摘要:

Mendeleev Communications Electronic Version, Issue 1, 1999 (pp. 1–44) Time-of-flight mass spectroscopic detection of new elemental and mixed small atomic clusters in the laser evaporation of carbon nitride Sergei I. Kudryashov,* Byong K. Kim, Jong I. Kim, Nikita B. Zorov and Yurii Ya. Kuzyakov Department of Chemistry, M. V. Lomonosov Moscow State University, 119899 Moscow, Russian Federation. Fax: +7 095 932 8846; e-mail: serge@laser.chem.msu.ru New, positively-charged elemental and mixed atomic carbon–nitrogen clusters have been detected by time-of-flight mass spectroscopy in the laser evaporation of carbon nitride C3N4.25.Many elemental and mixed atomic clusters of carbon and nitrogen — promising materials in the production of high-energy density materials — demonstrate relative stability, according to recent ab initio calculations.1,2 Meanwhile, up to date experimental studies on the structural, physical and chemical properties of these clusters are very rare because production of the clusters by laser vaporisation source techniques is impeded by a lack of nitrogen-rich carbonaceous targets.3–6 In this case, current mass spectroscopic studies of the gas-phase evaporation products of carbon–nitrogen materials, usually related to the laser deposition of superhard carbon nitride (b-C3N4) thin films,7,8 can also provide a clearer understanding of the nature of nitrogen and nitrogen–carbon atomic clusters.In this work positively charged products from the laser evaporation of carbon nitride C3N4.25 (a polymeric material with a symm-heptazine monomer produced by thermal decomposition of mercuric rhodanide9) were studied by time-of-flight mass spectroscopy.The sample was prepared from a purified powder of polymer carbon nitride (PCN) pressed at 50 bar to a pellet (diameter 0.7 cm, thickness 0.4 cm, bulk density 1.6 g cm–3). Time-of-flight studies were carried out using a commercial quadrupole MX-7304 mass spectrometer modified for time-offlight mass analysis (Figure 1).Operation conditions for the linear mass spectrometer were maintained by pumping the vacuum chamber with a high-vacuum discharge pump up to 10–7 Torr. The pulsed second harmonic output of a Q-switched Nd:YAG laser [laser wavelength l = 532 nm, pulse energy E = 4.5±0.3 mJ, pulse duration (FWHM) t = 10±1 ns, pulse repetition rate f = =0.92±0.05 Hz] attenuated by neutral calibrated filters and a LiNbO3 polariser was focused by a lens (F = 28 cm) onto the surface of a PCN target at a slight angle (10–15°).A small part (8%) of the laser radiation energy was directed by a beam splitter onto a photodiode and a pyroelectric plate to synchronise the detection system and to control the laser energy per pulse.Primary positive ions generated from the PCN target were extracted and accelerated by a pulsed voltage (–100 V, pulse width 0.1–10 ms) applied to a grid placed at a distance of 4 cm in front of the target with a ground graphite ring supporting the target. The cluster ion beam, with an adjustable mechanical momentum of the ions, was mass-analysed during drift in a field-free region (45 cm) to the input of a secondary electron multiplier with a time constant of 3 ns.After discrimination of noise and pre-amplification separate bursts of positive pulses of amplitude +4 V corresponding to individual cluster ions were digitised for 320 ms by a pulse counter connected to a PC via an interface. The digital signal was averaged over 100 laser shots and was then recorded.The time-of-flight spectra of positive cluster ions were obtained during laser evaporation of the PCN target with laser power density 0.3 GW cm–2. The calibration curve for the cluster ion mass versus flight time obtained for the cluster source conditions allowed the determination of ion mass values to an accuracy of 0.1% and reproducibility 0.3%. Maximum initial kinetic energies of ions measured by retarding potential technique were equal to 0.5 << eU ª 100 eV.Corresponding temperature of the cluster ion source T < 6000 K was favourable for preferable formation of singly charged ions. Cluster ion masses were calculated using equation (1) for the acceleration and drift of singly charged ions: where M is the cluster ion mass (a.m.u.) related to its charge (Z = 1), e is the charge of an electron (C), U is the accelerating voltage (V), l is the distance from the target to the accelerating grid (m), L is the drift tube length (m), m0 is one twelfth part of the carbon atom mass (carbon unit of mass, kg), t is the flight time (s) and t is the accelerating electric pulse width (s).According to the formula the maximum mass of a cluster ion detected with the mass analyser is proportional to the accelerating pulse width and increases from 200 to 1700 a.m.u., with growth of the latter in the range 1–10 ms.Pulsed acceleration of cluster ions to equal momentum is favourable in the resolution of high molecular mass ions due to the linear dependence of ion mass on flight time, as compared with monoenergetic ion beams which are described by a square root dependence. In this case, mass resolution of the cluster ion beam over the whole mass range was independent of cluster ion mass and was limited only by time resolution of the detection system tc = 1.25±0.05 ms (counting cycle of the pulse counter). Then mass resolution of the nearest peaks in the mass spectra of positive clusters with a difference in mass of D a.m.u.was provided by their detection within the next counting cycle: Taking into account equation (2), the numerical expression for mass resolution of the cluster ion beam is of the form D (a.m.u.) = 1.2t (ms) which corresponds to resolution of the two nearest peaks for mixed carbon–nitrogen cluster ions (with D = 2 a.m.u.) by using an accelerating electric pulse of width t < 2 ms.The abundances of the cluster ions detected were multiplied by coefficient M0.5, thus accounting for the dependence E/M0.5 of the ion-electron conversion efficiency of a secondary electron 1 2 3 4 5 6 7 8 9 10 11 12 11 Figure 1 Scheme for the linear time-of-flight mass spectrometer: 1 – laser, 2 – beam splitter, 3 – mirror, 4 – focusing lens, 5 – vacuum chamber, 6 – polymer carbon nitride target supported on a ground graphite ring, 7 – extracting/accelerating grid and ion drift tube of mass analyser, 8 – secondary electron multiplier, noise discriminator and pre-amplifier, 9 – photodiode and pyroelectric plate, 10 – computer with pulse counter and interface, 11 – prevacuum pump and high-vacuum discharge pump, 12 – thermocouple and ionisation gauges.M = (1) eUt(t – 0.5t) l(L + l)m0 t(N + D) – t(N) = D � 2tc Llm0 eUt (2)Mendeleev Communications Electronic Version, Issue 1, 1999 (pp. 1–44) multiplier on cluster ion mass (where E is the kinetic energy and M is the mass of the corresponding ion10). The time-of-flight spectrum obtained for the PCN target and recorded at an accelerating pulse width of 1 ms and laser power density 0.3 GW cm–2 (Figure 2) exhibits a variety of elemental and mixed carbon–nitrogen atomic positive cluster ions as well as molecular ions H2O+, N2 +, NO+ and O2 +.Most of the cluster ions appear in the 50–140 a.m.u. mass range due to the enhanced stability of their cyclic structure (for example, five-membered ring11) which is more favourable compared with a linear structure due to saturation of the valences.An unambiguous interpretation of chemical composition for the detected cluster ions Ck – 7xNm + 6x is possible only at the index values k < 7 and m < 6. Thus, in the mass range 50–110 a.m.u. new mixed carbon–nitrogen atomic positive cluster ions CN4 +, C2N4 +, C4N3 +, C5N3 +, C3N5 + and C5N4 + were identified.The chemical composition of the cluster ions with mass in the 110–140 a.m.u. range (also including masses 84, 96 and 98 a.m.u.) can be interpreted by many alternative combinations of carbon and nitrogen atoms [for example, an ion of mass 84 a.m.u. can be identied as C7 + (ref. 12) as well as N6 + (ref. 3)] (Figure 2). Unfortunately, an accurate identification of these Ck – 7xNm + 6x cluster ions with index values k > 7 or m > 6 (over 84–140 a.m.u.mass range) by an analysis of the natural isotopic distribution of carbon and nitrogen atoms was impossible, due to the discrimination of the low-intensity signals of isotopically-substituted cluster ions by the detection system of the mass spectrometer. Thus, time-of-flight mass spectroscopic detection of clusters generated by a laser vaporisation technique from the unique nitrogen-rich target of polymeric carbon nitride C3N4.25 carried out in this work shows the existence of several new elemental and mixed carbon–nitrogen small cluster ions incorporating up to 5 nitrogen atoms. The authors are grateful to Dr.O. V. Kravchenko for providing a carbon nitride sample, and the Russian Foundation for Basic Research for partial financial support (grant no. 97-03-33228). References 1 S. Shi, E. Chu, J. Yang, B. Xu and X. Hu, Theory Pract. Energ. Mater., 1996, 75. 2 M. L. Leininger, T. J. van Huis and H. F. Schaefer, J. Phys. Chem. A, 1997, 101, 4460. 3 R. Huang, Ch. Liu, F. Huang, F. Lin and L. Zheng, Huaxue Tongbao, 1995, 39, 61 (in Chinese). 4 Z.Tang, L. Shi, R.-B. Huang and L.-S. Zheng, Huaxue Xuebao, 1997, 55, 1191 (in Chinese). 5 T. Ruchti, T. Speck, J. P. Connelly, E. J. Bieske, H. Linnartz and J. P. Maier, J. Chem. Phys., 1996, 105, 2591. 6 Ch.-G. Zhan and S. Iwata, J. Chem. Phys., 1996, 104, 9058. 7 F. Kokai, Y. Koga, Y. Kakudate, M. Kawaguchi and S. Fujiwara, Appl. Phys. A., 1994, 59, 299. 8 P. T. Murray and M. Y. Chen, Surf. Rev. Lett., 1996, 3, 197. 9 S. I. Kudryashov, O. V. Kravchenko, G. M. Khafizova, K. P. Burdina and N. B. Zorov, Mendeleev Commun., 1998, 75. 10 A. A. Sysoev and M. S. Chupakhin, Vvedenie v mass-spektrometriyu (Introduction to mass spectrometry), Atomizdat, Moscow, 1972, p. 262 (in Russian). 11 Ch. Chen and K.-Ch. Sun, Int. J. Quantum Chem., 1996, 60, 497. 12 H. W. Kroto, A. W. Allaf and S. P. Balm, Chem. Rev., 1991, 91, 1213. Ion intensity (a.u.) 40 20 0 20 40 60 80 100 120 140 Ion mass (a.m.u.) H2O C2 CN N2 NO C3 O2 CN3N 4 C5 CN4 C4N N5 C4N2 C2N4 C7 (N6) C5N2 C4N3 C2N5 C8 (CN6) C7N (N7) C6N2 C5N3 C4N4 C3N5 C8N (CN7) C7N2 (N8) C5N4 C7N3 (N9) C11 (C4N6) C9N2 (C2N8) C8N3 (CN9) Figure 2 Time-of-flight spectrum of positive cluster ions (in the range 10–140 a.m.u.) generated from the polymer carbon nitride target. Received: Moscow, 8th July 1998 Cambridge, 8th October 1998; Com. 8/05582C

ISSN:0959-9436     

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 1, 1999 (pp. 1–44) Production of carbon nanoclusters supported on a graphite foil by laser ablation under supercritical conditions Sergei I. Kudryashov,* Sergei G. Ionov and Nikita B. Zorov Department of Chemistry, M. V. Lomonosov Moscow State University, 119899 Moscow, Russian Federation. Fax: +7 095 932 8846; e-mail: serge@laser.chem.msu.ru The most probable size of carbon nanoclusters produced under supercritical conditions by pulsed laser ablation from a graphite foil covered by a glass plate gradually decreased from 40 to 10 nm as the laser power density increased from 0.05 to 1.6 GW cm–2.Nanostructured materials consisting of nanoclusters less than 10 nm in diameter are expected to be different from common polycrystalline materials in the elastic, optical, magnetic and electrical properties.1–3 These cluster-assembled materials are usually produced by physical deposition using various sources of clusters (laser vaporisation, sputtering of solids by ion and electron beams or electric discharges).In this work, a new technique for the production of refractory nanocrystalline materials is proposed. It is based on clustering (spinodal decomposition) of a thermodynamically unstable (labile) liquid phase with near-critical parameters.4 The characteristic size of charged carbon nanoclusters that were generated from a polycrystalline graphite under laser-induced spinodal decomposition and detected in a gas phase by real-time electrostatic probe technique5 varied within the range 104–106 atoms per cluster (the diameter of clusters varied from 6 to 30 nm at a liquid carbon density of about 2 g cm–3).6 However, the preparation of nanocrystalline materials using this phenomenon has not been studied experimentally.In this work, carbon nanoclusters were produced from a lowdensity (r = 0.7 g cm–3) graphite foil, which was extensively studied earlier.7–10 Low bulk density and high porosity (near 70%) of the sample resulted in the black body absorbance (0.9) of visible light by a subsurface layer (100 nm) of the foil, very slow propagation of ultrasonic waves (Vus ª 450 m s–1)8 in the sample and low thermal diffusivity.Quasi-equilibrium laser evaporation of the subsurface layer of the sample occurred in pores with an average size of 10–20 nm (comparable to a free path length under conditions of intense evaporation) and was followed by subsequent formation of critical and supercritical states of carbon9,10 due to a relatively high bulk density of the foil sample (r > rcrit, where the critical carbon density rcrit is equal to 0.64 g cm–3).11 A considerable energy (about 70 kJ mol–1) released under laser irradiation of the graphite foil due to thermal annealing of non-equilibrium thermally and chemically induced defects (vacancies etc.) in graphite crystallites decreased the laser power density required for the generation of a supercritical carbon phase to 0.006 GW cm–2.9,10 A sample of the graphite foil (size of 1×1 cm, thickness of 0.06 cm) used for preparing carbon nanoclusters was placed on a square linear translation stage, which can be moved in the horizontal plane with a step of 5 mm.The surface of the lowdensity sample was accurately covered with a transparent glass plate to avoid any interface cavities. This glass plate also allowed laser irradiation of the target and prevented the removal of the material. The second harmonic output of a Q-switched Nd:YAG laser [wavelength 532 nm, pulse energy 4.5±0.3 mJ, pulse width (FWHM) 25±2 ns, angular divergence 0.6 mrad, pulse repetition rate 12.5±1.5 Hz] attenuated by neutral calibrated filters by factors of 2.5, 5, 13.5 and 32 was chopped and focused by a lens (F ª 28 cm) through the covering glass plate at the surface of the graphite foil sample with a focal aperture of the laser beam of about 160 mm.Fresh spots of the sample translated horizontally with a step of 150 nm were irradiated by 1–2 laser pulses. After irradiation, the sample was examined using a commercial JEM-2000FX scanning electron microscope (SEM) with a LaB6 cathode (the maximum magnification of the microscope was 800000, and the accelerating voltage of the electron beam was 200 kV).The instrument was operated in the mode of detecting of secondary electrons scattered at the right angle to the surface of the graphite foil sample. SEM images of the modified surface of the graphite foil irradiated with the laser power densities I0 0.05, 0.12, 0.3, 0.64 and 1.6 GW cm–2 were obtained. Laser-induced modification of the covered graphite foil surface proceeded as an isochoric ‘autoclave’ process in the subsurface foil layer with the thickness X of 0.3–5 mm, as measured by SEM (Table 1) [the initial depth of light absorption a–1 (532 nm) in the foil sample was equal to only 0.3 mm].That is, Figure 1 SEM image of carbon nanoclusters at the surface of a graphite foil sample irradiated with a laser power density of 0.05 GW cm–2 (magnification ×100000). Table 1 Experimental and calculated parameters of laser-induced modification of a graphite.Power density/ GW cm–2 Thickness of heated layer/mm Density of deposited energy/ kJ cm–3 Molar energy/ kJ mol–1 0.048±0.005 0.31±0.05 (3.4±0.8)×101 (6±2)×102 0.12±0.01 0.5±0.1 (5.6±1.7)×101 (9±3)×102 0.31±0.03 1.0±0.2 (7±2)×101 (12±4)×102 0.64±0.06 2.0±0.5 (7±2)×101 (12±4)×102 1.6±0.2 5±1 (7±2)×101 (12±4)×102 Figure 2 SEM image of carbon nanoclusters at the surface of a graphite foil sample irradiated with a laser power density of 1.6 GW cm–2 (magnification ×100000).Mendeleev Communications Electronic Version, Issue 1, 1999 (pp. 1–44) the pressure and temperature of the substance in the layer gradually increased during a laser pulse with the power density.Acoustic off-loading of the laser-heated layer with a thickness of 0.3–5 mm takes less than 10 ns at the velocity of ultrasonic waves Vus ª 450 m s–1. Thus, the main step of the graphite foil transformation occurs under quasi-static thermodynamic conditions of the ‘bath’. The volume density of the energy Edep deposited in the laser-heated layer of the sample was calculated on the assumption that the absorbance A of the sample is constant (0.9) during a laser pulse: where I(t) is the current laser power density with the peak value I0, min(a, X–1) is the effective absorption coefficient of the laserheated layer.The molar energies corresponding to the volume densities of deposited energy 30–70 kJ cm–3 are equal to 600– 1200 kJ mol–1 (Table 1) and increase as the laser power density increased from 0.05 to 1.6 GW cm–2.These calculated values of the molar energy deposited in the laser-heated subsurface layer of the graphite foil are considerably higher than the enthalpy of formation of critical carbon (about 270 kJ mol–1 as was estimated from the data12). This fact provides support to the conclusions of papers9,10 related to generation of a supercritical carbon state at a free (uncovered) surface of the same graphite foil (r > rcrit) under the same conditions of laser ablation (I0 = 0.006–1.6 GW cm–2).Carbon nanoclusters sized from 10 to 40 nm (Figures 1 and 2) were observed by SEM on the irradiated spots of the graphite foil surface. According to the calculations of a molar energy deposited in a laser-heated layer of the graphite foil, these particles were generated by laser ablation under supercritical conditions.It is assumed that this expanding supercritical fluid of carbon was quenched due to high porosity of the graphite foil and hence exhibited a microstructure of supercritical density fluctuations. The most probable size (diameter) of carbon nanoclusters generated (Figure 3) gradually diminishes with increasing molar energy deposited in a laser-heated layer of the graphite foil (Figure 4).This tendency is consistent with the current theory of supercritical fluids,4 which predicts a temperature-dependent decrease of the correlation radius as T–2/3 and thus provides a real opportunity to control the size of laser-generated nanoclusters by choosing proper operation conditions.Thus, carbon nanoclusters of size 10–40 nm were generated from a refractory material (graphite) through an intermediate laser-induced supercritical state of carbon. This work was supported in part by the Russian Foundation for Basic Research (grant no. 98-03-32679). References 1 G. W. Nieman, J. R. Weertman and R. W. Siegel, Nanostruct. Mater., 1992, 1, 185. 2 S. Gangopadhyay, G. C. Hadjipanayi and B. Date, Nanostruct. Mater., 1992, 1, 77. 3 A. Takami, H. Yamada, K. Nakano and S. Koda, Jpn. J. Appl. Phys., 1996, 35, 781. 4 V. P. Skripov, E. N. Sinitsyn and P. A. Pavlov, Termodinamicheskie svoistva zhidkostei v metastabil’nom sostoyanii (Thermodynamic Properties of Liquids in Metastable State), Atomizdat, Moscow, 1980, ch. 1 (in Russian). 5 S. I. Kudryashov, A. A. Karabutov and N. B. Zorov, Mendeleev Commun., 1998, 6. 6 S. I. Kudryashov and N. B. Zorov, Mendeleev Commun., 1998, 178. 7 V. A. Kulbachinskii, S. G. Ionov, S. A. Lapin and A. G. Mandrea, Phys. Chem. Solids, 1996, 57, 893. 8 S. I. Kudryashov, S. V. Sokolov, N. B. Zorov, A. A. Karabutov and Yu. Ya. Kuzyakov, Mendeleev Commun., 1997, 25. 9 S. I. Kudryashov, S.G. Ionov, A. A. Karabutov and N. B. Zorov, Mendeleev Commun., 1998, 212. 10 S. I. Kudryashov, S. N. Borisov, S. G. Ionov and N. B. Zorov, Mendeleev Commun., 1998, 214. 11 H. R. Leider, O. H. Krikorian and D. A. Young, Carbon, 1973, 11, 555. 12 Termodinamicheskie svoistva individual’nykh veshchestv (Thermodynamic Properties of Individual Substances), ed. V. P. Glushko, Nauka, Moscow, 1979, vol. 2, p. 10 (in Russian). Edep = Amin(a, X–1)òI(t)dt (1) 1.0 0.8 0.6 0.4 0.2 0.0 0 10 20 30 40 50 Abundance of nanoclusters (a.u.) Nanocluster diameter/nm Figure 3 Abundance of carbon nanoclusters (5 measurements, RSD < 25%) at the laser power densities (GW cm–2): 0.05, 0.12, 0.3, 0.64 and 1.6. Power density/GW cm–2 40 30 20 10 0 0.03 0.1 1 3 Probable diameter of nanoclusters/nm Figure 4 The most probable diameter of carbon nanoclusters versus laser power density. Received: Moscow, 15th July 1998 Cambridge, 26th October 1998; Com. 8/05584J

