摘要:

Russian Chemical Reviews 69 (9) 727 ± 746 (2000) Structural and dynamic properties of tetrahydroborate complexes V D Makhaev Contents I. Introduction II. General characteristics of tetrahydroborate complexes III. Transition metal complexes with terminal BH4 ligands IV. Possible reasons for the differences in the structural and dynamic properties of tetrahydroborate complexes V. Complexes of post-transition metals with terminal BH4 ligands VI. Complexes of Group IA and IIA metals and aluminium with terminal BH4 ligands VII. Tetrahydroborate complexes with bridging BH4 ligands VIII. Conclusion Abstract. complexes tetrahydroborate of structures the on Data Data on the structures of tetrahydroborate complexes from electron and neutron analysis, diffraction X-ray from X-ray diffraction analysis, neutron and electron diffraction, diffraction, and that, shown is It considered.are spectroscopy NMR and IR and IR andNMRspectroscopy are considered. It is shown that, in in addition to the diversity of stabilising ligands, complexes of addition to the diversity of stabilising ligands, complexes of Group III and IV transition metals are characterised by diverse Group III and IV transition metals are characterised by diverse types bridging r o terminal of coordination of types of coordination of terminal (Z1, Z2 and Z3) or bridging (between neutral molecular units) tetrahydroborate groups. Com- (between neutral molecular units) tetrahydroborate groups. Com- pounds with BH found be can denticities different with pounds with BH4 groups groups with different denticities can be found among these complexes.Most tetrahydroborates of these metals among these complexes. Most tetrahydroborates of these metals in the oxidation states +3, +4 comply with the principle of in the oxidation states +3, +4 comply with the principle of maximum occupancy of the coordination sphere. Tetrahydrobo- maximum occupancy of the coordination sphere. Tetrahydrobo- rates of Group III and IV metals exhibit clear-cut stereochemical rates of Group III and IV metals exhibit clear-cut stereochemical non-rigidity are the of protons the (fluxionality): non-rigidity (fluxionality): the protons of the BH BH4 ligand ligand are equivalent on the NMR time scale. Group I and V± VIII equivalent on the NMR time scale.Group I and V± VIII transition stereo- are which mainly metals transition metals mainly form form Z2-complexes, -complexes, which are stereo- chemically much more rigid than complexes of Group III and IV chemically much more rigid than complexes of Group III and IV metals. It is suggested that electronic configuration of the central metals. It is suggested that electronic configuration of the central atom is the crucial factor determining different structures and atom is the crucial factor determining different structures and stereochemical tetrahydroborate the of behaviours stereochemical behaviours of the tetrahydroborate complexes. complexes. Data post-transition of properties and structures the on Data on the structures and properties of post-transition and and s-element this with consistent also are tetrahydroborates s-element tetrahydroborates are also consistent with this assump- assump- tion. references 248 includes bibliography The tion.The bibliography includes 248 references. I. Introduction Coordination compounds containing the simplest hydroborate anion BH¡4 as a ligand possess many practically valuable proper- ties. Tetrahydroborate complexes are used as selective reducing agents, starting compounds in the syntheses of complex and organometallic derivatives, precursors for the production of borides, hydrides and other inorganic materials, and as catalysts of hydrogenation, isomerisation, oligomerisation, polymerisa- tion, etc. In addition, these complexes exhibit unusual structural and dynamic properties such as the ability to form metal ± hydro- gen ± boron multicentre bonds and exceptionally low barriers to V D Makhaev Institute of Problems of Chemical Physics, Russian Academy of Sciences, 142432 Chernogolovka, Moscow Region, Russian Federation.Fax (7-096) 576 40 09. E-mail: vim@icp.ac.ru Received 9 March 2000 Uspekhi Khimii 69 (9) 796 ± 816 (2000); translated by Z P Bobkova #2000 Russian Academy of Sciences and Turpion Ltd DOI 10.1070/RC2000v069n09ABEH000580 727 728 737 739 740 741 741 742 the intramolecular exchange of bridging and terminal hydrogen atoms in the tetrahydroborate ligand. In mononuclear complexes, a tetrahydroborate ligand either exists as the BH¡4 ion or is linked to the metal through one, two or three M7H7B bridges (Z1, Z2 and Z3 coordination, respec- tively).The forms of coordination of tetrahydroborate ligands in polynuclear complexes are even more diverse. The BH¡4 anion is isoelectronic with a methane molecule. Therefore, it has been suggested that tetrahydroborates can serve as structural models for some steps in the activation of C7H bonds in saturated hydrocarbons. The existence of complexes with Z1-, Z2- and Z3-BH4 ligands has been proven by various structural methods. However, the NMR spectra of many tetrahydroborate complexes point to the equivalence of the four protons in the BH4 group, which indicates that these compounds are fluxional, i.e. bridging and terminal protons of the tetrahydroborate ligands are able to undergo fast (on the NMR time scale) intramolecular exchange.1±5 Thus, most of tetrahydroborate complexes con- stantly exist in the state of fast dynamic transformation, resulting in the change of the mutual orientation of the complex-forming ion and the tetrahydroborate ligand.Numerous attempts have been made to explain the unusual properties of tetrahydroborate complexes by such factors as the influence of the Coulomb interaction between the metal ion and the anion, the tendency of the central atom to fill its outer electron shell, the influence of stabilising ligands and so on. A great deal of information on the structures of tetrahydroborates of light elements was provided by ab initio quantum-chemical calcula- tions.For transition metal complexes, these calculations are quite cumbersome and have now been performed for only a few objects. It is often impossible to determine unambiguously the most favourable configuration of the complex on the basis of calcu- lations because the energies of different types of coordination of BH4 ligands differ only slightly; therefore, the results of calcula- tions need to be compared with experimental data (see several papers 6±10 and references therein). Tetrahydroborate complexes of various metals have been studied to different extents. The structures of tetrahydroborates of Group III and IV transition metals, especially uranium derivatives, and titanium and zirconium compounds have been studied in most detail.Complexes formed by Group I andV± VIII transition metals and by alkali, alkaline earth and post-transition metals are less studied. Therefore, this review starts with a brief characterisation of the physical properties of tetrahydroborate728 complexes and methods for their investigation; the ability of various ligands to stabilise transition metal tetrahydroborates is also considered. This is followed by discussion of the structural and dynamic properties of tetrahydroborates of Group III and IV transition metals with terminal BH4 ligands and comparison of these properties with the properties of Group I and V± VIII transition metal complexes. An attempt is made to interpret the regularities observed. Data on the structure of alkali, alkaline earth and post-transition metal tetrahydroborates and polynu- clear complexes with bridging BH4 groups are used to verify and refine the correlations found.The results of determination of the structures of tetrahydroborate complexes by electron, X-ray and neutron diffraction analyses and by IR spectroscopy as well as the data on the complexes of Group IV ± VIII transition metals, for which the type of coordination of tetrahydroborate ligands has been characterised by NMR spectroscopy, are summarised in tables. Less reliable data on the ligand denticity based only on the results of IR spectroscopy and not confirmed by diffraction methods are cited as little as possible, only where it is necessary to complete the picture.Hundreds of studies dealing with the investigation or applica- tion of tetrahydroborate complexes are published annually. About 200 of these complexes have been structurally character- ised; new experimental results have been obtained for them in recent years. However, the questions of what factors determine the structural (type of coordination) and dynamic (stereochemical rigidity or fluxionality) properties of the BH4 ligands in the complexes and whether or not these properties are related to one another (if they are related, what is the type of relationship) still remain to be answered. In this review, we analyse the results of experimental determi- nation of the type of coordination ofBH4 groups in different types of complexes by electron, neutron and X-ray diffraction analyses and by NMR spectroscopy.The results are used to make an assumption about the main factors determining the structural and dynamic features of tetrahydroborate complexes. Table 1. Structurally characterised complexes with terminal BH4 ligands. Complex Complexes of Group III and IV transition metals Sc(Z2-BH4)(Z3-BH4)2(THF)2 [(Me3Si)2C5H3]2Sc(Z2-BH4) Y(Z2-BH4)(Z3-BH4)2(THF)3 Y(Z3-BH4)3 . 2THF [Y(Z3-BH4)4]7 [Y(Z3-BH4)2(THF)4]+ (MeOCH2CH2C5H4)2Y(Z2-BH4) (EtMe4C5)2Y(Z2-BH4)(THF) La(BH4)3 . 3.5THF [La(Z3-BH4)2(THF)5]+ [La(Z2-BH4)2(Z3-BH4)2(THF)2]7 (MeOCH2CH2C5H4)2Pr(Z3-BH4) (MeOCH2CH2C5H4)2Nd(Z3-BH4) (PriMe4C5)2Sm(Z3-BH4)(THF) (MeOCH2CH2C5H4)2Yb(Z2-BH4) [Me2Si(3-Me3SiC5H3)2]Yb(Z3-BH4)(THF) Yb(Z3-BH4)2(MeCN)4 Distance /A M_B B7Hb M7Hb 2.09(5) m 2.03(4) 2.330(5) (Z3) 1.08(5) m 2.229(5) (Z3) 2.551(5) (Z2) 2.52(2) 2.58 (Z3) 2.68 (Z2) 2.41 m 2.52 m 2.773 2.669(4) 772.26(5), 2.27(5) 2.35 m 2.730(2) 0.94, 1.15, 1.12 2.760(2) (Z3) 0.93, 0.96, 1.49 2.900(2) (Z2) 1.03, 1.13 2.757(18) 2.664(25) 2.624(3) 2.47, 2.59, 2.28 2.43, 2.74, 2.55, 2.40, 2.55 772.36(3), 2.51(3), 2.49(3) 72.25, 2.22, 2.06 2.46, 2.52, 2.33 2.800 2.470 2.666(6) V D Makhaev II.General characteristics of tetrahydroborate complexes Compounds containing tetrahydroborate ligands are known for all Group IA, IIA and IIIA metals and for all transition metals, except for technetium, promethium, platinum, gold, francium, radium and some actinides, mainly of the berkelium subgroup.1 ±5 The results of studies of metal tetrahydroborate complexes by various methods are summarised in Tables 1 ± 3.The formal oxidation numbers of metal atoms in these complexes vary from 0 to +4. Some metals can form the complexes in two different oxidation states [for example, Cu(I) and Cu(II), Yb(II) and Yb(III), U(III) and U(IV), etc.]; for titanium, complexes containing the central atoms in four different oxidation states (0, +2, +3, +4) have been structurally characterised. The ion radii of metal cations (r i) in tetrahydroborates can differ by a factor ofmore than six [for example, for Be(II), ri=0.27A, and for Cs(I), r i=1.67A],222 the surface area of the coordination sphere can differ by a factor of *40 and the volume of the central atom, by a factor of 200.The BH¡4 (r i=2.04A) anion markedly exceeds in size anymetal cation which forms complexes with it.1 Depending on the nature of the central atom, the stabilising ligands and conditions, the com- pounds in question can be solid, liquid, or gaseous. In the crystalline state, tetrahydroborates having ionic, molecular, islet (ion-molecular) and polymeric (oligomeric) structures are known. Diamagnetic and paramagnetic tetrahydroborate complexes, both low- and high-spin ones, for example (TMEDA)Cr(Z2-BH4)2 with mef = 4.8 mB,93 are known. Some tetrahydroborates are stable in air but most of them are unstable and decompose, often with an explosion, in the presence of air or moisture.An important characteristic of tetrahydroborate complexes is the manner in which the BH4 ligand is attached to the central atom. The structures of these complexes are determined most often using X-ray diffraction analysis. In many instances, this method can be used to determine the position of hydrogen atoms and establish the type of ligand coordination. The range of structurally characterised compounds includes complexes with Ref. Vibration frequencies of the BH4 group (IR spectroscopy) /cm71 11, 12 2508, 2420, 2377, 2304, 2242, 2154, 2061, 1348, 1250, 1185, 1124, 1044 1.19(4) 13 14 2415, 2125 7 7 7 12, 15 2482, 2300, 2220, 2165, 1250, 1192, 1103, 1042 7 2419, 2367, 2294, 2226, 2197, 2174 16 17 18 1.11(5), 1.11(5) 7 777 19 19 20 2404, 2254, 2190 2404, 2260, 2184 7 771.18(3), 1.13(3), 1.13(3) 7 7 1.15, 1.20, 1.09 1.19, 1.05, 1.09 19 21 22 2475, 2385 ± 2240, 2195, 2125 2363, 2310, 1604, 1176, 1109Structural and dynamic properties of tetrahydroborate complexes Table 1 (continued).Complex Complexes of Group III and IV transition metals Yb(Z3-BH4)2(Py)4 U(Z2-BH4)(Z3-BH4)2(DMPE)2 U(Z3-BH4)3(Ph2Ppy)2 (C6Me6)U(Z3-BH4)3 [U(Z3-BH4)3(DME)(m-H)]2 [Na(THF)6][Cp*U(Z3-BH4)3]2 [U(Z3-BH4)2(THF)5][(C7H7)U2(Z3-BH4)6] [U(Z3-BH4)2(THF)5]+ [U(Z3-BH4)3(C7H7)U(Z3-BH4)3]7 [[UCl5(Z2-BH4)][U(Z3-BH4)2(DC18C6)]2 [UCl5(Z2-BH4)]27 [U(Z3-BH4)2(DC18C6)]+ U(Z3-BH4)4(THF)2 U(Z3-BH4)2(Z2-BH4)2(Ph3PO)2(C6H6) U(Z3-BH4)3(Z2-BH4)(Ph3PO)2 CpU(Z3-BH4)3 Cp2U(Z3-BH4)2 [1,3-(Me3Si)2C5H3]2U(Z3-BH4)2 Cp2 U(Z3-BH4)2 Cp3U(Z3-BH4) (2,4-Me2C5H5)U(Z3-BH4)3 (C4Me4P)2U(Z3-BH4)2 Ind2U(Z3-BH4)2 U(Z3-BH4)3(OCHPh2)(THF)2 U(Z3-BH4)2(OCHPh2)2(THF)2 (tritox)U(Z3-BH4)3(THF) (tritox)3U(Z3-BH4) (COT)U(Z3-BH4)2(OPPh3) U{PhC[N(SiMe3)2]2}2(Z3-BH4)2 [(Me3SiNCH2CH2)3N]U(Z3-BH4)(THF) [(Me3Si)2N]3Th(Z3-BH4) Np(Z3-BH4)4 Ti(Z3-BH4)3 (see b) Distance /A M7Hb 7777777772.16, 2.39, 2.70 72.0, 2.5, 2.8 2.0, 2.5, 2.8 2.0, 2.5, 2.8 2.1 772.42, 2.36, 2.37 7772.29(20) 772.527(8), 2.483(9), 2.711(6) 2.63(1) 7772.26(3), 2.27(3), 2.31(3) 772.3(5), 2.1(3), 2.1(3) 1.984(5) 729 Ref.Vibration frequencies of the BH4 group (IR spectroscopy) /cm71 M_B B7Hb 22 2.692(4) 7 2390, 2360, 2342, 2337, 2248, 2216, 1948, 1876, 1869 2440, 2360, 2340, 2300, 2200, 1125 23 24 2420, 2300, 2200, 1120 25 2493, 2199, 2122 26 27 2480, 2210, 2140, 1195, 1170, 1095, 1035 7 7 2.84(3) (Z2) 7 2.68(4) (Z3) 2.61(2) 7 2.656(8) 2.54(3) 7 2.49(4) 2.69(3) 2.64(4) 7 2.53(6) 2.61(2) 2.57(2) 2.68(2) 2.58(2) 2.64(2) 2.59(2) 28 7 7 2.72(4) 2.71(4) 2.58(2) 29 7 77 30 31 32 33, 34 35 36 37 38 34 39 40 41 2.70(9) 7 2.44(12) 7 2.69(16) 2.53(6) 7 1.15, 1.15, 1.15 2.55(3) (Z3)7 7 2.75(3) (Z2) 2.51(5) (Z3)7 7 2.58(6) (Z3) 7 2.65(6) (Z3) 7 2.84(5) (Z2) 2.57(5) 7 2528, 2156, 2087, 1208 2.46(4) 2.61, 2.587 7 2.56(1) 1.13, 1.11, 1.20 2498, 2198, 2120, 1165 2.58(3) 7 7 2.48 7 7 2.50(3) 7 2518, 2156, 2090 2.52(3) 2.54(3) 2.553(1) 7 1.10(15) 2.51(2) 7 7 2.53(2) 2.629(7) 7 7 2.587(6) 2.599(9) 41 42 1.115(8), 1.02(3), 7 1.03(1) 7 7 42 43 7 7 7 7 44 2509, 2209, 2139 2.51(2) 2.65(2) 2.58(2) 2.57(1) 2.66(1) 7 45 46 47, 48 2452, 2256, 2201, 2163, 2135, 1246 2500, 2345, 2240, 2180 see a 2.68(2) 2.61(3) 2.46(3) 1.08(2), 1.11(3), 1.12(3) 771.0(3), 1.1(2), 1.1(2) 1.276(5) 49 2.175(4) 2585, 2400, 2230, 2155, 2030, 1345, 1265, 1210, 1075, 595, 450730 Table 1 (continued).Complex Complexes of Group III and IV transition metals Ti(Z2-BH4)3(DME) Ti(BH4)3(PMe3)2 Ti(Z2-BH4)2(DMPE)2 Cp*Ti(Z2-BH4){(Me2PCH2)3SiBut} Cp2Ti(Z2-BH4) Cp2Ti(Z2-BH4) (see b) [CpTi(Z3-BH4)(m-Cl)]2 {PhC[N(SiMe3)]2}2Ti(Z2-BH4) [(Me3Si)2N]2Ti(Z2-BH4)(THF) [(Me3Si)2N]Ti(Z2-BH4)2(Py)2 {[2,6-Pri2C6H3O]Ti(Z3-BH4)2}2 [2,6-Pri2C6H3O]Ti(Z2-BH4)2(PMe3)2 [2,6-Pri2C6H3O]3Ti(Z3-BH4) (ButNCH=CHNBut)TiCl(Z3-BH4) (ButNCH=CHNBut)Ti(Z2-BH4)(Z3-BH4) (BP)Ti(Z3-BH4)2 [Ti(CO)4(Z3-BH4)]7[K(18C6)]+ Zr(Z3-BH4)4 (see b) Zr(Z3-BH4)4 (BP)Zr(Z2-BH4)(Z3-BH4)(THF) Cp2Zr(Z2-BH4)2 [(MeC5H4)2Zr(Z2-BH4)(THF)][BPh4] (see c) 1.87(9), 1.93(9) 1.98(9), 1.92(8) 2.068, 2.036, 2.162 2.125, 1.999 Zr(Z3-BH4)2{Me3SiN[CH2CH2N(SiMe3)]2} 2.04(7), 2.10(5), Zr(COT)00(Z2-BH4)(Z3-BH4) Zr2(COT)200(m-H)(Z3-BH4) Zr(Z3-BH4)[(3,5-Me2Ph)N(Ad)]3 Zr2H4(Z2-BH4)2(PMe3)4 Zr3H6(Z2-BH4)2(Z3-BH4)4(PMe3)4 Zr2H3(Z2-BH4)(Z3-BH4)4(PMe3)2 Distance /A M7Hb 1.93(2), 2.01(2) 1.92(2), 1.87(2) 1.94(2), 1.88(2) 1.90(6) 1.73(7), 2.25(7), 2.62(7) 2.04(2), 2.09(2) 2.00(5) 1.75(8) 1.89(5) 1.89, 1.90, 2.05 1.92 2.08(5), 2.13(5) 1.98(4), 2.03(4) 2.059(4), 1.99(4) 777 2.207(13) 1.87(3), 1.90(4), 2.175(4) 1.97(3) 1.88(3), 1.87(3) 1.90(3), 1.89(3), 1.91(3) 1.978, 1.870, 1.951 2.184(7) 1.875, 2.013, 1.984 2.205(8) 2.02(4) m 2.158(7) 2.211(19) 2.308(3) 7 2.34(3) 72.03(6) 2.21(4), 2.20(8), 2.31(5) 772.16(4), 2.21(4), 2.23(4) 2.16(5), 2.20(5) 2.05(3), 2.11(4), 2.21(3) 2.14(3), 2.15(4) 2.19(3), 2.12(4) 2.08(3), 2.24(3), 2.08(3) 2.06(4), 2.10(4), 2.19(4) 2.35(4), 2.19(3), 2.11(4) 1.98(6), 2.23(7), 2.02(6) 2.05(5), 2.22(6) 2.05(7), 2.17(6), 2.14(5) M_B B7Hb 1.22(2), 1.16(2) 1.18(2), 1.16(2) 1.03(2), 1.16(2) 2.424(3) 2.403(2) 2.405(2) 1.03(7) 0.95(6) 2.40(1) 2.27(1) 1.15(2) 0.98(6), 1.05(5) 1.23(8) 7 7 1.07, 1.18, 1.11 1.14 7777 2.534(3) 2.445(7) 2.37(1) 2.31(4) 2.17(1) 2.426(4) 2.454(4) 2.476(6) 2.482(5) 2.166(4) 2.162(4) 2.457(6) 2.444(6) 71.12(4), 1.09(4), 1.18(4) 2.304(3) (Z2) 1.08(3), 1.06(3) 2.175(3) (Z3) 1.02(4), 1.07(3), 1.14(4) 771.11(2) m 1.272(16) 771.10(9), 1.30(9) 1.24(8), 1.27(9) 2.59(2) 2.54(1) 2.55(1) 2.354(8) (Z3) 7 2.572(12) (Z2) 7 2.370(7) 7 2.404(7) 77 7 2.605 (Z2) 2.368 (Z3) 2.33(2) 2.400(4) 7 72500, 2234, 2180, 2154 77 7 2.59(1) 2.352(4) (Z3) 1.17(4), 1.18(4), 1.19(3) 2.610(4) (Z2) 1.14(4), 1.14(4) 2.655(4) (Z2) 1.14(4), 1.13(4) 2.363(6) (Z3) 1.21(4), 1.10(4), 1.14(4) 2.361(6) (Z3) 1.08(4), 1.03(4), 1.19(4) 2.394(6) (Z3) 1.14(4), 1.12(4), 1.14(4) 2.336(7) (Z3) 1.27(6), 1.05(7), 1.05(6) 2.604(8) (Z2) 1.18(5), 1.03(7) 2.361(8) (Z3) 1.03(8), 1.23(6), 1.01(6) V D Makhaev Ref.Vibration frequencies of the BH4 group (IR spectroscopy) /cm71 50, 51 2470, 2420, 2130, 2060 52 2532, 2401, 2358, 2110 2332, 2297, 2220, 2138, 2062 72438, 2400, 2050, 1940, 1315, 1115 53, 54 55 56, 57 58 59 60 61 61 72420, 2395, 2110 2440, 2400, 2270, 2195 2380, 2340, 2220, 2140 62 2569, 2560, 2214, 2139, 2097, 2068 62 2412, 2401, 2377, 2368, 2164, 2130 62 63 2528, 2207, 2196, 2140 2546, 2173, 2121, 1213 63 2545, 2178, 2099, 1212 64 2559, 2542, 2209, 2118 65 66 67, 68 2495, 2132, 2058 72567, 2180, 2115, 1286, 1213, 1170, 1098, 1056, 575, 504 2431, 2375, 2216, 2137, 1294, 1108 2472 ± 2429, 2119 69, 70 71 64 2539, 2458, 2400, 2219, 2117, 2020 72 2519, 2494, 2236, 2214,2162 73 73 74 75 76 2530, 2495, 2430, 2400, 2385, 2320, 2230, 2135, 2095 77 2550, 2505, 2435, 2390, 2220, 2155, 2102, 1520Structural and dynamic properties of tetrahydroborate complexes Table 1 (continued).Complex Complexes of Group III and IV transition metals Zr2H3(Z2-BH4)(Z3-BH4)4(PMe3)2 Zr2H4(Z2-BH4)(Z3-BH4)3(DMPE)2 (ButNCH=CHNBut)Zr(Z3-BH4)2 Zr3S3(SBut)2(Z2-BH4)(Z3-BH4)3(THF)2 Zr6S6(SBut)4(Z2-BH4)2(Z3-BH4)6(THF)2 Hf(Z3-BH4)4 (see d) (MeC5H4)2Hf(Z2-BH4)2 (see d) Hf2H3(Z2-BH4)(Z3-BH4)4(PMe3)2 Hf2(NPP)2H3(Z3,Z2,Z1 -BH4)3 Complexes of Group I and V± VIII transition metals with a bidentate coordination of the BH4 ligands V(Z2-BH4)3(PMe3)2 [V(Z2-BH4)2(PMe3)2]2O CpV(Z2-BH4)(DMPE) [V(m-Cl)(m-DPPM)(Z2-BH4)]2 [(Me3Si)2N]2V(Z2-BH4)(THF) Cp2Nb(Z2-BH4) Me2C(C5H4)2Nb(Z2-BH4) Ta2(m-BH3)(m-DMPM)(Z2-BH4)2(C7H8) Ta2(m-BH3)(m-DMPM)(Z2-BH4)2 (TMEDA)Cr(Z2-BH4)2 CrH(Z2-BH4)(DMPE)2 Cr(CO)4(Z2-BH4)(Ph3P)2N Mo(CO)4(Z2-BH4)(Ph3P)2N MoH(Z2-BH4)(PMe3)4 W(Z2-BH4)(CO)(NO)(PMe3)2 Mn(Z2-BH4)2(THF)3 (Ph4P)[Mn(Z2-BH4)3(THF)] FeH(Z2-BH4)[MeC(CH2PPh2)3] RuH(Z2-BH4)(PMe3)3 OsH3(Z2-BH4)[P(cyclo-C5H9)3]2 Co(Z2-BH4)(terpy) Co(Z2-BH4)(terpy) (seed) CoH(Z2-BH4)[P(cyclo-C6H11)3]2 Co(Z2-BH4)[MeC(CH2PPh2)3] Distance /A M7Hb 2.17(5), 2.04(5), 2.21(5) 2.14(5), 2.09(6), 2.36(7) 2.32(8), 2.33(9) 2.08 ± 2.47 2.26(5), 2.03(4), 2.15(5) 2.09(5), 2.06(4), 2.30(5) 772.130(9) 2.069(7), 2.120(8) 2.553(6) 71.93, 2.43 2.01, 2.25 2.25 m 1.83(3) 1.97(3) m 1.78(4) 1.78(8), 1.69(10) 1.711(7), 1.916(6) 2.38(2) 2.26 2.36(1) 2.578(8) 2.539(10) 2.585(10) 2.44(1) 2.42(1) 2.45 2.0 1.94(7) 2.16(7), 2.17(7) 1.82(7), 2.25(8) 2.04(8), 2.12(7) 1.931, 2.094 1.875, 2.081 71.96(7), 1.80(6) 1.99(9), 2.04(8) 1.96 1.8(1), 2.1(1) 2.04(4) 1.95, 1.98 1.95, 2.18 1.87, 2.24 2.29(1) 2.413(20) 2.468(12) 2.36(2) 2.438(3) 2.368(3) 2.335(2) 2.362(2) 1.58(11), 1.65(11) 2.16(2) 1.81(4), 1.85(4) 1.90 1.81(5), 1.80(5) 1.707(12), 1.740(12) 1.80(8), 1.87(9) 1.55(15), 1.45(15) 2.21(3) M_B B7Hb 2.380(6) 2.331(5) 2.332(7) (Z3) 1.11(5), 1.19(5), 1.00(5) 2.357(8) (Z3) 1.07(6), 0.97(6), 1.11(7) 2.70(1) (Z2) 1.21(10), 1.06(9) 2.40(2) (Z3) 0.94 ± 1.38 1.16(5), 1.04(4), 1.04(6) 1.26(5), 1.11(4), 1.08(5) 2.53(2) (Z2)7 7 2.33(2) (Z3) 2.35(2) (Z3) 2.35(2) (Z3) 2.511(11) (Z2)7 7 2.279(2) (Z3) 2.325(11) (Z3) 2.276(12) (Z3) 2.281(8) 1.235(10) 2570, 2197, 2123, 1290, 1218, 1140, 480 1.255(9), 1.208(13) 2444, 2398, 2235, 2160, 1340, 1120 2545, 2505, 2430, 2385, 2210, 2160, 2105 2.33(2) (Z3) 7 2.53(3) (Z2) 2.34(3) (Z3) 2.35(3) (Z3) 2.32(3) (Z3) 2.636(7) (Z1) 1.29 m 2.583(6) (Z2) 2.322(8) (Z3) 2.365(5) 2.254(9) 2.314(14) 1.14(3) 2.383(6) m 1.15(4) aÊ á.1.17(4), 1.12(8) 1.24(9), 1.36(10) 0.943, 1.43(7) 1.1 1.25(7) 1.09(7), 1.09(7) 1.13(13), 1.19(7) 1.12(8), 1.16(7) 7771.25(8), 1.01(6) 1.26(11), 1.14(9) 1.12 71.1 m 1.10 m 7 71.31 1.44(6), 1.27(4) 2.30(1) 7 1.32(12),1.18(10) 2380, 2320, 1910, 1440, 1182 2383, 2369, 2318, 1932 2448, 2427, 2130, 2112 2420, 2403, 2325,2000, 1985, 1927, 1893, 1190 1.290(9), 1.287(10) 7 1.30(7), 1.39(9) 2.14(1) 2390, 2368, 1958, 1379 1.27(16), 1.38(17) 2370, 2330, 2015 731 Ref.Vibration frequencies of the BH4 group (IR spectroscopy) /cm71 77 2550, 2505, 2435, 2390, 2220, 2155, 2102, 1520 76 2490, 2370, 2300, 2200, 2175, 2140 63 2530, 2216, 2109, 1219, 1208 78 78 79 ± 81 82 77 83 2520, 2425, 2400, 2144 84, 85 85 86, 87 88 89 90, 57 91 92 92 2431, 2385, 2353, 2227, 2218, 2087 2408, 2397, 2359, 2229, 2106 2380, 2345, 2255, 1855 72434, 2396, 2240, 2134 2460, 2423,1745, 1628, 1382, 1162 2461, 2419 72376, 2341, 2203, 2122, 2090, 2052 93 2380, 2230, 2120 94 54 95 96 97 98 99 100, 101 2460, 2430, 2060, 1650, 1060, 2330, 2071, 1056 2415, 2365, 2210 2390, 2375, 2330, 2280 2340, 2290, 1935,1885,1360, 1165 2450, 2415 2390, 2140, 1115 2395, 2210, 2135, 1110 102, 103 104 105 106 106 107, 108 109V D Makhaev 732 Table 1 (continued).Ref. Complex Distance /A Vibration frequencies of the BH4 group (IR spectroscopy) /cm71 M_B B7Hb M7Hb Complexes of Group I and V± VIII transition metals with a bidentate coordination of the BH4 ligands 110 2421, 2383, 1383, 1138 7 7 Rh2(Z2-BH4)2(m-CO)(Ph2Ppy)2 1.27(5), 1.30(5) NiH(Z2-BH4)[P(cyclo-C6H11)3]2 Cu(Z2-BH4)(dmphen) Cu(Z2-BH4)(dmdphen) Cu(Z2-BH4)(phen)(PPh3) (seed) Cu(Z2-BH4)(phen)[P(OEt)3] Cu(Z2-BH4)(PPh3)2 Cu(Z2-BH4)(PBuPh2)2 Cu(Z2-BH4)(DPPF) 2.29(5) 2.23(5) 2.201(8) 2.08(2) 2.116(12) 7 2.29(2) 2.34(1) 2.185(6) 2.20(1) 2.158(8) 2.247(6) 111, 112 113 114 114 1153, 116 117, 118 119 120 2360, 2050, 2020, 1868, 1140 7 72385, 2350, 2245, 1982, 1128 2360, 2330, 2245, 2070, 1910, 1120 2360, 2330, 2080 2392, 2350, 2262,1984,1924 2404, 2394, 1995, 1937, 1139 2397, 2354, 2013, 1970 2365, 2010, 1120 1.24(3), 1.17(3) 0.81(16), 0.83(13) 1.07(3) 1.00(10), 1.40(10) 1.17(7), 1.23(6) 7 1.73(5), 1.76(6) 1.58(5) 1.50(5) 2.03(9), 1.63(7) 2.16(10), 1.98(10) 1.82(3) 1.60(10), 1.80(10) 1.65(6), 1.60(6) {CH2N(CH2CH2PPh2)2Cu(Z2-BH4)}2 1.683, 1.681 Complexes of Group I and V± VIII transition metals with a monodentate coordination of the BH4 ligand 54, 84 93 V(Z1-BH4)2(DMPE)2 Py4Cr(Z1-BH4)2 2.833 3.253 3.309 2.84 2341, 2312, 2110,2095, 1060 2300, 2180, 2120, 2080, 1080 ± 1040 2350, 2075, 1060 2320, 2265, 1055 2315, 2045, 1060 1.12(3) 1.16 0.98 1.14 2.895(16) 1.09 2.578(10) 1.19 1.170(5) 71.2(1) 0.98(9) 1.16(6) 1.88 1.94 2.33 1.72 1.83 1.83(8) 1.697(1) 1.60(4) 1.76(7) 2.28(1) 2.25(1) 121 ± 123 124 125, 126 125 109, 127 128, 129 130 ± 132 130 ± 132 2.518(3) 2.44(2) 2.80(1) 3.09(1) 3.18(1) 2300, 1980 2300, 2050, 1070 2315, 2227, 2156, 1104 2377, 2333, 2321, 2160, 1118 FeH(Z1-BH4)(DMPE)2 RuH(Z1-BH4){Me8[16]aneS4} Cu(Z1-BH4)(PMePh2)3 Cu(Z1-BH4)(PMePh2)3 (see d) Cu(Z1-BH4){MeC(CH2PPh2)3} Ag(Z1-BH4)(PMePh2)3 Cu(Z1-BH4)2(Cyclam) Cu(Z1-BH4)2(TMAC) Complexes of post-transition metals 71.1 133 134, 135 136 137 2422, 2368, 2207, 2120 2455, 2438, 2410, 2060 72445, 2420, 2100 Zn(Z2-BH4)(Cl)(TMEDA) (Ph4P)[Zn(Z2-BH4)3] V2Zn2(Z2-BH4)2(PMePh2)4 CpNb(CO)(m-H)Zn(Z2-BH4)2 72.29(2) 2.227 70.986, 1.056 1.067, 1.026 71.334 1.250 ± 1.456 1.983(2), 1.963(2) 1.80 1.78(5), 1.82(5) 1.839, 1.871 1.736, 1.942 71.826 1.762 ± 1.891 138 139 140 2.423(5) 2.179 2.172 Cd(Z2-BH4)[HB(3,5-Me2pz)3] H2Ga(Z2-BH4) (see c) HGa(Z2-BH4)2 (see c) 2426, 2384, 2037 72566, 2550, 2488, 2478, 2471, 2042, 2026, 1984, 1976, 1971, 1961 2540, 2470, 1980, 1922 141, 142 1.217 2.163 1.791 Me2Ga(Z2-BH4) (see c) 2419, 2376 2.374(3) 143 144 7 7 7 7 7 Cp(CO)2FeGa(Z2-BH4)(CH2)3NMe2 7 Me2In(Z2-BH4) (see e) Complexes of s-elements and aluminium 145 146 147 147 147 2.223(7) 2.36(3) 2.470(4) 2.478(6) 2.319(7) 71.1(2) 1.17(2) 71.15(3) Li(Z3-BH4)[HC(3,5-Me2pz)3] [Li(Z1-BH4)]2(18C6) Li(Z2-BH4)(DME)2 {Li(Z2-BH4)(TG)}? Li(Z3-BH4)(THF)3 148 148 148 148, 149 Li(Z2-BH4)(Py)3 Li(Z3-BH4)(py*)3 Li(Z2-BH4)(coll)2 Li(Z2-BH4)(PMDETA) (see c) 1.15, 1.07 1.15, 1.11, 1.12 1.17(2) 1.07(4), 1.07(4) 1.31(4), 1.30(4) 2.401(7) 2.279(4) 2.252(6) 2.29(2) 2.35(2) 2.70 1.88(1) 1.93(1) 2.451(4) 2.646(4) 150 2475, 2424, 2170,1477, 1131, 990, 823 151, 152 153 154 155 NaBH4(15C5) CpBe(Z2-BH4) (see a) (ButO)4Be3(Z2-BH4)2 Mg(Z2-BH4)2(THF)3 Ca(Z3-BH4)2(DME)2 72400, 2370, 2310, 2250, 2090 2370, 2350, 2250, 2210, 2160 2.665(5) 2374 ,2314, 2223, 2163 72373, 2281, 2258, 2134 2560, 2382, 2330, 2295, 2264, 2198 2382, 2334, 2272, 2237, 2176, 2127, 2111 2354, 2309, 2279, 2245 2231, 2230, 2209, 2149 2421, 2330, 2268 2324, 2284, 2210, 2198, 2181, 2163, 2102 7 7 1.29(5) 1.04(7), 0.89(7) 1.10(4), 1.23(4) 1.121(34), 1.086(34) 1.085(29) 1.135(33), 1.153(29) 1.041(31) 156 Ca(Z2-BH4)2(Z3-BH4)(Diglyme)2 (see c) 2.884(4) (Z2) 1.10(3), 1.15(3) 71.7(2) 1.7(2) 72.02(4), 2.10(5), 2.12(4) 2.06, 1.90 2.07, 2.08, 2.26 1.83 2.05(4), 2.04(4) 1.92(4), 1.92(2) 71.78(9) 1.47(8), 1.63(9) 2.09(4), 1.97(4) 2.474(30), 2.451(29) 2.475(28) 2.345(29), 2.429(31) 2.577(29) 2.41(4), 2.44(3), 2.60(3) 2.716(4) (Z3) 1.17(4), 1.11(4), 0.91(5) 7 2.32(3), 2.44(3) 2.41(4), 2.47(4), 2.51(4) 2.710(4) (Z3) 1.05(4), 1.03(4), 1.06(5) 2.38(4), 2.41(4) 2.873(5) (Z2) 1.04(5), 0.98(4)Structural and dynamic properties of tetrahydroborate complexes Table 1 (continued).Complex Complexes of s-elements and aluminium Sr(Z3-BH4)2(Diglyme)2 Sr(Z3-BH4)2(18C6) Ba(Z3-BH4)2(Diglyme)2 Ba(Z3-BH4)2(18C6) Al(Z2-BH4)3 (see b) Al(Z2-BH4)3 a-phase b-phase (PMePh3)[Al(Z2-BH4)4] Al(Z2-BH4)2Me (see b) Al(Z2-BH4)Me2 (see b) Al(Z2-BH4)3(NH3) Al(Z2-BH4)3(NMe3) [Al(Z2-BH4)2N(CH2)2]2 [Al(Z2-BH4)2(OSiMe3)]2 Note. The following designations are used: M7Hb is the distance between metal and bridging hydrogen; DMPE is 1,2-bis(dimethylphosphino)ethane; Ph2Ppy is 1-diphenylphosphinopyridine;DMEis 1,2-dimethoxyethane; Cp* is pentamethylcyclopentadienyl; DC18C6 is dicyclohexyl-18-crown-6; Cp is cyclopentadienyl; Ind is indenyl; tritox is tris(tert-butyl)methoxy group, But3CO; COT is cyclooctatetraene; BP is methylene-2,20-bis(4-methyl-6-tert- butylphenoxy) group, CH2(4-Me-6-ButC6H2O)2; 18C6 is 18-crown-6; (COT)00 is 1,4-bis(trimethylsilyl)cyclooctatetraene; Ad is adamantyl; NPP is bis[dimethyl(dimethylphosphinomethyl)silyl]amide, N(SiMe2CH2PMe2)2; DPPM is bis(diphenylphosphino)methane; DMPM is bis(dimethylphos- phino)methane; TMEDA is tetramethylethylenediamine; terpy is 2,20:6,200-terpyridyl; dmphen is 2,9-dimethyl-1,10-phenanthroline; dmdphen is 2,9- dimethyl-4,7-diphenyl-1,10-phenanthroline; phen is 1,10-phenanthroline; DPPF is 1,10-bis(diphenylphosphino)ferrocene; Me8[16]aneS4 is 3,3,7,7,11,11,15,15-octamethyl-1,5,9,13-tetrathiacyclohexadecane; Cyclam is 1,4,8,11-tetraazacyclotetradecane; TMAC is 5,7,12,14-tetramethyl- 1,4,8,11-tetraazacyclotetradecane; 3,5-Me2pz is 3,5-dimethylpyrazolyl; TG is dimethytl ether of triethylene glycol; py* is p-benzylpyridine; coll is 2,4,6- trimethylpyridine (collidine); PMDETA is N,N,N0,N00,N00-pentamethyldiethylenetriamine; 15C5 is 15-crown-5; Diglyme is diethylene glycol dimethyl ether. a Neptunium tetrahydroborate is isomorphous with plutonium tetrahydroborate; their IR spectra in the gas phase are similar to the IR spectrum of hafnium tetrahydroborate (see Ref.48). b Gas phase electron diffraction data. c Two independent molecules in the crystal cell.d Neutron diffraction data. e No structural or spectral data were reported. Table 2. 1H NMR and IR-spectroscopic data for some transition metal tetrahydroborates. Compound [Ti(CO)4(Z3-BH4)]7[K(18C6)]+ Cp2V(Z2-BH4) Cp2Nb(Z2-BH4) Cp2 Nb(Z2-BH4) Ind2Nb(Z2-BH4) Cp*CpTa(Z2-BH4) Ta(Z2-BH4)H2(PMe3)4 CrH(Z2-BH4)(DMPE)2 [Mo(CO)4(Z2-BH4)][(Ph3P)2N] MoH(Z2-BH4)(PMe3)4 WH3(Z2-BH4)(PMe3)3 W(Z2-BH4)(CO)(NO)(PPh3)2 W(Z2-BH4)(CO)(NO)[P(OPri3)]2 Mn2(CO)6(DPPM)(H)BH4 Distance /A M7Hb 2.64(4), 2.65(4), 2.77(4) 2.60(3) 2.69(3), 2.74(4), 2.95(5) 2.66(4) 1.801(6) 1.73(4) ± 1.76(4) 1.67(5) ± 1.75(4) 1.77(4), 1.82(4) 1.79(5), 1.80(4) 1.81(5), 1.83(4) 1.77(4), 1.83(5) 1.82(1) 1.77(3) 71.83(7), 1.75(8) 2.00(10), 1.95(11) 1.65(4), 1.71(4) 1.74(4), 1.84(4) 7d(BHt) /ppm d(MHxB) /ppm 74.87 0.044 7 724 5.7 5.2 5.7 6.27, 5.91 4.39 729.4 4.3 5.5 70.13 716.5 718.2 714.2 715.4 74.68 733.5 79.8 74.5 76.25 4.0 4.61 2.9 72.7,71.6 73.71,72.58 714.0 Vibration frequencies of the BH4 group (IR spectroscopy) /cm71 M_B B7Hb 2.916(7) 2.833(6) 3.058(7) 2.975(9) 2.143(3) 77771.283(12) 2347, 2247, 2167 2351, 2297, 2254, 2221, 2164 2340, 2274, 2233, 2204, 2153 2340, 2285, 2237, 2206, 2154 2555, 2490, 2218, 2135, 2030, 1930, 1505, 1425, 1112, 605 772472, 2411, 2240, 2150, 2127 2.10(2) ± 2.14(2) 1.13(4) ± 1.14(4) 2.10(1) ± 2.13(1) 1.12(3) ± 1.14(3) 0.95(5), 1.22(5) 1.09(5), 1.15(4) 1.07(4), 1.11(5) 1.04(4), 1.11(5) 1.241 2.261(5), 2.229(6) 2.254(7), 2.234(6) 2.152(16) 1.217(19) 2.128(8) 2550, 2485, 2035, 1950, 1496, 1414, 1117, 990 2545, 2475, 2035, 1960, 1540, 1465, 1127, 995 7 72515, 2439, 2150 7 1.18(7), 1.40(7) 1.16(11), 1.48(10) 1.26(5), 1.16(5) 1.20(5), 1.22(5) 7 7 2.229(8) 2.227(11) 2.257(12) 2.193(5) 2.181(5) 2.156(5) 2.143(4) T00 /K T0 /K DG6à /kcal mol71 Vibration frequencies of BH4 groups (IR spectroscopy) /cm71 8.8 7.6 210 185 178 77298 282 2448, 2412, 2311 14.6 16.4 13.2 16.4 7 7 2460, 2440, 1980, 1170 see a see a 710.0 7193 see a 2442, 2418, 1745, 1650, 1395, 1180 see a 2452, 2428, 1728, 1620, 1171 169 170 2478, 2437, 1747, 1621, 1160 171 172 94 96 97 94 346 388 331 298 b >323 193 7 72307 7 see a <2737 7 2495, 2420, 2060, 1800, 1275, 1140 7 7 2465, 2425 7 7 2470, 2430 2472, 2333 193 193 243 12.2 310 733 Ref.157 157 157 157 158 ± 160 160 161 162, 163 142, 162 164 165, 166 167 168 Ref. 65 57 169 98 98 173, 174V D Makhaev 734 Table 2 (continued). Ref. Compound T00 /K T0 /K DG6à /kcal mol71 d(MHxB) /ppm d(BHt) /ppm Vibration frequencies of BH4 groups (IR spectroscopy) /cm71 174 4.3 Mn2(CO)5(DPPM)(H)BH4 243 78.6,712.1 (Z2) 338 b7 7 2487 714.6 (Z1) 7 715.1 3.8 174 174 Mn2(CO)6(TEDIP)(H)BH4 Mn2(CO)5(TEDIP)(H)BH4 7 Re(Z2-BH4)(DPPE)2 ReH(Z1-BH4)(Z2-BH4)(PMe2Ph)3 ReH(Z2-BH4)(NO)(PPri3)2 7 7 223 7 7 7 7 7 7 <253 >306 197 b *240 2462, 2437, 1854, 1794 2463, 2437, 1843, 1790 2482, 2446, 1810 see a see a 722.4 712.8,715.8 713.4 7 7 2335 7 7 see a 2380 2.14,73.5 1.27 71.6 6.62 ReH(Z2-BH4)(NO)[P(cyclo-C6H11)3] 6.62 ReH(Z2-BH4)(NO)(PPh3)2 FeH(Z2-BH4)[MeC(CH2PPh2)3] FeH(Z1-BH4)(DMPE)2 FeH(Z1-BH4)(DEPE)2 FeH(Z1-BH4)(DPrPE)2 Fe3(CO)9H(BH4) (C6Me6)2Ru2H3(BH4) CpRu(PPh3)(Z2-BH4) Cp*Ru(PMe3)(Z2-BH4) Cp*Ru(PEt3)(Z2-BH4) Cp*Ru(PPri3)(Z2-BH4) <353 7 76.1 to 79.6 75.91,79.08 7 7 2475, 2360 79.5,712.4 (Z2) 295 b7 7 2466 713.45 (Z1) 79.0 75.3,713.7 73.91,76.12 73.59,76.39 7 72.92,74.53 7 714.7 0.87 721.2 7 721 3.2 3.27 717.2 711.05 2.10,73.22 711.63 7 712.54 7 712.36 7 712.0 7 712.03 6.1 ± 4.5 5.457 75.1,79.1 175 94 176 176 176 102, 103 123 123 123 177 178 179 180 180 180 180 180 180 104 181 182 2400, 2330, 1188 7 7 2390, 1100 77777 197 b7 7 7 197 b7 7 7 173 213 7 7 303 303 303 303 303 303 298 298 308 7 7 7 2438, 2362, 2282 7 7 2433, 2352, 2284 7 7 2393, 2318 7 7 2408, 2329 7 7 2387, 2292, 2223 7 7 2450, 2386, 2326 2383, 2369, 2318, 1932 7 7 2402, 2395, 2312 7 7 2395, 2375, 2315, 1945, Cp*Ru(PCy3)(Z2-BH4) Cp*Ru(PPh2Me)(Z2-BH4) Cp*Ru(PPh3)(Z2-BH4) RuH(Z2-BH4)(PMe3)3 RuH(Z2-BH4)(PMe2Ph)3 RuH(Z2-BH4)(PMePh2)3 7777 7 7 300 b 348 7 7203 3)2 RuH(Z2-BH4)(PPh3)3 RuH(Z2-BH4)[PhP(CH2)3PPh2)2] 5.1 RuH(Z2-BH4)[PhP(CH2)3PCy2)2] 4.8 RuH(Z1-BH4)[P(CH2CH2PPh2)3] 1.35 5.5 4.55 4.70 RuH(Z2-BH4)(PP3CyBH3) RuH(Z2-BH4)(CO)(PPri RuH(Z2-BH4)(CO)(PMeBut 77 2)2 1370, 1180 2382, 2340, 2080, 1119 2390, 2380, 2330, 1180 2395, 2330, 1960 2360, 2270, 1980, 1052 7 7 72420, 2320 2400, 2390 7 7 183 184 185 186 187 188 188 189 168 298 233 233 223 >243 >243 >243 [RuH(CO)(PPri3)2]2(BH4)(BF4) RuH(Z1-BH4)(PNP)(PPh3) 193 293 2.5 RuH(Z2-BH4)(S-biphemp)(PPh3) 5.8 7 7 2360, 2322, 1915, 1845, 1058 190 191 373 7 76.87 75.8,77.9 76.8,79.2 710.28 77.0 74.90,77.78 74.40,77.20 72.85 (2H) 75.76 (2H) 78.9 74.0,79.1 7 2403, 2380, 2333, 1966, 1331,1178 2447, 2351 10, 192 OsH3(Z2-BH4)(PPri3)2 105 300(Hb7OsH), 13.2 >360(Hb7Ht) >20 7 7 see a OsH3(Z2-BH4)[P(cyclo-C5H9)3]2 3)2 8.78 8.79 9.05 8.8 6.10 6.35 OsH(Z2-BH4)(CO)(PPri OsH(Z2-BH4)(CO)(PMeBut 77 2)2 213 360 193 363 233 233 223 188 188 189 2430, 2330 2440, 2420 7 7 >243 >243 >243 [OsH(CO)(PPri3)2]2(BH4)(BF4) >254 see a 2)2 2)2 Co(Z2-BH4)terpy Rh(Z2-BH4)H2(PMeBut Ir(Z2-BH4)H2(P MeBut Ir(Z2-BH4)H2(PBut3)2 76.78 710.14 76.83 710.19 b 76.50,78.10 76.15,77.75 73.82 (2H) 75.39 (2H) 79.97 74.53 76.86 77.76 714.2 193 194 2458, 2347, 2150, 1308, 1188 195 2460, 2367, 2142, 1330, 1196 195 196 11.1 7 7 7 77 7 7 see a 213 300 300 b >300 300 b >300 298 1.73 3.9 6.86 6.17 5.5 Cp2 Ir2H3(BH4) Note. The following designations are used: d(BHt) and d(MHxB) refer to the signals of the terminal and bridging hydrogen atoms of the tetrahydroborate ligands, respectively; T 0 is the temperature at which the spectrum was recorded; T00 is the temperature of coalescence of signals of bridging and terminal protons; DG= is the activation energy for the exchange of bridging and terminal protons; TEDIP is tetraethyl pyrophosphite; DPPE is 1,2- bis(diphenylphosphino)ethane; DEPE is 1,2-bis(diethylphosphino)ethane; DPrPE is 1,2-bis(dipropylphosphino)ethane; PP3Cy is tris(dicyclohexyl- phosphinoethyl)phosphine, P[CH2CH2P(cyclo-C6H11)2]3; PNP is 1,2-bis[N,N-bis(2-diphenylphosphinoethyl)amino]ethane; S-biphemp is (S)-2,20- dimethyl-6,60-bis(diphenylphosphino)biphenyl.a The data of IR spectroscopy are presented in Table 1. b The spectrum was recorded under the 1H±{11B} double resonance conditions. If the positions of hydrogen atoms cannot be determined by X-ray diffraction analysis, the denticity of the BH4 ligand can be estimated from the metal ± boron distance. These distances in mono-, bi- and tridentate terminal BH4 groups (Fig. 1) and polynuclear complexes with diverse tetrahydroborate bridges between metal atoms (Fig. 2).Structural and dynamic properties of tetrahydroborate complexes Table 3. Polynuclear complexes with bridging tetrahydroborate groups.Compound LiBH4(Et2O) LiBH4(MeOBut) LiBH4(C3H6O2) LiBH4(TMEDA) LiBH4(TMTA) [H2C(3,5-Me2pz)2]LiBH4 (4,40-Me2bipy)LiBH4 NaBH4(Diglyme) NaBH4(PMDETA) NaBH4(TMTCN) Be(BH4)2 Sr(BH4)2(THF)2 Ba(BH4)2(THF)2 Me2AlBH4 Y(BH4)3(DME) (1,3-But2C5H3)2CeBH4 (1,3-But2C5H3)2SmBH4 (COT)(THF)2NdBH4 ansa-Cp02Yb(BH4)2Li(THF)2 U(BH4)4 U(BH4)4(OMe2) U(BH4)4(OEt2) U(BH4)4(OPr2) [Cr2(PriNCHNPri)3]BH4 Cr3(BH4)6(TMEDA)2Py4 Mn2(CO)6(DPPM)(H)BH4 Mn2(CO)5(DPPM)(H)BH4 HFe3(CO)9(BH4) H[Fe(CO)3]3[(m2-H)(m02-H)(m00 2-H)BH] [RuH(CO)(PPri3)2]2(BH4)(BF4) [RuH(CO)(PPri3)2]2[(m2-H)2(m02-H)2B](BF4) [OsH(CO)(PPri3)2]2(BH4)(BF4) [OsH(CO)(PPri3)2]2[(m2-H)2(m02-H)2B](BF4) [HC(CH2PPh2)3RuH]2(BH4)(BPh4) [HC(CH2PPh2)3RuH]2[(m2-H)2(m02-H)2B]}(BPh4) (C6Me6)2Ru2H3(BH4) Co(DPPP)(BH4) Cp2 Ir2H3(BH4) {[(Ph3P)2Cu]2BH4}ClO4 MeZnBH4 [cyclo-(o-C6F4Hg)3](BH4)2(Bu4N)27 72300, 2035 219, 220 [cyclo-(o-C6F4Hg)3]2(BH4)(Bu4N)7 72159, 2052 219, 220 GaH2(BH4) Note.The following designations are used: TMTA is trimethylhexahydrothiazine; bipy is bipyridine; TMTCN is 1,4,7-trimethyltriazacyclononane; ansa- Cp0 is methylenebis(3-trimethylsilylcyclopentadienyl), CH2(Me3SiC5H3)2; DPPP is 1,5-bis(diphenylphosphino)pentane, Ph2P(CH2)5PPh2 . tetrahydroborate complexes are usually close to the sum of the Shannon ion radius of the metal 222 and the radius of the BH4 a b c H H M H B H M H H Figure 1. Coordination of terminal tetrahydroborate ligands in covalent complexes; (a) monodentate Z1; (b) bidentate Z2; (c) tridentate Z3.Type of coordination of the BH4 group according to X-ray diffraction data See Vibration frequencies of Fig. 2 the BH4 groups {(MeOBut)Li[(m2-H)(m0 {(Et2O)Li[(m2-H)(m02-H)(m00 2-H)(m4-H)B]}? 2-H)(m3-H)(m03-H)B]}? {(C3H6O2)Li[(m2-H)2(m02-H)2B]}? {(TMEDA)Li[(m2-H)(m02-H)(m3-H)BH]}2 {(TMTA)Li[(m2-H)(m02-H)(m3-H)BH]}2 cdabbb {[H2C(3,5-Me2pz)2]Li[(m2-H)(m02-H)(m3-H)BH]}2 b {(4,40-Me2bipy)Li[(m2-H)(m02-H)(m3-H)BH]}2 {(Diglyme)Na[(m2-H)(m02-H)(m3-H)BH]}? {(PMDETA)Na[(m2-H)(m3-H)2BH]}2 {(TMTCN)Na[(m4-H)BH3]}4 bjha {Be(Z2-BH4)[(m2-H)2(m02-H)2B]}? ii {(THF)2Ba[(m2-H)(m0 {(THF)2Sr[(m2-H)(m02-H0)(m3-H)2B]2}? 2-H0)(m3-H)2B]2}? {Me2Al[(m2-H)2(m02-H)2B]}? {(DME)Y(Z3-BH4)2[(m2-H)2(m02-H)2B]}? {(1,3-Bu2 tC5H3)2Ce[(m2-H)(m02-H)(m3-H)2B]}2 {(1,3-Bu2 tC5H3)2Sm[(m2-H)(m02-H)(m3-H)2B]}2 {(COT)(THF)2Nd[(m2-H)(m02-H)(m3-H)2B]}2 {ansa-Cp02Yb[(m2-H)2(m02-H)2B][(m2-H)2(m00 2-H)2B]Li(THF)2}? aaiiiaa {U(Z3-BH4)2[(m2-H)2(m02-H)2B]2}? {(Me2O)U(Z3-BH4)3[(m2-H)2(m02-H)2B]}? {(Et2O)U(Z3-BH4)3[(m2-H)2(m02-H)2B]}? {(Pr2O)2U(Z3-BH4)3[(m2-H)2(m02-H)2B]U0(Z3-BH4)4} [Cr2(PriNCHNPri)3][(m2-H)(m02-H)BH2] [(TMEDA)PyCr(Z2-BH4)[(m2-H)(m02-H)BH2]2[Py2Cr0(Z2-BH4)2] [Mn(CO)3]2(m2-DPPM)(m2-H)[(m2-H)(m02-H)BH2] [Mn(CO)3](m 2-DPPM)(m-H)][(m2-H)(m02-H)2BH][Mn0(CO)2] [(C6Me6)RuH2](m2-H)[(m2-H)(m02-H)BH2] {(DPPP)Co[(m2-H)(m02-H)(m3-H)BH]}2 [(Cp*IrH]2(m2-H)[(m2-H)(m02-H)BH2] [(Ph3P)2Cu]2[(m2-H)2(m02-H)2B]ClO4 aaafffegaaafbfaa {MeZn[(m2-H)2(m02-H)2B]}? e, f {GaH2[(m2-H)2(m02-H)BH](Ga0H2)(m02-H)2(m00 2-H)BH2] .