Table 3 for II. CCDC reference number 440/093. See http ://www.rsc.org/ suppdata/nj/1999/303/ for crystallographic –les in .cif format. Results Structure of I, [NH(CH2)2NH2(CH2)2NH3 ]2ë- [Zn5(PO4)4 ]2ó The asymmetric unit is represented on Fig. 1(a).The structure of I consists of a network of and moieties, ZnO4, PO4 ZnO3N connected by ZnwOwP bonds. The framework has the formula and charge neutrality is achieved by [Zn5(PO4)4]2~ the incorporation of the doubly protonated amine molecule. The Zn atoms are all tetrahedrally coordinated and connected to P atoms via oxygen links and to each other via threecoordinated oxygen atoms (bonding two Zn atoms and one P atom).Four of the 16 oxygen atoms in the asymmetric unit are three-coordinated. This is an unusually large percentage (25%) of such units in an open-framework material. The Zn atoms in I make three ZnwOwP bonds with four P neighbours, with a spread of ZnwOwP angles (ave. 127.3°) as given in Table 2. The fourth connections needed for a tetrahedral linkage in the case of Zn(1) to Zn(4) are obtained through ZnwOwZn linkages (ave. 115.7°), and via the terminal N atom of the amine in the case of Zn(5). Thus, there are in–nite one-dimensional ZnwOwZn chains in this structure. The average values of ZnwO bond lengths Mdav[Zn(1)wO\1.947 ”] ; dav[Zn(2)wO\1.956 ”] ; dav[Zn(3)wO\1.957 ”] ; dav[Zn(4)wO\1.958 ”] ; dav[Zn(5) wO\1.968 and OwZnwO bond angles [OwZn(1)w ”]N O\109.1° ; OwZn(2)wO\109.1° ; OwZn(3)wO\109.4° ; OwZn(4)wO\108.1° ; OwZn(5)wO\109.4°] indicate that the environment around the Zn atoms is tetrahedral and the values are in good agreement with those reported earlier.1h5,7 The four distinct phosphorus atoms are linked to Zn atoms via the oxygens.The PwO bond lengths are in the range 1.517»1.582 and the OwPwO bond angles are in the range ” 104.8»113.9°.The longest PwO distances, however, involve the three-coordinated oxygen atom. These values agree well with those reported in the literature. The polyhedral connectivity between the and ZnO4 PO4 units leads to the formation of in–nite ZnwOwZn chains and 3-membered rings via the three-coordinated oxygen atoms. The structure also posseses 4-membered units.It Zn2P2O4 should be noted that 4-membered rings appear to be the basic building blocks of many of the open-framework structures such as aluminophosphates and tin phosphates.10 The 3- and 4-membered rings in I share their edges, forming a onedimensional chain (Fig. 2). The individual ribbons (columns of one-dimensional chains) are joined together by tetra- ZnO3N hedra, giving rise to a 10-membered channel system along the a axis.Such a linkage between one-dimensional chains by the moiety requires the amine molecule to be in the ZnO3N middle of the 10-membered channel (10.36]4.62 oxygen- ”; to-oxygen contact distance excluding the van der Waals radii). 304 New J. Chem., 1999, 303»308Table 2 Selected bond lengths and angles (°) for I, (”) [NH(CH2)2NH2(CH2)2NH3]2`[Zn5(PO4)4]2~ Zn(1)wO(1) 1.950(8) Zn(5)wO(15) 1.922(9) Zn(1)wO(2) 1.919(9) Zn(5)wN(3) 2.022(10) Zn(1)wO(3) 1.997(8) P(1)wO(6) 1.518(9) Zn(1)wO(4) 1.922(9) P(1)wO(12) 1.522(9) Zn(2)wO(1)1 2.021(8) P(1)wO(14)4 1.536(9) Zn(2)wO(5) 1.938(9) P(1)wO(9) 1.561(9) Zn(2)wO(6) 1.931(9) P(2)wO(11) 1.510(10) Zn(2)wO(7) 1.934(10) P(2)wO(5) 1.518(9) Zn(3)wO(3) 1.953(8) P(2)wO(16) 1.535(10) Zn(3)wO(8) 1.956(8) P(2)wO(1)5 1.580(9) Zn(3)wO(9) 1.986(8) P(3)wO(15)6 1.525(10) Zn(3)wO(10)2 1.933(8) P(3)wO(7)7 1.526(10) Zn(4)wO(8) 2.020(9) P(3)wO(2) 1.527(8) Zn(4)wO(9)3 1.987(8) P(3)wO(3)3 1.582(9) Zn(4)wO(11) 1.913(9) P(4)wO(13) 1.517(10) Zn(4)wO(12) 1.911(9) P(4)wO(4) 1.521(10) Zn(5)wO(13) 1.977(9) P(4)wO(10) 1.554(8) Zn(5)wO(14) 1.952(9) P(4)wO(8) 1.580(9) O(2)wZn(1)wO(4) 111.3(4) O(15)wZn(5)wN(3) 105.5(4) O(2)wZn(1)wO(1) 119.7(3) O(14)wZn(5)wN(3) 119.8(4) O(4)wZn(1)wO(1) 111.5(4) O(13)wZn(5)wN(3) 112.1(4) O(2)wZn(1)wO(3) 106.6(4) O(6)wP(1)wO(12) 113.9(5) O(4)wZn(1)wO(3) 100.1(4) O(6)wP(1)wO(14)4 108.1(5) O(1)wZn(1)wO(3) 105.4(3) O(12)wP(1)wO(14)4 110.2(5) O(6)wZn(2)wO(7) 101.4(6) O(1)wP(1)wO(9) 108.5(5) O(6)wZn(2)wO(5) 120.5(4) O(12)wP(1)wO(9) 108.6(5) O(7)wZn(2)wO(5) 118.0(4) O(14)wP(1)wO(9) 107.5(5) O(6)wZn(2)wO(1)1 114.9(4) O(11)wP(2)wO(5) 112.2(5) O(7)wZn(2)wO(1)1 95.9(3) O(11)wP(2)wO(16) 108.1(6) O(5)wZn(2)wO(1)1 104.3(4) O(5)wP(2)wO(16) 109.9(5) O(10)2wZn(3)wO(3) 121.0(4) O(11)wP(2)wO(1)5 104.8(5) O(10)2wZn(3)wO(8) 106.1(3) O(5)wP(2)wO(1)5 109.6(5) O(3)wZn(3)wO(8) 102.5(4) O(16)wP(2)wO(1)5 112.2(5) O(10)2wZn(3)wO(9) 97.2(3) O(15)6wP(3)wO(7)7 109.3(6) O(3)wZn(3)wO(9) 118.2(4) O(15)6wP(3)wO(2) 111.1(5) O(8)wZn(3)wO(9) 111.5(4) O(7)7wP(3)wO(2) 112.9(5) O(12)wZn(4)wO(11) 124.1(4) O(15)6wP(3)wO(3)3 108.4(5) O(12)wZn(4)wO(9)3 111.4(4) O(7)7wP(3)wO(3)3 106.9(5) O(11)wZn(4)wO(9)3 107.2(4) O(2)wP(3)wO(3)3 108.0(5) O(12)wZn(4)wO(8) 102.6(4) O(13)wP(4)wO(4) 112.2(6) O(11)wZn(4)wO(8) 104.9(4) O(13)wP(4)wO(10) 107.3(5) O(9)3wZn(4)wO(8) 104.8(3) O(4)wP(4)wO(10) 111.4(5) O(15)wZn(5)wO(14) 114.0(4) O(13)wP(4)wO(8) 109.7(5) O(15)wZn(5)wO(13) 113.0(4) O(4)wP(4)wO(8) 108.9(5) O(14)wZn(5)wO(13) 92.2(4) O(10)wP(4)wO(8) 107.1(5) Organic Moiety N(2)wC(3) 1.46(2) C(3)wN(2)wC(2) 110.3(12) N(2)wC(2) 1.47(2) C(4)wN(3)wZn(5) 119.8(8) N(1)wC(1) 1.47(2) N(2)wC(2)wC(1) 111.2(12) N(3)wC(4) 1.52(2) N(1)wC(1)wC(2) 112.2(11) C(2)wC(1) 1.51(2) N(2)wC(3)wC(4) 111.2(13) C(3)wC(4) 1.48(2) C(3)wC(4)wN(3) 111.5(12) Symmetry transformations used to generate equivalent atoms: 1 x[1/2, [y[7/2, z[1/2. 2 x, y[1, z. 3 x, y]1, z. 4 x, [y[3, z[1/2. 5 x[1/2, [y[5/2, z[1/2. 6 x, [y[2, z[1/2. 7 x]1/2, [y[5/2, z]1/2. 8 x]1/2, [y[7/2, z]1/2. 9 x, [y[3, z]1/2. 