ISSN:0959-9436     

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 1, 1999 (pp. 1–44) Intriguing modes of addition of 1,3,4-triphenyl-4,5-dihydro-1H-1,2,4-triazol- 5-ylidene to bicyclopropylidene Armin de Meijere,*a Sergei I. Kozhushkov,a Dmitrii S. Yufitb and Judith A. K. Howardb a Institut für Organische Chemie der Georg-August-Universität Göttingen, D-37077 Göttingen, Germany. Fax: + 49 (0)551 39 9475; e-mail: ameijer1@uni-goettingen.de b Department of Chemistry, University of Durham, DH1 3LE, UK 1,3,4-Triphenyl-4,5-dihydro-1H-1,2,4-triazol-5-ylidene 1 reacts with bicyclopropylidene 2 to yield four unexpected products 3–6, none of which resembles the typical [2+1] mode of cycloaddition observed for 1 with electron-deficient alkenes.Bicyclopropylidene 2 is a uniquely strained and reactive tetrasubstituted alkene which has been shown to readily add electrophiles including organometallics1 and undergo various cycloadditions2 including [2+1] cycloadditions even of nucleophilic carbenes such as dimethoxycarbene.3 We have now tested the reactivity of 2 towards the stable carbene 1,3,4-triphenyl-4,5- dihydro-1H-1,2,4-triazol-5-ylidene 14 and found four unexpected products 3–6 resembling four unusual modes of addition (Scheme 1).† The structures of all new compounds 3–6 were unequivocally established by X-ray crystal structure analyses (Figure 1).‡ No mechanistic details of these additions and cycloadditions have been proved as yet and even the rationalisation of their formation is difficult except for compounds 3 and 5.Most † Compounds 3–7 were obtained by heating a solution of the heterocycle 14 (183 mg, 0.615 mmol) and bicyclopropylidene 25 (246 mg, 288 ml, 3.075 mmol) in anhydrous toluene (10 ml) at 100 °C under argon for 3 h in a sealed tube.The resulting mixture was concentrated under reduced pressure and chromatographed (3×15 cm column, 40 g of silica gel, CH2Cl2–hexane, 5:1) to give 28 mg (12%) of 5,7,8-triphenyl-5,6,8-triazadispiro[ 2.0.4.3]undeca-6,10-diene 3, 53 mg (23%) of 1,4-diphenyl-2- (1-cyclopropylcyclopropyl)-6,7-benzo-1,3,5-triazepine 4, 37 mg (19%) of 1,3,4-triphenyl-4,5-dihydro-1H-1,2,4-triazol-5-one 7 and 60 mg (25%) of the non-separable mixture of 4,5-dihydro-1,3-diphenyl[4,1' :5,1'' ]bis- (spirocyclopropane)[3a-H][1,2,4]triazolo[4,3-a]quinoline 5 and N(3)-[2- (1-cyclopropylcyclopropylcarbonyl)phenyl]-N(1)-phenylbenzamidrazone 6.Their relative ratio was determined from the 1H NMR spectrum of the mixture in comparison to the spectra of the individual compounds obtained by the selection of the crystals in accordance with their shape. For 3: mp 166–168 °C (decomp.) (hexane–ether), Rf 0.57. 1H NMR (250 MHz, CDCl3) d: 0.86–0.97 (m, 1H, cyclopropyl), 1.04–1.12 (m, 1H, cyclopropyl), 1.24–1.26 (m, 1H, cyclopropyl), 1.39–1.46 (m, 1H, cyclopropyl), 2.46 (dt, 1H, CH2, J 19.3 Hz, 2.0 Hz), 3.24 (dt, 1H, CH2, J 19.3 Hz, 2.3 Hz), 5.53 (dt, 1H, =CH, J 6.5 Hz, 2.0 Hz), 5.70 (dt, 1H, =CH, J 6.5 Hz, 2.3 Hz), 6.77–6.83 (m, 1H, Ph), 7.05–7.19 (m, 3H, Ph), 7.23–7.27 (m, 9H, Ph), 7.43–7.47 (m, 2H, Ph). 13C NMR (62.9 MHz, CDCl3) d: 12.79, 14.61, 39.38 (CH2), 125.61, 127.54, 128.15, 128.52, 128.67, 128.75 (2CH), 113.23, 117.93, 125.09, 125.87, 136.28 (CH), 38.96, 94.09, 129.07, 139.02, 142.04, 145.60 (C).HRMS (EI, 70 eV) m/z: 377.1891 [M]+. For 4: mp 136–138 °C (decomp.) (hexane–ether), Rf 0.51. 1H NMR (250 MHz, CDCl3) d: 0.08–0.12 (m, 2H, cyclopropyl), 0.40–0.55 (m, 2H, cyclopropyl), 0.65–0.88 (m, 2H, cyclopropyl), 1.41–1.43 (m, 2H, cyclopropyl), 1.71–1.80 (m, 1H, cyclopropyl), 6.59 (d, 2H, J 8.0 Hz), 6.72 (t, 1H, J 7.8 Hz), 7.02 (t, 2H, J 7.8 Hz), 7.26–7.88 (m, 7H, Ph), 8.04 (dd, 2H, Ph, J 7.2 Hz, 1.8 Hz). 13C NMR (75.5 MHz, 100 °C, C2D2Cl4) d: 2.82, 15.11 (2CH2), 127.70 (4CH), 112.43, 128.57 (2CH), 12.51, 119.84, 126.81, 128.35, 129.33, 129.64, 130.04 (CH), 27.62, 134.52, 137.15, 144.40, 145.56, 159.07, 169.51 (C).HRMS (EI, 70 eV) m/z: 377.1891 [M]+. For 5: Rf 0.40. 1H NMR (250 MHz, CDCl3) d: 0.62–0.66 (m, 4H, cyclopropyl), 0.92–1.02 (m, 4H, cyclopropyl), 5.95 (s, 1H, CH), 6.95– 7.52 (m, 12H, Ph), 7.85 (d, 2H, Ph, J 7.5 Hz). HRMS (EI, 70 eV) m/z: 377.1891 [M]+. For 6: 1H NMR (250 MHz, CDCl3) d: 0.12–0.15 (m, 2H, cyclopropyl), 0.33–0.39 (m, 2H, cyclopropyl), 0.90–0.99 (m, 2H, cyclopropyl), 1.04– 1.10 (m, 2H, cyclopropyl), 1.38–1.46 (m, 1H, cyclopropyl), 6.09 (s, 1H, NH), 6.33 (d, 1H, J 7.5 Hz), 6.90–7.53 (m, 11H, Ph), 8.24 (d, 2H, Ph, J 7.5 Hz, 1.8 Hz), 11.37 (s, 1H, NH).MS (EI, 70 eV) m/z: 395 [M]+. Compound 7 is known,4 Rf 0.23. probably, the nucleophilic carbene 14 first attacks the double bond in 2 to give the 1,3-zwitterion 8 which may be in an equilibrium with the ring-closed form, the dispiro[2.0.2.1]- heptane derivative 11.For some reason, possibly due to considerable ring strain inherent in the sterically congested skeleton, 11 must be unstable under the employed conditions (100 °C)§ and prefer to open the central ring either back to 8 or with the reverse polarity to give the 1,3-zwitterion 10. The latter can close a six-membered ring by electrophilic attack of the cationic end on one of the vicinal phenyl groups to give the product 5.The triazaspiro[4.4]octadiene 3 can only arise by ring closure of a 1,5-zwitterion like 9 which must have formed from 8 by opening of the anionic cyclopropyl group going along with a 1,2-hydrogen shift (Scheme 2). The formation of the benzotriazepine derivative 4 is particularly obscure as the connectivity of the atoms is changed on ‡ Crystal data: some details of the single-crystal X-ray experiments for compounds 3–6 and crystal data are given in Table 1.All the data were collected using MoKa radiation (l = 0.71073 Å) on a ‘Nonius KAPPACCD’ and a ‘SMART-CCD’ diffractometers for compounds 3 and 4–6, respectively. The structures were solved by direct methods and refined by full-matrix least-square against F2 with anisotropic displacement parameters for all non-hydrogen atoms.Hydrogen atoms in molecules 3–5 were located in the difference Fourier maps and refined isotropically. For compound 6 the positions of H atoms were calculated. For all compounds the maximum features on the final residual maps do not exceed 0.3 e/Å3. Full lists of bond angles, bons lengths, atomic coordinates and thermal parameters have been deposited at the Cambridge Crystallographic Data Centre (CCDC). For details, see ‘Notice to Authors’, Mendeleev Communications, 1999, Issue 1.Any request to the CCDC for data should quote the full literature citation and the reference number 1135/36. § Essentially the same distribution of products 3–6, yet with lower total yield (47%), was observed when carbene 1 was exposed to bicyclopropylidene 2 in THF solution under a pressure of 10 kbar at 20 °C for 24 h.Under the same conditions, but at ambient pressure, only 7% conversion of 1 to 3–6 was observed. N N N Ph Ph Ph toluene 100 ºC, 3 h N N N Ph Ph Ph N N N Ph Ph N N N Ph Ph HNPh Ph N N O H N N N Ph Ph Ph O 1 2 3 (12%) 4 (23%) 5 (12%) 6 (13%) 7 (19%) Scheme 1Mendeleev Communications Electronic Version, Issue 1, 1999 (pp. 1–44) Table 1 Crystal data for compounds 3–6. 3 4 5 6 Chemical formula C26H23N3 C26H23N3 C26H23N3 C26H25N3O Formula weight 377.47 377.47 377.47 395.49 T/K 100 150 120 150 Crystal system monoclinic triclinic triclinic monoclinic Space group P21/n P1 P1 P21 /c Z 4 2 2 4 a/Å 14.607(2) 9.583(1) 8.870(1) 10.715(1) b/Å 9.031(1) 10.053(1) 10.509(1) 19.578(1) c/Å 15.936(2) 12.805(1) 11.889(1) 10.415(1) a/° 90 98.12(1) 78.24(1) 90 b/° 110.05(1) 105.96(1) 70.70(1) 105.43(1) g/° 90 116.99(1) 72.67(1) 90 V/Å3 1974.9(5) 1004.6(1) 991.6(1) 2106.2(1) Dc/g cm–3 1.270 1.248 1.264 1.247 m/mm–1 0.075 0.074 0.075 0.077 Reflections measured 7471 8406 9014 11941 Unique reflections 3872 5241 4506 2748 R1 (I = 2s) 0.0575 0.0553 0.0863 0.0973 wR2 0.1654 0.1305 0.2299 0.1643 GOOF 0.990 1.076 0.972 1.196 3 4 5 6 C(1) C(2) C(3) C(4) C(5) C(6) C(7) C(8) C(9) C(10) C(11) C(12) C(13) C(14) C(15) C(16) C(17) C(18) C(19) C(20) C(21) C(22) C(23) C(24) C(25) C(26) Figure 1 Structures of compounds 3–6 in the crystals.N(1) N(2) N(3) C(1) C(2) C(3) C(4) C(5) C(6) C(7) C(8) C(9) C(10) C(11) C(12) C(13) C(14) C(15) C(16) C(17) C(18) C(19) C(20) C(21) C(22) C(23) C(24) C(25) C(26) N(1) N(2) N(3) C(1) C(2) C(3) C(4) C(5) C(6) C(7) C(8) C(9) C(10) C(11) C(12) C(13) C(14) C(15) C(16) C(17) C(18) C(19) C(20) C(21) C(22) C(23) C(24) C(25) C(26) N(1) N(2) N(3) C(1) C(2) C(3) C(4) C(5) C(6) C(7) C(8) C(9) C(10) C(11) C(12) C(13) C(14) C(15) C(16) C(17) C(18) C(19) C(20) C(21) C(22) C(23) C(24) C(25) C(26) N(1) N(2) N(3) O(1) N N N Ph Ph Ph N N N Ph Ph Ph N N N Ph Ph 1 2 Scheme 2 N N N Ph Ph Ph 8 N N N Ph Ph Ph 9 10 N N N Ph Ph Ph 11 3Mendeleev Communications Electronic Version, Issue 1, 1999 (pp. 1–44) going from 8 or 10 to 4.Formally, this could be brought about by opening of the five-membered heterocycle in 10 between the two adjacent nitrogens, a subsequent 1,3-shift of a phenyl group from the central to the terminal nitrogen and ring closure by intramolecular nucleophilic aromatic substitution.Similarly difficult to explain is the formation of compound 6. Formally, an intramolecular electrophilically assisted nucleophilic aromatic substitution in 10 could lead to a tricyclic benzazepine derivative which due to its ring strain might undergo hydrolysis to give 6 during column chromatography.Without any further evidence, all these mechanistic considerations, especially the last ones, are highly speculative. None the less, the observed reactivity of the stable carbene 1, which so far has been reported to react only with acceptor-activated C=C double bonds,4 towards the strained tetrasubstituted alkene 2 is quite remarkable, and so are the products 3–6.This work was supported by the Fonds der Chemischen Industrie, and the Engineering and Physical Science Research Council (EPSRC, UK). We are grateful to the companies BASF AG, Bayer AG and Hüls AG for generous gifts of chemicals, to Dipl.-Chem. R. Machinek for recording the high-temperature 13C NMR spectra and to Dr.B. Knieriem for his careful reading of the manuscript. We are very thankful to Dr. J. Steed (Imperial College, London) for the opportunity to use the Nonius KAPPACCD X-ray diffractometer and for valuable help in processing of the collected data. References 1 S. Braese and A. de Meijere, Angew. Chem., 1995, 107, 2741 (Angew. Chem., Int. Ed. Engl., 1995, 34, 2545) and references cited therein. 2 A. de Meijere, S. I. Kozhushkov and A. F. Khlebnikov, Zh. Org. Khim., 1996, 32, 1607 (Russ J. Org. Chem., 1996, 32, 1555). 3 A. de Meijere, S. I. Kozhushkov, D. S. Yufit, R. Boese, T. Haumann, D. L. Pole, P. K. Sharma and J. Warkentin, Liebigs Ann. Chem., 1996, 601. 4 D. Enders, K. Breuer, J. Runsink and J. H. Teles, Liebigs Ann. Chem., 1996, 2019. 5 A. de Meijere, S. I. Kozhushkov, T. Spaeth and N. S. Zefirov, J. Org. Chem., 1993, 58, 502. Received: Cambridge, 26th October 1998 Moscow, 11th November 1998; Com. 8/08251K