(Ga00H2)[(m00 2 2 -H)(m000-H) 2BH]}? H H B H M B H HH ligand with the corresponding denticity [for Z3-BH4, r i^1.36(6)A and for Z2-BH4, r i^1.6(1)A].223 In most cases, the results of this summation are in good agreement with experimental results. For example, for La(III) with a coordination number of 12 (r i=1.36A),222 the lanthanum ± boron distances calculated in this way (2.72 and 2.96A) are close to the values found experimentally (2.76 and 2.90A) in [La(Z2-BH4)2(Z3- BH4)2(THF)2]7 (see Ref. 18). However, when the coordination number is low and the complex contains bulky ligands, the calculated metal7boron distances deviate substantially from those found experimentally.In our opinion, these deviations point to a considerable influence of steric factors on the structural properties of the BH4 ligands in the complexes. The use of so- 735 Ref. (IR spectroscopy) /cm71 2364, 2286, 2234, 2230 2317, 2256, 2180 7 147, 197 2410, 2320, 2306, 2277, 2180 147 147 198 2367, 2343, 2271, 2225, 2154 148 145 2376, 2332, 2292, 2259, 2241, 2221, 2181, 2168, 2138 2390, 2293, 2227 145 199 148 148 200, 201 157 157 160 72316, 2249, 2184 2398, 2304, 2227 2510, 2451, 2330, 2113, 2075, 1995 2372, 2313 2353, 2308, 2241, 2216, 2186, 2169 72468, 2319, 2300, 2230, 2175 202 203 204 205 206 207 ± 210 211 211 212 93 93 173 174 177 189 189 213 178 214 2290, 2180, 2100 2420, 2280 2255 72552, 2538, 2262, 2182, 2087 7772401, 2365, 2236, 2101 72472, 2333 2487 2335 7772355, 2220 2415, 2030, 1985 2430, 2360, 2290, 2115, 2040 196 215, 216 217, 218 2190, 2140, 2050 2255, 2215 221 7736 M M0 M MMM Figure 2.Coordination of bridging tetrahydroborate ligands in poly- nuclear complexes; (a ± j) see the text. called effective coordination numbers, instead of the true values, has been proposed for compounds of this type.223 Valuable information concerning the type of the metal7li- gand bond in the compounds under consideration can be gained using vibrational (IR and Raman) spectroscopy.2, 3 The observed absorption bands of BH4 ligands with different denticities were assigned resorting to the studies of model compounds with known structures.Certain correlations between the spectral character- istics and structures of complexes were established. The most characteristic spectral region for the compounds in question is the region of stretching vibrations of the B7H bonds (*1650 ± 2600 cm71). As a rule, a tridentate BH4 group is responsible for an intense absorption band at *2450 ± 2600 cm71 [stretching vibrations of the terminal B7H bonds, n(B7Ht)] and for an intense doublet of bands at about 2100 ± 2200 cm71 [stretching vibrations of the B7H bridging bonds, n(B7Hb)]. A bidentate type of coordination shows itself as an intense doublet of bands in the region of 2400 ± 2600 cm71 and an intense band (perhaps, with a shoulder) at *1650 ± 2150 cm71.Monodentate complexes are characterised by two broadened absorption bands, *2300 ± 2450 and *2000 cm71 (see Refs 2 ± 5). Complexes in which the BH4 group is bound to the metal ion via one or several M7H7B bonds have been named covalent tetrahydroborates.2 Unlike X-ray diffraction and IR spectroscopy data, the data of NMR spectroscopy for many tetrahydroborate complexes, both ionic and covalent, point to the equivalence of the four protons in the BH4 ligand. The disagreement between the results obtained by different methods is due to the difference between the time scales of X-ray diffraction analysis and IR spectroscopy (characteristic time *10718± 10720 s), on the one hand, and NMR (characteristic time *1072± 1079 s), on the other hand, and to the fluxionality of these compounds on the NMR time scale, i.e., the ability of the bridging and terminal protons of tetrahydroborate ligands to undergo fast intramolecular exchange.2±5a b In terms of composition, tetrahydroborate complexes can be M0 M M0 c d formed commonly by metals of each Group.Neutral homoleptic M00 M0 M M00 e f thallium(I).1 ± 5, 224 Many neutral tetrahydroborates are able to M0 M0 M h g M M0 M0 M00 M00 j i M0 M0 M V D Makhaev divided into two groups, homoleptic (containing one type of ligand) and heteroleptic (containing different ligands). Analysis of the vast experimental information accumulated to date permits one to establish the main types of tetrahydroborate complexes tetrahydroborates Mn(BH4)n are typical of alkali and alkaline earth metals, beryllium, magnesium and aluminium. Among transition and post-transition metals, these complexes are known only for some lanthanides in the oxidation states +2 or +3, titanium subgroup metals, some actinides, zinc, cadmium and attach additional BH4 groups to give anionic complexes of the type [Mn(BH4)n+k]k7 (k=1 ± 3); however, this anion stabilises noticeably only tetrahydroborates of 3d metals of the beginning of the transition metal row, from Sc to Fe.No homoleptic anionic complexes of cobalt, nickel, copper or Group I and V± VIII 4d or 5d transition metals have been obtained in a pure state.225 In most cases, in addition to BH4 groups, tetrahydroborate complexes contain various stabilising ligands; the ligands found most often are ethers and polyethers, amines and polyamines (both cyclic and acyclic), cyclopentadienide anion derivatives and compounds of trivalent phosphorus (tertiary organic phosphines and phosphites). A large number of relatively stable complexes with ethers have been prepared for elements for which non- solvated neutral tetrahydroborates are known. A complex with tetrahydrofuran stable at room temperature is also formed by divalent manganese the ions of which possess a half-filled d shell.99 For other transition metals, these complexes are either unstable [for example, the chromium complex Cr(BH4)2(THF)2 decom- poses at720 8C] or unknown.3 Complexes with ligands such as halide, alkoxy or aryloxy anions and NO are rather rarely encountered in the chemistry of tetrahydroborates.Complexes with CO molecules as ligands are quite numerous but are mainly formed for Group VI ± VIII metals. Anionic ligands (amides, alkoxides, aryloxides, etc.) stabilise effectively only tetrahydroborates formed by 3d metals of the beginning of the transition metal row. Examples of stabil- isation of platinum metal complexes by cyclopentadienyl ligands are also documented. Unlike the above-mentioned ligands, amines and polyamines stabilise more efficiently complexes of metals from the middle to the end of each transition metal row.Organic tertiary phosphines are the most efficient stabilising ligands for tetrahydroborates. In many cases, complexes contain hydride ligands together with phosphines. Tetrahydroborate complexes with these ligands are known for most transition metals but not for rare earth elements. The broad range of stabilising capacity of phosphines is stipulated by the broad range of variation of steric and electronic properties of these compounds. To fill the valence shells of metal atoms of the beginning of a transition metal row, a large number of electron-donating ligands are needed; in this case, strong electron donors having relatively small sizes such as trimethylphosphine and bis(dimethylphosphi- no)ethane are effective.On the contrary, to stabilise Group VIII metal complexes, a few ligands should shield the central atom having many valence electrons from contacts with the outer- sphere electron acceptors. In this case, the stability of complexes increases when they contain bulky ligands [P(cyclo-C6H11)3, PBut3]. The most diverse types of tetrahydroborate complexes are formed by Group I, III and IV transition metals, although no complexes with CO ligands have been found for Group III metals and no complexes with CO or Cp ligands have been described for Group I transition metals. The tetrahydroborate complexes of nickel and palladium are least diverse (no complexes of this type are known for platinum), which is in sharp contrast with the extremely varied chemistry of coordination compounds of these metals with many other ligands (Ni, Pd, Pt are typical complex- forming elements).The typical stabilising ligands vary depending on the position of the metal in the Periodic Table. Whereas forStructural and dynamic properties of tetrahydroborate complexes Table 4. The ability of ligands of different types to stabilise transition metal tetrahydroborates. n L in the complexes M(BH4)nLx M M(BH4)n [Mn(BH4)n+1]7 ethers NR¡2 7+++(V) +++7+(Cr) +(Mn) +++7777+ 7 7++++ +(Cr) 7 7 7 7 7 +(Ru) 7 7 +(Ir) 7 7 7 7 +(Ni) 7 7 7 7 +(Cu) 7 7 7 7 7 7 7 7 +7 7 7 7 7 7 7 + ++++(V) +(Cr) +(Mn) +(Fe) 7 7 7 7 7 7 7 7+ 7 7 REE Actinides Ti, Zr, Hf V, Nb, Ta Cr, Mo,W Mn, Re Fe, Ru, Os Co, Rh, Ir Ni, Pd Cu, Ag Pt, Au Zn, Cd Hg Note.A plus sign means that for the given group of metals, tetrahydroborate complexes with ligands of this particular type are known, while a minus sign means that these complexes are unknown; the metal (or ion) to which the plus sign refers is given in parentheses. complexes of Group III and IV transition metals, ethers and amines are commonly found ligands, in the case of Group I and V± VIII metal tetrahydroborates, compounds with these ligands are unusual. The ligands found most often for the latter type of metals are derivatives of tertiary organic phosphines, which are represented by only a few examples in the case of Group III and IV transition metal complexes.The dependence of the types of stabilising ligands and com- plexes they form on the nature of the central transition metal atom can be followed by analysing the data of Table 4. It can be seen that the number of ligand types capable of efficient stabilisation of tetrahydroborate complexes decreases on passing from Group III and IV transition metals to Group VIII metals, especially to nickel and palladium for which only few examples of tetrahydroborate complexes are known. The decrease in the stability of tetrahydro- borate complexes is even more pronounced on passing from 3d to 4d and 5d metals. As a result of these two trends, no tetrahydro- borate complexes of 5d metals of the end of the transition row, viz., platinum or gold, have been characterised.III. Transition metal complexes with terminal BH4 ligands The complexes of Group III and IV transition metals are most diverse regarding the mode of addition of the BH4 ligands to the central atom (see Table 1). Compounds with any of the main types of coordination (see Fig. 1) can be found among complexes of these metals. Molecules containing BH4 ligands coordinated to the same metal atom in two different modes are also known, for example, Sc(Z2-BH4)(Z3-BH4)2(THF)2 ,11 Y(Z2-BH4)(Z3-BH4)2. .(THF)3 ,14 (BP)Zr(Z3-BH4)(Z2-BH4) . (THF).64 The Hf2. .(NPP)2H2(BH4)3 complex contains simultaneously mono-, bi- and tridentate BH4 groups.83 The structures of Group I and V± VIII metal complexes are much less studied (see Table 1); this can be explained by the sensitivity of many of them to oxygen and moisture and thermal instability and the difficulty of their synthesis and investigation arising thereupon.The structural chemistry of the tetrahydrobo- rates of Group I and V± VIII transition metals is much scantier than that for Group III and IV metals. No complexes with tridentate BH4 groups have been found for them; in the majority of their compounds, the BH4 groups are bidentate ligands. Among compounds of Group I and V± VIII transition metals, there are quite a few complexes with a monodentate coordination of tetrahydroborate ligands (*25% of the total number of structurally characterised complexes of the metals in question); this type of coordination is rather poorly studied.Nowadays, the 737 Other ligands H amines Cp derivatives CO PR3 ++ ++ 77+(Ti) +(Ta) +++ +(LnII) +(U) +7+(Cr) +(Mn) + +(Rh, Ir) +(Co) +++++ 7 7 +++ + + ++++++ + + +++++7 + 7 + + 7 existence of complexes with Z1-BH4 groups has been reliably established by X-ray diffraction analysis and neutron diffraction. Structures of about 10 compounds of this type determined by X-ray diffraction analysis are documented; one of these com- pounds has also been studied by neutron diffraction. Taking into account the data ofNMRspectroscopy, we would like to note that for Group I and IV ± VIII transition metals, at least one example of an Z1-complex can be given for each group.It is noteworthy that all Z1-complexes incorporate P-, N- or S-donor ligands (tertiary organic phosphines, amines and thia- and aza-macrocycles) possessing strong electron-donating prop- erties. In most of these compounds, the central atom is linked to three or more phosphorus, nitrogen, or sulfur atoms. In terms of the type of the stabilising ligand, the compounds considered can be divided into two groups, namely, complexes with P- and N-donor ligands, which differ markedly in physico- chemical characteristics. The length of the M7Hb bond in the former type of compound varies from 1.60 to 1.88A and the M7B distance ranges from 2.44 to 2.84A. TheM7H7B bond is bent, the MHB angle ranging from 121 to 164 8.The IR spectra of these compounds exhibit broad absorption bands in the ranges of 2300 ± 2350 cm71 [n(BHt)] and *200072100 cm71 [n(BHb)], which is consistent with the values predicted for Z1-complexes.3 Resorting to the above-mentioned correlation between the struc- tures and spectroscopic properties, the presence of monodentate BH4 groups in L3CuBH4 type complexes [L = PHPh2 , P(OPri )2Ph, PMe2Ph] was demonstrated by IR spectroscopy. Study of the relationship between the structure of the copper compounds LnCuBH4 and the electronic and steric properties of phosphine ligands L showed that the complexes L3Cu(Z1-BH4) with monodentate tetrahydroborate ligands are formed only when the electron-donating ligands L have a moderate size.A decrease in the electron-donating ability of the ligands, as well as an increase in their size, result in the complexes L2Cu(Z2-BH4).118 However, compounds with N-donor ligands follow a different type of regularities. X-ray diffraction data point to an unusually long metal7boron distance (3.09 ± 3.31A) in these compounds and IR spectra attest to the presence of BH¡4 anions.130, 131 In solutions, these complexes appear to have ionic structure.132 For compounds with P-donor ligands, the difference between the experimentally determined metal7boron distance and the ion radius of the central atom is 2.00.2A. In the complexes with N-donors, this difference is close to 2.4A, i.e. it substantially exceeds the radius of the BH¡4 anion (2.04A).This long metal7 boron distance implies apparently the absence of covalent inter- action between the central atom and the ligand, the monodentate738 coordination of the BH4 ligand in the crystals of these compounds being related to packing effects. The NMR spectra of some monodentate complexes exhibit separate signals for the terminal and bridging protons, i.e. these compounds are less fluxional than complexes with tridentate BH4 groups. It was shown in relation to iron compounds that an increase in the size of stabilising ligands increases the stereo- chemical rigidity of complexes, i.e. stabilises the Z1-coordina- tion.123 However, comparison of the structures of Ti(Z2-BH4)2. .(DMPE)2 (d 2-configuration of titanium) with V(Z1-BH4)2..(DMPE)2 (d 3-configuration of vanadium)53, 54 and CrH(Z2- BH4)(DMPE)2 (d 4-configuration of chromium) with FeH(Z1-BH4)(DMPE)2 (d 6-configuration of iron),94, 123 which contain Ti(II) orV(II) and Cr(II) or Fe(II) ions with closely similar radii, demonstrates that Z1-coordination is not determined by steric factors alone. The Z1-coordination may also be facilitated by the d 3-configuration of V(II) and the d 6-configuration of Fe(II), which are favourable for the formation of octahedral complexes. The tetrahedral geometry, preferred for d 10 ions for energy reasons, stipulates apparently the formation of the mono- dentate complexes Cu(Z1-BH4)(PMePh2)3 and Ag(Z1-BH4). .(PMePh2)3 , despite the fact that the ionic radius of Ag(I) is greater than that of Cu(I) and the volume of the silver coordina- tion sphere is sufficient for Z2-coordination.125, 129 A Z1-coordi- nation of BH4 has been suggested previously for RuH(Z2-BH4). .(PPh3)3 ;183 however, the fact that its IR spectrum differs from the IR spectra typical of Z1-complexes and resembles the spectrum of the complex RuH(Z2-BH4)(PMe3)3, which has been characterised by X-ray diffraction analysis (see Table 1), points to a bidentate coordination of the BH4 group in this complex.Thus, in all the known cases, monodentate coordination of the BH4 groups is due to the presence of stabilising ligands possessing definite properties (strong electron donors having a large size), i.e. this type of coordination is forced. Complexes with Z1-coordina- tion of BH4 groups might also be possible for ions with a different electron configuration provided that the coordination sphere contains a sufficient number of bulky electron-donating ligands.However, the currently available data do not allow one to predict unambiguously the appearance of Z1-coordination in each partic- ular case. The IR spectroscopic studies show that in most cases, exper- imental results are in good agreement with those predicted theoretically for the BH4 groups of the corresponding coordina- tion;2 ±5 however, in some cases, the absorption bands due to the stretching vibrations of terminal B7H bonds occur at a substan- tially lower frequency than expected (see Tables 1, 2). Experimen- tal data on the structures of tetrahydroborate complexes show that the idea of the three main types of coordination of the BH4 ligands in so-called covalent tetrahydroborates is an idealisation.Among complexes of both Group III, IV and Group I, V± VIII transition metals, there are compounds with a distorted geometry of coordinated BH4 ligands (unequal M7Hb distances and MHbB angles), which hampers unambiguous identification of the type of coordination. For example, in the structure of Ti(Z2-BH4)(Z3-BH4)2(PMe3)2, the tridentate coordination of the two BH4 ligands is so distorted that these groups can be regarded as being coordinated through the B7Hbonds (the Ti7B distance is 2.27A, Ti7Hb is 1.73A, the MHbB angle is 112 8).52 It has been noted that this coordination can be regarded as being characteristic of a hypothetical transition state in the activation of the methane molecule, isoelectronic with the tetra- hydroborate anion.A similar distortion of a symmetrical coordi- nation is also observed for some Z2-complexes, for example, for (phen)(PPh3)Cu(Z2-BH4) (the Cu7B distance is 2.29A, Cu7Hb is 1.63A).114 This type of coordination plays, apparently, an important role in the exchange of bridging hydrogen atoms and hydride ligands in the complexes MH3(BH4)L2 .10 It has been noted above that the M7Hb7B bridge in monodentate com- plexes is bent, and the MHbB angle varies over wide limits (from 120 to 164 8); hence, in some cases [for example, in V D Makhaev 4 Hf2(NPP)2H2(Z3,Z2,Z1-BH4)3], the coordination of BH4 groups found experimentally can be considered to be intermediate between mono- and bidentate coordinations.83 The distortion of the geometry of coordinated ligands affects substantially the spectral properties of the complexes; therefore, in some cases, the denticity of the BH4 groups cannot be identified unambiguously from the data of vibrational spectroscopy; for example, this is true for the copper and titanium compounds considered above.52, 114 Compounds of Group III and IV transition metals possess clear-cut fluxionality; therefore, all the protons of the BH4 ligands are equivalent on theNMRtime scale.2± 5 Until recently, for none of the complexes of this type, were separate signals for the terminal and bridging protons of tetrahydroborate ligands distinguished in the NMR spectra, not even at low temperatures (for some compounds, studies were performed at*150 K).3 The activation energy for the exchange of terminal and bridging protons in these compounds equals *5 kcal mol71, as estimated for Cp3U. .(Z3-BH4).3 For compounds of Group IB metals, separate signals for bridging and terminal protons have not been identified either.226 Only recently, was the first titanium compound reported {the complex [Ti(CO)4(Z3-BH4)]7[K(18C6)]+} for which the limit of slow proton exchange was attained at 178K and individ- ual signals for the terminal and bridging protons of Z3-BH4 were observed.65 The fact that the tetrahedral geometry of the BH¡ anion in this complex is virtually non-distorted indicates a low contribution of the covalent constituent to the interaction of the central atom with this ligand. Note that in the analogous chro- mium and molybdenum complexes [M(CO)4(Z2-BH4)]7, the tetrahydroborate ligand has a bidentate coordination.The unique structural and dynamic properties of the titanium complex (this is the only example of a tridentate tetrahydroborate group for which individual signals from the terminal and bridging protons were observed) are apparently due to the combination of the carbonyl groups (active p-acceptors and the most `powerful' strong-field ligands) with an `early' 3d metal in a low oxidation state in one molecule. Many tetrahydroborates of Group V± VIII metals are much less fluxional than complexes formed by Group III and IV transition metals (see Table 2).At low temperature, individual signals of the terminal and bridging protons for both Z2- and Z1-BH4 groups can be distinguished in the NMR spectra of these compounds. It is of interest that in some cases, bridging protons of the Z2-BH4 groups account for two separate signals and the exchange of these non-equivalent bridging protons with terminal protons occurs at different rates.171 In some complexes, this proton exchange does not proceed at a noticeable rate even at room temperature, or at higher temperatures. Thus for one of the least fluxional tetrahydroborates Os(Z2-BH4)H3(PPri3)2, this process cannot be observed even at 360 K.The activation energy for the exchange of the terminal and bridging protons in this compound cannot be determined experimentally because this compound decomposes at higher temperatures. According to estimates,10 this energy is greater than 20 kcal mol71. At 300 K, the bridging protons in this compound undergo fast exchange with the hydride ligands (Os)H3 (activation energy *13 kcal mol71). The results of experimental studies and quantum- chemical calculations for Os(Z2-BH4)H3(PPri3)2 provide grounds for suggesting that the exchange of bridging hydrogen atoms (B)Hb with the hydride ligands (Os)Ht without participation of the terminal hydrogen atoms (B)Ht proceeds via a transition state in which the metal is coordinated to the B7H bond.10 The data presented above demonstrate that the existing view about the exceptionally high fluxionality of transition metal tetrahydroborate complexes needs to be corrected.Individual signals for the terminal and bridging protons cannot be observed (with only one exception 65) only for compounds of Group I, III and IV transition metals. Among Group V± VIII transition metal compounds, many complexes do give rise to separate NMR signals for the terminal and bridging protons of the BH4 ligands. The fluxionality decreases with an increase in the size of theStructural and dynamic properties of tetrahydroborate complexes stabilising ligands. This has been shown in relation to the iron complex FeH(BH4)L2 (L=DMPE, DEPE, DPrPE) and a niobium complex with the Cp and Cp* ligands, respectively.123, 169 The results of analysis of the activation energy for the exchange of bridging and terminal protons suggest that the fluxionality of the complexes decreases also on passing from Group V metals to Group VIII metals and from 3d metals to 4d or 5d metals (see Table 2).Thus, the decrease in the fluxionality of tetrahydrobo- rates occurs in the same direction as the decrease in their thermal stability mentioned above. The correlation observed between these properties might be due to the fact that a decrease in the fluxionality is indicative of an increase in the degree of covalency of the M7Hb bond, which increases the probability of decom- position of tetrahydroborate complexes by the oxidative addition mechanism (MBH4?HMBH3).IV. Possible reasons for the differences in the structural and dynamic properties of tetrahydroborate complexes 4 What is the reason for the so sharp differences in the types of ligand coordination and in the structural and dynamic properties of the complexes formed by Group III and IV transition metals and those formed by Group I, V± VIII transition metals? The attempts to rationalise the structure of tetrahydroborate com- plexes in terms of the Coulomb interaction between the BH¡ anions and metal cations (with the assumption that enhancement of the Coulomb forces should increase the degree of covalence of the metal7boron bond and the denticity of the BH4 ligand 227) do not provide satisfactory results.The attractive force between the anion and the central atom in many complexes of Group I, V± VIII transition metals should be greater than that in Group III and IV metal complexes because the ratio of the ion charge to the ionic radius squared (z/r2) for, for example, Fe(II), Co(II) and Ni(II) is greater than this ratio for U(III) or cerium subgroup lanthanides. Nevertheless, the denticity of the BH4 ligand is lower in the former type of complexes. Since the oxidation number of the central atoms in most of the Group I, V± VIII transition metal tetrahydroborates is +2, while that in the complexes of Group III and IV transition metals is +3 or +4, one can suggest that the difference in the ligand denticity is related to the oxidation state of the central atom.However, this suggestion is at variance with the existence of a number of bidentate complexes of Group V± VIII transition metals contain- ing the central atom in the +3 and +4 oxidation state [for Ir(Z2-BH4)H2(PBut example, 3)2, WH3 (Z2-BH4)(PMe3)3 , OsH3(Z2-BH4)(PPri3)2]; conversely, complexes of divalent ytterbium contain tetrahydroborate groups coordinated in the tridentate manner (see Table 1). The most widespread hypothesis states that the tetrahydrobo- rate ligand in the complexes is linked to the central atom via covalent two-electron three-centre M7H7B bonds and is iso- lobal to the chloride, allyl or cyclopentadienyl anion, i.e. it acts as a two-, four- or six-electron donor, depending on the denticity.According to this hypothesis, the type of coordination (Zn) of the BH4 ligand corresponds to the number n of electron pairs provided by this ligand and the ligand denticity is determined by the tendency of the central atom to fill the outer electron shell and can be predicted using the effective atomic number (EAN) rule.228 ± 230 However, the attempts to apply this approach to a broad range of objects showed that predictions based on this hypothesis often do not comply with experimental data regarding both chemical and physical properties of tetrahydroborate complexes, or with the EAN rule. For example, in conformity with this hypothesis, Cu(Z2-BH4)(phen)(PPh3) and Cp2M(Z2-BH4)2 (M=Zr, Hf) should be 20-electron complexes, Y(Z2-BH4)(Z3- BH4)2(THF)3 should be 22-electron, M(Z3-BH4)4 (M=Zr, Hf) should be 24-electron, and [Zn(Z2-BH4)4]27 should even be a 26- 739 electron complex.231 This hypothesis cannot also interpret the change in the type of coordination in the pairs of isoelectronic complexes Sc(Z2-BH4)(Z3-BH4)2(THF)2, Y(Z3-BH4)3(THF)2 and (MeOCH2CH2C5H4)2Pr(Z3-BH4), (MeOCH2CH2C5H4)2Yb(Z2- BH4). The paramagnetism of the vanadium V(Z2-BH4)3(PMe3)2 (mef=2.5 mB),84 chromium (TMEDA)Cr(Z2-BH4)2 (mef= 4.8 mB),93 and cobalt Co(Z2-BH4)[MeC(CH2PPh2)3] (mef= 3.1 mB) 109 complexes considered in terms of this hypothesis does not agree with the denticity of the BH4 ligands found for these complexes. The Lewis acidity of the allegedly 18-electron complex Zn(Z2-BH4)2 (see Ref. 231) and the 24-electron complexes M(Z3-BH4)4 (M=Zr, Hf) cannot be explained either.232, 233 When the number of electrons is calculated in this way, deviations from the EAN rule are observed for many structurally charac- terised transition metal tetrahydroborates; the deviation increases with an increase in the number and denticity of the BH4 ligands in the complex, which indicates that this method overestimates the number of electrons donated by the tetrahydroborate ligand.It should also be noted that, in accordance with the views considered above,222 ± 230 the strength of the metal7tetrahydro- borate ligand bond should increase with an increase in the number of M7H `covalent' bonds, i.e. with an increase in the ligand denticity. This, in turn, should result in an increase in the stereo- chemical rigidity of complexes on passing from monodentate to bi- and then to tri-dentate coordination of BH4 groups.However, experimental results point to an opposite trend, namely, com- plexes with Z2- and even Z1-BH4 groups are stereochemically more rigid than complexes with Z3-BH4 ligands. Thus, the hypothesis that, depending on its denticity, the BH4 group is isolobal to chloro, allyl or cyclopentadienyl ligand cannot explain and predict correctly the chemical, physical or structural properties of transition metal tetrahydroborates. Data on a decrease in the fluxionality of complexes with Z3-BH4 groups with respect to the complexes with Z2-BH4 and Z1-BH4 groups point to a low contribution of covalent interaction to the bond of the metal with the BH¡4 anions, at least, for Group III and IV transition metals.This conclusion is in agreement with the results of calculations carried out for the L47nCu ±Hn±BH47n com- plexes (n=1, 2, 3), which showed a largely ionic character of the Cu7Hn7BH47n bond and electrostatic nature of the interaction between the coordinated copper atom and the BH¡4 anion with a relatively small contribution of the covalent bond.234 This con- clusion is also consistent with the results of a study of copper tetrahydroborate complexes by mass spectrometry with extrac- tion of ions from the solution (electro-spray mass spectrome- try).235, 236 The change in the type of coordination of the BH4 ligands in Group III and IV transition metal complexes is much better explained by the influence of purely steric properties of ligands if one assumes that these complexes tend to fill the coordination sphere of the central atom as fully as possible and that a tridentate coordination of the BH4 group to the metal atom, which implies the largest volume occupied by this group, is the most favourable. It is assumed that this type of coordination is always realised whenever it is allowed by the spatial conditions in the complex and that transition to other types of coordination is induced by the influence of steric factors. On this basis, a procedure for the calculation of the degree of occupancy of the metal coordination space was developed and convincing examples confirming the efficiency and predictive capacity of this approach were cited.237 { The trend to fill the coordination sphere of the central atom as fully as possible and the high coordination numbers typical of tetrahydroborates of Group III and IV transition metals attest to a high degree of ionicity of theM7BH4 bond, which is consistent with the substantial fluxionality of these compounds.Nevertheless, several examples demonstrate that the scope of this approach is also limited. It is mainly applicable to Group III { Unfortunately, only few steric constants of ligands are given in the study cited,237 which narrows down the scope of application of this method.740 and IV transition metal compounds, although with exceptions.For example, only half of the BH4 groups in the [La(Z3-BH4)2. .(Z2-BH4)2(THF)2]7 anion are coordinated as Z3-ligands, despite the fact that the metal coordination sphere is filled only to 86%.18 This approach cannot be used to interpret the structure of complexes formed by transition metals of other groups. For example, in the [Mn(Z2-BH4)3(THF)]7 and [Zn(Z2-BH4)3]7 anions, the BH4 ligands have Z2-coordination,100, 101, 134 although the coordination spheres of the central atom are filled only to 86% and 75%, respectively. Thus, in the general case, the difference between the types of coordination of the BH4 ligands in transition metal compounds cannot be explained by either the influence of the oxidation state of the central atom, or the different electrostatic interaction of the metal with the ligands, or by the `tendency' of the central atom to fill the outer electron shell, or by the influence of spatial factors.Analysis of the data presented in Tables 1, 2 suggests that the main factor determining the differences between the structures and dynamics of the tetrahydroborate complexes of Group III and IV transition metals, on the one hand, and Group I, V± VIII metals, on the other hand, is the difference between the electronic configurations of the central atoms. The electronic configuration of Group III and IV transition metals in compounds with tridentate tetrahydroborate ligands (except for the complex of zerovalent titanium with carbonyl groups 65) is d 0 (or d 1 in the case of complexes of trivalent titanium). In the complexes of Group I, V± VIII transition metals, the central atom with the d n-configuration (n = 2 ± 10) has at least two electrons not involved in the formation of the chemical bonds in the outer d shell.The fact that no examples of structurally characterised complexes of these metals with tridentate BH4 ligands are available suggests that the presence of two or more electrons in the outer d shell destabilises the Z3-coordination and, therefore, the Z2-coordination becomes more favourable from the energy standpoint. Thus, whereas the bidentate coordination is sterically forced in complexes with central atoms having the d 0- and d 1-configuration, in the case of ions with d n-configurations (n = 2 ± 10), it is the Z2-coordination that is realised without steric restrictions.The increase in the stereochemical rigidity of complexes on passing to compounds of Group V± VIII metals can be accounted for by the presence of electrons in the outer d shell which do not participate in the formation of chemical bonds and decrease the energetic profitability of the most probable rearrangement path- way Z2 ± Z3 ± Z2 0.57 Indeed, quantum-chemical calculations 10 showed that Z2 ± Z1 ± Z2 0 is an energetically more favourable pathway for the exchange of the bridging and terminal protons of the BH4 group in the complex Os(Z2-BH4)H3(PPri3)2. In conformity with the above-noted crucial influence of the electronic configuration of the central atom on the structural and dynamic properties of the tetrahydroborate ligands in the com- plex, derivatives of titanium group metals with the oxidation number +2 should contain bidentate BH4 ligands. In fact, no examples of Ti(II) compounds with tridentate BH4 groups are available; in the d 2-complexes, Cp*Ti(Z2-BH4)[(Me2PCH2)3..SiBut] and Ti(Z2-BH4)2(DMPE)2, the BH4 groups are coordi- nated as bidentate ligands.54, 55 However, in the d 0-complex Yb(Z3-BH4)2(Py)4, these ligands are tridentate,22 despite the substantially greater ionic radius of the central atom, which decreases the Coulomb interaction. In this connection, we would like to mention the well-known influence of lone electron pairs on the structure of coordination compounds explained by repulsion of the coordinated ligands from non-bonding electrons.238 It was noted that, when the number of non-bonding d electrons is 1 to 3, their effect on the structural properties of molecules is normally too low to be detected experimentally.However, the examples considered above demonstrate that the type of coordination and the dynamic behaviour of the BH4 ligand change crucially even in the presence of two non-bonding outer d electrons. V D Makhaev As any other corollary based on the generalisation of exper- imental data, this assumption leaves room for exceptions the discovery and analysis of which would result in further develop- ment of the views on the structures of the compounds considered. To verify the assumption concerning the reasons for the differ- ences in the modes of coordination and dynamic properties ofBH4 ligands, it is necessary, in particular, to study complexes of Group IV transition metals with the central atom in a low oxidation state and to develop methods for the synthesis and study of compounds of Group V± VIII metals in high oxidation states.V. Complexes of post-transition metals with terminal BH4 ligands The assumption that metal d electrons destabilise the Z3-coordi- nation of the BH4 groups more significantly than the Z2-coordi- nation relied on analysis of the results of quantum-chemical calculations for complexes with central d 10 ions.239, 240 It was suggested that this destabilisation should decrease the difference between the energies of Z2- and Z3-coordinated groups and thus enhance the fluxionality of tetrahydroborate complexes.Indeed, no data about non-equivalence of the terminal and bridging protons in the tetrahydroborate complexes with d 10-configura- tion of the central atom can be found in the literature. Moreover, experimental data point to specific properties of the complexes in which the outer d shell of the metal ions is either half or completely filled (d 5 or d 10). Thus Mn(II), Cu(I) and Zn(II) complexes are much more numerous and thermally stable than complexes of other Group V± VIII transition metals or post-transition metals. It could be suggested that the presence of a filled d shell influences not only the thermal stability of the complex but also the mode of coordination of the BH4 groups.However, it can be seen that both post-transition metals (Zn, Cd, Hg, Ga, In) having a closed outer d 10 shell and Group I and V± VIII transition metals having d n-configurations (n=2 ± 10) tend to be coordinated by BH4 in a bidentate fashion even in those cases where steric factors are favourable for Z3-coordination (see Table 1). Indeed, the complexes Zn(Z2-BH4)2 and [Zn(Z2-BH4)3]7 contain bidentate BH4 ligands, despite the fact that the metal coordination sphere is not 100% full {in [Zn(Z2-BH4)3]7, it is only 75% full}. Appa- rently, even a closed outer d-shell destabilises Z3-coordination, although, due to its higher symmetry, a less pronounced destabi- lisation of the complex should be expected in this case.The paucity of post-transition metal complexes the structures of which are known can be explained by their low thermal stability (see Table 1). The complex Cp(CO)2FeGa(Z2-BH4)[(CH2)3NMe2] is the only example, among gallium tetrahydroborate complexes with terminal BH4 groups, the structure of which has been established by X-ray diffraction analysis.143 The structural data for other gallium complexes presented in Table 1 have been obtained by gas phase electron diffraction analysis. No examples of structurally characterised tetrahydroborate complexes of mer- cury, indium or thallium can be found in the literature. However, the conclusion about bidentate coordination of the BH4 ligand in Me2In(Z2-BH4) has been drawn on the basis of IR-spectroscopic data.144 The IR spectra of the tetrahydroborate complexes of post- transition metals reported in the literature generally comply, in the region of stretching B7H vibrations, with those expected for bidentate BH4 groups (see Table 1).The synthesis of tetravalent germanium and tin tetrahydro- borates M(Z3-BH4)4 (M=Ge, Sn) has been reported;241 they were obtained as solid compounds stable in air up to 473K and containing tridentate (according to IR spectroscopy) tetrahydro- borate ligands. An estimate of the degree of filling of the coordination sphere in these compounds, which we performed using published data,223, 237 showed that in the case of Ge(IV), Z3-coordination of the four BH4 groups is unlikely because the coordination sphere of the central atom is overcrowded, while the Sn(IV) coordination sphere in Sn(Z3-BH4)4 should be filled to 100%.Covalent tetrahydroborates with filled coordinationStructural and dynamic properties of tetrahydroborate complexes spheres of the metal atoms should be volatile compounds unstable in air, like Zr(Z3-BH4)4 . Therefore, the properties described for tetravalent germanium and tin tetrahydroborates, and even the mere fact of existence of these compounds, appear doubtful. It is noteworthy that reports concerning the syntheses of unusual tetrahydroborate complexes of transition metals, such as Cp3CeBH4 242 and Cp2MO(Z2-BH4)2 (M=Mo, W),243 have been published and repeatedly cited, although subsequent attempts to reproduce them failed.244, 245 VI.Complexes of Group IA and IIA metals and aluminium with terminal BH4 ligands The structures of alkali and alkaline earth tetrahydroborate complexes are little studied. Among alkali metal complexes, lithium compounds have been studied most comprehensively. Table 1 presents several examples of complexes of univalent lithium with Z2- and Z3-coordinated BH4 ligands. Antsyshkina et al.146 believe that the complex LiBH4(18C6) in the crystalline state contains a Z1-coordinated BH4 group. The same structure was proposed 146 for NaBH4(15C5).150 However, the conclusions about the monodentate coordination of the tetrahydroborate ligands are inconsistent with the experimental data on the M_B distances in these complexes, which are close to the value expected for Z2-BH4.223 Despite the fact that no data on the structures of tetrahydroborates of heavier metals (K, Rb, Cs) are available to date, it can be suggested that these metals, like lithium, form compounds with all the main types of coordination of BH4.The difference between the structures of Li(Z3-BH4)[HC(3,5-Me2pz)3] and Cd(Z2-BH4)[HB(3,5-Me2pz)3], noted by Reger et al.,145 namely, the fact that theBH4 group is coordinated in the bidentate fashion to the Cd(II) ion having a greater size and a higher charge, whereas it is coordinated as a tridentate ligand to the Li(I) ion, can be understood in view of the assumption about destabilising influence of outer d electrons. The similarity of the structural and spectral properties of TlBH4 to those of alkali metal tetrahydroborates 1, 2 is also consistent with this assumption because the closed 5d 10 shell of the Tl(I) ion is shielded by the filled 6s2 shell, which diminishes the destabilising influence. The type of BH4 coordination in CpBe(Z2-BH4) could not be established unambiguously from electron diffraction data.The conclusion about the bidentate coordination of this ligand to the beryllium atom was drawn resorting to the data of IR spectro- scopy.151, 152 The denticity of BH4 ligands in alkaline earth metal complexes (Ca, Sr, Ba) corresponds to that expected on the basis of steric considerations D BH4 groups accomplish tridentate coordination whenever the spatial possibility exists.No structural data pointing to the possibility of Z3-coordination of BH4 groups in magnesium or aluminium complexes are currently available. In this connection, recall that the ionic radii of Al(III) and, especially, Be(II) are much smaller than those of transition metals.222 The type of coordination of tetrahydroborate ligands in the metal complexes in question cannot be determined from the data of NMR spectroscopy; these data attest to equivalence of all the four protons in the BH4 ligand.145, 147, 148, 157, 161 The data of IR spectroscopy for alkali and alkaline earth metal complexes do not conform to any particular type of coordination of the tetrahy- droborate groups either.145, 147, 148, 157 The structures of the tetrahydroborate complexes with the central d 0-metal ions in the oxidation state +1 as well as complexes with the d n-metal central ions (n=2 ¡À 10) cannot be explained in terms of the principle of the maximum occupancy of the coordination sphere.Calculations using this principle 237 provide the best results in the case of complexes with d 0(d 1)- metal central ions in the oxidation states +3 or +4 and with ionic radii ranging from 0.7 to *1A. Calculations performed in conformity with the data reported by Edelstein 223 show that tridentate coordination of BH4 groups in tetrahydroborates of metals with ionic radii of less than 0.68A (the difference between 741 4 the radii of the ionic and tridentate forms ofBH4) requires that the metal7boron distances be smaller than the radius of the BH¡¦ anion. Therefore, the deviations from the principle of the max- imum occupancy of the coordination sphere observed for these ions can be attributed to the mutual repulsion of ligands.For d 0-metal ions in the oxidation state +2, this principle can be used to estimate the probable denticity of theBH4 ligands; however, the maximum occupancy of the metal coordination sphere is seldom encountered. Apparently, under normal conditions, the number of neutral ligands (for example, ethers) these ions can hold is insufficient to fill completely their coordination sphere. Thus, the scope of application of the principle of the maximum occupancy of the coordination sphere of the central atom for predicting the structures of tetrahydroborate complexes is mainly limited to complexes with central d 0-metal ions in the oxidation states +3 and +4.VII. Tetrahydroborate complexes with bridging BH4 ligands The structurally characterised compounds with bridging tetra- hydroborate groups are presented in Table 3. Comparison of the data of Tables 1 and 3 shows that alkali metal complexes with bridging BH4 ligands are as prevalent as complexes with terminal tetrahydroborate ligands. On passing to alkali earth metals, lanthanides and actinides, the proportion of compounds with bridging BH4 groups in the whole set of structurally characterised compounds gradually decreases; no structurally characterised bi- or polynuclear complexes of Group IV and V transition metals with tetrahydroborate bridges between the metal atoms are known.The trend to form complexes with bridging BH4 groups then starts increasing again and, for Group VIII metals, the fractions of complexes with known structures containing terminal and bridgingBH4 groups are nearly equal. These data suggest that the formation of bridged compounds is facilitated by a decrease in the charge and an increase in the radius of the complex-forming ion.The most frequently encountered polynuclear complexes (14 of 39, see Table 3) are those in which two metal atoms are bound in the Z2-fashion to the opposite edges of a BH4 tetrahedron (see Fig. 2 a). None of the tetrahedron vertices belong simultaneously to both metal atoms.Each hydrogen forms m2-bridges between boron and the metal, i.e. the BH4 groups are generally tetraden- tate. Depending on the oxidation state of the metal, the number of BH4 tetrahedra, the nature of the stabilising ligands and crystal- lisation conditions, this type of coordination of bridging groups results in the formation of dimers,189, 212, 213, 216 infinite polymer chains,160, 200, 202, 217 a two-dimensional network,210 or a three- dimensional network.147, 207, 208 Apparently, three-dimensional structures of this type are also formed in some non-solvated compoundsMn(BH4)n (`ionic' tetrahydroborates). In a less abundant (six complexes) type of coordination, the BH4 group is bound by two edges to two metal atoms and the edges participating in the coordination share a vertex (m3-H).One hydrogen remains free from binding to the metal atoms; thus, the BH4 groups are in total tridentate (see Fig. 2 b). This pattern of coordination results most often in the formation of dimeric complexes 145, 148, 198, 214 and in some cases, polymer chains.199 There is only one compound (LiBH4 . Et2O) in which the BH4 group is coordinated to three metal atoms (to each of them, by an edge); these three edges share a vertex (m4-H, see Fig. 2 c).147, 197 In the crystals of LiBH4(MeOBut), the BH4 group is also coordi- nated by its edges to three metal atoms; however, there is no vertex common to these three edges (see Fig. 2 d ). In these cases (see Fig. 2 c,d ), all the four hydrogen atoms are involved in the formation of bridging bonds, i.e.the BH4 group is tetradentate. The Z1,Z2-bridging coordination is represented by two exam- ples,Mn2(CO)5(DPPM)H(BH4) andGaH2(BH4); in this case, one metal atom is coordinated to an edge of the tetrahedron, while the other one is bound to the opposing vertex (see Fig. 2 e).174, 221 The742 Z1,Z1-coordination according to which each of the two metal atoms is bound to one tetrahedron vertex (see Fig. 2 f) is repre- sented by six examples.93, 173, 178, 196, 221 This gives rise to a chain or a ring with the M7H7BH27H7M sequence of bonds; two hydrogen atoms remain in the terminal positions. The complex GaH2(BH4) in the crystalline state forms a polymer chain consisting of alternating BH4 tetrahedra with the types of coordination shown in Fig.2 e, f, so that the coordination numbers of the gallium ions are 4 and 6.221 There exists one example, HFe3(CO)9(BH4), of Z1,Z1,Z1- coordination of the BH4 ligand where three metal atoms of the cluster are bound to three vertices of the BH4 tetrahedron (three m2-H bonds; the overall denticity of the tetrahydroborate ligand is Z3, see Fig. 2 g).177 Conversely, in the Na4B4 heterocubane core of the [NaBH4(TMTCN)]4 tetramer, each three neighbouring sod- ium atoms are linked to the same vertex of the BH4 tetrahedron (m4-H bond; the total denticity of theBH4 group is Z1, Fig. 2 h).148 Among mercury complexes, two compounds, [cyclo-(o- C6F4Hg)3] . (BH4)2(Bu4N)2 and [cyclo-(o-C6F4Hg)3]2(BH4)..