摘要:

Mendeleev Communications Electronic Version, Issue 2, 2000 (pp. 43–82) Stabilization of the glycine zwitterionic form by complexation with Na+ and Cl–: an ab initio study Ruslan M. Minyaev,* Andrei G. Starikov and Vladimir I. Minkin Institute of Physical and Organic Chemistry, Rostov State University, 344090 Rostov-on-Don, Russian Federation. Fax: +7 863 228 5667; e-mail: minyaev@ipoc.rsu.ru DOI: 10.1070/MC2000v010n02ABEH001259 Ab initio [MP2(full)/6-31G**] calculations predict that the glycine zwitterionic form bridged with NaCl is more stable than the isomeric complexes of the neutral forms.The zwitterionic form of amino acids dominates in the aqueous media of biological systems.1 Gas phase experimental data2–4 and numerous ab initio calculations4–9 show that the simplest amino acid glycine exists only in the neutral forms 1, 2 and 3, whereas the zwitterionic form 4 is not observed; that is, it does not correspond to a minimum on the potential energy surface (PES).Both theoretical and experimental studies2–8 point to the crucial role of medium effects on the existence and stability of the zwitterionic form of glycine. Recent ab initio calculations predict that the glycine zwitterionic form can be stabilised, i.e., corresponds to the minimum on the PES, when it forms an H-bonded cluster with at least two water molecules.5,6 At the same time, it is well known1 that in biological systems amino acids form salts and various complexes with alkali metals (Na+, K+),9,10 copper (Cu+)11,12 or other ions,1,13 which stabilise the zwitterionic form.Here, we report on ab initio [MP2(full)/6-31G**]14 calculations of glycine complexes 5–7 with the cation Na+, complexes 8–10 with the anion Cl– and complexes 11–13 with undissociated sodium chloride. Our results indicate that only complexation of a glycine molecule simultanuously with the two counterions Na+ and Cl– results in the stabilization of zwitterionic form 13 relative to isomeric complexes 11, 12 formed by neutral glycine.According to the MP2(full)/6-31G** calculations, all structures 5–13 correspond to minima (l = 0, hereafter l designates the number of hessian negative eigenvalues at a given stationary point) on the PES. Various other possible structures of the complexes have also been studied but have been found to correspond to the stationary points with l � 1.Therefore, they are not considered here. The calculated molecular structures, geometries and energy parameters of complexes 5–13 are given in Figure 1 and Table 1. The counterions Na+ and Cl– stabilise glycine zwitterionic forms 7 and 10, respectively, i.e., calculations predict the correspondence to local minima on the PESs. However, complex 7 is less stable than the neutral form of glycine 5 and 6, and anionic complex 10 is less stable than 8, 9.Only the cooperative influence of the two counterions makes the zwitterionic form 13 preferred in energy (0.5 kcal mol–1) as compared to the most stable complex of the neutral form of glycine 11. Accounting for ZPE makes the zwitterionic form less stable than the neutral form 11 by 0.5 kcal mol–1.The same tendency is observed for relative enthalpy, whereas the behaviour of the relative free energy (DG) coincides with that of the relative free energy (DE). The complexation energy of 13 calculated without accounting for a basis set superposition error (BSSE)15 is 32.9 kcal mol–1. We will compare it with the value 32.4 kcal mol–1 for the complexation energy of the neutral form 11 (in both cases, complexation energies are estimated as the difference {Etot(13 or 11) – – [Etot(1) + Etot(NaCl)]}, where Etot is the total energy of 13, 11, 1 or NaCl, respectively). Note that the C1-structure of 13 is skewed toward a helical form (see Figure 1).The geometric parameters of glycine forms 1 and 2 are markedly changed under complexation with NaCl: all valent bonds participating in interactions with Na and Cl are elongated by ~0.03 Å, as is the NaCl N O O H H H H H N O O H H H H H N O O H H H H H N O O H H H H H 1, Cs 2, C1 3, C1 4 N O O H H H H H N O O H H H H H N O O H H H H H 5, C1 6, Cs 7, Cs Na+ Na+ Na+ N O O H H H H H N O O H H H H H N O O H H H H H 8, Cs 9, C1 10, C1 Cl– Cl– Cl– N O O H H H H H N O O H H H H H N O O H H H H H 11, Cs 12, C1 13, C1 Na Cl Cl– Na Cl Na+ N 2.010 2.466 99.9 2.245 1.24 1 112.1 178.6 1.006 C C O O Cl Na 1.314 1.517 122.0 116.0 1.447 11, Cs O C O N C Na Cl 1.011 1.012 1.466 1.089 2.480 1.522 0.997 1.232 1.321 2.244 2.452 N Na Cl O C O C 1.253 1.543 1.490 1.275 1.066 1.960 2.572 2.374 2.385 12, C1 13, C1 Figure 1 Geometry parameters of glycine complexes with NaCl 11, 12 and 13, as calculated by the MP2(full)/6-31G** method. Bond lengths and angles are given in angström units and degrees, respectively.Mendeleev Communications Electronic Version, Issue 2, 2000 (pp. 43–82) distance increased on 0.1 Å. It should be noted that the electric dipole moment of the zwitterionic form of complex 13 (3.84 D) is smaller than that of 11 (6.83 D) and 12 (5.10 D) by a factor of about two.This fact testifies to a partial screening of the charges on the NH3 + and CO2 – groups in the complex. In conclusion, the calculations show that the counterions not only stabilise the zwitterionic form of glycine but also convert it to the dominating form in the complex with NaCl. Obviously, the zwitterionic form of glycine in biological systems is stabilised by interactions with both water molecules and counterions. The cooperative medium effect can increase the zwitterionic stability.This work was supported by the Russian Foundation for Basic Research (grant no. 98-03-33169a). V. I. M. acknowledges the support of the Alexander von Humboldt Foundation (Humboldt Research Award 1999). References 1 L. Stryer, Biochemistry, W.H. Freeman and Comp., San Francisco, 1981. 2 S. J. McGlone, P. S. Elmer, R. D. Brown and P. D. Godffrey, J. Mol. Struct., 1999, 485–486, 225. 3 (a) R. D. Suenham and F. J. Lovas, J. Mol. Spectrosc., 1978, 72, 372; (b) R. D. Suenham and F. J. Lovas, J. Am. Chem. Soc., 1980, 102, 7180. 4 S. G. Stepanian, I. D. Reva, E. D. Radchenko, M. T. S. Rosado, M. L. T. S. Duarte, R. Fausto and L.Adamowicz, J. Phys. Chem. A, 1998, 102, 1041. 5 J. H. Jensen and M. S. Gordon, J. Am. Chem. Soc., 1995, 117, 8159. 6 F. R. Tortonda, J. L. Pascual-Ahuir and E. S. Tunon, Chem. Phys. Lett., 1996, 260, 21. 7 D. Chakraborty and S. Manogaran, Chem. Phys. Lett., 1998, 294, 56. 8 D. T. Nguyen, A. C. Scheiner, J. W. Andzelm, S. Sirois, D. R. Salahub and A. T. Hagler, J. Comput. Chem., 1997, 18, 1609. 9 T.Wyttenbach, J. Bushnell and M. T. Bowers, J. Am. Chem. Soc., 1998, 120, 5098. 10 B. A. Cerda, S. Hoyau, G. Ohanessian and C. Wesdemiotis, J. Am. Chem. Soc., 1998, 120, 2437. 11 S. Hoyau and G. Ohanessian, J. Am. Chem. Soc., 1997, 119, 2016. 12 J. Bertran, L. Rodriguez-Santiago and M. Sodupe, J. Phys. Chem. B, 1999, 103, 210. 13 S. Hoyau and G. Ohanessian, Eur. J.Chem., 1998, 4, 1561. 14 M. W. Schmidt, K. K. Baldridge, J. A. Boatz, S. T. Elbert, M. S. Gordon, J. H. Jensen, S. Koseki, N. Matsunaga, K. A. Nguyen, S. J. Su, T. L. Windus, M. Dupuis, J. A. Montgomery, J. Comput. Chem., 1993, 14, 1347 (package of ab initio programs ‘GAMESS’, Version 1996). 15 F. B. van Duijneveld, J. G. C. M. van Duijneveld-van deRijdt and J. H. van Lenthe, Chem.Rev., 1994, 94, 1873. aEtot and DE are the total and relative energies (1 a.u. = 627.5095 kcal mol–1); DEZPE is the relative energy including a harmonic zero-point correction; DH is the relative enthalpy; DG is the relative free energy; w1 is the smallest vibration frequency. bThermochemistry data are given at T = 298.15 K and P = 1 atm. Table 1 Ab initio MP2(full)/6-31G** data for the structures of 5–13.a Structure Etot/a.u. DE/kcal mol–1 DEZPE/kcal mol–1 DHb/kcal mol–1 DGb/kcal mol–1 w1/cm–1 5, C1 –445.40111 0 0 0 0 85 6, Cs –445.39513 3.75 2.94 3.04 4.84 80 7, Cs (ZW) –445.39686 2.67 2.80 2.80 2.51 102 8, C1 –743.36694 0 0 0 0.48 43, C1 –743.36631 0.39 1.34 1.16 0 34 10, C1 (ZW) –743.36347 2.18 2.71 2.44 1.62 80 11, Cs –905.26099 0.50 0 0 2.84 41 12, C1 –905.25422 4.75 4.73 4.70 6.82 40 13, C1 (ZW) –905.26178 0 0.47 0.01 0 92 Received: 29th December 1999; Com. 99/1585

ISSN:0959-9436     

出版商:RSC    

年代:2000

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 2, 2000 (pp. 43–82) Homochiral and pseudoracemic 3,3- and 1,2-dimethyldiaziridine–silver nitrate complexes Remir G. Kostyanovsky,*a Konstantin A. Lyssenkob and Vasily R. Kostyanovskya a N. N. Semenov Institute of Chemical Physics, Russian Academy of Sciences, 117977 Moscow, Russian Federation. Fax: +7 095 938 2156; e-mail: kost@center.chph.ras.ru b A.N. Nesmeyanov Institute of Organoelement Compounds, Russian Academy of Sciences, 117977 Moscow, Russian Federation. Fax: +7 095 135 5085; e-mail: kostya@xray.ac.ru DOI: 10.1070/MC2000v010n02ABEH001261 Diaziridine molecules in the title complexes 1 and 2, respectively, are bidentate ligands and form coordination polymers in which silver ions are coordinated to the trans-oriented nitrogen lone pairs of the ligands; complex 1 is homochiral (space group P212121), whereas 2 is a pseudoracemate (space group Pbca) in which the alternation of ligands with opposite configurations is statistically disordered.Molecular assembling by metal coordination and, in particular, by coordination polymerisation is an important tool in supramolecular chemistry1–4 and crystal engineering.5–11 Diaziridines (see ref. 12 for a recent review) are bidentate ligands suitable for coordination polymerisation because the trans-oriented lone pairs of nitrogen cannot be coordinated to the same metal ion. All monocyclic diaziridines are chiral (with the exception of cis-1,2-dimethyl-3-tert-butyldiaziridine13); we resolved them into enantiomers for the first time;14–21 however, there is no data concerning their spontaneous resolution.The data on coordinated complexes are limited (see refs. 22–25 and references therein). According to the X-ray diffraction analysis of mixed complexes such as A22 and B–D23 and of complex E,24 they do not form coordinated polymers in crystals. It should be emphasised that complex C is homochiral.23 In this work, we synthesised† complex 1 of 3,3-dimethyldiaziridine with AgNO3 for the first time and found by X-ray diffraction analysis of a single crystal‡ that 1 is a homochiral (space group P212121) coordinated polymer (Figure 1).Complex 2 of 1,2-dimethyldiaziridine with AgNO3 was prepared earlier; however, the relevant structural data reported were inadequate.25 † 1 was prepared by the procedure given below. 3,3-Dimethyldiaziridine was purified by freezing from n-hexane, mp 40 °C. 1H NMR (CD3CN) d: 1.31 (s, 6H, Me2C), 2.26 (br.s, 2H, 2HN). A mixture of 0.2 g of 3,3-dimethyldiaziridine and 0.4 g of AgNO3 in 5 ml of absolute MeOH was kept at 4 °C for 10 h; the precipitate (0.5 g, 88%) was separated and crystallised from absolute MeCN to give 0.3 g of colourless transparent bright crystals in 52.6% yield, mp 137 °C. 1H NMR (CD3CN) d: 1.44 (s, 6H, Me2C), 3.26 (br. s, 2H, 2HN). Found (%): N, 17.46. Calc. for C3H8N3O3Ag (%): N, 17.37. 2 was prepared by the known method,25 mp 136 °C (MeCN). The product with [a]D 20 = 27.9° (c 2.5, MeCN) was obtained from partly enriched (+)-1,2-dimethyldiaziridine,29 [a]D 20 = 9.5° (c 2.2, MeCN) after the treatment with a half-mole quantity of AgNO3 in MeCN, the separation of 2 and the distillation of the mother liquor into a cold trap (–80 °C).The repeated X-ray determination of the structure of 2 confirmed that the complex is a heterochiral coordination polymer (space group Pbca). However, in contrast to data,25 we found ‡ Crystallographic data for 1 and 2: at –80 °C, crystals of C3H8AgN3O3 1 are orthorhombic, space group P212121, a = 5.2023(9), b = 7.921(1), c = = 17.653(3) Å, V = 726.7(2)Å3, Z = 4, dcalc = 2.212 g cm–3, m = 2.778 mm–1, M = 241.99, F(000) = 472; crystals of C3H8AgN3O3 2 are orthorhombic, space group Pbca, a = 10.192(3), b = 10.678(4), c = 13.339(4) Å, V = = 1451.6(9) Å3, Z = 8, dcalc = 2.215 g cm–3, m = 2.731 mm–1, M = 241.99, F(000) = 944.The intensities of 1498 reflections for 1 and 2143 reflections for 2 were measured on a Syntex P21 diffractometer at –80 °C (lMoKa radiation, q/2q-scan technique, 2qmax < 60° and 70° for 1 and 2, respectively). The structures were solved by a direct method and refined by a full-matrix least squares technique against F2 in the anisotropic– isotropic approximation. The positions of hydrogen atoms were calculated from the geometrical point of view with the exception of the nitrogen atoms in 1, which were located from the difference Fourier synthesis and refined in the isotropic approximation. An analysis of difference electron density syntheses in the structure of 2 revealed additional electron density maxima which were interpreted as the disorder of diaziridine molecules.The refinement of the occupancies for two positions of nitrogen atoms resulted in the 1:1 ratio. The absolute S-configurations for the N(1) and N(2) atoms in 1 were confirmed by estimating the Flack absolute structure parameter x,30 is equal to zero with a rather small esd [–0.00(11)] in the case of the S-configuration for N(1) and N(2). The refinement converged to wR2 = 0.1668 and COF = 1.133 for 1396 reflections [R1 = 0.0699 was calculated against F for the 1308 observed reflections with I > 2s(I)] for the structure of 1 and to wR2 = = 0.1892 and COF = 1.10 for all independent reflections [R1 = 0.0689 was calculated against F for the 1852 observed reflections with I > 2s(I)] for the structure of 2.All calculations were performed using the SHELXTL PLUS 5.0 program on an IBM PC/AT.Atomic coordinates, bond lengths, bond angles and thermal parameters have been deposited at the Cambridge Crystallographic Data Centre (CCDC). For details, see ‘Notice to Authors’, Mendeleev Commun., Issue 1, 2000. Any request to the CCDC for data should quote the full literature citation and the reference number 1135/62. N N H M R R H N N H M R R H M N N H H NH2 NH2 Cd(ClO4)2 2 A: R2 = ButCH(CH2CH2)2 M = RhCl(cod) cod = cyclooctadiene space group P21/c (z = 4) B: R2C = 2,2-adamantylidene M = PtCl2(Et3P) space group Pbca (z = 8) C: R2 = (CH2)5 M = PtCl2(Et3P) space group P21 (z = 2) D: R2C = 2,2-adamantylidene M = PtCl2(Et3P) Space group P21/n (z = 4) E: space group Pbc21 (z = 4) Figure 1 The crystal structure of 1: (a) the Ag-coordinated homochiral polymeric chain directed along the crystallographic axis a; (b) the arrangement of chains into ‘walls’ laying in the crystallographic plane ab.Ag(1') C(2) N(2) C(1) N(1) C(3) Ag(1) N(2') O(3) O(3) N(3) O(3) H(1'') O(3) O(2) Ag(1'') N(1) H(1) O(3') O(1') Ag(1) 0 a b c (a) (b) O(1)Mendeleev Communications Electronic Version, Issue 2, 2000 (pp. 43–82) that diaziridine ligands are statistically disordered (each ligand position is randomly occupied by the opposite enantiomers) leading to a chain-type structure containing AgNO3 with the superposition of diaziridine enantiomers [Figure 2(a)]. Thus, complex 2 is a rare example of a pseudoracemate.26 The Ag–N bond lengths in complexes 1 and 2 are slightly different and equal to 2.228(6)–2.238(6) and 2.279(8)–2.336(5) Å, respectively.For comparison, the Ag–N bond lengths in the polymeric complexes of AgBF4 and AgSbF6 with pyrazine27 and of CF3SO3Ag with a methionine derivative28 are 2.459(9)– 2.519(8) and 2.219(4)–2.378(5) Å, respectively. In both structures, within the above chain directed along the crystallographic axis a, NO3 acts as a bidentate ligand with some shortening of the Ag–O bond lengths in the case of 2 [2.508(3), 2.590(3) Å], as compared with 1 [2.622(4), 2.643(6) Å].Additional chain-to-chain interactions between Ag and NO3 in 1 and 2 are different. In complex 1, NO3 acting as a monodentate ligand links adjacent chains to ‘walls’ laying along the crystallographic plane ab [Ag(1)–O(1') (x – 1, y – 1/2, z – 1/2) 2.528(6) Å] [Figure 1(b)]. The additional interaction of the O(1) atom with two silver atoms results in elongation of the N(3)–O(1) bond [1.264(6) Å], as compared with the N(3)–O(2) bond [1.243(6) Å].At the same time, the formation of similar ‘walls’ (also laying in the crystallographic plane ab) in 2 is accomplished by weak interactions of NO3, which acts as a bidentate ligand [Ag(1)– O(1') (3/2 – x, 1/2 + y, z) 2.853(4) Å and Ag(1)–O(2') (3/2 – x, 1/2 + y, z) 2.733(4) Å] [Figure 2(b)]. Thus, both of the structures are build up from the chains assembled into ‘walls’ by means of the Ag···(NO3) interactions.Note that the ‘walls’ in complex 1 are more compact because of the above chain-to-chain interactions. This can be illustrated by a comparison between the Ag···Ag distances in a chain [5.202(2) and 5.407(3) Å] and between chains [4.534(2) and 5.531(2) Å] in complexes 1 and 2, respectively.The higher density of 2 can be explained by more tightly packed ‘walls’ in this compound, as compared with 1. The observed difference in the layer architecture of complexes 1 and 2 is probably due to the presence of a bulky substituent at nitrogen in 2.In addition, the homochiral crystal packing in complex 1 is stabilised by the formation of hydrogen bonds [N(1)–H(1)···O(3') distances of 3.01(1) and 2.13 Å] between the diaziridine molecule of a chain and the NO3 ligand of the adjacent chain. Thus, favourable steric conditions of the ligand and hydrogen bonding probably lead to the formation of conglomerate 1. As was presumed earlier,25 the formation of racemic complex 2 can be used for increasing the optical purity of partially enriched (+)-1,2-dimethyldiaziridine. The treatment with a halfmole quantity of AgNO3 resulted in an increase in the optical purity by a factor of about three.† Similarly, we used chloral for increasing the optical purity of 1-methyl-3,3-pentamethylenediaziridine, which forms a racemic adduct with this compound. 14(c) The attempts to separate an optically active NH-diaziridine from homochiral complex 1 (and also perhaps from compound C23) were unsuccessful because of the easy racemization due to proton exchange. Thus, we attempt to prepare homochiral complexes of AgNO3 with 1,3-dimethyl- and 1,3,3-trimethyldiaziridines. This work was supported by the Russian Foundation for Basic Research (grant nos. 97-03-33021 and 98-03-04119) and INTAS (project no. 157). References 1 M. Fujita, N. Fujita, K. Ogura and K. Yamaguchi, Nature, 1999, 400, 52. 2 D. Whang and K. Kim, J. Am. Chem. Soc., 1997, 119, 451. 3 C. Piguet, G. Bemardinelli and G. Hopgartner, Chem. Rev., 1997, 97, 2005. 4 D. Philp and J. F. Stoddart, Angew. Chem., Int. Ed. Engl., 1996, 35, 1154. 5 A. K. Ghosh, P. Mathivanan and J. Cappiello, Tetrahedron Asymm., 1998, 9, 1. 6 D. A. Evans, E. J. Olhava, J. S. Johnson and J. M. Janey, Angew. Chem., Int. Ed. Engl., 1998, 37, 3372. 7 C.B.Aakeröy, Acta Crystallogr., 1997, B53, 569. 8 A. Williams, Chem. Eur. J., 1997, 3, 15. 9 T. L. Hennigar, D. C. MacQuarrir, P. Losier, R. D. Rogers and M. J. Zaworotko, Angew. Chem., Int.Ed. Engl., 1997, 36, 972. 10 P. Klüfers and J. Schuhmacher, Angew. Chem., Int. Ed. Engl., 1994, 33, 1742. 11 D. A. Evans, K. A. Woerpel and M. J. Scott, Angew. Chem., Int. Ed. Engl., 1992, 31, 430. 12 R. G. Kostyanovsky, R. Murugan and M. Sutharchanadevi, in Comprehensive Heterocyclic Chemistry, ed. A. Padwa, Pergamon Press, Oxford, 1996, vol. 1A, ch. 1.11, p. 347. 13 R. G. Kostyanovsky, G.V. Shustov, V. V. Starovoitov and I. I. Chervin, Mendeleev Commun., 1998, 113. 14 (a) R. G. Kostyanovsky, A. E. Polyakov and V. I. Markov, Izv. Akad. Nauk SSSR, Ser. Khim., 1974, 1671 (in Russian); (b) R. G. Kostyanovsky, A. E. Polyakov and V. I. Markov, Izv. Akad. Nauk SSSR, Ser. Khim., 1975, 198 (in Russian); (c) R. G. Kostyanovsky, A. E. Polyakov, G. V. Shustov, K. S.Zakharov and V. I. Markov, Dokl. Akad. Nauk SSSR, 1974, 219, 873 [Dokl. Chem. (Engl. Transl.), 1974, 219, 831]. 15 R. G. Kostyanovsky, A. E. Polyakov and G. V. Shustov, Tetrahedron Lett., 1976, 2059. 16 R. G. Kostyanovsky and G. V. Shustov, Dokl. Akad. Nauk SSSR, 1977, 232, 1081 [Dokl. Chem. (Engl. Transl.), 1977, 232, 71]. 17 R. G. Kostyanovsky, G. V. Shustov and N. L. Zaichenko, Tetrahedron, 1982, 38, 949. 18 G. V. Shustov, G. K. Kadorkina, R. G. Kostyanovsky and A. Rauk, J. Am. Chem. Soc., 1988, 110, 1719. 19 G. V. Shustov, S. N. Denisenko, M. A. Shochen and R. G. Kostyanovsky, Izv. Akad. Nauk SSSR, Ser. Khim., 1988, 1862 (Bull. Acad. Sci. USSR, Div. Chem. Sci., 1988, 37, 1665). 20 S. V. Konovalikhin, A. B. Zolotoi, L. O. Atovmyan, G. V. Shustov, S. N. Denisenko and R.G. Kostyanovsky, Izv. Akad. Nauk, Ser. Khim., 1995, 500 (Russ. Chem. Bull., 1995, 44, 483). 21 V. A. Korneev, G. V. Shustov, I. I. Chervin and R. G. Kostyanovsky, Izv. Akad. Nauk, Ser. Khim., 1995, 1396 (Russ. Chem. Bull., 1995, 44, 1346). 22 A. Adedapo, S. A. Benyunes, P. A. Chaloner, C. Claver, P. B. Hitchcock, A. Ruiz and N. Ruiz, J. Organomet. Chem., 1993, 443, 241. 23 A. Adedapo, A. G. Avent, D. Carmichael, P. A. Chaloner, P. B. Hitchcock and A. Wagnenaar, J. Chem. Soc., Chem. Commun., 1993, 186. 24 A. V. Shevtsov, V. Yu. Petukhova, S. A. Kutepov, V. V. Kuznetsov, N. N. Makhova, N. E. Kuzmina and G. G. Alexandrov, Izv. Akad. Nauk, Ser. Khim., submitted for publication. 25 G. V. Shustov, A. B. Zolotoi, S. V. Konovalikhin, L. O. Atovmyan and R. G. Kostyanovsky, Mendeleev Commun., 1995, 218. 26 J. Jacques, A. Collet and S. H. Willen, Enantiomers, Racemates, and Resolution, Krieger Publ. Co., Malabar, Florida, 1994, pp. 5, 104. 27 L. Carlucci, G. Ciani, D. M. Proserpio and A. Sironi, Angew. Chem., Int. Ed. Engl., 1995, 34, 1895. 28 J. F. Modder, K. Vrieze, A. L. Spek and G. van Koten, J. Org. Chem., 1991, 56, 5606. Figure 2 The crystal structure of 2: (a) the superposition of enantiomers in the Ag-coordinated heterochiral polymeric chain directed along the crystallographic axis a; (b) the arrangement of chains into ‘walls’ laying in the crystallographic plane ab. (a) (b) Ag(1' ) N(2B) N(1A) C(2A) C(2B) N(1B) N(2A) C(3A) C(3B) C(1) N(3) O(3) O(2) O(1) Ag(1) O(3) O(2) O(1) Ag(1) O(1' ) Ag(1' ) a b c O(3')Mendeleev Communications Electronic Version, Issue 2, 2000 (pp. 43–82) 29 R. G. Kostyanovsky, V. A. Korneev, I. I. Chervin, V. N. Voznesensky, Yu. V. Puzanov and P. Rademacher, Mendeleev Commun., 1996, 106. 30 H. D. Flack, Acta Crystallogr., 1983, A39, 876. Received: 30th December 1999; Com. 99/1587