ISSN:0959-9436     

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 1, 1999 (pp. 1–44) Fast crystallisation of amorphous water during the growth of a condensate film Mikhail Yu. Efremov, Aleksandr F. Batsulin and Gleb B. Sergeev* Department of Chemistry, M. V. Lomonosov Moscow State University, 119899 Moscow, Russian Federation. Fax: +7 095 939 0283; e-mail: gbs@cryo.chem.msu.su Fast spontaneous crystallisation of amorphous water in the course of the growth of a condensate film at 80 K was detected after the film thickness reached a constant critical value. Low-temperature condensates are films produced by condensation of vapours onto a cold surface.This process is characterised by an extremely high rate of cooling and can result in a metastable state of the substance and in special properties of the condensate films.Various size effects are typical of such objects. In this work, we describe fast crystallisation of an amorphous phase during the growth of a water condensate film. This effect occurs when the growing film thickness increases to a particular value. The experiments were carried out using an original lowtemperature thin-film differential scanning calorimeter. The reactor of the setup was evacuated with oil diffusion and cryogetter pumps to a vacuum of 0.1 Pa or lower.The measuring unit was a massive copper cylinder (thermostat) with two soldered copper plates (sample and reference plates) placed inside of the reactor. A differential thermocouple and microheaters were arranged on the plates. Thermal effects in the film sample caused a difference between the plate temperatures, which was compensated by the microheaters. The heat effects can be calculated from the voltage and the current in the heater.The whole measuring unit can be warmed up by the heater placed on the copper cylinder. The calorimeter was controlled by a computer. The operation precision in the programmed heating was ±0.06 or ±0.02 K at 80 or 300 K, respectively, when the heating rate was 3 K min–1.The reproducibility of calorimetric curves in an idle experiment during heating from 80 to 300 K at a rate of 3 K min–1 with an integration step of 1 K was as follows: the mean variation of heat power less than ±0.1 mW, the maximum variation ±0.3 mW. The absolute error of the temperature measurement during the heating at a rate of 3 K min–1 was within 3 K (the real temperature was always lower than the measured value) for the melting of 1-butanol condensate.The preparation of low-temperature condensate films with predefined shapes under isothermic conditions directly in the calorimetric cell and the performance of calorimetric experiments in a vacuum are important features of this calorimeter.The calorimeter was described elsewhere.1,2 Typical experiment includes three stages: i, cooling of the measuring unit to 80 K with liquid nitrogen; ii, condensation of the amorphous water film in the molecular beam mode onto the sample plate (film thickness from 0.5 to 470 mm, condensation rate 6×1014–5×1016 cm–2 s–1 or 0.01–1 mmmin–1) and iii, warming of the measuring unit at a rate of 3 K min–1.Spontaneous fast (several seconds) release of heat was observed when the water film thickness increased to a fixed (critical) value. The sharp exothermic effect was followed by a significant decrease of heat generation in the condensate. A typical calorimetric curve is shown in Figure 1. A repeated fast process was observed after the first one during the continuous condensation in some experiments.The second process was less pronounced. The film undergone the fast process might partially fall off the support during the sample heating after the condensation being completed. The critical thickness of fast process was equal to 4–6 mm for the films prepared at a condensation rate of 1 mm min–1. The critical thickness depends only slightly on the condensation rate: it increased no more than twice when the rate of condensation decreased by a factor of 50.The excess of the water film surface temperature was estimated. This value is lower than 4 K for the films of critical thickness. The heat release during the fast process varied in the range 0.2–0.7 kJ mol–1. These values are underestimated, because a portion of the released heat was masked by a subsequent decrease of background heat generation in the film.The background heat occurred due to exothermic condensation and absorption of environmental heat radiation. The condensation power is constant during the film formation; thus, it should be supposed that the decrease of the background heat generation was caused by changes in the optical properties of the film, namely, an increase of the transparency or reflectivity.In our opinion, such a significant heat release during the fast process may be caused only by the first-order phase transition in the condensate, namely, the crystallisation of amorphous water. The crystallisation of amorphous water films with the thickness lower than the critical value was observed during the sample heating after the film formation completed.The scanning rate was 3 K min–1. The slow crystallisation (near 1 min) of the amorphous film was observed in the temperature range 163– 167 K. The determined heat of crystallisation was equal to 1.2±0.1 kJ mol–1 (the given accuracy was calculated on the basis of 15 experiments for the confidence level 0.95). This value satisfactorily agrees with the published data, for example, 1.330±0.02 kJ mol–1.3 Thus, the heat of crystallisation of amorphous water was sufficient to provide the fast heat release during the film formation.On the other hand, the heat of the slow crystallisation, which occurred in the samples undergone the fast process during the formation, is significantly lower. The heat of the slow crystallisation in these experiments was 0.3–0.8 kJ mol–1.Thus, it should be concluded that the sharp heat release during the water film formation is caused by the spontaneous highvelocity crystallisation of the amorphous condensate which propagated through a considerable part of the sample. A similar process during the growth of amorphous condensates of Sb, Bi and some metals, so-called spontaneous explosive crystallisation, has been observed earlier.4 Furthermore, spontaneous explosive 0.09 0.08 0.07 0.06 0.05 0.04 0.03 260 270 280 290 300 310 320 Power/W t/s exothermic endothermic Figure 1 Dependence of the power of heat effects on the time of water condensation (a region of the fast exothermic process).The condensation rate is 0.9 mm min–1.Mendeleev Communications Electronic Version, Issue 1, 1999 (pp. 1–44) chemical reactions in the co-condensates of reagent vapours during their formation are well known.5 Previously, it has been demonstrated that spontaneous explosive reactions in the chemically active co-condensates can be initiated by the crack formation in the samples.6 As a rule, condensate growth was accompanied by the generation of an internal mechanical stress in the film.This stress causes the crack destruction of the condensate film with the thickness which is equal to or greater than the critical value. This fact can initiate a chemical reaction or crystallisation propagating in the autowave mode.7 The applicability of the above model to the effect described here was tested as described below.The same amorphous water films were prepared in the cryotensimetric setup described earlier.8 This setup allows the measurement the internal mechanical stress in growing low-temperature condensate films. The conditions of sample preparation in the calorimetric and tensimetric setups are comparable. The typical dependence of the internal mechanical stress on the thickness of a growing film is presented in Figure 2.A significant decrease of the internal stress was observed when the condensate film attained the critical thickness. At this time, the formation of a crack network was detected on the film. The critical thickness determined by the heat release and that one by the crack network generation differs from each other less than 2 times.Such difference may be caused either by an error in the absolute thickness of the condensate film, determined in the tensimeter (–50%+100%) or by the difference in the shapes of the condensate films produced in the mentioned setups. Thus, it should be concluded that the high internal mechanical stress causes the film degradation by cracks when the thickness of the growing water condensate reaches the critical value.The crack generation may, in turn, initiates the crystallisation of the amorphous film. It is reasonable to suppose that both of these processes (degradation and crystallisation) are interconnected by a positive feedback to generate autowave explosive crystallisation of the amorphous water condensate. This work was supported by the Russian Foundation for Basic Research (grant no. 96-03-33970a). References 1 M. Yu. Efremov, A. F. Batsulin and G. B. Sergeev, Vestn. Mosk. Univ., Ser. 2: Khim., 1999, 40 (3), 180 (in Russian). 2 G. B. Sergeev, M. Yu. Efremov and A. F. Batsulin, Proceedings of the European Conference on Thin Organised Films, Germany, 1998, p. 332. 3 A. Hallbrucker and E. J. Mayer, Phys. Chem., 1987, 91, 503. 4 V. A. Schklovskii and V. M. Kuz’menko, Usp. Fiz. Nauk, 1989, 157, 311 (Physics-Uspekhi, 1989, 32, 163). 5 V. S. Komarov, M. Yu. Efremov and G. B. Sergeev, Mol. Cryst. Liq. Cryst., 1992, 211, 445. 6 M. Yu. Efremov, V. S. Komarov and G. B. Sergeev, Mol. Cryst. Liq. Cryst., 1994, 248, 111. 7 V. V. Barelko, I. M. Barkalov, V. I. Gol’danskii, A. M. Zanin and D. P. Kiryukhin, Usp. Khim., 1990, 59, 353 (Russ. Chem. Rev., 1990, 59, 205). 8 M. Yu. Efremov, PhD Thesis, Moscow State University, Moscow, 1995 (in Russian). 12 10 8 6 4 2 0 2 4 6 8 10 12 14 16 18 20 Internal stress/MPa Film thickness/mm Figure 2 Dependence of the internal stress in the amorphous water condensate on the film thickness during the condensation. The condensation rate is 0.4 mmmin–1. Received: Moscow, 20th May 1998 Cambridge, 8th June 1998; Com. 8/04238A

ISSN:0959-9436     

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 1, 1999 (pp. 1–44) X-ray analysis of the structure of an assembly of cationic aminomethylated calix[4]resorcinarene and the zinc tetrachloride anion Aidar T. Gubaidullin, Yuliya E. Morozova, Asiya R. Mustafina, Ella Kh. Kazakova, Igor A. Litvinov and Alexander I. Konovalov* A. E. Arbuzov Institute of Organic and Physical Chemistry, Kazan Scientific Centre of the Russian Academy of Sciences, 420083 Kazan, Russian Federation.Fax: +7 8432 752 253; e-mail: aidar@iopc.kcn.ru The first X-ray crystallographic data on an ionic host–guest self-assembling system with water-soluble tetraprotonated macrocyclic 2,4,14,20-tetramethyl-5,11,17,23-(N,N-dimethylamino)methylenecalix[4]resorcinarene as a host and zinc tetrachloride as a guest are presented.Because of their high flexibility, host calixarene molecules are affected by the nearest environment, and this fact gives new unexpected opportunities for coordination chemistry. Metal complexes with host molecules can be effectively used for making self-assembling systems and offer the opportunity to compare the conformational modifications of the macrocycles induced by the complexation.Few crystal structures of actinide complexes with deprotonated calixarenes have been reported to date. The complexes described are the inclusion compounds of thorium in p-tert-butylcalix[8]arene1 and of the uranyl ion in p-tert-butylcalix[5]arene.2 In the latter case, one protonated triethylamine molecule is located inside and the other, outside the calixarene cavity. Lanthanide elements form complexes with water-soluble p-sulfonatocalix[5]arene; the crystal structure includes pyridine N-oxide and water as guest species.3 The transition metal complexes of p-sulfonatocalix[5]arene also include 2-N,N-dimethylacetamide and 15 water molecules as guest species.4 In the last two cases, an interesting clay-like bilayer supramolecular structure of solid compounds was discussed.In all of these examples, calixarenes occurred as anions. Heretofore only five X-ray entries for calix[4]resorcinarene with free OH groups5–9 and no data on their metal complexes are included in the Cambridge Crystallographic Database.10 Here we report the crystal and molecular structure of the ionic compound 2(C44H64O8N4)4+(ZnCl4)2–·6Cl–·4H2O, which was obtained from aminomethylated calix[4]resorcinarene and zinc chloride in an aqueous HCl solution.† This is the first example of X-ray data for a metal complex in which a calixarene exists in the protonated cationic form.The calixarene molecule has a twofold axis of symmetry and lies on the C2 symmetry axis of the crystal. The Zn atom is located at the intersection of two twofold axes.The asymmetric unit contains only one half of the macrocycle, one quarter of the anion [ZnCl4]2–, a Cl– anion and a water molecule. Water molecules are disordered (Figure 1). The anion [ZnCl4]2– is statistically disordered, as can be seen from the amplitudes and forms of thermal parameters of the Cl atom. The calixarene † The crystals exhibited extreme sensitivity to solvent losses and were protected in a glass capillary during the collection of X-ray diffraction data with an ‘Enraf-Nonius CAD4’ four-circle diffractometer (graphite monochromator, CuKa radiation, w/2q scan, q £ 74.3°).Twenty-fivecentered reflections gave a refined tetragonal unit cell of the dimensions a = 14.894(4) Å, c = 23.395(6) Å, V = 5190.0(2) Å3, Z = 4, r = 1.37 g cm–3.A total of 6027 reflections were measured, of which 4032 were unique with I > 3s(I). The structure was solved in the uniquely assignable space group P-4n2 by direct methods using SIR11 and difference Fourier syntheses. All non-hydrogen atoms were refined anisotropically, while the thermal and positional parameters of hydrogen atoms were fixed. The final R values were R = 0.054, Rw = 0.064 for 3633 unique reflections with F2� 3s(I).All calculations were carried out on a ‘DEC Alpha Station 200’ computer with the MolEN12 system. Atomic coordinates, bond lengths, bond angles and thermal parameters have been deposited at the Cambridge Crystallographic Data Centre (CCDC). For details, see ‘Notice to Authors’, Mendeleev Commun., Issue 1, 1999.Any request to the CCDC for data should quote the full literature citation and the reference number 1135/35. molecule exhibits a 1,3-diplanar shape with the two opposite phenolic rings bearing the dimethylamino groups almost parallel to one another. The least squares plane formed by the methyne carbon atoms C(1a)–C(1)–C(8a)–C(8) is generally chosen as a reference plane for the molecule.The dihedral angle between Figure 1 The crystal structure of the complex 2(C44H64O8N4)4+ (ZnCl4)2–· 6Cl–·4H2O. Dashed lines indicate hydrogen bonds. Zn(1) Cl(4) N(22) C(21) O(12) O(10) Cl(3) O(1) N(18) O(5) O(3) H(8) Cl(2) O(12a) C(1a) C(8a) C(1) A B (a) (b) Figure 2 (a) A fragment of molecular packing and (b) arrangement of sticks in the crystal.Mendeleev Communications Electronic Version, Issue 1, 1999 (pp. 1–44) opposite resorcinolic rings (A) lying in the least squares plane is 171.3(4)°. The other pair of resorcinolic rings (B) perpendicular to the least squares plane constitute the dihedral angle of 31.2(2)° with each other. The neighbouring resorcinolic rings make angles 70.5(2)° with each other. The dimethylamino groups of rings (B) together with OH groups form hydrophilic host cavities.The N(18)–H(18)···O(3) 2.35(3) Å (A rings), N(22)–H(22)··· O(10) 2.57(3) Å (B rings) distances suggest the formation of intramolecular hydrogen bonds between the amino protons and one of the hydroxyl oxygens of the resorcinol moiety (Figure 1). The remaining hydroxyl protons also take part in hydrogen bonding with the bridging methyne hydrogen C(8)–H(8)···O(5) 2.53(4) Å (A rings) and with the methylene hydrogen of CH2NH+Me2 groups C(21)–H(211)···O(12) 2.44(5) Å (B rings).The two Cl– anions participate in the intermolecular hydrogen bonding with resorcinol groups: O(3)–H(3)···Cl(2) 2.24(4) Å (A rings), O(12)–H(12)···Cl(2) 2.47(4) Å (B rings), O(5)–H(5)···Cl(3) 2.50(6) Å (A rings), O(10)–H(10)···Cl(3) 2.77(5) Å (B rings).The chlorine anions are also stabilized by additional weak interactions with the bridging methylene hydrogen C(1)–H(1)···Cl(2) 2.65(4) Å and the protons at the nitrogen atoms N(18)–H(18)··· Cl(3) 2.63(3) Å of the A rings. Protons at the nitrogen atom also interact with a water molecule N(22)–H(22)···O(1) 1.89(4) Å. The crystal superstucture is rather interesting (Figure 2).Two calixarene molecules are faced each other with their planar resorcinol A rings to form lipophilic regions [Figure 2(a)]. On the other hand the resorcinol B rings form a supercavity, which is large enough to encompass the anion [ZnCl4]2–. This motif is organized into infinite one-dimensional sticks, which are surrounded by a hydrophilic layer that consists of NH+Me2 and OH groups, Cl– and H2O, which form a solvate shell.The coplanar sticks are packed to form a monolayer (Figure 3). Each subsequent monolayer is swung 90° to the preceding layer [Figure 2(b)]. This is the reason why all guests have a lipophilic covering. One Cl– anion is placed on the C2 axis and forms an anion pseudochannel. Hydrogen bonding forms a three-dimensional network; the directions of hydrogen bonds do not coincide with the directions of sticks and monolayers.Thus, protonated aminomethylated calixarenes can serve as cationic host ligands for ionic self-assembling systems. In contrast to the complexes described in refs. 3 and 4 the aminomethylated calixarene fragments are not involved the first coordination sphere of the metal ion. Electrostatic forces and the size of an anion are important in this case.This work was supported by the Russian Foundation for Basic Research (grant no. 98-03-33051). References 1 J. M. Harrowfield, M. I. Ogden and A. H. White, J. Chem. Soc., Dalton Trans., 1991, 979. 2 P. Thuery and M. Nierlich, J. Incl. Phenom., 1997, 27, 13. 3 J. W. Steed, C. P. Johnson, C. L. Barnes, R. K. Juneja, J. L.Atwood, S. Reilly, R. L. Hollis, P. H. Smith and D. L. Clark, J. Am. Chem. Soc., 1995, 117, 11426. 4 C. P. Johnson, J. Athwood, J. W. Steed, C. B. Bauer and R. D. Rogers, Inorg. Chem., 1996, 35, 2602. 5 L. M. Tunstad, J. A. Tucker, E. Dalcanale, J. Weiser, J. A. Bryant, C. Sherman, R. C. Helgeson, C. B. Knobler and D. J. Cram, J. Org. Chem., 1989, 54, 1305. 6 G. Zahn, J. Sieler, K. Muller, L. Hennig and G. Mann, Z. Kristallogr., 1994, 209, 468. 7 H. Adams, F. Davis and C. J. M. Stirling, J. Chem. Soc., Chem. Commun., 1994, 2527. 8 T. Lippman, H. Wilde, M. Pink, A. Schafer, M. Hesse and G. Mann, Angew. Chem., Int. Ed. Engl., 1993, 32,1195. 9 D. A. Leigh, P. Linnane, R. G. Pritchard and G. Jackson, J. Chem. Soc., Chem. Commun., 1993, 389. 10 Cambridge Structural Database System. Version 5.14, November, 1997. 11 A. Altomare, G. Cascarano, C. Giacovazzo and D. Viterbo, Acta Crystallogr., Sect. A, 1991, 47, 744. 12 L. H. Straver and A. J. Schierbeek, MolEN. Structure Determination System. Program Description, Nonius B.V., 1994, vol. 1. Figure 3 The monolayer arrangement in the crystal. View along the c axis. Received: Moscow, 6th May 1998 Cambridge, 23rd July 1998; Com. 8/03645D

