摘要:

Mendeleev Communications Electronic Version, Issue 5, 1998 (pp. 169–205) Formation of singlet oxygen in the system tris(bipyridine)ruthenium(II)–dimethyldioxirane Dmitri V. Kazakov, Aleksandr I. Voloshin, Valeri V. Shereshovets, Valeri N. Yakovlev and Valeri P. Kazakov* Institute of Organic Chemistry, Ufa Scientific Centre of the Russian Academy of Sciences, 450054 Ufa, Russian Federation. Fax: +7 3472 35 6066; e-mail: chemlum@ufanet.ru Decomposition of dimethyldioxirane in the presence of tris(bipyridine)ruthenium(II) complex (RuII) is accompanied by chemiluminescence (CL) in the infrared spectrum region (1270 nm) due to the formation of singlet oxygen; reaction between RuIII and superoxide ion, which is formed in this process via a sequence of electron-transfer reactions, is assumed to be a CL step.Dioxiranes are known to be powerful yet selective oxidizing reagents. These three-membered ring cyclic peroxides have become an intensively studied species over the last decade mainly owing to their role in oxygen-transfer reactions. Epoxidation of alkenes and oxygen insertion into the ‘nonactivated’ aliphatic C–H bonds of alkanes are prominent examples of these oxidations.1–3 However, dioxiranes are also prone to electrontransfer (ET) reactions,2,4–10 some of which are accompanied by chemiluminescence (CL).2,7–10 Indeed, it is an ET mechanism, namely chemically induced electron exchange luminescence (CIEEL), that accounts for CL upon decomposition of methyl- (trifluoromethyl)dioxirane (TFMD) and dimethyldioxirane (DMD) in the presence of some aromatic hydrocarbons.2,7,10 The phenomenon of CL is a new and attractive property of dioxiranes which has begun to be explored only in recent years. Recently, we reported on CL in the visible spectrum region (both in solution9 and on a silipor surface10) in the RuIIpromoted decomposition of DMD (CIEEL mechanism): In this communication we report that, besides the visible CL (Scheme 1), the interaction between DMD and catalytic amounts of RuII is also accompanied by CL in the infrared spectrum region (1270 nm) due to radiative deactivation of singlet oxygen.The near infrared chemiluminescence of 1O2 was recorded using apparatus described previously.11 In a typical procedure, a solution of RuII in the appropriate solvent (1 ml, [RuII]0 = = 1×10–4 mol l–1) was poured into a cell placed above the photocathode of a photomultiplier.Then, a DMD solution in acetone (0.5 ml, [DMD]0 = 5.25×10–2 mol l–1) was rapidly injected. All reactions were carried out at room temperature and under continuous argon flow. The apparatus for CL recording was calibrated using the emission from 1O2 formed during thermal decomposition of triphenyl phosphite ozonide, using a known yield of singlet oxygen and decomposition rate.The yield of 1O2 formation per dioxirane molecule consumed was equal to 0.00084% (in acetone). The infrared CL at 1270 nm was solvent-dependent. Indeed, a significantly stronger signal was observed when the reaction was carried out in acetone–[2H6]acetone or acetone–[2H3]acetonitrile mixtures, which is in good agreement with the fact that in deuteriated solvents the lifetime of singlet oxygen is considerably higher than in non-deuteriated ones.12 Thus, the yield of 1O2 was equal to 0.0075% (in acetone–[2H6]acetone, 1:2) and 0.03% (in acetone–[2H3]acetonitrile, 1:2).However, the effect observed was much higher than that expected from the literature data on singlet oxygen life times (t) in these deuteriated solvents.12 Thus, the relative light sums were 1.0:13.9:56.0 in acetone, acetone–[2H6]acetone (1:2) and acetone–[2H3]acetonitrile (1:2), whereas the ratio of t in these solvent mixtures† is 1.0:2.7:2.8.One can assume that apart from an increase in the radiative efficiency of 1O2, an unexpectedly high isotope effect is associated with the interference of acetone or [2H3]acetonitrile during the process.It is likely that in the presence of [2H6]acetone or especially of [2H3]acetonitrile, where the isotope effect is even more pronounced, the reaction changes direction so that it favours more effective formation of singlet oxygen, which results in an additional increase in CL intensity. However, in order to establish exactly how this is occurring more experimental work needs to be done.† Since the reactions were carried out in a mixture of deuteriated and non-deuteriated solvents, the t values were estimated according to the Stern–Volmer equation taking into account the fact that the singlet oxygen lifetime is reduced upon adding acetone to [2H6]acetone or to [2H3]acetonitrile. C CH3 CH3 O O H3C C OCH3 O RuII + C CH3 CH3 O O RuII ...C CH3 CH3 O O RuIII ... RuIII ... H3C C OCH3 O Ru*II ... H3C C OCH3 + RuII + hn (630 nm) O H3C C OCH3 + RuII O (a) (b) Scheme 1 I (relative units) 9000 6000 3000 0 50 100 150 200 250 300 350 400 t/s Figure 1 Typical kinetic curve for the singlet oxygen chemiluminescence decay in the system RuII–DMD. Starting conditions: solvent acetone– [2H3]acetonitrile (1:2), 25 °C, argon atmosphere, [RuII]0 = 6×10–5 mol l–1, [DMD]0 = 1.75×10–2 mol l–1.Mendeleev Communications Electronic Version, Issue 5, 1998 (pp. 169-206) Luminescence decay follows a complex rate law and cannot be described with either first- or second-order kinetics (see Figure 1). One possible path for singlet oxygen formation is the reaction between RuIII and superoxide ion.In fact, it is known that 1O2 is produced quantitatively in the RuIII-superoxide reaction:13 Consequently, if these intermediates are really formed in our case, their interaction should lead to the generation of 1O2. To explain the formation of RuIII, superoxide and singlet oxygen, one can propose the following scheme for the process: Previously, Adam et al.6 established a similar series of reactions [path (a)] for iodide ion-promoted decomposition of TFMD.In contrast to that case, we suggest here that another option available to the superoxide in our system is path (b), namely reaction with RuIII, followed by the regeneration of the ruthenium to its initial oxidation state. It is this reaction that should lead to chemiluminescence of singlet oxygen in our system.It should be pointed out that this is already a secondary event when singlet oxygen is formed in the reactions of dioxiranes. It was previously reported on 1O2 generation in the reaction of DMD with N-oxides, but in that case quite another mechanism operates.14 Thus, DMD, at least partially, participates in the ET process presented in Scheme 2 [paths (a) and (b)].Perhaps this is one of the reasons [along with RuII-induced isomerization of DMD without formation of Ru*II, Scheme 1, (b)] why the chemiexcitation yield of RuII in the CIEEL reaction [Scheme 1, (a)] was found to be rather small (0.01)9 as compared with that in the case of 1,2-dioxetanes (0.2).15,16 One can suppose that interaction of the other transition metals with dioxiranes will also result in decomposition of the latter via the sequence of ET reactions presented in Scheme 2, and in some of these cases (if it is energetically allowed),‡ the formation of singlet oxygen seems to be very possible.The observation of CL of 1O2 in the system DMD–RuII opens a new and promising direction in the investigation of dioxirane properties and certainly awaits further experimentation.‡ We failed to detect IR-CL under decomposition of DMD in the presence of CeIII probably due to the high reduction potential of CeIV/CeIII (1.4 V).17 We were also unable to detect IR-CL during decomposition of DMD (adsorbed from the gas phase) on a silipor surface containing RuII. The diffusion limits, which inhibit the proceeding of the processes indicated in Scheme 2, seem to be responsible for this.However, the visible CL (Ru*II emitter, 630 nm) according to CIEEL (Scheme 1) is observed under these conditions.10 V. P. Kazakov and A. I. Voloshin are grateful to the Russian Foundation for Basic Research for financial support of this investigation (grant no. 96-03-33871). Special thanks are due to Professor W. Adam (University of Wurzburg, Germany) for critical reading of the manuscript.References 1 R. W. Murray, Chem. Rev., 1989, 89, 1187. 2 W. Adam, L. P. Hadjiarapoglou, R. Curci and R. Mello, in Organic Peroxides, ed. W. Ando, J. Wiley & Sons, New York, 1992, ch. 4, p. 195. 3 R. Curci, A. Dinoi and M. F. Rubino, Pure Appl. Chem., 1995, 67, 811. 4 R. Mello, F. Ciminale, M. Fiorentino, C. Fusco, T.Prencipe and R. Curci, Tetrahedron Lett., 1990, 31, 6097. 5 W. Adam, S. E. Bottle and R. Mello, J. Chem. Soc., Chem. Commun., 1991, 771. 6 W. Adam, G. Asensio, R. Curci, M. E. Gonzalez-Nunez and R. Mello, J. Am. Chem. Soc., 1992, 114, 8345. 7 D. V. Kazakov, N. N. Kabalnova, A. I. Voloshin, V. V. Shereshovets and V. P. Kazakov, Izv. Akad. Nauk, Ser. Khim., 1995, 2286 (Russ.Chem. Bull., 1995, 44, 2193). 8 A. M. Nazarov, A. I. Voloshin, G. A. Yamilova, V. D. Komissarov and V. P. Kazakov, Izv. Akad. Nauk, Ser. Khim., 1996, 2593 (Russ. Chem. Bull., 1996, 45, 2462). 9 D. V. Kazakov, A. I. Voloshin, N. N. Kabalnova, V. V. Shereshovets and V. P. Kazakov, Izv. Akad. Nauk, Ser. Khim., 1997, 1138 (Russ. Chem. Bull., 1997, 46, 1089). 10 D. V. Kazakov, A. I.Voloshin, N. N. Kabalnova, V. V. Shereshovets and V. P. Kazakov, Mendeleev Commun., 1998, 49. 11 V. V. Shereshovets, S. S. Ostakhov, N. M. Korotaeva, G. L. Sharipov, V. D. Komissarov, V. P. Kazakov and G. A. Tolstikov, Izv. Akad. Nauk SSSR, Ser. Khim., 1989, 2687 (Bull. Acad. Sci. USSR, Div. Chem. Sci., 1989, 38, 2460). 12 T. A. Jenny and N. J. Turro, Tetrahedron Lett., 1982, 23, 2923. 13 S. S. Miller, K. Zahir and A. Haim, Inorg. Chem., 1985, 24, 3978. 14 W. Adam, K. Briviba, F. Duschek, D. Golsch, W. Kiefer and H. Sies, J. Chem. Soc., Chem. Commun., 1995, 1831. 15 G. L. Sharipov, V. P. Kazakov and G. A. Tolstikov, Khimia i khemiluminestsentsiya 1,2-dioksetanov (Chemistry and chemiluminescence of 1,2-dioxetanes), Nauka, Moscow, 1990, p. 242 (in Russian). 16 A. I. Voloshin, G. L. Sharipov, V. P. Kazakov and G. A. Tolstikov, Izv. Akad. Nauk SSSR, Ser. Khim., 1991, 1316 (Bull. Acad. Sci. USSR, Div. Chem. Sci., 1991, 40, 1158). 17 G. J. Kavarnos and N. J. Turro, Chem. Rev., 1986, 86, 401. RuIII + O2 –· 1O2 + RuII C H3C H3C O O 2H3C C CH3 Scheme 2 C H3C H3C O O C H3 C H3 C O O C CH3 CH3 O O DMD O2 O path (a) DMD 3O2 RuII RuIII 1O2 3O2 hn (1270 nm) path (b) DMD Received: Moscow, 17th February 1998 Cambridge, 18th May 1998; Com. 8/01638K