ISSN:0959-9436     

出版商:RSC    

年代:2000

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 2, 2000 (pp. 43–82) Effect of proton donors on the mechanism of electroreduction of -radicals of linear and cyclic ethers Alexander G. Krivenko, Alexander S. Kotkin and Vladimir A. Kurmaz* Institute for Problems of Chemical Physics, Russian Academy of Sciences, 142432 Chernogolovka, Moscow Region, Russian Federation. Fax: +7 096 515 3588; e-mail: kurmaz@icp.ac.ru DOI: 10.1070/MC2000v010n02ABEH001254 Using laser photoemission, we found two routes of the electroreduction of a-radicals (Rads) adsorbed on a mercury electrode for linear and cyclic ethers in the presence of strong proton donors (BH+), i.e., by direct one-electron transfer and with the antecedent formation of the metastable complex [Rads·BH+].According to laser photoemission data, the irreversible electroreduction1 –4 and electrooxidation4,5 of radicals adsorbed on an electrode surface (Rads) proceeds via two parallel pathways, namely, by the direct one-electron transfer and by the electron transfer on the antecedently formed metastable complex6 of Rads with a proton donor/acceptor ([Rads·BH+]/ [Rads·BOH–]) The second pathway is predominant for radicals containing active functional groups, e.g., carbonyl,1,2 carboxyl7(a) and hydroxyl5 groups, or bifunctional radicals4,6 (k0 = 107–1010 dm3 mol–1 s–1; the second group of radicals6).Such complexes were not found for hydrocarbon radicals or their halogen derivatives1 (k0 < < 103 dm3 mol–1 s–1; the first group). To understand the nature of such differences, it seems important to investigate radicals which possess less active functional groups.For example, the reactivity of C–O bonds in ethers is much lower than that of C–O and O–H bonds in aliphatic alcohols. 8 Since the electronic structures and spectrophotochemical characteristics of these two classes of a-radicals are similar,9 it is likely that their electrochemical behaviour is also similar.Electrode reactions of a-radicals1,5,10,11 and b-radicals6,7(a),(b) of alcohols were studied in detail by photoemission and other techniques based on non-electrochemical generation of intermediates. There is few data on the electrochemistry of ether a-radicals. Although we found earlier using laser photoemission3 that the 1,4-dioxane radical belongs to the second group (k0 = = 107 dm3 mol–1 s–1), this sole example is probably not typical because 1,4-dioxane is a diether and therefore the radical may be more active than in other ethers, as was observed, e.g., in diols4 with relation to corresponding aliphatic alcohols.1,5 Therefore, we decided on more typical radicals of linear and cyclic ethers, namely, radicals of diethyl ether 1, 1,4-dioxane 2, tetrahydrofuran (THF) 3, 2,5-dimethyltetrahydrofuran 4 and tetrahydropyran (THP) 5 (Table 1).Only radicals 1–3 were studied previously by the polarography of pulse-radiolysis products11 and by photomodulated voltammetry.12 Radicals were generated according to the following reactions: where e– aq is hydrated electron; RH is an aliphatic ether; ka and kOH (kH) are the rate constants of reactions (3) and (4), respectively. The generation of radicals 1–5 only by reaction (4) is provided by considerable differences in the rate constants ka for the e– aq capture by N2O, H3O+ and ether molecules (6×109, 2.3×1010 and < 107 dm3 mol–1 s–1, respectively13) (Table 1).The N2O molecules are main scavengers of e– aq at pH � 3, and the H3O+ cations are main scavengers of e– aq in more acidic solutions.The measurements at pH no higher than ~2 are difficult to perform because of the appearance of the anodic–cathodic wave of H· near –0.6 V (SCE)1 and the dark discharge of H3O+ on mercury, leading to a deterioration in the signal-to-noise ratio. The radicals formed in reactions (3) and (4) diffuse to an electrode, become adsorbed on it and participate in electrode a R+ e– – Rads + e– R– Ve' (1) Rads + BH+ (or BOH–) [Rads·BH+] (or [Rads·BOH–]) + e– products k1 = k0[BH+] (or [BOH–]) k2 Ve (2) 2.0 1.5 1.0 0.5 0.0 1.0 1.5 1.4 1.6 1.8 1.0 1.5 j/j0 (arbitrary units) –E/V (SCE) Figure 1 Typical TRWs of radicals at UV illumination of the electrode at various modulation periods: (a) radical 5, tm = 30 ms; (b) radical 4, tm = = 300 ms; (c) radical 3, tm = 30 ms.Stationary mercury electrode. Supporting electrolytes: aqueous buffer solutions with the addition of 0.5 M KCl saturated with N2O; pH � 5. (a) (b) (c) N2 O (H+) + e– aq N2 + OH–+ OH· (H·) ka OH· (H·) + RH R·+ H2O (H2), kOH (kH ) (3) (4) 1 2 3 4 5 106 104 102 100 0.4 0.8 1.2 1.6 2.0 W2 , W3 /s–1 –E/V (SCE) Figure 2 Tafel plots for the rate constants of reduction and oxidation (1)– (5) of radicals 1–5, respectively. The data for the reduction of radical 2 at pH�7.0 were taken from ref. 3. The experimental conditions are the same as in Figure 1.Mendeleev Communications Electronic Version, Issue 2, 2000 (pp. 43–82) reactions to generate photocurrent j. The values of j were obtained by the measurements and numerical Fourier transformation of signals from a photoelectrochemical cell illuminated with modulated light at the period tm = 1.0–0.001 s.This results in the voltammetric time-resolved wave (TRW) of a radical. The position of the half-wave E1/2 on the E axis depends on tm and the type of R and is controlled by the competition between the characteristic time of irreversible reduction/oxidation of Rads with the rate constants W3/W2 and the electrode illumination time.At E = E1/2, W = ktm, where k = 5.31 or 10.88 for W = W3 or W2, respectively. Thus, the measurements of W3 values can be performed in the range 5–5.8×103 s–1. In the range 3×103– 6×106 s–1, the values of W3 were determined from measurements of the kinetics of emitted charge Q = f(t).† Typical TRWs for the oxidation and reduction of the radicals are presented in Figure 1.The experimental details were described elsewhere.1–7 It is well known that H· and OH· radicals are reactive and hence non-selective agents. Therefore, in addition to a-radicals, the formation of b-radicals or, moreover, g-radicals in the case of THP, by reaction (4) cannot be excluded a priori (such possibilities were repeatedly discussed in the literature).However, the TRWs of radicals 1–5 demonstrated the electroreduction or electrooxidation of only one type of intermediates. Consequently, under the experimental conditions, the fraction of b- and g-radicals does not exceed 5%, similarly to the EtOH–N2O system.1,5 Figure 2 shows the Tafel functions lg W2,3(E) for radicals 1–5 at pH varied from weakly acidic to strongly basic values.It can be seen that all of the functions are similar and have similar slopes (Table 1); W is independent of pH in the specified range. This fact is indicative of the similarity of mechanisms of electrode processes. Direct electron transfer via reaction (1) is predominant, i.e., the reduction of Rads to carbanions (analogously to the radicals of alcohols, hydrocarbons and their halogen derivatives1,5,10) and the oxidation to carbocations (as in the cases of methyl, ethyl1 and carboxyalkyl radicals7(a)).The lg W3(E) functions plotted for radicals 1–5 are shifted to more positive E values on the addition of NH4 + ions and at pH £ pH*, where pH* is a threshold value which is characteristic of each of the radicals, e.g., pH* ~ 3.3 for radical 3 (Figure 3).The (dE/dpH)W3 and {dE/d(–lg[NH4 +])}W3 values obtained by cross-section of the Tafel plots at W3 = const (Table 1) are close to the corresponding values of (2.3RT/F)/a. This fact characterises the electroreduction as a first-order reaction with respect to [BH+]. Thus, with increasing [BH+], a transition to the quasi-reversible reduction of the complex [Rads·BH+] occurred in all of the radicals.Simultaneously the obsrate constant was a linear or exponential function of [BH+] or E, respectively. In contrast to radical 2, the slopes of the Tafel plots for radicals 1 and 3–5 remained unchanged even at minimum [BH+]. Thus, the transition to the discharge, which is limited by the rate of complex formation, occurs at observed constants higher than Wm, i.e., at k1 � Wm ª 107 s–1 (Figure 3).Consequently, the value of k0 lies between 107–1010 and 109– 1010 dm3 mol–1 s–1 for the complexes with NH4 + and H3O+ ions, respectively. In terms of the model6 it is possible to determine the difference of overvoltages Dh for radicals 1–5 and their complexes‡ with BH+. For BH+ = H3O+, this difference decreased † With the use of laser photoemission, radicals were generated at distances of ~(10–100) Å from the electrode; this allowed us to measure W up to the maximum values Wm ª 107 s–1.as the Tafel plot was shifted towards the cathodic region from ~1.0 (radical 2) to ~0.45 V (radical 1); for BH+ = NH4 +, Dh is ~0.2–0.3 V, and the above correlation does not occur. Previously, 14–17 a much lower efficiency of NH4 + ions as proton donors, as compared with H3O+, was found in the bulk protonation of NO3 2– and CO2 ·– and in the electroreduction of stable anions.An analysis of the experimental results (Table 1) and a comparison with published data demonstrate that the position of Tafel plots for radicals 1–5 on the E axis correlates with the redox potentials of pairs§ E0 (R/R+), e.g., E0 2 >> E0 1, E0 3 (ref. 12), as well as with the rates of anodic methoxylation of ethers in the presence of solid polymer electrolytes19 and, generally, with the structure and reactivity of the corresponding molecules. For example, the insertion of an additional CH2 group into a cyclic ether molecule facilitates the electrooxidation, and the replacement of this group by an electronegative O atom leads to considerable inhibition of the process.18 The Tafel plot of the electroreduction is analogously shifted by ca. 0.40 V on the replacement of an O atom in radical 2 with a CH2 group (radical 5). As in the case of the parent cyclic ether molecules,18 the reduction rate of the radicals substantially increased (Figure 2) after the insertion of additional electronegative substituents (O atoms), and it was somewhat inhibited due to appearance of additional electropositive (CH2 or Me) groups (Table 1).This effect is most pronounced for structurally similar radicals, e.g., the shifts of ‡ The maximum values of k0 and k2 are diffusion-controlled constants of ~1010 dm3 mol–1 s–1 and the reciprocal of the lifetime ~1012 s–1 for a collisional complex.The following averaged k values were chosen for the calculation: k0 = 7×109 dm3 mol–1 s–1 and k2 = 1011 s–1. § The correspondence between the differences in the redox potentials of intermediates and the Tafel plots for their electroreduction was postulated previously.1 aData from ref. 3. bThe value was obtained by extrapolation of the Tafel plot to W2 = 103 s–1.cThe reduction wave of radical 5 is overlapped by the dark discharge of H3O+ at pH < 3.5. Table 1 Electrochemical properties of linear and cyclic ether radicals and the constants of capture13 of H atoms and OH radicals by the ether molecules. a-Radical kOH (kH)/dm3 mol–1 s–1 a (b), ±0.05 (dE/dpH)W3 (dE/d[NH4 + ])W3 –E±0.05/V at W3 = 103 s–1 –E±0.05/V at W2 = 103 Ethoxyethyl 1 (2.9–4.2)×109 (4.3×107) 0.59 (0.61) 0.12 (0.132) 1.74 0.79 1,4-Dioxanyl 2 (2.5–3.1)×109 (107) 0.47a (0.53) 0.13a 1.45 0.38b Tetrahydrofuran-2-yl 3 4×109 [(3.3–7.2)×107] 0.55 (0.58) 0.13 (0.114) 1.68 0.87 2,5-Dimethyltetrahydrofuran-2-yl 4 0.46 (0.63) 0.11 (0.13) 1.83 1.52 Tetrahydropyran-2-yl 5 0.54 (0.66) —c (0.142) 1.76 0.66 108 106 104 102 100 1.2 1.4 1.6 1.8 2.0 2.2 W3/s–1 –E/V (SCE) pH 1.7 pH 2.6 pH 3.1 pH 3.3–13.5 Figure 3 Typical relationship between W3 and H3O+ concentration.Radical 4; the experimental conditions are the same as in Figure 1. Dashed lines show the calculations of W3(E, pH) in terms of the model.6Mendeleev Communications Electronic Version, Issue 2, 2000 (pp. 43–82) the Tafel plots are ~0.25±0.03 V for radicals 2 and 5 and ~0.15±0.03 V for radicals 3 and 4.Thus, the a-radicals of linear and cyclic ethers can form complexes with proton donors at the rate constants of 107–1010 s–1, which depend on the nature of BH+ only slightly. In contrast, the reduction overvoltages for the [Rads·H3O+] and [Rads·NH4 +] complexes are substantially different. The electrochemical properties of adsorbed a-radicals of ethers and alcohols, i.e., the mechanisms and rates of electrode processes, were found to be similar. In both cases, the electrode reactions are described by the model,6 i.e., electron transfer proceeds by the two parallel pathways: directly to Rads or to the antecedently formed metastable complex of Rads with a proton donor or acceptor. The rate of electron transfer to the adsorbed radicals correlates with the structure and reactivity of the parent ether molecules.This work was supported by the Russian Foundation for Basic Research (grant no. 97-03-32265). References 1 V. A. Benderskii and A. G. Krivenko, Usp. Khim., 1990, 59, 3 (Russ. Chem. Rev., 1990, 59, 1). 2 V. A. Benderskii, A. G. Krivenko and V. A. Kurmaz, Elektrokhimiya, 1987, 23, 625 [Sov.Electrochem. (Engl. Transl.), 1987, 23, 577]. 3 V. A. Benderskii, A. G. Krivenko, A. S. Kotkin and V. A. Kurmaz, Elektrokhimiya, 1993, 29, 246 (Russ. J. Electrochem., 1993, 29, 221). 4 A. G. Krivenko, V. A. Benderskii, A. S. Kotkin and V. A. Kurmaz, Elektrokhimiya, 1993, 29, 869 (Russ. J. Electrochem., 1993, 29, 741). 5 V. A. Benderskii, A. G. Krivenko and V. A. Kurmaz, Elektrokhimiya, 1986, 22, 644 [Sov.Electrochem. (Engl. Transl.), 1986, 22, 607]. 6 A. G. Krivenko, A. S. Kotkin and V. A. Kurmaz, Mendeleev Commun., 1998, 56. 7 (a) A. G. Krivenko, A. P. Tomilov, Yu. D. Smirnov, A. S. Kotkin and V. A. Kurmaz, Zh. Obshch. Khim., 1998, 68, 292 (Russ. J. Gen. Chem., 1998, 68, 266); (b) V. A. Kurmaz, A. G. Krivenko, A. P. Tomilov, V. V. Turigin, A. V. Khudenko, N.N. Shalashova and A. S. Kotkin, Elektrokhimiya, 2000, 36, 344 (in Russian). 8 A. Hynes, in Comprehensive Organic Chemistry, ed. J. F. Stoddart, Pergamon Press, Oxford, 1979. 9 M. Ya. Melnikov and V. A. Smirnov, Handbook of Photochemistry of Organic Radicals, Begell House, New York, 1996. 10 Z. A. Rotenberg and N. M. Rufman, J. Electroanal. Chem., 1984, 175, 153. 11 A. Henglein, in Electroanal. Chem., ed. A. J. Bard, Marcel Dekker, New York, 1976, vol. 9, p. 163. 12 D. D. M. Wayner, D. J. McPhee and D. Griller, J. Am. Chem. Soc., 1988, 110, 132. 13 G. V. Buxton, C. L. Greenstock, W. Ph. Helman and A. R. Ross, J. Phys. Chem. Ref. Data, 1988, 17, 513. 14 N. V. Fedorovich, Zh. Anal. Khim., 1993, 48, 1006 [J. Anal. Chem. (Engl. Transl.), 1993, 48, 700]. 15 N. V. Fedorovich, Ross. Khim. Zh. (Zh. Ross. Khim. Ob-va im. D. I. Mendeleeva), 1996, 40, 86 (in Russian). 16 V. A. Benderskii, A. G. Krivenko, E. A. Ponomarev and N. V. Fedorovich, Elektrokhimiya, 1987, 23, 1435 [Sov. Electrochem. (Engl. Transl.), 1987, 23, 1343]. 17 D. Schiffrin, Disc. Faraday Soc., 1974, 56, 75. 18 J. L. Goldman, R. M. Mank, J. H. Young and V. R. Koch, J. Electrochem. Soc., 1980, 127, 1461. 19 L. Kröner, D. Hoormann, E. Steckhan, J. Jorissen and H. Putter, 21st Sandbjerg Meeting on Organic Electrochemistry, Sandbjerg, 1999, p. 47. Received: 23th December 1999; Com. 99/15