ISSN:0959-9436     

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 1, 1999 (pp. 1–44) Influence of the guest molecular size on the thermodynamic parameters of host–guest complexes between solid tert-butylcalix[4]arene and vapours of organic compounds Valery V. Gorbatchuk,* Alexey G. Tsifarkin, Igor S. Antipin, Boris N. Solomonov and Alexander I. Konovalov Department of Chemistry, Kazan State University, 420008 Kazan, Russian Federation.E-mail: valery.gorbatchuk@ksu.ru The shape of guest molecules has a significantly greater influence on the free energy of supramolecular effect in the formation of solid complexes of tert-butylcalix[4]arene than on their stoichiometry. Molecular recognition of ion species and small neutral molecules by calix[4]arenes has been intensively studied.1–4 Some information is currently available on the stoichiometry of calix[4]arene complexes with organic guests5–8 and the free energy of formation of these complexes in solution3,4 but a detailed structural understanding remains lacking.This provides an impetus to search for macroscopic structural parameters that may be used for describing and predicting the thermodynamic characteristics of host–guest complexes of calix[4]arene derivatives.In this paper we report the structural trends in the vapour guest binding by solid tert-butylcalix[4]arene 1. To examine the influence of the guest molecular structure on the molecular recognition thermodynamics the sorption isotherms of 12 organic compounds with solid 1 were determined. The sorption isotherms were obtained by the method of gas chromatographic headspace analysis described earlier.9,10 The host 111 was purified from volatile impurities by heating at 190–210 °C during 3–4 h in a vacuum (100 Pa).In the presence of guests no additional chromatographic peaks were observed in the headspace over purified 1. Sorbate purity tested by GLC was > 99.5%. The isotherms were corrected referred to the sorbate loss on equilibration which was estimated in blank experiments without solid host. This loss is equal to 0.5–2% of the total sorbate amount in the vial for most guests studied at P/P0 > 0.3 depending on their volatility.Maximum losses are 3% for cyclohexane and 8% for n-hexane. The accuracy of sorbate activity determination lies in the interval from 5% (for P/P0 > 0.5) to 10% (for P/P0 < 0.1).Examples of sorption isotherms obtained are presented in Figure 1. The sorption isotherms of benzene, toluene, acetonitrile and ethanol have been described earlier.9 The limiting activity coefficients of the guests in toluene solution g• were determined with a 10% precision by the same experimental method of headspace analysis10 for infinitely dilute solutions (0.2 vol.% for alcohols and 1 vol.% for other solutes).The absence of a concentration dependence for the g• values was checked in each case. The isotherms obtained have a guest threshold activity athr needed for incorporation of guest in solid cavitand. Up to this activity weak binding of the guests by solid 1 is observed. Above the guest threshold activity athr the guest sorption significantly increases, which indicates the formation of a stable inclusion complex.Analogous threshold phenomena were observed previously for the formation of solid host–guest complexes.12–15 At a guest activity 0.6 < P/P0 < 0.8 the saturation of the isotherms obtained takes place. The methanol isotherm with a very large threshold activity reaches saturation at the upper end of the activity interval studied.The values of the guest threshold activity athr and the guest–host molar ratio Ssat in the solid phase at the sorption saturation are summarized in Table 1. The value of athr is given as the guest activity at solid phase composition 0.25Ssat. It may be seen (Table 1) that the saturation is achieved at sorbate/cavitand molar ratio in the solid phase corresponding to 2:1, 1:1 or 1:2 complex formation.Comparison with the available X-ray data for solid calixarene 1 complexes (column 7) indicates a good agreement of complex stoichiometry and sorbate/cavitand saturation ratio Ssat. Thus, the last value (Ssat) can be used for the estimation of host–guest complex stoichiometry. The sorption isotherms observed cannot be described by monolayer (Langmuir) or multilayer (BET) adsorption models.For a description of such sorption behaviour the Hill equation (1) was used.17 MeOH Pinacolone n-Nonane 2.00 1.50 1.00 0.50 0.00 YS (mole of guest/mole of host) Guest activity (P/P0) 0.50 1.00 Figure 1 Vapour sorption isotherms for various guests on solid tert-butylcalix[ 4]arene at 298 K.The lines correspond to the isotherms calculated by the Hill equation (1). 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 2.50 2.00 1.50 1.00 0.50 0.00 30.0 80.0 130.0 180.0 S (mole of guest/mole of host) Molar volume VM/cm3 mol–1 Figure 2 Correlation between the stoichiometry S of saturated solid host– guest complexes of tert-butylcalix[4]arene and the molar volume VM of the guests. Point numbers correspond to the number of guests in Table 1.Empty points are S values obtained in this work. Filled points are X-ray data.5–8 YS = SC(P/P0)N 1 + C(P/P0)N (1)Mendeleev Communications Electronic Version, Issue 1, 1999 (pp. 1–44) In equation (1), Y is the complex saturation degree, S is the stoichiometry, C is the sorption constant, N is the cooperativity constant and YS is the experimentally determined solid phase composition (mole of guest per mole of host).Approximation of the sorption isotherms by the Hill equation gives two important characteristics of complex formation: the stoichiometry S and the ratio (ln C)/N. RT(ln C)/N is the total binding free energy DGc of the guest by solid host when ln (P/P0) is given by equation (1) as a function of Y: Term DGc is equal to the transfer free energy of 1 mole of guest from the standard state of pure liquid to the saturated solid host–guest complex at the guest activity P/P0 = 1 and corresponds to a complex formation free energy calculated for one mole of the guest.The values S (mole guest/mole host), ln C/N and DGc are listed in Table 1. Clearly there is agreement between the stoichiometry values obtained from the sorbate/ cavitand saturation ratio Ssat, the Hill equation and X-ray experiments.Analysis of the obtained data shows the dependence of the complex stoichiometry on the guest molecular size. Small molecules (like methanol) form 2:1 complexes whereas larger molecules form 1:1 or 1:2 complexes. The molar volume VM and molar refraction MRD are often considered as molecular size parameters.We tried to correlate the complex stoichiometry with both of them. Molar refraction was calculated by the Lorenz–Lorentz equation. The correlation of S values with the guest molar volume VM (Figure 2) is rather poor. In Figure 3 a plot of S vs. MRD values of the guests is presented. Obviously the dependence is stepwise.There are three definite areas in which 2:1, 1:1 and 1:2 complexes are formed. The stoichiometry changes are observed at 9–11 cm3 mol–1 (from 2:1 to 1:1 complex) and at 30–32 cm3 mol–1 (from 1:1 to 1:2 complex). This correlation includes linear, branched, cyclic aliphatic and aromatic compounds. It means that the shape of the guest molecule has hardly any effect on the complex stoichiometry.So, the dependence presented in Figure 3 allows us to predict the stoichiometry of solid calixarene 1 complexes with different organic guests. The specific features of the guest molecular structure may be essential within the interval of MRD where the stoichiometry changes take place. For example, guests with a compact molecular structure such as tert-butyl acetate (MRD = 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 S (mole of guest/mole of host) 2.50 2.00 1.50 1.00 0.50 0.00 5.0 15.0 25.0 35.0 45.0 Molar refraction MRD/cm3 mol–1 Figure 3 Correlation between the stoichiometry S of saturated solid host– guest complexes of tert-butylcalix[4]arene and the molar refraction MRD of the guests.Point numbers correspond to the numbers of guests in Table 1.Empty points are S values obtained in this work. Filled points are X-ray data.5–8 DGc = RTòln (P/P0)dY = –RT(ln C)/N 0 1 (2) aThe estimated error of athr is ±10%, the error of Ssat and S is ±5%; the error of DGc is ±0.4 kJ mol–1. bD(YS) is the solid phase interval composition YS for which S and (ln C)/N were calculated from equation (1). cd is the standard deviation of the approximation in the interval D(YS) for the shortest distances between experimental points and calculated line: d = (S{[(P/P0)calc – (P/P0)exp]2 + [(YS)calc – (YS)exp]2}/(n – 2))1/2.dData from ref. 16. eData from ref. 8. fData from ref. 6. gData from ref. 5. hData from ref. 7. Table 1 Sorption isotherm parameters for vapour guests on solid tert-butylcalix[4]arene at 298 K.a Entry Guest MRD/ cm3 mol–1 VM/ cm3 mol–1 athr Ssat S (Slit) (lnC)/N D(YS)b dc DGc/ kJ mol–1 g DGtrans/ kJ mol–1 1 Methanol 8.3 40.5 0.54 1.99 1.91 0.50 0.03–1.89 0.03 –1.2 21.8 –8.9 2 Acetonitrile 11.1 52.4 0.13 1.15 1.17 1.63 0.06–1.20 0.03 –4.0 4.33 –7.7 3 Ethanol 12.9 58.4 0.36 1.13 1.10 0.91 0.17–1.15 0.02 –2.3 17.4; 15.4d –9.3 4 Propionitrile 16.0 71.3 0.09 0.95 0.91 2.08 0.07–0.92 0.01 –5.2 2.92 –7.8 5 n-Propanol 17.5 74.8 0.29 0.93 1.05 1.00 0.06–1.11 0.04 –2.5 15.9 –9.3 6 n-Pentane 25.3 115.3 (1e) 7 Chlorobutane 25.5 104.5 (1e) 8 Benzene 26.2 88.9 0.04 1.07 1.08 (1f) 2.92 0.07–1.09 0.01 –7.3 0.97 –7.2 9 Cyclohexane 27.7 108.0 0.09 1.20 1.20 (1e) 2.20 0.08–1.24 0.02 –5.5 1.36 –6.3 10 n-Hexane 29.9 130.8 0.13 0.50 0.50 1.88 0.00–0.51 0.03 –4.7 1.50 –5.7 11 Pinacolone 30.0 125.0 0.19 1.05 1.07 1.29 0.11–1.01 0.04 –3.2 1.47 –4.2 12 Toluene 31.1 106.3 0.08 0.99 0.99 (1g) 2.24 0.03–1.00 0.02 –5.6 1.00 –5.6 13 tert-Butyl acetate 31.7 134.8 0.20 0.87 0.79 1.27 0.05–0.82 0.03 –3.2 1.40 –4.0 14 Nitrobenzene 32.8 102.9 (1e) 15 Anisole 32.9 108.7 0.06 0.63 0.62 (0.5h) 2.58 0.06–0.65 0.01 –6.4 1.31 –7.1 16 n-Heptane 34.5 146.5 0.12 0.49 0.44 1.94 0.06–0.45 0.01 –4.8 1.68 –6.1 17 o-Xylene 35.9 120.5 0.3 0.55 0.60 1.10 0.05–0.58 0.02 –2.7 1.09 –2.9 18 n-Octane 39.2 162.5 0.14 0.61 0.56 1.72 0.07–0.57 0.01 –4.3 1.69d –5.6 19 n-Nonane 43.8 178.6 0.35 0.52 0.52 0.98 0.07–0.53 0.02 –2.4 1.90 –4.0 1 2 3 4 5 8 9 10 11 12 13 15 16 17 18 19 –2.0 –4.0 –6.0 –8.0 –10.0 –12.0 0.0 10.0 20.0 30.0 40.0 50.0 DGtrans/kJ mol–1 (298 K) Molar refraction MRD/cm3 mol–1 Figure 4 Correlation between the free energy of guest transfer from toluene solution to a saturated solid complex with tert-butylcalix[4]arene DGtrans and molar refraction MRD of guests.Point numbers correspond to the numbers of guests in Table 1.Mendeleev Communications Electronic Version, Issue 1, 1999 (pp. 1–44) = 31.7 cm3 mol–1) and pinacolone (MRD = 30.0 cm3 mol–1) form a 1:1 complex, whereas linear n-hexane (MRD = 29.9 cm3 mol–1) forms a 1:2 complex.Encouraged by the results obtained for the complex stoichiometry, we tried to find an analogous dependence for DGc values. Unfortunately there is no definite correlation between DGc and the MRD or VM values of the guests studied. It may be due to the variation of the energy of molecular interactions in standard states of pure liquid guest that is unequal for the guests with different molecular composition.In order to estimate the difference between the energy of supramolecular host–guest interactions in the solid phase and molecular interactions in solution we determined the free energy of transfer DGtrans of the guest from the solution in toluene at infinite dilution to the saturated solid complex with 1: where g• is the guest limiting activity coefficient in toluene.The values of DGtrans and g• are given in Table 1. Toluene was chosen as a solvent that can model the hydrophobic environment of the guest molecule in a host–guest complex. The value of DGtrans can therefore be assumed to be the estimated free energy of the supramolecular effect distinguishing the host–guest complex from an ordinary solution.This supramolecular effect probably arises from the reduced energy of cavity formation in solid cavitand 1 with bowl-like molecules and may also include the energy of simultaneous conformational changes in neighbouring host molecules, because formation of the studied solid host–guest complexes is a cooperative process.In terms of this model the supramolecular effect for solid cavitand 1 must depend on the size of the guest molecules as well as on complex stoichiometry, since the studied host 1 has an internal molecular cavity of restricted size. The correlation between DGtrans and the guest molar refraction MRD (Figure 4) demonstrates the molecular recognition properties of solid tert-butylcalix[4]arene.There is a certain decrease in the observed supramolecular effect with the growth of the guest molecular size. The sharpest decrease in –DGtrans values of the studied guests is exhibited by branched molecules of pinacolone and tert-butylacetate and by alkyl substitution at the aromatic ring in the series benzene > toluene > o-xylene.For the series of studied n-alkanes only the largest, n-nonane, has a significantly lower –DGtrans value whereas for other alkanes it is approximately constant. Its value does not change within the studied sets of nitriles and alcohols. The reason for the slightly higher –DGtrans values for alcohols than for nitriles is probably due to the ability of alcohols to undergo weak hydrogen bonding with the host 1, which is underestimated by their limiting activity coefficients in toluene solution.The shape and the size of the guest molecules therefore has a significantly greater influence on the free energy of the supramolecular effect in the formation of solid complexes of tert-butylcalix[4]arene than on their stoichiometry. Support from the Russian Foundation of Basic Research (grant no. 98-03-32105a) is gratefully acknowledged. References 1 A. Ikeda and S. Shinkai, Chem. Rev., 1997, 97, 1713. 2 I. S. Antipin, I. I. Stoikov, E. M. Pinkhassik, N. A. Fitseva, I. Stibor and A. I. Konovalov, Tetrahedron Lett., 1997, 38, 5865. 3 A. Arduini, W. M. McGregor, D. Paganuzzi, A. Pochini, A. Secchi, F. Ugozzoli and R. Ungaro, J. Chem. Soc., Perkin Trans. 2, 1996, 839. 4 S. Smirnov, V. Sidorov, E. Pinkhassik, J. Havlichek and I. Stibor, Supramol. Chem., 1997, 8, 187. 5 G. D. Andreetti, R. Ungaro and A. Pochini, J. Chem. Soc., Chem. Commun., 1979, 1005. 6 R. Ungaro, A. Pochini, G. D. Andreetti and P. Domiano, J. Chem. Soc., Perkin Trans. 2, 1985, 197. 7 R. Ungaro, A. Pochini, G. D. Andreetti and V. Sangermano, J. Chem. Soc., Perkin Trans. 2, 1984, 1979. 8 E. B. Brouwer, J. A. Ripmeester and G. D. Enright, J. Inclusion Phenom. Mol. Recogn., 1996, 24, 1. 9 V. V. Gorbatchuk, I. S. Antipin, A. G. Tsifarkin, B. N. Solomonov and A. I. Konovalov, Mendeleev Commun., 1997, 215. 10 V. V. Gorbatchuk, S. A. Smirnov, B. N. Solomonov and A. I. Konovalov, Zh. Obshch. Khim., 1990, 60, 1200 [J. Gen. Chem. USSR (Engl. Transl.), 1990, 60, 1069]. 11 C. D. Gutsche and M. Iqbal, Org. Syn., 1989, 68, 234. 12 F. L. Dickert and M. Keppler, Adv. Mater., 1995, 7, 1020. 13 F. L. Dickert, M. Keppler, G. K. Zwissler and E. Obermeier, Ber. Bunsenges. Phys. Chem., 1996, 100, 1312. 14 L. J. Barbour, M. R. Caira and L. R. Nassimbeni, J. Chem. Soc., Perkin Trans. 2, 1993, 2321. 15 A. Coetzee, L. R. Nassimbeni and K. A. Achleitner, Thermochim. Acta, 1997, 298, 81. 16 J. H. Park, A. Hussam, P. Couasnon, D. Fritz and P. W. Carr, Anal. Chem., 1987, 59, 1970. 17 J. T. Edsall and H. Gutfreund, Biothermodynamics, Wiley, New York, 1983. DGtrans = DGc – RT ln g• Received: Moscow, 26th May 1998 Cambridge, 23rd September 1998; Com. 8/04728F