10 x, [y[2, z]1/2. Structure of II, [CN5H6 ]ë[Zn2(PO4)(HPO4) ]ó Fig. 1(b) represents the asymmetric unit.The structure is builtup from the vertex-linkage of and PO4, ZnO4 ZnO3N2 moieties. The framework is anionic and has the formula Charge neutrality is achieved by the [Zn2(PO4)(HPO4)]~. incorporation of the protonated amine molecule, with one protonated 1,3-diaminoguanidine, per framework [CN5H6]`, formula unit. Of the eight oxygens in the asymmetric unit, one is three-coordinated.There are two crystallographically distinct Zn as well as P atoms. Of the two Zn atoms in the asymmetric unit, Zn(1) is bound with two nitrogens and three oxygens, forming a dis- Table 3 Selected bond lengths and angles (°) for II, (”) [CN5H6]`- [Zn2(PO4)(HPO4)]~ Zn(1)wO(1) 1.954(5) P(1)wO(1) 1.514(5) Zn(1)wO(2) 2.004(4) P(1)wO(5)2 1.521(5) Zn(1)wN(3) 2.110(6) P(1)wO(4) 1.543(5) Zn(1)wN(5) 2.115(6) P(1)wO(7) 1.573(5) Zn(1)wO(2)1 2.126(4) P(2)wO(3)3 1.526(5) Zn(2)wO(3) 1.929(5) P(2)wO(8) 1.526(5) Zn(2)wO(4) 1.955(5) P(2)wO(2) 1.539(5) Zn(2)wO(5) 1.957(5) P(2)wO(6)4 1.553(5) Zn(2)wO(6) 1.968(5) O(1)wZn(1)wO(2) 111.2(2) O(4)wZn(2)wO(6) 101.6(2) O(1)wZn(1)wN(3) 101.0(2) O(5)wZn(2)wO(6) 94.6(2) O(2)wZn(1)wN(3) 100.5(2) O(1)wP(1)wO(5)2 112.1(3) O(1)wZn(1)wN(5) 127.4(2) O(1)wP(1)wO(4) 109.3(3) O(2)wZn(1)wN(5) 121.1(2) O(5)2wP(1)wO(4) 110.9(3) N(3)wZn(1)wN(5) 75.9(2) O(1)wP(1)wO(7) 107.1(3) O(1)wZn(1)wO(2)1 89.7(2) O(5)2wP(1)wO(7) 111.6(3) O(2)wZn(1)wO(2)1 81.2(2) O(4)wP(1)wO(7) 105.7(3) N(3)wZn(1)wO(2)1 167.6(2) O(3)3wP(2)wO(8) 110.9(3) N(5)wZn(1)wO(2)1 92.7(2) O(3)3wP(2)wO(2) 108.0(3) O(3)wZn(2)wO(4) 111.7(2) O(8)wP(2)wO(2) 109.9(3) O(3)wZn(2)wO(5) 113.2(2) O(3)3wP(2)wO(6) 109.5(3) O(4)wZn(2)wO(5) 107.5(2) O(8)wP(2)wO(6)4 109.7(3) O(3)wZn(2)wO(6) 125.9(2) O(2)wP(2)wO(6)4 108.7(3) Organic Moiety N(1)wC(1) 1.330(9) C(1)wN(2)wN(5) 118.6(6) N(2)wC(1) 1.359(9) C(1)wN(3)wN(4) 115.1(6) N(2)wN(5) 1.401(8) C(1)wN(3)wZn(1) 117.4(5) N(3)wC(1) 1.318(9) N(4)wN(3)wZn(1) 127.3(4) N(3)wN(4) 1.432(8) N(2)wN(5)wZn(1) 112.4(4) N(3)wC(1)wN(1) 126.7(7) N(3)wC(1)wN(2) 105.5(6) N(1)wC(1)wN(2) 117.7(6) Symmetry transformations used to generate equivalent atoms: 1 [x, [y]1, [z. 2 [x, [y]2, [z. 3 [x[1/2, y[1/2, [z[1/2. 4 x]1/2, [y]3/2, z[1/2. 5 [x[1/2, y]1/2, [z[1/2. 6 x[1/2, [y]3/2, z]1/2. torted trigonal bipyramidal unit, while Zn(2) is tetra- ZnO3N2 hedrally coordinated to four oxygens.The observation of the distorted trigonal bipyramidal environment for the Zn atom is unusual and only a few examples are known for such coordination of the Zn atom, for example in The Zn(acac)2 .11 average ZnwO bond distances Mdav[Zn(1)wO/N\2.062 ”] ; and OwZnwO bond angles dav[Zn(2)wO\1.952 ”]N [OwZn(1)wO/N\106.8° ; OwZn(2)wO\109.0°] are consistent with the coordination environment of the Zn atoms.Both P(1) and P(2) are connected by three oxygens to Zn atoms and the last neighbour is a terminal oxygen [P(1)wO(7) and P(2)wO(8)]. The average bond distances involving P atoms [P(1)wO\1.538 P(2)wO\1.536 and bond ”; ”] angles [OwPwO\109.5°] suggest a regular P tetrahedron, unlike Zn. Bond valence sum calculations12 indicate that the oxygen O(7) is protonated and O(8) has a double-bond character.As usual, the longest PwO [P(1)wO(7)\1.573 dis- ”] tance is the PwO… … …H one. The polyhedral connectivity between the ZnO4 , ZnO3N2 and units leads to the formation of the observed three- PO4 dimensional structure. Linking of two units results in ZnO3N2 the formation of a 2-membered ring [Fig. 1(b)], the N Zn2O2 atoms being part of the amine molecule.The 2-membered ring is connected to P atoms via the ZnwOwP links. The formation of a 2-membered ring in an open-framework solid is not common and to our knowledge, there is only one such report in the literature.13 The and units form two inter- ZnO4 PO4 connected 4-membered rings [Fig. 3(b)], which are threedimensionally connected through the units involving ZnO3N2 the 2-membered ZnwOwZn linkage.The linkages between these building blocks lead to the formation of two 8- New J. Chem., 1999, 303»308 305Fig. 1 (a) Asymmetric unit of I, (b) Asymmetric unit of II, [NH(CH2)2NH2(CH2)2NH3]2`[Zn5(PO4)4]2~. [CN5H6]`[Zn2(PO4)(HPO4)]~. (Thermal ellipsoids are shown at 50% probability). membered channels running along the [100] (7.67]5.64 ”) and [101] (7.30]5.81 directions (Fig. 3). The amine mol- ”) ecule protrudes from the Zn centres into the channels along the [101] direction. Discussion Two zinc phosphates, [NH(CH2)2NH2(CH2)2NH3]2`- I, and II, [Zn5(PO4)4]2~ [CN5H6]`[Zn2(PO4)(HPO4)]~ have been synthesized by hydrothermal methods in the presence of structure-directing agents. As with kinetically controlled processes, there is no relation between the starting stoichiometry and the –nal compostion of the product.Compound I has been prepared in the presence of HCl and it appears that the role of Cl~ is similar to that of F~ ions in some of the phosphate-based open-framework materials reported in the literature.14,15 We are presently evaluating the role of Cl~, and other similar ions in determining the (SO4)2~ structure of the products.The structures of I and II are both built up from the regular the unusual ZnO4, PO4 , ZnO3N tetrahedra and the distorted trigonal bipyramidal ZnO3N2 , leading to the three-dimensional connectivity. We may recall that most of the known zinc phosphates contain and ZnO4 tetrahedra only, although distinct diÜerences exist PO4 between the structures. The two structures described here are unusual in that they form ZnwOwZn linkages although the Zn : P ratio is 1 in II and 1.25 in I.The presence of more Zn than P in I may be responsible for the in–nite one-dimensional wZnwOwZnw linkages. In II, these linkages may be attri- 306 New J. Chem., 1999, 303»308Fig. 2 (a) Structure of I, [NH(CH2)2NH2(CH2)2NH3]2`- showing the one-dimensional 10-membered channel [Zn5(PO4)4]2~, system along the a axis.(b) Figure showing the connectivity between the chain units and unit leading to the formation of the 10- ZnO3N membered ring. buted to the presence of the rare trigonal bipyramidal coordination of the Zn atom. No PwOwP type linkages are seen in either of the materials. The wZnwOwZnw linkages are always accompanied by three-coordinated bridging oxygen atoms and the third coordination is to a phosphorus.The trigonal coordination of the oxygen in the wZnwOwZnw bridge is apparently an electrostatic valence requirement of bridging oxygen atoms. There are other examples where such trigonal coordination has been observed.3,16,17 The trigonal and tetrahedral coordinations of oxygen bridges are generally common when divalent tetrahedral atoms are involved and we would therefore expect such a feature in the zinc phosphates to yield novel open-framework topologies that have no analogues in aluminosilicates and aluminophosphates.Another interesting aspect of the Zn phosphates examined here relates to the protonation of the framework oxygen atoms. In II, only one oxygen atom is protonated in spite of the fact that the Zn:P ratio is 1, and no oxygens are protonated in I.This may be due to the presence of ZnwN linkages present in these materials. In the zinc phosphates reported in the literature, some of the oxygen atoms are invariably