(Bu4N), have been isolated in an individual state.219 Their structures have not been determined. Relying on the results of quantum-chemical calculations, it was assumed that the BH4 ligand in [cyclo(o-C6F4Hg)3]2 . (BH4)(Bu4N) is linked to one mercury macrocycle as shown in Fig. 2 g and to one more macro- cycle as shown in Fig. 2 h,220 i.e. this group is supposed to be connected simultaneously to six metal atoms. A type of coordination in polynuclear complexes represented rather widely (six structures) is the coordination according to which one or two metal atoms are bound to a face of the BH4 tetrahedron, which corresponds to Z3-coordination in mononu- clear complexes (see Fig. 2 i , total denticity Z4, and Fig. 2 j, total denticity Z3).However, as in the case of mononuclear complexes, this coordination pattern is encountered only for metal ions having no outer d electrons, i.e. sodium,148 strontium(II), barium(- II),157 cerium(III),203 samarium(III) 204 and neodymium(III),205 whereas no examples of face coordination (Z3) are known for d n ions (n=2 ± 10) (see Table 3). In the NMR spectra of complexes of Group I, II, III and IV metals with BH4 bridges, separate signals for the terminal and bridging protons have not been detected. The non-equivalence of protons of the bridging BH¡4 ligands has been observed in the NMR spectra of complexes of Group VII ± VIII transition met- als.173, 174, 177, 178, 188, 196 Thus, the structure and the dynamic behaviour of polynuclear compounds with BH4 bridging groups also does not contradict the assumptions made in this study.The data presented in Table 3 demonstrate a fundamental difference between the complexes with central d 0-metal ions and the complexes with d n transition metal ions (n = 2 ± 10), namely, in the case of alkali and alkaline earth metals, lanthanides and uranium, the tetrahydroborate bridges connect neutral mononu- clear molecular units, which results in a fuller occupancy of the metal coordination sphere and corresponds to the non-direction- ality and non-saturability of the mainly ionic M7BH4 bond in these compounds. In the case of Group I and V± VIII transition metals, the tetrahydroborate bridges are formed between cationic species or, in most cases, between metal atoms bound into a cluster and located close to each other.The example of the cobalt complex [Co(DPPP)(BH4)]2 seemingly contradicts this conclu- sion. However, the low magnetic moment of this complex points to the presence of a metal7metal bond.214 Hence, in this case, too, the formation of the tetrahydroborate bridges is favoured by the presence of a bond between transition metal atoms. Among post- transition metal complexes, there are compounds with polymer chains consisting of neutral molecules connected via tetrahydro- borate bridges, for example, [MeZnBH4]?217, 218 and [GaH2(BH4)]?.221 The fluxionality, Z3-coordination and the trend for the maximum occupancy of the coordination sphere are as typical of lanthanide and actinide complexes as of complexes formed by d metals with d 0 central ions.Therefore, it can be concluded that f V D Makhaev electrons do not influence the properties of tetrahydroborate complexes as substantially as outer d electrons. It has been suggested 5, 207 that f orbitals participate in the formation of the bonds of the central atom with the BH4 groups. However, in our opinion, the available data do not provide grounds for this suggestion. In the structures which have now been determined, one to three metal ions are connected to one BH4 group via its vertices or edges, or one or two metal ions are coordinated to this BH4 group via faces. The metal ions in one structure can be coordinated both in the same way and in different ways; however, not all the possible combinations of various types of coordination have been con- firmed experimentally.Thus a complex in which three metal ions are bound to the same vertex of the BH4 tetrahedron in the monodentate fashion is known but there are no examples of complexes in which two or more metal ions are coordinated in the bidentate fashion to the same edge or in the tridentate fashion to the same face of the tetrahedron. Examples of structurally characterised complexes in which four or more metal ions are linked to one BH4 group or examples of heteronuclear complexes in which this group serves as a bridge between ions of different metals cannot be given either. In the majority of polynuclear complexes (*60% of the total number), all the four hydrogen atoms of theBH4 groups are involved in coordination, i.e.they are formally tetradentate ligands (see Fig. 2 a,c,d,j). About one- fourth of the complexes studied (see Fig. 2 b,e,g) contain BH4 ligands with one non-coordinated hydrogen atom. There are only a few polynuclear complexes with tetrahydroborate bridges formed by one (see Fig. 2 h) or two (see Fig. 2 f ) hydrogen atoms. Unlike compounds with terminal tetrahydroborate ligands, for bridging complexes, no correlation has been found between the data of IR spectroscopy and the types of coordination of bridging groups. It has been noted 198 that the hypothesis of three-centre two- electron type of theM7H7B bond is inapplicable for calculation of the EAN in the compounds with bridging BH4 groups.VIII. Conclusion Analysis of experimental data on the structure of tetrahydrobo- rate complexes makes it possible to elucidate a series of correla- tions between the structural or dynamic properties of these compounds and the nature of the central atom and stabilising ligands. Compounds of Group III and IV transition metals (lantha- nides, U, Ti, Zr, Hf) have been studied most comprehensively. In addition to the diversity of stabilising ligands, these complexes are characterised by diverse modes of coordination of tetrahydrobo- rate ligands (Z1-, Z2- and Z3-BH4, various types of BH4 bridges between neutral species); complexes with BH4 ligands of different denticities at one metal atom are also known. Most of complexes containing these metals in the oxidation states+3and+4comply with the principle of the maximum occupancy of the metal coordination sphere.The known tetrahydroborates of these metals possess clear-cut fluxionality (the protons of the BH4 group are equivalent on the NMR time scale). Only for the complex [Ti(CO)4(Z3-BH4)]7[K(18C6]+, were the signals of the terminal and bridging protons of Z3-BH4 distinguished in low- temperature NMR spectra. The structures of alkali and alkaline earth metal tetrahydro- borates have not been adequately studied. As indicated by the available data, characteristic features of these compounds include fluxionality, diversity of the modes of coordinations of tetra- hydroborate ligands to metals, formation of BH4 bridges between neutral molecular units; the principle of the maximum occupancy of the coordination sphere is often violated.The structural chemistry of Group I and V± VIII transition metal tetrahydroborates is much less plentiful than that for the Group III and IV transition metals; the range of stabilising ligands is more limited; no complexes with Z3-BH4 groups or with BH4Structural and dynamic properties of tetrahydroborate complexes bridges between neutral molecular units have been found. The principle of the maximum occupancy of the coordination sphere is not obeyed for these compounds. In the complexes of Group I and V¡À VIII transition metals, BH4 groups tend to be coordinated as mono- and bidentate ligands.Monodentate coordination is found only when the coordination sphere contains bulky ligands having strong elec- tron-donating properties (phosphines, macrocyclic compounds), i.e. it is a forced type of coordination. The factors inducing the formation of complexes of these metals with mono- and bidentate coordination of theBH4 ligands are not entirely clear yet. Whereas alkali, alkaline earth and rare earth metals and uranium tend to form polynuclear complexes due to tetrahydroborate bridges between neutral mononuclear molecules, in the case of Group I and V¡À VIII metals, the BH4 bridges connect either cationic species or metal atoms bound into a cluster and located close to one another. Tetrahydroborates of Group V¡À VIII transition metals pos- sess higher stereochemical rigidities than complexes of Group IA, IIA, IIIB and IVB metals. For many of them, separate signals for terminal and bridging protons of mono- and bidentate BH4 groups can be detected in low-temperature NMR spectra. In some of the compounds considered, the exchange of terminal and bridging protons does not proceed at a noticeable rate even at room or higher temperature. In some cases, bridging protons give rise to two separate NMR signals and the exchange of non- equivalent bridging protons with terminal protons occurs at different rates.The absence of a trend for the maximum occupancy of the coordination sphere, the low coordination numbers and an increase in the stereochemical rigidity indicate that the degree of covalence of the M7BH4 bond in Group V¡À VIII metal tetrahy- droborates is higher than that in the Group IA, IIA, IIIB and IVB metal complexes.Neither resorting to the views on the three-centre two-electron M7H7B bond, widely employed now, nor allowing for steric factors can substantiate in the general case the observed structures and some physical and chemical properties of tetrahydroborate complexes. Analysis of the experimental data presented above led to the assumption that the crucial factor determining the differences in the structures and the dynamic properties of transition metal tetrahydroborate complexes is the electron configuration of their central ions, which is d 0 (d 1) for most of the Group III and IV metal compounds considered or d n (n=2 ¡À 10) for Group I and V¡À VIII metal complexes.High coordination numbers, the trend for the maximum occupancy of the coordination sphere and fluxionality of the tetrahydroborates of Group IA, IIA, IIIB and IVB metals indicate that the M7BH4 bonds in these compounds are mostly ionic. The fluxional behaviour of these compounds is apparently related to the fact that the spherically symmetrical electron shell of d 0 ions does not create barriers to free rotation of the BH¡¦4 anion in the metal coordination sphere. In the complexes with central d n ions (n = 2 ¡À 8), the mutual repulsion of the non- bonding d-electrons and coordinated BH¡¦4 anions destabilises the Z3-coordination of the BH4 groups, which makes the most probable rearrangement pathway (Z2 ¡ÀZ3 ¡À Z2 0) energetically less favourable and results in an increase of the stereochemical rigidity of complexes of Group V¡À VIII metals.In addition, this factor prevents the formation of tetrahydroborate bridges between neutral mononuclear molecules in d n-metal tetrahydro- borates (n=2 ¡À 8). Complexes with central d 10 ions, possessing completely filled d-electron shells, occupy an intermediate posi- tion D there are no complexes with Z3-groups among them but stereochemically rigid complexes have not been found either. The similarity of the structural and dynamic properties of tetraborate complexes formed by f elements and those with central d 0 ions (Z3-coordination, fluxional behaviour and the trend for maximum occupancy of the metal coordination sphere) leads to the conclusion that f electrons do not influence the properties of 743 tetrahydroborate complexes as significantly as outer d electrons and they apparently do not participate in the formation of bonds of the central atom with the BH4 ligand.The data considered above provide information not only on the dependence of the structural and dynamic properties of the tetrahydroborate ligand on the electronic configuration of the central atom but also on the dependence of the coordination and, correspondingly, catalytic properties of the central atom on the structure of its outer electron shell. Since the tetrahydroborate anion is isoelectronic with the methane molecule, tetrahydroborates can be regarded as models of some steps of C7H bond activation in saturated hydrocar- bons.3 It is noteworthy in this connection that the sharp difference between the coordination (and, hence, catalytic) properties of the complexes with central d 0 and d n ions (n = 2 ¡À 10) is also manifested in many systems related to the transformations of hydrocarbon ligands.Thus activation of methane and cyclo- metallation reactions with participation of d 0-ion complexes follow a mechanism which includes a multicentre transition state,246 whereas activation of C7H bonds involving complexes with central d n ions often proceeds according to an oxidative addition mechanism, similar to the exchange of hydride ligands and bridging hydrogen atoms in OsH3(BH4)(PPri3)2 .10 It is also known that transition metal metallocene complexes with central d 0 ions act, under certain conditions, as active catalysts for polymerisation of a-olefins according to the so-called metallocene catalysis mechanism.247 Similar complexes with d n ions do not exhibit pronounced catalytic activity under the same conditions; the catalysts of a-olefin polymerisation based on d n ions have a different composition, `work' under different con- ditions and by a different mechanism.248 By analogy with the BH4 ligand, hydrocarbon ligands might possess different stereochem- ical rigidities and different denticities in the coordination with d 0 or d n ions.Further studies of structural and dynamic properties of the tetrahydroborate complexes with central d 0 and d n ions are needed to understand the nature of the metal7ligand bond in these complexes.They are also required for development of the general views on the fluxional behaviour, for more extensive investigations of the dependence of the coordination and catalytic properties of metals on the electronic structures of their atoms and for modelling of the transition states in hydrocarbon activation processes. References 1. A F Zhigach, D S Stasinevich Khimiya Gidridov (The Chemistry of Hydrides) (Leningrad: Khimiya, 1969) 2. B D James, M G H Wallbridge Progr. Inorg. Chem. 11 99 (1970) 3. T J Marks, J R Kolb Chem. Rev. 77 263 (1977) 4. K B Gilbert, S K Boocock, S G Shore, in Comprehensive Organometallic Chemistry Vol.6 (Eds G Wilkinson, F G A Stone, E N Abel) (Oxford: Pergamon Press, 1982) p. 879 5. T N Dymova, U Mirsaidov Borogidridy Perekhodnykh Metallov (Borohydrides of Transition Metals) (Dushanbe: Donish, 1985) 6. A Lledos,MDuran, Y Jean, F Volatron Inorg. Chem. 30 4440 (1991). 7. F Volatron,M Duran, A Lledos, Y Jean Inorg. Chem. 32 951 (1993) 8. D G Musaev, K Morokuma Organometallics 14 3327 (1995) 9. Y Oishi, T A Albright, H Fujimoto Polyhedron 14 2603 (1995) 10. I Demachy,MA Esteruelas, Y Jean, A Lledos, F Maceras, L A Oro, C Valero, F Volatron J. Am. Chem. Soc. 118 8388 (1996) 11. E B Lobkovskii, S E Kravchenko, K N Semenenko Zh. Strukt. Khim. 18 389 (1977) a 12. O V Kravchenko, S E Kravchenko, V B Polyakova, K N Semenenko Koord.Khim. 6 76 (1980) b 13. M F Lappert, A Singh, J L Atwood, W E Hunter J. Chem. Soc., Chem. Commun. 206 (1983) 14. G Segal, S J Lippard Inorg. Chem. 17 844 (1978) 15. E B Lobkovskii, S E Kravchenko, O V Kravchenko Zh. Strukt. Khim. 23 (4) 111 (1982) a744 16. D A Laske, R Duchateau, J H Teuben, A L Spek J. Organomet. Chem. 462 149 (1993) 17. H Schumann, M R Keitsch, J Demtschuk, S H Muhle Z. Anorg. Allg. Chem. 624 1811 (1998) 18. V K Bel'skii, A N Sobolev, B M Bulychev, T Kh Alikhanova, A Kurbonbekov, U Mirsaidov Koord. Khim. 16 1693 (1990) b 19. L Deng, X F Zheng, C T Qian, J Sun, L Zhang J. Organomet. Chem. 466 95 (1994) 20. H Schumann,M R Keitsch, S H Muhle Acta Crystallogr., Sect. C 56 48 (2000) 21.A V Khvostov, V V Nesterov, BMBulychev, A I Sizov, MYu Antipin J. Organomet. Chem. 589 222 (1999) 22. J P White, H B Deng, S G Shore Inorg. Chem. 30 2337 (1991) 23. H J Wasserman, D C Moody, R R Ryan J. Chem. Soc., Chem. Commun. 532 (1984) 24. H J Wasserman, D C Moody, R T Paine, R R Ryan, K V Salazar J. Chem. Soc., Chem. Commun. 533 (1984) 25. D Baudry, E Bulot, P Charpin, MEphritikhine, MLance, MNierlich, J Vigner J. Organomet. Chem. 371 155 (1989) 26. D Baudry, P Charpin, M Ephritikhine,M Lance,M Nierlich, J Vigner J. Chem. Soc., Chem. Commun. 739 (1987) 27. R R Ryan, K V Salazar, N N Sauer, J M Ritchey Inorg. Chim. Acta 162 221 (1989) 28. T Arliguie, M Lance, M Nierlich, J Vigner, M Ephritikhine J. Chem. Soc., Chem.Commun. 847 (1994) 29. A Dejean, P Charpin, G Folcher, P Rigny, A Navaza, G Tsoucaris Polyhedron 6 189 (1987) 30. P Charpin, M Nierlich, D Vigner,MLance, D Baudry Acta Crystallogr., Sect. C 43 1630 (1987) 31. P Charpin, M Lance, E Soulie, D Vigner, H Marquet-Ellis Acta Crystallogr., Sect. C 41 1723 (1985) 32. P Charpin, M Nierlich, G Chevrier, D Vigner,M Lance, D Baudry Acta Crystallogr., Sect. C, 43 1255 (1987) 33. D Baudry, P Charpin, MEphritikhine, G Folcher, J Lambard, M Lance,M Nierlich, J Vigner J. Chem. Soc., Chem. Commun. 1553 (1985) 34. DBaudry, E Bulot, P Charpin,MEphritikhine,MLance,MNierlich, J Vigner J. Organomet. Chem. 371 163 (1989) 35. P Zanella, G DePaoli, G Bombieri, C Zanotti, R Rossi J. Organomet. Chem. 142 C21 (1977) 36.P C Blake, M F Lappert, R G Taylor, J L Atwood, W E Hunter, H Zhang J. Chem. Soc., Dalton Trans. 3335 (1995) 37. P Gradoz, D Baudry, M Ephritikhine, M Lance, M Nierlich, J Vigner J. Organomet.Chem. 466 107 (1994) 38. P Zanella, N Brianese, U Casellato, F Ossola,M Porchia, G Rossetto, R Graziani Inorg. Chim. Acta 144 129 (1988) 39. D Baudry,M Ephritikhine, F Nief, L Ricard, F Mathey Angew. Chem., Int. Ed. Engl. 29 1485 (1990) 40. J Rebizant,M R Spirlet, S Bettonville, J Goffart Acta Crystallogr., Sect. C 45 1509 (1989) 41. R Adam, C Villers,M Ephritikhine, MLance,M Nierlich, J Vigner New J. Chem. 17 455 (1993) 42. C Baudin, D Baudry,M Ephritikhine, M Lance, A Navaza, M Nierlich, J Vigner J. Organomet. Chem. 415 59 (1991) 43.D Baudry, E Bulot,M Ephritikhine, MNierlich, MLance, J Vigner J. Organomet. Chem. 388 279 (1990) 44. M Muller, V C Williams, L H Doerrer, M A Leech, S A Mason, M L H Green, K Prout Inorg. Chem. 37 1315 (1998) 45. P Scott, P B Hitchcock J. Chem. Soc., Dalton Trans. 603 (1995) 46. H W Turner, R A Andersen, A Zalkin Inorg. Chem. 18 1221 (1979) 47. R H Banks, NMEdelstein, B Spencey J. Am. Chem. Soc. 102 620 (1980) 48. R H Banks, N M Edelstein, R R Rietz, D H Templeton, A Zalkin J. Am. Chem. Soc. 100 1957 (1978) 49. C J Dain, A J Downs,M J Goode, D G Evans, K T Nicholls, D W H Rankin, H E Robertson J. Chem. Soc., Dalton Trans. 967 (1991) 50. E B Lobkovskii, V D Makhaev, A P Borisov, K N Semenenko Zh. Strukt. Khim. 27 (1) 173 (1986) a 51.A P Borisov, V D Makhaev, E B Lobkovskii, K N Semenenko Zh. Neorg. Khim. 31 86 (1986) c 52. J A Jensen, S R Wilson, G S Girolami J. Am. Chem. Soc. 110 4977 (1988) V D Makhaev 53. J A Jensen, S R Wilson, A J Schultz, G S Girolami J. Am. Chem. Soc. 109 8094 (1987) 54. J A Jensen, G S Girolami Inorg. Chem. 28 2107 (1989) 55. Y J You, S R Wilson, G S Girolami Organometallics 13 4655 (1994) 56. KMMelmed,DCoucouvanis, S J Lippard Inorg. Chem. 12 232 (1973) 57. T J Marks, W J Kennelly J. Am. Chem. Soc. 97 1439 (1975) 58. G I Mamaeva, I Hargittai, V P Spiridonov Inorg. Chim. Acta 25 L123 (1977) 59. K N Semenenko, E B Lobkovskii, A I Shumakov Zh. Strukt. Khim. 17 1073 (1976) a 60. D G Dick, R Duchateau, J J H Edema, S Gambarotta Inorg.Chem. 32 1959 (1993) 61. L Scoles, S Gambarotta Inorg. Chim. Acta 235 375 (1995) 62. H Noth,M Schmidt Organometallics 14 4601 (1995) 63. W A Herrmann,M Denk, W Scherer, F R Klingan J. Organomet. Chem. 444 C21 (1993) 64. F Corazza, C Floriani, A Chiesi-Villa, C Guastini Inorg. Chem. 30 145 (1991) 65. P J Fischer, V G Young Jr, J E Ellis Angew. Chem., Int. Ed. Engl. 39 189 (2000) 66. V Plato, K Hedberg Inorg. Chem. 10 590 (1971) 67. P H Bird, M R Churchill J. Chem. Soc., Chem. Commun. 403 (1967) 68. B E Smith, H F Shurwell, B D James J. Chem. Soc., Dalton Trans. 710 (1977) 69. B E Smith, B D James, J A Dilts J. Inorg. Nucl. Chem. 38 1973 (1976) 70. B Sztaray, E Rosta, Z Bocskey, L Szepes J. Organomet. Chem. 582 267 (1999) 71. R Choukroun, B Douziech, B Donnadieu Organometallics 16 5517 (1997) 72. F G N Cloke, P B Hitchcock, J B Love J.Chem. Soc., Dalton Trans. 25 (1995) 73. S C P Joseph, F G N Cloke, C J Cardin, P B Hitchcock Organometallics 14 3566 (1995) 74. A Kazani, S Gambarotta, C Bensimon Can. J. Chem. 75 1494 (1997) 75. S C Mayo,M Brown, V K Lloyd Acta Crystallogr., Sect. C 50 367 (1994) 76. J E Gozum, S R Wilson, G S Girolami J. Am. Chem. Soc. 114 9483 (1992) 77. J E Gozum, G S Girolami J. Am. Chem. Soc. 113 3829 (1991) 78. D Coucouvanis, A Hadjikiriakou, R Lester, G G Kanatzidis Inorg. Chem. 33 3645 (1994) 79. R W Broach, I S Chuang, T J Marks, J M Williams Inorg. Chem. 22 1081 (1983) 80. T A Keiderling, W T Wozniak, R S Gay, D Jurkowitz, E R Bernstein, S J Lippard, T G Spiro Inorg.Chem. 14 576 (1975) 81. N Davies,M G H Wallbridge, B E Smith, B D James J. Chem. Soc., Dalton Trans. 162 (1973) 82. P L Johnson, S A Cohen, T J Marks, J M Williams J. Am. Chem. Soc. 100 2709 (1978) 83. M D Fryzuk, A J Rettig, A Westerhaus, H D Williams Inorg. Chem. 24 4316 (1985) 84. J A Jensen, G Girolamy J. Am. Chem. Soc. 110 4450 (1988) 85. J A Jensen, G Girolamy Inorg. Chem. 28 2114 (1989) 86. B Hessen, J H Teuben, T H Lemmen, J C Huffman, K G Caulton Organometallics 4 946 (1985) 87. B Hessen, T H Lemmen, MG H Luttikhedde, J H Teuben, J L Petersen, J C Huffman, S Jagner, K G Caulton Organometallics 6 2354 (1987) 88. F A Cotton, S A Duraj,W J Roth Inorg. Chem. 23 4113 (1984) 89. P Berno,M Moore,R Minhas, S Gambarotta Can.J. Chem. 74 1930 (1996) 90. N I Kirillova, A I Gusev, Yu T Struchkov Zh. Strukt. Khim. 15 718 (1974) a 91. N J Bailey,M L H Green,MA Leech, J F Saunders, HMTidswell J. Organomet. Chem. 538 111 (1997) 92. F A Cotton, C A Murillo, XWang J. Am. Chem. Soc. 120 9594 (1998) 93. M Dionne, S Hao, S Gambarotta Can. J. Chem. 73 1126 (1995) 94. A R Barron, J E Salt, G Wilkinson,M Motewalli,M B Hursthouse Polyhedron 5 1833 (1986) 95. M Y Darensbourg, R Bau, M W Marks, R R Burch, J C Deaton, S Slater J. Am. Chem. Soc. 104 6961 (1982) 96. SWKirtley, M A Andrews, R Bau, GWGrynkewich, T J Marks, D L Tipton, B R Whittlesey J. Am. Chem. Soc. 99 7154 (1977)Structural and dynamic properties of tetrahydroborate complexes 97. J L Atwood, W E Hunter, E Carmona-Guzman, G Wilkinson J.Chem. Soc., Dalton Trans. 467 (1980) 98. A A H van der Zeiden, V Shklover, H Berke Inorg. Chem. 30 4393 (1991) 99. V D Makhaev, A P Borisov, T P Gnilomedova, E B Lobkovskii, A N Chekhlov Izv. Akad. Nauk SSSR, Ser. Khim. 1712 (1987) d 100. E B Lobkovskii, A N Chekhlov, V D Makhaev, A P Borisov Koord. Khim. 15 377 (1989) b 101. V D Makhaev, A P Borisov, T P Karpova, E B Lobkovskii, B P Tarasov,A N Chekhlov Izv. Akad. Nauk SSSR, Ser. Khim. 432 (1989) d 102. C A Ghilardi, P Innocenti, S Midollini, A Orlandini J. Organomet. Chem. 231 C78 (1982) 103. C A Ghilardi, P Innocenti, S Midollini,A Orlandini J. Chem. Soc., Dalton Trans. 605 (1985) 104. J A Statler, G Wilkinson, M Thornton-Pett,M B Hursthouse J.Chem. Soc., Dalton Trans. 1731 (1984) 105. P W Frost, J A K Howard, J L Spencer J. Chem. Soc., Chem. Commun. 1362 (1984) 106. E J Corey, N J Cooper,WMCanning,WN Lipscomb, T F Koetzle Inorg. Chem. 21 192 (1982) 107. M Nakajima, H Moriyama, A Kobayashi, T Saito, Y Sasaki J. Chem. Soc., Chem. Commun. 80 (1975) 108. M Nakajima, T Saito, A Kobayashi, Y Sasaki J. Chem. Soc., Dalton Trans. 385 (1977) 109. P Dapporto, S Midollini, A Orlandini, L Sacconi Inorg. Chem. 15 2768 (1976) 110. F Shafiq, R Eisenberg Inorg. Chem. 32 3287 (1993) 111. T Saito, M Nakajima, A Kobayashi, Y Sasaki J. Chem. Soc., Dalton Trans. 482 (1978) 112. M L H Green, H Munakata, T Saito J. Chem. Soc., A 469 (1971) 113. B E Green, C H L Kennard, C J Hawkins, G Smith, B D James, A H White Acta Crystallogr., Sect.B 36 2407 (1980) 114. B E Green, C H L Kennard, G Smith, MMElcombe, F H Moore, B D James, A H White Inorg. Chim. Acta 83 177 (1984) 115. V D Makhaev, A P Borisov, A S Antsyshkina, G G Sadikov, M A Porai-Koshits, N S Kedrova, N N Mal'tseva, S Ya Istomin Koord. Khim. 19 858 (1993) b 116. S J Lippard, K M Melmed Inorg. Chem. 6 2223 (1967) 117. E B Lobkovskii, V D Makhaev, A P Borisov Zh. Strukt. Khim. 25 (3) 172 (1984) a 118. V D Makhaev, A P Borisov, E B Lobkovskii, V B Polyakova, K N Semenenko Izv. Akad. Nauk SSSR, Ser. Khim. 1881 (1985) d 119. V D Makhaev, F M Dolgushin, A I Yanovskii, Yu T Struchkov Koord. Khim. 22 608 (1996) b 120. M M T Khan, P Paul, K Venkatasubramanian Polyhedron 10 1827 (1991) 121.M V Baker, L D Field J. Chem. Soc., Chem. Commun. 996 (1984) 122. R Bau,H S H Yuan,M V Baker, L D Field Inorg. Chim. Acta 114 L27 (1986) 123. M V Baker, L D Field Appl. Organomet. Chem. 4 543 (1990) 124. T Yoshida, T Adachi, T Ueda, H Akao, T Tanaka, F Goto Inorg. Chim. Acta 231 95 (1995) 125. F Takusagava, A Fumigalli, T F Koetzle, SG Shore, T Schmitkons, A V Fratini, K W Morse, C Y Wei, R Bau J. Am. Chem. Soc. 103 5165 (1981) 126. J C Bommer, K W Morse Inorg. Chem. 19 587 (1980) 127. C A Ghilardi, S Midollini, A Orlandini Inorg. Chem. 21 4096 (1982) 128. E B Lobkovskii,M Yu Antipin, A P Borisov, V D Makhaev, K N Semenenko, Yu T Struchkov Koord. Khim. 7 307 (1981) b 129. V D Makhaev, A P Borisov, E B Lobkovskii, K N Semenenko Izv.Akad. Nauk SSSR, Ser. Khim. 2614 (1980) d 130. V D Makhaev, A P Borisov, B I Saidov, N S Kedrova, A S Antsyshkina, N N Mal'tseva, Yu M Shul'ga Koord. Khim. 18 469 (1992) b 131. A S Antsyshkina,MA Porai-Koshits, V D Makhaev, A P Borisov, N S Kedrova, N N Mal'tseva Koord. Khim. 18 474 (1992) b 132. V D Makhaev, A S Antsyshkina, Yu M Shul'ga Zh. Neorg. Khim. 41 1022 (1996) c 133. G A Koutsantonis, F C Iee, C L Raston J Chem. Soc., Chem. Commun. 1975 (1994) 134. B I Saidov, A P Borisov, V D Makhaev, G N Boiko, A S Antsyshkina, N S Kedrova, N N Mal'tseva Zh. Neorg. Khim. 35 626 (1990) c 745 135. A S Antsyshkina,M A Porai-Koshits, B I Saidov, V D Makhaev, A P Borisov, N N Mal'tseva, N S Kedrova Koord. Khim.17 405 (1991) b 136. R L Bansemer, J C Huffman, K G Caulton J. Am. Chem. Soc. 105 6163 (1983) 137. MAPorai-Koshits,AS Antsyshkina,AAPasynskii,G G Sadikov, Yu V Skripkin, V I Ostrikova Inorg. Chim. Acta 34 L285 (1979) 138. D L Reger, S R Mason, A L Rheingold J. Am. Chem. Soc. 115 10 406 (1993) 139. C R Pulham, P T Brain, A J Downs, D W H Rankin, H E Robertson J Chem. Soc., Chem. Commun. 177 (1990) 140. A J Downs, T M Greene, L A Harman, P F Souter, P T Brain, C R Pulham, D W H Rankin, H E Robertson,M Hoffmann, P v R Schleyer Inorg. Chem. 34 1799 (1995) 141. A J Downs, P D P Thomas J. Chem. Soc., Dalton Trans. 809 (1978) 142. M T Barlow, A J Downs, P D P Thomas, D W H Rankin J. Chem. Soc., Dalton Trans. 1793 (1979) 143. R A Fischer, T Priermeier, W Scherer J.Organomet. Chem. 459 65 (1993) 144. S Aldridge, A J Downs, S Parsons J. Chem. Soc., Chem. Commun. 2055 (1996) 145. D L Reger, J E Collins, M A Matthews, A L Rheingold, L M Liable-Sanches, I A Guzei Inorg. Chem. 36 6266 (1997) 146. AS Antsyshkina,GGSadikov,MAPorai-Koshits, VNKonoplev, T A Silina, A S Sizareva Koord. Khim. 20 274 (1994) b 147. H H Giese, H Noth, H Schwenk, S Thomas Eur. J. Inorg. Chem. 1 941 (1998) 148. H H Giese, T Habereder, H Noth,W Ponikwar, S Thomas, M Warchhold Inorg. Chem. 38 4188 (1999) 149. J E Espidel, RKHarris, KWade Magn. Reson. Chem. 31 166 (1994) 150. A I Gorbunov, P A Storozhenko, L V Ivakina, B N Bulychev, A I Gusev Dokl. Akad. Nauk SSSR 285 129 (1985) e 151. D A Drew, G Gundersen, A Haaland Acta Chem.Scand. 26 2147 (1972) 152. D A Coe, J W Nibler, T H Cook, D Drew, G L Morgan J. Chem. Phys. 63 4842 (1975) 153. B Morosin, J Howatson J. Inorg. Nucl. Chem. 41 1667 (1979) 154. E B Lobkovskii, L V Titov, S B Psikha, M Yu Antipin, Yu T Struchkov Zh. Strukt. Khim. 23 (4) 172 (1982) a 155. E A Hanecker, J Moll, H Noth Z. Naturforsch., B Chem. Sci. 39 424 (1984) 156. E B Lobkovskii, A N Chekhlov, M D Levicheva, L V Titova Koord. Khim. 14 543 (1988) b 157. M Bremer, H Noth, M Thomann,M Schmidt Chem. Ber. 128 455 (1995) 158. A Almenningen, G Gundersen, A Haaland Acta Chem. Scand. 22 328 (1968) 159. D A Coe, J W Nibler Spectrochim. Acta A, Mol. Spectrosc. 29 1789 (1973) 160. S Aldridge, A J Blake, A J Downs, R O Gould, S Parsons, C R Pulham J.Chem. Soc., Dalton Trans. 1007 (1997) 161. D Dou, J P Lin, J A Krause Bauer, G T Jordan IV, S G Shore Inorg. Chem. 33 5443 (1994) 162. P R Oddy, MG H Wallbridge J. Chem. Soc., Dalton Trans. 869 (1976) 163. MT Barlow, C J Dain, A J Downs, P D P Thomas, DW H Rankin J. Chem. Soc., Dalton Trans. 1374 (1980) 164. E B Lobkovskii, A L Dorosinskii, K N Semenenko Zh. Strukt. Khim. 15 70 (1974) a 165. NABailey, PHBird,MGHWallbridge Inorg. Chem. 7 1575 (1968) 166. NDavies, PHBird,MG HWallbridge J. Chem. Soc. A 2269 (1968) 167. K N Semenenko, E B Lobkovskii, B L Tarnopol'skii, M A Simonov Zh. Strukt. Khim. 17 1076 (1976) a 168. P Bissinger, P Mikulcik, J Riede, A Schier, H Schmidbaur J. Organomet. Chem. 446 37 (1993) 169.R A Bell, S A Cohen, N M Doherty, R S Threlkel, J E Bercaw Organometallics 5 972 (1986) 170. M L H Green, A K Hughes J. Chem. Soc., Dalton Trans. 527 (1992) 171. M L H Green, L L Wong J. Chem. Soc., Chem. Commun. 571 (1989) 172. M L Luetkens, J C Huffman, A P Sattelberger J. Am. Chem. Soc. 107 3361 (1985)746 173. R Carreno, V Riera,M A Ruiz, Y Jeannin,M Philoche-Levissalles J. Chem. Soc., Chem. Commun. 15 (1990) 174. R Carreno, V Riera, M A Ruiz, C Bois, Y Jeannin Organometallics 12 1946 (1993) 175. A J L Pombeiro, R L Richards J. Organomet. Chem. 306 C33 (1986) 176. D Gusev, A Llamazares, G Artres, H Jacobsen, H Berke Organometallics 18 75 (1999) 177. J Vites, C F Housecroft, C Eigenbrot,M L Buhl, C J Long, T P Fehlner J.Am. Chem. Soc. 108 3304 (1986) 178. MJahncke, G Meister, G Rheinwald, H Stoeckli-Evans, G Suss-Fink Organometallics 16 1137 (1997) 179. T Blackmore,M I Bruce, F G A Stone J. Chem. Soc. A2376 (1971) 180. H Suzuki, D H Lee, N Oshima, Y Moro-oka J. Organomet. Chem. 317 C45 (1986) 181. J W Bruno, J C Huffman, K G Caulton Inorg. Chim. Acta 89 167 (1984) 182. R H Crabtree, A J Pearman J. Organomet. Chem. 157 335 (1978) 183. D G Holah, A N Hughes, B C Hui Can. J. Chem. 54, 320 (1976) 184. J B Letts, T J Mazanec, D W Meek J. Am. Chem. Soc. 104 3898 (1982) 185. G C Jia, S D Drouin, P G Jessop, A J Lough, R H Morris Organometallics 12 906 (1993) 186. C Bianchini, P J Perez, M Peruzzini, F Zanobini, A Vacca Inorg. Chem. 30 279 (1991) 187. G C Jia, D W Meek, J C Gallucci Inorg.Chem. 30 1409 (1991) 188. H Werner, M A Esteruelas, U Meyer, B Wrackemeyer Chem. Ber. 120 11 (1987) 189. M A Esteruelas, M P Garcia, A M Lopez, L A Oro, N Ruiz, C Schlunken, C Valero, H Werner Inorg. Chem. 31 5580 (1992) 190. C Bianchini, P Innocenti, D Masi, M Peruzzini, F Zanobini Gazz. Chim. Ital. 122 461 (1992) 191. A Mezzetti, L Costella, A Del Zotto, P Rigo Gazz. Chim. Ital. 123 155 (1993) 192. M A Esteruelas, Y Jean, A Lledos, L A Oro, N Ruiz, F Volatron Inorg. Chem. 33 3609 (1994) 193. D J Wink, N J Cooper J. Chem. Soc., Dalton Trans. 1257 (1984) 194. H D Empsall, E Mentzer, B L Shaw J. Chem. Soc., Chem. Commun. 861 (1975) 195. H D Empsall, E M Hyde, E Mentzer, B L Shaw,M F Uttley J. Chem. Soc., Dalton Trans. 2069 (1976) 196. T M Gilbert, F J Hollander, R G Bergman J. Am. Chem. Soc. 107 3508 (1985) 197. A Heine, D Stalke J. Organomet. Chem. 542 25 (1997) 198. D R Armstrong,WClegg, H MColquhoun, J A Daniels, R E Mulvey, I R Stephenson, K Wade J. Chem. Soc., Chem. Commun. 630 (1987) 199. A S Antsyshkina, G G Sadikov, M A Porai-Koshits, V N Konoplev, A S Sizareva, T A Silina Koord. Khim. 19 596 (1993) b 200. D S Marynick, W N Lipscomb J. Am. Chem. Soc. 93 2322 (1971) 201. J W Nibler, D F Shriver, T H Cook J. Phys. Chem. 54 5257 (1971) 202. V D Makhaev, A P Borisov, E B Lobkovskii Zh. Neorg. Khim. 29 1160 (1984) c 203. E B Lobkovsky, Y K Gunko, B M Bulychev, V K Belsky, G L Soloveichik, M Yu Antipin J. Organomet. Chem. 406 343 (1991) 204. Y K Gunko, B M Bulychev, G L Soloveichik, V K Belsky J. Organomet. Chem. 424 289 (1992) 205. G MCendrowski-Guillaume,M Nierlich, M Lance, M Ephritikhine Organometallics 17 786 (1998) 206. A V Khvostov, V K Belsky, A I Sizov, B M Bulychev, N B Ivchenko J. Organomet. Chem. 564 5 (1998) 207. E R Bernstein, WC Hamilton, T A Keiderling, S J LaPlaca, S J Lippard, J J Mayerle Inorg. Chem. 11 3009 (1972) 208. E R Bernstein, T A Keiderling, S J Lippard, J J Mayerle J. Am. Chem. Soc. 94 2552 (1972) 209. V V Volkov, K G Myakishev Zh. Neorg. Khim. 15 2861 (1970) c 210. P Charpin, M Nierlich, D Vigner, MLance, D Baudry 211. R R Rietz, A Zalkin, D H Templeton, N M Edelstein, 212. A Zalkin, R R Rietz, D H Templeton, N M Edelstein Acta Crystallogr., Sect. C 43 1465 (1987) L K Templeton Inorg. Chem. 17 653 (1978) Inorg. Chem. 17 661 (1978) V D Makhaev 213. L F Rhodes, A Albinati, C Sorato, L M Venanzi Inorg. Chem. 25 3335 (1986) 214. D G Holah, A N Hughes, S Maciaszek, V R Magnuson, K O Parker Inorg. Chem. 24 3956 (1985) 215. F Cariati, L Naldini J. Inorg. Nucl. Chem. 28 2243 (1966) 216. B E Green, C H L Kennard, G Smith, P C Healy, A H White Inorg. Chim. Acta 81 147 (1984) 217. S Aldridge, A J Blake, A J Downs, S Parsons, C R Pulham J. Chem. Soc., Dalton Trans. 853 (1996) 218. J W Nibler, T H Cook J. Phys. Chem. 58 1596 (1973) 219. E S Shubina, E V Bakhmutova, L N Saitkulova, I A Tikhonova, G G Furin, V I Bakhmutov, V B Shur, L M Epshtein Izv. Akad. Nauk, Ser. Khim. 885 (1997) d 220. L N Saitkulova, E V Bakhmutova, E S Shubina, I A Tikhonova, G G Furin, V I Bakhmutov, N P Gambaryan, A L Chistyakov, I V Stankevich, V B Shur, L M Epstein J. Organomet. Chem. 585 201 (1999) 221. A J Downs, S Parsons, C R Pulham, P F Souter Angew. Chem., Int. Ed. Engl. 36 890 (1997) 222. R D Shannon Acta Crystallogr., Sect. A 32 751 (1976) 223. N Edelstein Inorg. Chem. 20 279 (1981) 224. V D Makhaev, A P Borisov Zh. Neorg. Khim. 44 1489 (1999) c 225. A P Borisov, V D Makhaev, in Khimiya Neorganicheskikh Gidridov (The Chemistry of Inorganic Hydrides) (Ed. N T Kuznetsov) (Moscow: Nauka, 1990) p. 82 226. J C Bommer, K W Morse Inorg. Chim. Acta 74 25 (1983) 227. K N Semenenko, O V Kravchenko, V B Polyakova Usp. Khim. 42 3 (1973) [Russ. Chem. Rev. 42 1 (1973)] 228. J W Lauher, R Hoffmann J. Am. Chem. Soc. 98 1729 (1976) 229. A P Hitchcock, N Yao, N H Werstiuk, MJ McGlinchey, T Ziegler Inorg. Chem. 21 793 (1982) 230. M Mancini, P Bougeard, R C Burns,M Mlekuz, B G Sayer, J I A Thompson,MJ McGlinchey Inorg. Chem. 23 1072 (1984) 231. V I Mikheeva, N N Mal'tseva, N S Kedrova Zh. Neorg. Khim. 24 408 (1979) c 232. H Ehemann, H NoÈ th Z. Anorg. Allg. Chem. 386 87 (1971) 233. V D Makhaev, A P Borisov, G N Boiko, B P Tarasov Izv. Akad. Nauk SSSR, Ser. Khim. 1207 (1990)d 234. B Nissen-Sobocinska, B Jezowska-Trzebiatowska J. Organomet. Chem. 452 277 (1993) 235. R Colton, B DJames, I D Potter, J C Traeger Inorg. Chem. 32 2626 (1993) 236. I D Potter, B D James Polyhedron 18 1201 (1999) 237. E B Lobkovskii Zh. Strukt. Khim. 24 (2) 66 (1983) a 238. R J Gillespie Molecular Geometry (London: Van Nostrand Rheingold Company, 1972) 239. O P Charkin, D G Musaev, N M Klimenko Koord. Khim. 11 728 (1985) b 240. A S Zyubin, D G Musaev, O P Charkin Koord. Khim. 13 1329 (1987) b 241. K S Siddiqi, F M A M Aqra, S H Shah, S A A Zaidi Polyhedron 13 247 (1994) 242. S Kapur, B L Kalsotra, R K Multani, B D Jain J. Inorg. Nucl. Chem. 35 1689 (1973) 243. S P Anand, R K Multani, B D Jain J. Organomet. Chem. 26 115 (1971) 244. G B Deacon, T D Tuong, D G Vince Polyhedron 2 969 (1983) 245. R H Crabtree, G G Hlatky Polyhedron 4 521 (1985) 246. I P Rothwell Polyhedron 4 177 (1985) 247. W Kaminsky J. Chem. Soc., Dalton Trans. 1413 (1998) 248. W Keim Vysokomol. Soedin., Ser. A 36 1644 (1994) f a�Russ. J. Struct. Chem. (Engl. Transl.) b�Russ. J. Coord. Chem. (Engl. Transl.) c�Russ. J. Inorg. Chem. (Engl. Transl.) d�Russ. Chem. Bull. (Engl. Transl.) e�Dokl. Chem. Technol., Dokl. Chem. (Engl. Transl.) f�Polym. Sci. (En