ISSN:0959-9436     

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 1, 1999 (pp. 1–44) Novel synthesis of 3-monosubstituted furoxans Alexander N. Blinnikov and Nina N. Makhova* N. D. Zelinsky Institute of Organic Chemistry, Russian Academy of Sciences, 117913 Moscow, Russian Federation. Fax: +7 095 135 5328; e-mail: mnn@cacr.ioc.ac.ru The nitro group in 3-substituted 4-nitrofuroxans (4-nitro-1,2,5-oxadiazol 2-oxides) can be replaced by the hydride ion under the action of NaBH4 in EtOH, and this reaction is convenient for the preparation of 3-monosubstituted furoxans.Recently,1 we have proposed a general preparative synthesis of a-hydroxyalkyl(benzyl)furoxan and -furazan derivatives based on the reduction of acyl and ethoxycarbonyl substituents in these heterocycles with NaBH4 in EtOH. Acyl, methyl and amino groups were second substituents in the compounds studied. The reaction was completed within 10–30 min at 10–15 °C.The 4-amino groups in 4-amino-3-(a-hydroxymethyl- and ethyl)- furoxans 1a,b were oxidised to nitro groups to form corresponding 4-nitro-3-(a-hydroxymethyl- and ethyl)furoxans 2a,b (Scheme 1). In this work, we attempted to obtain nitroalcohol 2a by the reduction of a furoxan derivative containing a nitro group, 3-ethoxycarbonyl-4-nitrofuroxan 3,2 using the same reducting agent.Compound 3 was found to react with NaBH4 in EtOH under milder conditions (–10 °C, ~1 min); however, previously unknown 3-ethoxycarbonylfuroxan 4 was isolated (yield 51%) instead of expected 2a (Scheme 2). These conditions seem to be favourable to the nucleophilic substitution of the hydride ion for the nitro group in compound 3.It is likely that the ethoxycarbonyl group cannot be reduced to the hydroxymethyl group under mild conditions. The chemical shift of the hydrogen atom at C(4) in the 1H NMR spectrum of compound 4 is 8.2 ppm, and the chemical shift of the C(4) atom in the 13C NMR spectrum is 148.5 ppm. These values are consistent with the data published for the parent furoxan3 and 3-phenylfuroxan.4 The mass spectrum exhibits a peak of the molecular ion.The reactions of nitroaromatic compounds with NaBH4 can proceed via various pathways depending upon the reaction conditions and the type of substitution in the ring. For instance, azobenzenes, azoxybenzenes and anilines can be obtained by this reaction.5 Nitroaromatic compounds with electron-accepting substituents are reduced by NaBH4 to cyclohexene derivatives6 or form stable Meisenheimer complexes with hydride.7 The nucleophilic substitution of the hydride ion for the nitro group in aromatic compounds under the action of NaBH4 was also described.However, this reaction requires the presence of bulky substituents adjacent to the nitro group to prevent its conjugation with the ring and the presence of electron-withdrawing groups to activate the ring towards the hydride ion attack.8 In the oxadiazole series, this reaction was observed for the first time, and it could not be predicted in advance.Though the oxadiazole ring possesses a very high electron-withdrawing effect,9 only one substituent is adjacent to the nitro group in compound 3.Moreover, the nitro groups in both the 4- and 3-positions of the furoxan cycle are almost coplanar with the ring (16° in 3-methyl-4-nitrofuroxan10 and 11° in 3-nitro-4-phenylfuroxan11). 3-Monosubstituted furoxans in contrast to 4-monosubstituted ones are not easily accessible compounds. Only few examples of such structures have been described: 3-arylfuroxans 5 and 3-furazanylfuroxans 6 and isomeric furoxan-3-aldoximes 7a,b.These compounds were prepared by different methods. 3-Arylfuroxans 5 were synthesised by oxidation of the b-forms of corresponding monoarylglyoximes 8.12–15 3-Furazanylfuroxan 6 was obtained by dehydration of a-(furazanyl)nitrooxime 9.16 Furoxan-3-aldoximes 7a,b were synthesised as a mixture of isomers by transformation of a diacetyl derivative of nitromalonaldehyde 1017 (Scheme 3).In this connection, the replacement of the nitro group by the hydride ion in 3-R-4-nitrofuroxans can be useful for the preparation of 3-monosubstituted furoxans. To estimate the application field of this reaction, we studied the interaction of 3-alkyl-4-nitrofuroxans 11a,b and 3-phenyl-4- nitrofuroxan 12 with NaBH4 in EtOH.The following expected 3-monosubstituted furoxans were isolated in high yields in all cases: 3-methyl- and 3-ethylfuroxans 13a,b (which were unknown previously) and 3-phenylfuroxan 5a (Scheme 4).† Thus, the interaction of 3-substituted 4-nitrofuroxans with NaBH4 in EtOH is a new, general and convenient method for N O N R COR1 (O)n NaBH4 EtOH N O N R2 CH(OH)R3 (O)n N O N H2N CH(OH)R 90% H2O2/H2SO4 N O N O2N CH(OH)R O O a R = H b R = Me 1 2 Scheme 1 R = NH2, Me, COR1 R1 = Me, OEt, Ph R2 = NH2, Me, CH(OH)R3 R3 = H, Me, Ph n = 0, 1 N O N O2N CO2 Et O 3 i 2a i N O N H CO2 Et O 4 Scheme 2 Reagents and conditions: NaBH4 (2 mol), EtOH, –10 °C, 1 min, then HCl/H2O.HON Ar NOH H N O N Ar H O HNO3 (N2O4) a Ar = Ph 8 5 N O N H 9 NO2 H NOH N O N H N O N H O SOCl2 6 NOAc H NO2 K+ H NOAc N O N H 7a H N 10 i, H2O, 60–70 ºC ii, 0.1 M HCl O OH N O N H 7b H N O HO 2:1 Scheme 3Mendeleev Communications Electronic Version, Issue 1, 1999 (pp. 1–44) the synthesis of hitherto hardly accessible 3-monosubstituted furoxans, and the result of this reaction is independent of the second substituent (alkyl, aryl or alkoxycarbonyl). This work was supported by the NATO Linkage grant nos.DISRM LG961369 and CNS 970584. † All new compounds had satisfactory elemental analysis data, and their structures were confirmed by IR and NMR spectroscopy and mass spectrometry. IR spectra were recorded in KBr pellets. 1H, 13C and 15N NMR spectra (300, 75.5 and 30.4 MHz, respectively) were measured in CDCl3, internal standard SiMe4 for 1H and 13C and external standard MeNO2 for 15N.Preparation of 3-monosubstituted furoxans (general procedure). A solution of NaBH4 (5 mmol) in 50 ml of anhydrous ethanol was added to a solution of nitrofuroxan (2.5 mmol) at –10 °C, and the reaction mixture was stirred at corresponding temperatures (compound 3: –10 °C, 1 min; compounds 11a,b: 20 °C, 5–6 min; compound 12: 5–10 °C, 10 min).Next, the reaction system was cooled, 12 mmol of conc. HCl was added, EtOH was evaporated and 3-monosubstituted furoxans were purified by column chromatography on SiO2 (CHCl3–hexane eluent). 3-Ethoxycarbonylfuroxan 4: yield 53%, high-boiling liquid, Rf 0.62 (CHCl3). 1H NMR, d: 1.20 (t, 3H, Me, 3J 7.7 Hz), 4.17 (q, 2H, CH2, 3J 7.7 Hz), 8.20 (s, 1H, CH). 13C NMR ([2H6]acetone) d: 14.45 (q, Me, 1J 137 Hz), 83.47 (t, CH2, 1J 147 Hz), 109.18 (C-3 in furoxan ring, 2J 10.8 Hz), 148.05 (d, C-4 in furoxan ring, 1J 205 Hz), 158.80 (C=O). IR (n/cm–1): 1395, 1417, 1438, 1523, 1623 (furoxan ring), 1732, 1760 (CO), 2857, 2925, 2995 (CH in Et), 3145 (CH in furoxan ring). MS, m/z: 158 (M+). Phenylfuroxan 5a: yield 89%, mp 107–108 °C (lit.,10 108–109 °C). 3-Methylfuroxan 13a: yield 74%, bp 63–64 °C (2 Torr), Rf 0.2 (CHCl3– heptane, 1:4). 1H NMR, d: 2.18 (s, 3H, Me), 8.55 (s, 1H, CH). 13C NMR, d: 8.16 (dq, Me, 1J 132 Hz, 3J 6.0 Hz), 113.45 (m, C-3 in furoxan ring, 2J 6.0 Hz), 148.25 (dt, C-4 in furoxan ring, 1J 200 Hz, 3J 3.2 Hz). 15N NMR, d: –22.95 (N-2, 3JCH–N-2 4.0 Hz), –2.77 (N-5, 2J 11.8 Hz). IR (n/cm–1): 1380, 1495, 1620 (furoxan ring), 2930 (CH in Me), 3140 (CH in furoxan ring).MS, m/z: 100 (M+). 3-Ethylfuroxan 13b: yield 49%, bp 86–87 °C (2 Torr), Rf 0.29 (CHCl3– CCl4, 1:1). 1H NMR, d: 1.27 (t, 3H, Me, 3J 8.1 Hz), 2.54 (q, 2H, CH2, 3J 8.1 Hz), 8.14 (s, 1H, CH). 13C NMR, d: 10.49 (q, Me, 1J 121 Hz), 16.38 (t, CH2, 1J 141 Hz), 116.7 (m, C-3 in furoxan ring, 2J 11.4 Hz), 145.16 (d, C-4 in furoxan ring, 1J 182.7 Hz). 15N NMR, d: –23.19 (m, N-2, 3JCH–N-2 1.7 Hz), –1.56 (d, N-5, 2J 12.1 Hz). IR (n/cm–1): 1410, 1445, 1470, 1500, 1625 (furoxan ring), 2900, 2960, 2990 (CH in Et), 3135 (CH in furoxan ring). MS, m/z: 114 (M+). References 1 A. N. Blinnikov, A. S. Kulikov, N. N. Makhova and L. I. Khmel’nitskii, Izv. Akad. Nauk, Ser. Khim., 1996, 1782 (Russ. Chem. Bull., 1996, 45, 1692). 2 N. N. Makhova, V.G. Dubonos, A. N. Blinnikov, I. V. Ovchinnikov and L. I. Khmel’nitskii, Zh. Org. Khim., 1997, 33, 1216 (Russ. J. Org. Chem., 1997, 33, 1140). 3 T. I. Godovikova, S. P. Golova, Yu. A. Strelenko, M. Yu. Antipin, Yu. T. Struchkov and L. I. Khmel’nitskii, Mendeleev Commun., 1994, 7. 4 L. I. Khmel’nitskii, T. I. Godovikova, N. A. Ruleva, B. N. Khasapov and S. S. Novikov, Izv.Akad. Nauk SSSR, Ser. Khim., 1979, 2295 (Bull. Acad. Sci. USSR, Div. Chem. Sci., 1979, 28, 2118). 5 R. O. Hutchins, D. W. Lamson, L. Rua, C. Milewsky and B. Maryanoff, J. Org. Chem., 1971, 36, 803. 6 T. Severin, R. Schmitz and M. Adam, Chem. Ber., 1963, 96, 3076. 7 L. A. Kaplan and A. R. Siedle, J. Org. Chem., 1971, 36, 937. 8 D. W. Lamson, P. Ulrich and R. O. Hutchins, J. Org. Chem., 1973, 38, 2928. 9 E. Sedano, C. Sarasola, J. M. Ugalde, I. X. Irarabalbeitia and A. G. Guerrero, J. Phys. Chem., 1988, 92, 5094. 10 A. F. Cameron and A. A. Freer, Acta Crystallogr., B, 1974, 30, 354. 11 R. Calvino, A. Gasco, A. Serafino and D. Viterbo, J. Chem. Soc., Perkin Trans.2, 1981, 1240. 12 M. Milone, Gazz. Chim. Ital., 1929, 59, 266. 13 G. Ponzio and L. Avogadro, Gazz. Chim. Ital., 1927, 57, 124. 14 G. Ponzio, Gazz. Chim. Ital., 1925, 55, 698. 15 W. Jugelt, M. Tismer and M. Rauh, Z. Chem., 1983, 23, 29. 16 C. Grundmann, G. W. Nickel and R. K. Bansal, Liebigs Ann. Chem., 1975, 1029. 17 C. Grundmann, R. K. Bansal and P. S. Osmanski, Liebigs Ann. Chem., 1973, 898. N O N O2N R O 11 12 R = Ph i N O N H R O 13 5a R = Ph a R = Me b R = Et Scheme 4 Reagents and conditions: i, NaBH4 (2 mol), EtOH. Received: Moscow, 3rd June 1998 Cambridge, 11th September 1998; Com. 8/04733B