protonated to compensate the charge imbalance. It is also hypothesized that the oxygens are protonated due to the difficulty in Fig. 3 Structure of II, showing the [CN5H6]`[Zn2(PO4)(HPO4)]~, 8-membered channels (a) along [100] and (b) [101] directions. Note that the amine molecules protrude into the channels along the [101] direction. Hydrogens of the amine molecule are not shown for clarity. packing enough bulky organic cations into the extra framework pores to achieve charge balance.18 In the case of the aluminosilicate materials, the Si :Al ratio is variable (for the same framework topology) and aÜects the charge-balance requirement; the zinc phosphates appear to have a precisely de–ned Zn:P ratio for a particular topology.In the present study, the nitrogen atoms of the amine molecule covalently binding with Zn centres possibly create a situation where the bulky organic cation can be accommodated within the pores/ channels and achieve charge balance.This type of observation is unique as in many of the channel structures the structuredirecting organic amine molecules sit in channels and cavities and interact with the framework through hydrogen bonds only. The ìopennessœ of a structure is de–ned in terms of the tetrahedral atom density6 (framework density, FD) de–ned as the number of tetrahedral (T) atoms per 1000 In the ”3.present materials, the number of T atoms per 1000 (here ”3 Zn and P) is 18.6 for I and 16.0 for II. These values are in the middle of the range of FD values observed in aluminosilicate zeolites,6 where the presence of channels is common. Furthermore, the position of the amine is such that the linear chain DETA (compound I) forms a well-de–ned 10-membered New J.Chem., 1999, 303»308 307Table 4 Important hydrogen bond interactions in I and II I O(11)… … …H(1) 2.235(1) O(11)… … …H(1)wN(1) 136.5(1) O(16)… … …H(1) 2.185(1) O(16)… … …H(1)wN(1) 154.9(1) O(14)… … …H(2) 1.925(1) O(14)… … …H(2)wN(1) 156.0(1) O(10)… … …H(3) 1.968(1) O(10)… … …H(3)wN(1) 175.8(1) O(5)… … …H(8) 2.362(1) O(5)… … …H(8)wN(2) 140.7(1) O(16)… … …H(9) 1.930(1) O(16)… … …H(9)wN(2) 172.9(1) O(4)… … …H(12) 2.467(1) O(4)… … …H(12)wC(4) 161.9(1) II O(1)… … …H(1) 2.176(1) O(1)… … …H(1)wN(1) 137.7(1) O(7)… … …H(2) 2.447(1) O(7)… … …H(2)wN(1) 133.5(1) O(6)… … …H(3) 1.991(1) O(6)… … …H(3)wN(2) 170.6(1) channel system. In II, however, the distorted trigonal pyramidal unit links with another unit to form ZnO3N2 ZnO3N2 two wZnwOwZnw linkages (2-membered rings) that connect the double 4-membered rings, creating distortion of the 8-membered channels.Thermogravimetric analysis (TGA) of I and II was carried out in air from room temperature to 600 °C. The results show only one mass loss for I in the region 350»440 °C. The mass loss of 16.5% corresponds to the loss of the amine molecule from the structure (calcd 13%) and some adsorbed water.In the case of II, the mass loss of 22.6% occurring in the temperature range 350»450 °C corresponds to the loss of the amine molecule (calcd 21.5%). In both the cases the loss of the amine molecule resulted in the collapse of the framework structure, leading to the formation of largely amorphous weakly diÜracting materials (XRD) that corresponds to dense zinc phosphate phases consistent with their structures. In addition to the coordination of the amine with the metal atoms in the framework, both the structures show dominant hydrogen bond interactions between the amine and the framework (Table 4).In I, since the amine has a linear chain, hydrogen bonding is prominent. The strongest hydrogen bonding in I is between the hydrogens attached to the nitrogens N(1) and N(2) and the framework oxygens. The following hydrogen bond parameters reveal this : O(10)… … …H(3)\1.986 and ” O(10)… … …H(3)wN(1)\175.8° ; O(16)… … …H(9)\1.930 and ” O(16)… … …H(9)wN(2)\172.9°. Here, O(16) is one of the terminal double-bonded oxygen atoms attached to P.There is only one strong hydrogen bond in II as given by O(6)… … …H(3)\1.991 and O(6)… … …H(3)wN(2)\170.6°.” Compared to other open-framework zinc phosphates, I and II are also novel because of the three-dimensional connectivity arising from the linkage between the ZnO4 , ZnO3N, PO4 tetrahedra and distorted trigonal bipyramidal In ZnO3N2 . the work of Halasyamani et al.,7 where similar bonding of the amine with the Zn centres is reported, the linkages between the and tetrahedra lead to the formation of ZnO3N PO3 F layers.Covalent bonding between the nitrogen of the amine and the Zn atoms and the position of the amine molecule in the organic-inorganic hybrid structure described by these workers are reminiscent of pillared materials. Conclusions The synthesis and structure of a new family of threedimensional open-framework zinc phosphates, where the amine molecule acts not only as a structure-directing agent but also ful–ls the coordination sphere of Zn centres, are described.The noteworthy structural features of these materials are a 10-membered channel system in I as well as a distorted bipyramidal unit and a 2-membered ring ZnO3N2 along with two 8-membered channels in II. It would be of interest to explore other structures where the structuredirecting amine plays a dual role as in I and II.References 1 W. T. A. Harrison, T. E. Gier, T. E. Martin and G. D. Stucky, J. Mater. Chem., 1992, 2, 175; W. T. A. Harrison, T. M. NenoÜ, M. M. Eddy, T. E. Martin and G. D. Stucky, J. Mater. Chem., 1992, 2, 1127. 2 T. Song, M. B. Hursthouse, J. Chen, J. Xu, K. M. A.Malik, R. H. Jones, R. Xu and J. M. Thomas, Adv. Mater., 1994, 6, 679. 3 P. Feng, X. Bu and G. D. Stucky, Angew. Chem., Int. Ed. Engl., 1995, 34, 1745; J. Solid State Chem., 1996, 125, 243. 4 W. T. A. Harrison and L. Hannooman, Angew. Chem., Int. Ed. Engl., 1997, 36, 640; J. Solid State Chem., 1997, 131, 363. 5 W. T. A. Harrison and M. F. L. Phillips, Chem. Commun., 1996, 2771. 6 Atlas of Zeolite Structure T ypes, ed. W. H. Meier, D. H. Olson and Ch. Baerlocher, Elsevier, Boston, USA, 1996. 7 P. S. Halasyamani, M. J. Drewitt and D. OœHare, Chem. Commun., 1997, 867. 8 G. M. Sheldrick, SHEL XS-86 Program for Crystal Structure Determination, University of Goé ttingen, 1986; Acta. Crystallogr., Sect. A, 1990, 35, 467. 9 G. M. Sheldrick, SHEL XS-93 Program for Crystal Structure Solution and Re–nement, University of Goé ttingen, 1993. 10 W. T. A. Harrison, T. M. NenoÜ, T. E. Gier and G. D. Stucky, J. Solid State Chem., 1994, 113, 168; S. Ayyappan, X. Bu, A. K. Cheetham, S. Natarajan and C. N. R. Rao, Chem. Commun., 1998, 2181. 11 A. F. Wells, Structural Inorganic Chemistry, 5th edn., Oxford Science Publications, Oxford, 1986. 12 I. D. Brown and D. Aldermatt, Acta Crystallogr., Sect. B, 1984, 41, 244. 13 D. Whang, N. H. Hur and K. Kim, Inorg. Chem., 1995, 34, 3363. 14 A. M. Chippindale, S. Natarajan, J. M. Thomas and R. H. Jones, J. Solid State Chem., 1994, 111, 18. 15 J. L. Guth, H. Kessler and R. Wey, Stud. Surf. Sci. Catal., 1986, 28, 121; G. J. Feç rey, J. Fluorine Chem., 1995, 72, 187. 16 W. T. A. Harrison, T. E. Gier and G. D. Stucky, Angew. Chem., Int. Ed. Engl., 1993, 32, 1745. 17 T. You, J. Xu, Y. Zhao, Y. Yue, Y. Xu, R. Xu, N. Hu, G. Wei and H. Jia, J. Chem. Soc., Chem. Commun., 1994, 1171. 18 W. T. A. Harrison and M. L. F. Phillips, Chem. Mater., 1997, 9, 1837. Paper 8/09030K 308 New J. Chem., 1999, 303»308