ISSN:0036-021X    

出版商:RSC    

年代:2000

数据来源: RSC

摘要:

Russian Chemical Reviews 69 (9) 747 ± 768 (2000) Vibrational spectra, force constants and conformations of molecules of trivalent tricoordinate phosphorus compounds S A Katsyuba Contents I. Introduction II. Alkylphosphines III. Alkyl(halogeno)phosphines IV. Phosphines with unsaturated groups at the phosphorus atom V. Alkoxy derivatives of trivalent phosphorus VI. Dialkylphosphoramidites VII. Phosphorothioites VIII. Compounds with Si7P, Ge7P and P7P bonds IX. Rotational isomerism of trivalent phosphorus compounds and its manifestation in vibrational spectra X. Force constants of molecules containing trivalent phosphorus atoms XI. Conclusion Abstract. spectra Raman and IR the of interpretation the on Data Data on the interpretation of the IR and Raman spectra based on normal coordinate analysis, on the force constants and based on normal coordinate analysis, on the force constants and on the conformations of the molecules of trivalent tricoordinate on the conformations of the molecules of trivalent tricoordinate phosphorus existing The summarised. are compounds phosphorus compounds are summarised.The existing correla- correla- tions refined are structures molecular and spectra the between tions between the spectra and molecular structures are refined and and some bibliography The proposed. are correlations new some new correlations are proposed. The bibliography includes includes 171 references 171 references. I. Introduction Vibrational spectroscopy is widely used in the chemistry of organophosphorus compounds for the identification of functional groups, determination of the molecular structure and studies on the molecular conformations, inter- and intramolecular interac- tions.Monographs by Thomas 1, 2 and atlases by Shagidullin et al.3±5 are devoted to the interpretation of IR spectra of organo- phosphorus compounds. Unfortunately, Raman spectroscopy has not become a routine method in phosphorus chemistry as yet, though the scope of possible application of this experimental technique in structural studies is as large as that of IR spectro- scopy. This is to a great extent due to the lack of studies devoted to the generalisation of scarce information on the Raman spectra of different organophosphorus compounds. In this review, a com- parative analysis is given of a continuously growing body of data on both the IR and Raman spectra of phosphorus-containing molecules.An attempt to summarise information on the `finger- prints' of rotational isomerism in the molecules of tricoordinate phosphorus compounds as manifested in vibrational spectra is also undertaken. Adistinctive feature of the review is that here we consider only those molecules the spectra of which have been interpreted on the basis of normal coordinate analysis.Anormal coordinate analysis S A Katsyuba A E Arbuzov Institute of Organic and Physical Chemistry, Kazan Scientific Centre of the Russian Academy of Sciences, ul. Akad. Arbuzova 8, 420088 Kazan, Russian Federation. Fax (7-843) 275 22 53.Tel. (7-843) 276 74 83. E-mail: katsyuba@iopc.kcn.ru Received 13 June 2000 Uspekhi Khimii 69 (9) 817 ± 839 (2000); translated by AMRaevsky #2000 Russian Academy of Sciences and Turpion Ltd DOI 10.1070/RC2000v069n09ABEH000551 747 748 751 754 755 757 759 762 763 765 766 1 1<n1&n2<n 02. makes it possible to calculate the eigenvectors of normal vibra- tions, i.e., to assess the contribution of each valence bond and each bond angle to a particular molecular vibration.6, 7 This allows unambiguous assignment of the frequencies to the vibrations of particular structural fragments. Knowledge of the eigenvectors of normal vibrations is also necessary for the analysis of the interactions between vibrations of particular bonds in a molecule. It is known that effective coupling of vibrations of two molecular fragments is achieved if their individual frequencies are close (n1&n2) and the fragments either share a number of atoms or are connected by a valence bond.8 For the coupling of vibrations to occur, they should also be of the same symmetry.9 As a result, coupling of two individual vibrations gives rise to two new coupled vibrations.Each fragment of the molecule makes a contribution to each of the new vibrations. Their frequencies (n 0 and n 02) can differ appreciably from the frequencies of the `old' vibrations; as a rule, n 0 Thus in the molecule of methylphosphine (1), vibrations with a predominant contribution of the P7C bond occur at a fre- quency of about 670 cm71. Vibrations of two P7C bonds in the molecule of dimethylphosphine (2) give rise to two new vibrations, viz., a symmetric stretch of both bonds at a frequency nn 00 1&660 cm71 and an antisymmetric vibration at a frequency 2&700 cm71.These changes in the n(P7C) frequencies on going from molecule 1 to the molecule 2 are mainly due to the coupling of two individual oscillators sharing the phosphorus atom. Vibrations of the PCl2 group in the molecule of methyl phosphorodichloridite, CH3OPCl2 (3), are characterised by strong IR absorption bands at *460 and *500 cm71. The frequencies of bending vibrations of the butyl group lie in the same spectral region. However, analysis of the molecular structure of butyl phosphorodichloridite, C4H9OPCl2, shows that these two vibrations cannot interact, since the structural fragments under consideration share no atoms and are not connected by a covalent bond.Therefore, vibrations of the PCl2 group have identical patterns in the IR spectra of both C4H9OPCl2 and CH3OPCl2. These examples show that, strictly speaking, the lack of information on the eigenvectors of some vibrations makes it impossible to conclude whether they will be characteristic on going from one molecule to another. Therefore, spectral ± struc-748 tural correlations become much more reliable if they are based on theoretical interpretation of the spectra. A normal coordinate analysis also makes it possible to calculate the force constants of chemical bonds and bond angles in molecules.The force constants are primarily of interest as transferable parameters, which are used for calculating the vibra- tional spectra of related molecules.6 At the same time, they are also numerical characteristics of the potential surface of a mole- cule. It should be noted that for several reasons 6, 10 the force constant values depend on the type of computational procedures employed, so that small differences between the estimates of the force fields obtained by different authors for the same molecule are virtually inevitable. Because of this, here we compared the force constants assessed using the same approximations for series of related compounds. This makes it possible to judge the changes in the electronic structure of the molecules if analysis of their force fields reveals large and systematic deviations. The notations are given according to monograph 6 and the values of force constants are given in 102 N m71.Compounds of trivalent tricoordinate phosphorus were chosen to be the subject of this review primarily since they are most widely used in the syntheses of organophosphorus com- pounds. Additionally (probably, because of relatively low stability of tricoordinate phosphorus compounds), their vibrational spec- tra have been less studied than those of tetracoordinate phospho- rus compounds. Virtually all results of normal coordinate analyses of vibrational spectra of organophosphorus compounds reported to date have been analysed.The exceptions are the results of studies on the simplest tetraatomic molecules of phosphine (PH3) and its halogen derivatives (HnPHal37n). Spectral characteristics of all the compounds under discussion are listed in Tables. Each fundamental vibration was assigned a number as follows. The whole set of vibrations was divided into groups according to particular symmetry types of the vibrations. Within each symmetry type, the vibrations were numbered in descending order of their frequencies. The frequencies assigned to overtones and combination bonds are omitted. The notations of vibrations are n for stretching, d for bending, r for rocking, w for wagging, t for twisting and w for torsional vibrations; g denotes vibrations associated with the out- of-plane deviation of particular bonds. These notations describe the type of motion of particular structural fragments of molecules.In discussing any vibration of the entire molecule, that is, a simultaneous motion of several individual oscillators, the above- mentioned notations have a somewhat different meaning. For instance, the totally symmetrical fundamental vibration of the molecule 1 is composed of the contributions from not only the stretching vibrations n(P7C) of the P7C bond, but also the wagging vibrations of the phosphino group, w(PH2). However, the major contribution to the energy of the resultant vibration is make by the P7C bond; therefore, such a vibration is denoted below as n(P7C). All frequencies are given in cm71.The intensities (I ) and shapes of spectral bands, as well as the degree of depolarisation of the Raman lines (r), are qualitatively described using the follow- ing notations: w for weak, m for medium, s for strong, ms for medium strong, br for broad, v for very, sh for shoulder, p for polarised and dp for depolarised. Spectral data listed in Tables are presented in the following order: the vibrational frequency (in cm71), a qualitative or quantitative description of the intensity (in the latter case, the intensity of a line with respect to that of the strongest Raman peak, taken as 100) and a qualitative or quantitative description of the degree of depolarisation. II. Alkylphosphines Vibrational spectra and conformations of primary alkylphos- phines have been studied in most detail.Interpretation 11 ± 13 of the IR spectra 11 and the Raman spectra 14 of methylphosphine S A Katsyuba Table 1. Interpretation of vibrational spectra of methylphosphine. IR (g) Raman (l) Assignment No. Symmetry A00 113 A0 12 A00 2297, vs, p 2297, vs, p *1440, w *1440, w 1288, m, p 1086, m, p 4 A0 5 A0 6 A0 A00 137 A0 8 A0 974, m, p 740, s, p A00 nas(PH2) ns(PH2) das(CH3) das(CH3) ds(CH3) d(PH2) r(CH3), t(PH2) r(CH3), w(PH2), d(PH2) w(PH2), r(CH3), n(P7C) t(PH2), r(CH3) n(P7C), w(PH2) 2309, vs 2305, vs 1435, m 1429, m 1296, w 1092, s 1017, m 978, s 730, w 696, wa 676, s 149 A0 673, s, p Note. Hereafter, l stands for liquid and g for gas.Totally symmetric fundamental vibrations of the A0 type were labelled with figures from 1 to 9 and antisymmetric vibrations of the A00 type were labelled with figures from 10 to 15. Vibrations with the highest frequency and belonging to the A0 type (Nos 1 and 2) and A00 type (No. 10) and vibration No. 15 are not included, since they correspond to the C7H stretching vibrations of the bonds, which are insignificant for analytical purposes. For the same reason, some vibrations are also not included in Tables 2 ± 25. a Appears only in the spectrum of crystals. (Table 1) and its D-isotopomers CH3PD2, CD3PH2 and CD3PD2 has been given and the force constants have been estimated. Vibrational spectra of ethylphosphine (4) have been investigated in a number of studies.15 ± 21 Using vibrational 15 and micro- wave 16, 17 spectroscopy, the molecules of compound 4 were shown to exist in the liquid and gas phases in two rotameric forms, as trans (T) and gauche conformers (G) (Fig.1; X=H, Z P P XX XX Y YZ G T Figure 1. Conformers of molecules X2PYZ: trans (T) and gauche (G). Y=CH2, Z=CH3). Crystals of this compound are built of the molecules of the trans conformer only, so that the spectra are simplified owing to the disappearance of the bands of the gauche isomer. Interpretation of the spectra of ethylphosphine was based on the results of normal coordinate analysis 22 performed simulta- neously for both C2H5PH2 and C2H5PD2 molecules (Table 2).In studies of vibrational spectra of n-propylphosphine (5) 23 (Table 3) it has been shown that this compound exists in the liquid phase as a mixture of five conformers (Fig. 2; X=H, Y=CH2). Vibrational and microwave spectra of isopropylphos- P P P CH2 XX XX XX CH3 Y Y Y Gt CH2 H3C CH2CH3 Tg Tt CH3 H3C CH2 P P CH2 XX Y XX Y Gg Gg 0 Figure 2. Conformers of molecules X2PYCH2CH3: trans,trans (Tt), trans,gauche (Tg), gauche,trans (Gt), gauche,gauche (Gg) and gauche,gauche 0 (Gg 0).Vibrational spectra, force constants and conformations of molecules of trivalent tricoordinate phosphorus compounds Table 3. Interpretation of vibrational spectra of n-propylphosphine. Table 2. Interpretation of vibrational spectra of ethylphosphine.Raman (g) IR (g) Assignment No. Sym- Assignment metry No. Sym- metry 2304, vs 5 A0 17 A00 4 A0 nas(PH2) ns(PH2) 2298, vs 23 A00 11 A0 26 A00 1247, w 1085, m 12 A0 1056, m 982, w 835, w 13 A0 27 A00 2296, vs 2294, vs 1271, vw 1247, m 1086, w 1065, m, sh 1067, m 1055, m 979, w 833, m 817, m 8 A0 19 A00 9 A0 10 A0 20 A00 11 A0 21 A00 12 A0 22 A00 14 A0 28 A00 643, w 634, w 13 A0 14 A0 w(CH2), nas(PCC), d(CH2) t(CH2) d(PH2), n(C7C), w(PH2) n(C7C), d(PH2) r(CH3), r(CH2), t(PH2) r(CH3), w(CH2), d(PCC) t(PH2), r(CH3), r(CH2) w(PH2), w(CH2), d(CH2), d(PH2) 818, m 692, m 640, vs 636, vs 307, m 288, w 15 A0 245, w 183, w 23 A00 24 A00 162, vw r(CH2), t[PH2(G)] r(CH2), t(PH2) n(P7C), d(PCC) d(PCC) d[PCC(G)] w(CH3), w(PH2) w(PH2), w(CH3) w[PH2(G)] 16 A0 29 A00 Note.The number of vibration, symmetry type and assignment are given for the T conformer (see Fig. 1). Lines of the Gconformer were interpreted only if they did not coincide with the corresponding lines of the T conformer (their assignments are denoted by the character `G'). 17 A0 phine (6) and its D-isotopomer, iso-C3H7PD2, have been studied.24, 25 It was shown that the G and T conformers of these molecules are stable in the liquid and gas phases. 30 A00 18 A0 CH3 CH3 H H P P H CH3 C C H H HT CH3 G 31 A00 19 A0 32 A00 33 A00 Note. The number of vibration, symmetry type and assignment are given for the Tt conformer (see Fig.2). Vibrations of other conformers (for notations of their frequencies, see Caption to Fig. 2) were interpreted only if their frequencies did not coincide with the corresponding lines of the Tt conformer. a From the spectrum of n-propylphosphine solution in liquid krypton. b From the gas-phase spectrum. c From the low-temperature spectrum of solid n-propylphosphine. The calculated assignments 22 for isopropylphosphine 6 (see Table 4) are somewhat different from the empirical ones.24 Compared with the spectra of primary alkylphosphines, the interpretation of the spectra of secondary and tertiary alkylphos- phines is more complicated because of the necessity to take into account the coupling of vibrations of adjacent alkyl groups at the phosphorus atom. Therefore, it is not surprising that the spectra of such simple representatives of organophosphorus compounds as dimethylphosphine and its isotopomers have been interpreted in different ways.26 ± 29 The problem of contradictory assignments was alleviated only after performing a normal coordinate analy- sis.12, 13, 29 Different versions of empirical interpretation of the vibrational spectra of phosphirane (CH2)2PH (7) and its deutero derivatives have also been suggested.30, 31 Calculations 32 made it possible to refine the vibrational assignments and to assess the force constants of the molecule, its thermodynamic functions and the barrier to inversion of the phosphirane ring.been re-calculated 42 ± 44 using the results of ab initio quantum- chemical calculations; it should be mentioned that not only the spectra of deuterated specimens, but also those of 13C-isotopom- ers of trimethylphosphine 8 have been analysed 43 and that vibra- tional anharmonicity has also been taken into account.The molecule of triethylphosphine (9) can adopt four con- formations;45 however, it has been shown 45, 46 that vibrational spectra of compound 9 in the liquid phase (see Refs 46, 47) can be interpreted assuming the existence of only three conformers shown below. Trimethylphosphine (8) is the first organophosphorus com- pound for which a normal coordinate analysis has been per- formed.33 The data on both the eigenvectors of molecular vibrations of (CH3)3P and (CD3)3P and the force fields of the molecules have been repeatedly revised.13, 14, 34 ± 40 These mole- cules have a C3v symmetry, therefore the A2 modes are optically inactive and not observed in the spectra.Some of the modes were observed owing to the reduction of symmetry of the molecule as a result of crystallisation of the substance.39 Nevertheless, a number of vibrational assignments had to be revised after the spectra of low-symmetric D-isotopomers, (CH3)2CD3P and CH3(CD3)2P, had been reported.41 The force constants of the molecule 8 have 749 IR (g) Raman (l) ns[PH2(Tg,Gg 0)] ns(PH2) nas(PH2) w(CH2P) t(CH2P), r(CH3), r(CH2) t[CH2P(Gg,Gg 0)] d(PH2) d[PH2(Gt, Gg 0)] r(CH3), ns(CCC) r(CH3), t[CH2P(Gg, Gg 0)] nas(CCC) t(PH2) t(PH2), r(CH3), ns[CCC(Gt)] t(PH2), ns[CCC(Gg 0)] 2306, s a 2298, vs 2294, vs 1233, sh a 1221, wa 1207, sh 1104, sh 1096, wa 1085, s 1073, ma 1052, sh 1037, sh 910, sh 898, m 883, sh 898, m 845, m 2300, sh, p 2289, sh, p 2285, vs, p 1228, m, p 1218, m, p 1203, sh, p 1097, m, p 1094, m, p 1075, m, p r(CH3), t(CH2P), t(PH2), t(CH2) 1061, wb 1052, wc 1032, m, p 907, vw, p 898, m, p 880, m, p 898, m, p 846, mc 812, vw, dp 812, sh 891, sh 889, m, p ns(CCC), r(CH3), t(PH2) ns(CCC), w(PH2), n[P7C(Tg)] ns(CCC), t(PH2), n[P7C(Gt)] w(PH2), r(CH3), r(CH2P), ns[CCC(Gt)] w(PH2), ns[CCC(Tg,Gg)] w(PH2), ns(CCC), n(P7C) r(CH2), r(CH3), t(PH2) n[P7C(Gt)] n(P7C), w(PH2), d(CCC) n[P7C(Gg 0)] 883, sh 838, ma 770, m 700, vw a 680, m 640, sh 880, m, p 840, m, p 768, w, p 704, s, p 676, s, p 636, sh, p 628, sh, p 624, s, p 648, wc 357, s, p 348, mb 625, m 647, wc 355, s 348, s 285, w 284, w, p n[P7C(Gg)] n[P7C(Tg)] r(CH2P), t(PH2) d(CCC), n(P7C), d(PCC) d(CCC), n(P7C), d(PCC), ns[CCC(Gt)] w(C7CH3), w(P7C), w[C7C2H5(Tg,Gg 0)] 268, sh b 230, w, p 215, vw b 214, vw c w(C7CH3), d(PCC), w[C7C2H5(Gg)] w(C7CH3) d(PCC), d(CCC) d(PCC), d[CCC(Gt)] w(P7C) w[P7C(Gt)] w(C7C2H5) 239, w 227, sh 215, sh 185.4, s 177, s 105, m 126, vw c750 Table 4.Interpretation of vibrational spectra of isopropylphosphine. Raman (g) IR (g) No. Sym- Assignment metry 2299, vs 22 A00 2299, vs 2288, vs 2297, vs 5 A00 nas[PH2(G)] nas(PH2) ns[PH2(G)] ns(PH2) 9 A d(CH), w(PH2) A00 1256, m 1234, w 26 r(CH), nas(CC2), r(CH3) r(CH), nas(CC2), r[CH3(G)] nas(CC2), r(CH) r(CH3), d(CH), ns(CC2) d(PH2) *970 a 27 10 11 12 28 29 13 A00 A0 A0 A0 A00 A00 A0 892, m 813, m 792, w 14 30 A0 A00 606, s 1066, s *970 a 924, vw 892, w 813, s 791, s 809, m 628, m, sh 626, vw 606, w 388, w 325, w 15 16 17 A0 A0 A0 2297, vs 2286, vs 1255, m 1237, w 1231, w 1164, vw 1114, m 1077, m r(CH3), n(P7C), w(CC2), d(PH2) 1067, m r(CH3) r(CH3), d(CH) ns(CC2), r(CH3), d(CH) w(PH2), d[CH(G)] w(PH2), d(CH) t(PH2) n(P7C), w(CC2), d[CC2(G)] n(P7C), w(CC2), d(CC2) d(CC2), w(CH3) w(CC2), w(CH3) w(CC2), w[CH3(G)] r(CC2), w(CH3) w(CH3) w(CH3) w(P7C) 325, m 309, vw 322, w, sh 322, w 287, w 266, vw, br 163, s 31 18 32 33 A00 A0 A00 A00 163, vw Note.The number of vibration, symmetry type and assignment are given for the T conformer. The lines of the G conformer (denoted by character `G') were interpreted only if their frequencies did not coincide with the corresponding lines of the T conformer. a The frequency was suggested after normal coordinate analysis 22 (was not observed experimentally). CH3 CH3 CH3 CH2 CH2 CH2 P P H3C H3C CH3 CH2 PCH3 CH2 CH2 CH2 CH2 H2C CH3 CH3 GGG TG0G TGG The GGG and TG0G conformers are present simultaneously in the crystalline phase.45, 46 Theoretical interpretation of the spec- tra 46, 47 of triethylphosphine 9 was based on the results of normal coordinate analysis 45 performed using the force constants for molecules 8 and 4.Comparison of the spectra of primary alkylphosphines reveals three vibrations, viz., the nas(PH2), ns(PH2) and d(PH2) ones, which seem to be the most characteristic vibrations of these compounds. The first two vibrations appear as two very strong bands with similar frequencies near*2300 cm71 in the gas-phase spectra. On going from gases to liquids, these bands usually coalesce into one band observed in the frequency range 2275 ± 2300 cm71 (see, e.g, Refs 15, 18 ± 20, 23, 24, 48, 49).A normal coordinate analysis shows that the vibrations nas(PH2) and ns(PH2) are the characteristic vibrations, i.e., that the fre- quencies and intensities of the corresponding spectral peaks are determined solely by the properties of the P7H bonds. Scissoring vibrations of the phosphino group, d(PH2), can mix to a certain degree with vibrations of the alkyl radical (see, e.g., Table 2), which affects mainly the intensities of corresponding bands. At the same time, the frequency of the d(PH2) vibration depends only slightly on the vibration of the alkyl fragment of the molecule and lies in a relatively narrow frequency range between 1073 and 1097 cm71 (for liquids).15, 18 ± 20, 23, 24, 48, 49 Mixing of the scissoring deformation vibration d(PH2) with the wagging vibra- tion w(PH2) also occurs (see Table 2).Probably, that is why the former vibration has been interpreted 20 as w(PH2). The structure of the alkyl fragments of the molecules of primary alkylphosphines strongly affects the frequency of vibra- tion with predominant contribution of the P7C bond (Tables 1 ± 5). For instance, the frequency of the vibration Table 5. Stretching vibrations of P7C bonds in the spectra of alkylphos- phines and their substituted derivatives. Molecule CH3PH2 CH3PCl2 CH3PF2 C2H5PH2 C2H5PCl2 C2H5PF2 n-C3H7PH2 n-C3H7PCl2 iso-C3H7PH2 (CH3)2PH (CH2)2PH (CH3)2PCl (CH3)3P (C2H5)3P Note. Notations: cr stands for crystal and s for solution (in Kr). a Ring `breating' vibration accompanied by synchronous changes in the P7C and C7C bond lengths.b Symmetric ring deformation accompanied by synchronous lengthening of two P7C bonds and shortening of the C7C bond. c Non-symmetric ring deformation associated with nas(PC2). d Vibrations of the A0 symmetry type. e Vibrations of the A00 symmetry type. S A Katsyuba IR Raman Assignment n(P7C), w(PH2) n(P7C), ds(CH3) 673, w, p (l) 691, w, p (l) n(P7C), w(PF2), das(CH3) 709, w, p (l) 676, s (g) 690, s (cr) 727, w (cr), 711, m (cr) 634, w (g) 666, s (g) 632, m (g) 680, s (g) 651, m (g) 636, vs (g) 663, m, p (l) 631, m, p (l) 682, m, p (g) 655, s, p (g) 700, w (s) 680, m (g) 704, w, p (l) 676, w, p (l) 640, sh (g) n(P7C), d(PCC) n(P7C), d[CCP(G)] n(P7C), d[CCP(T)] n[P7C(G)] n(P7C), d(CCP), w[CH2(T)] n[P7C(Gt)] n(P7C), w(PH2), d[CCC(Tt)] n[P7C(Gg 0)] 636, sh, p (l) 628, sh, p (l) 624, w, p (l) 736, w, p (l) 687, w, p (l) 625, m (g) 738, w (g) 690, w (g) n[P7C(Gg)] n[P7C(Tg)] n[P7C(Gt)] n(P7C), d(CCC), d[CCP(Tt )] n(P7C), r[CH2(Gg 0)] 650, w (g) 626, vw (g) 650, w, p (l) 628, m, sh (g) 606, w (g) 606, s (g) 704, (g) n(P7C), w(CC2), d[CC2(G)] n(P7C), w(CC2), d[CC2(T )] nas(PC2) 660, m, p (g) 704.5, s (g) 667, w (g) 660, w (g) ns(PC2) 1058, vs (g) 1056, s, p (g) w(CH2), ring `breating',a r(PH), symmetric deformation b t(CH2), r(CH2), non-sym- 650, m, dp (g) 657, s (g) metric deformation c r(PH) nas(PC2) ns(PC2) 708, m, dp (l) 674, w, p (l) 708, m, dp (g) nas(PC3) 707, s (l) 675, s (l) 716, m (g) 709, m (g) 706, m (g) 653, vs, p (g) 690, s (l) 670, m (l) ns(PC3) nas(PC3), d[PCC(TGG0)] d 697, m (l) 669, s (l) 619, vs (l) nas(PC3), d[PCC(GGG,TGG)] nas(PC3) e ns(PC3), d[PCC(GGG)] ns(PC3), d[PCC(TGG0)] 659, m (l) 625, m (l) 619, m (l) 600 m (l) ns(PC3), d[PCC(TGG)]Vibrational spectra, force constants and conformations of molecules of trivalent tricoordinate phosphorus compounds n(P7C) decreases by*70 cm71 on going from the molecule 1 to the molecule 6 (see Tables 1 and 4).According to calculations, this is due to not only the `weighting' of the alkyl group, but also a decrease in the force constant KP7C . The n(P7C) vibration can mix with deformation vibrations of the phosphino group and alkyl radical.Because of the changes in the contributions of particular groups to the vibrations, the intensities of the corresponding bands in the IR spectra can vary over a wide range on going from one alkylphosphine to another, which severely hampers their assignment. It is much easier to use the Raman spectra for identification of the vibration n(P7C), since the corresponding lines in the spectra of compounds in question always are the most intense and polarised in the frequency range 550 ± 705 cm71. Thus, it is the Raman spectra that should be used for easy discrimination between different primary alkylphosphines. When assigning the n(P7C) vibration, it should be remem- bered that the corresponding frequency can be `sensitive' to molecular conformations.Therefore, a number of `extra' peaks is often observed in the vibrational spectra of compounds, whose molecules can adopt several conformations (see, e.g., Tables 2 and 5). Some of the `extra' peaks disappear as a result of crystallisation of the specimens. Like the n(P7H) vibrations of primary alkylphosphines, those of the secondary alkylphosphines are observed in nearly the same spectral region 2270 ± 2298 cm71. On the other hand, lines corresponding to two (rather than one) stretching vibrations of the P7C bonds are observed in the spectra of secondary phosphines (see Table 5). An intense, strongly polarised Raman line observed in the region 550 ± 670 cm71 is associated with symmetric ns(PC2) vibrations, while a somewhat weaker and weakly polarised line in the region 600 ± 710 cm71 is attributed to antisymmetric nas(PC2) vibrations.In the IR spectra, the latter band is usually stronger than the former. As for the primary phosphines, the n(P7C) frequencies are, as a rule, `conformation- ally sensitive' and decrease substantially as the chain length of alkyl radicals increases. All the aforesaid refers to acyclic dialkyl- phosphines and unstrained cyclic systems. Closure of three-, four- and five-membered strained rings leads to an increase in the ns(PC2) frequencies, so that they become higher than the nas(PC2) frequencies and lie beyond the upper bounds of the corresponding spectral regions.The ns(PC3) and nas(PC3) stretching vibrations of the P7C bonds in the molecules of tertiary phosphines have spectral features similar to those mentioned above (see Table 5). Since the lines corresponding to some characteristic vibrations such as n(P7H) or d(PH2) are absent in this case, identification of the spectra of tertiary phosphines should be performed using both the ns(PC3) and nas(PC3) lines and the alkyl group bands. The frequencies of some vibrations of the methyl or ethyl radicals at the phosphorus atom are sufficiently characteristic, which makes it possible to use them in the spectral analysis. For instance, the bands associated with symmetrical ds(CH3P) and antisymmetrical das(CH3P) deformation vibrations are observed in the IR spectra of any liquid alkylphosphines in the regions *1280 ± 1295 and *1410 ± 1445 cm71, respectively.The r(CH3P) rocking vibra- tions mix with vibrations of other groups at the phosphorus atom (see Table 1). Changes in the environment around the P atom lead to substantial changes in the frequency and eigenvector of the resultant complex vibration. Therefore, the corresponding bands in the IR spectra of liquid alkylphosphines are observed in a rather wide region between 940 and 985 cm71 and their intensities can vary from weak to strong. In the Raman spectra of alkylphos- phines, the lines attributed to all the three vibrations are usually of weak or medium intensity. If the ethyl radical is bound to the phosphorus atom, the d(CH2) bending and t(CH2) twisting vibrations of the methylene group can be considered characteristic.The corresponding peaks in vibrational spectra are usually of medium intensity and observed in the spectral regions *1410 ± 1430 and 751 *1230 ± 1250 cm71, respectively. A complex vibration with pre- dominant contribution of the wagging motion, w(CH2), is char- acterised by a slightly higher frequency (*1240 ± 1280 cm71). Since this vibration is strongly coupled, the intensity of the corresponding peak can vary from very weak to medium. On the contrary, vibrations of the C7C bond are characteristic and associated with rather strong peaks in the region *1020 ± 1065 cm71. Rocking and wagging vibrations of the methylene group mix with the motions of other atoms constituting the alkylphosphine molecule and usually give rise to several peaks in the spectral region*690 ± 820 cm71.At least one of the peaks is of rather high intensity. It should be noted that the above-mentioned frequencies are specific to not only alkylphosphines, but also other organophos- phorus compounds,1± 5 the molecules of which contain the CH3P or C2H5P fragments. This specificity is due to both the mass of the phosphorus atom and the peculiarities of the electronic structure of these fragments, which affect the force constant values. Bands corresponding to the bending and twisting vibrations of the methylene group at the phosphorus atom of an arbitrary coordi- nation are observed in the spectral regions *1430 ± 1380 and *1190 ± 1220 cm71, respectively, and can be used for the identi- fication of non-branched alkylphosphorus groups.It is more convenient to use the Raman spectra for identification of twisting vibrations, since the corresponding bands in the IR spectra are often weak. The above-mentioned frequency ranges for the `fingerprint' bands are given for the liquid-phase spectra, since they are most widely used in routine studies. However, these spectral regions are only slightly shifted on going to the spectra of gases and crystals. For alkylphosphines, the shifts of spectral peaks due to the change in the state of aggregation usually do not exceed 10 ± 15 cm71. The band intensities also vary only slightly, which indicates the absence of strong intermolecular interactions in the condensed phases.It should be taken into account that crystallisation of alkylphosphines, whose molecules can adopt several conforma- tions, can lead to disappearance of some spectral bands. III. Alkyl(halogeno)phosphines Almost no interactions between the vibrations involving relatively heavy halogen atoms and light hydrogen atoms occurs in the molecules of alkyl(halogeno)phosphines. Therefore, skeletal vibrations in these compounds are much less `sensitive' to sub- stitution of deuterium atoms for hydrogen atoms, so that deuter- ation, which often facilitates the interpretation of the spectra, appears to be ineffective. Strengthening of intermolecular inter- actions due to the presence of halogen atoms in the molecules is yet another objectionable circumstance.This can cause shifts of spectral peaks comparable in magnitudes with those caused by structural rearrangement, which precludes the interpretation of experimental results. Comparison of the spectral characteristics of the P7F stretching vibrations for different states of aggregation of difluo- ro(methyl)phosphine (10) 50, 51 makes it possible to assess the effect of intermolecular interactions on these vibrations (Table 6). Vibrational spectra of compound 10 and dichloro(me- thyl)phosphine (11) 52 ± 56 have been interpreted.51, 56, 57 The results obtained in the two last-mentioned studies differ mainly in the assignment of the n8 and n9 frequencies.In Ref. 57, they were assigned to the vibrations d(PCl2) and w(PCl2), respectively, while the opposite assignment has been given in Ref. 56. The latter attribution is in better agreement with the vibrational assignments for the molecule C2H5PCl2 (12) based on the results of ab initio quantum-chemical calculations.58 Difluoro(ethyl)phosphine (13) and dichloro(ethyl)phosphine exist in the gas and liquid phases as mixtures of the T and G conformers 21, 56, 58 ± 60 (see Fig. 1; X=F, Cl; Y=CH2; Z=CH3). Vibrational spectra of the former substance and its deuterio derivative have been interpreted 60 and it was shown that752 Table 6. Characteristic frequencies of the stretching vibrations of PF2 bonds in the spectra of several organophosphorus compounds.Assignment Molecule CH3PF2 ns(PF2), r(CH3) nas(PF2), r(CH3) C2H5PF2 ns(PF2) nas(PF2), r(CH2) ClCH2PF2 812, w, p (g) 795, w, p (l) 789, w (cr) 800, w, dp (g) 769, w, dp (l) 755, w (cr) 819, m, p (g) 796, w, dp (g) ns(PF2), n(CCl), n(P7C) 849, m, p (l) 790, m, p (l) CF3PF2 ns(PF2), n(CCl), n[P7C(G)] nas(PF2), r(CH2) ns(PF2) nas(PF2) crystallisation of the specimens leads to the `freezing out' of the gauche conformers. Unlike difluoro(ethyl)phosphine, both con- formers of dichloro(ethyl)phosphine exist in the crystalline state.21, 56 Vibrational spectra of C2H5PCl2 (see Refs 21, 56) and C2D5PCl2 (see Ref. 21) have been analysed repeatedly.21, 56, 58 It is believed that comparison of the results obtained in the ab initio calculations 58 of the vibrations of heavy atoms of molecule C2H5PCl2 with assignments 21 of the vibrations with predominant contribution of hydrogen atoms provides a version of the inter- pretation of the spectra of compounds 12, which does not contra- dict experimental results 21, 56, 58 (Table 7).The molecule of dichloro(propyl)phosphine, n-C3H7PCl2 (14), can adopt five stable conformations (see Fig. 2; X=Cl, Y=CH2). Analysis of the vibrational spectra of compound 14 showed 61 that the Tt, Gt and Gg 0 conformers exist in the liquid phase. The interpretation of the spectra of dichloro(propyl)phos- Table 7. Interpretation of vibrational spectra of dichloro(ethyl)phosphine. Assignment No. Symmetry 8 A0 9 b 19 b A0 A00 20 A00 10 11 21 12 A0 A0 A00 A0 13 22 A0 A00 14 23 A0 A00 24 n(CC) n[CC(G)] r(CH3), d(CCP) r(CH3), r(CH2) r(CH2), r[CH3(G)] r(CH2), r(CH3) n(P7C), d[CCP(G)] n(P7C), d(CCP) ns(PCl2), d(CCP) nas(PCl2) d(CCP), ns(PCl2) d[PCl2(G)] d(PCl2) t(PCl2) t[PCl2(G)] w[PCl2(G)] w(PCl2) w(CH3) w[CH3(G)] w(PCl2) A00 a Vibrations with frequencies higher than 1200 cm71 are not listed.b The reverse assignment of the n9 and n19 vibrations has also been reported.18 c From the gas-phase spectrum. IR Raman 798, s (cr) 767, s (cr) 830, vs (g) 819, vs (g) 859, vs (g) 826, vs (g) 799, s (g) 864, s (g) 848, sh, w (g) 813, sh (l) 861, m, p (g) 850, m, dp (g) IR (g) a Raman (l) a 1038, w, p 1022, w, p 990, vw, p 750, vw, p 663, m, p 631, m, p 499, vs, p 478, s, p 400, m, p 320, m, p 213, m, p 191, m, dp 165, w, p 286, m, p 178, w, p 1040, m 1026, m 994, w 975, vw, dp 976, vw 752, s 727, vw, dp 728, m 666, s 632, m 506, vs 498, vs 403, s 321, m 213, m 185, s 156, w 287, s 175, s, sh 184, w 180, w 113, w 83, vw c Table 8.Interpretation of vibrational spectra of dichloro(chloromethyl)- phosphine. No. Symmetry Assignment A00 124 A0 A00 5 A0 136 A0 7 A0 8 A0 14 A00 9 A0 15 r(CH2), t[CH2(G)] r(CH2) n(CCl), d(PCCl), n[P7C(G)] n(P7C), d(PCCl), d(CH2), d[PCl2(G)] n(P7C) nas(PCl2), ns[PCl2(G)] ns(PCl2), nas[PCl2(G)] d(PCCl) d(PCCl), d(PCl2), w(CH2), t[PCl2(G)] w(PCl2), d(PCl2), ns[PCl2(G)] w(PCl2) d(PCl2) t(PCl2), d(PCCl), w[CH2(G)] d(PCl2), n(P7C), d(PCCl), w[CH2(G)] w(P7C) A00 Note.The number of vibration, symmetry species and tentative vibrational assignments are given for the T conformer only. Detailed assignment is given for the G conformer only. phine was based on the results of ab initio quantum-chemical calculations.61 molecules The (chloromethyl)difluorophosphine of ClCH2PF2 (15) and (chloromethyl)dichlorophosphine ClCH2PCl2 (16) exist 62 ± 65 both in the liquid and gas phases as mixtures of the T and G conformers (see Fig. 1; X=F, Cl; Y=CH2; Z=Cl). Crystallisation of these substances leads to the `freezing out' of one isomeric form, so that the crystals of compound 16 comprise the molecules of only the gauche conformer,64 while the molecules of compound 15 comprise the trans conformer.63 The spectra of both compounds were interpreted using the force fields obtained from normal coordinate analysis for the corresponding deuterio derivatives.63, 64 The assignments suggested 64 for compound 16 (Table 8) differ from those given earlier,56, 66 which is likely due to impurities present in the specimens studied.56, 66 In studying the vibrational spectra of (CH3)2PCl (17) it has been shown 49, 67 that the molecules of this compound form asymmetrical dimers in the crystals.This leads to the doublet splitting of a large number of singlet (in the spectra of liquid) bands and lines (Table 9).The components of the doublets Table 9. Characteristic frequencies of the stretching vibrations of PCl bonds in the spectra of several organophosphorus compounds. Raman Assignment Molecule CH3PCl2 C2H5PCl2 483, vs, p (l) 483, vs, p (l) 499, vs, p (g) 478, s, p (g) n-C3H7PCl2 ClCH2PCl2 CF3PCl2 ns(PCl2), d(PCl2) nas(PCl2), t(PCl2) ns(PCl2), d(CCP) nas(PCl2) ns(PCl2), nas[PCl2(Gt)] ns[PCl2(Tt )] nas(PCl2) nas(PCl2), ns[PCl2(G)] ns(PCl2), noà s[PCl2(G)] nas(PCl2), das(CF3) ns(PCl2), das(CF3) 497, vs, p (l) 479, s, p (l) 494, vs (l) 494, vs, p (l) 523, w, dp (l) 516, vs, p (l) 462, s, p (l) (CH3)2PCl n(PCl), w(PC2) 301, w, p (l) w(PC2), n(PCl) S A Katsyuba IR (g) Raman (l) 812, w 816, m 776, vw, p 774, sh 757, m 688, s 746, w 683, s, p 620, vw, p 623, vw 508, vs 498, vs 365, w 274, vs 494, vs 494, vs, p 364, vw 274, w 256, vs 256, vs, p 228, vw 203, w, p 143, w 187, s 136, s 187, s 85, w 90, vwIR 479, vs (cr) 479, vs (cr) 506, vs (g) 498, vs (g) 512, vs (g) 498, vs (g) 482, s, sh (g) 508, vs (g) 498, vs (g) 526, vs (g) 520, vs (g) 533, vw, sh (cr) 460, s (cr) 306, m (cr)Vibrational spectra, force constants and conformations of molecules of trivalent tricoordinate phosphorus compounds correspond to the in-phase and antiphase molecular vibrations in the dimers.Normal coordinate analysis for the molecule 17 57 was based on the known experimental data,67 which are somewhat different from those reported earlier.49 Since vibrational spectra of CF3PH2 (18) and its deuterium- substituted analogue revealed no reduction of the C3v local symmetry of the CF3 group due to the addition of the PH2 group, a normal coordinate analysis for this molecule was performed assuming this symmetry.An interpretation of the vibrational spectra of molecule 18, based on the results obtained by BuÈ rger et al.,68 is given in Table 10; however, the description of the vibrations of the PH2 fragment is somewhat changed to accord with the true Cs symmetry of the entire molecule. Table 10. Interpretation of vibrational spectra of (trifluoromethyl)phos- phine. IR (g) Raman (g) Assignment No. Symmetry 10 A00 1 A0 2 A0 11 A00 3 A0 4 A0 12 A00 5 A0 6 A0 7 A0 13 A00 8 A0 9 A0 14 2348,vs 2305, vs 1187, vs 1187, vs 1154, vs 1072, s 833, vs 833, vs 743, m 525, m 525, m 419, s 284, w 284, w A00 nas(PH2) ns(PH2) nas(CF3), das(CF3) 1184, w nas(CF3), das(CF3) 1184, w ns(CF3), ds(CF3), n(P7C) 1140, w d(PH2) 1072, w t(PH2) 848, m w(PH2) ds(CF3), ns(CF3), n(P7C) 745, s das(CF3) 524, w das(CF3) 524, w n(P7C), ds(CF3) 422, vs r(CF3) 290, w r(CF3) 290, w Spectra of (CF3)2PH (19) and its deuterio analogue, as well as those of (CF3)3P (20) have been interpreted on the basis of normal coordinate analysis.69, 70 Vibrational spectra 54, 71 of compounds CF3PF2 (21), CF3PCl2 (22), CF3PBr2 (23) and CF3PI2 (24) have been analysed.72 Unlike vibrational spectra of unsubstituted alkylphosphines, those of alkyl(halogeno)phosphines are strongly dependent on the state of aggregation of the substances (see Tables 1 ± 10). This can be rationalised by self-association of the molecules.Since routine spectral experiments usually do not include strong heating or cooling of specimens, the spectral indications of particular vibra- tions presented below are given for those states of aggregation, in which the compounds under study exist at ambient temperature. Compared to unsubstituted compounds, spectral regions corresponding to vibrations of structural fragments constituting the molecules of alkyl(halogeno)phosphines become much wider. This is due to not only intermolecular interactions, but also relatively strong coupling of vibrations and strong substituents effects.Mention has been made 1 that the IR spectra of com- pounds of the type RPF2 reveal two strong bands in the frequency ranges 806 ± 869 and 860 ± 947 cm71, which correspond to the vibrations nas(PF2) and the ns(PF2), respectively (see Table 6). As can be seen in Table 6, only the frequencies in the spectra of molecule 21 fall within the above-mentioned limits. Probably, a tendency of compounds containing the PF2 group to undergo an autocatalytic disproportionation 73, 74 can be a reason for viola- tion of the known correlations.1 Analysis of the IR spectra of alkyl(difluoro)phosphines showed that both stretching vibrations of the PF2 group are observed in the same characteristic region *780 ± 830 cm71.When assigning the vibrations, one should take into account that the n(PF2) frequencies can increase sub- stantially on going from the liquid to vapour phase. In addition, our experience suggests that in the Raman spectra the ns(PF2) lines are of medium or weak intensity and are polarised, while the nas(PF2) lines are weak and depolarised (though quantitative 753 differences between the degrees of depolarisation of these lines can be rather small). Introduction of halogen atoms into mole- cules causes an increase in the force constants of the P7F bonds, so that the n(PF2) frequencies can be as high as 870 cm71. The frequencies of stretching vibrations of the PCl2 groups lie in the region 420 ± 526 cm71 (see the monograph 1).For alkyl(ha- logeno)phosphines, they are usually close to the upper bound of this spectral region (*450 ± 526 cm71). The frequencies of the vibrations ns(PCl2) of alkyl(dichloro)phosphines are somewhat higher than those of the vibrations nas(PCl2) [the opposite is observed for molecules 16 and 22 (see Table 9)]; in the spectra of liquids the corresponding peaks usually coalesce into an unre- solved contour. These compounds are characterised by an increase in the frequencies of stretching vibrations of the PCl2 group on going to the gas phase. However, both the IR spectral bands and the Raman lines remain very strong, which facilitates the identification of these n(PCl2) vibration in spite of their strong coupling with other vibrations (see Table 9).The vibrations n(P7Cl) of the molecule 17 are strongly mixed with the rocking vibrations of the (CH3)2P group, which gives rise to two co-operative vibrations. Among them, the vibration with the higher frequency can be rather arbitrarily associated with the n(P7Cl) stretch (see Table 9). Analysis of scarce spectral data for dialkyl(chloro)phosphines suggests that the frequencies of the vibrations n(P7Cl) of these compounds lie in nearly the same spectral region as those of the vibrations n(PCl2) (see above). Therefore, the most pronounced distinction between the corre- sponding spectra is reduced to the appearance of two vibrations, viz., ns(PC2) and nas(PC2), in the spectra of dial- kyl(chloro)phosphines instead of one n(P7C) vibration in the spectra of alkyl(dichloro)phosphines.The frequencies of the vibrations n(PC2) and n(P7C) in the spectra of alkyl(halogeno)- phosphines (see Table 5) are observed near the upper bounds of the corresponding spectral regions (see Section II) for alkylphos- phines. (The molecules containing the CH2Cl and CF3P groups are exceptions to this rule. Their spectra are discussed below.) The intensities of the corresponding bands in the IR spectra of alkyl(halogeno)phosphines are usually higher than those in the spectra of unsubstituted analogues. Other parameters of these spectral peaks are similar to those discussed above (see Section II). Spectral indications of theCH3P and C2H5P groups are nearly identical for alkylphosphines and their halogeno-substituted derivatives. The exceptions are the vibrations with predominant contribution of the r(CH3P) vibration, since their frequencies decrease from *940 ± 985 down to *880 cm71 upon introduc- tion of halogen atoms to the phosphorus atom in the last-named molecules.The intensities of the corresponding bands in the IR spectra are rather high. The peaks in the region 1380 ± 1410 cm71, corresponding to the d(CH2) vibrations, are characteristic of the CH2Cl group. (This spectral region is characteristic of the vibrations of CH2X groups, where X is a rather heavy atom; e.g., X=P or S). Stretching vibrations of the C7Cl bond of the CH2Cl group strongly interact with the vibrations n(P7C) and give rise to two new vibrations.Among them, the vibration with the lower frequency is observed in the region typical of the vibration n(P7C) and has all the specific spectral indications (see Section II). The peaks corresponding to the second stretching vibration of the PCCl group are observed in the region*740 ± 790 cm71 and have variable intensities. The vibrations of the CF3P group are readily identified in the spectra of any organophosphorus compounds. Mention has been made 4 that very strong IR absorption bands in the region 1100 ± 1300 cm71 correspond to the stretching vibrations of trifluoromethyl groups. For phosphines, this correlation has specific details. Two very strong bands in the region 1140 ± 1235 cm71 of the IR spectra of phosphines can be arbitrarily assigned to the vibrations nas(CF3) (Table 11), which result from the mixing of the stretching and deformation [das(CF3)] vibrations.They are754 Table 11. Vibration nas(CF3) in the spectra of organophosphorus com- pounds. IR Raman Assignment Molecule CF3PH2 1184, w (g) 1184, w (g) CF3PI2 CF3PBr2 CF3PCl2 CF3PF2 1184, m (l) 1156, sh (l) 1200, w (l) 1184, m (l) 1222, m (g) 1166, m (g) (CF3)2PH (CF3)3P 1176, w, br (g) 1187, vs (g) 1187, vs (g) 1177, vs (g) 1140, vs (g) 1190, vs (g) 1160, sh (g) 1200, vs (g) 1176, m (g) 1222, vs (g) 1170, sh (g) 1212, vs (g) 1180, vs (g) 1235, s (g) 1189, s (g) nas(CF3), das(CF3) (k) nas(CF3), das(CF3) (?) nas(CF3), das(CF3) (k) nas(CF3), das(CF3) (?) nas(CF3), das(CF3) (k) nas(CF3), das(CF3) (?) nas(CF3), das(CF3) (k) nas(CF3), das(CF3) (?) nas(CF3), das(CF3) (k) nas(CF3), das(CF3) (?) nas(CF3), das(CF3) nas(CF3), r(CF3) nas(CF3), das(CF3) (k) nas(CF3), das(CF3) (?) Note.Hereafter (k) and (?) respectively denote the vibrations, which occur parallel and perpendicular to the molecular symmetry plane. localised within one CF3P group and do not interact kinematically with the motions of adjacent fragments. Therefore, variations of the frequencies of the vibrations nas(CF3) are due to dynamic reasons alone and, hence, reflect the changes in the electronic structure of the phosphorus-containing fragment of the molecule. The nas(CF3) frequencies increase as the electronegativities of substituents at the phosphorus atom increase (see Table 11). Symmetrical stretching vibrations ns(CF3) mix with the bend- ing vibrations ds(CF3) and the vibrations n(P7C) (see Table 10).The involvement of the phosphorus atom in the vibrations provides the possibility for kinematic interactions between adja- cent groups to occur, which is particularly effective in the case of equal `individual' frequencies. That is why the number of bands corresponding to the vibration ns(CF3) in the IR spectra of compounds with one CF3P group and with two (or three) CF3P groups is different, namely, one very strong band in the region 1100 ± 1160 cm71 vs. two bands in the same spectral region, respectively.Previously (see Section II), particular attention was paid to the vibrations n(P7C) and it was shown that the corresponding Raman lines have distinctive spectral indications, which are common to various organophosphorus compounds with the P7C(sp3) valence bonds. Unfortunately, compounds with the CF3P group are exceptions to this rule, since all attempts to discriminate between the vibrations with predominant contribu- tion of the P7C bond and other normal coordinates have failed. The vibrations denoted 4 as n(P7C) are in fact coupled vibrations with nearly equal contributions of the vibrations ds(CF3) and ns(CF3) and with a some contributions of the vibrations n(P7C) (see Table 10). The corresponding bands in the vibrational spectra of organophosphorus compounds are observed in the region between 650 and 750 cm71.For phosphines, this region is much narrower, namely, one or two strong IR absorption bands are observed in the frequency range *720 ± 750 cm71. The corre- sponding Raman lines are polarised and intense. Vibrations of the P7C bonds make somewhat greater con- tributions to the coupled vibrations denoted 4 as d(C7F). The corresponding IR absorption bands and Raman lines (*410 ± 470 cm71) are rather strong, the latter being polarised, which facilitates vibrational assignments. We suggest that the Raman lines of medium or weak intensity, which are observed in the region 520 ± 570 cm71 and correspond to the vibrations with the predominant contribution of the vibrations das(CF3), can also be used for identification of com- pounds under consideration.The intensities of the das(CF3) bands in the IR spectrum can vary from very weak to strong. S A Katsyuba IV. Phosphines with unsaturated