ISSN:0959-9436     

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 1, 1999 (pp. 1–44) New approaches to the preparation of azoxyfuroxans Alexander N. Blinnikov, Nina N. Makhova* and Lenor I. Khmel’nitskii N. D. Zelinsky Institute of Organic Chemistry, Russian Academy of Sciences, 117913 Moscow, Russian Federation. Fax: +7 095 135 5328; e-mail: mnn@cacr.ioc.ac.ru A new convenient method for the synthesis of 4,4'-azoxyfuroxans by reductive condensation of 4-nitrofuroxans (4-nitro-1,2,5- oxadiazole 2-oxides) has been developed; the title compounds can also be synthesised by oxidation of 4-amino- and 4,4'-azofuroxans, and a general method for the synthesis of isomeric azofuroxans is suggested.The first representatives of azoxyfuroxans (4,4'-azoxy-3,3'-diphenyl- and -3,3'-dimethylfuroxans 1a,b) were synthesised by the oxidation of 4-trioctylphosphinimino-3-phenyl(methyl)furoxans under the action of MCPBA.1 However, this method is inconvenient, because the latter needs to be specially prepared.However, azoxy derivatives of the aromatic and heteroaromatic series exhibit various biological activities (antibacterial, anticancer, antituberculous, and nematocidal).2 Therefore, it was reasonable to search for new, more convenient approaches to the preparation of azoxyfuroxans with various substituents.The synthesis of aromatic azoxy derivatives is normally based on the transformation of amino-, azo-, or nitro-substituted derivatives using various oxidising or reducing agents. Until recently, it has been impossible to perform similar reactions in the furoxan series, because the corresponding starting furoxan derivatives were lacking.We have recently developed convenient methods for the synthesis of amino- and nitrofuroxans.3–7 In this work, we studied the possibility of preparing azoxyfuroxans 1† by the oxidative condensation of aminofuroxans, by oxidation of azofuroxans, and by reductive condensation of nitrofuroxans.It seemed probable that the first two reactions would be efficient, since similar transformations with aminoand azo-furazans proceed successfully.8,9 The possibility of reductive condensation of nitrofuroxans seemed less probable, because of the sensitivity of the furoxan ring towards reducing agents.10 The oxidation of 4- and 3-aminofuroxans 4 and 5 by KMnO4 in the presence of HCl (Scheme 1) was used to obtain the starting isomeric 4,4'- and 3,3'-azofuroxans 2 and 3.‡ The synthesis of compounds 2 and 3 is of independent interest, because only a few representatives of these structures are described in the literature.11,12 This reaction was established to apply to both 4- and 3-aminofuroxans 4 and 5 and to be almost independent of the second substituent in the aminofuroxan.The possibility of direct oxidation of aminofuroxans to azoxyfuroxans was studied using 4-aminofuroxans 4 only, since it is known11 that 3-aminofuroxans 5 give a mixture of compounds by the action of oxidants of the peroxide type, which are those usually used for these transformations in the furazan series.8 A mixture of hydrogen peroxide and H2SO4 was used as the oxidant.The studies showed that electron-withdrawing substituents in position 3 of the furoxan ring (for example, CON3, compound 4f) prevent the formation of azoxy derivatives. The oxidation † 4,4'-Azoxy-3,3'-diphenylfuroxan 1a: yield 52%, mp 190–192 °C (MeOH), (lit.,1 190–192 °C). 4,4'-Azoxy-3,3'-dimethylfuroxan 1b: yield 58%, mp 187–189 °C (CHCl3), (lit.,1 187–189 °C). 4,4'-Azoxy-3,3'-dihydroxymethylfuroxan 1g: yield 27%, mp 103–104 °C, Rf 0.45 (CHCl3:PriOH, 9:1). 1H NMR (CDCl3) d: 4.75 (d, 2H, CH2), 4.96 (d, 2H, CH2), 5.05 d (1H, OH), 5.07 d (1H, OH). 13C NMR (CDCl3) d: 54.04 and 54.72 (CH2), 111.42 and 111.59 (C-3 in furoxan ring), 156.12 and 159.48 (C-4 in furoxan ring). 14N NMR (CDCl3, internal standard MeNO2) d: –68.0 [N=N(O)]. IR (n/cm–1): 1315 [N=N(O)], 1600 (furoxan ring), 2800, 2920 (CH), 3320 (OH). 4,4'-Azoxy-3,3'-diethylfuroxan 1h: yield 49%, mp 129–130 °C (CHCl3), Rf 0.47 (hexane:CH2Cl2, 1:2). 1H NMR (CDCl3) d: 1.21 and 1.24 (t, Me), 2.63 and 2.96 (q, CH2). IR (n/cm–1): 1185 [N=N(O)], 1536, 1650 (furoxan ring). MS, m/z: 270 (M+). 4,4'-Azoxy-3,3'-bis(2-methoxyethyl)furoxan 1i: yield 17%, mp 72.5– 73 °C, Rf 0.11 (benzene:EtOAc, 20:1). 1H NMR (CDCl3) d: 2.99 and 3.27 (t, CH2), 3.28 and 3.29 (s, Me), 3.70 and 3.77 (t, CH2O). 13C NMR (CDCl3) d: 24.33 and 25.22 (Me), 58.76 (2CH2), 67.81 and 68.12 (CH2O), 110.77 and 110.94 (C-3 in furoxan ring), 156.90 and 160.18 (C-4 in furoxan ring). IR (n/cm–1): 1323 [N=N(O)], 1630 (furoxan ring); MS, m/z: 330 (M+). ‡ 4,4'-Azo-3,3'-diphenylfuroxan 2a: yield 96%, mp 160–162 °C, Rf 0.45 (CCl4:CHCl3, 1:1). 1H NMR ([2H6]DMSO) d: 7.5 (m, 3H, m- and p-CH in Ar), 7.65 (m, 2H, o-CH in Ar). IR (n/cm–1): 1330, 1465, 1610 (furoxan ring). UV (EtOH, lmax/nm): 227, 270. MS, m/z: 350 (M+). 4,4'-Azo-3,3'-dimethylfuroxan 2b: yield 96%, mp 156–158 °C, Rf 0.4 (CHCl3:CH2Cl2, 1:1). 1H NMR (CDCl3) d: 2.45 (s, Me). IR (n/cm–1): 1400, 1540, 1655 (furoxan ring). UV (EtOH, lmax/nm): 263, 352.MS, m/z: 226 (M+). 4,4'-Azo-3,3'-diethoxycarbonylfuroxan 2c: yield 68%, mp 128–129 °C, Rf 0.51 (CHCl3). 1H NMR ([2H6]acetone) d: 1.26 (t, Me, 3J 7.2 Hz), 4.40 (q, CH2, 3J 7.2 Hz). 13C NMR ([2H6]acetone) d: 14.17 (Me), 64.39 (CH2), 103.82 (C-3 in furoxan ring), 155.89 (C-4 in furoxan ring), 162.29 (C=O). IR (n/cm–1): 1620, 1630 (furoxan ring), 1720 (C=O), 2960 (CH); 3,3'-Diacetyl-4,4'-azofuroxan 2d: yield 22.5%, mp 123–125 °C, Rf 0.64 (benzene:methanol, 10:0.3). 1H NMR ([2H6]DMSO) d: 2.63 (s). IR (n/cm–1): 1330, 1370, 1500, 1630 (furoxan ring), 1650 (CO), 2940 (CH). UV (EtOH, lmax/nm): 261, 342. MS, m/z: 282 (M+). 4,4'-Azo-3,3'-bis(2-methoxyethyl)furoxan 2i: yield 7%, mp 89.5–91 °C, Rf 0.18 (benzene:EtOAc, 20:1). 1H NMR (CDCl3) d: 3.46 (t, CH2), 3.63 (s, Me), 4.02 (t, CH2O). 13C NMR (CDCl3) d: 25.24 (CH2), 58.54 (Me), 66.31 (CH2O), 107.72 (C-3 in furoxan ring), 165.3 (C-4 in furoxan ring). IR (n/cm–1): 1118, 1622 (furoxan ring), 2840, 2870, 2915, 2950, 3000 (CH). MS, m/z: 314 (M+). 3,3'-Azo-4,4'-diphenylfuroxan 3a: yield 96%, mp 196–197 °C, Rf 0.65 (hexane:ethylacetate, 3:1). 1H NMR ([2H6]DMSO) d: 7.65 (m, 3H, mand p-CH), 8.05 (m, 2H, o-CH).IR (n/cm–1): 1335, 1475, 1570 (furoxan ring). UV (EtOH, lmax/nm): 216, 252, 349. MS, m/z: 350 (M+). 3,3'-Azo-4,4'-dibenzoylfuroxan 3e: yield 67%, mp 143–144 °C, Rf 0.57 (acetone). 1H NMR ([2H6]DMSO) d: 7.42 (m, 2H, m-H in Ar), 7.7 (m, 1H, p-H in Ar), 7.9 (m, 2H, o-H in Ar). 13C NMR ([2H6]DMSO) d: 124.0 (C-3 in furoxan ring), 128.5, 129.2, 130.1, 132.8 (Ar), 148.5 (C-4 in furoxan ring), 167.3 (C=O).IR (n/cm–1): 1325, 1390, 1460, 1500, 1600, 1630 (furoxan ring), 1680, 1700 (C=O). UV (EtOH, lmax/nm): 263, 293, 380. 3-Hydroxymethyl-4-nitrofuroxan 6g: yield 16%, oil, IR and NMR spectral data are identical with those of 6g in ref. 17. N R O N NH2 O N R O N N O N R O N N O i N R O N NH2 N R O N N N R O N N i O O O 4a–d 2a–d 5a,e 3a,e aR = Ph b R = Me c R = CO2Et d R = COMe e R = COPh Scheme 1 Reagents and conditions: i, KMnO4 (1.5–2 mol), HCl/H2O/ CH2Cl2, 20 °C, then HOOC–COOH.Mendeleev Communications Electronic Version, Issue 1, 1999 (pp. 1–44) was carried out using hydrogen peroxide of different concentrations. With dilute hydrogen peroxide, the starting compound 4f is unchanged, while an increase in the concentration of hydrogen peroxide leads to the formation of 4-nitrofuroxan 6f.The oxidation of 4-aminofuroxans with electron-donating substituents 4b,g gives a mixture of azoxy derivatives 1b,g (predominantly) and nitrofuroxans 6b,g (Scheme 2). A peroxide type oxidant (peracetic acid) was also used for the oxidation of isomeric azofuroxans 2 and 3. The isomeric diphenylazofuroxans 2a and 3a were studied. It was found that only 4,4'-azo-3,3'-diphenylfuroxan 2a formed an azoxy derivative 1a (Scheme 3).The isomer 3a was not affected by the oxidant. Evidently, this transformation is sterically hindered by the N-oxide groups. An appropriate reducing agent should be chosen for the synthesis of compounds 1 by the reductive condensation of 4-nitrofuroxans. Various reagents (aldehydes, hydrazines, Mg and Zn metals, and SnII salts) are used in the aromatic series for this purpose.13 The reactions are usually carried out in an alkaline medium, but the reactions of compounds with electronwithdrawing substituents are performed in neutral or weakly acidic media.Since the furoxan ring possesses a strong electronwithdrawing effect,14 we used zinc dust in acetic acid for the transformation of 4-nitrofuroxans into 4,4'-azoxyfuroxans.Although this reagent can reduce the furoxan ring,15,16 it can be expected that, due to an increase in the reactivity of the nitro group under the action of the ring, its transformation will occur rapidly under very mild conditions, and the ring itself be untouched. First, the reaction was studied using 4-nitro-3-methylfuroxan 6b.It was carried out at low temperature with a small excess of the reducing agent. The reaction afforded a mixture of the expected 4,4'-azoxy-3,3'-dimethylfuroxan 1b (predominantly) and its azo analogue 2b. The mixture was separated by column chromatography on SiO2. The reduction of the azoxy fragment in compound 1b is the most probable reason for the production of 2b.To prevent this process and to obtain almost pure 1b, a 1:1 acetic acid–water mixture was used as the solvent. In this mixture the starting nitrofuroxan 6b is only slightly soluble, and the azoxy compound 1b is virtually insoluble. It precipitates and does not react further. The other azoxyfuroxans were synthesised under similar conditions in high yields (Scheme 4).§ In all cases, the AcOH:H2O ratio was selected according to the solubility of the starting and final compounds.Pure azoxyfuroxans 1 were obtained after recrystallisation from appropriate solvents. Only in the case of 3-methoxyethyl-4-nitrofuroxan 6i, the formation of the azo derivative was not avoided. Clearly, the MeOCH2CH2 group increases the solubility of 1i in the AcOH/H2O mixture, resulting in its partial reduction to 2i. 3-Nitrofuroxans 7a,b (4-methyl-3-nitro- and 3-nitro-4-phenylfuroxans were studied) do not form the expected 3,3'-azoxy derivatives with zinc dust in AcOH; they decompose under the reaction conditions. Thus, the reductive condensation of 4-nitrofuroxans 6 is a new and convenient method for the preparation of 4,4'-azoxyfuroxans 1. It is noteworthy that nitrofurazans (methyl- and phenylnitrofurazans were studied) do not form azoxy derivatives under the action of the Zn/AcOH mixture. This reaction is specific for 4-nitrofuroxans only.All new compounds had satisfactory elemental analysis data and their structures were confirmed by IR, NMR and mass spectroscopy.†,‡ This work was supported by the NATO Linkage grant nos. DISRM LG961369 and CNS 970584.References 1 O. A. Rakitin, O. G. Vlasova, A. N. Blinnikov, N. N. Makhova and L. I. Khmel’nitskii, Izv. Akad. Nauk SSSR, Ser. Khim., 1991, 523 (Bull. Acad. Sci. USSR, Div. Chem. Sci., 1991, 40, 455). 2 S. G. Zlotin and O. A. Luk’yanov, Usp. Khim., 1993, 62, 157 (Russ. Chem. Rev., 1993, 62, 143). 3 N. N. Makhova, A. N. Blinnikov and L. I. Khmel’nitskii, Mendeleev Commun., 1995, 56. 4 I. V. Ovchinnikov, A. N. Blinnikov, N. N. Makhova and L. I. Khmel’nitskii, Mendeleev Commun., 1995, 58. 5 V. G. Dubonos, I. V. Ovchinnikov, N. N. Makhova and L. I. Khmel’nitskii, Mendeleev Commun., 1992, 120. 6 N. N. Makhova and T. I. Godovikova, Ross. Khim. Zh., 1997, 41, 54 (in Russian). 7 N. N. Makhova, V. G. Dubonos, A. N. Blinnikov and L. I. Khmel’nitskii, Zh. Org.Khim., 1997, 33, 1216 (Russ. J. Org. Chem., 1997, 33, 1140). 8 T. S. Novikova, T. M. Mel’nikova, O. V. Kharitonova, V. O. Kulagina, N. S. Alexandrova, A. B. Sheremetev, T. S. Pivina, L. I. Khmel’nitskii and S. S. Novikov, Mendeleev Commun., 1994, 138. 9 M. A. Epishina, N. N. Makhova, L. V. Batog, L. S. Konstantinova and L. I. Khmel’nitskii, Mendeleev Commun., 1994, 102. 10 L. I.Khmel’nitskii, S. S. Novikov and T. I. Godovikova, Khimiya furoksanov: reaktsii i primenenie (Chemistry of Furoxans: Reactions and Application), 2nd edn., Nauka, Moscow, 1996, p. 51 (in Russian). 11 N. N.Makhova, I. V. Ovchinnikov, B. N. Khasapov and L. I. Khmel’nitskii, Izv. Akad. Nauk SSSR, Ser. Khim., 1982, 646 (Bull. Acad. Sci. USSR, Div. Chem. Sci., 1982, 31, 573). 12 I. V.Ovchinnikov, N. N. Makhova, L. I. Khmel’nitskii, L. N. Akimova and V. I. Pepekin, Dokl. Ross. Akad. Nauk, 1998, 359, 499 [Dokl. Chem. (Engl. Transl.), 1998, 359, 67]. 13 K. H. Schundenhulte, Aromatic Azoxycompounds, in Metoden der Organischen Chemie, Houben-Weyl, 1965, 10/3, 745. 14 E. Sedano, C. Sarasola, J. M. Ugalde, I. X. Irarabalbetia and A. G. Guerrero, J. Phys. Chem., 1988, 92, 5094. 15 A. R. Gagneux and R. Meier, Helv. Chim. Acta, 1970, 53, 1883. § Synthesis of 4,4'-azoxyfuroxans 1 by reductive condensation of 4-nitrofuroxans 6 (general procedure). 4-Nitrofuroxan (10 mmol) was added to a mixture of AcOH/water (6a, 20:15 ml and 15 ml of MeOH; 6b, 15:15 ml; 6h, 15:30 ml; 6i, 10:15 ml). The reaction mixture was cooled to 0–2 °C and Zn dust (1.5–2.5 mmol) was added at this temperature over 2–4 h.The mixture was then warmed to 20 °C and stirred at this temperature for 1 h; the product obtained was filtered off and crystallised. N R O N NH2 O N R O N N O N R O N N O i N R O N NO2 O O + b R =Me f R = CON3 g R = CH2OH 1b 33% 1f 0% 1g 27% 6b 12% 6f 65% 6g 16% Scheme 2 Reagents and conditions: i. H2O2 (85%) (2.5 mol)/conc. H2SO4, 20 °C, 30 min, then 30 °C, 30 min and 65 °C, 30 min. 4b,f,g N Ph O N N O N Ph O N N O O N Ph O N N O N Ph O N N O i 2a 1a Scheme 3 Reagents and conditions: i, 30% H2O2 (24 mol), AcOH (170 mol), Ac2O (110 mol), 80–90 °C, 7 h, then reflux with addition of H2O2 until colourless. N R O N N O N R O N N O i N R O N NO2 O O 6a,b,h,i 1a,b,h,i + 2 (only 2i) i N R O N NO2 7a,b O decomposition a R = Ph b R =Me h R = Et i R = CH2CH2OMe Scheme 4 Reagents and conditions: i, Zn dust (1.5–2.5 mol), AcOH/H2O.Mendeleev Communications Electronic Version, Issue 1, 1999 (pp. 1–44) 16 G. Alimenti, M. Grifantini, F. Gualtieri and M. L. Stein, Tetrahedron, 1968, 24, 395. 17 A. N. Blinnikov, A. S. Kulikov, N. N. Makhova and L. I. Khmel’nitskii, Izv. Akad. Nauk, Ser. Khim., 1996, 1782 (Russ. Chem. Bull., 1996, 45, 1692). Received: Moscow, 3rd June 1998 Cambridge, 17th July 1998; Com. 8/04734K