ISSN:1144-0546    

DOI:10.1039/a809030k

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

NH2 N N N N Zn -1.3 -4.5 Self-assembly of zinc aminoporphyrins Mark Gardner,b Andrea J. Guerin,a Christopher A. Hunter,*a Ulrike Michelsena and Carmen Rotger§ a Krebs Institute for Biomolecular Science Department of Chemistry University of Sheffield Sheffield UK S3 7HF b Discovery Chemistry P–zer L td Sandwich UK CT 13 9NJ Received (in Cambridge UK) 9th October 1998 Accepted 4th November 1998 The aniline»zinc porphyrin interaction is an order of magnitude weaker than the corresponding pyridine»zinc porphyrin interaction but it is still strong enough to cause self-assembly of zinc aminoporphyrins in solution. Three isomeric systems are reported two form self-assembled dimers but the geometry of the third forces it to form an open chain oligomer. The stabilities and structures of the complexes have been determined using 1H NMR spectroscopy.Introduction Self-assembly is the spontaneous generation of well-de–ned molecular assemblies via non-covalent interactions such as hydrogen-bonding and metal»ligand coordination.1h12 Porphyrins are popular building blocks because they are easily synthesized and functionalised they are large and rigid they are spectroscopically rich and they have interesting photochemical and redox properties.13h16 We have been investigating the use of pyridine»zinc porphyrin coordination interactions for the self-assembly of oligomeric arrays of these chromophores.17h19 In the course of this work we prepared a range of diÜerent aminoporphyrins and when these compounds were metallated with zinc they displayed some unusual spectroscopic properties.These are reported in this paper along with structural characterization of the complexes. We show that the aniline»zinc porphyrin coordination bond although weaker than the pyridine»zinc porphyrin interaction 20 is strong enough to yield stable self-assembled complexes. Results and discussion The three isomeric zinc aminoporphyrins 2 3 and 4 were all synthesised using the same procedure (Scheme 1).21,22 Starting from the appropriate nitrobenzaldehyde a statistical reaction with pyrrole and n-pentylbenzaldehyde gave a mixture of porphyrin products. These were separated by column chromatography to give the mononitroporphyrin or (H2L6 H2L7 and the tetraalkylporphyrin which was used for H2L8) (H2 L1) control binding experiments. Each of the nitroporphyrins was reduced with stannous chloride and metallated with zinc acetate to give 2 3 and 4 in essentially quantitative yields.was similarly metallated to give 1. H2L1 The –rst evidence for the self-assembly of the three zinc aminoporphyrins came from the 1H NMR spectra which were recorded at millimolar concentrations. The spectra were all concentration-dependent and surprisingly complex with many more signals than the corresponding free base porphyrins. At high concentrations there were particularly large up–eld shifts § Current address Department of Chemistry Universitat de les Illes Balears 07071 Palma de Mallorca Spain. for the signals due to the protons on the aniline ring which suggested that the lone pair of the aniline nitrogen of one porphyrin was coordinated to the vacant zinc binding site on the face of another porphyrin.This would bring the aniline into the shielding region of the coordinated porphyrinœs ring current. To test this hypothesis a 1H NMR titration was carried out to measure the strength of the zinc»aniline interaction a millimolar solution of 1 in d-chloroform was titrated with aliquots of a molar solution of L5. The association constant is 130^10 M~1 and the limiting complexation-induced changes in chemical shift are shown in Fig. 1. The aniline clearly coordinates to the zinc and lies over the face of the porphyrin. The self-assembly properties of the three zinc aminoporphyrins were therefore investigated by quantitative 1H NMR dilution experiments. The results for each system are discussed in turn. Complex 2 At a concentration of 25 mM the 1H NMR spectrum of 2 was surprisingly complicated and COSY and ROESY experiments were required to fully assign all of the signals.The complexity is caused by non-equivalence of the two faces of the porphyrin and slow rotation about all four meso-phenyl bonds Fig. 1 Structure of the 1 ÆL5 complex showing the limiting complexation-induced changes in chemical shift from the 1H NMR titration in chloroform (some of the porphyrin meso substituents are omitted for clarity). New J. Chem. 1999 309»316 309 NH2 N N N N Zn HN N NH N -0.3 -1.8 +0.1 -0.2 Scheme 1 on the NMR timescale. Clearly the rate of rotation about the meso-aniline bond is much slower than for the other three torsions due to the ortho substituent but nevertheless on the NMR timescale all are in slow exchange. In other systems such as and this slow exchange has no impact on H2L2 H2L6 the appearance of the 1H NMR spectrum because there is very little diÜerence between the environments on the two faces of the porphyrin.It is the self-assembly process that causes the diÜerence between these two enviroments in 2. A detailed 1H NMR dilution study was carried out on 2 in d-chloroform. The data could be –tted reasonably well to either a dimerisation isotherm or a non-cooperative linear polymerisation isotherm but the dimer model was marginally better. The dimerisation constant is 160^20 M~1 which is similar in magnitude to the value for the simple zinc»aniline interaction in 1 ÆL5 