ISSN:0959-9436     

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 5, 1998 (pp. 169–205) On-line sorption preconcentration and FIA-FAAS determination of palladium and platinum in solution Lubov’ V. Bogacheva, Igor A. Kovalev, Grigory I. Tsysin,* Andrew A. Formanovsky and Yury A. Zolotov Department of Chemistry, M. V. Lomonosov Moscow State University, 119899 Moscow, Russian Federation. E-mail: root@conc.chem.msu.su A flow injection flame atomic absorption method for palladium and platinum determination in solution, including sorption preconcentration, has been developed.Preconcentration of platinum metals as ionic associates with n-octyldiethylenetriamine was carried out on ‘reversed phase’ styrene–divinylbenzene copolymer SSPS. RSD 0.03–0.08, detection limits 5 and 3 mg l–1 (Pd and Pt) for 1 min of preconcentration and sampling frequency up to 40 h–1 were achieved. To determine low concentrations of platinum group metals (PGM) in minerals, preconcentration of ores and alloys is often needed.Sorption preconcentration is widely used for this purpose due to the large preconcentration factors obtained and the simplicity of the procedure.1–5 Dynamic sorption is the most promising method.It does not require phase separation, provides a possibility for automation of the preconcentration process and also for the development of a powerful integrated preconcentration –determination system. However, the variety and remarkable kinetic inertness of PGM complexes in solution complicate their dynamic preconcentration. The distribution coefficients of many metals are rather low and the recovery of PGM is not quantitative.Desorption of PM is slow and difficult; eluents often destroy the sorbent.1,4,5 Sorbent extraction combines the benefits of solvent extraction and solid phase preconcentration while eliminating some drawbacks of the two methods. Sorbent extraction applies a solid support of hydrophobic functionality to recover metal hydrophobic complexes, while elution is achieved by reversing the solvent polarity.This allows the use of a wide range of reagents forming even stronger complexes with metals than immobilized functional groups. Elution may be carried out under mild conditions without concentrated acids or other aggressive reagents.6 Preconcentration of PGM hydrophobic complexes and ionic associates on reversed-phase sorbents offers new possibilities.7,8 Desorption of PGM can easily be carried out with polar eluents, providing an opportunity for on-line combination of preconcentration and determination steps.However, the on-line systems including sorption preconcentration previously proposed for the determination of PGM feature low selectivity.7,9,10 Extraction of PGM anionic chloride complexes was shown to be an effective method for their recovery from complex solutions.11 Amines and quaternary ammonium reagents are the most promising for extraction of PGM.11,12 Extraction efficiency was observed to increase from monoamines to triamines.The possible effectiveness of dynamic sorption preconcentration of PGM was also demonstrated on a polystyrene-based sorbent containing immobilized diethylenetriamine groups (DETA-sorbent).Sorption 100 80 60 40 20 0 1 2 R (%) CHCl /M Figure 1 Effect of HCl concentration on the recovery of PdII on SSPS (circles), silica-C16 (triangles), XAD-2 (squares), XAD-8 (crosses) and PtIV on SSPS (diamonds). CPd = 2 mg ml–1, CPt = 5 mg ml–1, Creag = 5×10–3 M, v = 1 ml min–1, V = 10 ml.aCr eag = 5×10–4M. Table 1 Distribution coefficients of PdCl4 2– ionic associates with amino reagents on ‘reversed-phase’ sorbents. CPd = 2 mg ml–1, CHCl = 1M, Creag= = 5×10–3M, msorb = 20 mg, V = 50 ml, tcont = 1 h; n = 3, RSD£3%. Reagent Sorbent SSPS Amberlite XAD-2 Amberlite XAD-8 Silica C16 C8H17NH2 365 75 247 84 C10H21NH2 960 156 90 145 NH2(CH2)6NH2 150 88 30 62 CH3C(CH2NH2)3 200 20 42 53 C8H17N(CH2CH2NH2)2 8000 160 350 250 C10H21N(CH2CH2NH2)2 3900 790 246 175 C12H25N(CH2CH2NH2)2 a 4000 600 90 340 100 50 0 1 2 CHCl /M R (%) Figure 2 Effect of HCl concentration in eluent on palladium(II) and platinum(IV) desorption.Eluent: methanol (circles for Pd, diamonds for Pt), ethanol (squares for Pd, triangles for Pt), propan-2-ol (crosses for Pd). v = 1 ml min–1, V = 5 ml.Mendeleev Communications Electronic Version, Issue 5, 1998 (pp. 169-206) of PGM from hydrochloric acid solutions on the sorbent is caused by the formation of ionic associates of chloride PGM complexes and the protonated nitrogen of functional groups.13 The purpose of this study was to investigate the dynamic preconcentration of palladium and platinum as ionic associates with reagents containing amines (mono-, di-, tri-) and a hydrophobic part on non-polar sorbents and to develop a flow injection–flame atomic absorption (FIA-FAAS) method for the determination of metals in solutions of complicated composition.Distribution coefficients of ionic associates of PdCl4 2– with various reagents containing amino groups on some reversed-phase sorbents were determined (Table 1).Sorbents Amberlite XAD-2 and XAD-8 (Serva, USA); C16-bonded silica (BioKhimMak, Russia) and SSPS (styrene–divinylbenzene copolymer) (Diapak, Russia) were investigated. The reagents containing a diethylenetriamine part were shown to be essentially more effective than monoamines for recovery of Pd. We suppose that in this case formation of the most hydrophobic species (ionic associates) in solution is caused by binding of a polycharged metal complex with a polycharged protonated molecule of the reagent.The largest distribution coefficient for Pd was achieved with n-octyldiethylenetriamine (Table 1). This reagent was chosen for further investigations. The dependence of Pd and Pt recovery on various sorbents on HCl concentration in solution was studied.The solution was pumped through a sorbent microcolumn (15 mm×2.5 mm i.d.) with a peristaltic pump at a flow rate 1–2 ml min–1. The reagent concentration was 5×10–3 M. The recovery of metals was determined by measuring their residual concentrations by FAAS in solution after sorption. Determination of metals was carried out with a Quant-AFA atomic absorption spectrometer (Cortech, Moscow, Russia).Propane–butane–air and acetylene–air flames were applied for the determination of Pd (247.66 nm) and Pt (265.95 nm), respectively. Continuum spectrum D2–background correction was used. The recovery of Pd and Pt from 0.2–0.5 M HCl is quantitative on SSPS sorbent (Figure 1). The decrease in metal recovery at higher HCl concentrations could be explained by interference from a competitive chloride ion.Quantitative desorption of Pd is achieved with methanol or 1–2 M HCl ethanol solution and for Pt with 1–2 M HCl methanol or ethanol solution (Figure 2). aCpd = 50 mg l–1. bCpd = 100 mg l–1. Table 2 Influence of main matrix elements on FIA-FAAS determination of Pd and Pt. CPd = 200 mg l–1, CPt = 50 mg l–1, CHCl = 0.5M, Creag = 5×10–3 M (P = 0.95, n = 4).Matrix element Matrix element concentration/g l–1 Found Pd/mg l–1 Found Pt/mg l–1 Na 1.0 200±1 51±2 2.5 200±2 50±3 5.0 200±2 52±4 10.0 196±4 41±6 20.0 148±6 — 30.0 120±6 — 40.0 76±8 — Ca 0.5 200±2 53±2 1.0 194±2 49±2 2.0 198±2 55±3 4.0 199±3 57±5 10.0 196±4 54±5 20.0 181±4 53±5 FeIII 1.0 202±2 57±3 2.0 201±2 53±3 5.0 200±2 51±3 10.0 196±3 57±3 20.0 174±4 45±7 30.0 — 32±8 Nia 1.0 51±1 49±2 2.0 52±2 47±2 6.0 49±3 42±2 8.0 50±3 36±5 Cu 1.0 198±1 50±2 2.0 204±2 53±2 5.0 180±6 49±1 10.0 160±8 48±6 20.0 143±8 40±7 CrIII a 1.0 52±3 49±2 5.0 57±9 46±3 10.0 48±4 44±2 20.0 23±5 — 30.0 15±6 — CrIV b 0.1 107±3 50±7 0.3 101±3 36±5 0.5 102±4 25±1 1.0 30±5 — 1.5 20±5 — SO4 2–/Mb 0.03 — 47±2 0.05 97±3 63±6 0.1 101±2 60±4 0.2 103±6 63±6 0.4 94±7 — 0.6 — 65±4 water sample eluent rinsing solution AAS P I1 I2 column loopback waste (a) (b) (c) loopback water sample eluent rinsing solution AAS P I1 I2 column waste loopback water sample eluent rinsing solution AAS P I1 I2 column waste loopback loopback Figure 3 Manifold system for FIA-FAAS determination of Pd and Pt.(a) preconcentration, (b) rinsing, (c) elution; P is a peristaltic pump, I1 and I2 are injection valves.Sample is the analysed solution containing of reagent (0.5 M HCl), eluent: 1–2 M HCl in ethanol, rinsing solution: 5×10–3 M reagent solution in 0.5 M HCl.Mendeleev Communications Electronic Version, Issue 5, 1998 (pp. 169–205) The flow injection–FAAS determination of palladium and platinum was carried out using a commercial automated complex consisting of a Quant-AFA atomic absorption spectrometer, a flow injection device, an IBM compatible computer and the original software.An original manifold system for FIA-FAAS determination of metals including the on-line sorption preconcentration was proposed (Figure 3). The determination cycle consists of preconcentration, rinsing and desorption steps. The solution analysed containing reagent (0.5 M HCl, reagent concentration 5×10–3 M), rinsing solution (5×10–3 M reagent solution in 0.5 M HCl) and eluent (1–2 M HCl solution in ethanol) were sequentially pumped through a microcolumn. The tubes connecting the FIA device and nebulizer of the spectrometer were filled with eluent simultaneously with the column rinsing.After desorption, metal concentrate in ethanol (methanol) was introduced directly into the detector. The width of the concentrate zone was about 300 ml.Peak area was accepted as a value of the analytical signal. The dependence of the analytical signal on Pd and Pt concentration is linear in the metal concentration range 0–200 mg l–1. The influence of the main matrix elements on the determination of Pd and Pt was investigated (Table 2).The average concentrations of macrocomponents in the solutions obtained after digestion of alloys and ores are up to 8–10 g l–1 for Na, 4–5 g l–1 for Ca, 10–12 g l–1 for Fe, 2–3 g l–1 for Ni and 2–3 g l–1 for Cu. Thus the determination of Pd and Pt was found to be possible under high concentrations of the main components of ores and alloys (Table 2).The accuracy of the FIA-FAAS method proposed was proved by the determination of Pd and Pt in standard reference materials of ores and alloys (Table 3). Good agreement of the results obtained with the certified values was demonstrated. RSD of the determination is 0.03–0.08 and detection limits are 5 mg l–1 (propane–butane–air flame) and 3 mg l–1 (acetylene–air flame) for Pd and Pt for 1 min of preconcentration.The authors express their gratitude to the Russian Foundation for Basic Research (grant no. 97-03-33225a) for partial financial support. References 1 Z. Su, X. Chang, K. Xu, X. Luo and G. Zhan, Anal. Chim. Acta, 1992, 268, 323. 2 L. Elci, Anal. Lett., 1993, 26, 1025. 3 K. Kritsotakis and H. J. Tobschall, Fresenius’ Z. Anal. Chem., 1985, 320, 15. 4 X.Chang, X. Luo, G. Zhan and Z. Su, Talanta, 1992, 39, 937. 5 D. K. Singh and N. K. Mishra, Chromatographia, 1991, 31, 300. 6 H. L. Lancaster, G. D. Marshall, E. R. Gonzalo, J. Ruzicka and G. D. Christian, Analyst, 1994, 119, 1459. 7 M. L. Lee, G. Toelg, E. Beinrohr and P. Tschoepel, Anal. Chim. Acta, 1993, 272, 193. 8 G. I. Malofeeva, O. M. Petrukhin, L. S. Rojkova, B. Ya. Spivakov, G.K. Genkina and T. A. Mastryukova, Zh. Anal. Khim., 1996, 51, 1038 (J. Anal. Chem., 1996, 51, 949). 9 H. Mikai, Y. Ambe and M. Morita, J. Anal. At. Spectrom., 1990, 5, 75. 10 A. Cantarero, M. M. Gomez, C. Camara and M. A. Palacios, Anal. Chim. Acta, 1994, 296, 205. 11 V. V. Belova, A. I. Kholkin, T. I. Jidkova, T. P. Sidorova and N. G. Aleksandrova, Abstracts of International Symposium ‘The problems of complex ores utilization’, St. Petersburg, 1994, p. 31. 12 B. K. Tait and D. P. Shillington, S. Afr. J. Chem., 1992, 45, 17. 13 I. A. Kovalev, G. I. Tsysin and Yu. A. Zolotov, Mendeleev Commun., 1995, 111. aPropane–butane–air flame. bAcetylene–air flame. Table 3 Results of FIA-FAAS determination of Pd and Pt in standard reference materials of alloys and ores (P = 95%, n = 4). Sample Pda Ptb Certified/mg g–1 Found/mg g–1 RSD (%) Certified/mg g–1 Found/mg g–1 RSD (%) Copper alloy 2900±100 2800±100 3 1000±100 900±100 7 Copper alloy 1000±100 1000±100 6 300±20 310±40 8 VP-2 6.6±0.5 6.3±0.4 4 2.6±0.3 2.5±0.3 8 G-3 5.6±0.3 5.8±0.4 4 1.2±0.1 1.1±0.1 6 Received: Moscow, 2nd April 1998 Cambridge, 10th June 1998; Com. 8/02791I