ISSN:0959-9436     

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 1, 1999 (pp. 1–44) New version of mononuclear heterocyclic rearrangement Nina N. Makhova* and Alexander N. Blinnikov N. D. Zelinsky Institute of Organic Chemistry, Russian Academy of Sciences, 117913 Moscow, Russian Federation. Fax: +7 095 135 5328; e-mail: mnn@cacr.ioc.ac.ru Thermal recyclization of the 3-diazenofuroxanyl unit to form the 4-nitro-1,2,3-triazole fragment has been found in noncondensed 1,2,5-oxadiazol 2-oxide derivatives (3,3'-azofuroxans) with acetamido substituents in the 4,4'-positions.Recently,1 we have developed a general preparative methods for the synthesis of azo- and azoxyfuroxans. This work is devoted to the study of their reactivity. It has been found unexpectedly that 4,4'-bis(acetamido)-3,3'-azofuroxan 1b upon heating gives a new compound with the same molecular formula. According to the 1H and 13C NMR spectroscopic data, this compound contains two different acetamido groups.The 13C NMR spectrum contains four downfield signals (117.68, 144.73, 148.02 and 150.14 ppm), two of which (117.68 and one of the downfield signals at 148.02 or 150.14 ppm) are most likely assigned to the furoxan ring.2 Two other signals can be assigned to another aromatic heterocycle. In addition, a signal of the nitro group (–26.0 ppm) appeared in the 14N NMR spectrum.Therefore, a molecule of the new compound contains two aromatic heterocycles (one of which is the furoxan ring), two different acetamido groups and one nitro group. To obtain a simpler spectral pattern, the acetamido groups in the new compound were oxidized to nitro groups by a mixture of conc. H2O2 and conc.H2SO4. This reaction afforded a mixture of two new compounds in the 8:1 ratio, which were separated by chromatography on SiO2. According to the 14N NMR data, the prevailing compound contained three nitro groups, two of which had equal chemical shifts (–37.0 ppm), and the third nitro group had a chemical shift of –39.0 ppm.The 13C NMR spectrum exhibited only three signals at 124.48, 145.41 and 147.98 ppm, two of which were broadened and appeared as triplets due to spin–spin coupling with atoms of the 14N nitro groups. This shape of the signals is typical of C–NO2 fragments. The presence of the C–NO2 fragments was confirmed by narrowing of these signals in the 13C NMR spectrum after decoupling of 14N in both of the nitro groups.The ratio of the integral intensities showed that the carbon atom with a chemical shift of 124.48 ppm is bound to a nitro group, and two other nitro groups are linked to the carbon atoms with chemical shifts of 147.98 ppm. Obviously, the compound obtained contained two equivalent C–NO2 fragments, which are a part of a heteroaromatic ring. Two other signals in the 13C NMR spectrum belong to the furoxan ring.The comparison of the elemental analysis and NMR data suggests that the structure of this compound is 4-nitro-3-(4,5- dinitro-1,2,3-triazol-2-yl)furoxan 2b. The 15N NMR spectrum confirmed the presence of the triazole ring (two equivalent signals with the chemical shifts of 36.31 ppm and a signal with the chemical shift of 162.75 ppm; these data agree with the data published for 15N spectra of 2-substituted 1,2,3-triazoles3).The second compound is its isomer 2c. (It is known that the furoxan ring is prone to tautomerism,4 especially when two electronwithdrawing substituents are present). Thus, the primary product obtained by heating of azofuroxan 1b is the product of its thermal rearrangement, viz., 4-acetamido-3-(5-acetamido-4-nitro- 1,2,3-triazol-2-yl)furoxan 2a.Starting diazenofuroxan 1b was synthesised by the acetylation of 4,4'-diamino-3,3'-azofuroxan 1a with acetic anhydride in the presence of a catalytic amount of conc. H2SO4 (Scheme 1). Two rearrangements resulting in the 1,2,3-triazole ring are known in the furoxan series.The first rearrangement is the formation of 1,2,3-triazol 1-oxide derivatives under the action of amide anions on benzofuroxans5 or of primary aliphatic amines on 4-amino-3-nitrofuroxans.6 This transformation is probably initiated by the nucleophilic attack of an amide anion or amine on the N(5) atom of the furoxan ring. The second rearrangement (a version of the Boulton–Katritzky rearrangement7) is the recyclization of 4-aryldiazeno-5-nitrobenzofuroxan 4 to 2-aryl-4,7-dinitrobenzo-1,2,3-triazole 5.Starting benzofuroxan 4 was not isolated but appeared as an intermediate product upon heating of 2,6-dinitro-3-azidoaryldiazenobenzene 3. In this case, the reaction also starts from the nucleophilic N O N NH2 N N N O N NH2 O O N O N NHAc N N N O N NHAc O O i N N N O2N AcHN N O N AcHN O ii iii N N N O2N O2N N O N O2N O 1a 1b 2a 2b (4-NO2-furoxan) 2c (3-NO2-furoxan) 2b:2c = 8:1 Scheme 1 Reagents and conditions: i, Ac2O (20 mol), H2SO4 (cat.amount), 30 °C, 10 min; ii, AcOH:Ac2O = 4:1, 50 °C, 3 h; iii, conc. H2O2/conc. H2SO4, 22–24 °C, 40 min. O2N NO2 N3 N N Ar 3 155–160 ºC diglyme, 20 min O2N N N Ar N O N O O2N N N Ar N O N 4 NO2 N N N NO2 O 5 O Scheme 2 Ar N A B D X Y RZ N X Y Z A B DR Scheme 3Mendeleev Communications Electronic Version, Issue 1, 1999 (pp. 1–44) attack of a diazene unit on the N(5) atom of the furoxan ring (Scheme 2). The rearrangement found is formally similar to the latter case, but it occurs with the noncondensed furoxan system. Moreover, several mononuclear heterocyclic rearrangements are known,8 which correspond to the general Scheme 3.However, the rearrangement under consideration can be presented by none of these schemes, because, first, azofuroxans, in particular, 1a and its dinitro analogue 1h, are planar systems with the trans arrangement of furoxan rings with respect to the azo group9 and hence it is difficult to imagine the nucleophilic attack of the diazene unit on the N(5) atom in compound 1b.Second, the isomerisation of the furoxan ring as the first stage of the reaction should be ruled out, because the acetamido and nitro groups in compound 2a are bound to two different carbon atoms of the triazole ring. To obtain additional information on this reaction, we used other 3,3'-azofuroxans with various substituents in the 4,4'- positions, namely, 4,4'-diamino-, 4,4'-bis(dimethylamino)-, 4,4'- bis(N-methylacetamido)- and 4,4'-diphenyl-azofuroxans 1a,c–e, respectively.We found that amino derivatives 1a,c remained unchanged upon heating at 100 °C for 2 h. Heating of diphenyl derivative 1e at 110 °C for 10 h resulted in the formation of a mixture of the starting compound 1e and products of isomerisation of one and two furoxan rings 1f and 1g, respectively (Scheme 4).Only bis(acetamido) derivative 1d undergoes the aforementioned rearrangement to form a mixture of isomeric triazolylfuroxans 2d and 2e in the 1:2 ratio. The starting azofuroxan 1c was synthesised by nucleophilic substitution for the nitro groups in 3,3'-azo-4,4'-dinitrofuroxan 1h, and 1d was obtained by the acetylation of 1c (Scheme 5).† Thus, it was established that the presence of 4,4'-acetamido groups in the starting 3,3'-azofuroxan 1 is the key condition for this rearrangement.Based on this fact, we can suggest a hypothetical scheme of this reaction, which includes two successive rearrangements. The first rearrangement is the transformation of one of the acetamidofuroxan fragments in compound 1 into the 1,2,4-oxadiazole unit with the cleavage of the O(1)–N(5) bond of the furoxan ring and the release of a nitromethylene fragment (intermediate 6).The second rearrangement is the transformation of compound 6 into 2. † All new compounds exhibited satisfactory elemental analysis data and their structures were confirmed by IR, NMR and mass spectroscopy. Spectroscopic data: 1H NMR (300MHz), 13C NMR (75.47 MHz), standard TMS; 14N and 15N NMR (21.6 MHz), internal standard MeNO2. 3,3'-Azo-4,4'-bis(acetamido)furoxan 1b: yield 93%, mp 196–198 °C, Rf 0.49 (CHCl3:PriOH, 9:1). 1H NMR (CF3COOD) d: 2.3 (s, 3H, Me), 11.1 (br. s, NH). IR (n/cm–1): 1595 (furoxan ring), 1695 (C=O), 3233 (NH). MS, m/z: 312 (M+). 3,3'-Azo-4,4'-bis(methylamino)furoxan 1c: yield 75%, mp 201–203 °C, Rf 0.32 (benzene:EtOAc, 9:1). 1H NMR ([2H6]acetone) d: 3.0 (d, Me).IR (n/cm–1): 1516, 1604 (furoxan ring), 2865, 2940 (CH), 3433 (NH). MS, m/z: 256 (M+). 3,3'-Azo-4,4'-bis(N-methylacetamido)furoxan 1d: yield 91%, mp 125– 126.5 °C, Rf 0.20 (benzene:EtOAc, 3:1). 1H NMR ([2H6]DMSO) d: 2.12 (s, MeCO), 3.40 (s, NMe). 13C NMR ([2H6]DMSO) d: 21.35 (MeCO, 1J 146.3 Hz), 35.34 (NMe, 1J 145.5 Hz), 126.66 (C-3 in furoxan ring), 149.68 (C-4 in furoxan ring), 170.28 (C=O). 4-Acetamido-3-(5-acetamido-4-nitro-1,2,3-triazol-2-yl)furoxan 2a. A suspension of 1b (1.0 g, 3.2 mmol) in AcOH (40 ml) and Ac2O (5 ml) was heated at 48–50 °C for 3 h. The reaction mixture was evaporated to 5 ml and cooled, and the precipitate was filtered off. Yield 65%, mp 141–143 °C, Rf 0.21 (CHCl3:PriOH, 9:1). 1H NMR (CF3COOD) d: 2.21 (s), 2.36 (s, 6H, 2MeCO), 11.43 (br. s, 2H, 2NH). 13C NMR (CF3COOD) d: 29.19 and 29.94 (2Me), 117.68 (C-3 in furoxan ring), 144.74 (C-4 in triazole ring), 146.02 (C-5 in triazole ring), 150.14 (C-4 in furoxan ring), 178.04, 178.81 (C=O). 14N NMR (CH3COOD) d: –26.6 (NO2). IR (n/cm–1): 1333, 1580 (NO2), 1640 (furoxan ring), 1708 (C=O), 2995, 3050 (CH), 3235 (NH). 4-Nitro-3-(4,5-dinitro-1,2,3-triazol-2-yl)furoxan 2b: yield 22%, mp 94.5– 95 °C (CHCl3, decomp.), Rf 0.51 (benzene). 13C NMR (CDCl3) d: 113.6 (C-3 in furoxan ring), 146.3 (C-4 and C-5 in triazole ring), 154.8 (C-4 in furoxan ring). 14N NMR (CDCl3) d: –37.0 (2NO2 of triazole ring), –39.0 (NO2 of furoxan ring). 15N NMR (CDCl3) d: 6.75 (N-2 in furoxan ring), 21.11 (N-5 in furoxan ring), 36.41 (N-1 and N-3 in triazole ring), 38.77 (NO2 of triazole ring), 40.05 (NO2 of furoxan ring), 162.75 (N-2 in triazole ring).IR (n/cm–1): 1320, 1332, 1570 (NO2), 1685 (furoxan ring). MS, m/z: 288 (M+). 3-Nitro-4-(4,5-dinitro-1,2,3-triazol-2-yl)furoxan 2c: yield 3%, oil. 13C NMR (CDCl3) d: 124.48 (C-3 in furoxan ring, 1J13C–14N 21.3 Hz), 145.41 (C-4 in furoxan ring), 147.96 (C-4 and C-5 in triazole ring). 14N NMR (CDCl3) d: –37 (Dn1/2 16 Hz, NO2 of triazole ring), –43 (Dn1/2 3.0 Hz, NO2 of furoxan ring). MS, m/z: 288 (M+). The mixture of 4(3)-(N-methylacetamido)-3(4)-(4-nitro-5-N-methylacetamido- 1,2,3-triazol-2-yl)furoxans 2d and 2e (2:1): total yield 38%, oil. 2d: 1H NMR ([2H6]acetone) d: 2.26 (s, 3H, MeCO of triazole ring), 2.30 (s, 3H, MeCO of furoxan ring), 3.48 (s, 3H, NMe of triazole ring), 3.55 (s, 3H, NMe of furoxan ring). 13C NMR ([2H6]acetone) d: 21.23 (MeCO of furoxan ring, 1J 141.6 Hz), 21.90 (MeCO of triazole ring), 31.10 (NMe of furoxan ring), 35.44 (NMe of triazole ring, 1J 142.3 Hz), 115.15 (C-3 in furoxan ring), 145.2 (C–NAc of triazole ring), 148.34 (C–NO2 in triazole ring), 150.42 (C-4 in furoxan ring), 171.16 (CO in triazole ring), 172.73 (CO in furoxan ring). 14N NMR ([2H6]acetone) d: –28.0 (NO2, Dn1/2 77 Hz). 2e: 1H NMR ([2H6]acetone) d: 2.05 (s, 3H, MeCO of furoxan ring), 2.20 (s, 3H, MeCO of triazole ring), 3.20 (s, 3H, NMe of furoxan ring), 3.35 (s, 3H, NMe of triazole ring). 13C NMR ([2H6]acetone) d: 21.53 (MeCO of furoxan ring, 1J 129.3 Hz), 22.03 (MeCO of triazole ring, 1J 142.3 Hz), 30.34 (NMe of furoxan ring, 1J 137.2 Hz), 35.34 (NMe of triazole ring, 1J 143.3 Hz), 115.94 (C-3 in furoxan ring, 3J 3.0 Hz), 145.2 (C–NAc in triazole ring), 148.56 (C–NO2 in triazole ring), 150.07 (C-4 in furoxan ring), 170.26 (CO in furoxan ring), 171.16 (CO in triazole ring).N O N Ph N N O N Ph N O O N O N Ph N N O N Ph N O i i O N O N Ph N N O N Ph N O O 1e 1f 1g Scheme 4 Reagents and conditions: i, AcOH, 110 °C, 10 h.N O N NO2 N N N O N NO2 O O N O N NHMe N N N O N NHMe O O i N N N O2N AcN N O N AcN O N N N O2N AcN N O N AcN 1h 1c 2d N O N N N N O N NAc O O 1d AcN Me Me iii Me Me ii 1:2 2e Me Me O Scheme 5 Reagents and conditions: i, CHCl3:Et2O = 3:1, MeNH2 (in gas phase), 20 °C, 40 min; ii, Ac2O (2 mol), H2SO4 (cat. amount), 20 °C, 30 min; iii, AcOH (or EtOAc, dioxane), 80–100 °C, 1.5–2 h.Mendeleev Communications Electronic Version, Issue 1, 1999 (pp. 1–44) The scheme suggested is consistent with the known methods of synthesis and the reactivity of 1,2,4-oxadiazoles, for example, the preparation of 1,2,4-oxadiazoles by thermal cyclization of a benzamidine derivative.10 Moreover, the thermal cleavage of the O(1)–N(2) bond under the action of nucleophiles to form a new heterocycle, in particular, 1,2,3-triazole,11–13 is the typical reaction of 1,2,4-oxadiazoles, including intramolecular reactions.In addition, it is noteworthy that both rearrangements in Scheme 6 agree with the general scheme (Scheme 3) of mononuclear heterocyclic rearrangements. This work was supported by the NATO Linkage grant nos. DISRM LG961369 and CNS 970584.References 1 A. N. Blinnikov, N. N. Makhova and L. I. Khmel’nitskii, Mendeleev Commun., 1999, 15. 2 R. Calvino, R. Fruttero, A. Gasco, V. Mortarini and S. Aime, J. Heterocycl. Chem., 1982, 19, 427. 3 L. Stefaniak, J. D. Roberts, M. Witanowski and G. A. Webb, Org. Magn. Reson., 1984, 22, 215. 4 L. I. Khmel’nitskii, S. S. Novikov and T. I. Godovikova, Khimiya furoksanov: reaktsii i primenenie (Chemistry of Furoxans: Reactions and Application), 2nd edn., Nauka, Moscow, 1996, p. 13 (in Russian). 5 B. Gohrmann and H. J. Niclas, J. Prakt. Chem., 1990, 332, 1054. 6 T. I. Godovikova, S. P. Golova, S. A. Vozchikova, E. L. Ignat’eva, M. V. Povorin, V. S. Kuz’min and L. I. Khmel’nitskii, Mendeleev Commun., 1995, 194. 7 A. J. Boulton, P. B. Ghosh and A. R. Katritzky, J. Chem. Soc. (B), 1966, 1004. 8 A. J. Boulton, A. R. Katritzky and A. M. Hamid, J. Chem. Soc. (C), 1967, 2005. 9 I. V. Ovchinnikov, N. N. Makhova, V. S. Kuz’min, A. N. Akimova and V. I. Pepekin, Dokl. Akad. Nauk SSSR, 1988, 359, 499 [Dokl. Chem. (Engl. Transl.), 1988, 359, 67]. 10 T. Fuchigami and K. Odo, Chem. Lett., 1974, 1139. 11 M. Ruccia, N. Vivona and G. Cusmano, J. Heterocycl. Chem., 1971, 8, 137. 12 P. Gramantieri, Gazz. Chim. Ital., 1935, 65, 102. 13 V. G. Yashunsky and L. E. Kholodov, Usp. Khim., 1980, 49, 54 (Russ. Chem. Rev., 1980, 49, 28). N O N N N N O N N O O 1 N R R N N N O2N AcN N O N AcN O Me O Me N N O N N O R O Me N NO2 N N R O Me 6 O R R 2a R = H 2d + 2e R = Me Scheme 6 Received: Moscow, 11th June 1998 Cambridge, 17th July 1998; Com. 8/04741C