and suggests that there is no cooperative self-assembly process in this system. However this is not an appropriate comparison because there is signi–cant steric hindrance of the amine binding site in the ortho-derivative.In order to quantify this eÜect we measured the associ- 1 ÆH2L2 ation constant (Fig. 2). It is difficult to obtain accurate values for the binding constant and complexation-induced changes in chemical shift for this complex because it is very weakly bound and it is therefore not possible to reach saturation in the binding isotherm. We estimate the association constant to be 10^5 M~1 which is signi–cantly lower than the dimerisation constant for 2. Thus 2 self-assembles via a cooperative process which suggests that the structure of the complex is a closed macrocyclic dimer held together by two zinc»aniline interactions (Fig. 3).23 Of course larger macrocyclic structures are also consistent with this data but a dimer is entropically favoured and this structure is supported by the chemical shift data discussed below.Fig. 2 Structure of the complex showing the limiting 1 …H2L2 complexation-induced changes in chemical shift from the 1H NMR titration in chloroform (some of the porphyrin meso substituents are omitted for clarity). 310 New J. Chem. 1999 309»316 N N N N M C5H11 C5H11 C5H11 H11C5 N N N N M C5H11 C5H11 H11C5 NH2 N N N N M C5H11 C5H11 H11C5 NH2 N N N N M C5H11 C5H11 NH2 H11C5 NH2 H13C6 H2L1 M= 2H 1 M= Zn H2L4 M= 2H 4 M= Zn H2L3 M= 2H 3 M= Zn H2L4 M= 2H 2 M= Zn L5 The limiting complexation-induced changes in chemical shift (CIS) for and are shown in Fig. 2 and 3 1 ÆH2L2 (2)2 respectively. Although the magnitudes of the values for are subject to a large error due to the low stability of 1 ÆH2L2 the complex the pattern of shift changes is accurate and is quite diÜerent from that found for indicating that the anil- (2)2 ines are bound in diÜerent orientations in the two systems.Very large changes in chemical shift are found for the porphyrin b-pyrrole signals as well as the signals due to the protons on the aniline ring. CIS values can be used in a quantitative manner to obtain more detailed information about the structures of porphryin complexes of this type.24h26 We have recently developed a computational method for deriving high resolution three-dimensional structural information on supramolecular complexes from complexation-induced changes in chemical shift,27 and this was used to obtain the threedimensional structure of which is shown in Fig. 4 (see (2)2 Experimental section for details).The CIS values for the optimised dimer structure agree extremely well with the experimental data root mean square diÜerence (rmsd)\0.03 ppm. Complex 3 Like 2 the 1H NMR spectrum of 3 was surprisingly complicated at high concentrations and COSY and ROESY experiments were required to fully assign the spectrum. The two faces of the porphyrin are again diÜerent and rotation about the meso-phenyl bonds is slow on the NMR timescale. The low concentration spectrum and the spectra of the corresponding free base porphyrins and are relatively H2L3 H2L7 simple which implies that it is the self-assembly process which causes the two faces of the porphyrin to become nonequivalent. A 1H NMR dilution study was carried out in dchloroform. The data could be –tted equally well to either a dimerisation isotherm or a non-cooperative linear polymerisation isotherm but the association constants are an order of magnitude larger than the value for the simple zinc»aniline interaction in 1 ÆL5 (the association constant for dimerisation is 1080^90 M~1).This implies that formation of the complex is a cooperative process involving more than one coordination bond i.e. the complex is a closed macrocycle held together by two zinc»aniline interactions per monomer. Entropically the most favourable structure is a dimer (Fig. 5) and this is supported by the CIS data discussed below. The limiting complexation-induced changes in chemical shift for are shown in Fig. 5. The pattern of shift changes (3)2 is similar to that found for 2 except that the values for the porphyrin b-pyrrole signals are signi–cantly smaller.These CIS values were used in conjunction with the computational method discussed above to obtain a three-dimensional struc- New J. Chem. 1999 309»316 311 N N N N R R R Zn NH2 NH2 H2N N N N N Zn N N N N Zn -0.5 -2.1 +0.1 +0.3 +0.2 +0.2 +0.1 -0.5 -4.1 -0.7 -1.1 -0.1 -0.6 +0.1 -0.1 +0.1 N N N N R R R Zn H2N NH2 N N N N Zn N N N N Zn NH2 -0.9 -0.6 0.0 +0.2 +0.1 +0.1 +0.2 -3.3 -3.7 -0.2 -0.3 0.0 +0.1 0.0 -0.3 +0.1 Fig. 3 Limiting complexation-induced changes in chemical shift from the 1H NMR dilution of 2 (the signals due to the anilino protons were difficult to identify reliably and reproducibly). The structure of the self-assembled dimer is also shown. Some of the porphyrin meso substituents are omitted for clarity.Fig. 4 Two views of the three-dimensional structure of the selfassembled dimer The structure was determined by matching cal- (2)2 . culated complexation-induced changes in 1H NMR chemical shift with the experimental values. Fig. 5 Limiting complexation-induced changes in chemical shift from the 1H NMR dilution of 3 (the signals due to the anilino protons were difficult to identify reliably and reproducibly). The structure of the self-assembled dimer is also shown. Some of the porphyrin meso substituents are omitted for clarity. ture (Fig. 6). The calculated CIS values for the optimised dimer structure agree extremely well with the experimental data rmsd\0.02 ppm. The reason for the diÜerence in the pattern of chemical shift changes for 2 and 3 can easily be seen by comparing the structures in Fig.4 and 6 in 2 the bpyrrole protons on the edge of one porphyrin lie over the centre of the ring current of the neighbouring molecule but moving the substituent from the ortho to the meta position increases the lateral displacement of the porphyrin rings and signi–cantly reduces the extent of overlap. Complex 4 The behaviour of 4 was quite diÜerent from the other two systems. All of the signals in the 1H NMR spectrum of 4 were broad at a concentration of 25 mM but became much sharper at lower concentrations. The line-broadening suggests that self-assembly generates a high molecular weight oligomeric species (a chemical exchange process is an alternative explanation). A 1H NMR dilution study was carried out in