groups at the phosphorus atom To date, a normal coordinate analysis has been performed only for three groups of the trivalent phosphorus compounds with unsaturated fragments. These are the molecules with the P7C=C fragment, ethynylphosphines and cyano derivatives of phosphines. The first group comprises alkenyl(dichloro)phosphines. Their molecules can adopt both the cis (C) and anti,gauche (aG) conformations. CRR0 H P P CRR0 Cl Cl Cl Cl HC aG According to the vibrational spectroscopy data,75 the cis conformers of the molecules CH2=CHPCl2 (25) and (CH3)2C=CHPCl2 (26) are dominant in the condensed phase, whereas liquid dichloro(styryl)phosphine (27) is a mixture of both conformers.Because of steric hindrances, in the molecule of S-ethyl tert-butyl(1,2-dichlorovinyl)phosphinothioite (28) the C=C bond is in cis position with respect to the lone electron pair (LEP) of the phosphorus atom.76 However, it is believed that a mixture of conformers can exist in the liquid owing to internal rotation of the C2H5S fragment Cl Cl (CH3)3C (CH3)3C H H P PC2H5 Cl S Cl SC2H5 B A The interpreted vibrational spectra 75 of the simplest repre- sentative of this group of compounds, viz., compound 25, are listed in Table 12. Triethynylphosphine (29) is a representative of the second group of the trivalent phosphorus compounds with unsaturated fragments.Its molecule has a C3v symmetry and the vibrations of the A2 type are optically inactive and not observed in vibrational spectra. Table 13 lists the experimental 77 vibrational spectra of compound 29 and the assignments 78 based on the results of normal coordinate analysis. Unfortunately, neither the eigenvec- tors of the vibrations nor the potential energy distribution over the Table 12. Interpretation of vibrational spectra of dichloro(vinyl)phos- phine. IR (l) Raman (l) Assignment No. Sym- metry 4 A0 5 A0 n(P7C) d(CH2), n(C=C) d(PCH), r(CH2), d(CH2) g(CH2) r(CH2), r(P7C) 6 A0 13 A00 7 A0 14 A00 8 A0 15 A00 9 A0 n(C=C), d(CH2), d(PCH) 1605, sh, w, 0.20 1592, vw 1595, m, 0.17 1392, m, 0.22 1255, m, 0.21 1028, vw 1010, w w(C=C), g(CH2), g(PCH) 971, w 720, m, 0.42 587, w, 0.74 502, vs, 0.25 473, s, 0.42 348, vs, 0.15 255, m, 0.50 218, m, 0.86 190, s, 0.83 1391, m 1255, w 1037, vw 1011, m 965/976, w 719, s 585, s 504, vs 473, vs 347, w 258, m 218, m 188, m n(PC), r(CH2) g(PCH), t(PCl2) ns(PCl2), d(PCl2) nas(PCl2) r(C=C) d(PCl2) t(PCl2), g(PCH) w(PCl2) A00 A0 A0 A00 A0 16 10 11 17 12Vibrational spectra, force constants and conformations of molecules of trivalent tricoordinate phosphorus compounds Table 13. Interpretation of vibrational spectra of triethynylphosphine.IR (s) b Raman (s) a Assignment No.Symmetry n(C:C) d(C:C) d(C:CH) (k) d(C:CH) (k) 2054, 100, p 2054, 100, p 702, 2 702, 2 654, 1 643, 3 449, 2 d(C:CH) (?) nas(PC3) ns(PC3) d(PC:C) (k) d(PC:C) (k) 273, 4 16 d(PC:C) (?) ds(PC3) das(PC3) 2 A1 10 E 3 A1 11 E 12 E 13 E 4 A1 5 A1 14 E 15 E 6 A1 E 106, 71, dp 2053, m 2053, m 696, vs 696, vs 651, s, sh 639, vs 614, w 444, m 423, m 270, m 131, m, v.br 100.5, vw c a The frequencies were averaged over the spectra of solutions in CCl4 and furan. b The frequencies were averaged over the spectra of solutions in CCl4 and CS2. c From the gas-phase spectrum. vibrational coordinates have been reported,78 therefore the assignments listed in Table 13 seem to be tentative.Vibrational spectra of dicyano(methyl)phosphine (30) 79 and tricyanophosphine (31) 80 ± 82 have been studied. The spectra reported by Goubeau, Haeberle and Ulmer 80 differ appreciably from those reported in the other studies.81, 82 This is probably due to chemical transformations of the substances during the prepa- ration of the samples.82 In the condensed phase, both compounds can be involved in strong intermolecular interactions, which can cause rather large shifts of the spectral peaks (Table 14) and even be responsible for the appearance of new bands as compared to the gas-phase spectra. For instance, an intense peak at *630 cm71 (see Refs 79, 81, 82) in the spectra of liquid and crystalline cyanophosphines and their solutions in CH3CN is not observed in the spectra of vapours or solutions in other solvents.This peak results from intermolecular interactions involving the cyano groups.79, 82 Such strong effects hamper the interpretation of the spectra of these molecules in the liquid or crystalline phase. It should be noted that the molecule 31 has a C3v symmetry and vibrations belonging to the A2 type are optically inactive and not observed in vibrational spectra. The force field of the molecule 31 has been first evaluated by Unnikrishnan and Aruldhas.83 How- Table 14. Interpretation of vibrational spectra of phosphorus tricyanide. IR (cr) Raman (cr) Raman (s) a No. Sym- Assignment Raman (g) metry 2197, s, p 2196, s, p 1 A1 n(C:N) 2210, w 2205, w 2201.2, s 603, s 2190, vw 625, w, p 606, s, p 6 E n(C:N) 2 A1 604, s ns(PC3), d(PCN) (?), ds(PC3) 581, w, dp 577, w 7 E nas(PC3) 3 A1 453, w 452, w 452, w, p d(PCN) (?), 451, w, p 464, w, dp 469, m 468, w ns(PC3), 8 E d(PCN) (?), 442, vw das(PC3) 312, w, dp 318, w 314, w (148, w) (145, w) 128, m, dp 4 A1 E 10 105, s, dp (130, m, dp) (112, vw) 9 E d(PCN) (k), das(PC3) ds(PC3), d(PCN) (?) das(PC3), d(PCN) (k) Note.The frequencies assigned tentatively are given in parentheses. aA solution in CH3CN. 755 ever, calculations performed using this force field have led to unrealistic values of the root-mean-square vibrational ampli- tudes.84 Therefore, Table 14 lists the theoretical interpretation of the spectra of compound 31 according to other studies.84, 85 The results of normal coordinate analysis for the spectra of dicyano- (methyl)phosphine 30 have been reported.86 A common feature of the spectra of the three groups of compounds considered in this Section is that the frequencies of stretching vibrations of the multiple bonds at the trivalent phosphorus atom are lower than those for the corresponding organic molecules.For instance, the frequencies of stretching vibrations of the C=C(P) group (see Table 12) are lower than the lower bound of the spectral region 1620 ± 1680 cm71 typical of alkenes.87 As is known, the introduction of halogens also leads to a decrease in the frequency of the stretching vibrations n(C=C); however, this effect is not so strong as that observed after introduction of the phosphorus atom.For instance, the frequency of the vibrations n(C=C) in CCl2=CHCl is 1589 cm71 (cf. 1539 cm71 for the molecule 28). The same is also observed for the frequency of the vibrations n(C:C) in compound 29 (see Table 13), which is lower than the corresponding frequencies in acetylene and haloacetylenes (*2085 ± 2270 cm71).87, 88 The fre- quencies of the stretching vibrations n(C:N) of molecules 30 and 31 also lie far below the spectral region 2215 ± 2260 cm71 typical of organic cyanides.87 It should be noted that the peaks at *2180 ± 2217 cm71 are `fingerprint' of the PCN groups in the spectra of neat liquids or crystals (see monograph 1). A small overlap of the two spectral regions 1, 87 is of no concern for trivalent phosphorus compounds, since the corresponding fre- quencies of stretching vibrations of the PCN group always lie below 2215 cm71.At the same time, it should be remembered that the n(C:N) frequencies of halonitriles 89 lie nearly in the same region as the n(PCN) frequencies.The n(C=C), n(C:C) and n(C:N) bands in the IR spectra can be of weak intensity, therefore it is more convenient to identify these groups using the Raman spectra in which the corresponding lines always are sufficiently strong and polarised. Raman spectroscopy can also be used to assign the vibrations denoted 87 as d(CH2) and d(C7H), since the lines associated with them are polarised and are of medium intensity.The frequency of the vibration d(CH2) of the vinyl radical at the phosphorus atom is observed in the region 1390 ± 1410 cm71 (cf. 1410 ± 1420 cm71 in the spectra of terminal alkenes). The frequencies of the vibrations d(C7H) of monosubstituted styrenes 87 lie in the region 1295 ± 1310 cm71 [cf. 1335 cm71 for dichloro(styryl)phosphine 27]. The d(C7H) vibrations in the molecules of terminal alkenes are observed in the spectral region 1290 ± 1300 cm71 (see Ref. 87), while the frequency of the vibration d(PCH) of the vinyl group at the phosphorus atom varies between 1250 and 1260 cm71 (see Table 12). The g(CH2) and w(C=C) frequencies also `drop out' of the corresponding spectral regions typical of alkenes.87 Changes in the `fingerprint' vibrational frequencies of alkenylphosphines as compared to those of alkenes are due to the effect of the phosphorus atom on the force constants of the alkenyl fragment rather than kinematic effects.V. Alkoxy derivatives of trivalent phosphorus The spectra of both cyclic and acyclic alkoxy derivatives of trivalent phosphorus have been studied. According to ab initio calculations,90 ± 92 the simplest acyclic molecules such as methyl dimethylphosphinite (32), methyl phosphorodifluoridite (33) and methyl phosphorodichloridite (3) can adopt both the T and G conformations (see Fig. 1; X=CH3, F, Cl; Y=O; Z=CH3). Studies of the vibrational spectra of molecule 32 and its isotopo- mers CD3OP(CH3)2, CH3OP(CD3)2 and CD3OP(CD3)2 showed that both conformers were present in the liquid and gas states,93 while only the trans conformer was retained in crystals.It should be noted that the interpretation of incomplete spectra of methyl dimethylphosphinite 32 (see Seel and Velleman 94) is in agreement756 Table 15. Interpretation of vibrational spectra of methyl phosphorodi- chloridite. IR (g) Raman (l) No. Symmetry Assignment 1181, 15, p A00 A00 1044, 50, p 816, 100, p 778, 5, dp 778, 39, p 536, 40, p 370, 20, p 351, 10, p 218, 30, p 1181, s 1154, 50, dp 1151, ma 1057, vs 829, vs 797, vs 784, m 539, m 370, s 358, sh 214, s 120, m r(CH3), nas(POC) r(CH3) nas(POC), r(CH3) ns(PF2) nas(PF2) ns(POC) r(PF2), d(POC), d(PF2) t(PF2), w(PO) d(PF2) d(POC), r(PF2) w(PO), w(O7CH3), t(PF2) 128,10, dp 81, 14, dp A00 A0 A0 A0 A00 w(O7CH3), w(PO) 5 A0 146 A0 7 A0 158 A0 9 A0 16 10 11 17 18 a From the spectrum of crystals.with the assignments suggested on the basis of ab initio quantum- chemical calculations.90, 93 Both vibrational 91, 95 and microwave 96 spectral studies as well as a gas-phase electron diffraction study 97 revealed no indications of the gauche conformers of the molecule of methyl phosphorodi- fluoridite 33. Therefore, the interpretation of the IR absorption bands and Raman lines in the spectra of compound 33 (Table 15) is based on the results of ab initio calculations of the vibrations of the trans conformer.91 Data on rotational isomerism in the molecule of methyl phosphorodichloridite 3 are contradictory.92, 98 ± 103 In spite of this, the existence of trans conformers of molecule 3 is doubt- less,92, 102, 103 and the interpretation of the spectra of trans con- former was based on the results of ab initio calculations of the vibrations.92 The molecule of ethyl phosphorodichloridite (34) can adopt five stable conformations 104 (see Fig.2; X=Cl, Y=O); how- ever, this has not been confirmed experimentally. Using normal coordinate analysis, it has been shown 104 that the observed spectra can be attributed to the Tt and/or Gt conformers. The interpretation of the spectra (see Table 16) is based on the results of normal coordinate analysis for the Tt conformer.26 The molecule of b-thiocyanatoethyl phosphorodichloridite (35) can adopt twenty-three stable conformations.In the liquid state, compound 35 exists as a mixture of at least three con- formers, while in the crystalline state (at low temperatures) the Table 16. Interpretation of vibrational spectra of ethyl phosphorodichlor- idite. Raman (l) IR (l) a No. Sym- Assignment metry 1022, vw 727, m, 0.6 1095, vw, 0.4 1095, vw 1022, vs 963, s 735, m 735, m 500, s 450, s 430, sh 8 A0 9 A0 10 A0 22 A00 11 A0 12 A0 23 A00 13 A0 14 A0 24 A00 15 A0 25 A00 16 A0 n(OC), n(CC), d(POC), d(OCC) nas(POC), n(CC), w(CH2) n(CC), n(PO), w(CH3) r(CH2), r(CH3) ns(POC), n(CC), d(POC), w(CH2) 727, m, 0.6 493, vs, 0.3 443, m, dp 422, m, 0.8 300, vw 300, vw 192, s, 0.6 150, w, 0.67 125, vw, dp ns(PCl2) nas(PCl2) d(POC), d(OCC), w(PCl2) d(OCC), d(POC), w(PCl2) t(PCl2), w(PO), w(OC) d(PCl2) t(PCl2), w(CC) d(POC), o(PCl2), d(PCl2) a Spectral region below 400 cm71 has not been recorded.S A Katsyuba number of conformers equals at least two.105 Since conforma- tional analysis of compound 35 is too complicated, attempts to determine unambiguously the number and type of observable conformers of molecule 35 have failed. It has only been shown 105 that the mutual gauche-orientation of theO7CandC7S bonds is energetically favourable. The interpretation of the IR absorption bands and Raman lines in the spectra of b-thiocyanatoethyl phosphorodichloridite 35 is based on the results of normal coordinate analysis for all possible conformers.105 A few studies devoted to the normal coordinate analysis of cyclic alkoxy derivatives of trivalent phosphorus has been reported.Five-, six- and seven-membered phosphacycles with the P7O7C fragment have been studied. First, consider the spectra and conformations of five-membered organophosphorus compounds. According to the vibrational spectroscopy data,106 the molecule of 2-chloro-1,3,2-dioxaphospholane (36) adopts a single conformation. Gas-phase electron diffraction study 107 of compound 36 revealed a stable (in the gas phase) conformation of molecule 36 shown in Fig. 3 (a-EO,X=Y=O). Earlier, a normal coordinate analysis for 2-chloro-1,3,2-dioxaphospholane 36 has been performed for a symmetrical `envelope' conformation in which the phosphorus atom deviates from the plane passing through the other ring atoms 106 (a-EP).Next to the publication cited,107 a normal coordinate analysis for the non-symmetrical a-EO conformer (see Fig. 3) has been reported.108 It is this version of the interpretation of the spectra that is listed in Table 17. Y CH2 CH2 Cl P X Y CH2 Y CH2 Cl P X H2C CH2 P X Cl e-KC a-KC a-KY Figure 3. The `envelope' conformations of the molecules of 2-chloro- 1,3,2-dioxa(thia)phospholanes: with apical Y atom and pseudoaxial orientation of the P7Cl bond (a-EY), with apical C atom and axial orientation of the P7Cl bond (a-EC) and with apical C atom and equatorial orientation of the P7Cl bond (e-EC).2-Chloro-1,3,2-oxathiaphospholane (37) exists in the liquid state and in solutions as a mixture of two non-symmetrical conformers shown in Fig. 3 (a-EC and e-EC; X=O, Y=S).108, 109 Their vibrational spectra 108, 109 have been inter- preted theoretically.108 The molecule of 2-chloro-1,3,2-dioxaphosphorinane (38) adopts a single conformation.110 Based on the results of the gas- phase electron diffraction study,111 the conformation a was suggested for its molecule; at the same time, the band-shape Table 17. Interpretation of vibrational spectra of 2-chloro-1,3,2-dioxa- phospholane. Raman (l) Assignment IR (l) a No. 1009, 20, p 1048, sh 1010, s 984, sh 928, s 809, m 764, s 750, sh 597, m 445 s nas(POC) nas(CCO) r(CH2) n(CC), nas(POC), d(CCO) ns(POC), nas(POC), ras(CH2) nas(POC), r(CH2), d(POC) n(PO), r(CH2) das(CCO), das(POC) n(PCl) d(OPO), ds(POC) das(OPCl) 11 12 13 14 15 16 17 18 19 20 21 922, 20, p 805, 100, p 767, 5 754, 5 602, 10, p 443, 70, p 330, 60, p 258, 10, p a Spectral region below 400 cm71 has not been recorded.Vibrational spectra, force constants and conformations of molecules of trivalent tricoordinate phosphorus compounds analysis of the IR spectra of compound 38 in the vapour phase counts in favour of conformation e.110 Cl P P Cl O O CC O O CC C C a e A normal coordinate analysis for both conformers of the molecule of 2-chloro-1,3,2-dioxaphosphorinane 38 has been per- formed.110, 112 The molecule of 4,7-dihydro-2-dimethylamino-1,3,2-dioxa- phosphepin (39) adopts mainly an non-symmetrical conforma- tion,113 CH CH2 HC O H2C P O N CH3 H3C for which theoretical interpretation of the vibrational spectra has been carried out.113 Based on the results of normal coordinate analyses (see Tables 15 ± 17), it is possible to establish the origin of the bands used as spectral indications of the presence of different AlkOP groups in the molecules under study.It is known that the methoxy group at the phosphorus atom is responsible for the appearance of a strong band at*1010 ± 1050 cm71, a strong or medium band in the region *700 ± 800 cm71 and a medium or weak (often multiplet) band in the region *1150 ± 1200 cm71 in the IR spectra.The first two bands are associated with the antiphase and in-phase stretching vibrations of the P7O and O7C bonds, respectively. The former bond makes the major contribution to the symmetric vibration ns(POC), while the latter contributes largely to the antisymmetric vibration nas(POC). Mention may be made that the vibrations ns(POC) can also be readily identified using the Raman spectra, since the corresponding line is of rather high intensity and polarised. Weak Raman lines and IR absorp- tion bands in the spectral region *1150 ± 1200 cm71 are mainly due to the rocking vibrations of the methyl group. Spectral `fingerprints' of the C2H5OP group are well known.1 ±5 These are the strong bands in the IR spectra in the regions *900 ± 970, *1010 ± 1040 cm71 and the bands of medium or weak intensity in the regions *1090 ± 1110 and *1150 ± 1170 cm71.All of them are due to strongly coupled vibrations (Table 16). The vibration with the highest frequency is associated with the rocking motion r(CH2). The frequencies of stretching vibrations of the OCC fragment usually lie in the range *1090 ± 1110 cm71. Strong bands in the IR spectrum are to a certain degree due to the antiphase stretching vibrations of the P7O, O7C and C7C bonds (see Table 16, stretching vibrations Nos 9 and 10). The vibration ns(POC) gives rise to the appearance of rather intense spectral peaks both in the IR and Raman spectra (stretching vibration No.11). They are observed in the region *700 ± 800 cm71, the corresponding Raman line being polar- ised.This spectral region is characteristic of the frequencies of the vibrations ns(POC) for not only the ethoxy group, but almost any alkoxy groups at the trivalent phosphorus atom. Only for cyclic alkoxy derivatives does the region extend to *690 ± 820 cm71. Because of the possibility that the molecules of compounds under study adopt several conformations owing to internal rotation about the P7O bond, the bands corresponding to the vibrations ns(POC) can split into doublets. It should be noted that if a molecule contains two POC groups at the phosphorus atom, their stretching vibrations effectively interact through this atom (see Section I).For instance, the frequencies of the ring `breathing' 757 vibrations corresponding to the synchronous stretch or contrac- tion of the bonds in both POC groups (see Table 17, vibrations No. 15) and to the antiphase vibrations of these fragments (vibrations No. 16) are observed in this region of the spectra of 2-chloro-1,3,2-dioxaphospholane. The IR absorption bands asso- ciated with the antiphase vibrations ns(POC) are usually more intense than those corresponding to the in-phase vibrations. At the same time, the lines of the in-phase vibrations in the Raman spectra are symmetrical, very strong and strongly polarised, while those of the antiphase vibrations are, as a rule, of weak intensity. It should be noted that the ring `breathing' vibrational frequencies of 1,3,2-dioxaphospholanes are always higher than the frequencies of the in-phase vibrations ns(POC) of related acyclic organophosphorus compounds, as well as those of the ring `breathing' vibrations of 1,3,2-dioxaphosphorinanes 112 and other related unstrained phosphacyclanes (provided that the exocyclic groups in the dioxaphospholane and in the larger ring to be compared are the same).A remarkable feature of 1,3,2- dioxaphospholanes is that the frequencies of the ring `breathing' vibrations are higher than those of the antiphase vibrations n(POC). The opposite is observed for 1,3,2-dioxaphosphorinanes and acyclic dialkoxy derivatives. These distinctions can be used for analytical purposes. If the molecules under study have exocyclic P7Cl bonds, then the strong IR absorption bands and very strong polarised Raman lines in the spectral region *440 ± 460 cm71 provide reliable spectral indications of the 1,3,2-dioxaphospholane rings.These spectral bands and lines are due to the stretching vibrations of the P7Cl bond. In the molecules of 1,3,2-dioxaphosphorinanes, the vibrations n(P7Cl) interact with the ring deformation vibrations. Two resultant vibrations appear in the spectra as the intense IR absorption bands and Raman lines in the regions *400 ± 430 and *460 ± 490 cm71. The most characteristic of the spectra of acyclic phosphorochloridites are intense IR absorption bands in the region*490 ± 510 cm71. Thus, the readily identifiable vibra- tions n(P7Cl) can be used for the discrimination not only between 1,3,2-dioxaphospholanes and 1,3,2-dioxaphosphorinanes, but also between these cyclic and corresponding acyclic compounds.The vibrations of the PCl2 group (see Table 16) have marked spectral indications. Symmetrical stretching vibrations ns(PCl2) of the molecules of alkyl phosphorodichloridites appear as very strong IR absorption bands and Raman lines in the region *490 ± 510 cm71. The less intense peaks in the region *440 ± 480 cm71 are associated with antisymmetric vibrations. Identification of these vibrations using the Raman spectra is facilitated by the fact that the lines corresponding to ns(PCl2) are strongly polarised while those corresponding to nas(PCl2) are depolarised or nearly depolarised. It should be noted that the appearance of additional intense peaks in these spectral regions, which are due to the deformation vibrations of non-branched alkoxy groups at the phosphorus atom (see Table 16) prevents easy interpretation. The deformation vibration of the PCl2 group is also character- istic of alkyl phosphorodichloridites.The corresponding line is observed in the region *190 ± 220 cm71 and can be identified with ease in the Raman spectra, since it is always strong and has a depolarisation ratio of about 0.6. It should be noted that vibrational spectra of phosphoro- chloridites change as the state of aggregation of compounds under study changes or on dissolution, which is mainly due to intermo- lecular interactions.The frequencies of the vibrations involving the P7Cl bonds change to the greatest extent. All the above- mentioned characteristic spectral regions refer to the spectra of neat liquids at room temperature. VI. Dialkylphosphoramidites dimethylphosphoramidodifluoridite To date, a normal coordinate analysis has been performed only for four compounds containing Alk2NPIII groups. The simplest molecules, such as758 D-isotopomers have been reported,116 which are somewhat differ- ent from those published previously.114 The molecules of tetramethylphosphorodiamidochloridite and 2-chloro-1,3-dimethyl-1,3,2-diazaphospholane adopt single conformations.126, 127 The equilibrium conformation of the mol- ecule 43 has a Cs symmetry (see Refs 123, 127).CH3 N H2C H2C The vibrational spectra of compound 42 (Table 19) have been interpreted 127 assuming that both dimethylamino groups in the molecule are arranged so that they are nearly eclipsed by the LEP of the phosphorus atom. The results of normal coordinate analysis for this conformer with Cs symmetry are in good agreement with experimental data.127 At the same time, it should be noted that, according to the electron diffraction data,122 molecule 42 has an non-symmetrical conformation. The spectra of compound 43 listed in Table 20 are given in accordance with the known interpretation.127 Analysis of the spectra of phosphoramides based on the results of normal coordinate analysis (see Table 18 ± 20) made it possible to establish the nature of the bands characteristic of the (CH3)2NP group.It is known 1±5 that the bands of medium or high intensity in the frequency range *1260 ± 1320 cm71 are observed in the IR spectra of compounds with these groups. These bands are associated with the vibration in which a symmetric stretch of both N7C bonds [the ns(NC2) vibration] is accompa- nied by contraction of the P7N bond [the n(P7N) vibration]. The frequency of synchronous stretch of the three bonds lies in the range*650 ± 700 cm71. The corresponding IR absorption bands (40) 114 ± 116 and dimethylphosphoramidodichloridite (41),117 adopt single conformations. The structure of molecule 40 was studied by X-ray analysis,118 gas-phase electron diffraction,119 and microwave spectroscopy.116, 120 However, two structures of the (CH3)2NP fragment were proposed, a planar 118 and a pyr- amidal one.119 Later, the results obtained in the former study 119 were critically reviewed 116, 120 since structural parameters of the molecule 40 obtained in more recent studies appeared to be in agreement with the data of the X-ray study.118 The results of gas- phase electron diffraction investigations 121 ± 123 of molecules 41, [(CH3)2N]2PCl (42) and 2-chloro-1,3-dimethyl-1,3,2-diazaphos- pholane (43) also suggested that the dialkylamino group at the trivalent phosphorus atom has a nearly planar structure.It was found that this group in the molecule 40 is oriented parallel to the LEP of the phosphorus atom.116, 118, 120 Based on the results obtained in studies on the vibrational 117 and NMR spectra,124 a similar conformation was suggested for the molecule of com- pound 41.At the same time, it was stated 121 that the (CH3)2NP group is oriented perpendicular to the LEP of the phosphorus atom. Spectra of molecules 40, 41 and fully deuterated derivative of compound 41 have been studied 114, 117 and a normal coordinate analysis was performed 125 for the conformations with eclipsed arrangement of the plane of the dimethylamino groups and the LEP of the phosphorus atom. It was also assumed that the molecules have no symmetry elements because of the non-sym- metrical orientation of methyl groups. The initial empirical interpretation of the spectra of dimethyl- phosphoramidodifluoridite 40 (see Ref.117) was essentially dif- ferent from that based on the results of normal coordinate analysis (see Ref. 125). In addition, the experimental vibrational spectra reported by Durig and Casper 117 contained some `extra' Raman lines and IR absorption bands as compared with the spectra reported earlier.114 Probably, they are due to impurities formed as a result of partial hydrolysis. Because of this, Table 18 lists the experimental results 114 interpreted on the basis of normal coor- dinate analysis.125 After publication of the study cited (Ref. 125), new spectra of dimethylphosphoramidodifluoridite 40 and its Table 19. Interpretation of vibrational spectra of tetramethylphosphor- amidochloridite. Table 18.Interpretation of vibrational spectra of dimethylphosphorami- dodifluoridite. No. Sym- metry IR (g) Raman (l) Assignment No. 1306, m, p 1307, m 1195, m 1103, vw 1071, w, dp? 1073, w 1002, sh a 989, s 814, s 770, s 704, m 501, w, br r(CH3), n(PN), ns(NC2), d(NC2) nas(NC2), r(CH3) r(CH3) r(CH3) r(CH3,) nas(NC2) nas(PNC), r(CH3) ns(PF2) nas(PF2) ns(NC2), n(PN) d(PF2), r(NC2) d(NC2), d(PF2) t(PF2), jb, c r(NC2), d(PF2) w(PF2), r(NC2) j, b w b 989, m, dp? 792, m, p? 743, m, dp? 705, vs, p 495, m 393, w 393, w 336, m 239, w 173, w 97, vw w, b jb, d cA strong band at 343 cm71 in the IR spectrum and a weak Raman line at A0 A00 A0 A00 A00 A0 A0 A00 A0 A00 A00 A0 A00 A0 A0 A0 A00 A0 A00 A0 A00 A0 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 a From the spectrum of crystals.b The character j denotes the out-of-plane deviation of the P7N bond with respect to the CNC plane and the character w denotes torsional vibrations about the P7N bond. 346 cm71 in the spectra of low-temperature crystals were also associated with this vibration.116 A00 A00 13 41 14 42 43 15 16 44 17 45 46 18 47 19 20 21 48 22 49 23 50 24 51 52 d Weak peaks at *116 cm71 in the IR spectra were also associated with ns(NC2)+n(PN) ns(NC2)+n(PN) n(PCl), r(NC2) n(PCl), d(PN2) d(PN2), w(NC2) n(PCl), w(NC2) r(NC2), w(NC2) n(PCl), d(NC2) r(NC2), d(NC2) w(NC2), w(N7CH3) w(NC2), w(N7CH3) w(NC2) a Spectral region below 400 cm71 has not been recorded.this vibration.116 S A Katsyuba Cl P N CH3 Raman (l) Assignment IR (l) a r(CH3), n(PN)7ns(NC2) r(CH3), n(PN)7ns(NC2) 1275, 6, 0.3 1185, vw 1140, vw, 0.8 1090, vw 1060, 6, 0.8 nas(NC2), r(CH3) nas(NC2), r(CH3) r(CH3) r(CH3) r(CH3) r(CH3) r(CH3) r(CH3) ns(NC2)7n(PN), r(CH3) 980, 9, 0.8 ns(NC2)7n(PN), r(CH3) 960, 7, 0.8 695, w, sh 1280, sh 1271, s 1192, s 1140, w 1095, w 1082, vw 1061, m 1029, vs 1029, vw 1029, vs 980, vs 960, vs 698, s 670, 100, 0.12 671, m 488, m, br 415, m, br 490, vw 410, 9, 0.4 410, 9, 0.4 385, 6, 0.4 385, 6, 0.4 335, 67, 0.25 335, 67, 0.25 292, 52, 0.3 292, 52, 0.3 125, 3.7, dpVibrational spectra, force constants and conformations of molecules of trivalent tricoordinate phosphorus compounds Table 20.Interpretation of vibrational spectra of 2-chloro-1,3-dimethyl- 1,3,2-diazaphospholane. Raman (l) Assignment No. Sym- metry 1250, 1.6, p 1238, 1.8, p 1180, 3.3, p 1235, s 1217, s 1205, 2.9, dp 1205, w, sh 1180, vw 1148, s w(CH2), n(P7N) n(N7CH3), t(CH2) n(N7CH3), t(CH2) nas(PNCH3), t(CH2) nas(PNCH3), t(CH2) n(N7CH2), r(CH3), w(CH2) n(N7CH2), r(CH3), w(CH2) 1125, 1.6, 0.7 1085, 0.8, 0.8 1055, w 1030, 9.5, 0.4 1030, s 1010, m 946, s 855, br, sh 692, sh 10 A0 11 A0 36 A00 37 A00 12 A0 13 A0 38 A00 39 A00 14 A0 15 A0 40 A00 41 A00 16 A0 17 A0 18 A0 42 A00 43 A00 19 A0 20 A0 44 A00 45 A00 21 A0 22 A0 46 A00 25 A0 a Spectral region below 400 cm71 has not been recorded. r(CH3) r(CH3) r(CH3), n(C7C) r(CH2) n(N7CH2) n(N7CH2), n(C7C) r(CH2) n pulse n(P7N) ring d n(P7Cl), r(P7Cl) n(P7Cl) r(N7CH3) g(N7CH3), ring w ring d r(N7CH3) w(N7CH3) g(N7CH3) 1012, sh, dp 948, 2, 0.7 855, 1.4, 0.8 830, 0.3 710, 100, 0.08 706, m 692, m 597, 10.5, 0.86 595, w 490, 18.4, 0.34 490, m 392, 12.6, 0.3 355, 11, 0.83 310, sh 300, 29.5, p 268, 27.9, 0.25 245, br.sh 112, 30, 0.6 are, as a rule, rather weak, while the Raman lines are intense and strongly polarised. The vibration nas(NC2) with predominant contributions of the antiphase stretch and tension of theN7Cbonds interacts with the stretching vibration n(P7N).One of the two resultant vibration appears as a band of strong or medium intensity in the IR spectrum in the region 1140 ± 1210 cm71. For the trivalent phosphorus compounds, this band is shifted towards high fre- quencies in this spectral region. The other resultant vibration gives rise to a strong IR absorption band 1 in the region *935 ± 1008 cm71 (a narrower spectral region *940 ± 980 cm71 has also been reported 4, 5). The corresponding Raman lines are usually weak, as are the lines corresponding to the r(CH3) rocking vibrations, which appear in the IR spectra of molecules containing the (CH3)2NP groups as bands of medium intensity in the frequency range*1050 ± 1080 cm71.Two or even three peaks can be observed in these spectral regions for diamides and triamides (see Tables 19, 20). The bands due to the vibrations with the contributions of the vibrations n(P7N) and nas(NC2) in the frequency range 1140 ± 1210 cm71 or the vibration ns(NC2) in the frequency ranges 935 ± 1008 and 650 ± 700 cm71 are mainly split. The splitting is to a great extent caused by the kinematic interaction between the dimethylamino groups through the shared phosphorus atom.127 It is of interest to compare the spectra of acyclic compound 42 (see Table 19) and its cyclic analogue 43 (see Table 20). A distinctive feature of the vibrational spectra of 2-chloro-1,3- dimethyl-1,3,2-diazaphospholane 43 is that they contain the lines at 1370, 1342 and 830 cm71 corresponding to the methylene groups, as well as the lines at 855 and 597 cm71 caused by the ring vibrations.As in the case of acyclic compound 42, the strongest line in the Raman spectrum of the phospholane 43 is associated with the totally symmetrical vibration with the con- tribution of the P7N bonds. However, for compound 43 the frequency of this line (710 cm71) is shifted as compared with that of tetramethylphosphoramidochloridite 42 (670 cm71), whereas IR (l) a 759 the frequencies of the corresponding antisymmetrical vibrations (692 cm71) remain virtually unchanged. VII. Phosphorothioites A rather large number of studies dedicated to the interpretation of the spectra of phosphorothioites has been reported. Normal coordinate analyses were performed for cyclic and acyclic mole- cules with one, two and three RSP groups, where R are various alkyl, dithiocarbamoyl and dithiophosphoryl radicals.The sim- plest representative of phosphorothioites, methyl phosphorodi- chloridothioite (44), exists in the liquid and gas phases as a mixture of the T and G conformers (see Fig. 1; X=Cl, Y=S, Z=CH3).103, 128 Vibrational spectra of compound 44 have been interpreted 129 assuming that the molecule adopts a single con- formation. Then, it was established 103 that the molecule adopts several conformations, the previous assignments 129 were refined and a normal coordinate analysis was performed for the spectrum of compound 44.These results are listed in Table 21. Table 21. Interpretation of vibrational spectra of methyl phosphorodi- chloridothioite.a Raman (l) Assignment Symmetry IR (l) b No. 520, sh 496, s, br 7 A0 A00 472, s 466, s, br 0 158 A 9 A0 16 10 A00 A0 516, m 496, m 481, sh 474, s 461, m 318, s 262, s 202, vs 185, m 168, w 150, sh 11 A0 ns[PCl2(G)] ns(PCl2) nas[PCl2(G)] nas(PCl2) n(P7S) d(PSC) d[PSC(G)] t(PCl2) d(PCl2) w[PCl2(G)] w(PCl2) a See note to Table 2. b Spectral region below 400 cm71 has not been recorded. The molecule of ethyl phosphorodichloridothioite (45) can adopt five stable conformations 130 (see Fig. 2; X=Cl, Y=S). Vibrational spectra of compound 45 have been interpreted using the results of normal coordinate analysis.130 It has been shown that internal rotation about both the P7S and S7C bonds occurs in the neat liquid and in solutions.However, it was not established whether the Gg and Gg 0 conformations or at least one of them do exist (we only can say that the existence of these conformers does not contradict experimental data). Keeping this in mind, the spectra of ethyl phosphorodichloridothioite 45 can be interpreted in different ways, since some of the frequencies can be attributed to either the Tg conformers or the Tg, Gg and Gg 0 conformers simultaneously. These frequencies are denoted by a character `g'. The bands, which can be attributed to the Gt, Gg and Gg 0 conformers are denoted by a character `G'.A version 130, 131 of the interpretation of the spectra of compound 45 is listed in Table 22. The molecule of propyl phosphorodichloridothioite (46) can adopt 14 spectroscopically discernible conformations, of which three are shown in Fig. 4. It was established that the T, G, gt 0, tt 0, gg 0 and tg 0 conformers of the molecule 46 (for the meaning of notations, see Fig. 4) exist in the liquid and in solutions and a normal coordinate analysis was performed for all possible con- formers.132 Attempts to establish exact composition of such a complex mixture of conformers have failed. The molecule of isopropyl phosphorodichloridothioite (47) can adopt five spectroscopically discernible conformations (Fig. 5). The bands associated with internal rotation about the P7S and S7C bonds of the molecule are observed in the vibra-760 Table 22.Interpretation of vibrational spectra of ethyl phosphorodichlor- idothioite.a Raman (l) Assignment No. Symmetry IR (l) b 22 10 A00 A0 765, 1 655, 18, p 645, 21, p 510, 36, p 500, sh, p 11 12 A0 A0 23 A00 13 14 A0 A0 780, vw 765, w 655, w 645, w 514, vs 500, vs 474, 100, p 476, vs 493, sh, dp 460, sh, dp 340, 11, p 320, 4, p 260, 27, p 248, 32, p 190, 44, dp r[CH2(g)] r(CH2) n(SC), d(SCC), d(PSC) d(SC), d(SCC), d[PSC(g)] ns[Cl2PS(G)] ns(Cl2PS) ns(PCl2), n(P7S) nas[PCl2(Tg,Gt,Gg,Gg)] nas(PCl2) d(SCC), d[PSC(g)] d(SCC), d(PSC) d(PSC), d(SCC) d(PSC), d[SCC(G)] t(PCl2) d(PCl2) w(PCl2), d(SCC) A00 A0 A0 135, 10, p 24 15 16 a See note to Table 3.b Spectral region below 400 cm71 has not been recorded. P P Cl Cl S S Cl Cl P CH3 CH2 CH3 CH2 S CH2 Cl Cl H2C CH2CH2 CH3 Ttg 0 Tgt 0 Gtt 0 Figure 4. Conformers of the molecule of propyl phosphorodichlorido- thioite: Ttg 0 stands for trans,trans,gauche, Tgt 0 for trans,gauche,trans and Gtt 0 for gauche,trans,trans; T and G denote the trans and gauche conformers formed owing to rotation about the P7S bond; t and g denote the trans and gauche conformers formed owing to rotation about the S7C bond and t 0 and g 0 denote the trans and gauche conformers formed owing to rotation about the C7C bond. P P CH3 H3C Cl Cl Cl Cl S S C P H S Cl Cl C C H CH3 Gt HH3CTg H3CH3C Tt H H CH3 H3C C P C P Cl Cl CH3 S Cl Cl S CH3 Gg 0 Gg Figure 5.Conformers of the molecule of isopropyl phosphorodichlor- idothioite: Tt stands for trans,trans, Tg for trans,gauche, Gt for gauche,trans, Gg for gauche,gauche and Gg 0 for gauche,gauche 0. tional spectra of liquid 47.132 Only the Tg conformer is retained upon crystallisation. The number of possible conformers of the molecule of triethyl phosphorothioite (48) is as large as 36. Nevertheless, experimental spectra of neat (liquid) compound 48 and its solutions can be interpreted 130 assuming that its molecule has only two conform- ers. S A Katsyuba CH3 CH2 CH2 H3C CH3 CH2 S P S CH2 S PCH2 S CH3 S H3C S CH2CH3 GtGtGt TtGgG 0g 0 Here, the first two characters denote the conformation of one C2H5SP group of the molecule 48, while the next two pairs of characters denote the conformations of the other two identical groups.It has been reported that only the TtGgG 0g 0 conformer is retained upon crystallisation.130 In the crystalline phase, the equilibrium conformation of the molecule of S,S 0,S 00-tris(dimethoxythiophosphoryl) phosphoro- trithioite (49) has a C3 symmetry (see Ref. 133) CH3 O S CH3 P S CH3 P P S S O O O S O P H3C CH3 S OCH3 Other, non-symmetrical conformers appear in the liquid phase (either in a neat liquid or in solutions).134 O CH3 S P O CH3 S CH3 P S O CH3 O O P P CH3 P H3C P S CH3 O CH3 P S SOS OCH3 S P CH3 O S S SOS OCH3 O CH3 This gives rise to additional peaks in the spectra.Normal coordinate analysis for all possible conformers of the molecule 49 shows 134 that additional bands can be attributed to the non- symmetrical conformers. Some bands were assigned to those conformers which can be formed from all the three conformers shown above by changing the arrangement of the methoxy groups of the (CH3O)2P(=S) fragment from gauche 0, gauche with respect to the P=S bond to the gauche, gauche or gauche, trans with respect to the same bond. The presence of other conformers also does not contradict the results of spectral experiments.134 The equilibrium conformation of the molecule of S,S 0,S 00-tris- (N,N0-dimethylthiocarbamoyl) phosphorotrithioite (50) in the crystalline phase has a C3 symmetry group (see Ref.135). CH3 N H3C C S S S H3C C P CH3 C SS N N S CH3 CH3 According to the vibrational spectroscopy data,136 the mole- cule of compound 50 in solutions retains a single conformation. According to the results of the gas-phase electron diffraction study,137 the molecule of 2-chloro-1,3,2-dithiaphospholane (51) adopts a symmetrical `envelope' conformation with an apical phosphorus atom and an axial P7Cl bond. On the other hand, data obtained in the NMR138 and vibrational spectroscopic 108 studies indicate that the molecule of compound 51 adopts a non- symmetrical conformation. Calculations by the molecular mechanics method predict the a-EC conformation of molecule 51Vibrational spectra, force constants and conformations of molecules of trivalent tricoordinate phosphorus compounds Table 23.Interpretation of vibrational spectra of 2-chloro-1,3,2-dithio- phospholane. Raman (l) Assignment IR (l) a No. 990, w 942, m 830, w 665, w 647, m, s 482, w, br 442, w 420, w, br 352, ms 256, w 11 12 13 14 15 16 17 18 19 20 21 22 23 24 n(C7C) r(CH2) r(CH2), d(CCS) n(C7S), d(CCS), r(CH2) n(C7S), d(CCS) n(P7Cl), ns(PS2) d(PSC), d(CCS) n(P7S), d(CCS) n(PS2) d(PSC), d(SPS), d(SPCl) d(SPS), d(CCS) d(SPCl) ring puckering b ring puckering b 987, 0, 0.73 938, 0, 0.45 825, 0, 0.78 664, 10, 0.43 646, 50, 0.13 480, 100, 0.1 440, 40, 0.76 415, 40, 0.30 362, 50, 0.27 255, 10, 0.67 196, 30, 0.63 175, 40, 0.85 102, 20, 0.78 102, 20, 0.78 a Spectral region below 400 cm71 has not been recorded.b Nonplanar vibration. (see Fig. 3; X=Y=S). Table 23 lists the interpretation of the vibrational spectra of 2-chloro-1,3,2-dithiaphospholane 51 for this conformation.108 Experimental spectral parameters 108 of liquid compound 51 coincide with previously reported data,139 except for the degrees of depolarisation of the Raman lines. Stretching vibrations of the P7S and S7C bonds are most pronounced in the spectra of all the above-mentioned alkyl phosphorothioites. The vibrations n(S7C) can mix with the deformations of the alkylthio group and PSC fragment (Table 24), which is responsible for the variations in the intensities of the corresponding IR absorption bands (from weak to medium) and Raman lines (from medium to strong). The phosphorus- containing fragment of the molecule contributes negligibly to the potential energy of this vibration, therefore its frequency depends on the structure of the alkylthio group.Because of this, the n(S7C) `fingerprint' spectral regions 87 of phosphorothioites nearly coincide with those of different organosulfur compounds. It is more convenient to identify stretching vibrations of the S7C bond using the Raman spectra, since the corresponding lines are rather strong and polarised. If internal rotation of the alkylthio groups about the S7C bond can lead to inequivalent orientations of these groups with respect to the P7S bond, the Raman lines become multiplets. It is also desirable to identify the P7S bonds using the Raman spectroscopy. The stretching vibrations n(P7S) in the molecules of alkyl phosphorodichloridothioites are strongly coupled with symmetrical stretching vibrations of the PCl2 group.One of the two resultant vibrations appears as strong bands in the IR spectrum and polarised Raman lines of medium or strong intensity in the region *490 ± 530 cm71. The second vibration is associated with very strong polarised Raman lines and strong IR absorption bands in the region *460 ± 480 cm71. If the molecules of compounds under study can adopt several confor- mations, all these bands are, as a rule, multiplets; it is noteworthy that polarised Raman lines can have depolarised `shoulders'.The vibrations n(P7S) can also interact with vibrations of other groups at the phosphorus atom (see Table 24). Nevertheless, the presence of both a strong polarised Raman line and an IR absorption band of medium or strong intensity in the region *470 ± 540 cm71 can be considered to be a reliable spectral indicator of the single P7S bond, while multiplet IR and Raman spectra in this spectral region can be regarded as an indicator of the existence of different conformers due to the internal rotation about this bond. This spectral region is characteristic of acyclic molecules. For cyclic phosphorothioites, the frequencies of the stretching vibrations P7S can be substantially lower. Table 24.Vibrations of fragments of the molecules with contributions of the P7S, P7Cl and S7C bonds in the spectra of organophosphorus compounds. Assignment Molecule CH3SPCl2 C2H5SPCl2 n-C3H7SPCl2 iso-C3H7SPCl2 SS P Cl (C2H5S)3P [(CH3O)2P(S)S]3P ns[PCl2(G)] ns[PCl2(T)] nas[PCl2(G)] nas[PCl2(T )] n(P7S) n(S7C), d(SCC), d[PSC(t)] n(S7C), d(SCC), d[PSC(g)] ns[Cl2PS(G)] ns[Cl2PS(T)] ns(PCl2), n(P7S) nas[PCl2(Tg,Gt,Gg)] nas[PCl2(Tt )] n(S7C), d[SCC(tt 0)] n(S7C), d[SCC(tg 0)] n(S7C), d[SCC(gg 0)] ns[Cl2PS(G)] ns[Cl2PS(T )] ns(PCl2), n(P7S) nas[PCl2(Tg,Gt,Gg)] nas[PCl2(Tt )] n[S7C(g)] n[S7C(t )] ns[Cl2PS(G)] ns[Cl2PS(T)] ns(PCl2), n(P7S) nas[PCl2(G)] nas[PCl2(T )] n(S7C), d(SCC), d[PSC(t)] n(S7C), d(SCC), d[PSC(g)] n(P7S), r[PCCl(G)] n(P7S), r[PCCl(T )] n(C7S), d(CCS), d(CCO) n(P7Cl) n(P7S) n(C7S), d(CCS), r(CH2) n(C7S), d(CCS) n(P7Cl), ns(PS2) n(P7S), d(CCS) n(PS2) nas(PS3) nas(PS3) ns(PS3) ns[PS3(gg)], ns(PS3) ns(PSP) n(PVS) nas(PSP) [(CH3)2NC(S)S]3P ns(PS3) nas(PS3) a Melt.b Crystals. Effective coupling of the stretching vibrations of three P7S bonds in the molecules of trialkyl phosphorotrithioites gives rise to three new resultant motions (see Table 24). Symmetrical 761 IR (l) Raman (l) 520, sh 496, s, br 472 516, m 496, m 481, sh 474, s 461, m 655, m, p 466, s, br 655, w 645, w 645, m, p 514, vs 500, vs 476, vs 510, s, p 500, sh, p 474, vs, p 493, sh, dp 460, sh, dp 720, sh 645, w, p 635, w, p 510, s, p 500, sh 475, vs, p 492, sh, dp 455, s, dp 615, w, p 610, sh 527, s, p 505, sh, p 475, vs, p 490, sh, dp 450, sh, dp 665, sh, 0.5 650, w 650, s, 0.3 503, m 486, m 680, w 498, vs, 0.1 482, sh, 0.3 679, s, 0.35 469, vs, br, 0.3 470, vs 414, vs, br, 0.3 415, vs 665, w 664, w, 0.4 646, w, 0.1 647, ms 480, vs, br, 0.1 482, w, br 420, w, br 352, ms 513, s 485, vs 455, s 523, ma 498, sh a 415, s, br, 0.3 362, s, 0.3 515, sh, p? 485, s, dp? 485, vs, p 523, s a 498, vs, p a 485, wa 485, sh, dp a 485, vw b 465, mb 483, s b 464, s b762 stretching vibrations of all the three P7S bonds of the PS3 group appear in the spectral region*430 ± 470 cm71.The correspond- ing Raman lines are very intense and strongly polarised and the IR absorption bands are of medium or strong intensity. Two anti- symmetric vibrations nas(PS3) appear as very strong bands in the IR spectrum and strong Raman lines in the region *460 ± 500 cm71 and as peaks of medium intensity in the region *500 ± 570 cm71. All the above-mentioned vibrations can be `conformationally sensitive', therefore some `extra' peaks can be observed in these regions of the spectra of those trialkyl phos- phorotrithioites which can exist as a mixture of conformers. If the molecules of trialkyl phosphorotrithioites have a C3 symmetry, both antisymmetric vibrations nas(PS3) become degenerate, i.e., have equal frequencies. This is the case for S,S 0,S 00-tris-(N,N-dimethylthiocarbamoyl) phosphorotrithioite 50 (see Table 24), for which a singlet IR absorption band and a singlet Raman line corresponding to the vibration nas(PS3) are observed.Unlike the spectra of trialkyl phosphorotrithioites, the frequency of symmetric vibration ns(PS3) in these spectra is higher than that of antisymmetric vibration nas(PS3) (*485 vs. *465 cm71, respectively). This also holds for compound 49 (see Table 24) that is, the ns(PS3) frequency is as high as nearly 500 cm71, while the nas(PS3) frequency is 485 cm71. According to Thomas,1 the IR absorption bands and Raman lines characteristic of compounds with the PIIISP group are usually observed in nearly the same spectral region (*480 ± 510 cm71).As can be seen from the data listed in Table 24, they should be attributed to the stretching vibrations of this group. VIII. Compounds with Si7P, Ge7P and P7P bonds Silylphosphine (52), (trifluorosilyl)phosphine (53) and germyl- phosphine (54) are rare examples of the trivalent phosphorus compounds with Group IV elements for which normal coordinate analysis has been performed. The spectra of compound 52 and its fully deuterated analogue have been interpreted.140, 141 Analysis of the spectra of both these molecules 141 has led to some changes in the vibrational assignments suggested earlier.140 Vibrational spec- tra of compound 53 and its fully deuterated analogue were interpreted using the results of normal coordinate analysis 142 and the spectra of phosphine 54 and its three D-isotopomers were studied.143 A salient feature of the vibrational spectra of silyl- and germylphosphines is that they contain the peaks which correspond to the stretching vibrations n(Si7P) and n(Ge7P).The vibration n(Si7P) appears in the region *300 ± 530 cm71, while the vibration n(Ge7P) appears in the region *290 ± 410 cm71. Both stretching vibrations can mix with skeletal deformations of the molecules (Table 25). This probably results in substantial variations of the intensities of the corresponding peaks. The vibrations n(Si7P) in the molecules of acyclic organophosphorus compounds appear in a somewhat narrower spectral region *380 ± 530 cm71.The vibrations nas(PH2), ns(PH2), d(PH2), characteristic of all primary phosphines, appear in the spectra of compounds under consideration in nearly the same way as in the spectra of alkylphosphines. For primary phosphines, the n(PH2) frequencies are somewhat higher while the d(PH2) frequencies are somewhat lower than those indicated in Section II. In Table 25, the frequencies of these characteristic vibrations of the phosphino group are also given for the molecule of diphosphane (55). As can be seen, the frequencies of stretching and deformation vibrations of the PH2 groups of the molecule 55 lie far beyond the limits of the spectral regions characteristic of alkyl-, silyl- and germylphosphines and their halogen-substituted derivatives.Only the G conformers have been detected in diphosphane 55 (see Refs 144 ± 146). Vibrational spectra of diphosphane have been studied.144, 147 ± 149 Table 25 lists the interpretation 144 of the Table 25. Spectral indications of characteristic vibrations of molecules XnAPYm (A = Si, Ge, P). Molecule H3SiPH2 F3SiPH2 H3GePH2 H2PPH2 I2PPI2 (CF3)2PP(CF3)2 a Characters R and P denote the branches of the contours of the IR absorption bands. b The meaning of notationH0 is clearly seen from the Newman projections of possible conformations of diphosphane and its derivatives: H0 H P H0 HG H0 T c Symmetry species of vibrations. spectra, which is based on the results of normal coordinate analysis for molecules 55 and P2H2D2.Vibrational assignments made in the framework of ab initio calculations 146 of the force field of the molecule 55 differ only slightly from those listed above. Initially, the spectra of solutions of tetraiododiphosphane (56) were interpreted assuming the gauche conformation of the mole- cule.150 However, it was shown later 151 that the trans conforma- tion exists in fact both in the crystal and in solutions. Taking into account the results of normal coordinate analysis of the molecule 56 (see Ref. 152), some vibrational assignments in the low- frequency region were changed compared to those reported ear- lier.150 Vibrational spectra of tetrakis(trifluoromethyl)diphosphane (57) have been interpreted 153 assuming that the molecule 57 has only the trans-conformation with a C2h symmetry.Under this symmetry group, the vibrations of the Ag and Bg types should be Raman active while the vibrations of the Au and Bu types should be IR active (the so-called principle of alternative exclusion). However, the results of spectroscopic studies 153 indicate that this principle is not obeyed. The fact that the IR absorption band corresponding to the vibration n(P7P) (see Table 25) is weak is not always a consequence of alternative exclusion. It is known that these bands can also be of weak intensity if the molecules with the P7P bond adopt non-symmetrical conformations. Therefore, the conclusion 153 that the molecule 57 has a C2h symmetry, which is based on the weak intensity of the n(P7P) band, casts some doubts.Correspondingly, the results of theoretical interpretation of the spectra 153 can appear to be not sufficiently rigorous. S A Katsyuba IR Raman Assignment 2310, s (l) 507, vs (l) 2304, dp (l) 2293, p (l) nas(PH2) ns(PH2) d(PH2) n(Si7P) nas(PH2) ns(PH2) d(PH2) n(Si7P), ds(SiF3) nas(PH2) ns(PH2) d(PH2) 363, p n(Ge7P), w(PH2) 2312 (g) 2312 (g) 1072 (g) 454 (g) 2318, s (g) 2312, s (g) 1062, vw (l) 1069, s (g) 512, vs (g) 2320, s (g) 2310, s (g) 1080, R mw (g) a 1065, P vw (g) a 371, R? 352, P? 2299, vs (cr) 2295, vs (cr) 2312, vs (g) 2281, vs (cr) 1081 (g) nas(P7H) ns(P7H) nas(P7H0) b 2320, vs (g) ns(P7H0) b 2268, vs (cr) d(PH2), d(PPH) (B) c 1054, s (cr) d(PH2), d(PPH) (A) c 1037, s (cr) n(P7P) 438, s (g) n(P7P), n(P7I) 319, vw (cr) n(P7I), n(P7P) 307, vs (cr) n(P7P), ns(PC2) 486, m, p (l) *497, vw (g) H P H H0 .Vibrational spectra, force constants and conformations of molecules of trivalent tricoordinate phosphorus compounds It is convenient to identify the P7P bond in diphosphanes using Raman spectra where the line corresponding to the vibra- tion n(P7P) is observed in the region *300 ± 530 cm71, has a medium or strong intensity and is polarised (for gases, liquids or solutions). IX.Rotational isomerism of trivalent phosphorus compounds and its manifestation in vibrational spectra 1. General If interconversions between stable conformations of a molecule can occur owing to hindered internal rotation and if the con- formers are not mirror images, their vibrational frequencies can differ from one another.Because of this, the number of bands and lines in the vibrational spectra of organophosphorus compounds, whose molecules can adopt several conformations, is larger than it should be expected for a single molecular conformation. As a rule, the `sensitivity' to rotations of a particular molecular fragment about an X7Y bond with respect to another molecular fragment is characteristic of those vibrations to which the bond in question contributes. Usually, the frequencies of other vibrations depend only slightly on the molecular conformation, so that the corre- sponding IR absorption bands or Raman lines in the spectra of different conformers coincide.Therefore, an equilibrium between N conformers rarely, if ever, leads to the N-fold increase in the number of spectral peaks; rather, it causes multiplet splittings of particular peaks, e.g., the n(X7Y) or d(XYZ) peaks. However, the frequencies of vibrations with predominant contribution by a particular molecular fragment can also be dependent on the conformation of another molecular fragment. This dependence is due to changes in either or both the spatial mass distribution and the force constants of the molecule on interconversions between different conformations. Sometimes, the effect of conformations on the vibrational frequencies is so weak that it can be detected only in the low-temperature experiments.In this case, the spectral contours become much narrower and the peaks of different conformers which have been unresolved at room temperature, can be observed as multiplets or as separated bands. In other cases, shifts in the vibrational frequencies due to conformational effects are as large as tens of cm71. Usually, crystallisation leads to simplification of the conformational composition of most compounds and disappearance of particular spectral bands. The conformational composition depends also on the temperature and polarity of the medium. Therefore, the conditions of spectroscopic experiments can strongly affect the spectra of organophosphorus compounds, whose molecules can adopt several conformations. This should be taken into account when interpreting the spectra. For instance, the intensities of the bands of conformers of greater polarity usually increase in polar media and decrease on going to non-polar solutions. This correlation can be used for the assignment of spectral peaks to the conformers of different polar- ities.However, the assignments based on both the experimental results and the results of normal coordinate analysis for possible conformers are the most reliable. It is these vibrational assign- ments that are generalised in this review and serve as the basis for simple empirical rules which facilitate the interpretation of the spectra of organophosphorus compounds, whose molecules can adopt several conformations. 