ISSN:0959-9436     

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 5, 1998 (pp. 169–205) Kinetic mechanism and chemical oscillations in the branching chain decomposition of nitrogen trichloride Nikolai M. Rubtsov Institute for Structural Macrokinetics, Russian Academy of Sciences, 142432 Chernogolovka, Moscow Region, Russian Federation. Fax: +7 095 962 8045; e-mail: ab3590@mail.sitek.ru On the basis of numerical simulation of the branching chain process of the decomposition of nitrogen trichloride it has been shown that suitable conditions for providing oscillating regimes are the following: (a) inclusion of adsorption–desorption of NCl3 on the reactor walls; (b) inclusion of nonlinear chain branching and breaking reactions.The occurrence of nonlinear homo- and heterogeneous reactions determines the behaviour of the great majority of branching chain processes (BCP) as nonlinear dynamical systems.It shows up in the rise of autowave regimes (e.g. nonthermal flame propagation1) and structural organization as chemical oscillations in flow conditions and enclosed volumes. However, only the oscillations in liquid phase reactions in open systems have been adequately investigated.2 The trends in the initiation and development of oscillations in an enclosed volume are not clearly understood.One of the causes considered is heat evolution,3 however, in BCP at low pressures energy is accumulated in the active intermediates, and in this case the warming-up is negligible.1,4 The chemical isothermic oscillations are generally caused either by autocatalysis with active intermediates2,5 or by a nonstationary surface state4 which is due to nonlinear heterogeneous reactions involving adsorbed intermediates.Oscillating regimes are inherent in gaseous BCP such as phosphorus, CO,4 silane and dichlorosilane oxidation,6,7 and NCl3 decomposition.8,9 The consideration of gaseous BCP allows the use of deterministic equations (ODE) without regard for fluctuations.2 It is known that the thermal decomposition of NCl3 is an example of the branching chain low temperature decomposition of an individual substance in the gaseous phase, in which nonlinear chain branching plays an important role,10,11 leading in particular to nonthermal flame propagation11,12 and chemical oscillations.8,9 The number of elementary reactions in the kinetic mechanism is comparatively low, and the rate constants of most of them are known.9–15 It has also been shown that electronically excited Cl2 3P+ ou generated by this reaction takes part in chain branching, and in this case the fast quenching process Cl + Cl2 3P+ ou ® Cl + Cl2 1Sg – is of importance in chain breaking.11,16,17 This means that the decomposition of NCl3 does not exhibit the peculiarities of linear BCP as, for instance, oxidation of H2.1,4 Therefore, the revealing of common trends in nonlinear gaseous BCP is of interest with respect to the theory of chemical transformation.The practical utility of this BCP lies in laserochemical applications18 and safety in explosions.19 Recently it has been shown8,9 that mixtures of 3–5% NCl3 with He produce oscillatory self-ignitions in a closed vessel, if the surface is treated with NaCl.The warming-up does not exceed 5 °C, therefore the oscillations are chain in nature. The fact that NCl3 appears in the enclosed volume after each individual oscillation is unambiguous evidence that the adsorption and desorption processes involving NCl3 cause the oscillations.The chemical oscillations observed may be damped out, amplified or self-sustained with the position of the reaction mixture in the self-ignition area, and in this case the periods and amplitudes of oscillations strongly depend on the number of previous ignitions in the vessel. This work is aimed at a numerical calculation of the observed regimes of chemical oscillations in enclosed vessels treated with NaCl at low pressures (< 10 Torr) and 293 K in NCl3 decomposition (3–5% NCl3 in He).8,9 We tried to clarify whether the set of elementary steps known combined with an external y3 y0 y1 y2 1.0 0.5 0.0 6000 6200 6400 6600 6800 t 0.2 0.1 (a) (b) 6000 6200 6400 6600 6800 t 0.3 0.0 Figure 1 Numerical simulation of the self-ignition of NCl3: dimensionless coordinates t, time; y0, chlorine atoms; y1, Cl2 3P+ ou; y2, NCl2; y3, NCl3.(a) The dependence of y3 on t; (b) the dependence of y0, y1, y2 on t. The dimensionless parameters w = k0 /k1(NCl3)0; b = k2 /k1; f = k3 /k1; g = = g0(1 – mt) = k7/k1(NCl3)0; l = k4 /k1(NCl3)0; y = k5 /k1; d = k8 /k1(NCl3)0; t = k1(NCl3)0t; x = k6 /k1, c = k9 /k1(NCl3)0, a = a0(1 – ht) are also defined in equations (I) and (II).In subsequent Figures the definitions of the variables and parameters are the same. The calculated lower self-ignition limit occurs at g = 2.4×10–3 under these conditions. w º 5×10–8, c º 0.25, x º 100, a º 0, t0 = 0, t1 = 70000, b º 0.8×10–4, g º 2×10–4, y º 0.2, h º 0, m º 0, d º 7×10–4, f º 0.35, l º 0.4, N = 70000 w º10–6, c º0.25, x º100, a º3.5×10–5, t0 = 0, t1 = 240000, b º0.8×10–4, g º9×10–4, y º0.2, h º0.5×10–6, m º0.1×10–6, d º7×10–4, f º0.35, l º0.4, N = 240000 y3 0.2 0.1 0.0 0.00 0.02 0.04 0.06 0.08 y2 Figure 2 Calculated oscillatory regime with the depletion of surface and change in surface state through the desorption of NCl3 taken into account: m, h > 0; the phase portrait includes an unstable focus inside stable limit cycle.Mendeleev Communications Electronic Version, Issue 5, 1998 (pp. 169-206) source are sufficient for modelling the observed oscillation regimes, as well as whether the analysis of the one-dimensional system of ODE can give any new information about this process. The kinetic mechanism of the thermal decomposition of NCl3 at low pressures (the rates of termolecular chain break reactions are negligible) can be represented as follows:9,12,14–17,20 The nonlinear chain branching process [steps (3)–(5)] is shown in accordance with refs. 4 and 9–12. These steps take into account either the high probability of the depletion of the upper vibronic levels Cl2 3P (v > 13) in reaction (4), due to the ‘shallow depth’ of the potential-energy surface of this excited state (~7 kcal mol–1), or the fact that the energy released in reaction (5) is enough for dissociation of the NCl3 molecule.20 We determined the dimensionless variables as t = k1(NCl3)0t (t/s); Y0 = (Cl)/(NCl3)0; Y1 = (Cl2 3P)/(NCl3)0; Y2 = (NCl2)/(NCl3)0; Y3 = (NCl3)/(NCl3)0 and the dimensionless parameters as w = = k0 /k1(NCl3)0; b = k2 /k1; f = k3 /k1; g = k7 /k1(NCl3)0; l = = k4 /k1(NCl3)0; y = k5 /k1; d = k8 /k1(NCl3)0; x = k6 /k1, c = = k9/k1(NCl3)0.The corresponding ODE for the kinetic mechanism above are: Initial conditions were Y0 = Y1 = Y2 = 0, Y3 = 1. It is evident that the inclusion of the reaction gives another negative linear term in the second equation of system (I) of the eY1 type. Therefore the step (10) is incorporated implicitly by varying c over a rather wide range, see step (9).In addition, the transition 3P ® 1S is forbidden, i.e. the rate of step (10) is comparatively moderate. In fact, as shown in studies of Cl2 3P generated by rf-discharge in Cl2–He (Ar, Cl2) mixtures, reaction (6) is the predominant quenching process.20 The real situation for numerical modelling matched the experimental conditions:8,9 5 Torr, 3% NCl3 at 293 K.It is easy to verify that 1 s corresponds to ~7000 time units along the ‘t’ axis. ‘1’ along the ‘Y’ axis corresponds to ~4.5×1015 cm–3. t0 and t1 in Figures 1–4 correspond to initial and final values of ‘t’ and N is a number of solution values on [t0, t1]. System (I) was calculated by the fourth order Runge–Kutta method. Kinetic curves of the changes in concentrations of NCl3 and other intermediates are shown in Figure 1.As seen from Figure 1, the calculated concentrations of NCl3 and intermediates during the self-ignition and the lower self-ignition limit are in quantitative agreement with the experimental data, namely, the concentrations of NCl2 and Cl2 3P peak simultaneously,10 and the rate of decrease of NCl3 at its maximum coincides with the maximum of Cl2 3P; in this case the maximum concentration of chlorine atoms is achieved later.21 The maximum NCl2 concentration amounts to ~1015 cm–3 in accordance with ref. 10, and the concentration of chlorine atoms amounts to several tens of a percent of the initial NCl3 concentration.22 The Cl2 3P concentration is markedly low in comparison with that of other intermediates, which is in agreement with the low quantum yield of Cl2 3P measured in ref. 23. It should be noted that the author of ref. 23 formed the conclusion that Cl2 3P plays no marked role in NCl3 decomposition only on the basis of the low quantum yield values observed. However, as seen from the calculations performed, the low quantum yield is either a result of the occurrence of the fast reaction (6), which causes the overall concentration of Cl2 3P to decrease, or a consequence of the fact that the 3P ® 1S transition is forbidden (see above).The calculated value of the induction period (~1 s) is also in agreement with the experimentally observed value15 for the aforementioned conditions. From the above it might be assumed that the ODE (I) system based on the kinetic mechanism presented is entirely suitable for the fitting of experimental data.Experimentally observed desorption of NCl3 was simulated as follows. Into the fourth equation of system (I) for Y3 an additional term was included to represent the rate of desorption of NCl3:24 Evidently, for the adsorption rate of a substance A wads = = kads pA(1 – q), where pA is the partial pressure of A in the gas phase, and (1 – q) is the proportion of free surface; for desorption wdes = kdesq.If adsorption processes are fast, an equilibrium takes place: kads pA(1 – q) = kdesq, from which it follows that q = bA pA/(1 + bA pA), bA = kads /kdes, so wdes = = kdesbA pA/(1 + bA pA). The latter equation for bA pA ~ 1 might be approximated as wdes = kpA 1/m, m > 1. w º10–6, c º 0.25, x º 100, a º 3.5×10–5, t0 = 0, t1 = 240000, b º 0.8×10–4, g º 8.8×10–4, y º 0.2, h º 0.5×10–6, m º 0.1×10–6, d º 7×10–4, f º 0.35, l º 0.4, N = 240000 y3 0.2 0.1 0.0 0.00 0.02 0.04 0.06 0.08 y2 Figure 3 Calculated oscillatory regime under conditions similar to Figure 2, except g = 8.8×10–4 (the surface state has changed in the previous selfignition); the phase portrait includes an unstable focus inside a stable limit cycle.NCl3 NCl2 + Cl Cl + NCl3 NCl2 + Cl2 NCl2 + NCl3 N2 + Cl2 + 3Cl NCl2 + NCl2 N2 + Cl2 3P+ ou +2Cl Cl2 3P+ ou 2Cl Cl2 3P+ ou + NCl3 NCl2 + Cl + Cl2 Cl2 3P+ ou + Cl Cl2 1Sg – + Cl NCl2 reactor wall Cl reactor wall Cl2 3P+ ou + M deactivation k0 = 10–4–2×10–5 s–1,15 k1 = 1.6×10–12 cm3 s–1,15 k2 = 1.3×10–16 cm3 s–1, linear chain branching14,16 k3 = 6×10–13 cm3 s–1, nonlinear chain interaction12 k4 = 104 s–1,20 k5 = 10–12–10–15 cm3 s–1, expected reaction9 k6 = 1.6×10–10 cm3 s–1, nonlinear chain breaking20 k7 = 2–16 s–1,16,17 k8 = 1 s–1,12,16 k9 = 8×10–13–10–13 cm3 s–1,20 (1) (2) (0) (3) (4) (5) (6) (7) (8) (9) dY0/dt = wY3 – Y0Y3 + 3bY2Y3 + 2fY2 2 + yY1Y3 + lY1 – dY0 – xY0Y1 dY1/dt = fY2 2 – lY1 – yY1Y3 – xY0Y1 – cY1 dY2/dt = wY3 + Y0Y3 – bY2Y3 – yY1Y3 – gY2 dY3/dt = –wY3 – Y0Y2 – bY2Y3 – yY1Y3 (I) Cl2 3P Cl2 1S + hn (10) w º 10–6, c º 0.25, x º 100, a º 3.5×10–5, t0 = 0, t1 = 240000, b º 0.8×10–4, g º 9.1×10–4, y º 0.2, h º 0.5×10–6, m º 0.8×10–6, d º 7×10–4, f º 0.35, l º 0.4, N = 240000 y3 0.02 0 0.005 0.010 0.015 y2 Figure 4 Calculated regime of damped oscillations; the phase portrait includes a stable focus.dY3/dt = aY3 1/m, m = 2–6 (II)Mendeleev Communications Electronic Version, Issue 5, 1998 (pp. 169–205) Note that the change in m from 3 to 6 causes a displacement of the area of the existence of oscillatory solutions to lesser a values, i.e. the increase in m requires choice of another (lesser) a value. In addition, the change in k0 and k9 on the above intervals has a comparatively slight effect only on the scale of the calculated curves.It has been shown that in the absence of the desorption term the oscillatory solutions of (I) are missing, and in this case NCl3 reacts completely during the first self-ignition. On the other hand, when the source of NCl3 is included in (I), slowly damping oscillations arise at discrete values of a; their period and intensity in 5 – 10 s become fixed.This is evidence for the existence of the stable limit cycle. It should be noted that the following conditions are necessary for the oscillatory solutions to exist: k4, k6 > 0 (l, x > 0), k5 > 10–13 cm3 s–1 (y > 0.03), therefore the step (5) must be taken into account. These conditions are evident for experimentally investigated steps (4) and (6),20 however, the calculations predict the existence of reaction (5).Outside the interval of a values corresponding to oscillatory solutions, a ‘stationary combustion’ takes place after the first self-ignition. During this process the concentrations yi (0 < i < 3) are constant in time and do not exceed 4×10–3 on the scale of Figures 2–4. Evidently, the amount of NCl3 adsorbed on the reactor surface decreases in the course of the reaction and in doing so, the desorption rate of NCl3 also decreases.Thus, a was represented as a = a0(1 – ht), where h = (3–5)×10–7 and t is dimensionless time, i.e. essentially using the first term of the expansion in a series of some monotonic dependence that actually takes place. It has been shown that with this perturbation allowed for, two types of oscillations arise, depending on the k7 (g) value: amplified or damped oscillations with increasing or decreasing k7, respectively. Therefore, at h > 0 the above oscillatory solution falls into two types: the phase portrait of the first type includes an unstable focus located inside a stable limit cycle; and the phase portrait of the second type includes a stable focus.Notice that both regimes were experimentally observed.8,9 However, in this case the amplification of an oscillation leads to the attainment of an initial concentration of NCl3 (y3 = 1), but this does not occur experimentally. In this connection the evident variation in the surface state was taken into account. In fact, in the course of desorption, molecules of NCl3 leave the surface, and so if at the beginning of the oscillations the reactor surface was covered predominantly with NCl3, at the end of the oscillatory regime this surface would be chemically similar to NaCl.Therefore, the rate constants of the heterogeneous chain break of NCl2 and Cl must change. It was found that in the solution of system (I) the regularities of self-ignition depend on k7 more drastically than on k6 (at 0.5 < k6 < 3 s–1), hence only the change in k7 was considered.This change was included much as it was performed above. g was represented as g = g0(1 – mt), m = (1–10)×10–7, t is dimensionless time. The calculated curves are shown in Figures 2–4. It is seen from Figures 2–4 that the inclusion of the change in surface state leads to a qualitative agreement between calculated and experimentally observed oscillatory regimes.8,9 It is also seen from Figures 2 and 3 that the nature of the amplification of the oscillations changes with a change in the rate ‘constant’ of the heterogeneous chain break k7 in agreement with refs. 8 and 9. As this takes place, however, the change-over from damped to amplified oscillations is defined not only by h (h values are the same for Figures 2–4), but also by the value of m (see Figures 2 and 4), so the phase portrait is fairly complicated.Notice that the kinetic trends for the changes in concentrations of NCl3 and intermediates for each specific oscillation, regardless of the character of the regime, are identical to those presented in Figure 1.It is obvious that the negative sign of m is prescribed arbitrarily because of a lack of experimental data on the recombination of atoms and radicals on surfaces covered with NCl3. This means that further numerical fitting would be worth little, since various assumptions about either change in surface state or the real mechanism of NCl3 desorption [instead of the rather simple model (II)] would be required.Strictly speaking, treatment of a two-dimensional problem would be more correct. Therefore, it has been shown that the modelling of oscillatory regimes in NCl3 decomposition requires not only an external source to exist, but also a consideration of nonlinear chain branching and breaking as well as a change in the state of the reactor surface.I would like to thank Professor V. V. Azatyan (Institute for Structural Macrokinetics of the RAS) for many useful discussions. The work was supported by the Russian Foundation for Basic Research (grant no. 96 -03 -32791a). References 1 N. N. Semenov, O nekotorykh problemakh khimicheskoi kinetiki i reaktsionnoi sposobnosti (On some problems of chemical kinetics and reaction ability), Academy of Sciences of the USSR, Moscow, 1968, p. 686 (in Russian). 2 V. K. Vanag, Fluktuatsionnaya kinetika, kolebatel’nye reaktsii i khimicheskie nestabil’nosti v makroob’eme kak sisteme vzaimodeistvuyushchikh mikroob’emov (Kinetics of fluctuations, oscillatory reactions and chemical unstabilities in macrovolume as a system of interacting microvolumes), Dr. Sci.Thesis, Institute of Chemical Physics RAS, Moscow, 1997, p. 222 (in Russian). 3 D. A. Frank-Kamenetsky, Diffuziya i teploperedacha v khimicheskoi kinetike (Diffusion and heat transfer in chemical kinetics), Nauka, Moscow, 1974, p. 491 (in Russian). 4 N. N. Semenov, L. B. Soroka and V. V. Azatyan, Dokl. Akad. Nauk SSSR, 1977, 237, 152 [Dokl. Chem. (Engl. Transl.), 1977, 237, 217]. 5 S.K.Scott, Acc. Chem. Res., 1987, 20, 187. 6 R. G. Aivazyan and V. V. Azatyan, Kinet. Katal., 1994, 35, 17 [Kinet. Catal. (Engl. Transl.), 1994, 35, 257]. 7 V. V. Azatyan, N. M. Rubtsov, O. T. Ryzhkov and S. M. Temchin, Kinet. Katal., 1996, 37, 805 [Kinet. Catal. (Engl. Transl.), 1996, 37, 796]. 8 V. V. Azatyan, R. R. Borodulin and N. M. Rubtsov, Kinet. Katal., 1980, 21, 316 [Kinet. Catal.(Engl. Transl.), 1980, 21, 241]. 9 N. M. Rubtsov, Vzaimodeistvie reaktsionnykh tsepei v protsessakh goreniya i nizkotemperaturnogo osazhdeniya v otsutstvie i pri nalichii fizicheskogo stimulirovaniya (Chain interaction in the processes of combustion and low temperature chemical deposition both with and without physical stimulation), Dr. Sci. Thesis, Institute for Structural Macrokinetics, Chernogolovka, 1997, p. 350 (in Russian). 10 V. V. Azatyan, R. R. Borodulin and N. M. Rubtsov, Dokl. Akad. Nauk SSSR, 1979, 249, 1375 [Dokl. Chem. (Engl. Transl.), 1979, 249, 1265]. 11 V. V. Azatyan, R. R. Borodulin and N. M. Rubtsov, Fizika Goreniya i Vzryva, 1980, 5, 34 (in Russian). 12 Z. I. Kaganova and B. V. Novozhilov, Khim. Fiz., 1982, 1, 1110 (in Russian). 13 T. C.Clark and M. A. A. Clyne, Trans. Far. Soc., 1970, 66, 372. 14 V. V. Azatyan, R. R. Borodulin, E. A. Markevich, N. M. Rubtsov and N. N. Semenov, Dokl. Akad. Nauk SSSR, 1975, 224, 1096 [Dokl. Chem. (Engl. Transl.), 1975, 224, 1059]. 15 N. M. Rubtsov, V. V. Azatyan and R. R. Borodulin, Izv. Akad. Nauk, Ser. Khim., 1980, 1234 (Bull. Acad. Sci. USSR, Div. Chem. Sci., 1980, 29, 1165). 16 N. M. Rubtsov, R. R. Borodulin and S. S. Saidchanov, Khim. Fiz., 1984, 3, 521 (in Russian). 17 M. A. A. Clyne and D. H. Stedman, Trans. Far. Soc., 1968, 64, 2698. 18 R. W. Shwenz, J. V. Gilbert and R. D. Coombe, Chem. Phys. Lett., 1993, 207, 526. 19 A. A. Krasheninnikova, L. A. Furman and G. S. Yliankina, Zh. Prikl. Khim., 1971, 44, 2183 (in Russian). 20 M. A. A. Clyne and D. H. Stedman, Trans. Far. Soc., 1968, 64, 1816. 21 V. V. Azatyan, R. R. Borodulin, E. A. Markevich and N. M. Rubtsov, Fizika Goreniya i Vzryva, 1978, 14, 20 (in Russian). 22 R. R. Borodulin, E. A. Markevich, V. V. Azatyan and N. N. Semenov, Kinet. Katal., 1976, 17, 834 [Kinet. Catal. (Engl. Transl.), 1976, 17, 730]. 23 E. A. Markevich, Kinet. Katal., 1986, 27, 729 [Kinet. Catal. (Engl. Transl.), 1986, 27, 631]. 24 I. A. Semiochin, B. V. Strachov and A. I. Osipov, Kinetika khimicheskikh reaktsii (Kinetics of chemical reactions), Izd. Moskovskogo Universiteta, Moscow, 1995, p. 347 (in Russian). Received: Moscow, 12th March 1998 Cambridge, 18th June 1998; Com. 8/02188K