d-chloroform. The data could be –tted equally well to either a dimerisation isotherm or a non-cooperative linear oligomerisation isotherm but the association constants are very similar to the value for the simple zinc»aniline interaction in 1 ÆL5 (the association constant for open chain oligomerisation is 190^20 M~1).This implies that there are no cooperative processes in this system and suggests that an open linear oligomer is the most likely structure of the complex (Fig. 7).23 The limiting complexation-induced changes in chemical shift are shown in Fig. 7. The values for the aniline ring protons are comparable with those found for the 1 ÆL5 complex (Fig. 1) 312 New J. Chem. 1999 309»316 Fig. 6 Two views of the three-dimensional structure of the selfassembled dimer The structure was determined by matching cal- (3)2 . culated complexation-induced changes in 1H NMR chemical shift with the experimental values.indicating that the aniline is bound in a similar orientation in both systems. The pattern of shifts on the porphyrin ring are similar to 3 but the meso substituents experience much larger ring current-induced up–eld shifts. This suggests a more upright structure for this complex with the 10- and 20-meso substituents lying over the ring current of the neighbouring molecule. Attempts to use the CIS values to determine the three-dimensional structure of a closed dimer failed for this system the conformational search was unable to locate a compatible dimeric structure. Thus the CIS data also support formation of a simple oligomeric aggregate for 4. This behaviour can easily be rationalised on the basis of the geometry of the system it is not sterically possible to form a closed dimeric complex with the para- substituent.Further evidence for the structures of the assemblies was sought from mass spectrometry and vapor pressure osmometry but the complexes are not sufficiently stable to be characterized using these techniques. Similarly self-assembly does not take place at concentrations suitable for UV/Visible absorption spectroscopy. Conclusions The stability constant for coordination of a zinc porphyrin by aniline is approximately 100 M~1. This is one order of magnitude weaker than the stability of pyridine»zinc porphyrin complexes. Nevertheless stable oligomeric assemblies of zinc aminoporphyrins are formed in solution. The ortho and meta zinc amino porphyrin derivatives 2 and 3 form closed dimers with signi–cant cooperativity between the two zinc»nitrogen bonds which hold the macrocycle together.The geometry of the para derivative precludes the formation of macrocyclic dimers and so this compound forms an open chain linear oligomer with no cooperativity. The large ring current shifts in the 1H NMR spectra provide detailed information about the three-dimensional structures of the complexes and have been used to derive high resolution structures of the two dimers. Fig. 7 Limiting complexation-induced changes in chemical shift from the 1H NMR dilution of 4 (the signals due to the anilino protons were difficult to identify reliably and reproducibly). The structure of the self-assembled oligomeric chain is also shown. Some of the porphyrin meso substituents are omitted for clarity.New J. Chem. 1999 309»316 313 Experimental 1H NMR dilution experiments A sample of known concentration (of the order 10»100 mM) in was prepared and a 1H NMR spectrum was record- CDCl3 ed on a 0.8 ml sample. From this sample 0.4 ml was removed and replaced by 0.4 ml of solvent. After shaking to mix the solvents a second 1H NMR spectrum was recorded. This procedure was repeated until there was no further change in chemical shift or the sample was too dilute to record a spectrum. For signals that moved more than 0.01 ppm over the whole concentration range the chemical shifts at each concentration were recorded and –tted to a dimerisation or noncooperative linear polymerisation isotherm using purpose written software on an Apple Macintosh microcomputer or These programs use a NMRDil Dimer NMRDil Agg.Simplex procedure to –t the experimental data to the following equations to determine the optimum solutions for the association constant and the limiting bound and free chemical shifts. –ts the data to a dimerisation isotherm by NMRDil Dimer solving the following equations [AA]\ 1]4Kd[A]0[JM1]8Kd[A]0N 8Kd (1) [A]\[A]0[2[AA] (2) dobs\ 2[AA] [A]0 dd] [A] [A]0 df (3) where is the total concentration [A] is the concentration [A]0 of unbound free species [AA] is the concentration of dimer is the dimerisation constant is the free chemical shift Kd df and is the limiting bound chemical shift of the dimer. dd –ts the data to a non-cooperative linear NMRDil Agg oligomerisation/polymerisation isotherm by solving the following equations [Agg]\[A]0G1[ 2 1]JM1]4K[A]0NH (4) [A]\[A]0[[Agg] (5) dobs\ 2[Agg] [A]0 db] [A] [A]0 df (6) where is the total concentration [A] is the concentration [A]0 of sites which are unbound (this is the sum of the free species and the ends of the aggregate which are not bound) [Agg] is the concentration of sites involved in intermolecular interactions in the aggregate K is the association constant for chain extension of the aggregate is the free chemical shift df and is the limiting bound chemical shift of the bound sites db in the aggregate.Several signals were followed in each experiment and the value quoted for the association constant is the weighted average based on the observed changes in chemical shift. The errors quoted are twice the standard error (95% con–dence limit). Complexation-induced change in chemical shift calculations The method used to determine three-dimensional structures from complexation-induced changes in chemical shift has been described in detail elsewhere.27 In this work the porphyrin ring current shifts were calculated using the eight loop Haigh» Mallion model developed by Cross and Wright with ring current intensity factors of 1.99 for the six-membered rings and 0.55 for the –ve-membered rings.28 The conformational searches were carried out using the same protocol for each system.The structures of the zinc porphyrins were taken from the X-ray crystal structure analysis of the pyridine complex of zinc tetraphenylporphyrin and amine groups were added using standard bond lengths and angles in Macromodel.29 A genetic algorithm was used to optimise the conformation of the complex so that