ISSN:0959-9436     

出版商:RSC    

年代:2000

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 2, 2000 (pp. 43–82) Apparent molar volumes and enthalpies of solution of tetracyanoethylene in some solvents and of butan-1-ol in n-octane at different concentrations Vladimir D. Kiselev,* Elena A. Kashaeva, Mahdy S. Shihaab and Alexander I. Konovalov A. M. Butlerov Chemical Institute, Kazan State University, 420008 Kazan, Russian Federation. Fax: +7 8432 31 5578; e-mail: kiselev@butler.kazan.ru DOI: 10.1070/MC2000v010n02ABEH001176 The effects of specific solute–solvent and solute–solute interactions on the change in the apparent molar volumes and in the enthalpies of solution, were found to be proportional in the title solutions at 25 °C.The partial molar volume of a solute can be calculated at infinite dilution using the following equations: Here, MA and MS are the molecular masses of the solute and the solvent, respectively; d and d0 are the densities of the solution and the solvent, respectively; cA, mA and xA are the concentration scales in molarity, molality and mole fraction, respectively.The partial molar volume of a solute from equations (1)–(3) includes the structural volume of the solute in this solvent and the volume change of the solvent in the process of shell formation around the solute.1 The last part depends on the difference between solute–solvent and solvent–solvent interaction energies.In the case of strong ion–solvent interactions, the value of solvent electrostriction can be larger than the value of the structural volume of the solute, and the partial molar volume may be negative.1 The change of the heat of solution and the molar volume of a solute at infinite dilution through the range of solvents reflect the changes of properties of both solute and solvent in the dissolving process, except the case of ideal solutions.All these changes of solute–solvent and solute–solute interactions define the changes of the apparent molar volume and the enthalpy of solution.In general, specific interactions can sharply affect the activation volumes and reaction volumes, and from these details the nature of high-pressure effects on the rates and equilibria can be better estimated.2 Appreciable volume changes can be expected from the specific solute–solvent interaction of the strong p-acceptor, tetracyanoethylene (TCNE) (EA = 2.88 eV),3 with p-donor alkylbenzenes.It is well known that the heat of solution of all alcohols in nonpolar solvents has unusually high curvature, and for the range of concentrations less than 0.02 mol dm–3 there is saturation due to the total monomerization. Similarly, S-shaped curves can be expected for the apparent molar volume of alcohols in dilute alkane solutions.However, the apparent molar volumes of alcohols in the alkane solutions were determined4–7 for the range of concentrations above 0.02–0.04 mol dm–3. The calculation of partial molar volumes of alcohols from the data of this concentration range cannot be correct. Tetracyanoethylene (Merck) was sublimed in a vacuum (50 Pa) at 110 °C as white crystals, mp 198–200 °C (lit.,8 200 °C).Butan- 1-ol and all solvents were purified by known methods.9 The enthalpies of solution were measured at 25 °C using a differential calorimeter (the solvent volume was 180 cm3) as reported previously.10 Calibration of the calorimeter by the heat of solution of KCl in water at 25 °C gave 17.4±0.2 kJ mol–1 (precise data11 17.51±0.01 kJ mol–1). Three to five measurements with sequentially dissolving samples (30–100 mg) of TCNE were carried out.The integral heats of solution of butanol in n-octane were obtained by sequentially dissolving samples in three independent sets of experiments. Apparent molar volumes of solutes in solutions were determined by means of a digital vibratingtube densimeter (A. Paar, DMA 602) with an accuracy of (1–2)×10–6 g cm–3.The high level of thermostatic control [25± ±(1–2)×10–3 °C] was achieved using a triple cascade of water thermostats (22, 24.5 and 25 °C) with a reduced heater power in the last one (15 W, 20 dm3 of water).2 The densimeter was placed in a box at a constant temperature of air (25±0.2 °C). The apparent molar volume of TCNE in each of solutions (Table 1) was invariable in the concentration range 0.02–0.05 mol dm–3.The apparent molar volume of butan-1-ol in octane solutions depends on concentration in the whole range examined. Three to five measurements of the density were carried out for all concentrations of solutions. No change in the densities of TCNE and butanol solutions was observed in these solvents within few hours. Solute–solvent interactions. The correlation coefficient of the empirical relationship between the value of VTCNE and the enthalpies of solution of TCNE in alkylbenzenes [ TCNE = (100.76±0.63) + (0.413±0.021)DsolH; R = 0.9620, SV = 1.17, N = 6] is less than that with the values of ionization potentials of alkylbenzenes [ TCNE = (14.5±7.6) + (10.25±0.35)IP; R = = 0.9823, SV = 0.80, N = 6] and the free energy of complex formation [ TCNE = (109.40±0.66) + (1.536±0.062)DG0; R = = 0.9763; SV = 0.93, N = 6].Because of a large difference of the effects of chlorine and methyl groups on the IP values and complex formation (Table 1), chlorobenzene can be withdrawn VA = 1000(d0 – d)/cAd0 + MA/d0 VA = 1000(d0 – d)/mA dd0 + MA/d VA = (d0 – d)MS(1 – xA)/xAdd0 + MA/d (1) (2) (3) aFrom ref. 12 for alkylbenzenes and from ref. 13 for other solvents. bCalculated using the value DsublH 81.2 kJ mol–1.14 cFrom ref. 15. dFrom ref. 16. Table 1 Ionization potentials of solvents (IP), the partial molar volume of tetracyanoethylene (VTCNE), the enthalpies of solution (DsolH) and solvation (DsolvH), the free energy (DG0) of complex formation of TCNE with alkylbenzenes and the rate constants (k2) of the Diels–Alder reaction of TCNE with anthracene at 25 °C.Solvent IPa/eV VTCNE/cm3 mol–1 DsolH/kJ mol–1 –DsolvHb/kJ mol–1 DG0 a / kJ mol–1 k2 c/dm3 mol–1 s–1 Chlorobenzene 9.10 109.23±0.14 23.1±0.3 58.1±0.3 –0.65 1.82 Benzene 9.25 108.40±0.26 14.9±0.4 66.3±0.4 1.72 0.38 Toluene 8.82 104.56±0.37 9.7±0.5 71.5±0.5 3.24 0.13 o-Xylene 8.58 102.06±0.32 1.4±0.1 79.8±0.1 4.81 0.061 p-Xylene 8.48 101.46±0.25 0.0±0.5 81.2±0.5 5.04 — Mesitylene 8.14 98.07±0.10 –2.7±0.4 83.9±0.4 7.07 0.01 Acetonitrile 12.12 109.97±0.12 15.2±0.2 66.0±0.2 — 2.18 Ethyl acetate 9.54 112.09±0.06 9.2±0.5 72.0±0.5 — 0.24 Cyclohexanone 9.14 110.42±0.35 7.6±0.3 73.6±0.3 — 0.20 p-Dioxane 9.13 105.72±0.16 4.3±0.2 76.9±0.2 — 0.34 1,2-Dichloroethane 11.12 107.81±0.20 21.3±0.3 59.9±0.3 — 3.82 Dichloromethane 11.35 107.50±0.20d 23.4±0.5 57.8±0.5 — 4.28 V V VMendeleev Communications Electronic Version, Issue 2, 2000 (pp. 43–82) from consideration with alkylbenzenes.12 In this case a better correlation is observed [ TCNE = (22.43±1.4) + (9.299±0.073)IP; R = 0.9994; SV = 0.13, N = 5]. It is clear (Table 1) that the changes in the rate constants of the Diels–Alder reaction with TCNE, the heat of solution of TCNE in alkylbenzenes and the free energy of complex formation are connected with changes in the specific energy of interaction between TCNE and these aromatic solvents.From the correlations obtained, we can conclude that the variation of the partial molar volume of TCNE in alkylbenzenes is conditioned by the same cause. Inert solvents such as alkanes and carbon tetrachloride were excluded due to the very low solubility of TCNE.For chlorobenzene, dichloromethane and 1,2-dichloroethane, the heats of solution are more endothermic, and the enthalpies of TCNE solvation are only –(58–60) kJ mol–1 (Table 1). The heat of solution of TCNE is exothermic in mesitylene, the strongest p-donor among aromatic solvents of this series (Table 1), and the enthalpy of this complex formation is equal12 to –18.9 kJ mol–1.Even in this solution, the enthalpy of specific interaction contributes only one third of the enthalpy of nonspecific interaction (Table 1). Note that the enthalpies of interaction of TCNE in the crystal state and in a p-xylene solution are the same, and the molar volume of crystalline TCNE (lit.,17 102 cm3) is nearly the same as in a p-xylene solution (Table 1).The rate constants go down by a factor of 400 with the change from 1,2-dichloroethane or dichloromethane to mesitylene (Table 1). The correlation between the free energy and the enthalpy of complex formation12 of alkylbenzenes with TCNE with donor properties allows a prediction of the equilibrium constant (150 dm3 mol–1) and the enthalpy of complex formation (–30 kJ mol–1) for the complex between TCNE and the strongest p-donor, 9,10-dimethylanthracene (lit.,18 IP = 7.04 eV).The strongest specific intermolecular interaction between TCNE and 9,10-dimethylanthracene in inert solutions may be a reason for the observed negative temperature coefficient of the rate constants of the Diels–Alder reaction.19 The structure of n,p-complexes differs from the coplanar structure of p,p-complexes of alkylbenzenes with TCNE.20 All these types of correlation become poor in n-donor solvents (ethyl acetate, cyclohexanone, acetonitrile and dioxane).Solutions of TCNE in chloroalkanes are almost free from specific interactions, but partial molar volumes of TCNE in dichloromethane and dichloroethane are slightly lower than those in a benzene solution (Table 1).Solute–solute interactions. The increase in the proportion of monomer alcohol molecules can be seen in the increase of the apparent molar volume and the enthalpy of solution (Table 2). From experimental measurements4–7 in the concentration range above 0.02–0.04 mol dm–3 the apparent molar volumes of alcohols increase sharply with decreasing concentration.However, the S-shape of the saturation curve appears from calorimetric measurements at concentrations less than 0.02 mol dm–3. The fitting of experimental data in the usual way to the limit of concentration 0.0285 mol dm–3 gave an overestimated value for the partial molar volume of butanol (dotted line in Figure 1). It is difficult to do precise measurements of the density difference of solution and solvent at concentrations less than 0.01 mol dm–3 (Dd < 8×10–5 g cm–3), but more sensitive calorimetric measurements extended here up to 0.003 mol dm–3 (Table 2).In the low concentration range (< 0.02 mol dm–3), a greater part of n-butanol exists as the monomer, and the change in the heat of solution exhibits saturation.This S-shaped curve of the heat of solution of butan-1-ol in n-octane in the low concentration range is comparable to nearly the same S-shaped curve of the apparent molar volume of butan-1-ol in this concentration range. This saturation was realised taking into account an additional value of the apparent volume for a concentration of 0.0146 mol dm–3 (Figure 1), giving the partial molar volume of 99.27±0.20 cm3 mol–1.The limiting value of the heat of solution of butanol in n-octane evaluated here is 23.2 kJ mol–1. Nearly the same values (23.6) were obtained by us for the limiting heat of solution of butan-1-ol in n-hexane and for ethanol in n-hexane (23.5 kJ mol–1).21 With the assumption that the heat of solution and the change in the apparent molar volume of butanol caused by monomer formation, the monomer fraction of butanol in an octane solution can be calculated from the equation: where CM and C are the monomer and total butanol concentrations in solution, respectively; DsolHC and DsolH0 are the enthalpies of solution at the working concentration and at the limiting low concentration, respectively; DjC is the difference between the partial ( ) and apparent (jC) molar volumes of aAccuracy of ±0.15 kJ mol–1.bAccuracy of ±0.20 cm3 mol–1. cThe calculated limiting values (see the text) are given in parentheses. dFor pure butanol. eThe apparent molar volume of butanol in n-heptane at 20 °C from ref. 7. Table 2 Apparent molar volumes (j) and the integral enthalpies of solution (DsolH) of butan-1-ol in n-octane at 25 °C.Cbutanol / mol dm–3 DsolHa/ kJ mol–1 Cbutanol / mol dm–3 jb/ cm3 mol–1 Djb/ cm3 mol–1 0 (23.2)c 0 (99.27)c (7.30)c 0.00348 23.0 0.0146 98.95 6.98 0.00659 22.6 0.0285 98.50 6.53 0.0107 22.4 0.0353 98.20 6.23 0.0156 22.1 0.0536 97.64 5.67 0.0214 21.6 0.0845 96.81 4.84 0.0278 20.9 0.115 96.23 4.26 0.0349 19.8 0.524 94.08 2.11 0.0422 18.8 0.977 93.57 1.60 0.0537 17.2 10.87d 91.97 0 0.0681 15.5 0.0853 13.8 0.0332 97.19e 5.66 0.107 12.2 0.0766 96.08e 4.55 0.145 10.2 0.144 94.82e 3.29 0.190 8.7 0.182 94.58e 3.05 0.232 7.7 0.275 7.0 V 8 7 6 5 0 1 2 3 4 5 6 7 8 9 10 11 12 4 24 22 20 18 16 14 12 10 Cbutanol/10–2 mol dm–3 Dj/cm3 mol–1 DHsn/kJ mol–1 Figure 1 Concentration effect of butan-1-ol in an n-octane solution on the integral enthalpy of solution ( ) and on the apparent molar volume of butan-1-ol ( ).For dotted line, see the text. aThe values of DjC were calculated for the same butanol concentrations as those used for calculating the value of DsolHC. Table 3 Partial monomer fraction (a) of butanol in an octane solution and calculated equilibrium constants for dimer (KD) and trimer (KT) processes. Cbutanol / mol dm–3 a (DsolHC/ DsolH0) KD/ dm3 mol–1 KT/ dm6 mol–2 aa (DjC/ DV) KD/ dm3 mol–1 KT/ dm6 mol–2 0.00659 0.9741 2.1 215 0.980 1.6 165 0.0107 0.9655 1.7 112 0.969 1.5 96 0.0156 0.9526 1.7 75 0.951 1.7 76 0.0214 0.9310 1.9 62 0.922 2.1 71 0.0278 0.9009 2.2 59 0.890 2.5 67 0.0349 0.8534 2.9 64 0.853 2.9 65 0.0422 0.8103 3.4 67 0.819 3.2 62 0.0537 0.7414 4.9 82 0.774 3.5 56 0.0681 0.6681 5.5 80 0.712 4.2 58 0.0853 0.5948 6.7 88 0.651 4.8 58 0.107 0.5259 8.0 95 0.599 5.2 54 a = CM/C = DsolHC/DsolH0 = DjC/DV (4) VMendeleev Communications Electronic Version, Issue 2, 2000 (pp. 43–82) butanol in octane solution, respectively; DV is the volume difference between the partial molar volume of butanol in n-octane solution and the molar volume of pure butanol. The equilibrium constants for monomer–dimer (2M = D) and monomer–trimer (3M = T) equilibria can be calculated from the following equations: The value of a = DsolHC/DsolH0 is based on the additional assumption that there are slight differences in nonspecific solute– solute and solute–solvent interactions. To check the last assumption, the heat of solution of diethyl ether (as a homomorph of butanol) in octane was measured.This value (1.6±0.2 kJ mol–1) is low in comparison with the limiting value of the heat of solution of butanol in octane (23.2±0.2 kJ mol–1), but does not change the results of calculations summarised in Table 3. The attempt to distinguish these two types of equilibria has failed. We can suppose that in this range of concentrations both types of equilibria take place with different contributions.The data in Table 3 indicate that the value of KD is nearly constant only in the low concentration range (£ 0.02 mol dm–3) and at higher concentrations (0.02–0.1 mol dm–3) the trimer–monomer equilibrium can be the dominant process. This work was supported by the Russian Foundation for Basic Research (grant no. 98-03-33053). References 1 F. J. Millero, Chem.Rev., 1971, 71, 147. 2 V. D. Kiselev, E. A. Kashaeva and A. I. Konovalov, Tetrahedron, 1999, 55, 1153. 3 K. N. Houk and L. L. Munchausen, J. Am. Chem. Soc., 1976, 98, 937. 4 M. Costas and M. Caceres Alonso, Ber. Bunsenges. Phys. Chem., 1987, 91, 184. 5 M. R. Kumaran and G. C. Benson, J. Chem. Thermodyn., 1983, 15, 245. 6 A. J. Treszczanowicz and G. C. Benson, Fluid Phase Equilib., 1988, 41, 31. 7 L. A. Staveley and B. Spice, J. Chem. Soc., 1952, 406. 8 L. Cairns, R. A. Carboni, D. D. Coffman, V. A. Engelhardt, R. E. Heckert, L. Little, E. G. McCeer, B. C. McKusick, W. J.Middleton, R. M. Scribner, C. W. Theobald and H. E. Widberg, J. Am. Chem. Soc., 1958, 80, 2775. 9 A. Weissberger, Organic Solvents, Interscience, New York, 1955. 10 V. D. Kiselev, E. A.Kashaeva, N. A. Luzanova and A. I. Konovalov, Thermochim. Acta, 1997, 303, 225. 11 G. Somsen, J. Coops and M. W. Tolk, Recl. Trav. Chim., 1963, 82, 231. 12 R. E. Merrifield and W. D. Phillips, J. Am. Chem. Soc., 1958, 80, 2778. 13 N. Kondrat’ev, Energii razryva svyazei. Potentsialy ionizatsii i srodstvo k elektronu (Bond Energies. Ionization Potential and Electron Affinity), Nauka, Moscow, 1974 (in Russian). 14 J. D. Cox and G. Pilcher, Thermochemistry of Organic and Organometallic Compounds, Academic Press, London, 1970. 15 V. D. Kiselev and A. I. Konovalov, Zh. Org. Khim., 1974, 10, 6 [J. Org. Chem. USSR (Engl. Transl.), 1974, 10, 4]. 16 J. Jouanne, H. Kelm and R. Huisgen, J. Am. Chem. Soc., 1979, 101, 151. 17 F. K. Fleischmann and H. Kelm, Tetrahedron Lett., 1973, 39, 3773. 18 V. D. Kiselev and A. I. Konovalov, Usp. Khim., 1989, 58, 383 (Russ. Chem. Rev., 1989, 58, 230). 19 V. D. Kiselev and J. G. Miller, J. Am. Chem. Soc., 1975, 97, 4036. 20 R. Foster, Organic Charge-Transfer Complexes, Academic Press, New York, 1969, p. 216. 21 R. H. Stokes and C. Burfitt, J. Chem. Thermodyn., 1975, 7, 803. KD = (1 – a)/2Ca2, KT = (1 – a)/3C2a3. (5) (6) Received: 16th June 1999; Com. 99/1504