ISSN:0959-9436     

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 1, 1999 (pp. 1–44) Electrochemical transformations of alkylidenemalonates into substituted cyclopropanecarboxylates Michail N. Elinson,* Sergey K. Feducovich, Alexander A. Zakharenkov and Gennady I. Nikishin N. D. Zelinsky Institute of Organic Chemistry, Russian Academy of Sciences, 117913 Moscow, Russian Federation. Fax: + 7 095 135 5328 Electrolysis of alkylidenemalonates in acetonitrile in the presence of sodium iodide as a mediator in an undivided cell leads to various substituted cyclopropanetetracarboxylates depending on the conditions of the electrolysis and on the structure of the starting alkylidenemalonate.Alkylidenemalonates are well-known intermediates used in organic synthesis primarily owing to reactions of their activated double bond conjugated with two carboxyl groups.1 The known electrochemical transformations of alkylidenemalonates are associated with two types of the reactivity of the activated double bond: reductive hydrodimerisation2,3 and addition of electrochemically generated anions to the double bond.4,5 Nevertheless, to our knowledge, direct transformations of alkylidenemalonates into cyclopropanetetracarboxylates are unknown.In the last few years, mediators were widely used for the electrooxidation and electroreduction of organic compounds.6 Among a variety of mediators, the halide anion–halogen redox system is one of the most useful from the viewpoint of organic synthesis and large-scale industrial processes.7 Recently, in studies of the electrochemical oxidation of organic compounds in the presence of alkali metal halides, we have performed electrochemical cyclodimerisation of alkylidenemalonates into 3,4-disubstituted cyclobutane-1,1,2,2-tetracarboxylates3,8 and electrocatalytic transformation of alkylidenemalonates into 2-alkyl-3,3-dimethoxyalkane-1,1-dicarboxylates via an electrochemically induced oxidative rearrangement.9,10 Cyclopropane derivatives occupy a significant place in synthetic organic chemistry.11 Their structure and reactivity are responsible for the widespread use in the synthesis of naturally occurring products.Cyclopropanecarboxylic acid derivatives play an important role as effective agents in agriculture and medicine.12 Insecticidal pyrethrins (cyclopropanoid chrysanthemic acid derivatives) are perhaps the best known examples of their use.13 We have already used an electrochemical approach to the synthesis of substituted cyclopropanes by electrolysis of alkylidenemalonates and a malonate in methanol in the presence of halides as mediators in an undivided cell:14 Recently we have also synthesised substituted cyclopropanes by co-electrolysis of alkylidenecyanoacetates and cyanoacetic ester:15 Here we report a new unusual direct electrochemical transformation of alkylidenemalonates into functionally substituted cyclopropanes.We found that electrochemical transformation of ethylidenemalonate 1 in acetonitrile in the presence of NaI in an undivided cell depends on the conditions and gives rise to three types of substituted cyclopropanetetracarboxylates 2–4.† One of these compounds is formed as the main product of the reaction (Table 1).Note that the reactions described below are the first example of direct transformations of the alkylidenemalonates into functionally substituted cyclopropanes: Electrolysis of ethylidenemalonate in acetonitrile in the presence of only NaI leads mainly to the formation of cyclopropanetetracarboxylate 2 accompanied by a small amount of cyclopropanetetracarboxylate 3: Addition of ethylmalonate to the reaction mixture causes an increase in the cyclopropanetetracarboxylate 3 yield up to 57% (Table 1, experiment nos. 2–4). † All new compounds exhibited expected NMR spectra and data of elemental analysis or exact mass measurements. For 2: 1H NMR (CDCl3) d: 1.43 (d, 3H, Me), 1.57 (m, 1H, CH), 1.82 (m, 1H, CH), 2.43 (m, 1H, CH), 3.51 (m, 1H, CH), 3.71 (s, 6H, OMe), 3.73 (s, 6H, OMe). 13C NMR (CDCl3) d: 14.47 (t), 18.37 (q), 28.25 (d), 31.36 (d), 33.39 (s), 53.10 (q), 53.20 (q), 53.26 (q), 58.40 (s), 120.03 (s), 166.97 (s), 168.07 (s), 169.74 (s).For 3: 1H NMR (CDCl3) d: 0.92 (t, 3H, Me), 1.51 (dd, 1H, CHaHb– CHc, Jab –5.3 Hz, Jac 10.2 Hz), 1.81 (dd, 1H, CHaHb–CHc, Jab –5.3 Hz, Jac 8.9 Hz), 2.01 (m, 2H, CH2), 2.51 (dd, 1H, CHaHb–CHc, Jac 10.2 Hz, Jbc 8.9 Hz), 3.71 (s, 6H, OMe), 3.73 (s, 6H, OMe). 13C NMR (CDCl3) d: 8.96 (q), 18.44 (t), 27.94 (t), 27.96 (d), 32.76 (s), 52.26 (q), 52.42 (q), 57.14 (s), 167.97 (s), 168.32 (s), 169.56 (s), 170.23 (s). For 4: 1H NMR (CDCl3) d: 1.39 (d, 3H, Me), 2.46 (q, 1H, CH), 3.71 (s, 6H, OMe), 3.72 (s, 6H, OMe). 13C NMR (CDCl3) d: 8.32 (q), 31.41 (d), 43.94 (s), 52.72 (q), 53.13 (q), 165.02 (s), 167.01. CO2Me CO2Me H2C CO2Me CO2Me electrolysis NaHal, MeOH R1 R2 CO2Me CO2Me MeO2C MeO2C Hal = I, Br R1R2C CO2R CN H2C CO2R CN MHal R1 R2 CO2R CN RO2C NC M = Li, Na R1R2C electrolysis CN MeO2C CO2Me CO2Me CO2Me CO2Me CO2Me MeO2C CO2Me CO2Me CO2Me MeO2C CO2Me CO2Me MeO2C 1 4 2 3 CN MeO2C CO2Me CO2Me CO2Me CO2Me CO2Me MeO2C CO2Me CO2Me CO2Me 2 (62%) 3 (8%) electrolysis NaI, MeCNMendeleev Communications Electronic Version, Issue 1, 1999 (pp. 1–44) Small amounts of a protogenic solvent such as methanol or water also influence the ethylidenemalonate electrolysis (Table 1). Thus, in the presence of water, cyclopropanetetracarboxylate 4 was obtained in 60% yield.Electrolysis of ethylidenemalonate under the same conditions in the presence of malonate resulted in the formation of cyclopropane 4 in 92% yield (Table 1, experiment no. 7). Taking into consideration the data obtained, the following scheme of the formation of cyclopropanes 2–4 is proposed. Electrochemical reactions occur at both the anode and the cathode to generate intermediates which determine the overall course of the process.In all reactions studied, the anodic process is the oxidation of iodide anions into iodine: The formation of iodine at the anode is observed by a corresponding colour change when electrolysis is conducted without stirring of the reaction mixture. Electrolysis of ethylidenemalonate in acetonitrile in an undivided cell in the presence of NaI and in the absence of other additives leads to the formation of anion A from ethylidenemalonate by the action of the base electrogenerated at the cathode under the conditions of the electrolysis.The reaction of anion A with iodine generated at the anode gives rise to iodopropylidenemalonate 5: As it has been found previously, under the specified conditions, acetonitile reacts with Na generated at the cathode with the formation of methane and the cyanide anion:16,17 Thus, in this case, ethylidenemalonate can serve as a proton donor for the methylide anion generated in the last reaction.The cyanide anion formed adds to the double bond of ethylidenemalonate to produce anion B: The addition of the latter to the double bond of malonate 5 followed by cyclisation analogous to that reported previously,9 gives rise to cyclopropane 2 as the end product: The addition of ethylmalonate to the electrolytic system leads to the formation of anion C in the system (this ion is more stable than anion A) due to the interaction of ethylmalonate with any electrogenerated base that exists in the system.Among them, anion A can act as an electrogenerated base for ethylmalonate: The addition of anion C to the double bond of intermediate iodomalonate 5 and the following intermolecular cyclisation gives cyclopropane 3: The formation of small quantities of cyclopropane 3 in the absence of ethylmalonate takes place because of partial hydrogenation of ethylidenemalonate into ethylmalonate at the cathode under conditions of the electrolysis. The addition of protogenic solvents causes an additional route of the electrolytic process.Under these conditions, an electrochemically induced Knoevenagel retro-reaction takes place with the formation of a malonate ester which adds to the double bond of ethylidenemalonate with subsequent cyclisation to cyclopropane ester 4. In the presence of water, the above retroreaction becomes the main route of the process.The presence of small quantities of methanol also facilitates the hydrogenation of the double bond of ethylidenemalonate and thus increases the yield of cyclopropane ester 3. Electrolysis of propylidene- and butylidenemalonates in acetonitrile in the presence of NaI under conditions of experiment no. 1 (Table 1) leads to a more complex mixture of products in which cyclopropanes 6a,b were identified and isolated by column chromatography in 33 and 25% yields, respectively.CO2Me CO2Me MeO2C CO2Me CO2Me CO2Me 3 electrolysis , NaI, MeCN CO2Me CO2Me CO2Me CO2Me MeO2C CO2Me CO2Me MeO2C 4 electrolysis NaI, MeCN, 1 equiv. of H2O 2I– – 2e– I2 a 14 mmol of ethylidenemalonate; 7 mmol of NaI; 20 ml of acetonitrile; glassy carbon cathode; C-anode; current density 100 mA cm–2; 40 °C.b Determined by gas chromatography and NMR spectroscopy, yields of the isolated compounds are given in parentheses. Table 1 Electrolysis of ethylidenemalonate 1.a Experiment no. Additives Electricity passed/ F mol–1 Cyclopropane yield (%)b 2 3 4 1 — 2.1 62 (54) 8 — 2 Ethylmalonate (2 mmol) 2.1 33 22 — 3 Ethylmalonate (7 mmol) 2.1 9 57 (48) — 4 Ethylmalonate (14 mmol) 2.1 6 52 — 5 MeOH (14 mmol) 2.1 19 30 13 6 H2O (14 mmol) 2.1 — — 60 (47) 7 Malonate (1 mmol) 2.1 — — 92 (88) CO2Me CO2Me CO2Me CO2Me CO2Me CO2Me I B– – BH I2 5 A MeCN + 2Na + H+ CH4 + CN– + 2Na+ CO2Me CO2Me B NC CO2Me CO2Me + CN – CO2Me CO2Me B NC CO2Me CO2Me 5 I CO2Me CO2Me CN MeO2C CO2Me I CO2Me CO2Me CN MeO2C CO2Me 2 CO2Me CO2Me CO2Me CO2Me B– – BH C CO2Me CO2Me CO2Me CO2Me C I 5 MeO2C CO2Me CO2Me CO2Me 3 R CN MeO2C CO2Me CO2Me CO2Me 6a,b R a R = Me b R = EtMendeleev Communications Electronic Version, Issue 1, 1999 (pp. 1–44) Under the conditions studied, isobutylidene malonate is unable to form dimeric products apparently because of steric hindrances in the addition of anions B and C to the double bond owing to a bulky isopropyl substituent.In this case, the electrogenerated cyanide ion attacks the double bond of g-iodoisobutylidenemalonate 7, which is formed in a way analogous to the formation of intermediate 5 in the solution. The following cyclisation leads to 3,3-dimethyl-2-cyanocyclopropane-1,1-dicarboxylate 8 in 73% yield. This work was supported by the Russian Foundation for Basic Research (grant no. 97-03-33165a). References 1 G. Jones, in Organic Reactions, Wiley, New York, 1967, p. 204. 2 H. J. Shefer, Angew. Chem., Int. Ed. Engl., 1981, 20, 911. 3 M. N. Elinson, S. K. Feducovich, A. A. Zakharenkov, B. I. Ugrak, G. I. Nikishin, S. V. Lindeman and Yu. T. Struchkov, Tetrahedron, 1995, 51, 5035. 4 M. N. Elinson, S. K. Feducovich and G.I. Nikishin, Izv. Akad. Nauk SSSR, Ser. Khim., 1989, 352 (Bull. Acad. Sci. USSR, Div. Chem. Sci., 1989, 38, 301). 5 J.-C. le Menn, J. Sarrazin and A. Tallec, Can. J. Chem., 1991, 69, 761. 6 Novel Trends in Electroorganic Synthesis, ed. S. Torii, Springer–Verlag, Berlin, 1998. 7 T. Shono, Tetrahedron, 1984, 40, 811. 8 G. I. Nikishin, M. N. Elinson, S. K. Feducovich, B. I. Ugrak, Yu.T. Struchkov and S. V. Lindeman, Tetrahedron Lett., 1992, 33, 3223. 9 G. I. Nikishin, M. N. Elinson and S. K. Feducovich, Tetrahedron Lett., 1991, 32, 799. 10 G. I. Nikishin, M. N. Elinson and S. K. Feducovich, Tetrahedron, 1998, 54, 14529. 11 T. Tsuji and S. Nishida, The Chemistry of the Cyclopropyl Group, Wiley, New York, 1987. 12 L. A. Yanovskaya, V. A. Dombrovsky and A. Kh. Khusid, Tsiklopropany s funktsional’nymi gruppami. Sintez i primenenie (Cyclopropanes with functional groups. Synthesis and application), Nauka, Moscow, 1980 (in Russian). 13 J. Crosby, Tetrahedron, 1991, 47, 4789. 14 M. N. Elinson, S. K. Feducovich, S. G. Bushuev, A. A. Zakharenkov, D. V. Pashchenko and G. I. Nikishin, Mendeleev Commun., 1998, 15. 15 M. N. Elinson, S. K. Feducovich, S. G. Bushuev, D. V. Pashchenko and G. I. Nikishin, Izv. Akad. Nauk, Ser. Khim., 1998, 1165 (Russ. Chem. Bull., 1998, 47, 1133). 16 J. P. Billon, J. Electroanal. Chem., 1960, 1, 486. 17 J. Vedel and B. Tremillon, J. Electroanal. Chem., 1960, 1, 241. CO2Me CO2Me B– I2 CO2Me CO2Me I CN – – I – CO2Me CO2Me NC 7 8 Received: Moscow, 15th June 1998 Cambridge, 28th September 1998; Com. 8/05509B

ISSN:0959-9436     

出版商:RSC    

年代:1999

数据来源: RSC