ISSN:0959-9436     

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 5, 1998 (pp. 169–205) Line tension and capillary pressure in foams Valery V. Krotov and Anatoly I. Rusanov* Department of Chemistry, St. Petersburg State University, 199034 St. Petersburg, Russian Federation. Fax: +7 812 428 6939; e-mail: rusanov@rus.usr.pu.ru The line tension of Plateau borders is introduced and a general equation for the capillary pressure in a foam is derived, allowing for line tension and including the case of thin films typical of polyhedral foams and emulsions.Films and Plateau borders are the principal elements of polyhedral foams and concentrated emulsions. Introducing a unique dividing surface for each film and a dividing line for each Plateau border, one can consider a foam as a network of geometrical surfaces possessing film tension gf and lines possessing line tension k.The description of a foam in these terms and the derivation of a general equation for the foam-cell capillary pressure are the goals of this work. A rectilinear Plateau border is of the shape of a triangular column with concave surfaces. Assuming the foam films to be very thin, and neglecting the contact angle at the junction of the Plateau border and film surfaces, one can approximate a Plateau border as the space, filled with a liquid (phase a), between three touching cylinders of phase b of radius r, whose cross-section is shown in Figure 1.In this model, the cross-sectional area Ab of a Plateau border is If each film of a foam is depicted by a single dividing surface (the lines marked with gf in Figure 1), their meeting place (point O in Figure 1) should be a line, and a corresponding line tension can be derived.To do this, we first calculate the total force, fb, acting through the cross-section of a Plateau border. Such a force is the combination of surface and bulk contributions. The surface part is clearly prg since the perimeter of the circular triangle (Figure 1) is pr.According to the Laplace equation, the stress inside the Plateau border is –pa = –pb + g/r (p is pressure and g is surface tension) and the corresponding force is (–pb + g/r)Ab. So, the total force is where equation (1) has been used. Introducing the Plateauborder line tension as an excess quantity, we have to compare the above force (acting across the left part of Figure 1) with the force acting across the right part of Figure 1 and equal, evidently, to –pbAb + 3gflb where lb = r/30.5 is the length of the bisector of the circular triangle calculated from its centre (Figure 1).Subtracting this force from equation (2), we obtain the Plateau-border line tension k as Since the surface tension of liquids is always positive, equation (3) predicts a negative value for the Plateau-border line tension.It is of note that the latter has nothing to do with ordinary interfacial line tension (which we neglected in the above calculation) and is by several orders of magnitude larger in value: e.g. equation (3) yields k = –8×10–7 N at g = 50 mN m–1 and r = 0.1 mm. Generally, the Laplace equation gives the pressure difference in neighbouring phases or cells, speaking about a foam (then the surface tension is replaced by film tension), i.e.it defines the capillary pressure of a foam cell with respect to the pressure in the neighbouring cell. The latter, however, is not under control in the experiment, so it is convenient to introduce a more general definition of the capillary pressure of a foam cell, pc, as the difference between the pressure inside the cell and the outer (say, atmospheric) pressure.Derjaguin1 was the first to suggest an expression for such a capillary pressure of the foam cell as where s is the specific surface area of a foam. However, equation (4) is inexact in two respects. Firstly, it refers to thick films, whereas the presence of thin films is typical for polyhedral foams.Thin films possess a disjoining pressure and their tension is not 2g as is implied in equation (4). Secondly, equation (4) does not take into account linear phenomena in foams. Although the influence of line tension can be small, it should be represented in a rigorous relationship. To derive a more general expression for the foam-cell capillary pressure, let us consider the variation of the volume of a polyhedral monodisperse foam which is represented as part of a space compactly filled with polyhedrons.If l is a certain linear parameter of a single polyhedron, the total length of its edges (Plateau borders) Lp, the polyhedron surface area Ap and the polyhedron volume v are related to l by the relationships where k1, k2 and k3 are proportionality coefficients.Then, we have from equation (5) At a fixed temperature and amounts of all components, the fundamental thermodynamic equation for the free energy F of the system, including the polyhedral foam and a surrounding medium, is where p' is the outer pressure acting on the foam, V' is the volume of the surrounding medium, p is the pressure inside the foam polyhedrons, V is the foam volume (equal to the sum of the volumes of all polyhedrons), A is the total surface area of the polyhedral network and L is the total line length of the network.The Gibbs equilibrium principle is formulated in this case as Putting equation (7) in (8), we arrive at the mechanical equilibrium condition a b gf r b b gf gf O b b b gf gf gf lb Figure 1 The cross-section of a Plateau border and passing to its linear image.Ab = r2(30.5 – p/2) = 0.1612r2 (1) ~ fb = prg + (–pb + g/r)Ab = –pbAb + (30.5 + p/2)gr (2) k = (p/4 – 30.5/2)gfr = –0.0806gfr ~ (3) pc = 2gs/3 (4) Lp = k1l, Ap = k2l2, v = k3l3 (5) dlnL/dlnv = 1/3, dlnAf/dlnv = 2/3 (6) dF = –p'dV' – pdV + gfdA + kdL (7) dFV + V' = 0 (8) p – p' = gfdA + kdL = gfa + kl dlnA dlnV dlnA dlnV (9)Mendeleev Communications Electronic Version, Issue 5, 1998 (pp. 169-206) where a º A/V and l º L/V are the surface area and the line length per unit volume, respectively. Let us calculate the derivatives on the right-hand side of equation (9). If the number of polyhedrons is n, the foam volume is V = nv where v is the volume of a single polyhedron. Since each polyhedron’s face, except the outer faces, belongs simultaneously to two adjacent polyhedrons, the total surface area of the network is A = (n/2)Ap where Ap is the surface area of a single polyhedron.Similarly, since each line (except the outer lines) belongs simultaneously to three polyhedrons according to the Plateau rule, L = (n/3)Lp where Lp is the total length of edges of a single polyhedron (n is suggested to be large enough to neglect the outer foam cells).Therefore, for variations not influencing the number of polyhedrons, we have Then, according to equation (5), we have Now putting equation (11) in (9), we obtain Equation (12) is a general expression for the capillary pressure of the foam cell. Let us consider some particular and approximate forms of equation (12).Roughly regarding the films of a polyhedral cell as thick films, and setting gf ª 2g, equation (12) becomes where r is the curvature radius of the Plateau-border surface and the last term has been estimated with the aid of equation (3). Since typically lr < a in a polyhedral foam, the last term in equation (13) plays the role of a small correction. Passing to real polyhedral foams, a is approximately the specific film area in a foam.The corresponding specific surface area of a foam s is twice as much as a (since each film has two sides), so equation (13) can also be written, neglecting the last term, in the form of equation (4). As seen from equation (13), a more exact form of Derjaguin’s equation is It is of note that the above derivation of equation (12) is valid at an arbitrary outer phase surrounding the foam. The outer phase can be the same gas as inside the foam bubbles or another gas (say, the air) or a liquid phase including the case when the outer phase coincides with the continuous phase of the foam (a foam in the bulk of a solution). The concept of capillary pressure expressed in equation (12) is also applicable to polyhedral concentrated emulsions. References 1 B. V. Derjaguin, Zh. Fiz. Khim., 1931, 2, 745 (in Russian). dlnV = dlnv, dlnA = dlnAp, dlnL = dlnLp (10) dlnA dlnV dlnAp dlnv = =2 3 , dlnL dlnV dlnLp dlnv = =1 3 (11) pc º p – p' = 2gfa/3 +kl/3 (12) pc ª 4ga/3 + kl/3 ª 4ga/3 – 0.054glr (13) pc ª 2gfa/3 ª gfs/3 (14) Received: Moscow, 15th June 1998 Cambridge, 17th July 1998; Com. 8/05508D

ISSN:0959-9436     

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 5, 1998 (pp. 169–205) Electric probe detection of large cluster ions in spinodal decomposition of the laser-induced labile liquid phase of carbon Sergei I. Kudryashov* and Nikita B. Zorov Department of Chemistry, M. V. Lomonosov Moscow State University, 119899 Moscow, Russian Federation. E-mail: serg@laser.chem.msu.su The electric probe procedure developed here has made it possible to detect for the first time large cluster ions, the products of spinodal decomposition of the laser-induced labile state of the liquid phase of carbon, with a size up to a million atoms and relative content in the mass distribution of the order of ppm.In recent decades, interest in the problem of preparing nanocrystalline materials has significantly increased, because it has been found that a decrease in the crystallite size lower than some threshold value (of the order of magnitude of 10 nm) results in a considerable change in the optical, magnetic and elastic properties of the material.1 At the present time, compacting of isolated nanoclusters, crystallisation of amorphous alloys and intense plastic deformation of materials2–4 are the main methods for the preparation of nanocrystalline materials. Spinodal decomposition of the thermodynamically unstable (labile) state of the liquid phase of a substance is one of the methods used for preparing isolated nanoclusters formed as nuclei of the liquid phase.5 Macroscopic effects related to the spinodal decomposition of the laser-induced labile state of the liquid phase of carbon were considered previously;6 however, the composition of spinodal decomposition products has not been studied to date.So far, the study of the high mass distribution of cluster ions is restricted by the limited range of detected masses for most of the commercially available time-of-flight mass spectrometers. Upper values of detected masses M for the instruments are not usually higher than 104 a.m.u.due to the low sensitivity of the detection systems used (secondary electron multipliers, multichannel plates)7 in the high mass range due to the low efficiency of ion-electron conversion h ~ M–0.5.8 Electric probe detection has not previously been used for detecting large cluster ions because of low mass resolution; nevertheless, it has considerable advantages for studying the general modal character of the mass distribution: the h value is close to unity regardless of the mass of the detected ion and the position of the electric probe (collector) near the target ablated by laser radiation allows the sensitivity of the detection system to be further increased due to the corresponding increase in ion collection efficiency.9 These advantages of the electric probe method were used in this work to study the high mass cluster ion distribution of the gas phase spinodal decomposition products in the laser evaporation of graphite.A graphite target was evaporated using the radiation from the second harmonic of a Nd:YAG laser (532 nm, 25 ns, 12.5 Hz) with an energy of 5 mJ. After attenuating by neutral filters, the laser radiation was focused by a lens (F = 28 cm) through a quartz window to a vacuum chamber (the residual gas pressure was 10–7 Torr) onto the ground rotating ring-like graphite target, perpendicularly to the surface.Part of the laser radiation was directed to a photodiode and pyroelectric plate by a beam splitter in order to synchronise the system of detection and to control the energy of the laser radiation per pulse.The ion optics of a commercial quadruple mass spectrometer (MX-7304) were modified for recording the time-of-flight mass spectra and electric probe measurements.10 The extracting/accelerating grid of the instrument was used as a collector in the probe measurements and was placed at a distance of 4 cm from the target at a slight angle (10–15°) to the incident laser beam.For detecting negatively charged ions, a constant positive potential of +33 V was applied to the grid relative to the ground graphite target. During detection of the signal the negative pulsed component of the probe potential U(t) was determined by the image potential induced by the negatively charged component q0 of the evaporated substance in the detecting circuit (time constant of 0.11±0.02 ms).Time dependences of the probe potential (Figure 1) were observed and detected at various laser power density values using a S8-12 storage oscilloscope. The quantitative interpretation of pulsed image potentials measured by the electric probe technique is usually a very difficult 103 102 101 100 10–1 100 101 102 103 104 Probe potential /mV Time-of-flight/ms Figure 1 Time dependence of the induced pulsed negative image potential on the probe (extracting/accelerating grid) in the laser evaporation of polycrystalline graphite: laser power density (GW cm–2) 0.19 (dark squares) and 0.36 (light squares). 103 101 10–1 10–3 10–5 10–7 100 102 104 106 Ion intensity (arbitrary units) Cluster size (carbon atoms) Figure 2 Size (N'/z) distribution of negatively charged cluster ions from measurements of the probe potential [power density (GW cm–2)]: (1) 0.21, (2) 0.33 and (3) 1.2. 3 2 1Mendeleev Communications Electronic Version, Issue 5, 1998 (pp. 169-206) problem.9 In this work the approximation of a steady-state (slow) discharge of primary ions of a laser plume (the average power P released in the detecting circuit is close to zero) was developed and used for processing the time dependences obtained for the pulsed probe potential in its delay regions (Figure 1).Our assumption about the steady-state plume discharge was based on the following facts: i, the delay of the signal (discharge phase) lasts long enough (more than 7 ms) as compared with the arrival time of the low mass ions (~1–10 ms); ii, this region of the signal is smooth thus corresponding to a nearly monotonous change in the size distribution of the cluster ions.The condition of steady-state discharge of cluster ions (P ª 0) for the expression of the power (where W is the energy released in the detecting circuit, and q and U are the charge and potential, respectively, induced in the detecting circuit) can be written as the condition of energy balance on the surface of the detecting electrode where U0 and dU are the electrode potential and its change during the discharge of cluster ions with dq charge and q0 is the overall charge of all detected ions.Using equations (1) and (2), we can describe the dynamics of discharge for the packet of cluster ions with a size to charge ratio N'/z = N atoms and relative intensity I(N) by expressions (3) and (4) (where C1 is constant) Taking into account the equation of motion of a cluster ion of the effective size N atoms (M0 is the atomic mass of carbon) in the electric field of a capacitor from ground target plate to accelerating grid (collector) placed at a distance l apart, we obtain Combining (3) and (4), we obtain the resulting expression for I(N) in which C2 is constant. Equation (6) allows us to estimate the relative intensities of the detected cluster ions by differentiation of the time dependence of pulsed image potential U(t) over its delay regions.Accounting for the dependence (5) of the effective size of the detected monoenergetic cluster ions on the time of their arrival at the extracting/accelerating grid (probe), the I(N, t) function describes the distribution of cluster ions over their arrival time as in a common time-of-flight mass spectrum.The composition of the negatively charged products of laser evaporation was studied by the electric probe procedure over the laser power density range of I0 = 0.21–1.2 GW cm–2.Two characteristic groups of cluster ions with size to charge ratio N'/z of 2–20 and 104–106, respectively, were observed in the mass distribution (Figure 2). To study the processes responsible for the formation of ions in these groups, the angular slopes of ion intensity versus laser power density plots were determined using double logarithmic coordinates (Figure 3).The first group of ions (N'/z 2–20) over the range I0 = = 0.27–0.4 GW cm–2 (lower than the threshold laser power density value for spinodal decomposition of the labile liquid phase of carbon6) is characterised by the slope of 6.0±0.5 (Figure 3). This slope is close to the known value (5.6±0.1)11 for the yield of C1 + ion predominant in the mass spectra of primary carbon cluster ions over the indicated I0 range.Thus, the observed correlation of the slope values for the yield of C1 + and this group of ions explains the formation of this group by the recombination of C1 + ions and condensation of neutral carbon atoms followed by the capture of an electron by the cluster formed (electron affinity for the C2 – particle reaches 3.4 eV12 and increases as the cluster size further increases), and so the step of formation of the primary C1 + ion is limiting.The second group of negative ions with an N'/z ratio of 104–106 was observed at power densities higher than 0.33±0.11 GW cm–2, which coincides with the threshold laser power density value for the spinodal decomposition of carbon.6 It was found that the relative intensities for both groups of ions (small and large) differed dramatically (0.1–10 and 10–7–10–5, respectively) but the total amount of evaporated substance which is estimated as the relative cluster intensity multiplied by N'/z was comparable for both groups of ions.The slope value for the yield of these ions (for example, for cluster ions with N'/z = 104 and N'/z = 2×105) was 1.0±0.1, indicating the similarity of the microscopic mechanism of the ion yield and the macroscopic mechanism of laser-induced mass removal6 (at power density values higher than 0.3–0.4 GW cm–2).Thus large carbon cluster ions detected seem to be spinodal decomposition products of labile carbon liquid phase. The saturation of ion intensities for both groups was observed as the laser power density increased in the range 0.5–1.0 GW cm–2 (higher than the threshold laser power density value 0.3–0.4 GW cm–2 for the spinodal decomposition process for the laser-induced labile liquid phase of carbon) (Figures 2 and 3).Also taking into consideration the saturation of the crater depth plot in the I0 range6 (characterising the overall amount of substance evaporated per laser pulse) this phenomenon can be related to laser heating of the spinodal decomposition products resulting in their dissociation and ionisation (Figure 2) and to a change in the form of the initial high mass cluster distribution.Thus, in this work large cluster (nanocluster) ions of carbon with size to charge ratio (N'/z) up to 106 and relative intensities up to ppm (as compared with the same values for small cluster ions with N'/z = 2–20) were detected with a new electric probe procedure. The high sensitivity of the probe procedure was related to the high ion-electron conversion efficiency of probe detection and measurement of the integral image potential induced by the total charge of negatively charged species at the laser plume of carbon.dW dt = P= =U +q d(qU) dt dq dt dU dt (1) U0 = –q0 dq dt dU dt (2) = –q0U0 –1 dq(N) dt dU dt (3) = =C1 I(N)t dq(N) dt dN dt (4) dq(N) dN x(t) = eU0 (lM0) t2 N (5) I(N, t) ª C2 1 t dU dt (6) Ion intensity (arbitrary units) Power density/GW cm–2 1 2 3 0.1 1 3 102 100 10–2 10–4 10–6 10–8 Figure 3 Ion intensity versus laser power density plot [N'/z ratio: (1) 2; (2) 104 and (3) 2×105].Mendeleev Communications Electronic Version, Issue 5, 1998 (pp. 169–205) References 1 A. I.Gusev, Usp. Fiz. Nauk, 1998, 168, 55 (Physics-Uspekhi, 1998, 41, 49). 2 H. Gleiter, Nanostruct. Mater., 1992, 1, 1. 3 R. W. Siegel, Ann. Rev. Mater. Sci., 1991, 21, 559. 4 R. W. Siegel, Nanostruct. Mater., 1993, 3, 1. 5 V. P. Skripov, E. N. Sinitsyn and P. A. Pavlov, Termodinamicheskie svoistva zhidkostei v metstabil’nom sostoyanii (Thermodynamic Properties of Liquids in Metastable State), Atomizdat, Moscow, 1980, ch. 1 (in Russian). 6 S. I. Kudryashov, A. A. Karabutov and N. B. Zorov, Mendeleev Commun., 1998, 6. 7 H. Y. So and C. L. Wilkins, J. Phys. Chem., 1989, 93, 1184. 8 A. A. Sysoev and M. S. Chupakhin, Vvedenie v mass spektrometriyu (Introduction in Mass Spectrometry), Atomizdat, Moscow, 1977 (in Russian). 9 N. B. Delone, Vzaimodeistvie lazernogo izlucheniya s veshchestvom (Interaction of Laser Radiation with Matter), Nauka, Moscow, 1989 (in Russian). 10 S. I. Kudryashov, A. A. Karabutov, N. B. Zorov and Yu. Ya. Kuzyakov, Mendeleev Commun., 1997, 22. 11 J. J. Gaumet, A. Wakisaka, Y. Shimizu and Y. Tamori, J. Chem. Soc., Faraday Trans., 1993, 89, 1667. 12 Termodinamicheskie svoistva individual’nykh veshchestv (Thermodynamic Properties of Individual Substances), ed. V. P. Glushko, Nauka, Moscow, 1979, vol. 2, no. 2, p. 21 (in Russian). Received: Moscow, 3rd April 1998 Cambridge, 10th June 1998; Com. 8/02792G