2.Internal rotation of Alk7P fragments Alkylphosphines and their derivatives with the general formula AlkPX2 (Alk=C2H5, n-C3H7, iso-C3H7; X=H, F, Cl, CH3), which are liquids or gases at room temperature, are characterised by the existence of an equilibrium between theGand T conformers formed owing to internal rotation about the P7C bond (see Figs 1, 2). The T conformers are more energetically favourable than the G conformers,15 ± 21, 23, 56 ± 61 except for the molecule of ethyl(dimethyl)phosphine;46, 154, 155 the energy `gain' is nearly 763 1.6 kJ mol71 in the liquid phase and 0.8 kJ mol71 in the gas phase. It should be noted that the enthalpy differences (DH) between the T and G conformers are small: they vary from *0.4 to *4.2 kJ mol71 and decrease on going from the liquid to the gas phase.Therefore, liquid or gaseous alkylphosphines exist at room temperature as mixtures of the G and T conformers in comparable proportions. Many of the vibrations of the T and G conformers have different frequencies (see Tables 2 ± 4). The simplest rule is valid for the conformational `sensitivity' of the P7C stretching vibra- tions (Alk=C2H5, iso-C3H7): the frequency of the vibrations n[P7C(T )] is always lower than that of the vibrations n[P7C(G)]. The n[P7C(G)]7n[P7C(T )] difference for compounds C2H5PX2 (X=CH3, Cl, F) varies between *30 and 60 cm71. For the molecule 6, this value decreases to*20 cm71 and is only 7 cm71 for ethylphosphine 4, so that the doublet band corresponding to the vibrations n(P7C) can only be observed in the spectra of a solution of ethylphosphine 4 in liquid xenon.20 The appearance of gauche and trans conformers in n-C3H7PX2 is due to not only internal rotation about the P7C bond (the G and T conformers, respectively, see Fig.2;Y=CH2), but also the rotation with respect to the C7C bond (the g and t conformers, see Fig. 2). The t conformers are *1.3 to 2.9 kJ mol71 more energetically stable than the g conformers, the DH values decrease on going from the liquid to the gas phase or to solutions in liquid xenon. It is noteworthy that the enthalpy difference between the gauche and trans conformers of butane measured in the gas phase [*4.2 kJ mol71 (see Ref.156)] is much larger than the DH values for the corresponding n-propyl- phosphine conformers. Thus, substitution of a methyl group in the butane molecule by a bulky PX2 group causes a decrease rather than an increase in the probability of the appearance of the trans conformers owing to internal rotation about the C7C bond. The frequencies of the vibrations n(P7C) of propylphos- phines are `sensitive' to conformational transitions of both the G>T and g>t types. For G?T transition, the n(P7C) fre- quency decreases, as is the case of C2H5PX2, whereas the g?t transition leads to an increase in this frequency. As a result, the highest-frequency IR absorption band (or Raman line) in the spectra of propylphosphines, corresponding to the vibrations n(P7C), is characteristic of the Gt conformers, while the frequen- cies of the peaks of the Tt conformers are *20 ± 50 cm71 lower and the peaks of the Gg and Tg conformers are even more shifted towards the low-frequency region (see Table 5).All these correla- tions between the changes in the frequency of the vibrations n(P7C) and the conformations of the ethylphosphino and pro- pylphosphino groups are caused mainly by kinematic factors. Orientations of alkyl groups (gauche and trans) with respect to the lone electron pair of the phosphorus atom are also character- istic of the molecules of the Alk2PX or Alk3P types. In the liquid or gas phase, the molecules of secondary and tertiary phosphines adopt usually several conformations, which is responsible for the appearance of multiplet IR absorption bands and Raman lines corresponding to the symmetrical and antisymmetrical stretching vibrations of the P7C bonds and to some other vibrations (see Table 5).However, all attempts to find general correlations between the molecular conformations and frequencies of any vibrations have failed. 3. Internal rotation of AlkOP fragments The molecules of compounds with the general formula ROPX2 can adopt either trans or gauche conformations, which is due to internal rotation about the P7O bond (see Fig. 1; X=CH3, F, Cl; Y=O; Z=R). The gauche conformers are more favourable for X=CH3, though a detectable amount of trans conformers is also present in the liquid phase. On the contrary, the trans conformers of alkyl phosphorodichloridites or alkyl phosphoro- difluoridites are much more stable.According to the data of ab initio quantum-chemical calculations,91, 92 the energy difference between the G and T conformers for X=F is larger than for764 X=Cl. In the former case, it is so large that the G conformer cannot be detected experimentally.91, 95 ± 97 No spectral indications of the existence of several conformers was observed for alkyl phosphorodichloridites ROPCl2 (R=CH3, C2H5) in the liquid or gas phase at room temper- ature.98 ± 100, 103, 104 Based on the results of a gas-phase electron diffraction study 101 and measurements of the dipole moments,99 it was assumed that the molecule 3 adopts a nearly gauche con- formation. However, a more detailed gas-phase electron diffrac- tion study showed that this molecule adopts the trans conformation in the gas phase.102 Later, additional bands which cannot be attributed to the trans conformer were found in the spectra of glassy films of compound 3 and its solutions.103 The high-frequency component of the doublet at 748/754 cm71 cor- responding to the vibrations ns(POC) was attributed to the gauche conformer.The enthalpy difference between the gauche and trans conformers determined in CS2 from the temperature dependence of the relative intensities of the bands at 748 and 754 cm71 was found to be *0.8 kJ mol71 in favour of the latter. However, studies of methyl phosphorodichloridite 3 dissolved in liquid xenon revealed 92 disappearance of some of the bands attributed to the gauche conformer.103 Probably, these bands belong to molecular associates.However, the doublet band at 773/754 cm71 corresponding to the stretching vibrations ns(POC) retained in the solution in liquid xenon. It was suggested that this doublet is due to the Fermi resonance, since the relative intensities of the bands at 773 and 754 cm71 were found to be temperature-independent.92 It should be emphasised that the enthalpy differences between the G and T conformers of molecules AlkOPX2 (X=CH3, F, Cl) have not been measured as yet. The same is also true for the enthalpy differences between the g and t conformers, which are formed due to internal rotation about the O7C bond in this type of molecules (see Fig.2, Y=O). These differences seem to be so large that only t conformers are observed in the liquid phase at room temperature.104 This is also typical of the dialkyl ether molecules for which the trans conformers formed by internal rotation about the O7C bond are much more energetically favourable (DH'4.6 ± 6.3 kJ mol71).157 ± 160 However, their gauche conformers are also present in detectable amounts both in the liquid and gas phase at room temperature. Relative stabilities of the gauche and trans conformers change drastically upon introduction of the SCN group into the alkyl substituent of the ROPCl2 molecules. The enthalpies of the G and g conformers of the molecule 35 become much lower than those of the T and t conformers, respectively.105 The fact that the `twist' conformers of the molecule 35 are more energetically favourable was attributed 105 to the effects of hypervalent intramolecular 1 ± 5-attraction between the phosphorus and sulfur atoms.It should be noted that a doublet at 730 to*715 cm71 is observed in the spectra of liquid b-thiocyanatoethyl phosphorodichloridite 35 in the region characteristic of vibrations with predominant contribution of the ns(POC) vibration and that the low-frequency line of the doublet disappears upon crystallisation. Doublet IR absorption bands and Raman lines, assigned to the P7O stretching vibrations, are also observed in the spectra of liquid 2-chloro-1,3,2-oxathiaphospholane 37 (*760 to *750 cm71), whose molecule can adopt several conformations owing to internal rotation about the P7O bond (see Fig.3; conformers a-EC and e-EC, X=O, Y=S).108 The a-EC con- former, which is an analogue of the T conformers of acyclic alkoxy derivatives of the trivalent phosphorus compounds, was found to be *2 kJ mol71 more stable than the e-EC conformer (an analogue of the G conformers). Thus, despite some uncertainty in the data on the spectral indications of interconversions of the molecules containing one Alk7O7P group, one can say that they can be detected using the readily identifiable band corresponding to the POC stretching vibrations. However, the corresponding effects are weak and often remain unnoticed in the spectra of liquids measured at room S A Katsyuba temperature.These effects can be observed if the spectral reso- lution is improved compared to that of the routine experiments. This can be attained in the low-temperature experiments. The kinematic effect of the G ?T transition results in a decrease in the ns(POC) frequency. However, for 2-chloro-1,3,2-oxathiaphos- pholane 37 an increase rather than a decrease in this frequency was observed, which is probably due to changes in the force field of the molecule in the course of interconversion of its conformers. 4. Internal rotation of AlkSP fragments Alkyl phosphorodichloridothioites AlkSPCl2 exist in the liquid and gas phases as mixtures of T and G conformers 103, 128, 130 ± 132 which are formed owing to internal rotation about the P7S bond (see Fig.1; X=Cl, Y=S, Z=Alk). In the liquid phase, the T conformers are more energetically favourable 130 ± 132 (e.g., DH*3.3 kJ mol71 for Alk=C2H5). Nevertheless, the amount of the G conformers at room temperature is comparable with that of the T conformers. For the molecule 28, it is possible that great steric hindrances to the formation of the trans conformers (con- former A, see Section IV) lead to a decrease in their concentration in the equilibrium mixture. This can be assessed from the intensity ratio of the Raman lines corresponding to the vibrations n(P7S) of the gauche and trans conformers (see Table 25). Rotation about the S7C bond can also be responsible for the formation of the g and t conformers (see Fig.2; Y=S; Fig. 4, 5), the trans arrangement of the P7S and C7C bonds being more favourable for liquid alkyl phosphorodichloridothioites 130 ± 132 (e.g., DHg ± t'1.6 kJ mol71 for Alk=C2H5). For Alk= n-C3H7, the g 0 and t 0 conformers were also detected in the liquid phase 132 which appear due to internal rotation about the C7C bond (see Fig. 4). The molecules of alkyl phosphorodichloridothioites can adopt several conformations. A distinctive feature of their spectra is that the number of the IR absorption bands and Raman lines is much larger than one could expect. In particular, a complicated spectrum is observed in the region characteristic of the `co- operative' stretching vibrations of the Cl2PS fragment at *450 ± 530 cm71. The totally symmetric stretching vibration ns(Cl2PS) is `sensitive' only to rotation about the P7S bond and the frequencies of this vibration for the T conformers are always lower than those of the G conformers (*495 ± 505 vs.*510 ± 530 cm71, respectively). A similar type of conforma- tional `sensitivity' was also observed for the frequencies of the vibrations n(P7S) of the molecule 28 (see Table 25).Adecrease in the frequencies of vibrations with contribution of the P7S bond on going from the G to T conformers is mainly due to changes in the kinematic interactions in the molecule upon this transition. The frequencies of the antisymmetric vibrations nas(PCl2) of alkyl phosphorodichloridothioites depend on the conformations of the Cl2PSC and PSCC fragments.The conformation in which the lone electron pair of the phosphorus atom lies in the plane of the PSCC fragment (the Tt conformation, see Figs 2 and 4; or the Tg conformation, see Fig. 5) is characterised by a frequency in the range between*450 and 460 cm71. For all other conformations, the characteristic frequency nas(PCl2) is observed at &490 to 495 cm71 (see Table 25). The corresponding bands in the IR spectrum are masked by a very strong band corresponding to the antiphase stretching vibration of the P7S bond and PCl2 frag- ment (*460 ± 480 cm71), which is conformationally `insensitive'. Because of this, the vibrations nas(PCl2) of different conformers can be identified only in the Raman spectra in which they appear as depolarised shoulders on either side of a very strong polarised line at *460 ± 480 cm71 corresponding to the above-mentioned antiphase stretching vibration of the P7S bond and PCl2 group.It is also convenient to use the Raman spectra for the attribution of the lines corresponding to the stretching vibrations S7C, the frequency of which depends on not only the structure of the alkyl radical at the sulfur atom, but also the conformation of the PSCC fragment. The conformer with trans arrangement of the P7S and C7C bonds (t, see Figs. 2 and 4; g, see Fig. 5) isVibrational spectra, force constants and conformations of molecules of trivalent tricoordinate phosphorus compounds characterised by a higher frequency n(S7C) than the gauche conformer (g, see Figs. 2 and 4; t, see Fig.5). Therefore, in the case of a mixture of the g and t conformers, the vibrations n(S7C) appear as doublets. For S-propyl phosphorodichloridothioite, a triplet rather than a doublet corresponding to the vibrations n(S7C) is observed (see Table 25). This may be due to the fact that the frequency of this vibration depends on internal rotation about not only the S7C bond, but also the C7C bond. Never- theless, the high-frequency components of the triplet correspond to the t conformation of the PSCC fragment. The molecules of S,S0,S 00-trialkyl phosphorotrithioites (AlkS)3P are also characterised by the G and T orientations of the S7C bonds with respect to the LEP of the phosphorus atom and either a g or t arrangement of the C7C bonds with respect to the P7S bonds.Usually, S,S0,S 00-trialkyl phosphorotrithioites exist in the liquid or gas phase as mixtures of the TGG and/or TGG0 conformers with the GGG conformers.130, 131, 161 The latter are less energetically favourable, but are present in considerable amounts at room temperature. Since the molecules of S,S0,S 00- trialkyl phosphorotrithioites can adopt several conformations, some of the bands in their vibrational spectra become split into multiplets. However, only correlations between the conforma- tions of the C2H5SP fragments and the frequencies of the CH2 rocking vibrations were found.130, 131 The g arrangement of the C7C bond with respect to the P7S bond corresponds to the r(CH2) frequency &780 cm71, while the t arrangement corre- sponds to the r(CH2) frequency *760 cm71.The same is also true for the molecule 45 (see Table 22). X. Force constants of molecules containing trivalent phosphorus atoms Among different classes of organophosphorus compounds, the force constants of alkylphosphines and their halogen-substituted derivatives have been studied in most detail. The potential fields of all these molecules share the common property that the effect of the phosphorus atom is confined to the distance to the a-C atom. For instance, the force constants of the CH3 group of ethyl radical at the phosphorus atom remain virtually the same as in the ethane molecule. Because of this, only frequencies of the vibrations with predominant contribution of the P7C bonds, and those of the methyl and methylene groups bound to the phosphorus atom are specific to alkylphosphines and, hence, can be regarded as having an analytical value.On the other hand, variation of the alkyl radical in the AlkPX2 molecules affects not only the force constants of the AlkP group, but also the parameters of the potential field of the PX2 fragment. For instance, the force constants of the P7C and P7X bonds decrease simultaneously as the chain length of the alkyl radical increases. Force constant C2H5PCl2 CH3PCl2 C2H5PF2 CH3PF2 2.65 2.94 KPC 2.54 4.15 2.76 4.34 KPF 1.94 2.08 KPClFor X=H, this dependence can be expressed quantitatively in the form of linear relationships (recall that the force constants are given in 102 N m71) 22 KPH=0.126 s*+3.063, KPC=0.235 s*+2.664, where s* is the Taft constant of the alkyl group.162 Introduction of halogens into the AlkP fragment leads to a pronounced decrease in the force constant KPC and simultaneous increase in KPX.For instance, the KPC andKPCl values respectively change from 2.94 and 2.08 for CH3PCl2 57 to 2.34 and 2.20 for ClCH2PCl2.64 For the same alkyl radical, the force constant KPC depends on the type of substituents X at the phosphorus atom; however, its 765 magnitude changes only slightly and does not obey simple rules. Thus for the CH3PX2 molecules the KPC value increases from 2.66 to 2.94 on going from X=H,13, 14 to CH3,13, 14, 43 F 51 and Cl.57 The force constants of phosphorus-containing molecules can also be conformation dependent.For instance, the gauche?trans transition in the molecule 6 causes a decrease in the force constant KPH from 3.05 to 3.01.22 Despite the small difference between these values, it was found to be reliable, since the P7H stretching vibrations are known to be characteristic.22 Changes in the conformation of the polar ClCH2P group has a more pronounced effect on the potential field of the molecule 15: KPF(gauche)=4.43; KPF 0(gauche)=4.51; KPF(trans)=4.08; KPC(gauche)=2.81; KPC(trans)=2.84; KCCl(gauche)=3.09; KCCl(trans)=2.97. Thus, analysis of the force constants of the molecules of alkylphosphines and their halogen-substituted compounds sug- gests that the potential field of the phosphorus-containing frag- ment of the molecule strongly depends on the type and spatial orientation of each substituent at the phosphorus atom.On the other hand, the effect of the phosphorus-containing fragment on the force constants of adjacent groups is confined to the nearest environment of the phosphorus atom. For phosphines with unsaturated groups at the phosphorus atom, the effect of the latter can likely extend to distances longer than the distance to the a-C atom. The force constants of the fragment =C(CH3)2 of molecules Cl2PCH=C(CH3)2 (see Ref. 75) and CH2=C(CH3)2 (see Ref. 160) differ appreciably, thus providing an explanation for the specific values of the characteristic vibrational frequencies of the alkenyl groups at the phosphorus atom.The potential fields of all these molecules share the common property that the force constants of the multiple carbon ± carbon bonds at the trivalent phosphorus atom are smaller than those for the corresponding hydrocarbon molecules. The force constant KC=C for the molecules of alkenyl(dichloro)- phosphines is much smaller than those for ethylene and various alkenes and halogen-substituted alkenes [7.9 (Ref. 75) vs. 8.3 ± 10.0 (Ref. 163), respectively]. On the other hand, the con- stants KPC'3.1 ± 3.5 (see Ref. 75) and KPCl'2.3 (see Ref. 75) are larger than the corresponding force constants of alkyl(dichloro)- phosphines [e.g., KPC=2.94 and KPCl=2.08 for the molecule 11 (see Ref. 57)]. A similar situation is also observed for the molecule 29: the value of the constant KC:C=13.85 (see Ref.78) `drops out' of the range between 14.22 and 15.62 (see Refs 164 ± 167), which is characteristic of acetylene and its alkyl- and halogen-substituted derivatives; while the value of the constant KPC&3.3 (see Ref. 78) far exceeds the corresponding value [2.9 (see Ref. 43)] for the molecule 8. On the other hand, numerical values of the force constants KPC and KC:N for molecules 30 (see Ref. 79) and 31 (see Refs 84, 85) lie within the limits of conventional ranges, which are characteristic of trialkylphosphines and cyanogen halides.167 Probably, this indicates that the phosphorus atom affects the electronic systems of adjacent C:N and C:C groups in a different manner.For the AlkOP groups, this effect is most likely confined to the distance to the a-C atom of the ester group. In any case, the force constants of the ethyl fragments of molecules C2H5OPCl2 (see Ref. 104) and C2H5OC2H5 (see Ref. 168) are nearly equal. On the other hand, the force constant KPO increases by about one unit on going from the ethyl to the methyl substituent.103, 104 Usually, variation of the nearest environment of the trivalent phosphorus atom has little effect on the force field of the AlkOP groups. For instance, KPO=5.97 and KOC=4.91 for CH3OPCl2,92 while for CH3OPF2 these values are KPO=5.99 and KOC=4.51.91 For the molecules of dialkylphosphoramides, the dependence of the force constant KPN on the nature of substituents at the phosphorus atom is much more pronounced; for instance, the KPN value increases from 4.5 to 5.1 on going from (CH3)2NPCl2 to (CH3)2NPF2 .125766 Analysis of the force fields of the molecules of S-alkyl phosphorothioites shows that the phosphorus atom has no effect not only on the alkyl fragment of the molecule, but also on the force constant of the S7C bond.This seems to be reason why the spectral characteristics of the AlkS groups in the molecules of organophosphorus compounds coincide with the spectral indica- tions of these groups in organosulfur compounds. Alkyl phos- phorothioites are also characterised by the absence of pronounced effect of the alkyl radical on the force field of the phosphorus- containing fragment of the molecule.All acyclic compounds studied belonging to this class are characterised by the same force constant KPS=2.0.130 ± 132 It should be noted that even passage to strained five-membered rings with endocyclic P7S bond 108 or replacement of the alkyl radical in the AlkSP group by the dimethoxythiophosphoryl fragment 133, 134 decrease the KPS value by only*10%. As for the force fields of the molecules with the Si7P, Ge7P and P7P bonds, the corresponding data are too scarce to be used as a basis for any generalisation. Closing the discussion of the force constants of the molecules of organophosphorus compounds, it should be emphasised that they can be used to predict the molecular spectra of related compounds for which studies by conventional methods of vibra- tional spectroscopy give no unambiguous solutions.169 Addition- ally, rapid progress of computer industry opened the way to routine quantum-chemical calculations of the force constants.The first systematic studies of the so-called scaled `quantum- chemical' potential fields of phosphines gave encouraging results.170 Taken together, this allows one to hope that com- puter-assisted predictions of vibrational spectra of various orga- nophosphorus compounds will become a routine procedure already in the near future. XI. Conclusion Trivalent tricoordinate phosphorus compounds contain no struc- tural elements with such distinct spectral indications as, e.g., the phosphoryl group. However, analysis of the published data shows that the efficiency of vibrational spectroscopy in studying the structure of trivalent phosphorus compounds can be as high as in the case of pentavalent phosphorus compounds.This requires that the following salient features of the vibrational spectra of organo- phosphorus compounds be taken into account. The effect of the phosphorus-containing fragment of the molecule is confined to the force constants of adjacent molecular fragments. This leads to a nearly complete additivity in that part of the spectral region which is due to the vibrations of more distant groups. For instance, the spectra of molecule [(CH3O)2P(S)S]3P nearly coincide with those of the corresponding O,O0-dimethyl phosphoroditioate (CH3O)2P(S)SH (see Ref.171) in the whole frequency range, except for the region corresponding to the stretching (485 ± 585 cm71) and bending vibrations (108 ± 150 and 350 ± 405 cm71) of the PSP fragments. Therefore, the spec- trum of a phosphorus-containing molecule can be represented as the sum of the spectra of the fragments constituting the molecule minus the region corresponding to the vibrations of the nearest environment of the phosphorus atom. Therefore, the use of the data listed in Tables 1 ± 25 can make the interpretation of vibra- tional spectra of trivalent phosphorus compounds much easier. For instance, spectral parameters of the C2H5OP fragment of an arbitrary molecule will coincide with the corresponding parame- ters of the O-ethyl fragment of ethyl phosphorodichloridite 34.Vibrations of the POC fragment can appear to be an exception to this rule, since they can interact with vibrations of some other substituents through the shared phosphorus atom. This example shows that consideration of the interactions between individual oscillators is of great importance to the analysis of the `fingerprint' spectral regions. The data on the characteristic vibrations of different phosphorus-containing groups collected in this review will also help in predicting these S A Katsyuba effects correctly. Stretching vibrations of the C7C bond in the C2H5P fragment of dichloro(ethyl)phosphine may serve as an example. This bond is connected to the PCl2 group through the P7C valence bond. Nevertheless, no interaction between the vibration n(C7C) and the vibrations of the PCl2 group occurs (see Table 7).The reason is that the vibrational frequencies of the PCl2 fragment are much lower than the frequency of the vibrations n(C7C). A similar situation will be observed for any other molecules with heavy substituents at the phosphorus atom. There- fore, the frequencies of the vibrations n(C7C) in the molecules 9, 12 and 13 are equal. On the other hand, this frequency can vary on introduction of the C2H5P fragment into a molecule with light substituents at the phosphorus atom, since the character of the interaction between vibrations in the new molecule may be differ- ent. For instance, characteristic frequencies of the bending vibra- tions of the PH2 group (1073 ± 1097 cm71) are close to the frequency of the vibrations n(C7C).Therefore, one can expect that in the molecule of ethylphosphine 4 these individual oscil- lators will interact, thus changing the frequency of the vibration n(C7C) as compared to those of the same vibrations of molecules 9, 12 and 13. The data listed in Table 2 confirm this assumption. Vibrations with contributions of the nearest environment of the phosphorus atom can change on going from one compound to another since the force constants of the phosphorus-containing molecular fragment depend on the types and conformations of all substituents at the phosphorus atom. Turning our attention to the above-mentioned example with the C2H5P fragment, one can say that the frequency of the stretching vibrations n(P7C) will change from 636 cm71 for ethylphosphine 4 to 631/663 cm71 for the molecule 12 or to 655/682 cm71 on going to ethyldifluorophos- phine 13.Reliability of spectral ± structural correlations for the `finger- print' region increases substantially, if the Raman spectra are used in combination with the IR spectra. For instance, several IR absorption bands in the region 550 ± 710 cm71 can indicate two or three P7C bonds in the molecule under study. However, if the molecules of compounds under study contain one P7C bond but can adopt several conformations, more than one IR absorption band will be observed in the above-mentioned spectral region. It is easy to discriminate between these two variants using the Raman spectrum.In the first case, an intense, strongly polarised Raman line will correspond to only one of the bands in the IR spectrum [see, e.g., (CH3)2PCl or (CH3)3P in Table 5], while the number of polarised lines in the Raman spectrum of the second compound will be equal to the number of peaks in the IR spectrum (e.g., n-C3H7PCl2 or n-C3H7PH2, see Table 5). In conclusion, the author expresses his gratitude to Prof. R R Shagidullin, Prof. V A Al'fonsov, Prof. B V Lokshin and Prof. B J van der Veken for valuable comments which were taken into account in the course of preparation of the manuscript. References 1. L C Thomas Interpretation of the Infrared Spectra of Organo- phosphorus Compounds (London: Heyden, 1974) 2.L C Thomas The Identification of the Functional Groups in Organophosphorus Compounds (London: Academic Press, 1974) 3. R R Shagidullin, F S Mukhametov, R B Nigmatullina, V S Vinogradova, A V Chernova Atlas IK Spektrov Fosfor- organicheskikh Soedinenii (Atlas of the Infrared Spectra of Organo- phosphorus Compounds) (Moscow: Nauka, 1977) 4. R R Shagidullin, A V Chernova, V S Vinogradova, F S Mukhametov Atlas IK Spektrov Fosfororganicheskikh Soedinenii (Interpretirovannye Spektrogrammy) [Atlas of the Infrared Spectra of Organophosphorus Compounds (Interpreted Spectro- grams)] (Moscow: Nauka, 1984) 5. R R Shagidullin, A V Chernova, V S Vinogradova, F S Mukhametov Atlas of IR Spectra of Organophosphorus Compounds (Interpreted Spectrograms) (Dordrecht: Kluwer Academic, 1990)Vibrational spectra, force constants and conformations of molecules of trivalent tricoordinate phosphorus compounds 6.M V Vol'kenshtein, L A Gribov, M A El'yashevich, B I Stepanov Kolebaniya Molekul (Molecular Vibrations) (Moscow: Nauka, 1972) 7. E B Wilson, J C Decius, P C Cross Molecular Vibrations (New York: McGraw-Hill, 1955) 8. N B Colthup, L H D Daly, S E Wiberley Introduction to Infrared and Raman Spectroscopy (New York: Academic Press, 1964) 9. R C Lord, F Miller Appl. Spectrosc. 10 115 (1956) 10. P Pulay J. Mol. Struct. 347 293 (1995) 11. J A Lannon, E R Nixon Spectrochim. Acta, Part A 23 2713 (1967) 12. S A Katsyuba, Candidate Thesis in Physicomathematical Sciences, Kazan State University, Kasan, 1980 13.S A Katsyuba, I S Pominov, B P Khalepp Zh. Prikl. Spektrosk. 36 783 (1982) 14. J R Nielsen, J D Walker Spectrochim. Acta 21 1163 (1965) 15. J R Durig, A W Cox Jr J. Chem. Phys. 63 2303 (1975) 16. J R Durig, A W Cox Jr J. Chem. Phys. 64 1930 (1976) 17. P Groner, R D Johnson, J R Durig J. Chem. Phys. 88 3456 (1988) 18. J R Durig, M S Lee, R J Harlan, T S Little J. Mol. Struct. 200 41 (1989) 19. J R Durig, C M Whang, R J Harlan Vib. Spectrosc. 1 19 (1990) 20. J R Durig, J B Robb, T K Gounev J. Mol. Struct. 405 45 (1997) 21. J R Durig, C G James, A E Stanley, T J Hizer, S Cradock Spectrochim. Acta, Part A 44 911 (1988) 22. S A Katsyuba, V N Shegeda, R R Shagidullin Zh. Prikl. Spektrosk. 55 547 (1991) 23. J R Durig, T K Gounev,M S Lee, T S Little J.Mol. Struct. 327 23 (1994) 24. J R Durig, A W Cox Jr J. Phys. Chem. 80 2493 (1976) 25. J R Durig, Y S Li J. Mol. Spectrosc. 70 27 (1978) 26. J R Durig, S D Hudson,M R Jalilian, Y S Li J. Chem. Phys. 74 772 (1981) 27. J R Durig, J E Saunders J. Raman. Spectrosc. 4 121 (1975) 28. A J F Clark, J E Drake Spectrochim. Acta, Part A 34 307 (1978) 29. D C McKean, G P McQuillan J. Mol. Struct. 63 173 (1980) 30. S Chan, H Goldwhite, H Keyzer, R Tang Spectrochim. Acta, Part A 26 249 (1970) 31. R W Mitchell, L J Kuzma, R J Prikle, J A Merritt Spectrochim. Acta, Part A 25 819 (1969) 32. J Bragin, L W Dennis J. Mol. Struct. 18 75 (1973) 33. E J Rosenbaum,DJ Rubin,CRSandberg J. Chem. Phys. 8 366 (1940) 34. R K Shelline J.Chem. Phys. 18 602 (1950) 35. N Siebert Z. Anorg. Allg. Chem. 273 161 (1953) 36. D A Long, R B Gravenor Spectrochim. Acta 19 961 (1963) 37. G Bouquet, M Bigorgne Spectrochim. Acta, Part A 23 1231 (1967) 38. P J D Park, P J Hendra Spectrochim. Acta, Part A 24 2081 (1968) 39. H Rojhantalab, J W Nibler, C J Wilkins Spectrochim. Acta, Part A 32 519 (1976) 40. B Beagley, A R Medwid J. Mol. Struct. 38 229 (1977) 41. DCMcKean,GP McQuillan Spectrochim. Acta, PartA39 293 (1983) 42. W Schneider,W Thiel, A Komornicki J. Phys. Chem. 92 5611 (1988) 43. D C McKean, G P McQuillan,W F Murphy, F Zerbetto J. Phys. Chem. 94 4820 (1990) 44. K K Chatterjee, J R Durig, S Bell J. Mol. Struct. 265 25 (1992) 45. S A Katsyuba, V N Shegeda, R R Shagidullin Zh.Prikl. Spektrosk. 57 455 (1992) 46. C Crocker, P L Goggin J. Chem. Soc., Dalton Trans. 388 (1978) 47. J H S Green Spectrochim. Acta, Part A 24 137 (1968) 48. H Schindlbauer, E Steininger Monatsh. Chem. 92 868 (1961) 49. J R Durig, A W Cox, Jr J. Mol. Struct. 38 77 (1977) 50. E G Codding, R A Creswell, R H Schwendeman Inorg. Chem. 13 856 (1974) 51. J R Durig, A E Stanley,MR Jalilian J. Raman Spectrosc. 10 44 (1981) 52. J Goubeau, R Baumgartner, W Koch, U MuÈ ller Z. Anorg. Allg. Chem. 337 174 (1965) 53. J R Durig, F Block, I W Levin Spectrochim. Acta 21 1105 (1965) 54. J E Griffiths Spectrochim. Acta 21 1135 (1965) 55. J R Durig, J M Casper J. Phys. Chem. 75 1956 (1971) 56. A I Fishman, A B Remisov, I Ja Kuramshin, I S Pominov Spectrochim.Acta, Part A 32 651 (1976) 57. J R Durig, M-Sh Cheng, J F Sullivan Indian J. Phys. 60B 285 (1986) 58. J R Durig, J B Robb II, T R Honeycutt, F F D Daeyaert, W A Herrebout, B J van der Veken J. Mol. Struct. 348 213 (1995) 59. P Groner, J S Church, Y S Li, J R Durig J. Chem. Phys. 82 3894 (1985) 767 60. J R Durig, J S Church, C M Whang,R D Johnson, B J Streusand J. Phys. Chem. 91 2769 (1987) 61. J R Durig, T K Gounev J. Mol. Struct. 350 19 (1995) 62. P Coppens, B J van der Veken, J R Durig J. Mol. Struct. 142 367 (1986) 63. B J van der Veken, P Coppens, R S Sanders, F F Daeyaert, J R Durig J. Mol. Struct. 272 305 (1992) 64. B J van der Veken, R S Sanders, J R Durig J. Mol. Struct. 216 113 (1990) 65. A B Remizov, I Ya Kuramshin, A I Fishman Zh.Obshch. Khim. 43 1406 (1973) a 66. R A Nyquist Appl. Spectrosc. 22 452 (1968) 67. J R Durig, J E Saunders J. Mol. Struct. 27 403 (1975) 68. H BuÈ rger, J Cichon, R Demuth, J Grobe Spectrochim. Acta, Part A 29 943 (1973) 69. H BuÈ rger, J Cichon, J Grobe, R Demuth Spectrochim. Acta, Part A 29 47 (1973) 70. H BuÈ rger, J Cichon, J Grobe, F HoÈ fler Spectrochim. Acta, Part A 28 1275 (1972) 71. J D Brown, R C Dobbie, B P Straughan J. Chem. Soc., Dalton Trans. 1691 (1973) 72. H BuÈ rger, R Eujen, R Demuth Spectrochim. Acta, Part A 31 1955 (1975) 73. R Schmutzler, O Stelzer, J F Liebman J. Fluorine Chem. 25 289 (1984) 74. L Heuer, R Schmutzler J. Fluorine Chem. 39 197 (1988) 75. A V Chernova, GMDoroshkina, S A Katsyuba, R R Shagidullin, I A Aleksandrova, N A Khailova, I Kh Shakirov Zh.Obshch. Khim. 63 2603 (1993) a 76. O G Sinyashin, V A Zubanov, S A Katsyuba, V N Shegeda, R Z Musin, N V Ivasyuk, E S Batyeva Zh. Obshch. Khim. 61 2513 (1991) a 77. F A Miller,D H Lemmon Spectrochim. Acta, Part A 23 1099 (1967) 78. W M A Smit, G Dijkstra J. Mol. Struct. 7 223 (1971) 79. H G M Edwards, J S Ingman, D A Long Spectrochim. Acta, Part A 32 739 (1976) 80. J Goubeau, H Haeberle, H Ulmer Z. Anorg. Allg. Chem. 311 110 (1961) 81. F A Miller, S G Frankiss, O Sala Spectrochim. Acta 21 775 (1965) 82. H G M Edwards, J S Ingman, D A Long Spectrochim. Acta, Part A 32 731 (1976) 83. N V Unnikrishnan, G Aruldhas Acta. Chim. Acad. Sci. Hung. 78 393 (1973) 84.V A Naumov, V Yu Nesterov, V P Novikov, S A Katsyuba Zh. Strukt. Khim. 25 (6) 129 (1984) b 85. H G M Edwards, V Fawcett Spectrochim. Acta, Part A 43 1345 (1987) 86. HGMEdwards, VFawcett Spectrochim. Acta, Part A45 147 (1989) 87. L J Bellamy Infra-Red Spectra of Complex Molecules (New York: Wiley, 1954) 88. L M Sverdlov, M A Kovner, E P Krainov Kolebatel'nye Spektry Mnogoatomnykh Molekul (Vibrational Spectra of Polyatomic Molecules) (Moscow: Nauka, 1970) 89. G Herzberg Molecular Spectra and Molecular Structure (New York: D van Nostrand Co., 1945) 90. J R Durig, F F D Daeyaert J. Mol. Struct. 261 133 (1992) 91. J R Durig, J B Robb II J. Mol. Struct. 375 53 (1996) 92. J R Durig, J B Robb II, T S Little J. Mol. Struct. 328 37 (1994) 93.B J van der Veken, T S Little, Y S Li,M E Harris, J R Durig Spectrochim. Acta, Part A 42 123 (1986) 94. F Seel, K D Velleman Chem. Ber. 105 406 (1972) 95. J R Durig, B J Streusand Appl. Spectrosc. 34 65 (1980) 96. E G Codding, C E Jones, R H Schwendeman Inorg. Chem. 13 178 (1974) 97. MJ Davis,DWHRankin, S Cradock J. Mol. Struct. 238 273 (1990) 98. J R Durig, J S Di Yorio J. Chem. Phys. 48 4154 (1968) 99. O A Raevskii, F G Khalitov Izv. Akad. Nauk SSSR, Ser. Khim. 843 (1971) c 100. N Fritzowsky, A Lentz, J Goubeau Z. Anorg. Allg. Chem. 386 203 (1971) 101. V A Naumov,N M Zaripov, V G Dashevskii Zh. Strukt. Khim. 12 (1) 158 (1971) b 102. N M Zaripov Zh. Strukt. Khim. 23 (2) 142 (1982) b768 103. A B Remizov, F S Bilalov, S A Katsyuba, E N Ofitserov Zh.Prikl. Spektrosk. 37 410 (1982) 104. S A Katsyuba, O N Nadyseva, V N Shegeda, G S Stepanov Zh. Prikl. Spektrosk. 56 725 (1992) 105. S A Katsyuba, R M Kamalov, O N Scherba, G S Stepanov, V A Alfonsov J. Mol. Struct. 435 281 (1997) 106. R R Shagidullin, D F Fazliev, L I Gurarii, E T Mukmenev Zh. Obshch. Khim. 45 1257 (1975) a 107. N M Zaripov, R M Galiakberov Zh. Strukt. Khim. 25 (4) 143 (1984) b 108. R R Shagidullin, I Kh Shakirov, P N Sobolev, I I Vandyukova, A Kh Plyamovatyi, O N Nuretdinova Izv. Akad. Nauk SSSR, Ser. Khim. 1793 (1986) c 109. K Bergesen, MBjoroy, T Granstad Acta Chem. Scand. 26 2156 (1972) 110. A B Remizov, D F Fazliev, R R Shagidullin, T G Mannafov Opt. Spektrosk. 34 252 (1973) 111. V A Naumov, N M Zaripov Zh. Strukt. Khim. 13 (5) 768 (1972) b 112. D F Fazliev, Candidate Thesis in Physicomathematical Sciences, Kazan State University, Kasan, 1975 113. RRShagidullin, IKh Shakirov,AKh Plyamovatyi,RP Arshinova, R A Kadyrov, B A Arbuzov Izv. Akad. Nauk SSSR, Ser. Khim. 1803 (1984) c 114. S M A Fleming, R J Wyma, R C Taylor Spectrochim. Acta 21 1189 (1965) 115. J R Durig, J M Casper J. Phys. Chem. 75 3837 (1971) 116. J R Durig, R J Harlan, P Groner J. Phys. Chem. 93 3041 (1989) 117. J R Durig, J M Casper J. Cryst. Mol. Struct. 2 1 (1972) 118. E D Morris, C E Nordman Inorg. Chem. 8 1673 (1969) 119. G C Holywell, D W Rankin, B Beagley, J M Freeman J. Chem. Soc., A 785 (1971) 120. P Forti, D Damiani, P G Favero J. Am. Chem. Soc. 95 756 (1973) 121. L V Vilkov, L S Khaikin Dokl. Akad. Nauk SSSR 168 810 (1966) d 122. N M Zaripov, V A Naumov, L L Tuzova Phosphorus 4 179 (1974) 123. V A Naumov, N A Gulyaeva, M A Pudovik Dokl. Akad. Nauk SSSR 203 590 (1972) d 124. A H Cowley,M J S Dewar,W R Jackson,W B Jennings J. Am. Chem. Soc. 92 1085 (1970) 125. S A Katsyuba, R R Shagidullin Zh. Obshch. Khim. 54 694 (1984) a 126. R R Shagidullin, I Kh Shakirov, O N Nuretdinova, S A Katsyuba, E A Filippova Izv. Akad. Nauk SSSR, Ser. Khim. 2232 (1985) c 127. S A Katsyuba, E A Filippova, R R Shagidullin, I Kh Shakirov, O N Nuretdinova Izv. Akad. Nauk, Ser. Khim. 391 (1998) c 128. V A Naumov, O N Kataeva, O G Sinyashin Zh. Strukt. Khim. 25 (3) 79 (1984) b 129. N Fritzowsky, A Lentz, J Goubeau Z. Anorg. Allg. Chem. 386 67 (1971) 130. E A Filippova, S A Katsyuba, R R Shagidullin, O G Sinyashin Zh. Prikl. Spektrosk. 46 761 (1987) 131. E A Filippova, Candidate Thesis in Physicomathematical Sciences, Kazan State University, Kasan, 1989 132. E A Filippova, S A Katsyuba, V N Shegeda, R R Shagidullin, A G Trusenev Izv. Akad. Nauk SSSR, Ser. Khim. 1531 (1989) c 133. V A Al'fonsov, D A Pudovik, I A Litvinov, S A Katsyuba, E A Filippova, M A Pudovik, V K Khairullin, E S Batyeva, A N Pudovik Dokl. Akad. Nauk SSSR 296 149 (1987) d 134. S A Katsyuba, E A Filippova, V A Al'fonsov, R R Shagidullin, V N Shegeda, D A Pudovik Zh. Obshch. Khim. 58 1744 (1988) a 135. R W Light, L D Hutchins,R T Paine, C F Campana Inorg. Chem. 19 3597 (1980) 136. S A Katsyuba, V A Al'fonsov, V N Shegeda, D A Pudovik Zh. Obshch. Khim. 66 772 (1996) a 137. G Y Schultz, I Hargittai, J Martin, J B Robert Tetrahedron 30 2365 (1974) 138. J P Albandt, J Gagnaire, J Martin, J B Robert Org. Magn. Reson. 5 33 (1973) 139. G Davidson, K P Ewer Spectrochim. Acta, Part A 39 419 (1983) 140. H R Linton, E R Nixon Spectrochim. Acta 11 146 (1959) 141. J E Drake, C Riddle Spectrochim. Acta, Part A 26 1697 (1970) 142. R Demuth Spectrochim. Acta, Part A 31 233 (1975) 143. K M Mackay, K J Sutton, S R Stobart, J E Drake, C Riddle Spectrochim. Acta, Part A 25 925 (1969) S A Katsyuba 144. J D Odom, C J Wurrey, L A Carreyra, J R Durig Inorg. Chem. 14 2849 (1975) 145. R Ahlrichs, R Heinzman, C Zirc Theor. Chim. Acta 43 29 (1976) 146. J R Durig, Zh Shen, W Zhao J. Mol. Struct. 375 95 (1996) 147. E R Nixon J. Phys. Chem. 6 1054 (1956) 148. M Baudler, L Schmidt Z. Anorg. Allg. Chem. 289 219 (1957) 149. S G Frankiss Inorg. Chem. 7 1931 (1968) 150. A H Cowley, S T Cohen Inorg. Chem. 5 1200 (1965) 151. S G Frankiss, F A Miller, H Stammreich, Th Teixeira Sans Spectrochim. Acta, Part A 23 543 (1967) 152. A B Lele, P L Kanitkar, K Sathianandan Indian J. Pure Appl. Phys. 8 151 (1970) 153. H BuÈ rger, J Cichon, R Demuth, J Grobe, F HoÈ fler Z. Anorg. Allg. Chem. 396 199 (1973) 154. J R Durig, T G Hizer J. Raman. Spectrosc. 17 97 (1986) 155. J R Durig, J F Sullivan, S Cradock J. Mol. Struct. 145 127 (1986) 156. A L Verma, W F Murphy, H J Bernstein J. Chem. Phys. 60 1540 (1974) 157. J P Perchard Spectrochim. Acta, Part A 26 707 (1970); 158. T Kitagawa, K Kusaki, T Miyazawa Bull. Chem. Soc. Jpn. 46 3685 (1973); 159. T Kitagawa, K Ohno, H Sugeta, T Miyazawa Bull. Chem. Soc. Jpn. 45 969 (1972); 160. J P Perchard, J C Monier Spectrochim. Acta, Part A 27 447 (1971) 161. I I Patsanovskii, E A Ishmaeva, A B Remizov, F S Bilalov, E N Ofitserov, A N Pudovik Dokl. Akad. Nauk SSSR 254 414 (1980) d 162. Spravochnik Khimika (Handbook for Chemist) Vol. 3 (Moscow: Khimiya, 1964) 163. L A Gribov Vvedenie v Molekulyarnuyu Spektroskopiyu (Introduc- tion in Molecular Spectroscopy) (Moscow: Nauka, 1976) 164. J L Duncan Spectrochim. Acta 20 1197 (1964) 165. G R Hunt,M K Wilson J. Chem. Phys. 34 1301 (1961) 166. K Venkateswarlu, M P Mathew Z. Naturforsch. B, Chem. Sci. 23 1296 (1968) 167. E E Aynsley, R Little Spectrochim. Acta 18 667 (1962) 168. H Wieser,W J Laidlaw, P J Krueger,H Fuhrer Spectrochim. Acta, Part A 24 (1968) 1055 169. S A Katsyuba, in The XXIVth European Congress on Molecular Spectroscopy (Abstracts of Reports), Prague, 1998 p. 193 170. L S Khaikin, O E Grikina, in Tez. Dokl. Vseros. Konf. po Teor- eticheskoi Khimii, Kazan, 1997 (Abstracts of Reports of the All- Russian Conference on Theoretical Chemistry, Kazan, 1997) p. 100 171. E A Filippova, S A Katsyuba, R R Shagidullin, V N Shegeda Zh. Prikl. Spektrosk. 49 454 (1988) a�Russ. J. Gen. Chem. (Engl. Transl.) b�Russ. J. Struct. Chem. (Engl. Transl.) c�Russ. Chem. Bull. (Engl. Transl.) d�Dokl. Chem. Technol., Dokl. Chem. (Engl.