ISSN:0959-9436     

出版商:RSC    

年代:2000

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 2, 2000 (pp. 43–82) A theoretical study of the catalytic cycle of ethylene hydrogenation by a bipalladium cluster Viktor M. Mamaev,*a Igor P. Gloriozov,a Dmitrii A. Lemenovskiia and Yurii V. Babinb a Department of Chemistry, M. V. Lomonosov Moscow State University, 119899 Moscow, Russian Federation. Fax: +7 095 938 8846; e-mail: vmam@nmr.chem.msu.su b Far Eastern State Academy of Economics and Management, 690091 Vladivostok, Russian Federation.Fax: +7 4232 40 6634 DOI: 10.1070/MC2000v010n02ABEH001216 In terms of the reaction-path Hamiltonian formalism, the catalytic cycle of ethylene hydrogenation by the Pd2 cluster has been found to involve 5 reactions with 10 stationary points on the potential-energy surface.In the last 20 years, a great body of data concerning quantumchemical studies of catalytic reactions with the participation of organometallic compounds were published. Among these reactions, which are important steps in the catalytic cycles of hydrocarbon conversion, are olefin addition,1 oxidative addition and reductive elimination involving the activation of H–H, C–H and C–C bonds.2–7 Recently, two reviews concerning the theoretical studies of catalytic reactions were issued.8 However, a study of the full catalytic cycle involving a series of elementary reactions remains to be difficult to perform.Morokuma and co-workers performed theoretical studies of the full cycles of homogeneous olefin hydrogenation,9 hydroboration10 and hydroformylation11 with a RhI complex catalyst. They also theoretically examined alkyne and alkene diboration with diphosphine complexes of Pd and Pt.12 The reasons for the inactivity of C2H4 (a high barrier in the insertion of the hydrocarbon into a Pt–B bond and an endothermic effect) and of the palladium complex in alkyne diboration [the impossibility of the oxidative addition of a B–B bond to Pd(PH3)2 because of an endothermic effect and a very low barrier of 0.1 kcal mol–1 for a reverse process] were found. Albert et al.13 examined a catalytic cycle of ethylene addition to phenyl bromide with the participation of a palladium complex with two diaminocarbene ligands. It was found that ethylene insertion into the Pd–Ph bond and b-hydride elimination (barriers from 8.3 to 11.5 kcal mol–1) are the key steps of the catalytic cycle.The above studies involved calculations of the stationary points of elementary steps using ab initio and density functional theory (DFT) methods; based on these calculations, the potential-energy profile of the full catalytic cycle was chosen. The aim of this work was to study theoretically the catalytic cycle of ethylene hydrogenation with the Pd2 cluster.Previously, 14 we used the reaction-path Hamiltonian (RPH) approximation15 for studying in detail the oxidative addition of the H2 molecule to the Pd2 cluster. We found that the potential-energy surface of this reaction exhibits a complex shape with several valleys and stationary points. A pseudo-square-planar complex with the energy of –34 kcal mol–1 with respect to separated reactants corresponds to the global minimum.A reaction path (RP) with no barrier leads to this complex. Another RP, via a bifurcation point and a transition state (TS), leads to the trans-product with the energy higher by 18 kcal mol–1. In this work, we supplemented this reaction to the catalytic cycle of ethane formation from ethylene: The CNDO/S2 semiempirical technique16 was used for quantum- chemical calculations of PESs for complex molecular systems (MSs) including transition metal atoms on the basis of the RPH approach.The technique was parametrised on the basis of both experimental data and ab initio high-level calculations. We repeatedly used the above technique for studying the activation of H–H and C–H bonds by the Pd atom and the Pd2 cluster.7,14 Additional calculations were performed for the main stationary points of the PES using the DFT method in two approximations for exchange correlation energy (DFT-BLYP17 and DFT-PBE18), which were implemented in the program.19 A detailed study of the PESs of reactions (1) and (2) by the CNDO/S2 technique based on the RPH approach allowed us to separate five interrelated RPs (RP1, RP2, RP3, RP4 and RP5), elementary steps of the full catalytic cycle.Figure 1 shows the schematic diagram of the cycle, Figure 2 — the potential-energy profile, and Figure 3 demonstrates the structures of the stationary points. As mentioned above, RP1 with no barrier from the separated reactants (SR) H2 and Pd2 leads to bridged pseudo-square-planar complex Pd2H2 I (Figure 3).The RPs are represented as the normal coordinate s with dimensions of Å u1/2 (i.e., the mass of proton is taken as unity). Reaction (1) was studied in detail elsewhere.14 In further studies, we found that species I cannot add an ethylene molecule with hydrogen atom transfer, and it forms only an adsorption complex. trans-Form II was found to be active in the propagation of the catalytic cycle.The calculations of product I by the DFT-BLYP method are consistent with the CNDO/S2 calculations, whereas DFT-PBE results in a lower energy, which can be explained by overestimated exchange and correlation contributions in the bridged structure (Figure 3). Pd2 + H2 Pd2H2, Pd2H2 + C2H4 Pd2 + C2H6. (1) (2) Pd Pd H H Pd Pd H H Pd Pd H H C C H H H H Pd Pd H C C H H H H H Pd Pd H C C H H H H H Pd Pd H C C H H H H H Pd Pd C C H H H H H H Pd2 I PC V II III IV' IV oxidative addition H2 isomerization adsorption C2H4 addition rotation isomerization reductive elimination elimination – C2H6 Figure 1 Schematic diagram of the catalytic cycle of ethylene hydrogenation by the Pd2 cluster.Mendeleev Communications Electronic Version, Issue 2, 2000 (pp. 43–82) RP2 corresponds to a change from pseudo-square-planar complex I to trans-reactant II in the separated reactants Pd2H2 and C2H4.The RP2 vector involves a shortening in the Pd–Pd distance and an increase in the Pd–Pd–H plane angles. The energy of the system increased by 17.3 kcal mol–1 with a barrier of 21.5 kcal mol–1 (TS1). The TS1 saddle point exhibits the second imaginary frequency (w2), which corresponds to the rotational vibration of Pd–H bonds with respect to the Pd–Pd bond.The motion along the reaction path that corresponds to the rotational vibration leads to the cis-product which is higher than the transproduct in terms of energy.14 The DFT calculations also gave similar results for the structures of stationary points (Figure 2).Note that they also gave two imaginary frequencies in TS1. At the same time, for the planar trans-product, the DFT-BLYP method gave no minimum, and the potential barrier was 19.7 kcal mol–1. On going to the DFT-PBE method, the potential barrier increased up to 28.2 kcal mol–1 with a decrease in the energy in product II by 1.3 kcal mol–1. Note that a frequency (~50i cm–1) corresponding to the RP3 vector (approach of Pd2H2 and C2H4) was present in all stationary points calculated by the CNDO/S2 and DFT methods.RP3 consists of two portions: a gentle portion (coordination of an ethylene molecule to a palladium atom with the formation of intermediate III) with a decrease in energy (by 9.1, 9.3 or 10.8 kcal mol–1 according to the CNDO/S2, DFTBLYP or DFT-PBE method, respectively) at a long segment of 9 Å u1/2 and a steep portion associated with synchronous rupture of the Pd–H bond, formation of a new C–H bond and transfer of the second H atom from a Pd atom to the other.The second region of RP3 includes the transition state (TS2) with a barrier of 4.3, 6.5 or 5.8 kcal mol–1 according to the CNDO/S2, DFT-BLYP or DFT-PBE method, respectively.Next, the energy dramatically decreased to product IV (by 30.0, 20.9 or 18.6 kcal mol–1 ac- 0 –10 –20 –30 –40 –50 –60 0 10 20 30 40 Pd2 + H2 + C2H4 Pd2 + C2H6 I II III IV, IV' V DE/kcal mol–1 Reaction coordinate s/Å u1/2 TS1 TS2 TS3 TS4 PC Figure 2 Potential-energy profile along the normal coordinate (s) of the catalytic cycle of ethylene hydrogenation by the Pd2 cluster.SR FP H3 H4 H7 H8 H9 H10 Pd2 Pd1 1.30 (1.35) [1.34] 1.70 (1.71) [1.70] 2.71 (2.77) [2.88] C6 C5 I DE –34.1 (–31.6) [–46.4] Pd2 Pd1 H7 H8 H9 H10 1.30 (1.34) [1.34] C6 C5 1.55 (1.55) [1.53] 2.42 (2.61) [2.61] H3 H4 –12.4 (–11.9) [–18.2] TS1 H7 H8 H9 H10 1.30 [1.34] C6 C5 Pd2 Pd1 1.54 [1.56] 2.26 [2.65] H3 H4 II –16.6 [–19.5] H7 H8 H9 H10 C6 C5 Pd2 Pd1 H3 H4 III 1.53 (1.55) [1.54] 2.29 (2.40) [2.31] 2.16 (2.34) [2.22] 2.30 (2.64) [2.62] 1.58 (1.61) [1.62] –25.7 (–21.2) [–30.3] H7 H8 H9 H10 C6 C5 Pd2 Pd1 H3 H4 TS2 –21.4 (–14.7) [–24.5] DE 2.22 (2.57) [2.57] 1.58 (1.54) [1.54] 2.30 (2.65) [2.56] H7 H8 H9 H10 C6 Pd2 Pd1 H3 H4 IV 1.50 (1.54) [1.52] 1.56 (1.71) [1.60] 2.27 (2.72) [2.56] C5 (1.75) 1.90 (2.22) [2.04] –51.4 (–36.5) [–43.1] H8 H9 H10 C4 Pd2 Pd1 H3 H6 IV' 1.60 (1.69) [1.59] 2.26 (2.80) [2.65] C5 (1.73) 1.89 (2.12) [2.04] –51.4 (–35.6) [–44.0] H8 H9 H10 C4 Pd2 Pd1 H3 H6 TS3 1.71 2.57 C5 1.89 –39.8 1.89 H7 2.58 1.52 H7 H8 H9 H10 C4 C5 Pd2 Pd1 H3 V –46.5 (–38.5) [–52.5] DE 1.88 (2.07) [2.04] 1.51 (1.52) [1.52] 2.53 (2.77) [2.71] H6 1.79 (1.75) [1.75] 1.81 (1.69) [1.66] H7 H8 H10 C4 C5 Pd2 Pd1 H3 TS4 –37.7 (–25.9) [–32.2] 1.87 (1.80) [1.70] 1.51 (1.54) [1.54] 2.62 (2.72) [2.73] H6 1.95 (1.79) [2.00] 2.03 (2.22) [2.16] H7 H8 H10 C4 C3 Pd2 Pd1 H5 PC –46.3 (–39.5) [–49.0] 1.49 (1.54) [1.53] 2.85 (2.81) [2.79] H6 H9 Figure 3 Structures of stationary points calculated by the CNDO/S2, DFT-BLYP (in parentheses) and DFT-PBE (in square brackets) methods (bond lengths/Å; relative energy DE/kcal mol–1).Mendeleev Communications Electronic Version, Issue 2, 2000 (pp. 43–82) cording to the CNDO/S2, DFT-BLYP or DFT-PBE method, respectively). This latter is the trans-product of C–H bond activation in ethane (the CNDO/S2 and DFT-PBE methods) or a bridged product (DFT-BLYP). The next portion of the catalytic cycle PES is associated with the vibrational mode analogous to RP1. The vector corresponding to RP4 rearranges trans-product IV' (formed from IV by almost barrierless rotation about the Pd–C bond) to product V with the bridging bond Pd–H–Pd turned through 90° with respect to the C–Pd–Pd plane (Figure 3).As calculated by the CNDO/S2 method, product V is higher than the trans-product by 4.7 kcal mol–1 in terms of energy; the barrier towards it (TS3) is 11.6 kcal mol–1, and the width is about 4 Å u1/2.This barrier is an artefact of calculations by the CNDO/S2 method, which overestimated the stability of products IV and IV' (according to DFT calculations, a barrier was absent in this process). However, even the potential barrier TS3 ª 10 kcal mol–1 obtained using the CNDO/S2 method is no restriction for proceeding the catalytic cycle. Product V exhibit a vibrational mode that results in the final step of the catalytic cycle, reductive elimination of ethane.The RP5 vector involves an approach of the H3 proton to the C4 atom by rotation about the Pd–Pd bond from product V to the saddle point (TS4) with an increase in energy by 9.4 kcal mol–1 (Figure 3). Next, from TS4, the rupture of the bridging Pd–H–Pd bond and the formation of a new C–H bond occur synchronously to result in the pre-reaction complex (PC) of separated Pd2 and C2H6 molecules (FP) with the energy higher than that in V by 0.4 kcal mol–1. The total positive energy effect of reactions of the catalytic cycle is 42.6 kcal mol–1.In conclusion, note that the results obtained by calculations using the CNDO/S2 and DFT methods are in good agreement. Moreover, according to our data, the ethylene conversion into ethane proceeds only after the formation of trans-product II from pseudo-square-planar complex I (see RP2).In this case, ethylene can initially interact only with one activated hydrogen atom (see RP3). This fact is consistent with the experimental data20 that hydrogen atoms escaped from Ni(111) bulk metal to the surface easily hydrogenate ethylene adsorbed on the surface to ethane.At the same time, according to experimental data, surface boundary hydrogen atoms do not exhibit hydrogenating activity; this is consistent with our conclusion on the inactivity of pseudo-square-planar complex I. It is believed that bimetallic complexes that easily produce a trans-form by adding H2 will be the most effective catalysts.The new catalytic system21 based on the coordination of bimetallic transition metal clusters on metal porphyrins exhibits this property. References 1 (a) A. Dedieu, Inorg. Chem., 1981, 20, 2803; (b) C. A. Jolly and D. S. Marynick, J. Am. Chem. Soc., 1989, 111, 7968; (c) H. Kawamura- Kuribayashi, N. Koga and K. Morokuma, J. Am. Chem. Soc., 1992, 114, 2359; (d) P. E. M. Siegbahn, J.Am. Chem. Soc., 1993, 115, 5803; (e) N. Koga, S. Obara, K. Kitaura and K. Morokuma, J. Am. Chem. Soc., 1985, 107, 7109. 2 N. Koga and K. Morokuma, Chem. Rev., 1991, 91, 823. 3 (a) A. L. Sargent and M. B. Hall, Inorg. Chem., 1992, 31, 317; (b) A. L. Sargent, M. B. Hall and M. F. Guest, J. Am. Chem. Soc., 1992, 114, 517. 4 (a) S. Sakaki and M. Ikei, J. Am. Chem. Soc., 1991, 113, 5063; (b) S.Sakaki and M. Ikei, J. Am. Chem. Soc., 1993, 115, 2373; (c) N. Koga and K. Morokuma, J. Am. Chem. Soc., 1993, 115, 6883; (d) S. Sakaki, B. Bismas and M. Sugimoto, Organometallics, 1998, 17, 1278. 5 (a) V. I. Avdeev and G. M. Zhidomirov, Kinet. Katal., 1996, 37, 775 [Kinet. Catal. (Engl. Transl.), 1996, 37, 722]; (b) M.-D. Su and S.-Y. Chu, J. Am. Chem. Soc., 1997, 119, 5376; (c) M.-D.Su and S.-Y. Chu, Organometallics, 1997, 16, 1621; (d) M.-D. Su and S.-Y. Chu, J. Phys. Chem., A, 1997, 101, 6798. 6 (a) E. Clot and O. Eisenstein, J. Phys. Chem., A, 1998, 102, 3592; (b) S. A. Macgregor, O. Eisenstein, M. K. Whittlesey and R. N. Perutz, J. Chem. Soc., Dalton Trans., 1998, 291. 7 (a) V. M. Mamaev, I. P. Gloriozov, S. Ya. Ischenko, V. V.Simonyan, E. M. Myshakin, A. V. Prisyajnyuk and Yu. A. Ustynyuk, J. Chem. Soc., Faraday Trans., 1995, 91, 3779; (b) V. M. Mamaev, I. P. Gloriozov, A. V. Prisyajnyuk and Yu. A. Ustynyuk, Mendeleev Commun., 1996, 203; (c) V. M. Mamaev, A. V. Prisyajnyuk, I. P. Gloriozov, S. Ya. Ischenko, Yu. A. Ustynyuk and L. N. Alekseiko, Kinet. Katal., 1998, 39, 178 [Kinet. Catal. (Engl. Transl.), 1998, 39, 162]; (d) V.M. Mamaev, I. P. Gloriozov and A. V. Prisyajnyuk, Mendeleev Commun., 1998, 53. 8 (a) S. Niu and M. B. Hall, Chem. Rev., 2000, 100, 353; (b) M. Torrent, M. Solá and G. Frenking, Chem. Rev., 2000, 100, 439. 9 (a) N. Koga, C. Daniel, J. Han, X. Y. Fu and K. Morokuma, J. Am. Chem. Soc., 1987, 109, 3455; (b) C. Daniel, N. Koga, J. Han, X. Y. Fu and K. Morokuma, J.Am. Chem. Soc., 1988, 110, 3773. 10 D. G. Musaev, A. M. Mebel and K. Morokuma, J. Am. Chem. Soc., 1994, 116, 10693. 11 T. Matsubara, N. Koga, Ya. Ding, D. G. Musaev and K. Morokuma, Organometallics, 1997, 16, 1065. 12 (a) Q. Cui, D. G. Musaev and K. Morokuma, Organometallics, 1997, 16, 1355; (b) Q. Cui, D. G. Musaev and K. Morokuma, Organometallics, 1998, 17, 742. 13 K. Albert, Ph. Gisdakis and N. Rösch, Organometallics, 1998, 17, 1608. 14 V. M. Mamaev, I. P. Gloriozov, V. V. Simonyan, E. V. Zernova, A. V. Prisyajnyuk and Yu. A. Ustynyuk, Mendeleev Commun., 1997, 246. 15 W. H. Miller, J. Phys. Chem., 1983, 87, 3811. 16 M. J. Filatov, O. V. Gritsenko and G. M. Zhidomirov, J. Mol. Catal., 1989, 54, 452. 17 (a) A. D. Becke, Phys. Rev., 1988, A38, 3098; (b) C. Lee, W. Yang and R. G. Parr, Phys. Rev., 1988, B37, 785. 18 J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1996, 77, 3865. 19 D. N. Laikov, Chem. Phys. Lett., 1997, 281, 151. 20 S. P. Daley, A. L. Utz, T. R. Trautman and S. T. Ceyer, J. Am. Chem. Soc., 1994, 116, 6001. 21 V. M. Mamaev, I. P. Gloriozov, D. A. Lemenovskii and E. V. Zernova, Kinet. Katal., 2000, 41, 33 (in Russian). Received: 27th October 1999; Com. 99/1544