ISSN:0959-9436     

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 5, 1998 (pp. 169–205) Infrared and EPR spectra of F2NO radicals stabilised in solid argon Eugenii Ya. Misochko,*a Alexander V. Akimov,a Ilya U. Goldschlegera and Charles A. Wightb a Institute for Chemical Physics Research, Russian Academy of Sciences, 142432 Chernogolovka, Moscow Region, Russian Federation. E-mail: misochko@icp.ac.ru b Department of Chemistry, University of Utah, 84112 Salt Lake City, Utah, USA The F2NO radical has been formed by addition reactions of two F atoms with NO in solid argon, and using a combination of EPR and infrared absorption spectroscopy, we have identified this species and characterised its vibrational spectrum for the first time.The free radical F2NO is believed to have a pyramidal structure, and therefore cannot be easily classified as belonging to simple s or p types.The geometry and electronic structure of this radical has been the subject of intense study by quantum chemical methods.1–3 However, experimental data concerning this radical are virtually nonexistent, and this makes it difficult to evaluate the accuracy of the calculations or even to choose an appropriate computational method for calculating the properties of the radical.The F2NO radical is an intermediate species in the reaction of fluorine atoms with nitric oxide: Although this scheme was used successfully to synthesise F3NO,4 attempts to detect the intermediate F2NO in mixtures of F2 and NO in the gas phase and in solids were completely unsuccessful.4,5 In this communication, we describe a method which permits the stabilization of F2NO radicals formed in reaction (2).We present, for the first time, the infrared absorption spectra and EPR spectra of this radical isolated in a rare gas matrix. The basis of this method is the ability of fluorine atoms to diffuse in crystalline argon at temperatures well below its melting point. The barrier to thermal diffusion of F atoms in solid argon is 1.1–1.3 kcal mol–1.At temperatures less than 18 K, fluorine atoms are essentially immobile in the matrix, but at 20–26 K, the atoms are able to diffuse on a length scale of 100 Å on a time scale of 102–104 s.6,7 The ability to control the thermal diffusion of F atoms in this way provides a unique opportunity to carry out addition reactions and stabilization of the resulting intermediate species.Using this method combined with infrared and EPR spectroscopic detection, we have recently determined the spectral characteristics of intermediates formed in the reactions F + CH4,7,8 F +H2 9 and F + C2H4.10 In the present study, we have attempted to detect F2NO in ternary solid mixtures of F2, NO and Ar, in which F2 is used as a photolytic precursor of fluorine atoms.The experimental technique used in this study is similar to that described in our earlier papers.7–10 Solid argon films with impurity molecules were formed by vapour deposition of the reagent gases through separate gas inlets onto the surface of a cold substrate at 14 K. In all of the experiments, the mole fraction of reactants (F2 and NO) was less than 10–3.Dissociation of F2 was performed using 337 nm laser photolysis for the EPR experiments, and 355 nm for the infrared experiments. Fluorine photolysis at these wavelengths at temperatures less than 20 K leads to formation of stabilised F atoms in argon with a photochemical quantum yield close to unity.6,7 Infrared spectra were recorded using an FTIR spectrometer at 0.5 cm–1 resolution over the region 500–2000 cm–1.EPR spectra of freshly prepared samples exhibit no lines due to paramagnetic species. Although nitric oxide is paramagnetic, its lines are strongly broadened in the solid phase, and therefore its spectrum is not detected under our experimental conditions. Neither annealing of the samples to 24–30 K nor extended photolysis at 16 K leads to the appearance of any new lines in the EPR spectrum.Heating of photolysed samples to temperatures higher than 20 K leads to appearance of lines due to the FO2 radical, which forms by reaction of diffusing F ×2 (3) (2) (1) FO2 F2NO aN aF 300 320 340 H/mT Figure 1 EPR spectra of a sample (Ar:NO:F2 = 2000:1:1) after exhaustive photolysis at 16 K and subsequent annealing for 100 min at 24 K.Spectra recorded after annealing showed reversible temperature dependence: 30 K (1), 20 K (2) and 15 K (3). F + NO FNO F + FNO F2NO F + F2NO F3NO (1) (2) (3) 0.02 FNO F3NO FO2 B (NO)2 1900 1800 1700 1600 800 600 n/cm–1 FNO FNO F3NO A A A A F3 NO F3NO NOF Figure 2 Infrared spectra of a sample (Ar:NO:F2 = 2000:1:1) after deposition at 16 K (1) and after exhaustive photolysis at 16 K and subsequent annealing for 100 min at 24 K (2).Both spectra were recorded at 16 K. The series of lines labelled A and B are assigned to the F2NO radical and F···FNO complex, respectively (see text). B (2) (1)Mendeleev Communications Electronic Version, Issue 5, 1998 (pp. 169-206) atoms with O2 molecules that are always present in samples at concentrations of about 10–4–10–5.Annealing of samples at 24 K for 20–60 min leads to the appearance of a new series of 9 lines. Their lineshapes exhibit a strong reversible temperature dependent broadening, but the concentration of radicals remains constant during subsequent annealing cycles from 15 to 30 K. As shown in Figure 1, this series of 9 lines consists of a main triplet with relative intensities 1:2:1, each of which is further split into triplets with 1:1:1 relative intensities.This means that the radical contains two equivalent F nuclei (with nuclear spin I = 1/2) and one N nucleus (with I = 1). The observed magnetic parameters are aF = 14.3 mT, aN = 9.3mT and g = 2.007. These values are close to the values previously given for F2NO radicals produced by radiolysis of F3NO molecules in solid SF6 and in neat F3NO.11,12 The additional splitting of the most intense lines (relative intensity 2) is due to second-order correction and equals DH = aF 2 /H0 = 0.625 mT.Figure 1 shows the temperature dependence of the spectrum of stabilized F2NO radicals. At 30 K, the spectrum is isotropic, with a linewidth less than 0.2 mT.Lowering the temperature to 20 K causes a broadening of the lines. At temperatures below 20 K, the spectra are strongly anisotropic. A detailed analysis of the anisotropic spectra will be presented in a separate paper. We simply note now that the anisotropy of the g factor is small (Dg < 10–3) and the anisotropy of the hyperfine constant for the F atoms is very large (DaF ~ 10 mT).Rapid hindered rotation of the radicals at temperatures greater than 20 K is the mechanism for averaging the anisotropy of the hyperfine constants, and results in an isotropic spectrum. The correlation time of rotation is estimated from the spectral linewidths13 to be tc ~ 10–9 s at 25 K. The next experiments were performed in order to identify the infrared spectrum of the F2NO radical.We were unable to find any literature references that give the fundamental vibrational frequencies of this species. The spectra were assigned using the well-known vibrational frequencies of FNO and F3NO, as well as from the kinetic behaviour of new IR lines which appear during the diffusion and reaction of F atoms at T > 20 K. Figure 2 shows the infrared spectra of samples following deposition.It shows a set of four lines due to NO molecules stabilised in different local sites of the matrix: 1877, 1872, 1867 and 1862 cm–1, and two relatively weak lines due to NO dimers:14 1863 and 1776 cm–1. Also shown are the spectral lines of products that are formed upon condensation of the gases: FNO at 1849.6, 751 and 509.8 cm–1, and its isomer ONF5 at 1884 and 735 cm–1.Photolysis at 16 K leads to a small increase in the FNO lines and a decrease in the intensity of the NO lines. Heating of these photolysed samples at T > 20 K leads to a rapid decrease in the NO lines and growth of the FNO lines. Extended annealing at 24 K leads to growth of new lines of products (see Figure 2): molecules F3NO, lines of which were assigned earlier in ref. 15, and lines of two other products which are labelled A and B in Figure 2. Product B exhibits two doublet lines at 1880 (1884) and 719 (714) cm–1. Product A exhibits 5 lines, the frequencies of which are presented in Table 1. It was possible to distinguish between the lines of these two products due to the different kinetic behaviour shown during the annealing period.Figure 3 shows the kinetics of consumption of the reactant molecules and changes in the product concentrations. We determined the branching ratio between the products using a mass balance equation that assumes that each product contains only one NO functional group. In order to determine the absorption coefficients of the four products, it was necessary to make at least four different measurements of the relative band intensities during the course of the reaction while the relative concentrations were changing.As shown in Figure 3, the principal products are FNO, which is formed in reaction (1), and the species labelled A. The product B grows in the initial stages but is practically destroyed by the end of the annealing period. It follows from the fact that F3NO is formed by sequential addition of three F atoms to NO that we can assign the lines of A and B to intermediate products that contain 2 fluorine atoms.Kinetic support for the conclusion that A and B are secondary products of the sequential addition reactions come from the fact that in the initial stages of the reaction the sum of the concentrations of A and B is proportional to the square of the FNO concentration: [A] + [B] = k[FNO]2.Product B exhibits only two lines, which lie close in frequency to the two lines of FNO; therefore, we assign this species to a weakly bound complex F···FNO. Product A, which exhibits 5 intense lines, is assigned to the F2NO radical, which forms in a subsequent reaction in the complex F···FNO. This assignment is consistent with the kinetic behaviour of product B, because the final equilibrium concentrations of the products are determined by the ratio of the forward (k+) and reverse (k–) reaction rate constants in the complex, K = k+/k– > 1: We believe that the aforementioned analysis, combined with the EPR data, allows a reliable assignment of the product A to the stabilized radical F2NO.We note that although two radical intermediates (A and B) were observed in the infrared experiments, only one of these (F2NO) was observable in the EPR spectra.The F···FNO complex, when stabilised in solids, is not observable by EPR due to the strongly broadened hyperfine structure of the F atom. We carried out preliminary calculations of the vibrational frequencies and hyperfine constants of F2NO radical using ab initio and density functional quantum chemical methods.All of the calculations were performed using the GAUSSIAN 94 suite of codes.16 Table 1 shows the calculated structural data, hyperfine constants and vibrational frequencies of this species. Table 1 Optimised geometries and corresponding isotropic HF constants and frequencies of the F2NO radical, using various level of theory and basis sets.Method Basis RNF/Å RNO/Å �FNF/° �FNO/° aiso(N)/mT aiso(F)/mT w/cm–1 MP2 6-311+G(3df,2p) 1.421 1.152 101.5 119.5 7.52 9.01 1730, 900, 818, 616, 505, 436 B3LYP aug-ccpVDZ 1.453 1.162 102.0 117.0 9.08 11.01 1640, 815, 745, 580, 430, 370 EPR-III17 1.457 1.158 101.9 116.9 9.45 13.10 1625, 806, 730, 569, 418, 368 Exp 9.28 14.35 1573, 803, 761, 705, 553 tc ~ 18H0DH/w0DaF 2 (4) [F···FNO] F2NO (5) k+ k– 0.5 0.4 0.3 0.2 0.1 0 10 20 30 40 50 60 t/min F···FNO FNO F2NO F3NO D[Ni]/[NO]0 Figure 3 Kinetics of consumption of reactant molecules (DNO) and accumulation of reaction products during 24 K annealing of a sample (Ar:NO:F2 = 2000:1:1) that was exhaustively photolysed at 16 K.All concentrations are given relative to the initial concentration of NO molecules in the sample.DNOMendeleev Communications Electronic Version, Issue 5, 1998 (pp. 169–205) The data in Table 1 shows that the B3LYP density functional method gives better agreement with the experimental data than the MP2 calculations. This supports the conclusion in refs. 1 and 17 that density functional methods are a convenient method for calculating the electronic properties of radicals of this type.We hope that the next calculations, which will use more sophisticated methods to account for electron correlation, will provide even better agreement with the experiments. The detailed analysis and assignment of IR spectra will be given in a separate publication. This work was supported by the Russian Foundation for Basic Research (grant no. 98-03-33175) and the US National Science Foundation (grant no. CHE-9526277). References 1 V. Barone, F. Lelj, N. Russo, Y. Ellinger and R. Surba, Chem. Phys., 1983, 76, 385. 2 M. Yu. Balakina, M. B. Zuev, I. D. Morozova and A. V. Il’jasov, Izv. Akad. Nauk SSSR, Ser. Khim., 1988, 587 (Bull. Acad. Sci. USSR, Div. Chem. Sci., 1988, 37, 490). 3 I. L. Shamovsky, I.Yu. Yarovsky and Yu. M. Gershenzon, Theochem., 1993, 103, 43. 4 W. B. Fox, B. Sukornick, J. S. Mackenzie, R. L. Sturtevant, A. F. Maxwell and J. R. Holmes, J. Am. Chem. Soc., 1970, 92, 5240. 5 R. R. Smardzewski and W. B. Fox, J. Chem. Phys., 1974, 60, 2104. 6 J. Feld, H. Kunti and V. A. Apkarian, J. Chem. Phys., 1990, 93, 1009. 7 E. Ya. Misochko, V. A. Benderskii, A. U. Goldschleger and A.V. Akimov, Mendeleev Commun., 1995, 198. 8 E. Ya. Misochko, V. A. Benderskii, A. U. Goldschleger, A. V. Akimov, A. V. Benderskii and C. A. Wight, J. Chem. Phys., 1997, 106, 3146. 9 A. U. Goldschleger, E. Ya. Misochko, A. V. Akimov, I. U. Goldschleger and V. A. Benderskii, Chem. Phys. Lett., 1997, 267, 288. 10 V. A. Benderskii, A. U. Goldschleger, A. V. Akimov, E.Ya. Misochko and C. A. Wight, Mendeleev Commun., 1995, 203. 11 N. Vanderkooi, J. S. Mackenzie and W. B. Fox, J. Fluorine Chem., 1976, 7, 415. 12 K. Nishikida and F. Williams, J. Am. Chem. Soc., 1975, 97, 7168. 13 (a) J. H. Freed and J. K. Fraenkel, J. Chem. Phys., 1963, 39, 326; (b) J. K. Fraenkel, J. Phys. Chem., 1967, 71, 139. 14 W. A. Guillory and C. E. Hunter, J. Chem. Phys., 1969, 50, 3516. 15 R. R. Smardzewski and W. B. Fox, J. Chem. Phys., 1974, 60, 2193. 16 M. J. Frisch, G. W. Trucks, H. B. Schlegel, P. M. W. Gill, B. G. Johnson, M. A. Robb, J. R. Cheeseman, T. Keith, G. A. Petersson, J. A. Montgomery, K. Raghavachari, M. A. Al-Laham, V. G. Zakrzewski, J. V. Ortiz, J. B. Foresman, J. Cioslowski, B. B. Stefanov, A. Nanayakkara, M. Challacombe, C. Y. Peng, P. Y. Ayala, W. Chen, M. W. Wong, J. L. Andres, E. S. Replogle, R. Gomperts, R. L. Martin, D. J. Fox, J. S. Binkley, D. J. Defrees, J. Baker, J. P. Stewart, M. Head-Gordon, C. Gonzalez and J. A. Pople, Gaussian 94, Revision C.4, Gaussian, Inc., Pittsburgh PA, 1995. 17 V. Barone, A. Grand, C. Minichino and R. Subra, J. Phys. Chem., 1993, 97, 6355. 18 V. Barone, Chem. Phys. Lett., 1996, 262, 201. Received: Moscow, 27th May 1998 Cambridge, 17th July 1998; Com. 8/04