ISSN:0036-021X    

出版商:RSC    

年代:2000

数据来源: RSC

摘要:

Russian Chemical Reviews 69 (9) 769 ± 782 (2000) Crown ethers in radiochemistry. Advances and prospects S V Nesterov Contents I. Introduction II. Methods for immobilisation of crown ethers III. Radiation-chemical transformations in extraction systems based on crown ethers IV. Areas of application of crown ethers V. Conclusion Abstract. and dicyclohexano- of immobilisation the for Methods Methods for the immobilisation of dicyclohexano- and non-substituted the on Data reviewed. are ethers crown non-substituted crown ethers are reviewed. Data on the radiation radiation chemistry of influence the on and compounds these of chemistry of these compounds and on the influence of irradiation irradiation on systems extraction crown-containing of properties the on the properties of crown-containing extraction systems pub- pub- lished systematically.discussed are decade last the over lished over the last decade are discussed systematically. Examples Examples of radiochemistry of fields different in ethers crown of use the of the use of crown ethers in different fields of radiochemistry are are presented. references 199 includes bibliography The presented. The bibliography includes 199 references. I. Introduction Intense development of nuclear power resulted in the accumula- tion of huge volumes of liquid and solid radioactive waste, which present serious hazard to the environment.1, 2 An approach to the reduction of the amount of radioactive wastes consists in upgrad- ing of the existing technologies for the reprocessing of spent nuclear fuel as well as in the decrease in the bulk of waste subject to disposal and utilisation of the already available waste. The major procedure for the reprocessing of spent nuclear fuel is liquid extraction in which the transuranium elements are separated from fission fragments with subsequent regeneration of uranium and plutonium.Organophosphorus compounds used currently in extraction processes, such as tributyl phosphate, have a number of disadvantages, in particular low resistance to radiation and insufficient selectivity in binding radionuclides. In this context, the search for more efficient cation-binding compounds is topical. Since the mid-1970s, macrocyclic polyethers or the so-called crown ethers have been regarded as promising extractants for the reprocessing of spent nuclear fuel and utilisation of radioactive wastes.3 Their feature is high selectivity of complexation with cations of metals possessing similar chemical properties; in many cases, the magnitude of the complex stability constant is charac- terised by the ratio of the size of the macrocycle cavity and the cation radius (see, e.g., Refs 4 and 5).The existing methods for the synthesis of crown ethers allow preparation of ligands with the predetermined ability to bind cations of specific metals.6 Macro- cyclic polyethers are chemically stable and can be easily regen- erated. These properties of crown ethers stimulated studies on the S V Nesterov N S Enikolopov Institute of Synthetic Polymeric Materials, Russian Academy of Sciences, ul.Profsoyuznaya 70, 117393 Moscow, Russian Federation. Fax (7-095) 420 22 29. Tel. (7-095) 917 43 30. E-mail: neste@cc.nifhi.ac.ru Received 21 March 2000 Uspekhi Khimii 69 (9) 840 ± 855 (2000); translated by V D Gorokhov #2000 Russian Academy of Sciences and Turpion Ltd DOI 10.1070/RC2000v069n09ABEH000586 769 769 772 775 780 possibility of their use in the reprocessing of spent nuclear fuel, utilisation of radioactive waste in different stages of the nuclear fuel cycle and in radiochemical analysis. The most effective macrocyclic compounds for the separation of plutonium and uranium from fission fragments proved to be non-substituted and dicyclohexanocrown ethers.However, a substantial disad- vantage of these extractants is their rather high solubility in water, which leads to their washout from the extraction and sorption systems.7 In this connection, studies are underway which are aimed at the development of methods for the synthesis of grafted and polymeric macrocyclic polyethers capable of binding selec- tively radionuclides and at the elaboration of novel technological processes based on them. The present review discusses the methods for immobilisation and modification of non-substituted and dicyclohexanocrown ethers and synthesis of polymers based on them, generalises the published data on radiation chemistry of crown ethers and crown- containing extraction systems and presents examples of the use of macrocyclic polyethers in different fields of radiochemistry.II. Methods for immobilisation of crown ethers The efficiency of the use of crown ethers in extraction and sorption systems depends, among other factors, on the extent of losses associated with solubility of these compounds in an aqueous phase. The rather high solubility of crown ethers in water leads to the transfer of extractants from the organic to the aqueous phase, to transfer of water to the organic phase in extraction processes 8 and the washout of complexants from the surface of porous supports in sorption processes.9 The necessity of an increase in hydrophobicity of non-substituted and dicyclohexa- nocrown ethers has been emphasised in the review by McDowell 7 where he first considered in detail major problems to be solved for practical application of crown ethers in radiochemical technology.Appreciable progress has been achieved in the search for ways of decreasing the solubility of benzo- and azacrown ethers. The methods of immobilisation and synthesis of polymeric aza- and benzo-substituted crown ethers were considered in detail in a number of papers and monographs.10 ± 15 A highly promising method for the synthesis of polymeric crown ethers, which can be used for the solution of radiochemical problems, is the anodic oxidation of dibenzocrown ethers on the surface of an electrode, which makes it possible to obtain crown-containing polymers of the type 1 in one step.16 ± 19 In the case of non-substituted and dicyclohexanocrown ethers, immobilisation and synthesis of polymeric compounds is much more difficult,20 since these macrocycles are rather inert770 O O O m O O O n O n O O m O O O n, m=0±2.chemically. The basic methods for the reduction of their solubil- ities in water developed to date can be divided into several groups. 1. Synthesis of macrocyclic compounds with hydrophobic aliphatic substituents. 2. Synthesis of polymers containing fragments of non-sub- stituted and cyclohexanocrown ethers or grafting of crown ethers onto porous supports. 3. Immobilisation (without covalent binding) of crown ethers on a matrix which is inert with respect to metal cations or on the surface of a porous support.The first group includes the methods of synthesis of dicyclo- hexanocrown ethers containing branched alkyl and/or other hydrophobic substituents. Introduction of branched alkyl chains into the cyclohexane ring leads to decrease in solubilities of crown ethers in water and simultaneously increases their solubilities in organic solvents. One of the most popular compounds of this type is 40,40(50)-di(tert-butylcyclohexano)-18-crown-6 (2, DTBDCH- 18-C-6).21, 22 But OO The crown ether 3 (DTODB-18-C-6) containing C8 tertiary alkyl groups was obtained 23, 24 by cyclisation of alkylated pyro- catechol 4 with 2,20-dichlorodiethyl ether (5, chlorex). Subsequent hydrogenation of dibenzocrown ether 3 leads to substituted dicyclohexanocrown ether 6 (DTODCH-18-C-6).The crown ether 3 is soluble in all organic solvents (including alkanes) and is virtually insoluble in water and saline solutions. The solubility of the crown ether 6 in water proved to be close to that of ordinary dibenzocrown ethers. It is important that the introduction of tert-alkyl substituents did not change the ability Me3SiOSi(Me)H OSi(Me)H OSi(Me)HOSiMe3 + n=1, 2. O n O O m O O O O nO O O O O O m O O m O O O n 1 O O But O O2 HOCH2 O O m O O n S V Nesterov Et OH cat, PhMe +Me(CH2)3CHCH2OH OH OH Et Cl(CH2)2O(CH2)2Cl (5) OH Me(CH2)3C KOH 4 Me O Et Et O O Me(CH2)3C H2 C(CH2)3Me cat Me O O Me O 3 O Et Et O O Me(CH2)3C C(CH2)3Me Me O O Me O 6 for complexation compared to that of non-substituted dicyclo- hexanocrown ethers.McDowell et al.25 described a method for the preparation of di-tert-butyldibenzo-21-crown-7 (7); its hydrogenation gives rise to the corresponding dicyclohexanocrown ether. O O But But O O O O O7 A promising method for the modification of non-substituted and cyclohexanocrown ethers was proposed.26, 27 In the example of 1,4-dioxane and morpholine, which are low-molecular-weight analogues of crown ethers, the possibility of radical-chain addi- tion of an aromatic hydrophobic group was demonstrated. Selective addition of allylbenzene to these cyclic ethers with the formation of compounds 8a,b takes place upon g-irradiation of a solution of allylbenzene (0.05 ± 0.15 mol litre71) in 1,4-dioxane or morpholine.The degree of conversion of allylbenzene exceeds 30% at absorbed doses<60 kGy. X (CH2)3 Y 8a,b X=Y=O(a), X=NH, Y =O (b). The second group includes methods of synthesis of polymeric crown ethers and methods of grafting of macrocycles on porous polymeric supports. Chauhan and Boudjouk 28 developed a one- Scheme 1 Me Me Me O O Si Me3SiOSi SiOSiMe3 O O O m 2%± 3%RhCl(PPh3)3 C6D6, 80 8C, 24 h O O O O O O O O O O O O n n n 9Crown ethers in radiochemistry. Advances and prospects O O O K2CO3 (or KOH), H2O P CH2Cl+ R O R O O O O O XCH2 P O O CH2X O P is polymer; R=NH2, CH(OH)Me, P(O)(OAlk)2; X=NH, C(Me)O, P(O)OAlk. O HNC O O O n O O NH2 HO2C O O+H2N O O O O O n O O O n O O N N O O O O O m n 10a,b n = 1 (a), 2 (b).P step method for the synthesis of polysiloxanes 9 containing frag- ments of non-substituted crown ethers and their oligomers. This method consists in selective catalytic oxidation of poly(methyl)- siloxane and addition of 2-hydroxymethylcrown ether to the siloxane chain (Scheme 1). The reaction proceeds under mild conditions, oxidation of Si7H bonds occurs without disproportionation and scission of the main chain. The polymers obtained are fairly well soluble in common organic solvents. n=1, 2. A method was proposed 29 ± 31 for immobilisation of a number of dicyclohexanocrown ethers on a support which is a chloro- methylated copolymer of styrene with divinylbenzene (Scheme 2), the content of macrocyclic groups ranging from 0.75 to 1.6 mg-equiv g71.It is presumed that the groups X through which grafting is effected influence insignificantly the distribution of electron density in the crown-ether fragment. The electron- donor properties of the oxygen atoms are not weakened and the capacities for complexation of the grafted and free crown ethers differ insignificantly. Amethod of radical polymerisation in the presence of a larger- size crown ether, viz., 42-crown-14, was employed to synthesise poly[(styrene)rotaxacrown ether] in which polymeric chains pass through the polyether rings.34 Polyrotaxanes based on polyur- ethanes and dibenzo-24-crown-8 were obtained and their sorption properties with respect to a number of metals were studied.35 The third group comprises methods for the preparation of compounds in which there are no covalent bonds between the crown ether molecules and the matrix. Solid extraction systems (SESys) and solid extraction solutions (SESol) synthesised by Yakshin et al.36 ± 40 are characteristic representatives of crown- containing sorbents obtained by this method.Kurmanaliev et al.32 described synthesis of polyimides 10a,b containing the structural moiety of diaminodicyclohexano-18- crown-6 or diaminodicyclohexano-24-crown-8 and studied the transport of alkali metal cations through thin films of these polymers (Scheme 3). For polymer 10a, the selectivity of the cation transport changes in the order K+>Cs+>Na+ and for polymer 10b it changes in the order Cs+>K+>Na+.The sequences obtained match the binding capacities of free crown ethers.4 Thus, polymerisation of crown ethers does not change the selectivity of complexation. Solid extraction systems based on copolymers of styrene with divinylbenzene containing 40%± 43% of a crown ether were obtained by adding a homogeneous mixture of styrene, divinyl- benzene, crown ether and a polymerisation initiator to an aqueous solution of starch and subsequent polymerisation of the monomers.36 This resulted in the formation of ball-shaped Watanabe et al.33 synthesised a novel crown ether 11 contain- ing the pyrrolidine ring and grafted it onto a chloromethylated copolymer of styrene with divinylbenzene.P CO2H O CNH O O O CH2Cl+HN O 11 O P CH2N O 771 Scheme 2 Scheme 3 O n OO O m n O NaHCO3, MeCN O 70 ± 80 8C, 5 days O n O O O n772 granules. Their stability was studied under conditions of contact with water for two days. The losses amounted to 90% for dicyclohexanocrown ethers such as dicyclohexano-18-crown-16 (DCH-18-C-6, 12), DTBDCH-18-C-6 (2), dicyclohexano-21- crown-7 (DCH-21-C-7, 13) and dicyclohexano-24-crown-8 (DCH-24-C-8, 14). O n O O R R O O O m 2, 12 ± 14 R=But, n=m = 1 (2); R=H: n=m = 1 (12), n =1, m = 2 (13), n=m = 2 (14). Washout of poorly water-soluble dibenzocrown ethers, viz., dibenzo-18-crown-6 (15), dibenzo-21-crown-7 (16), dibenzo-24- crown-8 (17), dibenzo-30-crown-10 (18) and di(tert-butylbenzo)- 18-crown-6 (19), occurred to lesser extent and the losses ranged from 10% to 30%.And finally, the losses of crown ethers O n O O R R O O O m 15 ± 19 R=H: n=m = 1 (15), n =1, m = 2 (16), n=m = 2 (17), n=m = 3 (18); R=But, n=m = 1 (19). containing hydrophobic isooctyl substituents in the aromatic and dicyclohexane rings, DTODB-18-C-6 (3) and DTODCH-18-C-6 (4), did not exceed 5%± 10%. In order to reduce losses of crown ethers, Yakshin et al.36 proposed to presaturate the aqueous phase with the corresponding crown ether. Solid extraction solutions containing from1%to 50% (w/w) of extractants were obtained 37 ± 39 by heating different liquid and low-melting crown ethers [DB-18-C-6 (15), cis-syn-cis-DCH-18- C-6 (12), DTBDCH-18-C-6 (2), DTODCH-18-C-6 (4) and others] with naphthalene, biphenyl or paraffin to their melting temperatures with subsequent dispersion of homogeneous melts in an aqueous solution of starch.The ball-shaped granules formed were separated and their sorption abilities were studied. The onset of equilibrium in solutions in contact with SESys and SESol obtained by the methods described above takes a very long time (from 0.5 h to a few days 38, 40) since under these conditions the binding of metal cations is limited by diffusion of ions (including anions) to the macrocyclic ligand through an inert matrix. For this reason, the use of these extractants under dynamic conditions is not advisable. Solid extraction systems and solutions could not absorb metal cations from solutions in nitric acid.It was suggested that the macrocyclic centres are blocked in the nitric acid medium;40, 41 however, the nature of this effect has not been clarified. It should be noted that such a block does not occur for solutions in hydrochloric acid. The third group of methods includes also immobilisation of crown ethers on porous supports either by their application onto the surface or by impregnation. The advantage of the specimens obtained by this method is the accessibility of macrocyclic groups localised on the surface of the support for contact with metal ions. Chuang and Lo 42 described the preparation of a crown- containing sorbent by application of DCH-18-C-6 (12) onto silica gel.Silica gel was preliminarily dried at 110 8C, treated with dichlorodimethylsilane for substitution of free hydroxy groups S V Nesterov and then a solution of the crown ether in chloroform was added. After removal of the solvent, a sorbent was obtained which contained 3.8% of DCH-18-C-6 (12) and was sufficiently resistant to washout of the crown ether. Horwitz et al.21, 22, 43, 44 proposed a simple and efficient method for the preparation of a crown-containing sorbent which enables isolation of Sr2+ from nitric acid solutions. A solution of DTBDCH-18-C-6 (2) in octan-1-ol is immobilised on an inert porous support (Amberlite XAD-7 or Amberchrom CG-71 ms).The presence of hydrophobic tert-butyl substituents in the crown ether molecule provides good fixation of the macrocycle on the surface of the support on contact with aqueous solutions. This method for immobilisation of crown ethers has been patented; the commercial specimen called Sr-Spec may be purchased from the EIChrom Industries Inc. (USA). Another impregnate was pre- pared at the V I Vernadsky Institute of Geochemistry (Russia) by impregnation of a porous copolymer of styrene and divinylben- zene with a 10% solution of DCH-18-C-6 (12) in chloroform or tetrachloroethane.2, 45 Nazarenko et al.46 described a method of synthesis of crown- containing sorbents by coprecipitation of 18-crown-6 and phos- phomolybdic heteropolyacid. These authors 46 also studied the influence of sorbent composition and medium acidity on the sorption properties with respect to rare-earth elements of the cerium group.The capacity of sorbents for the metal was 10 ± 41 mg g71 depending on the specimen composition and the nature of the metal to be sorbed. The data on the resistance of sorbents to washout are not presented. III. Radiation-chemical transformations in extraction systems based on crown ethers High resistance of an extractant to irradiation is one of the main factors which determine the efficiency of its use in radiochemical processes. It is presumed 47, 48 that the ionising irradiation initiates radiation-chemical transformations which may result in consid- erable changes in properties of the extraction system.1. Decomposition of extractant leads to lower extraction efficiency. 2. The formation of radiolysis products affects unfavourably the selectivity of extraction. 3. The formation of active compounds prevents the partition of phases. 4. A third solid phase precipitates. 5. Formation of corrosive acids is possible. 6. Oligomers may be formed, which will unfavourably affect the extractant regeneration. It is known 49 that macrocyclic polyethers form more stable complexes with metal cations in solutions than their non-cyclic analogues having the same number of donor atoms. This phe- nomenon was called the `macrocyclic effect'. Upon irradiation of crown ethers, two main processes may occur, viz., abstraction of a hydrogen atom from the methylene group of the polyether ring or from the hydrocarbon substituent (if present) and the cleavage of the ether bond of the macrocycle.The relationship between the extraction ability of a macrocycle and its cyclic structure was confirmed experimentally. Draye et al.50, 51 synthesised products of the ether bond cleavage which can be formed from crown ethers upon irradiation. It turned out that the capacity for complexation of non-cyclic compounds mimicking the radiolysis products of DCH-18-C-6 (12) is much lower than that of the initial crown ether. Thus, the cleavage of the ether bond of the macrocycle upon radiolysis leads to a decrease in the efficiency of complexation. The first studies of the radiation chemistry of macrocyclic polyethers were initiated in the 1980s.52 ± 55 The major attention has been paid to the g-radiolysis products of individual crown ethers in the gas and condensed phases.52 ± 54, 56 Radiation-chem- ical yields of molecular hydrogen [G(H2)] and gaseous hydro- carbons have been calculated (Table 1).The yields obtained by different authors for the same compounds exhibited substantialCrown ethers in radiochemistry. Advances and prospects Table 1. Radiation-chemical yields of radicals [G(R)] and molecular hydrogen [G(H2)] in g-irradiated crown ethers. Compound 1,4-Dioxane Morpholine 12-Crown-4 (20a) 15-Crown-5 (20b) BeCl2 . 15-Crown-5 MgCl2 . 15-Crown-5 CaCl2 . 15-Crown-5 18-Crown-6 (20c) BeCl2 .18-Crown-6 MgCl2 . 18-Crown-6 CaCl2 . 18-Crown-6 SrCl2 . 18-Crown-6 BaCl2 . 18-Crown-6 Dicyclohexano-18-crown-6 (12) Dicyclohexano-24-crown-8 (14) Benzo-12-crown-4 (21a) Benzo-15-crown-5 (21b) Benzo-18-crown-6 (21c) Dibenzo-18-crown-6 (15) 3-Methyldibenzo-18-crown-6 (22a) 3,12-Dimethyldibenzo-18-crown-6 (22b) 3-Ethyldibenzo-18-crown-6 (22c) 3-Butyldibenzo-18-crown-6 (22d) 3,12-Dibutyldibenzo-18-crown-6 (22e) 3,12-Dihexyldibenzo-18-crown-6 (22f) [4 : 6]-Dibenzo-30-crown-10 (22g) Dibenzo-24-crown-8 (17) Dibenzo-30-crown-10 (18) [2.1]-Cryptofix (23) [2.2]-Cryptofix (24) Hexathia-18-crown-6 (25) [2.2.2]-Cryptand (26) Note: n is the number of radicals, m is the number of molecules. differences. Later, papers were published 60, 61 in which intermedi- ate species formed as a result of interaction of water radiolysis products with 18-C-6 and DCH-18-C-6 were studied by pulsed radiolysis and EPR methods.Based on the experimental results available at that time, it was concluded 52 ± 56, 60 ± 62 that the abstraction of hydrogen atoms from the methylene groups of the polyether ring is predominant in the radiolysis. Later, investigators began to give much attention to the identification of intermediate products of radiolysis of crown ethers. The results obtained provided evidence that it is the cleavage of the macrocycle that dominates in the radiolysis of crown ethers and extraction systems based on them. 1. Radiolysis of crown ethers Information about the nature of the intermediate and final radiolysis products of crown ethers, radiolysis mechanism and radiation-chemical yields is extremely important since this allows assessment of the macrocycle resistance to scission.Unfortu- nately, at present there are no data on primary radiolysis products (radical cations, excited molecules) or the contribution of ionisa- tion and excitation processes to the mechanism of radiolysis of polyethers. Furthermore, there is no information that studies G(R) /n (100 eV)71 2.70.7 75.30.8 3.70.7 5.40.8 5.90.7 11.52.0 6.50.8 8.11.1 6.01.1 5.71.1 3.21.1 5.60.8 7.91.1 8.30.7 9.30.3 10.10.5 3.0 72.90.5 1.10.2 0.70.15 1.70.3 0.60.1 0.160.02 70.60.1 1.00.2 1.00.4 0.80.1 1.20.2 1.40.3 0.70.17 1.30.2 70.90.15 3.70.7 5.30.8 1.10.3 3.60.6 Ref.64 72 64 61 64 61 66, 67 66, 67 66, 67 64 61 67 67 67 67 67 63 61 70 70 70 70 61 70 70 70 70 70 70 70 70 70 72 72 72 72 along these lines are being carried out systematically. To some extent, this is explained by the complexity of experiments which require investigation of radiolysis of macrocyclic systems under conditions of matrix isolation of crown ethers or in adsorbed monolayers, e.g., on zeolites. An exception are the studies of Kuruc et al.63, 64 who attempted to identify radicals and radical cations in g-irradiated crown ethers by the methods of spin traps and isolation in Freon matrices.However, it was impossible to interpret unambiguously the results obtained because of poor resolution of the EPR spectra. Much attention has been paid to studies of radical products of radiolysis of crown ethers stabilised at the temperature of liquid nitrogen. Schemes of the formation of radicals were proposed and assessment was made of the stability of g-irradiated (at 77 K) non-substituted crown ethers 20a ± d,58, 59, 65 complexes of 15-crown-5 (20b) and 18-crown-6 (20c) with alkaline-earth metal chlorides,66 ± 69 benzo- (21a ± c) and dibenzocrown ethers (22a ± g),58, 70, 71 aza- and thiacrown ethers (23 ± 25), as well as of [2.2.2]-cryptand (26).72, 73 It was proposed 68, 71 to use the values of radiation chemical yields of radicals G(R) as characteristics of the radiation stability (see Table 1).The data presented in Table 1 make it possible to G(H2) /m (100 eV)71 1.26 1.41 7771.60 70.90.2 0.70.2 0.90.2 1.62 0.3 0.70.2 0.70.1 1.00.1 1.10.2 70.86 0.12 770.0014 70.175 0.01 0.0017 77777770.01 0.0019 77777 773 Ref. 57 56 56 58, 59 58, 59 58, 59 56 53 58, 59 58, 59 58, 59 58, 59 56 53 54 56 53 54 53 54S V Nesterov 774 O O O O n n O O O O 21a ± c 20a ± d n = 1 (a), 2 (b), 3 (c). n = 1 (a), 2 (b), 3 (c), 4 (d). R1 O n O O O O NH HN O O O O O 23 R2 m 22a ± g n=m=1: R1=Me, R2 = H (a); R1=R2=Me (b); R1=Et, R2=H(c); R1=Bun, R2= H (d); R1=R2=Bun (e); R1=R2=C6H13 (f); n =2, m=4:R1=R2= H (g).their use under real conditions. Under the action of ionising radiation on an extraction system, both the processes of radiolysis of individual components of the system and the reactions of the products of solvent radiolysis (solvated electrons, hydrogen atoms and radicals) with the extractant contribute to the formation of products of radiation-chemical transformations. Due to these processes, the composition and properties of the system may be changed considerably. We will consider below the papers in which changes in the extraction capacity of crown-containing systems under the action of irradiation were studied and the stable products formed in the condensed phase were analysed. Thus irradiation of DCH-18-C-6 (12) and 18-C-6 (20c) in chloroform solutions (c=0.05 ± 0.1 mol litre71) resulted in a 20%decrease in Sr2+ extraction from aqueous solutions of picric acid at absorbed doses of 30 ± 50 kGy.74 No new signals were present in the 1H NMR spectra of irradiated solutions. This fact and the results of experiments in which preirradiated chloroform was used indicate that the coefficient of strontiumdistribution (DSr, i.e., the ratio of the concentration ofmetal ions in the organic phase to that in the aqueous phase) decreases due to formation of chloroform radiolysis products, viz., HCl and C2Cl6.And indeed, the addition of these compounds to the extraction system led to a decrease in DSr.O O S S O HN O O O N S S N HN O O S S O 24 26 25 Neither UV and IR spectra were changed nor new peaks in a chromatogram were observed following g-irradiation of a 10% solution of DCH-18-C-6 (12) in chloroform to absorbed dose of 168 kGy.75 Upon prolonged contact of DCH-18-C-6 with a-emit- ters (Pu4+ and U6+) during which the total absorbed dose corresponded to 1.6 MGy, the extraction capacity of the macro- cycle changed within acceptable limits: extraction of Pu4+ and U6+decreased by *2% and 23%, respectively. Irradiation of tributyl phosphate under analogous conditions leads to the formation of considerable amounts of mono- and dibutyl phos- phate, due to which the selectivity of plutonium extraction in the process of plutonium and uranium extraction (PUREX) decreased considerably.state that the radiation-chemical behaviour of crown ethers on the whole is similar to that of their low-molecular-weight non-cyclic analogues. Thus the yields of radicals [G(R)] in non-substituted crown ethers are several times higher than those in thiacrown ethers. The same regularity was observed earlier in the radiolysis Shukla et al.76 revealed a strong influence of irradiation on the of aliphatic alcohols, ethers and thiols.74 Benzo- and dibenzo- ability ofDCH-18-C-6 (12) solution in toluene (a mixture of isomers crown ethers show a negative deviation from the additivity rule in at a concentration ranging from 0.1 to 0.3 mol litre71) to extract the formation of radicals, which is due to the transfer of absorbed Pu4+ andU6+ fromsolutions in nitric acid.Irradiation was carried energy from the macrocycle to the aromatic group, i.e., a out in the presence of air up to an absorbed dose of 710 kGy. The protection effect is observed. In addition, in benzocrown ethers distribution coefficients of uranium and plutonium decreased two- the value ofG(R) depends on the size of the macrocycle, number of and tenfold due to decomposition of the extractant and toluene. aromatic groups in the polyether ring and the length of the At present, there is no unequivocal answer to the question hydrocarbon chain of the alkyl substituent.71 The yields of radicals for complexes of non-substituted crown ethers with alkaline-earth metal chlorides are considerably higher than for `free' polyethers .In this case, a substantial role is played by the matching of the macrocycle cavity size to the radius of the metal cation. Thus the resistance of 15-crown-5 complexes to radiation changes in the sequence Be2+&Mg2+>Ca2+ (see Refs 67, 68), whereas for 18-crown-6 complexes the sequence is Be2+>Mg2+>Ca2+>Sr2+>Ba2+ (see Ref. 68). Apparently, the dominant process of radiolysis of the majority of crown ethers is the cleavage of the ether bond rather than the abstraction of a hydrogen atom, as was believed earlier.62 This conclusion is supported by the complexity and multicomponent pattern of the EPR spectra of g-irradiated (at 77 K) crown ethers, which is explained 59, 66, 69 by the superposition of signals belong- ing to several paramagnetic species and also by the absence of correlation between the radiation-chemical yields of radical prod- ucts and of molecular hydrogen, G(R)=2G(H2) (see Table 1).An exception is hexathia-18-crown-6 where abstraction of hydrogen from the methylene groups of the polyether ring predominates.73 2. Radiolysis of model extraction systems based on crown ethers about the influence of ionising irradiation on the properties of membranes based on crown ethers. On the one hand, preirradia- tion of a membrane based on 18-C-6 (20c) in chloroform (c=1073 mol litre71) with a dose of 0.45 kGy in the presence of air results in a three- and twofold increase in the transport of Cs+ and Sr2+ cations, respectively, (see Ref.77) and the shift of selectivity of cation binding (fromK+ to Cs+). Vijayavergiya and Mookerjee 77 explain this effect by the reaction of crown ether 20c with chlorine atoms which are the radiolysis product of chloro- form in the presence of HCl resulting from radiolysis of chloro- form in air. These authors note the occurrence of a shift to the longwave region of the UV spectrum; however, they do not indicate to which absorption band this refers. On the other hand, the transport of U6+and Pu4+ cations decreases after g-irradi- ation of a membrane composed of a 0.2 M solution of DCH-18-C-6 in toluene on a polypropylene support.76 The influence of irradiation becomes apparent after absorbed doses of 460 kGy for uranium and 100 kGy for plutonium.Above 710 kGy, the fluxes of U6+ and Pu4+ are decreased ca. twofold and ca. sixfold, respectively. The radiation resistance of DCH-18-C-6 (12) immobilised on silica gel (3.8%) is estimated by comparing the sorption capacities of irradiated and non-irradiated samples with respect to Sr2+ (see Ref. 42). At absorbed doses ranging from 0 to 1.0 MGy, the sorption ability is virtually unchanged. Investigation of the behaviour of extraction systems based on crown ethers under irradiation sheds light on the possibility ofCrown ethers in radiochemistry. Advances and prospects Analysis of the final radiolysis products of extraction systems based on crown ethers is a rather complicated task because of low radiation-chemical yields.Successful solution of this problem requires chromatographic separation of the decomposition prod- ucts from the initial compounds. Thus irradiation (50 ± 400 kGy) of B-15-C-5 (21b) in chloroform (c=0.001 ± 0.15 mol litre71) gave rise to the formation of the chloroform radiolysis products and mono- and dichloro derivatives of B-15-C-5 (21b) which represented ionic associated complexes (B-15-C-5)H+Cl7 and (B-15-C-5)H+Cl¡2 . These were isolated chromatographically and identified by mass-spectometry.78 The products of the reaction of the chlorine atoms, which were derived from radiolysis of chloro- form, with the aromatic ring of the crown ether were absent. The IR spectra of the radiolysis products revealed a band at 1740 cm71 which was attributed to the carbonyl group.On this ground, a conclusion was made about the macrocycle opening. Myasoedova and Dmitrieva 79 studied radiation-chemical transformations in a system containing 0.1 mol litre71 of DCH-18-C-6 (12) in a 1 : 1 mixture of 1,1,2,2-tetrachloroethane and 3M HNO3. The organic and aqueous phases were stirred and irradiated in air. The absorbed dose was*105 Gy at a dose rate of 2 Gy s71. After distillation of tetrachloroethane, the residue was analysed by chromato-mass spectrometry. More than 40 radiolysis products were found. The probable products corresponding to the main peaks in the chromatogram are shown below: O NO2 OH O O O O O OH O NO2 O m/z=374 m/z=292 O O Cl O O O O O O O O Cl O O m/z=366 m/z=400 A scheme of radiation-chemical transformations of DCH- 18-C-6 (12) was proposed.The major processes of radiolysis are presumed 79 to involve oxidation of the crown ether leading to the macrocycle opening and formation of acyclic products, addition of the products of tetrachloroethane and HNO3 radiolysis to the initial crown ether, elimination of the oxyethylene fragment from the macrocycle with preservation of the cyclic structure (analo- gous to the fragmentation observed in the electron-impact mass spectrometry) and dehydrogenation of the cyclohexane ring. It was noted that at least two of the most abundant peaks in the chromatogram (m/z 366 and 400) were apparently due to the presence of an impurity derived from DB-18-C-6 and intermediate products appearing during its formation.Draye et al.80 studied the g-radiolysis products of a solution of DCH-18-C-6 (12) and uranyl nitrate in 1M nitric acid using NMR and IR spectroscopy and chromato-mass spectrometry. Samples were irradiated without preliminary evacuation up to the absorbed dose of 3.29 MGy. New bands corresponding to the OH and C=O groups were observed in the IR spectra. At the same time, the NMR spectra of irradiated and non-irradiated samples remained identical. Of the seven basic peaks, the most prominent were the peaks with m/z 160 and 204 which are attributed to compounds 27 and 28, respectively. OH OH O(CH2)2O(CH2)2OH O(CH2)2OH 28 (G=0.23) 27 (G=0.29) 775 Unfortunately, Draye et al.80 neither presented the experi- mental NMR and mass-spectra nor reported whether or not the samples were dehydrated before the analysis; therefore, the identification of the radiolysis products of crown ethers remains ambiguous.Abashkin et al.48 studied the radiolysis of cis-syn-cis- and cis-anti-cis-isomers of DCH-18-C-6 (12) and their complexes with strontium nitrate. At the absorbed doses below 500 kGy, no radiolysis products were found in free crown ether. Increase in the absorbed dose to 1500 kGy led to the appearance of a radiolysis product with a carbonyl group which was detected by chromatography. Presumably,48 radiolysis resulted in a symmet- rical scission of the macrocycle over two C7O bonds with the formation of a compound giving a peak at m/z 186 in the mass spectrum.OO(CH2)2OEt m/e=186 It was found that cis-syn-cis-12 is more stable than cis-anti-cis- 12. The complex of DCH-18-C-6 (12) with strontium nitrate is less resistant to the radiation than the free crown ether. At 500 kGy, complex decomposition products are formed, which is evidenced by the appearance of the absorption bands of the unbound nitrate in the IR spectrum and by the decrease in the intensity of the absorption bands of the complex. The macrocycle of the complex also undergoes decomposition yielding a product containing a carbonyl group. Thus, DCH-18-C-6 possesses high resistance to irradiation in the range of doses received by extractants during the spent nuclear fuel reprocessing cycle, and decomposition of the macrocycle may be disregarded.Irradiation of extraction systems based on crown ethers with larger doses leads mainly to macrocycle scission, while the formation of cross-linking or polymerisation products is hardly probable. A decrease in the extraction ability of the systems based on crown ethers under the action of ionising irradiation is asso- ciated with the formation of the solvent radiolysis products; for this reason, the use of chlorine-containing solvents is undesirable. In studies of the radiolysis of dicyclcohexanocrown ethers and their complexes with metal cations, the purity of samples is highly important, since even insignificant amounts of dibenzocrown ethers may lead to erroneous identification of the radiolysis products.48, 79 IV. Areas of application of crown ethers 1.Reprocessing of spent nuclear fuel and extraction of fission fragments from radioactive waste Reprocessing of spent nuclear fuel is a stage of the fuel cycle in which uranium and plutonium are separated from fission frag- ments. The major role in this process is played by liquid extraction. The use of crown ethers as extractants, in particular of dicyclohex- anocrown ethers, ensures higher selectivity of the extraction of uranium and transuranium elements compared to that provided by extractants used currently and allows separation of these radionuclides from the fission fragments.3, 8 New extraction proc- esses based on the use of macrocyclic polyethers were proposed which allow selective isolation of Sr and Tc,81 ± 85 U(VI) and Pu(IV) 51, 86 from solutions in nitric acid with high specific activity.Dicyclohexanocrown ethers are efficient in the utilisation of liquid radioactive waste formed in the processing of nuclear fuel and deactivation of equipment. Depending on their origin, the radioactive wastes can have acidic or alkaline reaction. Concen- trates of such wastes possess high activities due to the presence of long-living cesium and strontium isotopes with half-life times (t1/2) of 28.5 and 30.1 years for 90Sr and 137Cs, respectively.87 These concentrates are stored in geological storage pits. Selective extrac- tion of radionuclides from them makes it possible to decrease considerably the bulk of waste, which simplifies further reprocess- ing and storage.In addition, 90Sr and 137Cs are widely used as776 sources of b-radiation and energy generators.88 At present, two processes of radioactive waste treatment are under testing in the USA. Thus in the Argonne National Laboratory, a process for selective extraction of 90Sr (the so-called SREX, i.e., Strontium Extraction) from solutions in nitric acid was developed. This method employs DTBDCH-18-C-6 (2) (c=0.15 mol litre71) as a selective extractant of Sr2+. The extraction systemis supplemented with tributyl phosphate to prevent formation of a third phase and with kerosene Isopar L1 as the solvent. Pilot technological testing has shown that crown ether 2 allows 99.98% extraction of strontium in a single cycle, the extract containing virtually no Na, Fe, Ca ions and other ballast elements.84, 85 A disadvantage of the SREX process is co-extraction of Pb2+ and K+ cations, and this fact requires that the process included an additional stage for the separation of these cations.Researchers of the OakRidge National Laboratory devised an SRTALK process for the extraction of technetium and strontium from alkaline radioactive waste.81 ± 83 The extraction system consists of DTBDCH-18-C-6 (2) solution (c=0.02 mol litre71) in a 1 : 1 (v/v) mixture of kerosene IsoparL1 and tributyl phosphate as a modifier. Selective ligands with respect to cesium cations are dibenzo- crown ethers and calixarenes containing fragments of crown ethers.At present, intensive studies of extraction and sorption systems based on them and aimed at extracting Cs from radio- active waste are underway.89 ± 94 Photosensitive monocrown-6- azobenzocalix[4]arene (29) was synthesised.89, 95 When it is used for the extraction of Cs+ ions, the distribution coefficient increases threefold under the action of UV light. MeO OMe O O O O O O O O R1 O R1 O O O O O O O R2 O O R2O O N N O O 29 30a ± c R1=COOH, R2= H (a); R1=R2=SO3Na (b); R1=R2=SO2N(CH2CH2OH)2 (c). The preparation of water-soluble calix[4]arene-bis(crown ethers) 30a ± c has been described. These compounds were shown 91 to have high affinities for cesium ions.Calix[4]arene[- R R O O O O O O O O OO OO O O O O O O O O O O O O 31 32, 33R=H(32); R2=CH=CHCH=CH (33). S V Nesterov bis(tribenzo)-crown-6] (31) 90 and non-symmetrical calix[4]arene[- bis-crown ethers] (32, 33) 96 were synthesised. These were shown to possess high cation-binding abilities for cesium ions and to be virtually inert with respect to sodium ions. High selectivities of compounds 31 ± 33 are determined by the presence of aromatic substituents in the polyether ring resulting in a change in the sp3 hybridisation of carbon atoms of the oxyethylene moiety to sp2 hybridisation.90 It was proposed to use crown-containing calixarenes as supports in membranes for selective extraction of cesium ions from nuclear waste 97 ± 99 and in ion-selective electrodes.100 Lamare et al.101 described a method for selective extraction of traces of cesium by calixarene-crown ethers from sodium nitrate solutions. a.Effect of solution acidity and the crown ether structure on the extraction efficiency Quantitative parameters of the extraction process are character- ised by the distribution coefficient D and the extraction constant Kex.As a rule, extraction of metal ions by a macrocycle from acidic and neutral solutions is described 102 ± 107 by the scheme nMm++kCE+mAn7 Mn(CE)kAm , whereMis the metal cation, m is the metal oxidation state, CE is a crown ether, k is the number of macrocyclic molecules involved in complexation, A is an anion and n is the anion charge.Mn(CE)kAm is the complex of the crown ether with the cation. For most systems, k=1 or 2. For k=2, a sandwich-like complex is formed. The extraction constant may be calculated using the following relation: (1) Kex= âM âMmáänâCEäkâAn¡äm , nÖCEÜkAmä where [CE] and [Mn(CE)k Am]* are the concentrations of a crown ether and its complex in the organic phase, [Mm+] and [An7] are the concentrations of a cation and an anion in the aqueous phase. The distribution coefficient is determined from the relation(2) D=âMnÖCEÜkAmä . âMmáän Substituting Kex from the relation (1) gives D=Kex[CE]k[An7]m (3) or (4) logD=logKex+k log[CE]+mlog[An7]. The compositions of crown-containing extraction systems and the distribution coefficients of Cs, Sr, U and Pu ions obtained for them are listed in Table 2.The dependence of the distribution coefficient of a metal on the concentration of crown ether at a constant concentration of a counterion allows determination of the extraction constant and the composition of the crown ether complexes with metal cations (Table 3). It should be noted that the above-cited scheme of extraction of metal ions by macrocycles does not take into account a number of factors. As a rule, transition of a metal salt into the organic phase in the absence of a macrocycle and dissociation of a macrocyclic complex in the organic phase may be disregarded;113 however, the solution acidity is one of the crucial factors determining quanti- tative parameters and the mechanism of the extraction process.In an acidic solution, complexation is complicated by the involve- ment of protonated ligands.119, 122, 123 In addition, it was found that DCH-18-C-6 (12) forms complexes effectively with nitric and picric acids. It is noteworthy that up to 10 molecules of nitric acid per macrocycle can be extracted from the aqueous to the organic phase.3, 8, 124, 125 The effect of concentration of nitric and picric acids on the extraction of strontium,87, 102 ± 104, 119 ura- nium,105, 107, 114, 126, 127 and plutonium 8, 105, 106, 120 by dicyclo- hexanocrown ethers was investigated in numerous studies. It wasCrown ethers in radiochemistry. Advances and prospects Table 2. Distribution coefficients of Cs, Sr, U and Pu ions in extraction systems based on crown ethers.Crown ether Concentration of crown ether /mol litre71 DCH-18-C-6 (12) see b see b see b see b see c see c see b see b see c see c see c see c see c see c see c see c see c see c see c see c see b see b see b see b DTBDCH-18-C-6 (2) 18-C-6 (20c) DB-21-C-7 (16) B-15-C-5 (21b) DCH-24-C-8 (14) 4-Methylbenzo-15-crown-5 0.1 0.1 0.1 0.1 0.1 0.05 0.05 0.05 0.05 0.014 0.05 0.1 0.1 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.1 0.1 0.1 0.1 0.1 0.2 0.537 0.537 0.1 0.1 0.1 0.1 0.1 0.1 0.11 0.1 0.15 0.1 0.11 0.1 0.1 0.1 0.1 tert-Butylbenzo-21-crown-7 or n-decylbenzo-21-crown-7 DB-21-C-7 Tribenzo-21-crown-7 DB-24-C-8 Tribenzo-24-crown-8 Tetrabenzo-24-crown-8 DB-24-C-8 (17) 0.025 0.025 0.025 0.025 0.025 0.01 0.01 0.01 0.01 Di(tert-butylbenzo)-21-crown-7 0.01 0.01 0.01 0.01 9610± 3 2610± 3 3610± 3 6610 ±3 2610± 2 ????2.17610 ±4 1.12610 ±3 4.62610 ±3 1.64610 ±2 a 5M HCl was added; b cis-syn-cis-isomer, c mixture of isomers, dpH=3, e neutral medium, f alkaline medium, [OH7]=1.75 g-ion litre71.found that the efficiency of the extraction depends on the nature of the solvent, metal, counterion, etc. Thus in the extraction of Sr2+ by DCH-18-C-6 (12) in a butanol ± octanol mixture (80 : 20), the distribution coefficient DSr depends on the concentration of nitric Solvent chloroform n-hexylbenzene o-nitrophenyloctyl ether n-octanol 1,1,2,2-tetrachloroethane "1,2-dichloroethane nitrobenzene chloroform 1,1,2,2-tetrachloroethane chloroform "nitrobenzene toluene benzonitrile o-dichlorobenzene dichloromethane 1,2-dichloroethane benzene chloroform xylene toluene 1,2-dichloroethane benzene 1,2-dichloroethane chloroform toluene benzonitrile "n-octanol n-hexylbenzene o-nitrophenyloctyl ether 1,2-dichloroethane benzene 1,2-dichloroethane chloroform """""benzene 1,2-dichloroethane n-hexylbenzene 1,2-dichloroethane """"chloroform dichloromethane 1,2-dichloroethane nitrobenzene o-xylene 1,2-dichlorobenzene 1,2-dichloroethane nitrobenzene acid and increases as the HNO3 concentration is increased from 1 to 8 mol litre71 (see Ref.102). When 1,1,2,2,-tetrachloro- ethane 87 and chloroform 111 are used as solvents, the distribution coefficient of Sr2+ reaches maxima for HNO3 concentrations of Acid Cation Concentration /mol litre71 Sr2+ Sr2+ Sr2+ Sr2+ Sr2+ Sr2+ Sr2+ Sr2+ Sr2+ Sr2+ Sr2+ Sr2+ Sr2+ Pu4+ Pu4+ Pu4+ Pu4+ Pu4+ Pu4+ Pu4+ Pu4+ Pu4+ Pu4+ U6+ U6+ U6+ U6+ Pu4+ U6+ Sr2+ Sr2+ Sr2+ Pu4+ 2114.05 31.5 2.0 2.0 3.0 1.0 0.01 3.0 3.0 3.0 3.0 3.0 3.0 3.0 3.0 3.0 3.0 5.0 4.0 0.04 0.04 5.0 7.2 4.5 4.5 4.05 118.0 0.04 0.04 7.0 10.0 2.0 U6+ U6+ U6+ U6+ Sr2+ Sr2+ U6+ U6+ U6+ U6+ HNO3 a HNO3 HNO3 HNO3 HNO3 HNO3 HNO3 HNO3 HNO3 HNO3 picric acid HNO3 HNO3 HNO3 HNO3 HNO3 HNO3 HNO3 HNO3 HNO3 HNO3 HNO3 HNO3 picric acid "HNO3 HNO3 HNO3 HNO3 HNO3 HNO3 HNO3 HNO3 picric acid "picric acid d HNO3 HNO3 HNO3 picric acid d HNO3 picric acid "HNO3 7.0 0.04 0.04 1 Cs+ 0.01 0.01 0.01 0.01 see e see e see e see e see e picric acid """see f see f see f see f Cs+ Cs+ Cs+ Cs+ Cs+ Cs+ Cs+ Cs+ Cs+ Cs+ Cs+ Cs+ Cs+ 777 D Ref.108 109 109 102 110 111 111 111 111 87 47 112 112 105 105 105 105 105 105 105 105 1068 113 113 114 107 86 86 102 109 1098 113 113 115 114 103 112 115 114 113 113 116 720 2.66 1.58 3.31 7.8 24.4 8.7 2.6 24.9 8.7 1000 62.3 45.7 64.28 48.41 18.04 17.93 17.78 14.84 6.92 0.72 13 >10 0.142 0.237 1.27 0.52 98 5.6 15.8 1.99 0.25 *10 0.028 0.533 0.092 0.16 0.63 0.035 0.116 1.62 0.204 0.146 0.1 117 117 117 117 117 118 118 118 118 93 93 93 93778 Table 3.Composition of complexes of crown ethers with metal cations. Ref. Crown ether Cation Ratio of components in the complex Sr2+ Pu4+ U4+ 103 102, 119 3, 105, 120 121 114 114 105 1 : 1 1 : 2 1 : 2 1 : 3 1 : 2 1 : 1 1 : 1 18-C-6 (20c) DCH-18-C-6 (12) DCH-18-C-6 (12) 15-C-5 (20b) B-15-C-5 (21b) 18-C-6 (20c) DCH-18-C-6 (12) 1.25 and 3.0 mol litre71, respectively.For the U6+ ion, the DU increases as the concentration of HNO3 is increased and reaches a plateau for its concentration of 5 ± 8 mol litre71 (Refs 105, 114). The dependence of DPu on solution acidity has a maximum which is shifted from 4 ± 5 to 3 ± 4 mol litre71 HNO3 as the macrocycle concentration is increased from 0.05 to 0.2 mol litre71 (see Refs 105, 120). Presumably,8, 105 the initial enhancement of the Pu(IV) extraction is due to salting out of the metal by nitric acid. Subsequent decrease in DPu as the acidity is increased was explained by the displacement of the metal from the organic phase due to the reaction of the ligand with the acid.8 However, an alternative explanation was also proposed:105 this effect is believed to be related to the formation of less extractable acidic complexes of the type HPu(NO3)5 and H2Pu4(NO3)6.The structure of crown ether, in particular the size of the polyether ring and the nature of the substituent, exert a strong effect on the efficiency of extraction. Relevant studies were devoted to the investigation of the influence of the crown ether structure on the extraction of cesium,118, 128 ± 130 strontium,47, 87, 103 uranium and plutonium8, 105, 131 and other actinides.8 In the extraction of ions of alkali and alkali-earth metals, crown ethers behave in conformity with the concept of geometric correspond- ence of the size of the cation to the size of the polyether ring cavity.Thus the optimum macrocycles for the binding of strontium are 18-C-6 (20c) and DCH-18-C-6 (12), whereas dicyclohexano- and dibenzo-21-crown-7 and -24-crown-8 are selective ligands for cesium cations.132, 133 The effect of the macrocycle size on the extractability of uranium and plutonium ions is of a more complex character. The extraction ability of crown ethers in toluene 105 or chloroform 114 in regard of UO2(NO3)2 and Pu(NO3)4 in a nitric acid medium proved to be changed in parallel with the macrocycle size and had the following order: DCH-18-C-6 (12)>DB-24-C-8 (17)>DB-18-C-6 (15) B-15-C-5 (21b). However, these sequences for U6+ and Pu4+ ions do not coincide in the case of extraction with solutions of crown ethers in dichloroethane from nitric acid media.The extractability of U6+ decreases in the order: DCH-18-C-6 (12)>DCH-24-C-8 (14)>DB-24-C-8 (17)>DB-18-C-6 (15)>15-C-5 (20b)> 18-C-6 (20c), while that of Pu4+ decreases in the order: DCH-24- C-8 (14)&DCH-18-C-6 (12)>15-C-5 (20b)>18-C-6 (20c)>DB-24-C-8 (17)>DB-18-C-6 (15).3 In the extraction of uranyl ions from solutions in picric acid, 2 the increase in the size of the macrocycle cavity leads to a decrease in DUO2á in the order: B-15-C-5 (20b) 18-C-6 (20c)>DB-18-C-6 (15) DB-24-C-8 (17).115 Neutral inorganic complexes of uranium and plutonium were shown 134, 135 to be linked to the macrocycle by means of hydrogen bonds through coordinated water molecules.These characteristic features of the structures of uranium and plutonium complexes could explain why the principle of geometric correspondence does not operate in this case. b. Methods for increasing extraction efficiency and non- conventional extraction systems The major routes to increasing the distribution coefficients in the extraction systems based on crown ethers are as follows: introduc- tion of an additional extractant in order to achieve an extraction S V Nesterov synergism effect, addition of a complexation agent (EDTA, acetylacetone, etc.) to the aqueous phase, addition of a polar water-miscible solvent (acetonitrile, propylene carbonate, dioxane, etc.) to the aqueous phase, extraction by crown ethers dissolved in liquid CO2, the use of biphasic aqueous systems for extraction.The synergism of extraction is understood as a joint action of two extractants, which results in the distribution coefficients exceeding those for each extractant used separately. Systems involving crown ethers were developed and the synergetic effects in the extraction of ions of cobalt,136 manganese,137 ± 139 palla- dium,140 uranium(VI),141 lanthanides and actinides,142 ± 149 stron- tium and barium,150 ± 153 cesium 25 and radium 154, 155 were studied. Unfortunately, in the studies of synergism the major attention has been paid to quantitative parameters of the extrac- tion process (calculation of extraction constants and distribution coefficients), while the reasons for this effect and its mechanism have rarely been discussed.An exception are the studies by McDowell et al.128 and Meguro et al.148 who investigated the influence of the macrocycle structure on extraction synergism. In the extraction of lanthanides and actinides, which have high coordination numbers, the reason for the synergism seems to consist in the replacement of water molecules of coordination by the first extractant, which facilitates the transfer of the adduct with the second extractant to the organic phase. In a number of cases, it was possible to increase substantially the distribution coefficients by adding a polar water-miscible organic solvent to the extraction system.It was shown 106, 107, 156 that organic solvents less polar than water (acetonitrile, propylene carbonate) favour formation of neutral complexes, e.g., of UO2(NO3)2 or Pu(NO3)4, with crown ethers, which enhances strongly the extraction efficiency. Rather recently,157 liquid CO2 has been proposed as a solvent for crown ether-assisted extraction. The main advantage of this method consists in a substantial decrease in the amount of liquid waste containing as a rule toxic organic solvents. Furthermore, the high diffusive ability and low viscosity of liquid CO2 allows direct extraction of dissolved substances from the solid phase (e.g., from samples of ground, soil, sand, etc.). In his review, Wai considered the effect of solubility of ligands and metal complexes in CO2, pH of the aqueous phase, temperature and pressure, the chemical nature of metal compounds and other factors on the extraction efficiency.157 An interesting feature of this extraction technique consists in the possibility of separation of metal compounds in organic and inorganic forms by selecting appropriate ligands.2 2 Biphasic aqueous extraction systems represent mixtures of water-soluble polymers, inorganic salts and extractants which separate and form immiscible phases.158 ± 160 The advantages of these systems reside in the use of inexpensive, non-toxic and easily accessible water-soluble polymers and in the possibility of using novel water-soluble complexants, which is particularly attractive in the case of dicyclohexano- and non-substituted crown ethers.The application of water-soluble 15-C-5 (20b) and 18-C-6 (20c) as extractants for cations of metals of groups I and II, UO2á 2 , Pu4+, Th4+ and Am3+ in a biphasic aqueous system containing poly(ethylene glycol) (PEG-2000) and (NH4)2SO4 has been described.135, 160 The distribution coefficient of UO2á (DUO2á) exceeded unity at a rather high concentration of 18-C-6 (20c) (1.25 mol litre71).135 2. The use of crown ethers in radiochemical analysis a. Determination of the content of radionuclides in water, soil and biological materials Isotopes of uranium and plutonium as well as their fission products, viz., the radionuclides 90Sr and 137Cs, are extremely toxic. In this connection, the determination of concentrations of these radionuclides in various natural specimens, which can find their way into human and animal organisms, is of great impor- tance.This analysis is difficult to perform due to the presence of a great amount of interfering cations (Na+,K+,Mg2+,Ca2+, etc.).Crown ethers in radiochemistry. Advances and prospects From the biological viewpoint, the isotope 90Sr is one of the most hazardous fission products, since it can replace calcium in bones and has a long half-life time. Numerous procedures have been developed to date for the determination of the content of 90Sr in water, soil and biological samples. These include precipita- tion,161 liquid extraction,162 ion exchange,163 ± 165 thin-layer chro- matography 166 and other methods.All these techniques have serious disadvantages, which prevents their wide uses. Thus in the case of inadvertent release of radionuclides resulting fromaccidents at an atomic energy power plant or from those related to tests and production on nuclear weapons, a fast, simple and efficient method is required. However, most of the above-mentioned methods of analysis are lengthy.167 The use of macrocyclic compounds allows considerable simplification of the analytical procedure. Using the sorbent Sr-Spec, Horwitz et al.21, 22 succeeded in developing a number of competitive methods for the determina- tion of strontium against the background of a large amount of cations which are not biologically hazardous. Gjeci 168 proposed an extraction-chromatographic method for the determination of the content of radioactive strontium in soil, grasses, milk and animal bones.A sample under analysis is subjected to extraction, filtered and then strontium is precipitated from the filtrate as oxalate, the precipitate is dissolved in 3M HNO3 and the resulting solution is passed through a column with Sr-Spec. The strontium sorbed selectively is eluted and quantitated. Vajda et al.169 reported a method which allows determination of up to 80% of radioactive strontium in soil. This method consists in the separa- tion of potassium, sorption of strontium on Sr-Spec and subse- quent precipitation of strontium as oxalate. The Sr-Spec sorbent was also used 170 for rapid quantitation of 90Sr in radioactive waste solutions in nitric acid in the presence of radionuclides of Am, Cd, Co, Ce, Sn, Cs and Y with separation factors >99%.A contin- uous automated method for the analysis of 90Sr (see Ref. 171) and a procedure for the determination of 90Sr in water and sedimentary rocks based on the sorbent Sr-Spec were proposed.172 In this case, 90Y(90Sr) is determined by liquid scintillation after separation of strontium on a column packed with Sr-Spec.172 The Sr-sorption abilities of two DCH-18-C-6 (12)-based solvents prepared by different techniques were compared and a new method for the quantitation of strontium in natural waters was proposed.173 One of them, viz., SESys, was obtained by introducing a crown ether into a matrix of a styrene ± divinylben- zene copolymer during polymer isolation, while the other sorbent (impregnate) was prepared by impregnating a macroporous styrene ± divinylbenzene copolymer with the same crown ether. Higher values of the distribution coefficient DSr were obtained on this porous polymeric sorbent.Extraction-chromatographic and extraction methods were proposed 45, 174, 175 for rapid quantitation of radioactive strontium in water, soil and agricultural products. An impregnate containing a 10% solution of DCH-18-C-6 (12) was used as the sorbent,45 while the extraction system was composed of 10% solution of the crown ether 12 in chloroform.174, 175 The analytical procedure involved drying of the sample, its calcination at 500 ± 600 8C for several hours and subsequent treatment of the ash with HNO3 or HCl.Then the solution obtained was passed through a sorption column for binding strontium or recourse was made to extraction, strontium was isolated and its b-activity was measured. In this method, analysis of a sample after its pretreatment required no more than 30 ± 50 min, the chemical yield was >95% with the product purity>98%.45 A rapid method for the quantitation of Sr in milk was developed.176 Milk was treated with a chelate-forming resin, Sr was eluted with a dilute nitric acid, separated from Ca by extraction of the eluate with DCH-18-C-6 (12) in chloroform with subsequent transfer of Sr to the aqueous solution and removal of Ba by extraction with a solution of 21-C-7 in dichloro- methane.This analytical procedure takes about 3 h and more than 90% of strontium is determined. 779 Separation and concentration of a number of radionuclides based on co-precipitation of metal ions with silicotungstic acid (H4SiW12O40 . 2H2O) from 1M HNO3 in the presence of 15-C-5 (20b) and 18-C-6 (20c) were carried out.177 ± 180 The selectivity of separation of radionuclides depends on the ratio of the stability constants of crown-ether complexes with the metal cations present in the solution. Precipitation was used to separate and concentrate Sr in the presence of Ca;179 Pb in the presence of Ca, Sr and Ba,178 Ra in the presence of Ca and Sr.177 This method allows estimation of the stability constant of a radionuclide complex with a crown ether even at a very low radionuclide content in solution.177 A method was devised 181 for the determination of the content of alkali and alkaline-earth metals in natural water using ion- exchange chromatography with a conductometric detector.Non- modified silica gel was used as the ion-exchange material, and the eluent contained oxalic acid (0.001 mol litre71) and 18-C-6 (20c) (0.003 mol litre71). Miura et al.182 proposed a method for the determination of 210Pb and 210Po in sedimentary rocks, clay and biological speci- mens. The analytical procedure includes four stages: sample decomposition, co-precipitation of 210Pb and 210Po with CuS, separation of isotopes on a column with Sr-Spec and quantitation of 210Pb and 210Po.Calix[4]arene-crown-6 and calix[4]arene-crown-5 containing two allyl groups were synthesised and grafted onto silica gel.92 The macrocycle/silica gel ratio was 1 : 10 (w/w). These sorbents were successfully used for the chromatographic separation of Cs+from cations of other alkali metals. Mohite et al.183 proposed a chromatographic method for the determination of barium in minerals. A copolymer of DB-18-C-6 (15) with formaldehyde was used as the sorbent. The absorbed barium was eluted with 1M acetic acid and its content was determined spectrophotometrically with Sulfonazo III. The capacity for barium was equal to 1.475 mmol g71. This method allows separation of barium from lead, thorium, uranium, cesium, cerium and molybdenum.The procedure is reproducible with an error of 2%. The last-mentioned authors 126, 127 have also described the application of extraction and column chromatography for the quantitation of uranium in geological specimens and animal bones. Dibenzocrown ethers were used as the extractants, while the sorbent was a copolymer of DB-18-C-6 (15) with formalde- hyde with a sorption capacity of 2.5 mmol g71. Both methods are distinguished by their simplicity, rapidity and good reproducibil- ity and allow separation of uranium from nuclear fission products. b. Separation of isotopes In recent years, considerable progress has been observed in the separation of metal isotopes using extraction and sorption systems based on crown ethers.A monograph by Hiraoka 184 published about 15 years ago cited only one report describing a method of separation of calcium isotopes with the aid of crown ethers. A review by Tsivadze et al.185 published in 1996 generalises the results of over 40 similar studies and presents examples of the use of crown-containing systems for the separation of isotopes of hydrogen, lithium, sodium, potassium, strontium, barium, zinc, cerium and uranium. Kim et al.186 reported recently the applica- tion of a polymeric resin based on an azacrown ether for the separation of lithium isotopes. Methods are known for the extraction-based separation of potassium isotopes with the aid of DCH-18-C-6 (12) 187 and chromatographic enrichment of calcium isotopes using 18-C-6 (20c) grafted onto a macroporous solid support.188, 189 An anomalous isotope effect for the isotopes with even and odd mass numbers was observed in the extraction of isotopes of zinc,190, 191magnesium,192 samarium 193 and stron- tium 194 with crown ethers. The development of methods for the isolation of a daughter isotope 90Y from the parent isotope 90Sr is an important and interesting direction. The isotope 90Y has found wide uses in biology and radiotherapy owing to its short half-life time (64 h)780 and stability of its daughter isotope 90Zr. From the viewpoint of clinical application of 90Y, its purification from 90Sr is particularly important.It was found that this problem may successfully be solved using certain crown ethers.Thus 90Y with the purity >99.9% was obtained by liquid extraction based on benzocrown ethers having a carboxy substituent in their polyether ring.195 Chuang and Lo 196 studied the extraction of strontium and yttrium cations by solutions of DCH-18-C-6 (12), 18-C-6 (20c), B-15-C-5 (21b) and DB-18-C-6 (15) in chloroform and showed that DCH-18-C-6 (12) is the most efficient in the separation of these isotopes. This method of separation is remarkable for its simplicity. The separation proceeds with quantitative yields and the purity achieved is above 99.9%. The isotopes 90Y and 90Sr were separated chromatographically on a paper impregnated with carboxy-substituted benzocrown ethers 197 and on a column packed with a dibenzocrown ether ± formaldehyde copolymer.198 A method of extraction-chromatographic separation of 90Y from 90Sr by sorption with DCH-18-C-6 (12) immobilised on silica was proposed.42 Quantitation of strontium and separation of its daughter isotope Y3+ by thin-layer chromatography on SiO2 with DCH-18-C-6 (12) applied on it was described.199 V.Conclusion Crown ethers possess high selectivity and complexation efficiency, resistance to irradiation and a number of other advantages. The progress in the chemistry of macrocyclic polyethers made it possible to develop methods of quantitation of highly toxic radionuclides in various mineral and biological specimens, as well as a number of promising processes for the reprocessing of spent nuclear fuel and utilisation of radioactive waste.A new direction of the use of crown ethers in radiochemistry is the extraction with the aid of liquid CO2. Upon irradiation of both free crown ethers and extraction systems based on them, the main process of radiolysis is the macrocycle scission; however, the radiation-chemical yield of the ring opening is insignificant at absorbed doses corresponding to real conditions. However, the studies of radiation chemistry of crown ethers and the development of methods for their immobilisation on supports face some unsolved problems. Thus details of the mechanism of radiolysis of crown ethers remain virtually obscure, there is no information about the structure and transformations of primary products. The radiation chemistry of calixarene-crown ethers is virtually uninvestigated. Definite advances have been achieved in the synthesis of impregnates based on dicyclohexanocrown ethers.A number of procedures for the determination of strontium, lead, polonium and other elements have been developed based on the sorbent Sr-Spec. Unfortunately, the high solubility of non-substituted crown ethers in water apparently will not allow creation of impregnates based on them, unless the polyether ring of crown ethers is modified by hydrophobic groups. The methods of chemical immobilisation of dicyclohexano- and non-substituted crown ethers developed to date are mostly based on the conden- sation of the macrocycles containing reactive groups. These methods are complicated, include many steps and are time- consuming; therefore, their perfection is needed.In this context, of high interest might be the use of methods of radiation-chemical synthesis, which would allow selective single-step formation of active centres and their addition to the reactive groups of a support or a monomer. References 1. W S Fyfe Eng. Geol. 52 159 (1999) 2. B F Myasoedov, E G Drozhko J. Alloys Compd. 271 ± 273 216 (1998) S V Nesterov 3. V V Yakshin, E A Filippov, V A Belov, G G Arkhipova, V M Abashkin, B N Laskorin Dokl. Akad. Nauk SSSR 241 159 4. R M Izatt, J S Bradshaw, S A Nielsen, J D Lamb, J J Christensen, 5. R M Izatt, K Pawlak, J S Bradshaw, R L Bruening Chem. Rev. 91 (1978) a D Sen Chem. Rev. 85 271 (1985) 1721 (1991) 6.A A Gakh, B G Sumpter, D W Noid, R A Sachleben, B A Moyer J. Incl. Phen. Mol. Recogn. Chem. 27 201 (1997) 7. W J McDowell Sep. Sci. Technol. 23 1251 (1988) 8. A M Rozen, Z I Nikolotova, N A Kartasheva, N G Luk'yanenko, A V Bogatskii Dokl. Akad. Nauk SSSR 263 1165 (1982) a 9. B S Mohite, C D Jadage, S R Pratar Analyst 115 1367 (1990) 10. K Li, L P Ferrari, C Brown, H D H Stover Macromol. Chem. Phys. 11. A Favre-Reguillon, N Dumont, B Dunjic, M Lemaire Tetrahedron 12. E I Grigor'ev, S V Nesterov, L I Trakhtenberg Zh. Fiz. Khim. 69 13. V A Barabanov, S L Davydova Vysokomol. Soedin., Ser. A 24 899 14. Yu A Zolotov, N M Kuz'min Makrotsiklicheskie Soedineniya v 195 391 (1994) 53 1343 (1997) 1735 (1995) b (1982) c Analiticheskoi Khimii (Macrocyclic Compounds in Analytical Chemistry) (Moscow: Nauka, 1993) 15. U Tunca, Y Yagci Progr.Polym. Sci. 19 233 (1994) 16. V Le Berre, L Angely, N Simonet-Gueguen, J Simonet J. Electroanal. Chem. 206 115 (1986) 17. J Simonet, J Raultberthelot Progr. Solid State Chem. 21 1 (1991) 18. C Belloncle, B Fabre, P Cauliez, J Simonet Synth. Met. 93 115 (1998) 19. B Fabre, J Simonet Coord. Chem. Rev. 178 ± 180 1211 (1998) 20. G Zirnhelt,M J F Leroy, J P Brunette, Y Frere, Ph Gramain Sep. Sci. Technol. 28 2419 (1993) 21. E Ph Horwitz, M L Dietz, R Chiarizia J. Radioanal. Nucl. Chem., Artic. 161 575 (1992) 22. E P Horwitz, M L Dietz, D E Fisher Anal. Chem. 63 522 (1991) 23. V V Yakshin, N A Tsarenko, N G Zhukova, B N Laskorin Dokl. Akad.Nauk 326 650 (1992) a 24. USSR P. 1 689 377; Byull. Izobret. (41) 87 (1991) 25. W J McDowell, G N Case, J A McDonough, R A Bartsch Anal. Chem. 64 3013 (1992) 26. S V Nesterov, S KuÈ cuÈ kyavuz, A M OÈ nal J. Radioanal. Nucl. Chem. 240 953 (1999) 27. S V Nesterov, S KuÈ cuÈ kyavuz, A M OÈ nal J. Radioanal. Nucl. Chem. 242 91 (1999) 28. B P S Chauhan, P Boudjouk Tetrahedron Lett. 40 4123 (1999) 29. V V Yakshin, N A Tsarenko, N G Zhukova, B N Laskorin Dokl. Akad. Nauk 325 1190 (1992) a 30. V V Yakshin, N A Tsarenko, N G Zhukova, B N Laskorin Dokl. Akad. Nauk 329 59 (1993) a 31. A G Dedov, L E Tskhai, E A Karakhanov, V I Kal'chenko, O A Aleksyuk, L N Markovskii Vestn. Mosk. Univ., Ser. 2, Khim. 33 171 (1992) d 32. M Kurmanaliev, E E Ergozhin, A A Kalmuratova, F B Bizhanov Dokl.Akad. Nauk 323 1109 (1992) a 33. H L Watanabe, T Iijima,W Fukuda,M Tomoi React. Funct. Polym. 37 101 (1998) 34. H Gibson, P T Engen, S-H Lee Polymer 40 1823 (1999) 35. V V Yakshin, A R Korigodskii, OMVilkova, V V Kireev, N N Sanina Koord. Khim. 23 477 (1997) e 36. V V Yakshin, N A Tsarenko, N G Zhukova, B N Laskorin Dokl. Akad. Nauk 325 748 (1992) a 37. V V Yakshin, O M Vilkova, N A Tsarenko, B N Laskorin Dokl. Akad. Nauk 324 144 (1992) a 38. V V Yakshin, N A Tsarenko, B N Laskorin Dokl. Akad. Nauk 325 970 (1992) a 39. V V Yakshin, O M Vilkova, N A Tsarenko Zh. Analit. Khim. 48 184 (1993) f 40. V V Yakshin, O M Vilkova, N A Tsarenko, B N Laskorin Dokl. Akad. Nauk 323 334 (1992) a 41.V V Yakshin, O M Vilkova, N A Tsarenko, N G Zhukova, B N Laskorin Koord. Khim. 19 492 (1993) e 42. J T Chuang, J G Lo J. Radioanal. Nucl. Chem., Artic. 204 83 (1996) 43. E P Horwitz, R Chiarizia, ML Dietz Solv. Extr. Ion Exch. 10 313 (1992)Crown ethers in radiochemistry. Advances and prospects 44. R Chiarizia, E P Horwitz,M L Dietz Solv. Extr. Ion Exch., 10 337 (1992) 45. N Yu Kremlyakova J. Radioanal. Nucl. Chem., Artic. 170 189 (1993) 46. A Yu Nazarenko, S A Zaitsev, V V Sukhan Izv. Vyssh. Uchebn. Zaved., Khim. Khim. Tekhnol. 36(5) 38 (1993) 47. L Matel, T Bilbao J. Radioanal. Nucl. Chem., Lett. 137 183 (1989) 48. V M Abashkin, D W Wester, J A Campbell, K E Grant Radiat. Phys. Chem. 48 463 (1996) 49. F VoÈ gtle, E Weber (Eds) Host Guest Complex Chemistry Macrocycles (Berlin: Springer, 1985) 50.M Draye, A Favre-Reguillon, J Foos, A Guy, M Lemaire Radiochim. Acta 78 105 (1997) 51. M Draye, A Favre-Reguillon, R Chomel, R Faure, A Guy, J Foos, M Lemaire Bull. Soc. Chem. Fr. 133 183 (1996) 52. T G Myasoedova, A V Ponomarev, P A Zagorets, E A Filippov Khim. Vys. Energ. 18 110 (1984) g 53. T G Myasoedova, P A Zagorets Khim. Vys. Energ. 20 436 (1986) g 54. E J Langrock, H Langguth, B Plisch ZFI-Mitte 98 548 (1984) 55. H Langguth, E J Langrock Isotopenpraxis 21 211 (1985) 56. S V Zatonsky, O V Povolotskaya J. Radioanal. Nucl. Chem., Lett. 117 361 (1987) 57. J H Baxendale,M A J Rodgers Trans. Faraday Soc. 63 2004 (1967) 58. E I Grigor'ev, T G Myasoedova, S V Nesterov, L I Trakhtenberg Khim.Vys. Energ. 21 437 (1987) g 59. E I Grigor'ev, S V Nesterov Khim. Vys. Energ. 26 483 (1992) g 60. B G Ershov, A F Seliverstov, N L Sukhov, A V Gordeev Khim. Vys. Energ. 21 119 (1987) g 61. A F Seliverstov, B G Ershov Khim. Vys. Energ. 22 203 (1988) g 62. V V Makhlyarchuk, S V Zatonskii Usp. Khim. 61 883 (1992) [Russ. Chem. Rev. 61 484 (1992)] 63. J Kuruc, F Sersen J. Radioanal. Nucl. Chem., Lett. 145 197 (1990) 64. J Kuruc, V E Zubarev, L T Bugaenko Chem. Listy 91 775 (1997) 65. J Kuruc J. Radioanal. Nucl. Chem., Lett. 145 205 (1990) 66. E I Grigor'ev, S V Nesterov, P S Vorontsov, O V Mikhalitsyna, L I Trakhtenberg Radiat. Phys. Chem. 46 553 (1995) 67. E I Grigor'ev, O V Mikhalitsyna, S V Nesterov, L I Trakhtenberg Mendeleev Commun.66 (1994) 68. E I Grigor'ev, O V Mikhalitsyna, S V Nesterov, L I Trakhtenberg Khim. Vys. Energ. 31 18 (1997) g 69. O V Mikhalitsyna, S V Nesterov, in Proceedings of `RadTech Asia'95' Conference, Guiline, 1995 p. 466 70. E I Grigor'ev, S V Nesterov, O V Mikhalitsyna, L I Trakhtenberg, T G Myasoedova Khim. Vys. Energ. 26 311 (1992) g 71. S V Nesterov, O Yu Panov J. Radioanal. Nucl. Chem. 231 87 (1998) 72. S V Nesterov J. Radioanal. Nucl. Chem., Lett. 230 273 (1998) 73. S KuÈ cuÈ kyavuz, S V Nesterov, A M OÈ nal J. Radioanal. Nucl. Chem. 230 39 (1998) 74. V K Milinchuk, V I Tupikov Radiatsionnaya Stoikost' Organiche- skikh Materialov. Spravochnik (Radiation Strength of Organic Materials. Handbook) (Moscow: Energoatomizdat, 1986) p.40 75. M Lemaire, A Guy, R Chomel, J Foos J. Chem. Soc., Chem. Commun., 1152 (1991) 76. J P Shukla, A Kumar, R K Singh J. Nucl. Sci. Technol. 31 1066 (1994) 77. V Vijayavergiya, A Mookerjee J. Radioanal. Nucl. Chem., Artic. 134 39 (1989) 78. L Matel J. Radioanal. Nucl. Chem., Lett. 146 85 (1990) 79. T G Myasoedova, G V Dmitrieva Khim. Vys. Energ. 25 244 (1991) g 80. M Draye, R Chomel, P Doutreluingne, A Guy, J Foos,M Lemaire J. Radioanal. Nucl. Chem., Lett. 175 55 (1993) 81. B A Moyer,R A Leonard, C Conner,M Liberatore, P V Bonnesen, D J Presley, Bruce A Moyer, G J Lumetta Sep. Sci. Technol. 34 1043 (1999) 82. P V Bonnesen, D J Presley, B A Moyer, in Proceedings of the International Solvent Extraction Conference `ISEC'96' Melbourne, 1996 (Eds D C Shallcross, R Paimin, L .M Prvcic) p.299 83. P V Bonnesen, B A Moyer, T J Haverlock, V S Armstrong, R A Sachleben, in Emerging Technologies in Hazardous Waste Management VI (Eds DW Tedder, F G Pohland) (Annapolis: American Academy of Environmental Engineers, 1994) p. 245 84. J D Law, D J Wood, R S Herbst Sep. Sci. Technol. 32 223 (1997) 85. D J Wood, J D Law Sep. Sci. Technol. 32 241 (1997) 86. V Guyon, J Foos, A Guy, T Moutarde, R Chomel,M Draye, M Lemaire Sep. Sci. Technol. 30 1961 (1995) 781 87. E Blasius, W Klein, U SchoÈ n J. Radioanal. Nucl. Chem., Artic. 89 389 (1985) 88. W W Schulz, L A Bray Sep. Sci. Technol. 22 191 (1987) 89. N Reynier, J-F Dozol,MSaadioui, Z Asfari, J Vicens Tetrahedron Lett.39 6461 (1998) 90. Z Asfari, V Lamare, J-F Dozol, J Vicens Tetrahedron Lett. 40 691 (1999) 91. L Nicod, S Pellet-Rostaing, F Chitry, M Lemaire Tetrahedron Lett. 39 9443 (1998) 92. G Arena, A Casnati, A Contino, L Mirone, D Sciotto, R Ungaro Chem. Commun. 2277 (1996) 93. T J Haverlock, P V Bonnesen, R A Sachleben, B A Moyer Radiochim. Acta 76 103 (1997) 94. R Ungaro, A Casnati, F Ugozzoli, A Pochini, J-F Dozol, C Hill, H Rouquette Angew. Chem., Int. Ed. Engl. 33 1506 (1994) 95. M Saadioui, Z Asfari, J Vicens, N Reynier, J F Dozol J. Incl. Phen. Molec. Recogn. Chem. 28 223 (1997) 96. Z Asfari, P Thuery, M Nierlich, J Vicens Tetrahedron Lett. 40 499 (1999) 97. Z Asfari, C Bressot, J Vicens, C Hill, J-F Dozol, H Rouquette, S Eymard, V Lamare, B Tournois Anal.Chem. 67 3133 (1995) 98. C Perez-Jimenes, L Escriche, J Casabo Anal. Chim. Acta 371 155 (1998) 99. C Hill, J F Dozol, V Lamare, H Rouquette, S Eymard, B Tournois, J Vicens, Z Asfari, C Bressot, R Ungaro, A Casnati J. Incl. Phen. Molec. Recogn. Chem. 19 399 (1994) 100. J S Kim, A Ohki, R Ueki, T Ishizuka, T Shimotashiro, S Maeda Talanta 48 705 (1999) 101. V Lamare, C Bressot, J F Dozol, J Vicens, Z Asfari, R Ungaro, A Casnati Sep. Sci. Technol. 32 175 (1997) 102. A Kumar, P K Mohapatra, P N Pathak, V K Manchanda Talanta 45 387 (1997) 103. F A Shehata J. Radioanal. Nucl. Chem., Artic. 185 411 (1994) 104. A Kumar, P K Mohapatra, V K Manchanda Radiochim. Acta 85 113 (1999) 105. J P Shukla, R K Singh, A Kumar Radiochim. Acta 54 73 (1991) 106.J P Shukla, A Kumar, R K Singh Radiochim. Acta 60 103 (1993) 107. J P Shukla, A Kumar, R K Singh Talanta 40 1261 (1993) 108. V V Yakshin, O M Vilkova, S A Kotlyar Dokl. Akad. Nauk 355 214 (1997) a 109. J F Dozol, J Casas, A M Sastre Sep. Sci. Technol. 29 1999 (1994) 110. TGMyasoedova,MYu Ikonnikov,A V Ponomarev, PA Zagorets, E A Filippov Radiokhimiya 29 180 (1987) h 111. V V Yakshin, B F Myasoedov, O M Vilkova, A M Tuzova, A T Fedorova, I M Rodionova Radiokhimiya 31(2) 67 (1989) h 112. V V Yakshin, O M Vilkova, B N Laskorin Dokl. Akad. Nauk 325 967 (1992) a 113. W Wenji, C Bozhong, J Zhong-Kao,W Ailing J. Radioanal. Chem. 76 49 (1983) 114. V V Yakshin, O M Vilkova Radiokhimiya 34 68 (1992) h 115. P K Mohapatra, V K Manchanda Talanta 47 1271 (1998) 116.J F Dozol, J Casas, A M Sastre Sep. Sci. Technol. 30 435 (1995) 117. R A Sachleben, Y Deng, B Moyer Sep. Sci. Technol. 32 275 (1997) 118. R G Vibhute, S M Khopkar J. Radioanal. Nucl. Chem., Artic. 152 487 (1991) 119. E A Filippov, V V Yakshin, V M Abashkin, V G Fomenkov, I S Serebryakov Radiokhimiya 24 214 (1982) h 120. J P Shukla, A Kumar, R K Singh Sep. Sci. Technol. 27 447 (1992) 121. A G Godbole, N V Thakur, R Swarup, S K Patil J. Radioanal. Nucl. Chem., Lett. 108 89 (1986) 122. B N Laskorin, V V Yakshin Zh. Vsesoyuz. Khim. O-va. im. D I Mendeleeva 30 579 (1985) j 123. V V Yakshin, O M Vilkova Koord. Khim. 22 260 (1996) e 124. V V Yakshin, V M Abashkin, N G Zhukova, N A Tsarenko, B N Laskorin Dokl.Akad. Nauk SSSR 247 1398 (1979) a 125. M L Dietz, A H Bond,M Clapper, J W Finch Radiochim. Acta 85 119 (1999) 126. B S Mohite, J M Patil, D N Zambare, U G Mali Bull. Chem. Soc. Jpn. 66 3639 (1993) 127. B S Mohite, J M Patil, D N Zambare J. Radioanal. Nucl. Chem., Artic. 170 215 (1993) 128. W J McDowell, B A Moyer, G N Case, F I Case Solv. Extr. Ion Exch. 4 217 (1986)782 129. M L Dietz, E P Horwitz,M P Jensen, S Rhoads, R A Bartsch, A Palka, J Krzykawski, J Nam Solv. Extr. Ion Exch. 14 357 (1996) 130. R A Sachleben, Y Deng, D R Bailey, B A Moyer Solv. Extr. Ion Exch. 14 995 (1996) 131. B N Laskorin, D I Skorovarov, E A Filippov, V V Yakshin Radiokhimiya 27 156 (1985) h 132. Y P Deng, R A Sachleben, B A Moyer J. Chem.Soc., Faraday Trans. 23 4215 (1995) 133. B A Moyer, Y P Deng, Y F Sun, R A Sachleben, A K Batra, R B Robinson Solv. Extr. Ion Exch. 15 791 (1997) 134. A Yu Tsivadze, A A Varnek, V E Khutorskii, in Koordinatsionnye Soedineniya Metallov s Kraun-Efirami (Coordination Compounds of Metals with Crown Ethers) (Moscow: Nauka, 1991) p. 81 135. R D Rogers, C B Bauer, A H Bond J. Alloys Compd. 213/214 305 (1994) 136. S I El-Dessouky, J A Daoud, H F Aly Radiochim. Acta 85 79 (1999) 137. B A Moyer, L H Delmau, G J Lumetta, C F Baes Solv. Extr. Ion Exch. 11 889 (1993) 138. G J Lumetta, B A Moyer, P A Johnson Solv. Extr. Ion Exch. 8 457 (1990) 139. R B Chadwick, W J McDowell, C F Baes Sep. Sci. Technol. 23 1311 (1988) 140. V Guyon, A Guy, J Foos, R Chomel, T Moutarde, G Lebuzyte, M Lemaire J.Radioanal. Nucl. Chem., Lett. 187 19 (1994) 141. P Thakur, V Chakravortty, K C Dash, T R Ramamohan, M L P Reddy Radiochim. Acta 80 155 (1998) 142. Y Meguro,W Cheng, H Imura, Z Yoshida J. Radioanal. Nucl. Chem., Artic. 160 435 (1992) 143. F A Shehata, S M Khalifa, H F Aly J. Radioanal. Nucl. Chem., Artic. 159 353 (1992) 144. V V Proyaev, V V Romanovskii, I G Maslennikov Radiochemistry 36 176 (1994) 145. A Ramadan,M Mahmoud, S M Khalifa, N Souka J. Radioanal. Nucl. Chem., Artic. 176 457 (1993) 146. M L P Reddy, R L Varma, T R Ramamohan Radiochim. Acta 80 151 (1998) 147. Y Meguro, Z Yoshida Radiochim. Acta 65 19 (1994) 148. Y Meguro, Y Kitatsuji, T Kimura, Z Yoshida J. Alloys Compd.271 ± 273 790 (1998) 149. D D Ensor, G R McDonald, C G Pippin Anal. Chem. 58 1814 (1986) 150. P Vanura, E Makrlik Coll. Chech. Chem. Commun. 63 338 (1998) 151. Z Valentova, P Vanura, E Makrlik J. Radioanal. Nucl. Chem. 224 45 (1997) 152. B A Moyer, S A Bryan, G N Case, W J McDowell, R J Ontko Solv. Extr. Ion Exch. 4 83 (1986) 153. S A Bryan, C F Baes, G N Case, W J McDowell, B A Moyer Solv. Extr. Ion Exch. 5 717 (1987) 154. R Chiarizia, E P Horwitz,M L Dietz, Y D Cheng React. Funct. Polym. 38 249 (1998) 155. W J McDowell, B A Arndsten, G N Case Solv. Extr. Ion Exch. 7 377 (1989) 156. M L Dietz, E P Horwitz, S Rhoads, R A Bartsch, J Krzykawski Solv. Extr. Ion Exch. 14 1 (1996) 157. C M Wai J. Chromatogr. 785 369 (1997) 158. B F Myasoedov, N P Molochnikova, V M Shkinev Solv.Extr. Ion Exch. 10 687 (1992) 159. RDRogers, AHBond, C B Bauer Sep. Sci. Technol. 28 1091 (1993) 160. R D Rogers, A H Bond, C B Bauer Pure Appl. Chem. 65 567 (1993) 161. Measurement of Radionuclides in Food and Environment. Guidebook (Technical Report Series No. 295) (Vienna: IAEA, 1989) p. 70 162. S V Krouglov, A D Kurinoy, R M Alexakhin J. Environ. Radioact. 38 (1) 59 (1998) 163. V V Milyutin, V M Gelis, R A Penzin Radiokhimiya 35 76 (1993) h 164. M Adabbo, D Caputo, B de Gennaro,M Pansini, C Colella 165. Z Granek, I Eskinja, K Kosutic, S Lulic, K Kvastek Anal. Chim. 166. E Gattavecchia,D Tonelli J. Radioanal. Nucl. Chem., Artic. 152 391 167. J D Lamb, R R Nordmeyer, P A Drake, M P Elder, R W Miles Micropor. Mesopor. Mater. 28 315 (1999) Acta 379 107 (1999) (1991) J. Radioanal. Nucl. Chem. 134 317 (1989) S V Nesterov 168. E Gjeci J. Radioanal. Nucl. Chem., Lett. 213 165 (1996) 169. N Vajda, A Ghods-Esphahani, E Cooper, P R Danesi J. Radioanal. Nucl. Chem., Artic. 162 307 (1992) 170. M Rodriguez, J A Suarez, A G Espartero Nucl. Instrum. Methods Phys. Res. A 369 348 (1996) 171. J W Grate, R Strebin, J Janata, O Egorov, J Ruzicka Anal. Chem. 68 333 (1996) 172. Z Grahek, N Zecevic, S Lulic Anal. Chim. Acta 379 237 (1999) 173. G B Maslova, V M Gelis, S I Ushakova, N I Polyakova Radiokhimiya 34 108 (1992) h 174. AMTuzova, OM Vilkova, N Yu Kremlyakova, IMRodionova, A T Fedorova, B F Myasoedov, V V Yakshin Radiokhimiya 35 125 (1993) h 175. M Pimpl J. Radioanal. Nucl. Chem., Artic. 194 311 (1995) 176. D Tait, A Wiechen J. Radioanal. Nucl. Chem., Artic. 159 239 (1992) 177. P Rajec, V Mikulaj, P Husarcik, V Svec J. Radioanal. Nucl. Chem., Lett. 219 123 (1997) 178. V Svec, V Mikulaj, R Hanzel J. Radioanal. Nucl. Chem., Artic. 208 487 (1996) 179. V Mikulaj, V Svec J. Radioanal. Nucl. Chem., Artic. 183 127 (1994) 180. P Rajec, V Svec, V Mikulaj, R Hanzel J. Radioanal. Nucl. Chem. 229 9 (1998) 181. K Ohta, K Tanaka Anal. Chim. Acta 381 265 (1999) 182. T Miura, K Hayano, K Nakayama Anal. Sci. 15 23 (1999) 183. B S Mohite, D N Zamare, B E Mahadik Analyst 119 2033 (1994) 184. MHiraoka (Ed.) Crown Compounds, Their Characteristics and Applications (Tokyo: Kodansha Ltd, 1982) 185. A Yu Tsivadze, V I Zhilov, S V Demin Koord. Khim. 22 243 (1996) e 186. D W Kim, B K Kim, S R Park, N S Lee, Y S Jeon, K Y Choi, Y I Lee J. Radioanal. Nucl. Chem. 232 257 (1998) 187. A s 1623738; Byul. izobret. (4) 26 (1991) 188. B E Jepson,M A Novotny,W F Evans Sep. Sci. Technol. 28 507 (1993) 189. B E Jepson,M R Clager, J L Green Pure Appl. Chem. 65 489 (1993) 190. K Nishizawa, T Miki, T Satoyama, T Fujii, T Yamamoto, M Nomura Sep. Sci. Technol. 33 991 (1998) 191. K Nishizawa, Y Maesono, T Yamamoto, T Fujii, T Hirata Sep. Sci. Technol. 33 2101 (1998) 192. K Nishizawa, T Nishida, T Miki, T Yamamoto,M Hosoe Sep. Sci. Technol. 31 643 (1996) 193. T Fujii, T Yamamoto, K Nishizawa, J Inagawa, K Gunji, K Watanabe Solv. Extr. Ion Exch. 16 985 (1998) 194. K Nishizawa, T Satoyama, T Miki, T Yamamoto, M Hosoe J. Nucl. Sci. Technol. 32 1230 (1995) 195. D J Wood, S Elshani,H S Du,N R Natale,C M Wai Anal. Chem. 65 1350 (1993) 196. J T Chuang, J G Lo J. Radioanal. Nucl. Chem., Artic. 189 307 (1995) 197. C M Wai, H S Du Anal. Chem. 62 2412 (1990) 198. D J Wood, S Elshani, C M Wai, R A Bartsch, M Huntley, S Hartenstein Anal. Chim. Acta 284 37 (1993) 199. V A Ryabukhin, M P Volynets, B F Myasoedov Radiokhimiya 34 88 (1992) h a�Dokl. Chem. Technol., Dokl. Chem. (Engl. Transl.) b�Russ. J. Phys. Chem. (Engl. Transl.) c�Polym. Sci. (Engl. Transl.) d�Moscow Univ. Bull. (Engl. Transl.) e�Russ. J. Coord. Chem. (Engl. Transl.) f�Russ. J. Anal. Chem. (Engl. Transl.) g�High Energ. Chem. (Engl. Transl.) h�Radiochemistry (Engl. Transl.) j�Mendeleev Chem.