ISSN:0959-9436     

出版商:RSC    

年代:2000

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 2, 2000 (pp. 43–82) First direct calculation of the partial quadrupole splitting of ligands for the prediction of Mössbauer spectra parameters in low-spin iron(II) complexes Victor N. Nemykin,*a,b Anna E. Polshina,a,c Ernst V. Polshind and Nagao Kobayashi*a a Department of Chemistry, Graduate School of Science, Tohoku University, Sendai 980-8578, Japan.Fax: +81 222 17 7719; e-mail: victor_nemykin@yahoo.com b V. I. Vernadsky Institute of General and Inorganic Chemistry, National Academy of Sciences of Ukraine, 252680 Kiev, Ukraine. Fax: +380 44 444 3070 c Institute of Sorption and Endoecological Problems, National Academy of Sciences of Ukraine, 252680 Kiev, Ukraine. Fax: +380 44 452 9327 d Institute of Geochemistry, Mineralogy and Ore Formation, National Academy of Sciences of Ukraine, 252680 Kiev, Ukraine.Fax: +380 44 444 0342 DOI: 10.1070/MC2000v010n02ABEH001207 A semi-empirical quantum mechanics method and a cone angle conception were used to factorise partial quadrupole splitting parameters for different ligands in axially coordinated macrocyclic complexes. The 57Fe Mössbauer spectroscopy is one of the most powerful techniques for probing the metal centre in inorganic and organometallic compounds of iron.1 In spite of large progress in the theoretical analysis of Mössbauer spectra parameters on the basis of modern quantum theory methods, such as DFT and ab initio,2 the partial quadrupole splitting (p.q.s.) and partial central shift (p.c.s.) models for the prediction of Mössbauer spectra parameters are still popular and useful in the case of low-spin iron(II) compounds.3 The p.q.s.parameters have been introduced, developed and successfully applied over the last 20 years.3,4 The p.q.s. model is attractive because of the following: (i) tabulated p.q.s. values already contain the sum of all ligand properties on which the experimental quadrupole splitting (D) depends; (ii) the experimental geometries of compounds are not required; (iii) experimental D values can be easily estimated ‘by hand’ {see, for instance, equations (1) and (2) for the estimation of D in trans-[FeA2B4] and trans-[FeACB4] complexes}.3 The application of equations (1) and (2) to the pcFeL2, pcFeL' L'', nx2FeL2 and nx2FeL'L'' complexes, where [B] is pc/4, dmg/2 or nx/2, [A] is L or L' , and [C] is L'' , leads to the estimation of D values for the complexes of interest.The p.q.s. values can be found from experimental D values3 or from structural parameters of the corresponding complexes.4 However, the factorisation of the p.q.s. of new ligands without a wide range of experimental data is a difficult problem. Here, we consider the relationship between the electronic and geometric structures of ligands and the p.q.s.values in low-spin iron(II) complexes. The first direct calculations† of p.q.s. given below can be easily used for the calculations of p.q.s. values of unknown ligands. In general, the p.q.s. values include the following three main contributions: (i) A ‘lattice’ contribution, which depends on the ligand charge or, more correctly, the ligand coordination atom (LCA) and on the distance between the LCA and the iron atom.This interaction leads to a change in the lattice contribution to p.q.s. Usually, for low-spin iron(II) complexes, this lattice contribution is small and can be neglected.1 However, recent DFT calculations in a † All computations were performed using the HyperChem 5.1 Pro program (HyperCube Inc.) on a Pentium PC.All ligand structures were fully optimised by the gradient Polak–Ribiere method at the semi-empirical AM1 level.7 The molecular electrostatic potential (VMEP) and the ‘molecular back-bonding potential’ (Vb) were evaluated as described previously.5 The ligand resonance integrals were calculated in the NDO formalism.8 Steric parameters of axial ligands were evaluated from a cone angle conception.6 The effective van der Waals radius of the p-system was chosen as 1.7 Å.9 Multiparameter regression analysis was carried out by the Powell quadratic convergence method.10 local density approximation for the oepFe(pme3)2 complex2(a) have shown that at least in some cases the lattice contribution can play a dominant role.(ii) A ‘valence’ contribution, which arises from the s- and p-interaction between iron and LCA atom orbitals and leads to a change of the effective population of iron atom orbitals. For example, the s-donation of electron density by a ligand changes the effective population of 3d and 4p orbitals of the iron atom. The relationship between the s- and p-donor and p-acceptor properties of ligands and the and p.q.s.values have been widely discussed.3 (iii) A ‘steric’ contribution of the ligand, on which the metal– ligand distance and, correspondingly, the metal–ligand interaction depends. The influence of cone angle on p.q.s. parameters for phosphorus-containing ligands was discussed recently.3( f ) D = 4p.q.s.(A) – 4p.q.s.(B) D = 2p.q.s.(A) + 2p.q.s.(C) – 4p.q.s.(B) (1) (2) aAbbreviations: 14en, 1,4-diaminobutane; 4ohpy, 4-hydroxypyridine; 4chopy, pyridine-4-carbaldehyde; bu, butyl; dabco, diazabicyclooctane; meim, N-methylimidazole; mepip, 4-methylpiperidine; mph, morpholine; nh3, ammonia; ph, phenyl; pr, propyl; py, pyridine; pzl, pyrazole; tht, tetrahydrothiophene; tz, 1,2,4,5-tetrazine; tzl, 1,2,4-triazole; .bp.q.s.in mm s–1. The p.q.s.(Cl–) = –0.27 mm s–1 scale has been used for all values, see ref. 3. cAll p.q.s. values from this work are based on the pcFeL2, (Hnx)2FeL2 and (Hdmg)2FeL2 complexes, where pc2– is the phthalocyanine dianion, Hnx– is the nioximate anion and Hdmg– is the dimethylglioximate anion.11 Table 1 Ligand parameters estimated from experimental data and calculated p.q.s.values. La –VMEP/eV Vb/eV C×102 T p.q.s. (calc.) p.q.s. (exp.)b [Ref.] nh3 3.09 0.035 –1.0 93 –0.54 –0.52 [3(b), 4] n-prnh2 3.24 0.036 –0.65 110 –0.46 –0.48 [4] s-bunh2 3.28 0.035 –0.75 113 –0.44 –0.45 [this workc] t-bunh2 3.35 0.036 –0.77 129 –0.36 –0.36 [this workc] phnh2 2.45 0.036 1.4 110 –0.44 –0.43 [this workc] mepip 3.13 0.036 –0.27 123 –0.39 –0.39 [this workc] mph 3.02 0.038 –0.25 123 –0.39 –0.38 [this workc] 14en 3.84 0.03 –0.82 110 –0.5 –0.48 [this workc] dabco 2.99 0.041 0.05 145 –0.27 –0.24 [this workc] py 2.86 0.093 –0.87 100 –0.51 –0.48 [this workc] 4ohpy 2.99 0.091 –1.46 100 –0.51 –0.51 [this workc] 4chopy 2.53 0.11 –0.76 100 –0.5 –0.5 [this workc] tz 2.2 0.219 –0.47 100 –0.51 –0.52 [this workc] meim 3.32 0.079 –1.6 98 –0.53 –0.53 [4] pzl 2.67 0.087 –1.4 98 –0.51 –0.51 [this workc] tzl 2.91 0.096 –2.05 98 –0.51 –0.53 [this workc] t-bunc 2.07 0.319 1.9 76 –0.68 –0.72 [3(d)] phnc 2.04 0.362 1.6 76 –0.68 –0.7 [3(b)] cn– 8.59 0.165 –2.5 76 –0.86 –0.84 [4] co 1.51 0.633 5.5 78 –0.75 –0.74 [3(d)] pet3 1.9 0.12 26.29 118 –0.59 –0.59 [this workc] pobu3 0.59 0.368 32.96 109 –0.69 –0.68 [this workc] tht 1.28 0.1 8.5 122 –0.4 –0.41 [this workc] pc2– 11.97 0.06 –1.57 85 –0.93 –0.96 [4] Hdmg– Hnx– 12.07 0.048 –0.32 85 –0.94 –0.92 [4]Mendeleev Communications Electronic Version, Issue 2, 2000 (pp. 43–82) Thus, an increase in the cone angle of a ligand probably leads to the elongation of the Fe–LCA bond. This elongated Fe–LCA bond decreases the overlap between the lone-pair orbital of the ligand and the 3d and 4p orbitals of iron and hence leads to a change in D.Note that this statement is true only for conformationally rigid ligands. From the above, we can conclude that the p.q.s. values for each ligand depend on the s- and p-donor and p-acceptor properties, the LCA charge and the cone angle. Recently, Fielder et al.5 have developed a model for the calculation of the s- and p-donor and p-acceptor capabilities of ligands. In this model, the s- and p-donor properties were evaluated from the molecular electrostatic potential function, VMEP, while the p-acceptor properties were derived using the secondorder perturbation theory, and we adopt these two parameters (VMEP and Vb) for factorisation of the electronic structure of ligands. The ‘lattice’ contribution of ligands to p.q.s.can be roughly proportional to Where ZLCA is the LCA charge, and r is the probe atom–LCA distance. The steric factors of ligands were evaluated using the cone angle conception.6 The final expression for evaluating the p.q.s. values is the following: where T is the cone angle of the ligand,6 C is the ‘lattice’ contribution of LCA and VMEP and Vb were derived as described previously.5 The results obtained for 25 ligands are presented in Table 1.The final parameters are a0 = –0.96172 mm s–1, a1 = –0.03752 mm s–1 eV–1, a2 = –0.20849 mm s–1 eV–1, a3 = = –0.75555 mm s–1 au–1, a4 = 0.0057 mm s–1 deg–1 with the correlation coefficient 0.995 for 25 compounds with a root-meansquare error of 0.018 mm s–1. Our correlation line combines different classes of ligands (alkyl- and arylamines, six- and fivemembered heterocycles, isonitriles, carbon monoxide, sulfides, phosphines and phosphites). An increase in the s- and p-donor or p-acceptor properties of ligands leads to a decrease in the p.q.s.values, this is consistent with qualitative assumptions.3,4 As can be concluded from the results of the above regression analysis, an increase in the steric effect of ligands leads to an increase in the p.q.s.value, in accordance with experimental data.3( f ) It is interesting that the cone angle is more important for most of the ligands, as compared to VMEP and Vb, which are important only for ligands with strong s-donor and/or p-acceptor properties, while C is important only for phosphorus-containing ligands with a good agreement with recent DFT calculations.The values of D, both calculated from equations (1) and (2) and experimental, for macFeL2 and macFeL'L'' complexes, where mac is pc2–, Hdmg– or Hnx–, are presented in Table 2. The correlation coefficient r2 = 0.975 with a root-mean-square error of 0.10 mm s–1 for 27 complexes has been observed. The computed D values are lower than the experimental data.Probably, this fact suggests that the cone angle conception for pc2– and Hnx– is not a good estimation. Really, the recomputed D values, in which experimental p.q.s. values were used for macrocyclic ligands, are in better agreement with experimental D values (Table 2). Thus, we have proposed a new technique for the estimation of generally usable ligand p.q.s. values. The correlation coefficient obtained with a small root-mean-square error for p.q.s.values indicates that the proposed technique is a simple and correct tool for predicting D values in inorganic and organometallic compounds. VNN is grateful to JSPS for financial support (grant no. P98418). This work was partially supported by INTAS (grant no. 97-0791). References 1 Chemical Mössbauer Spectroscopy, ed.R. H. Herber, Plenum Press, New York, 1984. 2 (a) M. Grodzicki, H. Flint, H.Winkler, A. F. Walker and A. X. Trautwein, J. Phys. Chem., 1997, 101, 4202; (b) R. H. Havlin, N. Godbout, R. Salzmann, M. Wojdelski, W. Arnold, C. E. Schulz and E. Oldfield, J. Am. Chem. Soc., 1998, 120, 3144. 3 (a) L. M. D. R. S. Martins, M. T. Duarte, A. M. Galvao, C. Resende, A.J. L. Pombeiro, R. A. Henderson and D. J. Evans, J. Chem. Soc., Dalton Trans., 1998, 3311; (b) G. M. Bancroft, Coord. Chem. Rev., 1973, 11, 47; (c) J. M. Bellerby, M. J. Mays and P. L. Sears, J. Chem. Soc., Dalton Trans., 1976, 1232; (d) D. J. Evans, M. Jimenez-Tenorio and G. J. Leigh, J. Chem. Soc., Dalton Trans., 1991, 1785; (e) G. M. Bancroft and E. T. Libbey, J. Chem. Soc., Dalton Trans., 1973, 2103; ( f ) J.Silver, Inorg. Chim. Acta., 1991, 184, 235. 4 A. Y. Nazarenko, E. V. Polshin and Ya. Z. Voloshin, Mendeleev Commun., 1993, 45. 5 S. S. Fielder, M. C. Osborne, A. B. P. Lever and W. J. Pietro, J. Am. Chem. Soc., 1995, 117, 6990. 6 C. A. Tolman, Chem. Rev., 1977, 77, 313. 7 M. J. S. Dewar, E. G. Zoebisch, E. F. Healy and J. J. P. Stewart, J. Am. Chem.Soc., 1985, 107, 3902. 8 M. J. S. Dewar and W. Thiel, J. Am. Chem. Soc., 1977, 99, 4839. 9 A. J. Gordon and R. A. Ford, The Chemist’s Companion, Willey, New York, 1972. 10 R. P. Brent, Algorithms for Minimization without Derivatives, Prentice– Hall, Endlewood Cliffs, New Jersey, 1973. 11 (a) B. W. Dale, R. J. R. Williams, P. R. Edwards and C. E. Johnson, Trans. Farad. Soc., 1968, 64, 620; (b) R.Taube, Pure Appl. Chem., 1974, 38, 427; (c) V. N. Nemykin, V. Y. Chernii, E. V. Polshin and S. V. Volkov, Ukr. Khim. Zh., 1997, 63, 75 (in Russian); (d) M. Hanack, U. Keppeler, A. Lange, A. Hirsch and R. Dieing, in Phthalocyanines: Properties and Applications, eds. A. B. P. Lever and C. C. Leznoff, VCH Publishers, New York, 1989, vol. 1, part 2. aCalculated using p.q.s.values of –0.96 and –0.92 for pc2– and Hnx–, respectively. Table 2 Calculated and experimental D values for macFeL2 and macFeL'L'' complexes. Complex D (calc.) D (calc.)a D (exp.) [ref.] pcFe(n-prnh2)2 1.88 2.00 1.97 [11(b)] pcFe(s-bunh2)2 1.96 2.08 2.04 [11(c)] pcFe(t-bunh2)2 2.32 2.40 2.38 [11(a)] pcFe(phnh2)2 1.96 2.08 2.11 [11(b)] pcFe(mepip)2 2.20 2.28 2.28 [11(c)] pcFe(mph)2 2.20 2.28 2.31 [11(c)] pcFe(dabco)2 2.68 2.76 2.89 [11(d)] pcFe(14en)2 1.80 1.84 1.84 [11(c)] pcFe(py)2 1.68 1.80 2.02 [11(d)] pcFe(4chopy)2 1.70 1.84 1.84 [11(d)] pcFe(4ohpy)2 1.72 1.84 1.80 [11(d)] pcFe(tz)2 1.64 1.80 1.79 [11(d)] pcFe(meim)2 1.60 1.72 1.71 [this work] pcFe(pzl)2 1.68 1.80 1.79 [this work] pcFe(tzl)2 1.64 1.80 1.73 [this work] pcFe(t-bunc)2 0.96 1.12 0.80 [11(d)] pcFe(phnc)2 0.96 1.12 0.68 [11(d)] pcFe(co)2 0.72 0.84 0.82 [11(d)] pcFe(pet3)2 1.40 1.48 1.47 [11(d)] pcFe(pobu3)2 0.96 1.08 1.09 [this work] pcFe(tht)2 2.12 2.24 2.20 [11(d)] pcFe(nh3)(co) 1.12 1.24 1.02 [11(d)] pcFe(py)(co) 1.20 1.34 1.19 [11(d)] (Hnx)2Fe(n-bunh2)2 1.92 1.84 1.83 [11(a)] (Hnx)2Fe(py)2 1.72 1.64 1.79 [11(a)] (Hnx)2Fe(4chopy)2 1.76 1.68 1.68 [11(a)] (Hdmg)2Fe(py)2 1.72 1.64 1.67 [1] C = ZLCA/r3 (3) p.q.s. (calc.) = a0 + a1VMEP + a2Vb + a3C + a4T, (4) Received: 22nd September 1999; Com. 99/1535