ISSN:0959-9436     

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 5, 1998 (pp. 169–205) Direct synthesis of 17a-ethoxyimino-8-aza-D-homogonanes by annelation of 3,4-dihydroisoquinolines with 2-acetyl-5,5-dimethyl-3-ethoxyiminocyclohexanone Olga V. Gulyakevich, Irene L. Rubinova, Dmitry B. Rubinov* and Alexander L. Mikhal’chuk Institute of Bioorganic Chemistry, Belarus National Academy of Sciences, 220141 Minsk, Belarus.Fax: + 7 017 263 7274; e-mail: iboch@ns.igs.ac.by The annelation reaction of Schiff bases by b,b'-tricarbonyl compounds has been extended to 3-ethoxyimino derivatives of 2-acylcyclohexane- 1,3-diones. The first direct synthesis of 8-aza-D-homogonanes with a modified carbonyl group at the pharmacologically significant C(17a) position has been carried out. The 8-aza-steroids (benzo[a]cycloalkano[f]quinolizines) 1, being structurally similar to the important bioregulators of both the animal and plant kingdoms, represent a wide class of condensed nitrogen-containing heterocycles, which have both steroid and alkaloid isosteric fragments.1 These compounds possess valuable biological properties.For example, 8-aza-Dhomogona- 12,17a-diones 2 act as immunomodulators; moreover, both the degree and the direction of their effect may be modulated by transformations in the CD fragment of the ABCD tetracyclic 8-azasteroidal skeleton.2 This, unambigously, makes 8-azasteroids very interesting objects as a basis for the development of safe remedies for the correction of human and animal immunity.However, the possibilities for regioselective transformation of the C(12,17a)-b-dicarbonyl group of 2 are very limited,3 and there are no methods for selective conversion of the C(17a) carbonyl group, for example, into the imino or oxyimino functions.The annelation of Schiff bases by 2-acetylcycloalkanones,4 2-acylcycloalkane-1,3-diones5 and 2-(1-aminoethylidene)cyclohexane- 1,3-diones6 is well documented. Thus, among several approaches to the (17a)-modified derivatives of 2 the most attractive is the cyclocondensation of 3,4-dihydroisoquinolines 3 with accessible oxyimines 4 (Scheme 1),7 which contain the b-dicarbonyl fragment necessary for the condensation.Advantageously, and contrary to the unsubstituted oximes of b-di- and b,b'-tricarbonyl compounds, they fail to convert into isoxazoles by means of intramolecular cyclodehydration.8 Cyclocondensation of 3,4-dihydroisoquinolines 3a,b with an equimolar quantity of b,b'-oxyiminodiketone 4 has been carried out in boiling methanol or ethanol (inert atmosphere, 11–20 h, TLC control) as described for the general method of annelation for b-di- and b,b'-triketones.4,5 In contrast to the latter reaction, which yields derivatives of type 2, annelation of 3a,b with oximine 4 leads to dienones 6a,b (yields 69–71%) or to a mixture of the derivatives 5a,b and 6a,b,† the latter being predominant.If contact with atmospheric and acidic catalysis are avoided, both in the course of the reaction, and during the product isolation, only enones 5a,b are obtained in 83–89% yields. Studies of the properties of enones 5a,b reveal their sensitivity to an acidic environment and to atmospheric oxygen, as well as their thermal lability.Keeping the sample of 5a in a solution containing catalytic quantities of p-toluenesulfonic acid in air, and filtration of the solution of 5b through acidic silica gel, as well as raising the temperature of the reaction (boiling in isopropanol or butanol), afforded the corresponding dienones 6.Thus, the derivatives 6 are the products of dehydration of the initially formed enones 5. It is noteworthy that there are no molecular peaks in the mass spectra of 5, so they are identical with those of 6. This indicates the lability of † Satisfactory elemental analyses, as well as IR, UV, 1H and 13C NMR spectra, were obtained for all new compounds.For 5a: mp 167–171 °C (decomp.); 1H NMR (200 MHz, CDCl3) d: 1.04 [s, 3H, C(16)–Me], 1.10 [s, 3H, C(16)–Me], 1.30 (t, 3H, NOCH2Me, J 7.0 Hz), 2.31 [d, 1H, C(17)HB, J 16.5 Hz], 2.40 [d, 1H, C(15)HB, J 16.0 Hz], 2.43 [d, 1H, C(17)HA, J 16.5 Hz], 2.60 [t, 1H, C(11)HB, J 15.5 Hz], 2.64 [d, 1H, C(15)HB, J 16.0 Hz], 2.78 [tt, 1H, C(6)He, J 15.0, 4.0, 4.0 Hz], 2.87 [dd, 1H, C(11)HA, J 15.5, 4.0 Hz], 3.02 [m, 1H, C(6)Ha, J 15.0, 12.0, 4.0 Hz], 3.22 [ddd, 1H, C(7)He, J 12.0, 12.0, 4.0 Hz], 3.86 (s, 3H, OMe), 3.88 (s, 3H, OMe), 4.14 [tt, 1H, C(7)Ha, J 12.0, 4.0, 4.0 Hz], 4.23 (q, 2H, NOCH2Me, J 7.0 Hz), 4.70 [dd, 1H, C(9)HX, J 15.5, 4.0 Hz], 6.60 [s, 1H, C(4)H], 6.66 [s, 1H, C(1)H]; 13C NMR (90 MHz, CDCl3) d: 14.79 (q, NOCH2Me), 28.04 [q, C(18)], 29.36 [s, C(16)], 29.60 [q, C(19)], 29.76 (t), 36.14 (t), 41.53 (t), 43.96 (t), 47.09 (t), 55.97 (q, OMe), 56.03 (q, OMe), 56.91 [d, C(9)], 69.10 (t, NOCH2Me), 104.66 [s, C(13)], 108.39 (d), 110.96 (d), 125.61 (s), 126.13 (s), 148.08 (s), 148.35 (s), 151.99 (s), 163.06 (s), 187.47 (s).IR (KBr, n/cm–1): 3000–2830, 1645, 1522, 1466–1447, 1357, 1330, 1275, 1222, 1205, 1130, 1062, 866; UV (MeOH) lmax/nm (e): 201 (41.770), 234 (10.485), 275 (13.685), 339 (9.190), lmin/(e): 219 (8.005), 248 (7.035), 302 (3.940).For 6a: mp 87–92 °C (decomp.); 1H NMR (200 MHz, CDCl3) d: 1.07 [s, 6H, MeC(16)Me], 1.31 (t, 3H, NOCH2Me, J 7.0 Hz), 2.62 [s, 2H, C(17)H2], 2.64 [s, 2H, C(15)H2], 3.02 [t, 2H, C(6)H2, J 6.0 Hz], 3.94 [s, 6H, C(2)OMe, C(3)OMe], 4.08 [t, 2H, C(7)H2, J 6.0 Hz], 4.32 (q, 2H, NOCH2Me, J 7.0 Hz), 6.74 [s, 1H, C(11)H], 6.95 [s, 1H, C(4)H], 7.18 [s, 1H, C(1)H]; 13C NMR (90 MHz, CDCl3) d: 14.73 (q, NOCH2Me), 27.84 (t), 28.87 [q, C(18), C(19)], 29.99 [s, C(16)], 35.94 (t), 41.49 (t), 44.82 (t), 58.09 (q, OMe), 58.22 (q, OMe), 69.65 (t, NOCH2Me), 108.38 (d), 109.87 (d), 113.626 (d), 118.27 (s), 121.00 (s), 127.48 (s), 144.15 (s), 148.80 (s), 149.75 (s), 151.23 (s), 151.76 (s) , 175.08 [s, C(12)].IR (KBr, n/cm–1): 3000–2830, 1639, 1618, 1525 (sh), 1515, 1475, 1362, 1275, 1218, 1159, 1066, 876; UV (MeOH), lmax/nm (e): 232 (21.240), 275 (21.040), 324 (13.700), lmin/nm (e): 218 (19.125), 260 (15.745), 302 (11.255). For 5b: mp 110–113 °C (decomp.). For 6b: mp 95–100 °C (decomp.). N (CH2)n N Z O O R2 R2 R2 R1 R1 A B C D 1 2 n= 1, 2 R1 = H, OH, OMe R2 = H, alkyl ...Z = bond, CH2, CHMe, CMe2 ... 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 17a N R1 R2 O NOEt O N R1 R2 O NOEt 3 N R1 R2 O NOEt or + 4 5 6 a R1 = R2 = OMe b R1 = H, R2 = OMe Scheme 1 17a 18 19Mendeleev Communications Electronic Version, Issue 5, 1998 (pp. 169-206) these compounds under the conditions of mass spectral analysis. In order to obtain the derivatives 5 it was necessary to use an inert reaction atmosphere and to avoid acidic catalysis in all stages during the synthesis and isolation.Thus, the present work has demonstrated the possibility of annelation of the 2-acylcyclohexane-1,3-diones with a modified ring carbonyl group with Schiff bases. This provides a simple, one-step synthesis of 8-azasteroid derivatives with a modified carbonyl group at the pharmacologically significant C(17a) position.The authors express their thanks to Academician Aphanasy A. Akhrem for his kind attention to this work and useful discussions. References 1 A. A. Akhrem and Yu. A. Titov, Total Steroid Synthesis, Plenum Press, New York, 1970, p. 362. 2 (a) A. A. Akhrem, B. B. Kuz’mitsky, F. A. Lakhvich, V. A. Khripach and Yu.L. Zhuravkov, Khimiya i biologiya bioregulyatorov (Chemistry and Biology of Immunoregulators), Zinatne, Riga, 1985, p. 265 (in Russian); (b) B. B. Kuz’mitsky, I. G. Dad’kov, Yu. L. Zhuravkov, N. A. Konoplya, G. A. Shafranskaya, O. V. Gulyakevich, V. N. Pshenichny and V. A. Khripach, Vesti Akad. Nauk BSSR, Ser. Biol. Nauk, 1987, 79 (in Russian); (c) B. B. Kuz’mitsky, I.G. Dad’kov, Yu. L. Zhuravkov, N. A. Konoplya, G. S. Lyubin, A. E. Mashkovich, V. M. Nasek, O. V. Gulyakevich, V. N. Pshenichny and V. A. Khripach, Vesti Akad. Nauk BSSR, Ser. Khim. Nauk, 1989, 64 (in Russian); (d) N. A. Konoplya, O. V. Gulyakevich, A. L. Mikhal’chuk and B. B. Kuz’mitsky, Vesti Akad. Nauk BSSR, Ser. Khim. Nauk, 1994, 91 (in Russian). 3 (a) A. L. Mikhal’chuk, O. V.Gulyakevich and A. A. Akhrem, Zh. Obshch. Khim., 1993, 63, 1917 (Russ. J. Gen. Chem., 1993, 63, 1338); (b) O. V. Gulyakevich, A. L. Mikhal’chuk and A. A. Akhrem, Khim. Geterotsikl. Soedin., 1995, 187 [Chem. Heterocycl. Compd. (Engl. Transl.), 1995, 160]; (c) O. V. Gulyakevich, A. S. Lyakhov and A. L. Mikhal’chuk, Dokl. Ross. Akad. Nauk, 1996, 349 (2), 202 [Dokl. Chem. (Engl. Transl.), 1996, 172]. 4 M. von Strandtmann, M. P. Cohen and John Shavel, Jr., J. Org. Chem., 1966, 31, 797. 5 (a) A. L. Mikhal’chuk, O. V. Gulyakevich, A. A. Zenyuk, A. V. Korchik, L. G. Lis, V. A. Khripach and L. I. Ukhova, Dokl. Akad. Nauk SSSR, 1991, 317, 1397 [Dokl. Chem. (Engl. Transl.), 1991, 106]; (b) A. L. Mikhal’chuk, O. V. Gulyakevich, A. A. Zenyuk, Yu. V. Shklyaev, V. S. Shklyaev and A.A. Akhrem, Zh. Obshch. Khim., 1993, 63, 1891 (Russ. J. Gen. Chem., 1993, 63, 1319); (c) A. L. Mikhal’chuk, O. V. Gulyakevich, D. B. Rubinov and A. A. Akhrem, Khim. Geterotsikl. Soedin., 1993, 374 [Chem. Heterocycl. Compd. (Engl. Transl.), 1993, 313]; (d) O. V. Gulyakevich, A. L. Mikhal’chuk and A. A. Akhrem, Zh. Obshch. Khim., 1994, 64, 1544 (Russ. J. Gen. Chem., 1994, 64, 1382). 6 O. V. Gulyakevich, A. L. Mikhal’chuk and V. A. Khripach, Zh. Org. Khim., 1991, 27, 213 [J. Org. Chem. USSR (Engl. Transl.), 1991, 27, 187]. 7 F. A. Lakhvich, L. G. Lis, D. B. Rubinov, I. L. Rubinova, V. Z. Kurbako and A. I. Bykhovets, Vesti Akad. Nauk BSSR, Ser. Khim. Nauk, 1989, 51 (in Russian). 8 (a) H. Smith, J. Chem. Soc., 1953, 803; (b) M. V. Ablovatskaya, E. Yu. Gudriniece and A. Ya. Strakov, Izv. Akad. Nauk Latv. SSR, Ser. Khim., 1989, 601 (in Russian). Received: Moscow, 26th March 1998 Cambridge, 19th June 1998; Com. 8/02401D