the calculated CIS values matched the experimental values as closely as possible.We allowed intermolecular translation (^10 and Aé ) rotation (^180°) as well as intramolecular torsional changes (^180°) for all four meso-phenyl bonds in each molecule. To restrict the search space the amine nitrogens were constrained to be within 2.1^0.5 of the zinc atoms and van der Waals Aé clashes were penalised at distances of less than 3 for inter- Aé molecular clashes and 2 for intramolecular clashes for non- Aé hydrogen atoms. The searches for both 2 and 3 converged to values of of 9.0 and 13.5 respectively in about 2000 Rexpt/R*d generations for a population of 100 is the rms of the (Rexpt experimentally observed CIS values and is the rms diÜer- R*d ence between the calculated and experimental values).The 4 search failed to converge to a satisfactory solution the maximum value reached was 1.9. Rexpt/R*d Preparation of porphyrins Freshly distilled pyrrole (1.39 ml 0.02 mol 1 equivalent) the appropriate nitrobenzaldehyde (1.207 g 0.008 mol 0.4 equivalent) 4-n-pentylbenzaldehyde (2.112 g 0.012 mol 0.6 equivalent) dry EtOH (13 ml) and dry dichloromethane (2 l) were stirred under for 5 min. (1 ml 6.6 mmol N2 BF3OEt2 0.33 equivalent) was added via a septum and the reaction mixture was protected from light and stirred under nitrogen for 70 min. 2,3-Dicyano-5,6-dichloro-1,4-benzoquinone (DDQ) (4.54 g 0.02 mol 1 equivalent) was added and the reaction mixture was stirred for a further 90 min before the addition of triethylamine (2.8 ml). The solvent was removed in vacuo and the black residue was washed with methanol in a Soxhlet to remove polymeric material.The solid residue was dissolved in dichloromethane and passed through a short plug of Florisil eluting with light (1 1 bp range petroleum»CHCl3 for light petroleum; 40»60 °C) to remove any remaining polymer. The solvent was removed in vacuo and the porphyrin products were separated by column chromatography on silica eluting with light (2 1). The –rst petroleum»CHCl3 band was which was recrystallised from H2L1 to give purple plates (0.3»0.8 g 8»16%). The CHCl3»CH3OH second band was the mononitroporphyrin which was recrystallised in the same manner to give a dark purple powder. 5,10,15,20 - Tetrakis(4 - pentylphenyl) - 21H,23H - porphine Mp 313»316 °C (Found C 85.66 ; H 7.88 ; N 6.24.H2L1. Calc. for C 85.86 ; H 7.88 ; N 6.26%); C64H70N4 jmax 420 (e/dm3 mol~1 cm~1 530 000) 517 (25 000) (CH2Cl2)/nm 551 (16 000) and 598 (9 300) ; 8.87 (8H s b-pyrrolic dH(CDCl3) H) 8.12 (8H d J 7 7.56 (8H d J 7 2.96 (8H t ArHo) ArHm) J 7 1.93 (8H qn J 7 1.52 (16H m ArCH2) CH2) 8]CH2) 1.03 (12H t J 7Hz [2.75 (2H s NH); m/z (FAB) 895 CH3) (M` requires 895.2962). C64H70N4 5 - (2-Nitrophenyl) - 10,15,20 - tris(4 - pentylphenyl) - 21H,23Hporphine Yield 0.44 g 10%; mp 199»201 °C; H2L6. jmax 423 (e/dm3 mol~1 cm~1 640 000) 518 (13 400) (CH2Cl2)/nm 554 (6 810) 593 (3 990) and 649 (3 070) ; 8.88 (2H d dH(CDCl3) J 7 b-pyrrolic H) 8.86 (4H s b-pyrrolic H) 8.64 (2H d J 7 b-pyrrolic H) 8.45 (1H d J 7 8.24 (1H d J 7 5-H6) 5-H3) 8.10 (6H m 10- 15- and 7.95 (2H m and 20-Ho) 5-H4 5-H5) 7.50 (6H d J 7 10- 15- and 2.94 (6H t J 7 20-Hm) ArCH2) 1.95 (6H qn J 7 1.50 (12H m 1.05 (9H t J CH2) 6]CH2) 7Hz [2.72 (2H s NH); m/z (FAB) 870 (M` CH3) requires 870.1793).C59H59N5O2 314 New J. Chem. 1999 309»316 5- (3 - Nitrophenyl) - 10,15,20 - tris(4 - pentylphenyl) -21H,23Hporphine Yield 0.22 g 5%; mp 259»261 °C; H2L7. jmax 421 (e/dm3 mol~1 cm~1 360 000) 518 (19 000) (CH2Cl2)/nm 554 (9 590) 592 (5 890) and 648 (4 710) ; 9.10 (1H s dH(CDCl3) 8.92 (2H d J 7 b-pyrrolic H) 8.87 (4H s b-pyrrolic 5-H2) H) 8.70 (2H d J 7 b-pyrrolic H) 8.68 (1H d J 7 8.58 5-H4) (1H d J 7 8.15 (6H m 10- 15- and 7.95 (1H t 5-H6) 20-Ho) J 7 7.60 (6H d J 7 10- 15- and 2.97 (6H t J 5-H5) 20-Hm) 7 1.94 (6H qn J 7 1.50 (12H m ArCH2) CH2) 6]CH2) 1.05 (9H t J 7Hz [2.75 (2H s NH); m/z (FAB) 870 CH3) (M` requires 870.1793).C59H59N5O2 5-(4-Nitrophenyl)-10,15,20- tris(4-pentylphenyl) -21H,23Hporphine Yield 0.56 g 13%; mp 243»245 °C; H2L4. jmax 421 (e/dm3 mol~1 cm~1 330 000) 518 (18 000) (CH2Cl2)/nm 554 (11 000) 593 (6 500) and 649 (5 300) ; 8.92 (2H dH(CDCl3) d J 7 b-pyrrolic H) 8.88 (4H s b-pyrrolic H) 8.73 (2H d J 7 b-pyrrolic H) 8.63 (2H d J 7 and 8.40 (2H d 5-H2 5-H6) J 7 and 8.12 (6H d J 7 10- 15- and 7.56 5-H3 5-H5) 20-Ho) (6H d J 7 10- 15- and 2.96 (6H t J 7 1.93 20-Hm) ArCH2) (6H qn J 7 1.54 (12H m 1.05 (9H t J 7Hz CH2) 6]CH2) [ 2.75 (2H s NH); m/z (FAB) 870 (M` CH3) C59H59N5O2 requires 870.1793). General procedure for reduction of nitroporphyrins The mononitroporphyrin (200 mg 0.230 mmol 1 equivalent) was dissolved in 1,4-dioxane (45 cm3). (600 mg SnCl2 … 2H2O 2.66 mmol 12 equivalents) and concentrated HCl (70 cm3) were added.The reaction vessel was protected from light and heated for 60 min under argon using a preheated oil bath (70 °C). After removing the heat source the hot reaction mixture was basi–ed by adding concentrated and then NH3 allowed to cool to room temperature. The product was extracted with ethyl acetate (3]50 cm3). The organic fractions were combined washed with water dried over Na2SO4 and –ltered. The solvent was removed in vacuo and the product was puri–ed by column chromatography on silica eluting with light (1 4). The product was petroleum»CH2Cl2 recrystallised from giving a purple powder CH2Cl2»CH3OH (0.18 g 95%). 5 - (2 -Aminophenyl)-10,15,20 -tris(4 -pentylphenyl) -21H,23Hporphine Mp 169»171 °C; (cyclohexane)/nm 423 H2L2.jmax (e/dm3 mol~1 cm~1 94 500) 517 (19 200) 553 (9 240) 592 (5 550) and 647 (4 540) ; 8.80 (8H s b-pyrrolic H) dH(CDCl3) 8.05 (6H m 10- 15- and 7.84 (1H d J 7 7.53 20-Ho) 5-H6) (1H t J 7 7.50 (6H d J 7 10- 15- and 7.25 5-H4) 20-Hm) (1H t J 7 7.10 (1H d J 7 3.50 (2H s 5-H5) 5-H3) NH2) 2.94 (6H t J 7 1.95 (6H qn J 7 1.50 (12H m ArCH2) CH2) 1.05 (9H t J 7Hz [2.80 (2H s NH); m/z 6]CH2) CH3) (FAB) 840 (M` requires 840.1644). C59H61N5 5- (3 -Aminophenyl)-10,15,20 - tris(4-pentylphenyl) -21H,23Hporphine Mp 153»155 °C; 421 (e/dm3 H2L3. jmax (CH2Cl2)/nm mol~1 cm~1 328 000) 517 (13 000) 552 (6 890) 593 (3 840) and 648 (3 500) ; 8.93 (2H d J 7 b-pyrrolic H) 8.84 dH(CDCl3) (4H s b-pyrrolic H) 8.82 (2H d J 7 b-pyrrolic H) 8.10 (6H d J 7 10- 15- and 7.62 (1H t J 7 7.50 (7H m 20-Ho) 5-H6) 10- 15- and and 7.47 (1H d J 7 7.10 (1H 20-Hm 5-H2) 5-H5) d J 7 3.94 (2H s 2.94 (6H t J 7 1.95 5-H4) NH2) ArCH2) (6H qn J 7 