ISSN:0959-9436     

出版商:RSC    

年代:2000

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 2, 2000 (pp. 43–82) Unexpected isomerization in the series of fluorine-containing 2,3-dihydro-1H-1,4- diazepines with a 2-aminoethyl group at one of the nitrogen atoms Vyacheslav Ya. Sosnovskikh,*a Ivan I. Vorontsov,b Valentin A. Kutsenkoa and Yurii G. Yatluka a Department of Chemistry, A. M. Gor’ky Urals State University, 620083 Ekaterinburg, Russian Federation.Fax: +7 3432 61 5978; e-mail: Vyacheslav.Sosnovskikh@usu.ru b A. N. Nesmeyanov Institute of Organoelement Compounds, Russian Academy of Sciences, 117813 Moscow, Russian Federation. Fax: +7 095 135 5085; e-mail: ivorontsov@xray.ineos.ac.ru DOI: 10.1070/MC2000v010n02ABEH001209 Under kinetic control conditions, 6-methoxy-2-(1,1,2,2-tetrafluoroethyl)chromone reacts with diethylenetriamine to form 5-(2-hydroxy- 5-methoxyphenyl)-7-(1,1,2,2-tetrafluoroethyl)-1,4,8-triazabicyclo[5.3.0]dec-4-ene, and the latter undergoes isomerization to the title compound, which is more thermodynamically stable.Previously,1 we described the reaction between 2-polyfluoroalkylchromones and diethylenetriamine with the formation of a 1,4,8-triazabicyclo[5.3.0]dec-4-ene system, which can be considered as the cyclic form of 2,3-dihydro-1H-1,4-diazepines with a 2-aminoethyl group at the nitrogen atom nearest to the polyfluoroalkyl substituent.This work was devoted to an unusual behaviour of 6-methoxy-2-(1,1,2,2-tetrafluoroethyl)chromone 1a in a reaction with diethylenetriamine. We found that if chromone 1a reacts with diethylenetriamine in an alcoholic solution for a day, expected dihydrodiazepine 2a which exists in the cyclic form 1,4,8-triazabicyclo[5.3.0]dec-4-ene 3a (a kinetic control product) is formed.† However, a high-melting isomer (a thermodynamic control product) was isolated in place of 3a when the reaction was performed for a week.Based on the spectral characteristics, the structure of dihydrodiazepine 4 with a 2-aminoethyl group at the nitrogen atom remote from the tetrafluoroethyl substituent was ascribed to this isomer.‡ We found in special experiments that compound 3a undergoes isomerization to 4 on keeping its alcoholic solution at room temperature for a week.Note that in the case of trifluoromethylated chromone 1b the reaction stopped at a step of the formation of bicyclic compound 3b§ and did not result in the isomerization product under similar conditions.This is likely due to a higher capability of the CF3 group to stabilise an imidazoline ring, as compared with the CF2CF2H group.2 The signals of aliphatic protons in the 1H NMR spectrum of the high-melting isomer are complicated, and alternative struc- † 5-(2-Hydroxy-5-methoxyphenyl)-7-(1,1,2,2-tetrafluoroethyl)-1,4,8-triazabicyclo[ 5.3.0]dec-4-ene 3a: yield 56%, mp 127–128 °C. 1H NMR (250 MHz, CDCl3) d: 2.11 (br. s, 1H, NH), 2.89–3.25 [m, 5H, CH2(9), CH2(10), CHH(2)], 3.34 [AB system, Dd 0.16 ppm, 2H, CH2(6), JAB 15.5 Hz], 3.40–3.54 [m, 1H, CHH(2)], 3.78 (s, 3H, MeO), 3.92 [ddd, 1H, CHH(3), 2J 14.8 Hz, 3J 6.8 and 4.8 Hz], 4.14 [dt, 1H, CHH(3), 2J 14.8 Hz, 3J 6.0 Hz], 6.11 (tdd, 1H, CF2CF2H, 2JH,F 53.5 Hz, 3JH,F 8.1 and 4.5 Hz), 6.89 [d, 1H, H(3'), oJ 9.0 Hz], 6.96 [dd, 1H, H(4'), mJ 2.7 Hz], 7.07 [d, 1H, H(6')], 15.19 (br.s, 1H, OH). IR (Vaseline oil, n/cm–1): 3300 (NH), 1620 (C=N), 1585, 1505 (C=N, arom.). Found (%): C, 53.26; H, 5.21; N, 11.77. Calc. for C16H19F4N3O2 (%): C, 53.18; H, 5.30; N, 11.63. ‡ 1-(2-Aminoethyl)-5-(1,1,2,2-tetrafluoroethyl)-7-(2-hydroxy-5-methoxyphenyl)- 2,3-dihydro-1H-1,4-diazepine 4: yield 42%, mp 182–183 °C. 1H NMR (250 MHz, CDCl3) d: 2.7–4.0 (m, 8H, 3CH2, NH2), 3.77 (s, 3H, MeO), 4.0–4.3 (m, 2H, CH2–N=), 5.18 (s, 1H, CH=), 6.26 (tt, 1H, CF2CF2H, 2JH,F 53.4 Hz, 3JH,F 5.6 Hz), 6.69 [d, 1H, H(6'), mJ 2.3 Hz], 6.82–6.90 [m, 2H, H(4'), H(3')]. IR (Vaseline oil, n/cm–1): 3380, 3345 (NH2), 1610 (C=N), 1540, 1510 (C=C, arom.).Found (%): C, 53.21; H, 5.49; N, 11.75. Calc. for C16H19F4N3O2 (%): C, 53.18; H, 5.30; N, 11.63. § 5-(2-Hydroxy-5-methoxyphenyl)-7-trifluoromethyl-1,4,8-triazabicyclo- [5.3.0]dec-4-ene 3b: yield 37%, mp 100–101 °C. 1H NMR (250MHz, CDCl3) d: 2.14 (br. s, 1H, NH), 2.93–3.23 [m, 5H, CH2(9), CH2(10), CHH(2)], 3.33 [AB system, Dd 0.06 ppm, 2H, CH2(6), JAB 15.7 Hz], 3.42–3.51 [m, 1H, CHH(2)], 3.77 (s, 3H, MeO), 3.96 [ddd, 1H, CHH(3), 2J 15.1 Hz, 3J 6.8 and 4.8 Hz], 4.16 [dt, 1H, CHH(3), 2J 15.1 Hz, 3J 6.0 Hz], 6.89 [d, 1H, H(3'), oJ 8.8 Hz], 6.95 [dd, 1H, H(4'), mJ 2.8 Hz], 7.02 [d, 1H, H(6')], 15.21 (br.s, 1H, OH). IR (Vaseline oil, n/cm–1): 3345 (NH), 1625 (C=N), 1580, 1510 (arom.). Found (%): C, 54.62; H, 5.38; N, 12.94. Calc. for C15H18F3N3O2 (%): C, 54.71; H, 5.51; N, 12.76.tures cannot be eliminated on this basis. For this reason, we examined the crystals of compound 4 using X-ray diffraction analysis.¶ Figure 1 demonstrates the general view of a molecule of 4 and the numbering of atoms. The central seven-membered heterocyclic ring is nonplanar: the N(2), C(8), C(9) and C(10) atoms are arranged in a plane [the average deviation from a root-mean-square plane is 0.025(5) Å], and the N(1), C(11) and C(12) atoms are out of the plane by –0.356(7), –0.738(9) and ¶ Crystallographic data for 4: C16H19F4N3O2, monoclinic crystals.At 300 K, a = 7.821(6), b = 9.394(6), c = 23.785(16) Å, b = 80.64(6)°, V = = 1724(2) Å3, dcalc = 1.392 g cm–3, absorption coefficient m = 0.12mm–1, space group P21/n, Z = 4.The intensities of 3720 independent reflections (Rint = 0.06) were measured on a Siemens P3/PC automatic four-circle diffractometer (MoKa radiation, l = 0.71093 Å, graphite monochromator, q/2q scan, 2qmax = 55°). The structure was solved by the direct method with the use of the SHELXTL PLUS 5.0 program package.3 Nonhydrogen atoms were refined by the full-matrix least-squares procedures (with F2) in an anisotropic approximation.Atoms of MeO and CF2CF2H groups are disordered between two positions with equiprobable occupation; in the course of refining, the scatter of O(1)–C(7A), O(1)–C(7B) and C–F bond lengths was restricted by 0.02 Å. The positions of hydrogen atoms bonded to carbon atoms were calculated and included in the refinement by the rider model with fixed C–H distances (0.97 Å) and isotropic shift parameters Uiso = 1.5Ueq for methyl groups and 1.2Ueq for the other (Ueq are equivalent isotropic shift parameters of corresponding C atoms).The positions of hydrogen atoms of amino and hydroxyl groups were found by a difference Fourier synthesis and refined in an isotropic approximation. The final discrepancy factors R1 = 0.074, wR2 = 0.19, GOOF = 1.172 for 2486 reflections with I > 2s(I).Atomic coordinates, bond lengths, bond angles and thermal parameters have been deposited at the Cambridge Crystallographic Data Centre (CCDC). For details, see ‘Notice to Authors’, Mendeleev Commun., Issue 1, 2000. Any request to the CCDC for data should quote the full literature citation and the reference number 1135/61.O MeO O RF N N NH2 RF OMe OH 1a,b 2a,b (NH2CH2CH2)2NH N N RF OMe OH 3a,b NH 1 2 3 4 5 6 7 8 9 10 N N 4 4 3 2 1 7 6 5 OH MeO H2N a RF = CF2CF2H b RF = CF3 Scheme 1 CF2CF2HMendeleev Communications Electronic Version, Issue 2, 2000 (pp. 43–82) +0.268(8) Å, respectively. The planar fragment of the heterocycle and the plane of the benzene ring form an angle of 79.4(3)°.Disordered methoxy groups are turned about the C(4)–O(1) bond by 21° relative to each other. The positions of disordered CF2CF2H groups are characterised by a turn of 18° about the C(10)–C(15) bond. The terminal CF2H groups have different orientations with respect to the C(10)–C(15) bond: the torsion angles C(10)–C(15)–C(16)–H(16) and C(10)–C(15)–C(16A)– H(16A) are equal to 38.9(8) and –44.9(8)°, respectively. The formation of dihydrodiazepine 4 from bicyclic compound 3a, which is a cyclic form of dihydrodiazepine 2a, can be schematically represented as follows: It is likely that the isomerization begins with C(7)–N(1) bond rupture in kinetic control product 3a to form triazacyclodecadiene 5, which is shown in Scheme 2 as the most probable two tautomers.The subsequent intramolecular transamination at the carbon atom bonded to the aryl substituent proceeds via bicyclic nucleophilic-addition product 6, which then undergoes ring opening to form dihydrodiazepine 4. The latter compound is not prone to ring–chain tautomerism and, unlike dihydrodiazepine 2a, exists only in the open form because of the impossibility of the aryl group to stabilise the imidazolidine ring.4 It is well known5,6 that b-aminovinyl ketones (X = O) and thiones (X = S) with the geminate arrangement of an amino group and a polyfluoroalkyl substituent undergo irreversible isomerization to compounds with the g-arrangement of the above groups. Analogous isomerization in the series of b-aminovinyl imines (X = NH) was not reported in the literature, and the reaction under consideration is the first example of isomerization of b-aminovinyl imines with the geminate arrangement of amino and RF groups to form b-aminovinyl imines with the g-arrangement of the above groups.This work was supported by the Russian Foundation for Basic Research (grant nos. 99-03-32960, 97-03-33783, 96-15-97367 and 96-07-89187). References 1 V.Ya. Sosnovskikh, Yu. G. Yatluk and V. A. Kutsenko, Izv. Akad. Nauk, Ser. Khim., 1999, 1825 (Russ. Chem. Bull., 1999, 48, 1800). 2 V. Ya. Sosnovskikh and V. A. Kutsenko, Izv. Akad. Nauk, Ser. Khim., 1999, 546 (Russ. Chem. Bull., 1999, 48, 540). 3 G.M.Sheldrick, SHELXTL-Plus. Release 5.0, Siemens Analytical Instruments Inc., Madison, Wisconsin, USA, 1994. 4 K. N. Zelenin and I.V. Ukraintzev, Org. Prep. Proced. Int., 1998, 30, 109. 5 V. I. Filyakova, K. I. Pashkevich and I. Ya. Postovskii, Izv. Akad. Nauk SSSR, Ser. Khim., 1981, 2651 (Bull. Acad. Sci. USSR, Div. Chem. Sci., 1981, 30, 2207). 6 V. I. Filyakova, I. G. Busigin and K. I. Pashkevich, Zh. Org. Khim., 1989, 25, 1865 [J. Org. Chem. USSR (Engl. Transl.), 1989, 25, 1685]. N(3) C(14) C(13) N(2) C(12) C(11) N(1) C(10) C(15) F(2) C(16A) F(4A) F(3A) F(3) F(4) C(16) F(1A) F(1) F(2A) C(9) C(8) O(2) C(1) C(2) C(3) C(4) C(5) C(6) O(1) C(7A) C(7B) Figure 1 Molecular structure of 4.Selected bond lengths (Å): N(1)–C(10) 1.293(4), N(1)–C(11) 1.447(5), N(2)–C(8) 1.358(4), N(2)–C(12) 1.450(5), N(2)–C(13) 1.453(5), N(3)–C(14) 1.435(5), C(2)–C(8) 1.493(4), C(8)–C(9) 1.363(5), C(9)–C(10) 1.431(5), C(10)–C(15) 1.511(5), C(11)–C(12) 1.509(6), C(13)–C(14) 1.528(5); selected bond angles (°): C(10)–N(1)–C(11) 115.9(3), C(8)–N(2)–C(12) 119.6(3), C(8)–N(2)–C(13) 123.4(3), N(2)–C(8)–C(9) 125.2(3), N(2)–C(8)–C(2) 116.9(3), C(9)–C(8)–C(2) 117.7(3), C(8)–C(9)– C(10) 131.1(3), N(1)–C(10)–C(9) 132.8(3), N(1)–C(10)–C(15) 110.7(3), C(9)–C(10)–C(15) 116.6(3), N(1)–C(11)–C(12) 113.3(4), N(2)–C(12)–C(11) 113.0(4), N(2)–C(13)–C(14) 111.6(3); selected torsional angles (°): N(1)– C(11)–C(12)–N(2) –89.7(5), C(10)–N(1)–C(11)–C(12) 50.7(5), C(11)– N(1)–C(10)–C(9) 1.9(7), C(1)–C(2)–C(8)–N(2) –79.1(4), C(12)–N(2)– C(13)–C(14) –74.5(4), N(2)–C(8)–C(9)–C(10) –9.7(6), N(2)–C(13)–C(14)– N(3) 170.6(3), C(8)–C(9)–C(10)–C(15) 166.9(4).Ar N N NH2 RF 2a Ar N N RF 3a NH Ar N N 4 H2 N R F = CF2CF2H, Ar = Scheme 2 RF NH N N RF H Ar N HN N RF H Ar Ar N N R F HN 5 6 HO OMe X R RF NH2 NH2 R RF X X = O, S, NH Scheme 3 Received: 30th September 1999; Com. 99/1537