ISSN:0959-9436     

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 5, 1998 (pp. 169–205) Reaction of thioacetic acid with ethenyl- and ethynyl chlorides Svetlana G. Dyachkova,a Elena A. Beskrylaya,a Alexander I. Albanov,a Lidia M. Sinegovskaya,a Anastasia G. Malkina,a Terje A. Skotheimb and Boris A. Trofimov*a a Irkutsk Institute of Chemistry, Siberian Branch of the Russian Academy of Sciences, 664033 Irkutsk, Russian Federation.Fax: +7 395 235 6046; e-mail: bat@irioch.irk.ru b Moltech Corporation, Tucson, Arizona, USA Thioacetic acid (TAA) has been subjected to dichlorovinylation with trichloroethene (TCE) under free-radical conditions to form 1-acetylthio-2,2-dichloroethene 1 (yield 70%) which reacts with TAA at room temperature under phase-transfer catalysis conditions to afford a mixture of E- and Z-isomers of 1,2-bis(acetylthio)-2-chloroethene 2 (total yield 71%); ethylthio(chloro)ethyne with sodium thioacetate gives acetylthio(ethylthio)ethyne 3.Functional ethylenes and acetylenes containing a readily hydrolysable acetylthio group at the multiple bond are highly reactive synthons and intermediates for fine organic synthesis as well as potential monomers for the preparation of polyethenyl- and polyethynylthiols: redox and complexing polymers which can be used in advanced technologies, in particular, as cathode materials for lithium batteries.1–3 At the same time, the number of such compounds currently known is rather limited, and the methods for their synthesis are, as a rule, multi-stage and time-consuming processes.4,5 In the present work we studied some new convenient approaches to the synthesis of acetylthioethenes and -ethynes by the reaction of TAA with available chloroethenes and -acetylenes.6 The literature contains no data on the reaction of TAA with TCE, although the latter is known to dichlorovinylate aliphatic and aromatic thiols under radical conditions7 as well as in the presence of a superbasic KOH/DMSO medium.8 Our studies demonstrate that UV irradiation of a heated (80–85 °C) mixture of TAA in excess TCE leads to 1-acetylthio- 2,2-dichloroethene 1 in 70% yield (Scheme 1, i).The reaction (Scheme 1, i) seems to occur as an addition– elimination process, where the addition is of a chain radical nature, while the elimination follows an ionic monomolecular mechanism (E1): The HCl elimination from the adduct B seems to be reversible, which is typical of E1 reactions, and is facilitated by nitrogen blowing through the reaction mixture and completed by distillation in vacuo.The observed regioselectivity of the process corresponds to the expected greater steric hindrance of acetylthiyl radical attack at TCE from two chlorine atoms (repulsion of the lone electron pairs of the sulfur and the chlorine atoms), as well as to the greater stability of the radical A compared with that of the alternative radical A', due to the participation of the electron shells of two chlorine atoms in the spin density distribution. Treatment of ethene 1† with an equimolar amount of TAA in a superbasic aqueous–organic emulsion in the presence of phase-transfer catalyst gave 1,2-bis(acetylthio)-2-chloroethene 2,‡ in an approximately 1:1 mixture of E- and Z-isomers (1H NMR) (Scheme 1, ii).Our numerous attempts to carry out dehydrochlorination of chloroethene 2 with the aim of preparing bis(acetylthio)ethyne failed. Under phase-transfer conditions only black polymeric products were isolated from the reaction mixture irrespective of the organic phase type (toluene, diethyl ether) and dehydrochlorinating agent (KOH, NaOH, K2CO3).The reaction of ethylthio(chloro)ethyne with sodium thioacetate in ether at 20 °C leads to acetylthio(ethylthio)ethyne 3 in 20% yield (not optimized).§ An attempt to perform this process in a superbasic KOH/DMSO suspension successfully employed previously for thiylation of organylthio(chloro)ethynes with thiols9 failed: polymeric products were mainly formed in this case along with a small quantity of acetylthioethyne.¶ † IR spectra were recorded on a Specord IR-75 spectrometer in a microlayer. 1H NMR spectra were recorded on a Jeol FX-90 Q instrument (90 MHz) in CDCl3, with HMDS as an internal standard. Commercial grade TCE was purified by distillation.Ethylthiochloroacetylene was prepared by a procedure described in ref. 6. All the operations were performed in an argon atmosphere. For 1: bp 45–50 °C (3 mmHg), nD 20 1.5274. 1H NMR d: 2.41 (s, 3H, Me), 6.97 (s, 1H, =CH). IR (n/cm–1): 600, 642 (C–S), 784 (C–Cl), 810, 900, 956, 1110, 1342, 1410 (Me, =C–H), 1580 (C=C), 1700 (C=O), 2850, 2910, 2990, 3030 (=C–H, C–H). Found (%): C 28.97; H 2.16; Cl 40.02; S 19.81.Calc. for C4H4Cl2OS (%): C 28.07; H 2.33; Cl 40.52; S 18.71. ‡ For 2: a viscous liquid, nD 20 1.5192, which decomposes on heating (35–50 °C) during distillation in a vacuum, was prepared. 1H NMR d: 2.09 (s, 3H, Me), 2.43 (s, 3H, Me), 6.47 (s, 1H, =CH), 6.70 (s, 1H, =CH). IR (n/cm–1): 655, 680 (C–S), 765 (C–Cl), 815, 860, 945, 978, 1140, 1370, 1425 (Me, =C–H), 1590 (C=C), 1720 (C=O), 2850, 2920, 2955, 3050 (=C–H, C–H).Found (%): C 33.61; H 3.31; Cl 17.84; S 31.36. Calc. for C6H7ClO2S2 (%): C 34.20; H 3.32; Cl 16.86; S 30.40. § For 3: bp 85–90 °C (5 mmHg), nD 20 1.5382. 1H NMR d: 1.13 (t, 3H, Me), 2.21 (s, 3H, MeCO), 2.51 (q, 2H, CH2). IR (n/cm–1): 810, 830, 900, 950, 1100 (C–H), 1350, 1430, 1460 (Me), 1700 (C=O), 2130 (CºC), 2870, 2910, 2970 (C–H).Found (%): C 44.80; H 5.40; S 39.55. Calculated for C6H8S2O (%): C 44.98; C 5.03; S 40.02. AcS Cl H SAc Cl2C CHSAc HSAc i – HCl 1 1 ii 2 E : Z ~ 1:1 Scheme 1 Reagents and conditions: i, TCE (2 mol), reflux, UV irradiation, 10 h, 70% yield of 1; ii, 50% aq. KOH, TAA, PhCH2N+Et3Cl– (cat.), toluene, 20–22 °C, 2 h, 71% yield of 2. HSAc H + SAc h n A Cl2C CHCl + SAc Cl2C CHClSAc AcSC(Cl 2) CHCl A' Cl2CHCHClSAc + SAc etc.HSAc B B 1 + HCl Scheme 2 EtSC CCl EtSC CSAc i 3 Scheme 3 Reagents and conditions: i, AcSNa, Et2O, room temperature, 3 h, 20% yield of 3.Mendeleev Communications Electronic Version, Issue 5, 1998 (pp. 169-206) References 1 A. M. Richter and E. Fangh nel, Sulfur Lett., 1985, 3, 25. 2 P. G. Degott, Polymère Carbone-Soufre.Synthèse et Propriétés Electrochimiques. Thèse pour obtenire titre de Docteur, L’Institut National Polytechnique de Grenoble, 1986, p. 168. 3 K. P. Stiehl, A. M. Richter, E. Fangh nel and K. Wiesener, DDR Patent, 274709, A1, C1H, 1989 (Chem. Abstr., 1990, 113, 100925a). 4 R. Raap and R. G. Micetich, Can. J. Chem., 1968, 46, 1057. 5 L. Brandsma, Preparative Acetylenic Chemistry, Elsevier, Amsterdam– Oxford–New York–Tokio, 1988, p. 321. ¶ Reaction of ethylthio(chloro)ethyne with TAA in a KOH/DMSO suspension. To a KOH (1.39 g, 25 mmol) suspension in 25 ml of DMSO, 1.26 g (16.5 mmol) of TAA and 2 g (16.5 mmol) of ethylthio(chloro)- ethyne were successively added, dropwise with stirring. The reaction mixture was stirred for another 1 h and filtered. All the above procedures were performed at room temperature.The filtrate was poured into 70 ml of cold water with ice. The precipitated oil-like product of brown color was separated. The aqueous layer was extracted with diethyl ether, the extract was combined with the above oil-like product, dried over Na2SO4, passed through an Al2O3 packed column, the solvent was filtered off and the residue was dried under a vacuum. The product obtained (0.8 g) was a viscous brown oil. 1H NMR d: 1.18, 1.26 (m, Me), 2.61 (s, 3H, MeCO) 2.71 (m, 2H, SCH2). IR (n/cm–1): 800, 880, 960, 1100, 1180 (C–H), 1250, 1314, 1380, 1442 (Me), 1710 (C=O), 2130 (CºC), 2860, 2914, 2964 (C–H). Found (%): C 41.66, H 5.53, S 40.69. Calc. for C6H8S2O (%): C 44.98, H 5.03, S 40.02. The product resinified and decomposed during distillation in a vacuum (40–50 °C) in the presence of hydroquinone under argon.When carried out in a KOH/DMSO suspension at 13–15 °C, the reaction of ethylthio(chloro)ethyne with TAA led to an analogous result. 6 A. N. Mirskova, S. G. Seredkina and M. G. Voronkov, USSR Inventors Certificate, 1204616, C07, 1986 (Chem. Abstr., 1986, 105, 20846y). 7 A. N. Mirskova, A. V. Martinov and M. G. Voronkov, Zh. Org. Khim., 1980, 16, 2076 [J. Org. Chem. USSR (Engl. Transl.) 1980, 16, 1895]. 8 B. A. Trofimov, A. S. Atavin, N. K. Gusarova and A. I. Mikhaleva, USSR Inventors Certificate, 1819934, C07, 1982 (Chem. Abstr., 1982, 97, 91733h). 9 S. G. Seredkina, V. E. Kolbina, V. G. Rosinov, A. N. Mirskova, V. I. Donskikh and M. G. Voronkov, Zh. Obshch. Khim., 1982, 52, 2694 [J. Gen. Chem. USSR (Engl. Transl.), 1982, 52, 2375]. a a Received: Moscow, 28th May 1998 Cambridge, 23rd July 1998; Com. 8/04732D