1.50 (12H m 1.05 (9H t J 7Hz CH2) 6]CH2) [2.80 (2H s NH); m/z (FAB) 841 (M]1 CH3) C59H61N5 requires 840.1644).5- (4 -Aminophenyl)-10,15,20 -tris(4 -pentylphenyl)-21H,23Hporphine Mp 147»149 °C; 422 (e/dm3 H2L4. jmax (CH2Cl2)/nm mol~1 cm~1 290 000) 519 (15 000) 557 (10 000) 594 (4 600) and 650 (5 300) ; 8.91 (4H m b-pyrrolic H) 8.87 dH(CDCl3) (4H s b-pyrrolic H) 8.10 (6H d J 7 10- 15- and 8.00 20-Ho) (2H d J 7 and 7.55 (6H d J 7 10- 15- and 5-H2 5-H6) 7.10 (2H d J 7 and 4.05 (2H s 20-Hm) 5-H3 5-H5) NH2) 2.95 (6H t J 7 1.90 (6H qn J 7 1.4»1.6 (12H ArCH2) CH2) m 1.05 (9H t J 7Hz [2.75 (2H s NH); m/z 6]CH2) CH3) (FAB) 840 (M` requires 840.1644). C59H61N5 General procedure for the metallation of porphyrins The free base porphyrin (36 mg 0.04 mmol 1 equivalent) was dissolved in (3 1 30 ml) and zinc acetate CH2Cl2»CH3OH (94 mg 0.4 mmol 10 equivalents) was added.The reaction mixture was protected from light and stirred at room temperature for 60 min. The solvent was removed in vacuo and the product was puri–ed by column chromatography on basic alumina eluting with (99 1). Rec- CH2Cl2»CH3OH rystallisation from yielded the product as a CHCl3»CH3OH purple powder (37 mg 96%). [ 5,10,15,20-Tetrakis(4-pentylphenyl)-21H,23H-porphinato ] - zinc 1. Mp 309»311 °C (Found C 80.23 ; H 7.25 ; N 5.87. Calc. for C 80.19 ; H 7.15 ; N 6.26%); C64H68N4Zn jmax 422 (e/dm3 mol~1 cm~1 650 000) 551 (26 000) (CH2Cl2)/nm and 592 (8 300) ; 8.98 (8H s b-pyrrolic H) 8.12 (8H dH(CDCl3) d J 7 7.55 (8H d J 7 2.96 (8H t J 7 ArHo) ArHm) ArCH2) 1.93 (8H qn J 7 1.52 (16H m 1.03 (12H t CH2) 8]CH2) J 7Hz m/z (FAB) 957 (M` requires CH3) ; C64H68N4Zn 957.51).[ 5-(2-Aminophenyl)-10,15,20-(4-pentylphenyl)-21H,23H-porphinato ] zinc 2. Mp 227»230 °C; 423 (e/dm3 jmax (CH2Cl2)/nm mol~1 cm~1 61 500) 549 (29 100) and 589 (6 760) ; dH(CDCl3) concentration-dependent 9.04 (2H d J 7 b-pyrrolic H) 8.92 (2H d J 7 b-pyrrolic H) 8.70 (2H d J 7 b-pyrrolic H) 8.28 (1H d J 7 8.20 (1H d J 7 8.17 (2H d J 7 15-Ho) 15-Ho) 10- and 8.08 (2H d J 7 b-pyrrolic H) 7.93 (2H d J 7 20-Ho) 10- and 7.75 (1H d J 7 7.60 (1H d J 7 20-Ho) 15-Hm) 7.60 (2H d J 7 10- and 7.48 (2H d J 7 10- 15-Hm) 20-Hm) and 7.48 (1H d J 7 7.08 (1H t J 7 6.94 20-Hm) 5-H6) 5-H4) (1H t J 7 4.91 (1H d J 7 2.85 (6H t J 7 5-H5) 5-H3) 1.88 (6H qn J 7 1.48 (12H m 1.00 ArCH2) CH2) 6]CH2) (9H t J 7Hz 0.60 (2H s m/z (FAB) 903 (M` CH3) NH2) ; requires 903.4886).C59H59N5Zn [ 5-(3-Aminophenyl)-10,15,20-(4-pentylphenyl)-21H,23H-porphinato ] zinc 3. Mp 169»172 °C; 422 (e/dm3 jmax (CHCl3)/nm mol~1 cm~1 307 000) 549 (14 600) and 589 (2 940) ; dH(CDCl3) concentration-dependent 9.03 (2H d J 7 b-pyrrolic H) 8.95 (2H d J 7 b-pyrrolic H) 8.71 (2H d J 7 b-pyrrolic H) 8.25 (2H m b-pyrrolic H) 8.29 (2H d J 7 10- and 8.20 20-Ho) (1H d J 7 8.16 (1H d J 7 7.94 (2H d J 7 15-Ho) 15-Ho) 10- and 7.69 (1H d J 7 7.62 (2H d J 7 10- 20-Ho) 15-Hm) and 7.57 (1H d J 7 7.46 (2H d J 7 10- and 20-Hm) 15-Hm) 7.32 (1H d J 7 6.62 (1H t J 7 3.83 (2H 20-Hm) 5-H6) 5-H5) m and 2.96 (6H t J 7 1.94 (6H qn J 7 5-H2 5-H4) ArCH2) 1.50 (12H m 1.04 (9H t J 7Hz m/z CH2) 6]CH2) CH3) ; (FAB) 903 (M` requires 903.4886).C59H59N5Zn [5-(4-Aminophenyl)-10,15,20-tris(4-pentylphenyl)-21H,23Hporphinato ] zinc 4. Mp 101»104 °C; 425 jmax (CH2Cl2)/nm (e/dm3 mol~1 cm~1 100 000) 551 (35 400) and 592 (12 900) ; concentration-dependent 9.11 (2H d J 7 b- dH(CDCl3) pyrrolic H) 8.98 (2H d J 7 b-pyrrolic H) 8.45 (2H d J 7 b-pyrrolic H) 8.36 (2H d J 7 b-pyrrolic H) 8.30 (2H d J 7 7.64 (2H d J 7 7.86 (4H d J 7 10- and 15-Ho) 15-Hm) 7.08 (6H m 10- and and and 3.71 20-Ho) 20-Hm 5-H2 5-H6) (2H d J 7 and 3.04 (2H t J 7 2.75 (4H 5-H3 5-H5) ArCH2) t J 7 2.00 (2H qn J 7 1.73 (4H qn J 7 ArCH2) CH2) 1.45 (2H m 1.30 (4H m 1.11 (3H t J CH2) CH2) 2]CH2) 7 0.93 (6H t J 7Hz 0.35 (2H s m/z CH3) CH3) NH2) ; (FAB) 903 (M` requires 903.4886). C59H59N5Zn New J. Chem. 1999 309»316 315 Acknowledgements thank P–zer and the EPSRC (AJG) the Spanish govern- We ment (CR) and the Lister Institute (CAH) for –nancial support.References 1 J. S. Lindsey New J. Chem. 1991 15 153. 2 M. Fujita J. Yazaki and K. Ogura J. Am. Chem. Soc. 1990 112 5645. 3 M. Fujita S. Nagao and K. Ogura J. Am. Chem. Soc. 1995 117 1649. 4 S. C. Zimmerman and B. F. Duerr J. Org. Chem. 1992 57 2215. 5 S. C. Zimmerman F. W. Zeng D. E. C. Reichert and S. V. Kolotuchin Science 1996 271 1095. 6 D. P. Funeriu J. M. Lehn G. Baum and D. Fenske Chem. Eur. J. 1997 3 99. 7 P. N. W. Baxter J. M. Lehn J. Fischer and M. T. Youinou Angew. Chem. Int. Ed. Engl. 1994 33 2284. 8 N. Kimizuka S. Fujikawa H. Kuwahara T. Kunitake A. Marsh and J. M. Lehn J. Chem. Soc. Chem. Commun. 1995 2103. 9 P. J.Stang N. E. Persky and J. Manna J. Am. Chem. Soc. 1997 119 4777. 10 J. Yang J. L. Marendaz S. J. Geib and A. D. Hamilton T etrahedron L ett. 1994 35 3665. 11 N. Branda R. Wyler and J. Rebek Science 1994 263 1267. 12 G. M. Whitesides E. E. Simanek J. P. Mathias C. T. Seto D. N. Chin M. Mammen and D. M. Gordon Acc. Chem. Res. 1995 28 37. 13 C. M. Drain R. Fischer E. G. Nolen and J. M. Lehn J. Chem. Soc. Chem. Commun. 1993 243. 14 C. M. Drain and J. M. Lehn J. Chem. Soc. Chem. Commun. 1994 2313. 15 H. L. Anderson Inorg. Chem. 1994 33 972. 16 P. J. Stang J. Fan and B. Olenyuk Chem. Commun. 1997 1453. 17 C. A. Hunter and L. D. Sarson Angew. Chem. Int. Ed. Engl. 1994 33 2313. 18 X. L. Chi A. J. Guerin R. A. Haycock C. A. Hunter and L. D. Sarson J. Chem. Soc. Chem. Commun. 1995 2567. 19 C.A. Hunter and R. K. Hyde Angew. Chem. Int. Ed. Engl. 1996 35 1936. 20 J. K. M. Sanders Compr. Supramol. Chem. 1996 9 131. 21 J. S. Lindsey and R. W. Wagner J. Org. Chem. 1989 54 828. 22 J. S. Lindsey K. A. Maccrum J. S. Tyhonas and Y. Y. Chuang J. Org. Chem. 1994 59 579. 23 X. L. Chi A. J. Guerin R. A. Haycock C. A. Hunter and L. D. Sarson J. Chem. Soc. Chem. Commun. 1995 2563. 24 R. J. Abraham and K. M. Smith J. Am. Chem. Soc. 1983 105 5734. 25 K. M. Smith F. W. Bobe D. A. GoÜ and R. J. Abraham J. Am. Chem. Soc. 1986 108 1111. 26 P. Leighton J. A. Cowan R. J. Abraham and J. K. M. Sanders J. Org. Chem. 1988 53 733. 27 C. A. Hunter and M. J. Packer submitted. 28 K. J. Cross and P. E. Wright J. Magn. Reson. 1985 64 220. 29 F. Mohamadi N. G. J. Richards W. C. Guida R. Liskamp M. Lipton C. Cau–eld G. Chang T. Hendrickson and W. C. Still J. Comput. Chem. 1990 11 440. Paper 8/07876I 316 New J. Chem. 1999 309»316

ISSN:1144-0546    

DOI:10.1039/a807876i

出版商:RSC    

年代:1999

数据来源: RSC