ISSN:0959-9436     

出版商:RSC    

年代:2000

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 2, 2000 (pp. 43–82) Interaction of 5-methoxy-1,2,4-triazines with ureas as a new route to 6-azapurines Dmitry G. Beresnev,a Gennady L. Rusinov,*a Oleg N. Chupakhina and Hans Neunhoefferb a Institute of Organic Synthesis, Urals Branch of the Russian Academy of Sciences, 620219 Ekaterinburg, Russian Federation. Fax: +7 3432 74 1189; e-mail: rusinov@ios.uran.ru b Institute of Organic Chemistry, Technical University of Darmstadt, Darmstadt D-64287, Germany DOI: 10.1070/MC2000v010n02ABEH001258 A tandem of SN H and SN ipso reactions leading to imidazo[4,5-e]-1,2,4-trazines (6-azapurines) has been observed in the interaction of 5-methoxy-1,2,4-triazines with ureas in the presence of acylating agents.Synthetic routes to azapurines (imidazo[4,5-e]-1,2,4-triazines) described in the literature suggest multistage syntheses from uracil derivatives1,2 or involve rearrangements of 7-azalumazines. 3,4 Another approach of considerable current use is based on condensation of 5,6-diamino-1,2,4-triazines with C1-synthons. 5–7 At the same time, the high electrophilicity of a triazine ring and its tendency to ortho-cyclization reactions,8,9 suggest the possibility for direct annelation of the imidazole ring by interactions of triazines with 1,3-N,N'-binucleophiles due to a nucleophilic attack on unsubstituted carbon atoms of the triazine ring.We have developed a very simple one-step procedure for the synthesis of azapurines from easily accessible 5-methoxy- 3-phenyl-1,2,4-triazine10 and ureas (Scheme 2). These transformations belong to interactions of p-deficient heteroaromatic compounds bearing good leaving groups X (in this case, it is the methoxy group) with nucleophiles.In addition to the replacement of the group X,11 which is a common phenomenon in this case, nucleophilic substitution of hydrogen (SN H) can be reached under appropriate conditions.12 We found a tandem of SN H and SN ipso reactions leading to azapurines by studying the interaction of 5-methoxy-1,2,4-triazines with ureas in the presence of acylating agents.The presence of an acylating agent is a decisive factor. Thus, 5-methoxy-3-phenyl-1,2,4-triazine reacts with ureas in the presence of acetic anhydride to afford imidazo[4,3-e]-1,2,4-triazine (6-azapurine) derivatives.This reaction involves a number of successive steps and, depending on the reaction conditions and the nature of reagents and acylating agents, it affords either open-chain adducts or cyclization products. Note that the nucleophilic attack takes place first at the unsubstituted 6-position, although the replacement of a methoxy group in the 5-position can be expected in accordance with the classic theory of SN i pso reactions and quantum-chemical calculations.Thus, the reaction of urea with 5-methoxy-3-phenyl-1,2,4-triazine in acetic anhydride under mild conditions (20 °C) affords compound 2 (Scheme 1).† Unlike the reaction with urea, the interaction of N,N'-dimethylurea with triazine 1 at 20 °C did not stop at the stage of mono-adduct formation, and the only product is cyclic com- † 1H NMR spectra were measured on Bruker WM-300 and ARX-300 spectrometers, TMS was used as a standard. 1-Acetyl-5-methoxy-3-phenyl-6-(ureido)-1,6-dihydro-1,2,4-triazine 2. A mixture of 5-methoxy-3-phenyl-1,2,4-triazine (50 mg, 0.27 mmol) and urea (16 mg, 0.27 mmol) in acetic anhydride (0.7 ml) was stirred for 48 h at 20 °C. The precipitate formed was filtered off and washed with hot methanol to yield 51 mg (65%) of 2; mp 210–212 °C. 1H NMR ([2H6]DMSO) d: 2.33 (s, 3H, Ac), 3.99 (s, 3H, OMe), 5.60 (br. s, 2H, NH2), 6.39 (d, 1H, CH, J 8.5 Hz), 7.22 (d, 1H, NH, J 8.5 Hz), 7.45–7.48 (m, 3H, Ph), 8.09–8.12 (m, 2H, Ph). Found (%): C, 53.58; H, 5.27; N, 24.25. Calc. for C13H15N5O3 (%): C, 53.97; H, 5.23; N, 24.21. pound 3 isolated as crystals.‡ This is a rare case of isolation of a crystalline sipso-adduct.Upon refluxing a solution of compound 3 in CHCl3 for a long time, a molecule of methanol was eliminated to afford dihydrotriazine 4 in 85% yield.§ The reaction time can be shortened using acetic anhydride at a higher temperature. Thus, heating a mixture of equimolar amounts of 1,2,4-triazine 1 and dimethylurea in acetic anhydride at 70 °C gives dihydrotriazine 4 in 45% yield without isolation of any intermediates.‡ 1-Acetyl-5,7-dimethyl-4a-methoxy-3-phenyl-4,4a,5,6,7,7a-hexahydro- 1H-imidazo[4,5-e]-1,2,4-triazin-6-one 3. A solution of 5-methoxy-3- phenyl-1,2,4-triazine (50 mg, 0.27 mmol) and N,N'-dimethylurea (24 mg, 0.27 mmol) in acetic anhydride (1 ml) was stirred for 48 h at 20 °C. After evaporation of the solvent in vacuo, the oil residue was dissolved in diethyl ether (1 ml).The precipitated colourless crystals were filtered off to yield 55 mg (64%) of compound 3; mp 110–112 °C (decomp.). 1H NMR ([2H6]DMSO) d: 2.34 (s, 3H, Ac), 2.59 (s, 3H, N–Me), 2.78 (s, 3H, N–Me), 3.27 (s, 3H, OMe), 6.12 (s, 1H, CH), 7.45–7.55 (m, 3H, Ph), 7.80–7.83 (m, 2H, Ph). Found (%): C, 56.62; H, 5.99; N, 22.13.Calc. for C15H19N5O3 (%): C, 56.77; H, 6.03; N, 22.07. § 1-Acetyl-5,7-dimethyl-3-phenyl-5,6,7,7a-tetrahydro-1H-imidazo[4,5-e]- 1,2,4-triazin-6-one 4. Method 1: A solution of 3 (100 mg, 0.31 mmol) in CHCl3 (2 ml) was refluxed for 48 h. After evaporation of the solvent in vacuo, the residue was dissolved in CH2Cl2 and chromatographed on silica gel. Yield 77 mg (87%); mp 141–143 °C. 1H NMR ([2H6]DMSO) d: 2.43 (s, 3H, Ac), 3.26 (s, 3H, N–Me), 3.32 (s, 3H, N–Me), 4.72 (s, 1H, CH), 7.40–7.44 (m, 3H, Ph), 8.07–8.11 (m, 2H, Ph). Found (%): C, 59.05; H, 5.26; N, 24.41. Calc. for C14H15N5O2 (%): C, 58.94; H, 5.30; N, 24.55. Method 2: A mixture of 5-methoxy-3-phenyl-1,2,4-triazine (200 mg, 1.08 mmol), N,N'-dimethylurea (95 mg, 1.08 mmol) and acetic anhydride (3 ml) was stirred for 1.5 h at 85–90 °C.The solvent was removed in vacuo. The colourless crystals were washed with diethyl ether to yield 140 mg (45%) of compound 4; mp 144–146 °C. N N N Ph OMe N N N Ph OMe Me O NH NH2 O Ac2O H2N NH2 O Scheme 1 1 2 N N N PhOMe N N N PhN N O Me Me Me O MeO H N N N PhN N Me Me O MeO N N N PhN N R R O F3CO N N N PhN N R R O (RNH)2 CO (CF3 CO)2 O 20 °C (MeNH)2 CO Ac2 O 20 °C Ac2 O 70 °C 1 3 4 5a,b a R = H b R = Me Ac2O 110 °C Scheme 2Mendeleev Communications Electronic Version, Issue 2, 2000 (pp. 43–82) Increasing the temperature to 110 °C facilitates the aromatization of dihydrotriazine to imidazo[4,3-e]-1,2,4-triazine 5. Trifluoroacetic anhydride is a stronger activator for the reaction of 1,2,4-triazine with ureas. Moreover, the aromatization of acylated compounds proceeds more easily with trifluoroacetic anhydride.The reaction of 5-methoxy-1,2,4-triazine 1 with ureas under mild conditions afforded aromatic azapurines 5a,b.¶ This work was supported by the Volkswagen Foundation (grant no. 1/68 782). ¶ 3-Phenyl-6,7-dihydro-5H-imidazo[4,5-e]-1,2,4-triazin-6-one 5a. To a solution of 5-methoxy-3-phenyl-1,2,4-triazine (80 mg, 0.43 mmol) in dry dichloromethane (1 ml) trifluoracetic anhydride was added.The mixture was stirred for 15 min at 20 °C. Next, urea (26 mg) was added, and the mixture was stirred for 48 h. The reaction mixture was concentrated at a reduced pressure. The oil residue was washed with diethyl ether. Next, the precipitate was filtered off and washed with diethyl ether to yield 73 mg (80%) of compound 5a; mp 232–233 °C. 1H NMR ([2H6]DMSO) d: 7.33 (br. s, NH), 7.56–7.65 (m, 3H, Ph), 8.00–8.12 (m, 2H, Ph). Found (%): C, 56.09; H, 3.34; N, 32.71. Calc. for C10H7N5O (%): C, 56.33; H, 3.31; N, 32.85. 5,7-Dimethyl-3-phenyl-6,7-dihydro-5H-imidazo[4,5-e]-1,2,4-triazin-6- one 5b. Method 1: A solution of 5-methoxy-3-phenyl-1,2,4-triazine (50 mg, 0.27 mmol) and N,N'-dimethylurea (24 mg, 0.27 mmol) in acetic anhydride (1.5 ml) was heated at 110 °C for 6 h with stirrting.The solvent was evaporated in vacuo, and the colourless precipitate was filtered off and washed with diethyl ether to give 31 mg (48%) of compound 5b; mp 198–200 °C. 1H NMR (CDCl3) d: 3.47 (s, 3H, NMe), 3.53 (s, 3H, NMe), 7.40–7.46 (m, 3H, Ph), 8.34–8.40 (m, 2H, Ph).Found (%): C, 59.50; H, 4.42; N, 28.88. Calc. for C12H11N5O (%): C, 59.74; H, 4.60; N, 28.88. Method 2: To a solution of 5-methoxy-3-phenyl-1,2,4-triazine (50 mg, 0.27 mmol) in dry CH2Cl2 trifluoracetic anhydride (0.2 ml) was added, and the mixture was stirred for 15 min at 20 °C. N,N'-Dimethylurea (24 mg, 0.27 mmol) was added, and the solution was stirred for 48 h. The reaction mixture was concentrated under a reduced pressure, and the oil residue was filtered off and washed with diethyl ether.Next, the precipitate was filtered off and recrystallised from methanol to give 30 mg (46%) of compound 5b. References 1 T. Kametani, M. Higuchi, M. Noguchi, Y. Hashiguchi and F. Yoneda, Heterocycles, 1980, 14, 1295. 2 F. Yoneda, M. Higuchi and Y. Nitta, J. Heterocycl.Chem., 1980, 17, 869. 3 F. Yoneda, M. Kawamura, T. Nagamatsu, K. Kuretani, A. Hoshi and M. Iigo, Heterocycles, 1976, 4, 1503. 4 S. V. Shorshnev, S. E. Esipov, N. I. Yakushkina, N. A. Klyuev, V. G. Zhil’nikov and A. I. Chernyshev, Khim. Geterotsikl. Soedin., 1987, 1252 [Chem. Heterocycl. Compd. (Engl. Transl.), 1987, 1002]. 5 K. Kaji and M. Kawase, Chem. Pharm. Bull., 1976, 24, 2274. 6 C.-C. Tzeng, D.-C. Wei, L.-C. Hwang, M.-C. Cheng and Yu. Wang, J. Chem. Soc., Perkin Trans. 1, 1994, 2253. 7 J. Riand, M.-T. Chenon, C.-C. Tzeng and R. P. Panzica, J. Chem. Soc., Perkin Trans. 2, 1986, 931. 8 V. L. Rusinov and O. N. Chupakhin, Zh. Org. Khim., 1998, 34, 327 (Russ. J. Org. Chem., 1998, 34, 297). 9 V. N. Charushin, O. N. Chupakhin and H. C. van der Plas, Adv. Heterocycl. Chem., 1988, 43, 301. 10 J. Daunis, R. Jacquir and C. Piegiere, Tetrahedron, 1974, 30, 3171. 11 J. Miller, Aromatic Nucleophilic Substitution, Elsevier, Amsterdam, 1986. 12 O. N. Chupakhin, V. N. Charushin and H. C. van der Plas, Nucleophilic Aromatic Substitution of Hydrogen, Academic Press, New York, 1994. Received: 28th December 1999; Com. 99/1584

ISSN:0959-9436     

出版商:RSC    

年代:2000

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 2, 2000 (pp. 43–82) Hydrogenolysis of C–F bonds in fluorinated aromatic hydrocarbons catalysed by nickel complexes Nicolai Yu. Adonin and Vladimir F. Starichenko* N. N. Vorozhtsov Novosibirsk Institute of Organic Chemistry, Siberian Branch of the Russian Academy of Sciences, 630090 Novosibirsk, Russian Federation. Fax: +7 3832 34 4752; e-mail: adonin@nioch.nsc.ru, vstar@nioch.nsc.ru DOI: 10.1070/MC2000v010n02ABEH001171 Reactions of hexafluorobenzene, pentafluorobenzene, octafluoronaphthalene and pentafluoropyridine with the NiCl2–2,2'-bipyridine (or 1,10-phenanthroline)–Zn reductive catalytic system in DMF (or DMA) in the presence of water or ammonium chloride resulted in the products of C–F bond hydrogenolysis.The activation of C–F bonds by transition metal complexes is an urgent problem.1,2 However, only a few of examples of the catalytic activation of C–F bonds are known. Aizenberg and Milstein3 reported on the hydrodefluorination of hexafluorobenzene 1 and pentafluorobenzene 2 under catalysis with trimethylphosphine complexes of rhodium.Kiplinger and Richmond4 found that the interaction of fluorinated aromatic compounds with the Cp2MX2–HgCl2–Mg and Cp2MX2–PMe3–Mg reductive systems (M = Ti or Zr; X = Cl or F) results in the hydrogenolysis of aromatic C–F bonds.Deacon et al.5 suggested a procedure for the synthesis of 2,3,4,5-tetrafluorobenzoic acid by defluorination of pentafluorobenzoic acid in the presence of an YbCp2(dme) catalyst. We found that the hydrogenolysis of aromatic C–F bonds in polyfluoroaromatic compounds easily proceeds under the action of the NiCl2–2,2'-bipyridine (bipy) [or 1,10-phenanthroline (phen)]–Zn reductive system in DMF or DMA in the presence of H2O or NH4Cl.† No reaction was observed in the absence of H2O or NH4Cl.The same result was achieved without the NiCl2– bipy (or phen) catalytic complex. The reduction of hexafluorobenzene 1 for 3 h resulted in the complete conversion into pentafluorobenzene 2, 1,2,3,4- and 1,2,4,5-tetrafluorobenzenes 3 and 4, 1,2,3- and 1,2,4-trifluorobenzenes 5 and 6, which were isolated in 19, 52, 14, 5 and 2% yields, respectively.Note that the yield of 1,2,3,4-tetrafluorobenzene 3 was higher than that of 1,2,4,5-tetrafluorobenzene 4 by a factor of about † General procedure: NiCl2·6H2O (0.12 g, 0.5 mmol), 0.08–0.16 g (0.5– 1 mmol) of bipy [or 0.1–0.2 g (0.5–1 mmol) phen] and zinc dust (6.5 g, 100 mmol) were placed in a flask.Next, DMF (or DMA) (10 ml) and H2O (2 ml) (or 1 g of ammonium chloride) were added, and the mixture was heated at 70 °C for 30 min with stirring. A substrate (10 mmol) solution in DMF or DMA (5 ml) was added dropwise, and the resulting reaction mixture was stirred for several hours (see the text).The reaction was monitored by reversed-phase HPLC (eluent: aqueous acetonitrile; UV detection at 230 and 254 nm). Next, the solution was filtered and diluted with water, and the products were stream distilled. The products were analysed by 19F NMR spectroscopy and GC–MS. The identification of compounds was made by comparison of the retention times and mass spectra with those for standard samples.four, and 1,2,3,5-tetrafluorobenzene 7 was absent. The replacement of bipy with phen in the catalytic complex exerted almost no effect on the reaction. The use of DMF as a solvent, resulted in a decrease in the conversion of 1 (to 62% in 5 h). In this case, pentafluorobenzene 2 was the main product of the reaction, and compounds 3 and 4 were formed in minor amounts.In the presence of NH4Cl as a source of protons, the effective amount of a reducing agent (zinc) was 6 equiv. per 1 equiv. of the substrate, whereas about 10 equiv. of zinc per 1 equiv. of the substrate were required with the use of water. The latter reduction system was as good as that with ammonium chloride, and reproducibility of the results was better.For this reason, the reduction of other polyfluoroaromatic compounds was studied with the use of the NiCl2–bipy–Zn reduction system in aqueous DMF. The reaction of 2 with the NiCl2–bipy–Zn reduction system at 100% conversion (4 h) gave a mixture of fluorinated benzenes 3–6 in 45, 11, 18 and 15% yields, respectively.In addition, 1,2- and 1,4-difluorobenzenes 8 and 9 (ca. 1%) were formed. In this case, the ratio between isomers 3 and 4 was practically the same as in the case of hexafluorobenzene; 1,2,3,5-tetrafluorobenzene 7 was also not detected. For comparison, note that the reduction of 2 with lithium aluminium hydride gave 1,2,4,5-tetrafluorobenzene 4 (92%) with impurities of 1,2,3,4-tetrafluorobenzene 3 (7%) and 1,2,3,5-tetrafluorobenzene 7 (1%).6 The reactions of 1 and 2 with titanium, zirconium4 and rhodium3 complexes resulted only in compound 4.Of three possible isomers, only the formation of 1,2,4,5-tetrafluorobenzene 4 was observed in the reaction between pentafluorobenzene and tetrakis(triethylphosphine) nickel(0).7 Note that the catalytic activity of nickel complexes with phen is almost the same as that of NiCl2–bipy complexes.The reduction of octafluoronaphthalene 10 by the NiCl2– bipy–Zn system in aqueous DMF at 60 °C afforded polyfluorinated naphthalenes 11–13. At low conversion (10–15%) of substrate 10, the main product was 1,2,3,4,5,6,8-heptafluoronaphthalene 11. At 70% conversion of 10 (2 h), in addition to 11 (70%), 1,2,4,5,6,8-hexafluoronaphthalene 12 (25%), 1,2,3,4,5,6,7-heptafluoronaphthalene 13 F F F F F F i F F F F F F F F F F F F F F F F F F F 1 2 3 6 5 4 Reagents and conditions: i, NiCl2, bipy, Zn, DMA, NH4Cl, 70 °C, 3 h.Reagents and conditions: i, NiCl2, bipy, Zn, DMF, H2O, 50–60 °C. F F F F F F F F F F F F F F F F F F F F F F F F F F F F i 10 11 12 13Mendeleev Communications Electronic Version, Issue 2, 2000 (pp. 43–82) (ca. 5%) and unidentified fluoronaphthalenes (ca. less 1%) were formed. The further reaction led to the complete conversion of 10 into a complex mixture of fluorine-containing naphthalenes. Note that the interaction of octafluoronaphthalene 10 with lowvalence titanium and zirconium4 complexes and with zinc in aqueous ammonia8 resulted only in compounds 11 and 12. The interaction of pentafluoropyridine 14 with NiCl2–bipy–Zn gave 2,3,5,6-tetrafluoropyridine 15, 2,3,5-trifluoropyridine 16 and 3,5-difluoropyridine 17.Compound 14 was completely consumed within 1 h to give compounds 15–17 in 45, 27 and 18% yields, respectively. In contrast to the reaction between compound 14 and lithium aluminium hydride with deep hydrodefluorination,9 in our case, fluorine atoms were not replaced by hydrogen in 3,5-difluoropyridine 17.Note that the reduction of pentafluoropyridine with a zinc–copper couple (aqueous DMF, 70 °C, 10 h) gave only tetrafluoropyridine 15.10 Thus, hexafluorobenzene, pentafluorobenzene, octafluoronaphthalene and pentafluoropyridine react with a catalytic system in situ generated from nickel chloride, 2,2'-bipyridyl (or 1,10- phenanthroline) and zinc in DMF (or DMA) in the presence of water or ammonium chloride to form hydrodefluorination products resulting from the hydrogenolysis of aromatic C–F bonds.This work was supported by the Russian Scientific Centre ‘Applied Chemistry’. References 1 J. L. Kiplinger, T. G. Richmond and C. E. Osterberg, Chem. Rev., 1994, 94, 373. 2 J. Burdeniuc, B. Jedlicka and R. H. Crabtree, Chem. Ber./Recl., 1997, 130, 145. 3 (a) M. Aizenberg and D. Milstein, Science, 1994, 265, 559; (b) M. Aizenberg and D. Milstein, J. Am. Chem. Soc., 1995, 117, 8674. 4 (a) J. L. Kiplinger and T. G. Richmond, J. Chem. Soc., Chem. Commun., 1996, 1115; (b) J. L. Kiplinger and T. G. Richmond, J. Am. Chem. Soc., 1996, 118, 1805. 5 G. B. Deacon, C. M. Forsyth and J. Sun, Tetrahedron Lett., 1994, 35, 1095. 6 (a) G. Brooke, J. Burdon and J. Tatlow, J. Chem. Soc., 1962, 3253; (b) L. Wall, W. Pummer, J. Fearn and J. Antonucci, J. Res. NBS, 1963, 67A, 481. 7 L. Cronin, C. L. Higgitt, R. Karch and R. N. Perutz, Organometallics, 1997, 16, 4920. 8 S. S. Laev and V. D. Shteingarts, J. Fluorine Chem., 1998, 91, 21. 9 R. D. Chambers, C. W. Hall, J. Hutchinson and R. W. Millar, J. Chem. Soc., Perkin Trans. 1, 1998, 1705. 10 V. E. Platonov and V. I. Krasnov, Zh. Org. Khim., 1994, 30, 1271 (Russ. J. Org. Chem., 1994, 30, 1336). N F F F F F N F F F F N F F F N F F i 14 15 16 17 Reagents and conditions: i, NiCl2, bipy, Zn, DMF, H2O, 60 °C. Received: 8th June 1999; Com. 99/1497