ISSN:0959-9436     

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 5, 1998 (pp. 169–205) The first synthesis of methyl- and methoxy-substituted metal(III) diphthalocyanines Maria A. Ovseevich, Larisa G. Tomilova,* Evgeniya G. Kogan and Nikolai S. Zefirov Department of Chemistry, M. V. Lomonosov Moscow State University, 119899 Moscow, Russian Federation. Fax: +7 095 939 0290; e-mail: tom@org.chem.msu.su Synthesis of previously unknown methyl- and methoxy-substituted Lu and Tb diphthalocyanines has been developed and their spectral characteristics have been compared.Considerable attention is currently being paid to phthalo- and diphthalocyanine complexes of rare earth elements (REE) due to their unique optical and conductive properties. It was supposed in 1971 that the introduction of bulky substituents (such as tert-butyl) prevents intermolecular interaction.This results in an increase in solubility and the emergence of a determinable melting point.1 That is the reason for the interest demonstrated by a number of recent studies in compounds based on phthalocyanine with bulky substituents and polar groups in the benzene rings.2–9 Compounds with weakly-polar, non-bulky substituents (ethyl, methylethoxy, pentyloxy, heptyl) are rarely described.9,10 Moreover, all these substituted metal complexes are usually monophthalocyanines while only a few papers deal with diphthalocyanines (Pc2M).8,10 In this paper we describe the synthesis and spectral characteristics of the hitherto unknown diphthalocyanines MePc2M and MeOPc2M (M = Lu, Tb) based on octamethyl- and octamethoxy-phthalocyanines (4,5-Me2)4PcH2 and (4,5-MeO2)4PcH2.Due to the high coordination number of REE not only planar, but also sandwich-type REE complexes can be obtained. We have established that diphthalocyanines (Pc2Ln) can be obtained via the corresponding monophthalocyanines (PcLnX), transformable into the neutral stable free radical form of diphthalocyanines [Pc2–Ln3+Pc–·]0. Sandwich-type diphthalocyanines were synthesized by fusion of a mixture of the corresponding REE salt and 4,5-dimethyl- or 4,5-dimethoxy-phthalonitrile (molar ratio 1:8).The temperature was gradually raised from 150 to 250 °C and held at 250 °C for 3.5 h for Lu and 2 h for Tb. The first stage of complexation is formation of monophthalocyanine with further transformation into diphthalocyanine. Neither the increase of reaction time to 9 h, nor the increase of temperature up to 350 °C involves formation of the so-called blue form that was observed in the case of non-substituted or tert-butyl-substituted diphthalocyanines. 11 Purification of the diphthalocyanines obtained was carried out by recrystallisation from H2SO4 and DMF and then by thin layer chromatography on Silufol. Moreover, Pc2Ln may be obtained by reaction of PcLnX with the corresponding free phthalocyanines.We have developed a method of synthesis for (4,5-Me2)4PcH2 and (4,5-MeO2)4PcH2 via the corresponding PcLi2 in isoamyl alcohol and found that the template synthesis is not good enough for these compounds. Compounds (4,5-Me2)4PcH2 and (4,5-MeO2)4PcH2 were purified by recrystallisation from H2SO4 and washing with hot DMF.The elemental analysis, UV/VIS and chromatographic data confirmed the purity of the substances. The spectrum of (4,5-Me2)4PcH2 and (4,5-MeO2)4PcH2 is characterized by a Q-band split (670 and 700 nm) attributed to D2h symmetry in contrast to D4h for metallic complexes. While metal-free phthalocyanines (4,5-Me2)4PcH2 and (4,5- MeO2)4PcH2 are soluble only in a-chloronaphthalene, their diphthalocyanine complexes with REE are readily soluble even in CHCl3.Therefore, diphthalocyanines can be purified by column chromatography on alumina.† The electronic absorption spectra (EAS) demonstrate that the green diphthalocyanine forms [Pc–· Ln3+Pc2–]0 may be obtained both by the first and the second methods of synthesis.The absorption spectra of MePc2Lu and MeOPc2Lu show a Q-band at about 670 nm corresponding to electron transfer between the p and p* orbitals of the phthalocyanine ring, and a band at about 480 nm attributed to the presence of an unpaired electron in the Pc–· fragment. The Soret band is in the region of 350 nm (Figure 1). The results of the elemental analysis, spectral and chromatographic investigations of the compounds obtained indicate the formation of sandwich-type phthalocyanine complexes. The UV/VIS and near IR spectra of all the substances obtained were studied. We compared the spectra of the corresponding Tb and Lu complexes taking into account the influence of the REE ionic radii on the Q-band position: e.g.the Q-band of MePc2Tb is observed at 680 nm whereas the MePc2Lu band is observed at 674 nm.We have detected the influence of the phthalocyanine ring substituents on the position of the Q-band in the spectra of the REE complexes (e.g. the Q-band of Pc2Lu is located at † For MePc2Lu. Found (%): C 67.12, 67.21; H 4.30, 4.22; N 15.85, 15.94. Calc. for C80H64N16Lu (%): C 67.46; H 4.53; N 15.73. For MeOPc2Lu.Found (%): C 57.42, 57.53; H 4.02, 4.18; N 13.48, 13.22. Calc. for C80H64N16O16Lu (%): C 57.18; H 3.84; N 13.34. For MePc2Tb. Found (%): C 68.38, 68.42; H 4.65, 4.72; N 15.40, 15.48. Calc. for C80H64N16Tb (%): C 68.22; H 4.58; N 15.91. For (4,5-Me2)4PcH2. Found (%): C 76.95, 76.84; H 5.58, 5.66; N 17.60, 17.64. Calc for C40H34N8 (%): C 76.65; H 5.47; N 17.88. M N R R N N N N N R R R R N N R R N R R N N N N N R R R R N N R R N R R N N N N N R R R R N N M R R X R = Me, OMe M = Lu, Tb X = OAc 0.8 0.6 0.4 0.2 300 400 500 600 700 l/nm 1 2 Figure 1 Absorption spectra of MeOPc2Lu (1) and MePc2Lu (2) in CHCl3. DMendeleev Communications Electronic Version, Issue 5, 1998 (pp. 169-206) 654 nm while that of MeOPc2Lu is observed at 664 nm, and that of MePc2Lu at 674 nm).We have demonstrated that these complexes can be successfully used in sensors, as they are capable of forming ordered Langmuir–Blodgett monolayers. The thermal stability of many diphthalocyanine complexes allows us to prepare evaporated thin solid films, which can also be used in sensors. Now we are studying the absorption of various gases on Langmuir–Blodgett and evaporated thin solid films.The authors are grateful to the Russian Foundation for Basic Research for financial support (grant no. 97-03-33239). References 1 E. A. Luk’yanets, S. A. Mihalenko and E. I. Kovshev, Zh. Obshch. Khim., 1971, 41, 934 [J. Gen. Chem. USSR (Engl. Transl.), 1971, 41, 942]. 2 P. Vasudevan, N. Phougat and A. V. Shuklat, Applied Organometallic Chem., 1996, 10, 591. 3 R. J. Reeves, R. C. Powell, Y. H. Chang, W. T. Ford and W. Zhu, Optical Materials, 1996, 5, 43. 4 J. Vacus, G. Memetzidis, P. Doppelt and J. Simon, J. Chem. Soc., Chem. Commun., 1994, 697. 5 Ch. Sun, Y. Sun, X. Zhang, H. Xu and J. Shen, Anal. Chim. Acta, 1995, 312, 207. 6 R. W. Boyle and J. E. van Lien, Synthesis, 1995, 1079. 7 E. O. Tolkacheva, A. Y. Tsivadze, Sh. G. Bitiev, Yu. G. Gorbunova, V. I. Jilov and V. V. Minin, Zh. Neorg. Khim., 1995, 40, 984 (Russ. J. Inorg. Chem., 1995, 40, 949). 8 Yu. G. Gorbunova, E. O. Tolkacheva and A. Yu. Tsivadze, Koord. Khim., 1996, 22, 944 (Russ. J. Coord. Chem., 1996, 22, 884). 9 R. Dieing, G. Schmid, E. Witke, C. Feucht, M. Dreßen, J. Pohmer and M. Hanack, Chem. Ber., 1995, 128, 589. 10 J. Jiang, R. C. W. Liu, T. C. W. Mak, T. W. D. Chan and D. K. P. Ng, Polyhedron, 1997, 16, 515. 11 L. G. Tomilova, Y. G. Gorbunova, M. L. Rodriguez-Mendez and J. A. De Saja, Mendeleev Commun., 1994, 127. Received: Moscow, 20th February 1998 Cambridge, 21st May 1998; Com. 8/01640B

ISSN:0959-9436     

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 5, 1998 (pp. 169–205) A porphyrin chlorination reaction Andrei F. Mironov,* Valentina D. Rumyantseva and Olga N. Ponamoreva M. V. Lomonosov Moscow State Academy of Fine Chemical Technology, 117571 Moscow, Russian Federation. Fax: +7 095 434 8711 Treatment of nickel and palladium porphyrin complexes with thionyl chloride readily affords products of meso- and b-chlorination; further reaction leads to chlorination of macrocyclic methyl groups.A reaction for the chlorination of porphyrin metal complexes has been found. In this study, instead of traditional chlorinating agents (e.g. HCl and hydrogen peroxide,1,2 sulfuryl chloride3 and chlorosulfonic acid4), we tried thionyl chloride, as used in the case of tetraazaporphyrin.5 This reagent is known to readily replace the hydroxyl group in alcohols and carboxylic acids but has scarcely been used in reactions involving C–H linkages.6 Thionyl chloride has been widely used in porphyrin chemistry for transforming carboxylic acids into the corresponding chlorides.However, on attempting the activation of the carboxylic acids in palladium coproporphyrin III with thionyl chloride, we noted an abrupt colour change from orange–red to deep green. The electron absorption spectrum of the product showed a significant bathochromic shift of both the Soret band and the a- and b-bands, with the intensity ratio of the a- and b-bands being considerably lower, evidencing disturbance of the porphyrin macrocycle.Unfortunately, full characterisation of the compound thus obtained was unsuccessful, presumably due to the formation of a highly reactive chloride.Further study of this reaction was performed using the palladium coproporphyrin III tetramethyl ester 1a. This was dissolved in SOCl2, kept for 2 h at 20 °C, poured into ice and the resulting precipitate was filtered to give almost pure products (TLC assay) in 92% yield.Prior to elemental analysis, the porphyrin obtained was passed through alumina and recrystallised from chloroform–methanol. Elemental analysis data showed that the product contained four additional chlorine atoms. The mass spectrum also provided evidence in favour of four chlorine atoms (m/z 952).† The 1H NMR spectrum showed the disappearance of four meso-protons. Therefore, the structure 2a was assigned to the new compound.‡ In the case of palladium deuterioporphyrin 1b, not only the meso-protons, but also both b-positions were replaced to give the hexachloro-substituted porphyrin 2c.§ This showed an even greater colour change and bathochromic shift of absorption bands.Similar behaviour was observed for nickel porphyrins. In the case of coproporphyrin 3a, the meso-tetrachloro-derivative 4a¶ was obtained at 4 °C for 15 min in 93% yield.Deuterioporphyrin 3b was transformed into the hexachloro-derivative 4c.†† However, nickel porphyrins demonstrated higher reactivity. Prolonged treatment of 3a led to substitution of not only the meso-protons, but also the methyl groups. Heating for 1 h † Mass spectra were measured on a MSBKh instrument (SELMI, Sumy, Ukraine).Ionisation was effected by 252Cf fission products and a timeof- flight monitoring ion analyser was employed. ‡ Data for 2a, methyl ester: mp 199–202 °C. 1H NMR (CDCl3) d: 4.08 (t, 8H, CH2CH2CO2Me), 3.81 (s, 6H, COOMe), 3.78 (s, 6H, COOMe), 3.26 (s, 6H, Me), 3.24 (s, 6H, Me), 2.95 (m, 8H, CH2CH2CO2Me). UV [CHCl3, lmax/nm (e×10–3)]: 440 (146), 561 (9.6), 606 (6.4).MS, m/z: 952 (M+). Found (%): C 50.23, H 4.18, Cl 15.41, N 5.81. Calc. for C40H40Cl4N4O8Pd (%): C 50.41, H 4.23, Cl 14.88, N 5.88. § Data for 2c, ethyl ester: mp >300 °C. 1H NMR (CDCl3) d: 4.22 (m, 8H, CH2CH2CO2Me and CH2Me), 3.30 (s, 12H, Me), 3.00 (t, 4H, CH2CH2CO2Me), 1.30 (t, 6H, CH2Me). UV (CHCl3, lmax/nm): 452, 561 (b), 618 (a) (a/b = 0.70). MS, m/z: 878 (M+).¶ Data for 4a: mp 111–113 °C. 1H NMR (CDCl3) d: 4.41 (m, 8H, CH2CH2CO2Me), 3.78 (s, 12H, COOMe), 3.24 (s, 12H, Me), 2.92 (m, 8H, CH2CH2CO2Me). UV [CHCl3, lmax/nm (e×10–3)]: 442 (121), 577 (9.7), 621 (5.0). MS, m/z: 905 (M+). Found (%): C 53.44, H 4.24, Cl 15.01, N 5.81. Calc. for C40H40Cl4N4O8Ni (%): C 53.07, H 4.45, Cl 15.67, N 6.19. resulted in chlorination of all four methyl group to produce the octachloroporphyrin 5.‡‡ Shortening the reaction time makes it †† Data for 4c: mp 164–166 °C. 1H NMR (CDCl3) d: 4.14 (m, 4H, CH2CH2CO2Me), 3.77 (s, 6H, COOMe), 3.23 (s, 3H, Me), 3.18 (s, 3H, Me), 3.14 (s, 3H, Me), 3.12 (s, 3H, Me), 2.83 (m, 4H, CH2CH2CO2Me). UV [CHCl3, lmax/nm (e×10–3)]: 442 (156), 585 (7.0), 632 (4.9). MS, m/z: 802 (M+). Found (%): C 47.53, H 3.75, N 6.63.Calc. for C32H26Cl6N4O4Ni (%): C 47.92, H 3.27, N 6.99. ‡‡ Data for 5: mp 120–122 °C. 1H NMR (CDCl3) d: 5.75 (s, 4H, CH2Cl), 5.71 (s, 4H, CH2Cl), 4.12 (m, 8H, CH2CH2CO2Me), 3.76 (s, 12H, COOMe), 3.05 (m, 8H, CH2CH2CO2Me). UV [CHCl3, lmax/nm (e×10–3)]: 459 (130), 593 (10.4), 641 (6.6). MS, m/z: 1042 (M+). Found (%): C 45.64, H 3.39, Cl 26.91, N 5.12. Calc. for C40H36Cl8N4O8Ni (%): C 46.06, H 3.48, Cl 27.19, N 5.37.N N N N CO2R' CO2R' Me Me Me R Me R Pd N N N N CO2R' CO2R' Me Me Me R Me R Pd Cl Cl Cl Cl SOCl2 1a,b 2a,c N N N N CO2Me CO2Me Me Me Me R Me R Ni 3a,b SOCl2 N N N N CO2Me CO2Me Me Me Me R Me R Ni 4a,c Cl Cl Cl Cl N N N N CO2Me CO2Me Ni 5 Cl Cl Cl Cl Cl Cl CO2Me Cl Cl CO2Me R' = Me, Et a R = CH2CH2CO2Me b R = H c R = Cl Scheme 1Mendeleev Communications Electronic Version, Issue 5, 1998 (pp. 169-206) possible to obtain products with partially chlorinated methyl groups: we succeeded in isolating heptachloro-substituted nickel coproporphyrin III tetramethyl ester 6.§§ The number of chlorine atoms in 4a, 4c, 5 and 6 was proved by mass spectrometry, 1H NMR spectroscopy and elemental analysis data. The presence of a transition metal ion in the porphyrin molecule presumably plays the determining role in this reaction.On treatment of deuterioporphyrin dimethyl ester and coproporphyrin tetramethyl ester with thionyl chloride under similar conditions, no chlorination was observed. It seems likely that coordination of thionyl chloride with a central metal atom followed by a chain of redox reactions leads to the reduction of sulfur and to the formation of a highly reactive chlorine species (possibly, the chlorine radical), which attacks the macrocycle.Finding crystalline sulfur in the reaction mixture obtained during the synthesis of 2c provides evidence in favour of this assumption. §§ Data for 6: mp 190–192 °C. 1H NMR (CDCl3) d: 5.76 (s, 6H, CH2Cl), 4.18 (m, 8H, CH2CH2CO2Me), 3.77 (s, 12H, COOMe), 3.25 (s, 3H, Me), 3.00 (m, 8H, CH2CH2CO2Me).UV [CHCl3, lmax/nm (e×10–3)]: 448 (157), 582 (13.1), 627 (8.5). MS, m/z: 1007 (M+). Found (%): C 47.31, H 3.93, Cl 24.31, N 5.93. Calc. for C40H37Cl7N4O8Ni (%): C 47.63, H 3.70, Cl 24.60, N 5.56. This work was supported by the Russian Foundation for Basic Research (grant no. 96-15-97-709). References 1 H. Fischer and W. Klendauer, Ann. Chem., 1941, 547, 123. 2 R. Bonnett, P. Brewer, K. Noro and T. Noro, Tetrahedron, 1978, 34, 379. 3 D. Dolphin, The Porphyrins, Academic Press, New York, 1978, vol. 2, p. 153. 4 E. Samuels, R. Shuttleworth and T. S. Stevens, J. Chem. Soc. C, 1968, 145. 5 O. G. Khelevina, S. V. Timofeeva and B. D. Berezin, Zh. Org. Khim., 1994, 30, 295 (Russ. J. Org. Chem., 1994, 30, 312). 6 L. Fieser and M. Fieser, Reagents for Organic Synthesis, Wiley, New York, 1968. Received: Moscow, 23rd March 1998 Cambridge, 1st June 1998; Com. 8/02397B

ISSN:0959-9436     

出版商:RSC    

年代:1998

数据来源: RSC