ISSN:0959-9436     

出版商:RSC    

年代:2000

数据来源: RSC

摘要:

Mendeleev Communications Electronic Version, Issue 2, 2000 (pp. 43–82) Synthesis and electrochemical properties of the N-isocyanurate derivative of azahomo[60]fullerene Oleg G. Sinyashin,* Irina P. Romanova, Gulshat G. Yusupova, Adilya A. Nafikova, Valery I. Kovalenko, Nail M. Azancheev, Vitaly V. Yanilkin and Yuliya G. Budnikova A. E. Arbuzov Institute of Organic and Physical Chemistry, Kazan Scientific Centre of the Russian Academy of Sciences, 420088 Kazan, Russian Federation.Fax: +7 8432 75 22 53; e-mail:oleg@iopc.kcn.ru DOI: 10.1070/MC2000v010n02ABEH001159 The cyclic voltammogram of the title compound exhibits three reversible reduction peaks which are less negative than that of [60]fullerene. The development of new fullerene materials such as semiconductors, superconductors, ferromagnetics, optoelectronic materials and artificial photosynthesis systems are based on the electron- accepting properties of fullerenes.1 In general, functionalization of [60]fullerene with a large number of substituents leads to a decrease in the electron affinity of the fullerene moiety.As a result, [60]fullerene derivatives are reduced electrochemically at more negative potentials than the parent C60.2 This problem has been successfully solved by attaching electron-withdrawing groups or ‘periconjugation’ fragments.3–7 To increase the electron affinity of the fullerene derivatives, we proposed to add an isocyanuric acid moiety, which contains three electron-withdrawing carbonyl groups, to C60.We describe here the synthesis and electrochemical properties of the first example of isocyanurate-containing fullerene, 5-[5'-(azahomofullereno)pentyl]-1,3-diallyl-1,3,5-triazine-2,4,6- (1H,3H,5H)-trione 3, obtained by the [3 + 2]-cycloaddition of 5-[5'-(azidopentyl)]-1,3-diallyl-1,3,5-triazine-2,4,6-(1H,3H,5H)- trione 1 to C60.Compound 1 was prepared by the reaction of 5-(5'-bromopentyl)- 1,3-diallyl-1,3,5-triazine-2,4,6-(1H,3H,5H)-trione with sodium azide in boiling dry acetone (8 h).The product was isolated as light yellow viscous liquid by column chromatography (75% yield). The structure of azide 1 was studied by 1H NMR, 13C NMR and IR spectroscopy, and the composition was determined by elemental analysis.† The identification of 1H and 13C NMR signals of methylene groups of the pentyl fragment in azide 1 was performed by comparing the experimental and calculated chemical shifts, taking into account the influence of substituents in alkanes.The chemical shifts of the 13C signals were evaluated by the empirical Grant–Pole equation,8 and the d values of CH2(13) and CH2(17) proton signals were calculated according to the Shuleri equation.9 The CH2(14) and CH2(16) proton signals were identified by an analysis of the selective double resonance of CH2(14)-{CH2(13)}, CH2(13)- {CH2(14)}, CH2(16)-{CH2(17)} and CH2(17)-{CH2(16)}.The reaction of [60]fullerene (0.1 mmol) with azide 1 (0.3 mmol) was carried out in boiling dry o-dichlorobenzene under a nitrogen atmosphere. After 4 h, the solvent was removed by distillation, and the reaction mixture was separated by column chromatography on silica gel using toluene as an eluent.Unreacted fullerene (8% in terms of the initial amount) and products 2 (1–2%), † 1: 1H NMR (250 MHz, CDCl3) d: 1.48 [m, 2H, CH2(15)], 1.68 [m, 2H, CH2(14)], 1.90 [m, 2H, CH2(16)], 3.43 [m, 2H, CH2(17), AA'XX' system, 3JHH 7.4 Hz], 3.91 [m, 2H, CH2(13), AA'XX' system, 3JHH 7.4 Hz], 4.46 [d, 4H, CH2(7), CH2(10), 3JHH 6.0 Hz], 5.23 [d, 2H, =CHcis(9), =CHcis(12)], 5.26 [d, 2H, =CHcis(9), =CHcis(12)], 5.85 [ddt, 2H, CH(8), CH(11), 3J cis HH 9.0 Hz, 3JHH trans 18.0 Hz]. 13C NMR (100 MHz, CDCl3) d: 24.78 [tm, C(15), 1JCH 125.6 Hz], 26.48 [tm, C(14), 1JCH 124.5 Hz], 31.73 [tm, C(16), 1JCH 130.4 Hz], 42.28 [tm, C(13), 1JCH 142.3 Hz, 2JCH 4.6 Hz, 3JCH 8.5 Hz], 44.46 [tm, C(7), C(10), 1JCH 143.6 Hz, 2JCH 5.2 Hz, 3JCH cis 7.9 Hz, 3JCH trans 13.3 Hz], 50.80 [tm, C(17), 1JCH 141.7 Hz], 118.41 [tm, C(9), C(12), 1JCH 157.5 Hz], 130.71 [dm, C(8), C(11), 1JCH 159.60 Hz], 148.04 [m, C(4), 3JCH 3.6 Hz], 148.28 [m, C(2), C(6), 3JCH 3.6 Hz].IR (KBr, n/cm–1): 1691, 766 (C=O), 1645 (C=C), 933, 993 (=CH), 2943, 2852, 1460 (CH), 2098 (N3). Found (%): C, 52.38; H, 6.72; N, 26.20. Calc.for C14H20N6O3 (%): C, 52.50; H, 6.25; N, 26.25. 3 (20%) and 4 (5%) were obtained. The removal of the eluent in a vacuum resulted in compounds 2–4 as dark brown viscous liquids. To obtain the products as powders, they were dissolved in diethyl ether and then dried in a vacuum for 5 h. The structure of compound 3 was examined by 1H NMR, 13C NMR, IR and UV spectroscopy, and the composition was determined by elemental analysis.According to the elemental analysis data, 3 is a monoadduct of C60 and azide 1.‡ The 13C NMR spectrum of compound 3 shows that the addition of azide 1 to C60 takes place at the 6,5-bond. This spectrum contains 32 signals in the region between d 133 and 148 ppm, which is typical of the sp2 carbon signals of fullerene derivatives. The intensities of four signals correspond to one carbon atom, and the intensities of 28 signals correspond to two carbon atoms.In accordance with the proposed structure, no sp3 carbon signals of the fullerene fragment were observed.10 This 13C NMR ‡ 3: 1H NMR (250 MHz, CDCl3) d: 1.47 [m, 2H, CH2(15)], 1.67 [m, 2H, CH2(14)], 1.89 [m, 2H, CH2(16)], 3.40 [m, 2H, CH2(17), AA'XX' system, 3JHH 7.4 Hz], 3.88 [m, 2H, CH2(13), AA'XX' system, 3JHH 7.4 Hz], 4.46 [d, 4H, CH2(7), CH2(10), 3JHH 6.0 Hz], 5.23 [d, 2H, =CHcis(9), =CHcis(12)], 5.27 [d, 2H, =CHcis(9), =CHcis(12)], 5.82 [ddt, 2H, CH(8), CH(11), 3J cis HH 9.0 Hz, 3JHH trans 18.0 Hz]. 13C NMR (100 MHz, CDCl3) d: 21.35 (MePh), 22.60 [C(15)], 24.42 [C(14)], 28.90 [C(16)], 42.31 [C(13)], 44.90 [C(7), C(10)], 51.30 [C(17)], 119.00 [C(9), C(12)], 125.21, 128.13 and 128.94 (MePh), 130.91 [C(8), C(11)], 133.69, 135.78, 136.18 (1C), 137.11, 137.29 (1C), 137.76, 137.79, 138.00, 138.45, 139.17, 140.70, 141.37, 142.57, 142.65, 142.74, 142.84, 143.05, 143.33, 143.48 (1C), 143.59, 143.78, 144.05, 144.09, 144.22, 144.26, 144.39, 144.49, 144.67, 144.97, 145.09 (1C), 146.89, 147.74 (1C), 148.41 [C(4)], 148.67 [C(2), C(6)]. UV–VIS [CH2Cl2, lmax/nm (lg e/dm3 mol–1 cm–1)]: 261 (5.62), 330 (5.04), 403 (4.23), 430 (4.02), 546 (3.63).IR (KBr, n/cm–1): 1691, 763 (C=O), 1645 (C=C), 937, 990 (=CH), 2920, 2852, 1453 (CH). 526, 570, 1176 (C60). Found (%): C, 87.30; H, 2.10; N, 5.48. Calc. for C74H20N4O3·(C6H5Me)0.2 (%): C, 87.81; H, 2.09; N, 5.43. N N N All All (CH2)5Br N N N CHCH2 CH2CH H2C CH2 O O O O O O N3 NaN3 – NaBr 1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 N N N CHCH2 CH2CH H2C CH2 O O O N C60 2 + 3 + 4 All = CH2CH=CH2Mendeleev Communications Electronic Version, Issue 2, 2000 (pp. 43–82) spectrum is consistent with the CS symmetry of the molecule with an open annulene structure (the structure of azahomo[60]- fullerene).10,11 In addition, the 13C NMR spectrum of 3 showed signals of two equivalent allyl groups, five methylene and three carbonyl groups, two of which are equivalent.The UV–VIS spectroscopy data also confirm the azahomofullerene structure of 3.‡ The low-intensity broad absorption bands in the region 350–700 nm are assigned to the open annulene structure.10 Parent azide 1 does not exhibit absorption bands in this region. The signals of protons of two equivalent allyl substituents at the isocyanurate ring and of five methylene groups were observed in the 1H NMR spectrum of 3.Note that the chemical shifts of the signals due to methylene protons at the nitrogen atom of 3 do not differ from the chemical shifts of protons of this group in the parent azide, although it is well know that [60]fullerene shifts the signals of protons of neighbouring groups to the weak field region.12 The IR spectrum of compound 3 exhibits bands of both isocyanurate and fullerene fragments and no absorption bands in the azide region at 2100 cm–1.‡ Thus, all structure data indicate that the [3 + 2]-cycloaddition of 5-[5'-(azidopentyl)]-1,3-diallyl-1,3,5-triazine-2,4,6-(1H,3H,5H)- trione 1 to [60]fullerene results in the formation of a derivative having an open annulene structure.Elemental analysis§ indicates that compound 4 is a diadduct of azide 1 and [60]fullerene. Cyclic voltammograms of azahomo[60]fullerene 3 exhibited three reversible reduction peaks regardless of the scanning rate (20–250 mV). Each of these peaks corresponds to the transfer of one electron onto the fullerene.Note that azide 1 was not reduced under analogous conditions, and C60 exhibited three reversible reduction peaks (Table 1, Figure 1). It is of importance that all of the reduction potentials of compound 3 are less negative than the corresponding potential of [60]fullerene. This unusual behaviour of azahomo[60]fullerene 3 arises from the strong electron-withdrawing influence of isocyanurate on the fullerene moiety through the methylene chain or though the intramolecular interaction of isocyanurate carbonyl groups and the fullerene moiety.Thus, compound 3 is the first example of a [60]fullerene monoadduct which is reversibly reduced at less negative potentials than C60.5–7 § 4: IR (KBr, n/cm–1): 1691, 766 (C=O), 1645 (C=C), 933, 993 (=CH), 2941, 2861, 1460 (CH).Found (%): C, 80.49; H, 3.42; N, 7.30. Calc. for C88H40N8O6 (%): C, 80.98; H, 3.07; N, 7.36. We are grateful to S. G. Fattakhov for the synthesis of 1-(5'- bromopentyl)-3,5-diallyl-1,3,5-triazine-2,4,6-(1H,3H,5H)-trione. This work was supported by the Russian Foundation for Basic Research (grant no. 99-03-33001) and by the Ministry of Science of the Russian Federation (grant no. 99031). References 1 D. V. Konarev and R. N. Lyubovskaya, Usp. Khim., 1999, 68, 23 (Russ. Chem. Rev., 1999, 68, 19). 2 A. Hirsch, The Chemistry of the Fullerenes, Thieme, Stuttgart, 1994, p. 80. 3 F. Zhou, G. J. van Berked and B. T. Donovan. J. Am. Chem. Soc., 1994, 116, 5485. 4 M. Keshavarz-K, B. Knight, G. Srdanov and F. Wudl, J. Am. Chem. Soc., 1995, 117, 11371. 5 M. Keshavarz-K, B.Knight, R. C. Haddon and F. Wudl, Tetrahedron, 1996, 52, 5149. 6 T. Suzuki, Y. Maruyama, T. Akasaka, W. Ando, K. Kobayashi and S. Nagase, J. Am. Chem. Soc., 1994, 116, 1359. 7 T. Ohno, N. Martin, B. Knight, F. Wudl, T. Suzuki and H. Yu, J. Org. Chem., 1996, 61, 1306. 8 E. Breitmaier and W. Voelter, Carbon-13 NMR Spectroscopy High- Resolution Methods and Applications in Organic Chemistry and Biochemistry, VCH, Weinheim, 1987, p. 513. 9 B. I. Ionin, B. A. Ershov and A. I. Kol’tsov, YaMR spektroskopiya v organicheskoi khimii (NMR Spectroscopy in Organic Chemistry), Khimiya, Leningrad, 1983, p. 140 (in Russian). 10 A. B. Smith and H. Tokuyama, Tetrahedron, 1996, 52, 5257. 11 M. Prato, Q. C. Li, F. Wudl and V. Luchini, J. Am. Chem. Soc., 1993, 115, 1148. 12 C. J. Hawker, P. M. Saviller and J. W. White, J. Org. Chem., 1994, 59, 3503. aSolution concentrations, 1×10–3 mol dm–3; supporting electrolyte, 0.1 M Bu4NBF4; scanning rate, 50 mV s–1, working electrode, Pt; reference electrode, Ag/AgNO3 in 0.01 M MeCN. Table 1 Reduction of C60 and compound 3 [E1/2 in a mixture of o-dichlorobenzene and MeCN (3:1) at 25 °C].a Compound E1 red /V E2 red /V E3 red /V C60 –0.90 –1.25 –1.77 3 –0.82 –1.14 –1.68 I 1 mA 1 2 0.4 0.8 1.2 1.6 –E/eV 1.8 Figure 1 Cyclic voltammograms of (1) C60 and (2) azahomo[60]fullerene 3. Received: 7th May 1999; Com. 99/1486

ISSN:0959-9436     

出版商:RSC    

年代:2000

数据来源: RSC