ISSN:0036-021X    

出版商:RSC    

年代:2000

数据来源: RSC

摘要:

Russian Chemical Reviews 69 (9) 783 ± 794 (2000) Arene complexes of rare-earth metals MN Bochkarev Contents I. Introduction II. Complexes of rare-earth metals with derivatives of benzene and its homologues III. Complexes with naphthalene derivatives IV. Complexes with anthracene V. Pyrene and benzoanthracene derivatives VI. Complexes with heterocyclic aromatic ligands VII. Reactivity of arene lanthanide complexes VIII. Conclusion Abstract. scan- of reactivities and structures syntheses, on Data Data on syntheses, structures and reactivities of scan- dium, complexes lanthanide and lanthanum yttrium, dium, yttrium, lanthanum and lanthanide complexes with with p aro- six-membered several or one containing ligands -bonded -bonded ligands containing one or several six-membered aro- matic rings are surveyed and systematised. The bibliography matic rings are surveyed and systematised.The bibliography includes references 107 includes 107 references. I. Introduction Early investigations in the field of chemistry of organometallic compounds containing rare-earth metals were aimed at synthesis- ing and studying compounds typical of organometallic chemistry, such as s-alkyl(aryl), cyclopentadienyl, indenyl, fluorenyl, cyclo- octadienyl, etc. derivatives. Complexes containing p-bonded arene ligands were studied later.1 However, the chemistry of these compounds was extensively investigated in the last 12 years after the first benzene derivatives of lanthanides had been synthe- sized. The accumulated data allow some general conclusions to be drawn about the characteristic features of the synthesis, struc- tures, the character of metal ± ligand bonding and reactivity of p-arene lanthanide complexes belonging to a class of compounds unique in many respects.Particular types of p-arene lanthanide complexes were surveyed in the reviews.2, 3 The present review is devoted to comprehensive and critical consideration of the available data on the synthesis, structures, the character of bonding, reactivity and catalytic activity of rare-earth metal complexes containing p-bonded six-membered aromatic rings, which were published before January 2000. Derivatives of stil- bene, tolan and analogous compounds fall outside the scope of this review because p-electrons of multiple bonds of substituents in aromatic rings of these compounds are generally involved in Ln ligand interactions.MN Bochkarev G A Razuvaev Institute of Organometallic Chemistry, Russian Academy of Sciences, ul. Tropinina 49, 603600 Nizhnii Novgorod, Russian Federation. Fax (7-831) 266 14 97. Tel. (7-831) 266 67 95. E-mail: mboch@imoc.sinn.ru Received 26 April 2000 Uspekhi Khimii 69 (9) 856 ± 868 (2000); translated by T N Safonova #2000 Russian Academy of Sciences and Turpion Ltd DOI 10.1070/RC2000v069n09ABEH000601 783 783 785 788 789 790 791 793 II. Complexes of rare-earth metals with derivatives of benzene and its homologues 1. Arene chloroaluminate complexes The chloroaluminate complex of samarium, (C6Me6)Sm(AlCl4)3, was the first complex of a rare-earth metal with a p-bonded arene ligand synthesised under conditions of the Friedel ± Crafts reac- tion.4 ±6 This procedure for the synthesis was extended to other complexes of the type 1.The reactions were carried out with 1,3-Me2C6H4,7 C6H6,8± 10 Me4C6H2,11, 12 MeC6H5 (see Ref. 13) and a large series of lanthanides (Y, Pr, Nd, Gd,12 La,8 Eu and Yb 13). The compositions and structures of complexes of the type 1 indicate that the aromatic moiety is Z6-coordinated to the trivalent metal atom and serves as a neutral p-donor ligand. The average Ln7C distances [2.93 A (Nd),11 2.90 A (Sm),7 2.999 A (Eu) 14 and 2.978 A (Yb) 15] are noticeably shorter than the sums of the van der Waals radius of the carbon atom and the ionic radius of the corresponding metal atom, but are substantially longer than the Ln7C contacts in the cyclopentadienyl deriva- tives (2.60 ± 2.82 A for the whole series of lanthanides).Rn Ln Cl Cl Cl Cl AlCl2 Cl2Al Cl Cl Al Cl Cl 1 Ln=Y, Pr, Nd, Gd, La, Eu, Yb; R=H, Me; n=0, 1, 2, 4. The replacement of one of the bridging chlorine atoms by the alkyl group, for example, in the complexes (C6H5Me)Ln. .(AlCl3Me)3 (Ln=Y or Nd) and (C6H5Me)Nd(AlBr3Et)3,12 does not lead to a change in the character of the arene7metal bond. The divalent europium complex (C6Me6)Eu(AlCl4)2 has an analogous structure.14 2. Arene aryloxide complexes In aryloxides of lanthanides, there are nonvalent interactions between the metal atom and the benzene ring identical to those found in arene chloroaluminate complexes. The presence of the phenyl substituent at the ortho position of the aryloxy group [for784 example, in the complex Nd(OC6H3Ph2-2,6)3 (2)] results in an intramolecular Z6-interaction between the lanthanide atom and one of the phenyl substituents of the phenoxy ligand.The phenyl substituent in the second phenoxy ligand is Z1-bonded to the Ln atom, whereas the third phenoxy ligand remains monoden- tate.15 ± 17 In the analogous complex Nd(OC6H3Ph2-2,6)3(THF) (3), which differs by the presence of one THF molecule coordi- nated to the metal atom, only one OC6H3Ph2-2,6 substituent serves as a chelate ligand, the multiplicity of the Ln7Ph p-bond decreasing (becoming triple).16 THFOAr Ph Ph O ArO Nd O Nd O OAr Ph 2 3 Ar=2,6-Ph2C6H3 . In lanthanide triphenoxides containing the tert-butyl substitu- ents at positions 2 and 6, the metal atom is Z6-bonded to the phenoxide ligand to form the dimers (2,6-But2C6H3O)2Ln(m-2,6- But2C6H3O)2Ln(OC6H3But2-2,6)2 (4).18, 19 But O ArO La But OAr ArO But La O OAr 4 But Ar=2,6-But2C6H3 .The Ln7Arn contacts (Arn is arene) are retained upon dissolution of complexes of the type 4 in nonsolvating solvents. However, p-bonds are cleaved in THF or upon addition of ammonia and the coordination site of the Z6-arene ligand in the resulting complexes is occupied by solvent molecules, which is indicative of low strength of the Ln7Arn bond.The Ln7C bond lengths in these fragments (the average distances are 2.986 ± 3.061 A) are virtually identical to the corresponding values in arene chloroaluminate complexes. 3. Bisarene complexes Co-condensation of benzene and neodymium vapours on a sur- face cooled to 7196 8C afforded a binuclear complex with composition Nd2(C6H6)3,20 which was isolated as black insoluble crystals. The magnetic moment (3.42 mB) corresponds to the Nd(III) cation (3.68 mB 21) and the dianionic form of all three benzene ligands. The complex was not studied by X-ray diffrac- tion analysis, but it is reasonable to suppose that its structure is similar to that of the thulium naphthalene analogue [(C10H8)Tm(DME)]2(m-C10H8) considered below.Sandwich bisarene complexes of rare-earth metals of the (1,3,5-But3C6H3)2Ln type (5) analogous to the well-known bisar- ene complexes of d transition metals were prepared by coconden- But But But Ln But But5a ± p But Ln=Gd (a), Y (b), Ho (c), Nd (d), Tb (e), Dy (f ), Er (g), Lu (h), Ce (i), Eu ( j ), Yb (l), La (m), Pr (n), Sm (o), Sc (p). M N Bochkarev sation of 1,3,5-tri-tert-butylbenzene and metal vapours at 75 K.22 ± 25 The structures and magnetic moments of the com- plexes 5 confirm the Z6-character of the bond between the neutral aromatic ring and the formally zero-valent metal atom. In gadolinium complex 5a characterised by X-ray diffraction analysis, the Gd7C distances are in the range of 2.585 ± 2.660 A (the average distance is 2.630 A).These distances are virtually identical to the Ln7C bond lengths in cyclopentadienyl com- plexes of the Cp2GdBr type (2.635, 2.639 and 2.630 A26). The yttrium (5b) and holmium (5c) complexes were found to have analogous structures.3, 22 The stability of the compounds 5 depends on the nature of lanthanide. Thus the Yb, Nd, Gd, Tb, Dy, Ho, Er and Lu complexes are stable (some of them can be sublimed in vacuo without decomposition), whereas the Ce, Eu, Tm and Yb com- plexes do not form stable products. Derivatives of La, Pr and Sm are intermediate in stability. In the synthesis of the scandium complex, the formation of the expected sandwich (1,3,5- But3C6H3)2Sc (5p) was accompanied by the formation of the complex (Z6-But3C6H3)[Z6,Z1-But2(CMe2CH2)C6H3]ScH (6) con- taining Sc(II) due to the insertion of Sc at the C7H bond of the methyl group of the substituent.27 But But H But Sc But CH2 Me Me But 6 Since a strong arene7lanthanide p-bond can be formed if the metal atom adopts the d 1s2 configuration, the difference in thermal stability of the complexes 5 was attributed to the differ- ence in energy of promotion of f ns27f n71d 1s2 transitions for different lanthanides.Thermodynamic studies 25 confirmed the high stability of Ln7Arn bonds in bisarene complexes. Thus the enthalpies of cleavage of these bonds in the Yb, Gd, Dy, Ho and Er complexes are 72, 68, 47, 56 and 57 kcal mol71, respectively.25 The stability of bisarene complexes increases as the number and size of substituents in the aromatic ring increase as was exemplified by scandium derivatives.3 Calculations by molecular mechanics 28 for (C6H6)2Gd and (1,3,5-But3C6H3)2Ln (5) [Ln=Gd (a), Y (b) orYb (l)] confirmed that the benzene complex is substantially less stable than the complexes 5a,b,l containing the tert-butyl groups in the benzene ring.Quantum-chemical calcu- lations 29, 30 demonstrated that the metal atoms are bound to the benzene ligands primarily through back donation from the occupied d2 orbitals of the metal atom to the unoccupied p-orbitals of the benzene ligands. 4. Complexes with benzene radical anions and dianions Reduction of the cyclopentadienyl complex of lanthanum con- taining two tert-butyl substituents in the cyclopentadienyl rings, viz., Cptt 3 La, where Cptt=Z5-C5H3But2-1,3, with an excess of potassium in benzene in the presence of 18-crown-6 (18-C-6) afforded a dark-red product.Subsequent heating of the latter in benzene at 70 8C gave rise to the dark-green ate-complex [K(18-C-6)(Z2-C6H6)2][(Cptt 2 La)2(m-Z6 :Z6-C6H6)] (7).31 X-Ray diffraction analysis demonstrated that the bridging six-membered ring in the anionic moiety of the complex 7 is planar and the C7C bond lengths in this ring are almost identical. All La7C(benzene) bond lengths vary over a narrow range (2.75 ± 2.79 A). Based on these data, the structure of a salt containing two Cptt 2 La(II) units Z6-bonded to the benzene anion was assigned to the complex 7.Arene complexes of rare-earth metals But But O But But O O La La K But O O But O But But 7 An analogous reaction of potassium with the complexes Cpts 3 Ln [Cpts=C5H3(SiMe3)2-1,3; Ln=La, Ce, Pr or Nd] also afforded dark-red ionic complexes [K(18-C-6)][Cpts 2 Ln(C6H6)] (8a ± d).However, the cations in these complexes do not contain coordinated benzene molecules and the C6H6 ligand in the anionic moiety exists as the dianion of cyclohexa-1,4-diene Z6-bonded to the Cpts 2 Ln unit with the trivalent Ln atom.32, 33 In the complex 8a, two short La ± C(cyclohexadiene) distances are 2.617 and 2.652 A. In analogous complexes of cerium and neodymium 8b,c, the corresponding bond lengths are 2.588 and 2.612 A (Ce) and 2.555 and 2.572 A (Nd).SiMe3 SiMe3 [K(18-C-6)] Ln SiMe3 SiMe3 8a ± d Ln=La (a), C (b), Nd (c), Pr (d). ESR studies demonstrated 33 that reduction of Cpts 3 La with potassium in benzene in the presence of 18-C-6 afforded at least four metal-centred free-radical species. The authors believed that the latter are La(II) derivatives. In this stage, the reaction mixture was dark-blue-violet in colour, which then changed to dark-red. 5. Biphenyl complexes Treatment of the yttrium(III) chloride complex with composition LYCl, where L=PhP[CH2(SiMe2)N(SiMe2)CH2]2PPh, with phenyllithium gave rise to the dark-blue binuclear product ({PhP[CH2(SiMe2)N(SiMe2)CH2]2PPh}Y)2(m-PhPh) (9) { con- taining the biphenyl dianion as the bridging ligand.34 The complex 9 was also prepared by the reaction of the s-alkyl complex LYCH2SiMe3 {L=PhP[CH2(SiMe2)N(SiMe2)CH2]2PPh} with benzene.34 Ph Ph Si Si P Si Si P Si Si Si Si P P N N N N Ph Ph Y Y Me Me Y Ph Y Ph N N P Si N Si N P Si Si Si P Si Si P Si Ph Ph 10 9 In the complex 9, both rings in the biphenyl ligand remain virtually planar and the Y7C distances vary over a narrow range (2.675 ± 2.738 A), which are evidence of Z6-interaction.These { Hereinafter, the methyl groups at the silicon atoms are omitted. 785 distances (the average distance is 2.723 A) are close to the Y7C bond lengths in the complex Cp3Y(THF) (the average distance is 2.71 A) 35 and are somewhat larger than those in the complexes [Cp2YCl]2 (2.60 A) 36 and (C5Me5)2YCl(THF) (2.659 A).37 The reaction of LYCl with 3-MeC6H4Li also resulted in the formation of a C7C bond.According to the data of 1H NMR and UV spectroscopy, the dark-blue reaction product is structur- ally similar to the complex 9. However, 4-MeC6Me4Li under the same conditions gave the dark-brown compound ({PhP[CH2. .(SiMe2)N(SiMe2)CH2]2PPh}Y)2(m-MePhPhMe) (10) in which both YL fragments are bound to the same ring of the [MeC6H4C6H4Me]27 anion.34 The average Y7C distances in the fragments of 10 (2.676 and 2.699 A) are somewhat shorter than the corresponding bond lengths in the complex 9. 1H NMR spectral studies of the complex 10 demonstrated that slow (within theNMR time scale) migration of one of the YL groups from one benzene ring to another occurs in solution.It was assumed 34 that such behaviour in solutions is also typical of other complexes with bridging Ar7Ar ligands. Samarium and ytterbium powders activated with diiodo- ethane reacted with biphenylene at room temperature to form dark-coloured solutions.38 Attempts to isolate reaction products in the individual state failed. However, their reactions with oxygen (giving rise to 1,10-biphenyl-2,20-diol), deuteromethanol (yielding a mixture of biphenyl and dideuterobiphenyl) and Me2SiCl2 (forming a mixture of biphenyl and 1,1-dimethylsilafluorene) were carried out. These results testify that the reactions of lanthanides with biphenylene lead to the cleavage of the four- membered ring giving rise to the lanthanide s-complexes with biphenyl (11).It was suggested 38 that the first stage involved reduction of biphenylene to form the ionic intermediates C12H27 8 Ln2+, which abstracted hydrogen from the solvent to form biphenyl and the complex 11. Ln O O 11 Apparently, an analogous product was formed in the reaction of octamethylbiphenylenedilithium with SmI2.38 III. Complexes with naphthalene derivatives It was found that reduction of anhydrous ytterbium diiodide with two equivalents of lithium naphthalenide in THF or dimethoxy- ethane (DME) afforded insoluble black pyrophoric powders with composition (C10H8)Yb(THF)3 or (C10H8)Yb(DME).39 ± 44 Based on the facts that the resulting compounds are diamagnetic [Yb(0) is diamagnetic] and their reactivities are surprisingly high (even for organolanthanides), it was originally conceived that these prod- ucts are complexes of zero-valent ytterbium p-bonded to the neutral naphthalene ring.However, additional data on tetrahy- drofuran complex 12 obtained more recently demonstrated that these compounds in fact contain the naphthalene dianion and the Yb2+ cation. THF YbI2+C10H8Li (C10H2¡ 8 )Yb2+(THF)3 12 Analogous naphthalene complexes with samarium and euro- pium 43, 45 ± 48 as well as 1-methylnaphthalene complex with ytterbium 40, 48 were synthesised. The naphthalene complexes of samarium, europium and ytterbium, like alkali metal naphthale- nides, are powerful reducing agents. However, the former com- plexes in contrast to naphthalenides do not react with naphthalene [the reactions could form radical-anionic derivatives of the type (C10H¡8 )2Ln(THF)x].786 The reactions of equimolar amounts of LnI2 (Ln=Eu or Yb) and sodium naphthalenide resulted in the replacement of only one iodide anion and afforded binuclear derivatives 13 containing the bridging naphthalene dianion rather than radical-anionic com- plexes (C10H¡8 )LnI(THF)x .49, 50 DME LnI2+C10H8Na [ILn(DME)2]2(m-C10H8 ) 13 Ln=Eu, Yb.The Ln(m-C10H8)Ln moiety containing the naphthalene bridg- ing dianion (complex 14) was also formed in the reaction of sodium naphthalenide with lanthanum iodide.49 Analogous com- plexes were obtained for Ce, Pr, Nd and Gd.50 THF I2La(THF)3(m-C10H8)LaI2(THF)3 LaI3+C10H8Na 14 The stability of complexes of the type 14 decreases on going from light to heavy lanthanides.From Tb and onwards, attempts to isolate such complexes failed even at low temperature. The neodymium derivative 15 was synthesised by oxidation of divalent neodymium iodide with naphthalene in a solution in THF.51 THF NdI2(THF)5+C10H8 I2Nd(THF)3(m-C10H8)NdI2(THF)3 15 The naphthalene complexes of europium and lanthanum, viz., [I2Eu(DME)2)]2(m-C10H8) (16) and [I2La(THF)3)]2(m-C10H8) (17), respectively, were studied by X-ray diffraction analysis.49 In these complexes, the metal atoms are located on the opposite sides of the plane of the bridging naphthalene ligand and are Z4-bonded to different rings.The naphthalene moiety is nonplanar and the dihedral angles between the rings are 5.8 8 and 15.2 8 in the complexes 16 and 17, respectively. The shortest La7C and Eu7C distances are in the ranges of 2.782 ± 2.808 and 2.815 ± 2.858 A, respectively. O O O I THFI La THF I THF Eu OO THF La Eu I O THFI O O I THF17 16 With the aim of preparing halogen-free naphthalene com- plexes of lanthanides, the reactions of triiodides LnI3 with a threefold excess of C10H8Na were carried out.43, 52 After removal of the solvent, solid black or dark-brown amorphous substances were obtained from the resulting dark-brown solutions. Attempts to isolate individual products from these substances failed. The individual compound [C10H8Tm(DME))]2(m-C10H8) (18) was obtained only for thulium starting from the salt TmI2(DME)3 .53 TmI2(DME)3+C10H8Na DME O O Tm O O Tm 18 It is worthy of note that thulium was oxidised in the course of the reaction to the trivalent state in spite of the fact that C10H8Na M N Bochkarev is a strong reducing agent.The trivalent state of thulium was confirmed by magnetic measurements and by the fact that the colour of the reaction mixture changed from green to red-brown. The compound 18 has the structure of a triple-decker sandwich. According to the valence balance, all three naphthalene groups in the complex exist as dianions. The central binuclear unit with the Z4 :Z4-bonded bridging naphthalene ligand is structurally similar to those in the iodine-containing complexes 16 and 17.The lengths of the short Tm7C(bridging C10H8) bonds are in the range of 2.59 ± 2.62 A. The folding angle of the rings along the C(1)7C(4) line is 19.28 8. Based on the shortest Tm7C distances to the terminal naphthalene moieties [2.41 and 2.42 A for C(a) and 2.54 and 2.54 A for C(b)], both types of interactions can be considered as Z4 and Z2. However, the dihedral angle (26.22 8) and the redistribution of the C7C bond lengths in the coordinated rings correspond to cyclohexa-1,4-diene. The absolute values of the shortest Tm7C distances are identical (taking account of the difference in the ionic radii and coordination numbers of the metal atoms) to the Ln7C s-bonds in the complexes [Li(THF)3]. .{[(Me3Si)2CH]3YbCl} (the average distance is 2.375 A),54 (Cp*=Z5-C5Me5, Cp*CeCH(SiMe3)2 2.535 A) 55 and Cp*NdCH(SiMe3)2 (2.517 A) 56 It was assumed 46, 57 that black insoluble finely dispersed products formed upon reduction of SmBr3 and YbBr3 with sodium naphthalenide are metallic lanthanides. However, there are strong grounds to believe that these reactions, like the analogous reactions with iodides, afforded naphthalene com- plexes of lanthanides.Apparently, the results of reduction of NdCl3 with lithium naphthalenide were also misinterpreted.58 Reinvestigation 59 of reduction of NdCl3 with insufficient lithium in THF in the presence of naphthalene demonstrated that the reaction performed under these conditions gave rise to a mixture of naphthalene complexes with composition [NdCl2(THF)2LiCl]n(C10H8) (n=4 ± 7) rather than divalent neo- dymium chloride NdCl2(THF)2 as has been reported previously.60 Binuclear complexes of the general formula ({PhP[CH2. .(SiMe2)N(SiMe2)CH2]2PPh}Ln)2Arn (19a ± d) [Arn=C10H8, Ln=Y (a) or Lu (b); Arn=C10H7Me, Ln=Y (c) or Lu (d)] containing the Ln(m-Arn)Ln group were prepared by the reactions of a mixture of potassium-intercalated graphite (KC8) and naph- thalene or 1-methylnaphthalene with the complex ({PhP[CH2..(SiMe2)N(SiMe2)CH2]2PPh}YCl)2 (9) or its lutetium analogue.61 X-Ray diffraction analysis demonstrated that the fragments containing the phosphorus ligand bound to Y in the compound 19a, like those in the complex [I2La(THF)3)]2(m-C10H8) (17), are located on the opposite sides of the bridging naphthalene ligand Z4-bonded to the Y atoms.The shortest Y7C(1,2,3,4)(naphtha- lene) distances are 2.696, 2.684, 2.686 and 2.652 A, respectively. The rings in the naphthalene ligand are folded along the C(1)7C(4) and C(5)7C(8) lines. According to the preliminary data, the lutetium analogue has a similar structure. Ph Si P Si Si P Si N N Ph Y Y N N P Ph Si Si Si P Si 19a Ph The 31P{1H} NMR spectra of all the four compounds are similar (the spectrum of the yttrium derivative has a doublet atArene complexes of rare-earth metals d 21, 1JYP=60 Hz), which is indicative of the equivalence of the phosphorus ligands. Based on this fact, it was suggested 61 that intramolecular migration of the {PhP[CH2(SiMe2)N..(SiMe2)CH2]2PPh}Ln moieties between the naphthalene rings (structures A±D) occurs in solution. Apparently, the moieties migrate independently and, as a consequence, A is converted into B through the intermediates C and D. Ln Ln Ln Ln B A Ln Ln C Ln Ln D R=H, Me; Ln=Y, Lu. lanthanide The complexes [CpLn(THF)(m-Z4 : Z6- C10H8)VCp]x (20a ± c) containing the bridging naphthalene ligand were prepared by the reactions of naphthaleneytterbium, -euro- pium or -samarium with vanadocene.62, 63 O Ln (C10H8)Ln(THF)2+Cp2V THF V x 20a ± c Ln=Sm (a), Eu (b), Yb (c). According to the data of ESR spectroscopy and magnetic measurements, the bridging naphthalene ligand in the complexes 20a ± c is neutral and the negative charge is located at the vanadium atom as a result of which the latter becomes formally zero-valent.Compounds of this type can be considered as dicy- clopentadienyl complexes of lanthanides in which one of the cyclopentadienyl anions is coordinated to the metal atom through the neutral (C10H8)V group. Analogous lanthanide ± vanadium complexes [CpV(m-Z6 : Z6- C10H8)]2Ln(THF)(DME) (21) were synthesised in THF by the reactions of samarium or europium diiodides with cyclopentadi- enylnaphthalenevanadium anions prepared in situ.63 The reac- tions in hexane gave rise to monomeric complexes in which the rare-earth atom is coordinated to two CpV(C10H8) units. LnI2(DME)3 , THF CpK+[C10H8VCp]K O V V LnO O 21a,b Ln=Sm (a), Eu (b).The naphthalene ligands in the complexes 20 and 21 are planar. The Ln7C(naphthalene) bond lengths in the complex 20b [Eu7C(a), 2.828 and 2.869 A; Eu7C(b), 3.037 and 3.098 A] are indicative of Z4-interaction in the CpLn(C10H8)VCp moieties. In the quadruple-decker compound 21b, the Eu7C bond lengths are in the range of 2.872 ± 3.084 A, i.e., Z6-coordination occurs in the (C10H8)Eu(C10H8) moieties. 787 The reaction of lithium naphthalenide with Cp2LuCl pro- ceeded through another pathway. In this case, the resulting intermediate radical-anionic unstable) (apparently, Cp2Lu(C10H¡8 ) (22) underwent rapid conversions involving met- allation of naphthalene and formation of hydride and naphthyl derivatives.64 THF Cp2LuCl+C10H8Na 8 ) Cp2Lu(C10H¡ 22 THF Cp2Lu +[(Cp2LuH)3H][Na(THF)6] Reduction of CpLuCl2 with sodium naphthalenide afforded the mononuclear complex (Z5-C5H5)Lu(C10H8)(DME) (23).65 O DME Lu +C10H8 CpLuCl2+C10H8Na O 23 The molecular geometry of the complex is indicative of the s-character of the bond between the lutetium atom and the naphthalene ligand whose coordinated ring loses aromaticity. The Lu7C(a) bond lengths (2.406 and 2.397 A) are compa- rable with the Lu7C s-bond lengths in the complexes Cp2LuBut(THF) (2.47 A) 66 and Cp2LuCH2SiMe3(THF) (2.376 A).67 The Lu7C(b) bond lengths (2.579 and 2.562 A) are close to the Lu7C(Cp) bond lengths (2.65 A), which suggests the presence of additional p-interaction of the lutetium atom with the double bond (1.36 A) between the C(b) atoms of the coordinated ring.The folding angle in the coordinated ring is 31.5 8.68 There- fore, the bond between the naphthalene ligand and the metal atom in this case may be characterised as 2Z1 :Z2(2s,p) interac- tion.64, 65, 68 Analogous Y, Gd, Er and Tm complexes were also prepared.69 In the reactions with C10H8Na performed under comparable conditions, lanthanide derivatives characterised by lower poten- tials of Ln(II)/Ln(III) transitions (such as Sm and Yb) were reduced to form the binuclear Ln(II) complexes containing the bridging naphthalene group, viz., (CpLn)2(m-C10H8)4 (24).69 CpLnCl2+C10H8Na (CpLn)2(m-C10H8)4+C10H8 24 Ln=Sm, Yb. The reaction with the use of the pentamethylcyclopentadienyl derivative of lutetium, Cp*LuCl2, as the starting compound yielded the diamagnetic ate-cluster {[(Cp*Lu)3(C10H8)(C10H7)]..[Na(THF)3]}2+ (25) whose trinuclear cation contains the naph- thalene dianion and the naphthyl group with a charge of73.70 A B Lu2 Lu1 Lu3 C O Na O O 25 The naphthalene dianion, which is not coordinated to the metal atoms, represents the anionic moiety of the complex. The ring A of one of the naphthalene ligands in the cationic moiety is Z4-bonded to the Lu(1) atom. As in the complex 23, the Lu7C788 bond lengths of this fragment in 25 [Lu7C(a), 2.58 and 2.60 A; Lu7C(b), 2.67 and 2.77 A] and the folding angle along the C(a)7C(a) line (the dihedral angle is 172.5 8) correspond to 2Z1 :Z2-interaction.The B ring remains planar and is Z6-bonded to the Lu(2) atom (the Lu7C bond lengths are in the range of 2.61 ± 2.88 A). The 2Z1 :Z2-interaction is most pronounced in the Lu(3)7C10H7 group. In this case, the dihedral angle along the C(a) line is 142.5 8. The Lu(3)7C(a,b) distances are 2.44, 2.47, 2.54, and 2.59 A, respectively. The naphthyl ring C is s-bonded to the Lu(2) atom. The short Lu(1)7ring C (2.81 A) and Lu(3)7 ring A (2.72 A) distances are, apparently, indicative of some degree of bonding in these fragments. It is not inconceivable that the anionic cluster contains a bridging proton bound to the Lu(1) and Lu(2) atoms. However, the presence of this proton has not been unambiguously established.It was reported 71 that thin films of cerium and europium readily reacted with naphthalene in diethyl or dipropyl ethers at room temperature to form colourless solutions. Based on the spectral characteristics, it was concluded that pale-brown sub- stances obtained after removal of the solvent and excess naph- thalene are bis(naphthalene) p-complexes of zero-valent lanthanide of the bis(naphthalene)chromium type. However, this conclusion is in contradiction with the following facts: the products are pale in colour, which is untypical of neutral and charged arene complexes; under these conditions, naphthalene does not react with other lanthanides (Pr, Nd, Gd, Tb, Dy, Ho or Er); the reaction does not proceed in THF; the reaction with anthracene proceeds more slowly than that with naphthalene.The reaction of acenaphthylene with permethylsamarocene readily proceeded in benzene giving rise to a dark-blue compound. According to the data of NMR and IR spectroscopy and elemental analysis, the compound can be described by the chemical formula (Cp2 Sm)2(C12H8).72 The structure of the prod- uct was not established. IV. Complexes with anthracene Anthracene which possesses the higher electron affinity than naphthalene forms complexes with lanthanides in the direct reactions with ytterbium, samarium and cerium activated with diiodoethane.73 ± 75 Attempts to isolate these complexes in an individual state failed. However, structure 26 was assigned to the resulting complexes by reasoning from the products of reactions with MeOD, MeI and Me3SiCl.Ln(solv) 26 A film of metallic ytterbium prepared by concentrating a solution of the metal in liquid ammonia also reacted with anthracene in the presence of LiCl, NaBr, LiI, KI or Bun4 NI.76 The authors believed that the reactions gave rise to complexes containing the anthracene radical anion with composition Yb(C14H¡10 )2 . nMX. However, attempts to identify the reaction products were also unsuccessful. After washing of the substance with benzene, a violet powder containing the anthracene dianion, ytterbium and NaI in a ratio of 1 : 1 : 6 was obtained. Anthracene complexes were prepared by the reactions of anthracene with naphthaleneytterbium in THF in the presence of alkali metal halides or by the reaction of YbI2 with C14H10Na.40, 77 These complexes were isolated as black insoluble powders, which are similar in properties to naphthalene ana- logues.ESR spectral studies of these compounds and their hydrolysis products suggested the existence of an equilibrium between the radical-anionic (A) and dianionic (B) forms, with the latter predominating: M N Bochkarev (C10H8)Yb(THF)3+C14H10+LiX (C10H8)YbX +(C14H¡10 )Li+ [(C10H8)Yb(X)C14H¡10 ]Li+ or MX C14H10 Yb+C14H10 (C14H¡10 )2Yb . nMX A (C14H2¡ 10 )Yb+n MX. B Since the resulting complexes are poorly soluble, their high- quality crystals were not prepared and the structures could not be established by X-ray diffraction analysis.However, analogous investigations were successfully performed by the use of cyclo- pentadienyllanthanide derivatives. It appeared that rare-earth metals form two types of anthracene complexes through addition at either an edge or a plane; however, the arene ligand in both cases carries a negative charge72. The reaction of CpLuCl with the dianion or two radical anions of anthracene gave rise to heterobimetallic ate-complex 27 con- taining the anthracene dianion.78 CpLuCl +(C14H¡10 )Na+ THF CpLu(C14H2¡ 10 )Na+ CpLuCl +[C14H2¡ 10 ]Naá2 27 2La)2(m-Z3 : Z3-C14H10) . 2PhMe (29) with Neutral complexes of samarium (Cp2 Sm)2(m-Z3 : Z3-C14H10) (28) and lanthanum (Cp the anthracene dianion were prepared by oxidative addition of decamethylsamarocene to anthracene 72 or by metathesis of the lanthanum potassium complex with disodium anthracene.79 Sm Cp2 Sm+C14H10 Sm 28 PhMe Cp2 La(Cl)K(DME)2+Na2C14H10 (Cp2 La)2(m-Z3 : Z3-C14H10) .2 PhMe 29 Both products are structurally similar. The planar anthracene dianion is Z3 : Z3-bonded to two Cp2 Ln units located on the opposite sides of its plane. Apparently, the product of the reaction of Cp2 Sm with 9-methylanthracene, viz., (Cp2 Sm)2C15H12, has an 10 ) . analogous structure.72 The reaction of C14H27 10 with CpLuCl2 afforded a complex with different composition and structure, viz., CpLu(C14H27 2THF (30).80, 81 O THF Lu CpLuCl2+Na2(C14H10) 72NaCl OOO 30 In the complex 30, the character of the bond between the Lu atom and the arene ligand is analogous to that observed in the complex 23, i.e., it can be interpreted as s,s-interaction althoughArene complexes of rare-earth metals the Lu7C(9,10) distances (2.44 and 2.45 A, respectively) are somewhat longer than the analogous contacts in the naphthalene derivative 23, but are substantially shorter than those in the binulcear anthracene compound (2.683, 2.850 and 2.882 A).The folding angle of the anthracene molecule along the C(9)7C(10) line is 38.7 8 (29.9 8 for the second symmetrically independent molecule in the crystal). The complex 30 containing the anthracene dianion was also synthesised by the reaction of CpLuCl2 with the anthracene radical anion. In this case, free anthracene was detected in the reaction mixture.80, 81 CpLuCl2+[C14H¡10 ]Na+ THF 7NaCl Unlike the lanthanum and samarium complexes, lutetium di- cyclopentadienyl chloride gave the ate-complex [CpLu(C14H2¡ 10 )]..[Na+(THF)] in the reactions both with C14H¡10 and C14H210¡.81 The replacement of coordinated THF by diglyme afforded com- plex 31. The authors managed to grow perfect crystals of 31 and to establish its structure by X-ray diffraction analysis.81 It appeared that the character of the bond between anthracene and the lanthanide atom in this compound is similar to that in the complexes CpLu(C10H8)(DME) (23) and CpLu(C14H2¡ 10 ) (30). The arene ligand is folded along the C(9)7C(10) line (the dihedral angle is 40.1 8).The shortest distances between the anthracene ligand and the lutetium atom [Lu7C(9,10) are 2.473 and 2.482 A, respectively] are only 0.03 A larger than the corresponding distances in the complex 30. O OO Na OO O 31 The reaction of the aminophosphine yttrium complex 9 with anthracene in the presence of KC8 yielded the binuclear complex ({PhP[CH2(SiMe2)N(SiMe2)CH2]2PPh}Y)2(m-C14H10) (32). The latter, like the anthracene derivatives 30 and 31, contains the C14H2¡ 10 dianion.61 The metal atoms in the compound 32 are located on the opposite sides of the aromatic ligand and are Z4-bonded to the central ring and one terminal ring. Ph Si P Si Si P Si N N Ph Y Y N Ph Si P Si Si P Si 32 Ph The coordinated rings are folded along the C(1)7C(4) and C(9)7C(10) lines.The shortest Y7C distances (the average values are 2.694 and 2.751 A) are virtually identical to the corresponding distances in the naphthalene analogue 19a. Based on the data of 1H and 31P{1H} NMR spectroscopy, it was assumed 61 that in solution the metal-containing moieties in the CpLu[C14H2¡ 10 ] . 2THF+C14H10 30 Lu N 789 complex 32, as those in the naphthalene derivative 19a, migrate along the arene plane. It is significant that intermolecular migra- tion of the {PhP[CH2(SiMe2)N(SiMe2)CH2]2PPh}Ln moieties in the anthracene complexes or in mixtures of anthracene and naphthalene analogues does not occur. This is evidence that the complexes do not dissociate in solution in spite of the ionic character of the Ln7Arn bond.V. Pyrene and benzoanthracene derivatives Pyrene can form three types of complexes with rare-earth metals.82 The reaction of Cp2 La(m-Cl)2K(THF)2 (33) with the pyrene dianion in THF afforded red crystals with composition [Cp2 La(THF)2](C16H11) (34) in low yield. The 1H NMRspectrum of 34 has nine signals with equal intensities and one double- intensity signal (at d 4.12). The authors believed 82 that these signals are indicative of the presence of the hydropyrene anion C16H¡11, which was, apparently, formed as a result of detachment of protons from the solvent. THF [Cp2 La(THF)2](C16H11) Cp2 La(m-Cl)2K(THF)2+C16H211¡ 34 33 It was assumed 82 that the [Cp2 La(THF)2]+ cation does not form a strong bond with the aromatic anion, i.e., the complex in solution exists as a solvent-separated ion pair.The reaction of the ate-complex 33 with pyrene and potassium in toluene taken in a ratio of 2 : 2 : 1 gave rise to black-green crystals. The NMR spectrum of these crystals corresponds to the binuclear complex (Cp2 La)2(m-Z2 : Z2-C16H10) (35). It was sug- gested 82 that the Cp2 La units in 35, like those in the anthracene derivatives 29, are located on the opposite side of the plane of the aromatic ligand and are bound to the opposite rings. La La 35 The samarium analogue (Cp2 Sm)2(m-Z3 : Z3-C16H10) (36), which was prepared by the reaction of free pyrene with Cp2 Sm, differs in structure (according to the X-ray diffraction data) from 35.72 Both metallocene moieties in 36 are coordinated at the edge of pyrene and are located above and below its plane, but they form three short contacts each with the adjacent rings rather than with the opposite rings.Sm Sm 36 The angles between the samarium atoms and the plane of the polycyclic ligand are 117 8 and 120 8. The Sm7C distances are in the range of 2.660 ± 2.806 A, which is somewhat larger than the790 corresponding distances in the allylic complexes Cp2 Sm(Z3- CH2CHCHR) (2.551 ± 2.643 A),83 but are substantially shorter than the Sm7C bonds in the arene chloroaluminate compounds [ArnSm(AlCl4)3] (2.89 ± 2.91 A).5, 7, 11 In the reaction with the use of the reagent ratio Cp2 Sm: pyr- ene :THF=3 : 3 : 1 or in the reaction of pyrene and potassium with the monocyclopentadienyl complex Cp*LaCl(m-Cl)2 .Li(THF)2, pyrene was reduced to the trianion yielding the red- violet trinuclear complex (Cp*LaCl)2[Cp*LaCl(THF)](C16H10) (37).82 According to the X-ray diffraction data, all three Cp*LaCl moieties are located above the plane of the pyrene ligand. Two of these moieties are Z6-bonded to the arene ligand, whereas the third moiety containing an additional coordinated THF molecule forms only two short La7C(pyrene) contacts. O Cl Cl LaCl La La 37 The lengths of the La7C bonds with the participation of the Z6-coordinated rings are in the range of 2.766 ± 3.072 A and the lengths of the La7C bonds with the participation of the Z2-coor- dinated rings are 2.823 and 2.841 A, which are 0.1 A larger than the corresponding contacts in the naphthalene complex [I2La(THF)3]2(m-C10H8) (14).2,3-Benzoanthracene, like pyrene, readily reacts with Cp2 Sm to give green crystals of the complex (Cp2 Sm)2(m-Z3 : Z3-C18H12) (38),72 which is structurally similar to the pyrene complex 36. Two Cp2 Sm fragments are coordinated at the edge of the arene ligand and are located on the opposite sides of its plane. The distance from the Sm atoms to this plane is 2.063 A. Both metal atoms are bound to the same central ring. The Sm7C(Arn) bond lengths (2.688 ± 2.828 A) are similar to those in the pyrene analogue 36. Sm Sm 38 Interestingly, all permethylsamarocene complexes with are- nes, unlike other arenelanthanides, are readily decomposed upon dissolution in THF to form Cp2 Sm(THF)2 and the corresponding polycyclic hydrocarbon.72 This lability of the Sm7Arn bond may be evidence for its nonvalent character, i.e., the formation of these derivatives is not accompanied by redox processes and the coordination complexes [Cp2 Sm(II)]2(Arn0) are formed.How- ever, the 1H and 13C NMR spectra of the reaction products are unambiguously indicative of the change in the valence state of the metal atom from Sm(II) to Sm(III) in the course of the reaction and, consequently, the arene ligand in the complexes exists as the dianion. The Sm7Cp* distances in these complexes and their electronic spectra also correspond to trivalent samarium.The characteristic feature of the permethylsamarocene complexes with arenes is the allylic type of bonding between the arene ligand and the Cp2 Sm unit.72 M N Bochkarev VI. Complexes with heterocyclic aromatic ligands 2Sm)2(m- The great majority of complexes of rare-earth metals with hetero- cyclic aromatic ligands are formed through coordination inter- action between the lone electron pair of the heteroelement and the Ln atom.1 In these complexes, the ligands can be either neutral or charged (71 or 72). The charges are localised primarily on heteroatoms. However, examples of compounds similar in nature to typical bisarene complexes are also available. One of them, viz., (2,4,6-But3C5H2N)2Sc, was prepared under conditions of the cryogenic synthesis.3 It was suggested that this compound con- tains the formally zero-valent scandium atom and the neutral tert- butyl-substituted pyridine ligands.The molecule was represented as a planar-parallel sandwich. The reaction of Cp2 Sm with phenazine, like that with anthracene, afforded the binuclear complex (Cp Z3 :Z3-C12H8N2) (39) in which the samarocene fragments are in the transoid orientation with respect to the [C12H8N2]27 dia- nion.72 The presence of three short contacts between the sama- rium atoms and the phenazine ligand are indicative of the allylic type of bonding. Sm N N Sm 39 The Sm7N bond length (2.360 A) is slightly larger than the Sm7N s-bond length in the complex Cp2 SmN(SiMe3)2 (2.306 A),84 which is evidence for substantial localisation of the charge on the nitrogen atoms, as in the complexes of the Cp3Ln(py) 1 and Ln(bipy)4 types (bipy is 2,20-bipyridyl).85 The reaction of permethylsamarocene with acridine was accompanied by dimerisation of the latter to form aC7C bond.73 N Cp2 Sm+ SmCp2 N N Cp2 Sm 40 The character of the metal7arene bond in the complex 40 is analogous to that in the phenazine derivative, i.e., corresponds to Z3-interaction, with the Sm7N component predominating.Arene lanthanide complexes of an unusual type were prepared by cocondensation of vapours of scandium and tert-butylphos- phoalkyne ButC:P at 77 K.86 The reaction was accompanied by cyclisation of the phosphoalkyne giving rise to two products, viz., dark-violet scandocene (Z5-But3C3P2)2Sc containing divalent scandium and the green triple-decker complex [(Z5-But2C2P3)Sc]2..(m-Z6 : Z6-But3C3P3) (41) containing formally monovalent scan- dium. The valence states of the scandium atoms in both complexes were unambiguously established by UV and ESR spectroscopy and magnetic measurements. The structure of the triple-decker complex was confirmed by X-ray diffraction analysis.87 The central six-membered ring in the complex 41 is located on a crystallographic mirror plane. All P7C bond lengths in this ring have closely similar values (1.793 ± 1.807 A).Arene complexes of rare-earth metals P But But P P ScBut P P But P But Sc P But But P P41 It is worthy of note the distance from Sc to the central ring in the compound 41 is noticeably shorter (1.787 A) than the analo- gous distance in the bisarene complex (But3C6MeH2)2Sc (1.983 A).87 In spite of the valence state +1 untypical of scan- dium, the compound 41 is quite stable and can be even sublimed in vacuo at 250 8C.The magnetic moment (3.98 mB) of 41 is somewhat smaller than the value calculated for four unpaired electrons, which was attributed to the presence of an orbital contribution.87 VII. Reactivity of arene lanthanide complexes The chemical properties of complexes of rare-earth metals with p-bonded arene ligands were studied primarily using naphthalene derivatives as examples. For compounds with other aromatic hydrocarbons, only the most common characteristics are known. All arene lanthanide complexes exist as dark-coloured solids, which are readily oxidised in air (often with ignition).It was found that bisarene complexes of formally zero-valent Sc, Y, Nd, Gd, Tb, Dy, Ho, Er and Lu are thermally stable under an inert atmosphere or in vacuo. Some of them can even be sublimed with insignificant decomposition at 100 8C (at 1074 mmHg). These complexes are readily soluble in aromatic and aliphatic hydro- carbons. Complexes with arene dianions are virtually insoluble. The yttrium complex (1,3,5-But3C6H3)2Y (5b) readily reacted with phenol, 2,4,6-But3C6H2OH and hexamethyldisilazane to form tri(aryloxide) (42) and yttrium tris(trimethylsilyl)amide (43) in good yields.The reaction of the holmium complex 5c with bi-tert-butyl-1,3-diazabuta-1,3-diene (DAD) also resulted in com- plete replacement of the arene ligands.3 But (Me3Si)2NH [Me3Si)2N]3Y 42 But But 2,4,6-But3C6H2OH Ln (2,4,6-But3C6H2O)3Y But But DAD 43 Ln(DAD)3 5b,c But Ln=Y (b), Ho (c); DAD=ButN CHCH NBut. All (Arn)2Ln complexes are efficient catalysts of ethylene polymerisation. Naphthalene lanthanide complexes of all types exhibit very high reactivities with respect to many inorganic, organic and organometallic compounds. Essentially in all cases, these com- plexes act as strong reducing agents comparable in reducing ability to alkali metal naphthalenides. The reactions were carried out in THF orDMEat room or lower temperature.Generally, the yields of the products many of which were characterised by X-ray diffraction studies are higher than 60%± 70%. Hence, these reactions may be recommended as preparative (often the only 791 possible) procedures for the synthesis of organolanthanide com- pounds. The important advantage of these syntheses is the fact that the second product, viz., naphthalene, is unreactive and can be readily separated from the target compound. The reactions with the participation of the complexes (C10H8)Ln(THF)x are listed in Table 1. The compounds (C10H8)Ln(THF)x also exhibit high reactivity in catalytic processes. Thus the complex (C10H8)Yb(THF)3 catal- yses hydrogenation of hex-1-ene, stilbene, isoprene and piperylene at room temperature under atmospheric pressure of H2 yielding the corresponding hydrocarbons.41 Tolan was hydrogenated to form trans-stilbene and diphenylethane.In the presence of 1%± 5% of (C10H8)Yb(THF)3, styrene, methyl methacrylate, ethyl acrylate, isoprene and piperylene formed polymers at room temperature.48 The conversion reached 100% in several hours. Piperylene produced more than 80% of the trans-polymer, whereas polyisoprene was obtained as a mixture of virtually equal amounts of cis and trans forms. Naphthaleneytterbium catalyses polymerisation of epoxides and their reactions with CO2. In the latter case, a mixture of monomeric and polymeric alkyl carbonates was obtained.48 It is interesting to note that naphthalene complexes of `trivalent' lanthanides, which were not isolated in the individual state, appeared to be efficient catalysts of polymerisation of conjugated dienes.43 The reactions of the latter [like that of (C10H8Tm)2(C10H8)] with cyclopentadiene afforded the corresponding Cp3Ln in 15%± 60% yield depending on the metal.52, 53 It was found 58 that the products of the reaction of NdCl3 with C10H8Li (apparently, the reaction afforded naphthalene neody- mium complexes rather than finely dispersed neodymium as has been suggested in Ref. 58) readily reacted with ketones and aldehydes to yield the corresponding diols or ethylene derivatives as a results of coupling of the substrate.Generally, mixed-ligand complexes containing naphthalene as well as the iodine atom, the cyclopentadienyl substituent or the double-decker CpV(C10H8) moiety bound to the rare-earth ele- ment react analogously to compounds of the (C10H8)Ln(THF)3 type, i.e., the attack occurs primarily on the (C10H8)Ln group. Due to the presence of the LnI or LnCp groups along with the Ln(C10H8) moiety, these complexes are convenient starting com- pounds for the preparation of complexes of the RLnI or RLnCp types.The most typical reactions are given in Table 2. The chemical properties of anthracene lanthanide complexes are analogous to the properties of their naphthalene analogues. However, the activities of the former are substantially lower. Thus the reactions of anthraceneytterbium with cyclopentadiene, fluo- rene, 9-tert-butylfluorene, tetraphenylpropene and 9-phenylxan- thene in solutions afforded the corresponding anions, i.e., the R2Ln compounds apparently formed, but the reaction rates were low (from several days to weeks).Triphenylmethane did not enter into the reaction at all.76 Deuterolysis of the complexes (C14H10)Ln(THF)x (Ln=Sm or Yb) gave rise to 9,10-dideutero- 9,10-dihydroanthracene,48, 74 whereas their reactions with MeI afforded a mixture of 9,10-dimethyl-9,10-dihydroanthracene (14% ± 20%), MeC14H9 (1% ± 4%) and 9,10-dihydroanthracene (13% ± 25%). It is noteworthy that attempts to alkylate naphtha- lene in the reactions of naphthaleneytterbium with different alkyl halides have not met with success. In all cases, only unsubstituted naphthalene was obtained.{ For anthracene mixed-ligand complexes of lanthanides, only the reaction of CpLu(C14H10)(THF)2 with iron pentacarbonyl giving rise to the heterobimetallic product with composition [CpLu(THF)]Fe(CO)4 was reported.106 { Unpublished data of the author of the present review.M N Bochkarev 792 Table 1. Reactions of the naphthalene complexes (C10H8)Ln(B)x (Ln=Sm, Eu or Yb; B=THF or DME, x=2 or 3). Ref. Products Substrate 48 88, 89 48 see a 48 48 see a see a 90 90 90 90 91 45 92 93 88, 94 95, 96 96 62, 63, 97 O2 Ln2O3+C10H8 H2 LnH2+C10H8 H2O C10H10+Ln(OH)3 ButOH (ButO)2Ln+C10H8 R2NH [R=Et, N(SiMe3)2] (R2 N)2Ln+C10H8+C10H10+H2 CpH IndHb FluHc CpCH2CH2OH CpCH2CH(Me)OH CpCH2CH(CH2OBu)CH2OH CpSiMe2NHBut C5Me4HSiMe2OSiMe2OH HC:CH PhC:CHd Ph2Hg e Ph3GeHe Ph4Sn e Ph6Sn2 (see e) Cp2Vd 46 Cp2Ln+C10H8+H2 Ind2Yb Flu2Yb [(Z5-Cp)CH2CH2(Z1-O)]Yb(B)+C10H8+H2 [(Z5-Cp)CH2CH(Me)(Z1-O)]Yb(THF)+C10H8+H2 [(Z5-Cp)CH2CH(CH2OBu)CH2(Z1-O)]Yb(THF)+C10H8+H2 [(Z5-Cp)SiMe2(Z1-NBut)]Yb(THF)+C10H8+H2 {[(Z5-Cp)Si(Me2)OSi(Me2)(Z1-O)]Yb(THF)}2+C10H8+H2 LnC2(THF)x+C10H8+H2 (PhCH=CHCH=CHPh)Ln(DME)2+C10H8 Ph2Yb(THF)(m-Ph)3Yb(THF)3+Ph3Yb+Hg+C10H8 (Ph3Ge)2Yb(THF)4+C10H8 (Ph3Ge)2Eu(DME)3+C10H8 Ph3SnYb(THF)(m-Ph)3Yb(THF)3+(Ph3Sn)2Yb(THF)4+C10H8 (Ph3Sn)2Yb(THF)4+C10H8 CpVC10H8VCp [CpLn(THF)C10H8VCp]n Cp2Yb 97 Cp2Yb 97 Cp2Yb+(C10H8)2Cr 97 Yb(DAD)3+C10H8 98 [Yb(m2-bipy)(THF)3]3 99 [Yb4(m2-Z2 : Z2-Ph2N2)4(m3-PhN)2(THF)4] 39 [Sm4(m2-Z2 : Z2-Ph2N2)4(m3-PhN)2(THF)6] 47 [CpLu(DME)]2[1,1-m-4,4-m-PhCC(Ph)=(Ph)CCPh]+C10H8 100 SmS(THF)+C10H8 101 RLnX (Sm, Yb) Cp2Ni e Cp2Co e Cp2Cr e DADe bipy e, f PhN=NPhe see f PhC:CPh h S8 (see f) RX (R=Ph, C6F5 , X=Br, I) , X=Br) S RX (R=Ph, 4-BrC6H4 , Ph2CO (H+) PhCHO (H+) (H+) O CO2 CO 57 48 48 48 48 48 RLnX (Sm, Yb) (Ph2COH)2 (PhCHOH)2 HOCH2(CH2)2CH2OH+C10H8 C10H8(COOH)2 C10H8(COOH)2 a Unpublished data of the author of the present review; b Ind is 9-indenyl; c Flu is 9-fluorenyl; d for Eu and Yb; e for Yb; f for Sm; g for Lu.Table 2. Reactions of the naphthalene mixed-ligand lanthanide complexes. Ref. Products Substrate Complex [LaI2(bipy)2(THF)3]2[C10H8] bipy bipy bipy C60 [YbI(DME)2]2(C10H8) CpLu(C10H8)(DME) CpLu(C10H8)(DME) CpLu(C10H8)(THF)2 (C10H8)[LnI(DME)2]2 (see a) (C10H8)[EuI(DME)2]2 [CpV(C10H8)YbCp(THF)]n LaI2(bipy)2(DME)+C10H8 102 LaI2(bipy)(DME)2+C10H8 102 YbI(bipy)(DME)2 103 CpLu(C60)(DME) 104 Lu(OH)3+C10H10+CpH 65 Lu2O3+C10H8 65 Cp3Lu+C10H10+C10H8+H2 65 [CpLu(DME)]2(1,1-m-4,4-m-PhCC(Ph)=(Ph)CCPh]+C10H8 99 [CpLu(THF)]2(Ph2N2) 99 [PhC:CLnI(DME)2]2+C10H8+H2 95 Ph3GeEuI(DME)2+C10H8+H2 97 CpYbOH(THF)2+CpVC10H8 105 Cp2Yb+CpVC10H8 105 Cp2Yb+CpVC10H8 105 Cp2Yb+CpVC10H8+Cp2V+CpVC10H8VCp+C10H8 105 Cp2Yb+CpVC10H8 105 [CpYb(THF)2]2(Ph2N2)2+CpVC10H8 105 H2O O2 CpH PhC:CPh Ph2N2 PhC:CH Ph3GeH H2O CO CO2 Cr(CO)6 CpTl Ph2N2 a For Eu and Yb.Arene complexes of rare-earth metals VIII.Conclusion Based on analysis of the available data, the characteristic features of complexes of rare-earth metals with p-bonded arene ligands were revealed. First, it should be noted that these compounds exist in the two major forms, viz., with neutral arene or with the arene dianion. Forms with arene carrying a charge 71 or 73 are untypical of these complexes. Apparently, these forms can be obtained and exist only under particular conditions. It should be noted that the tendency of lanthanides to form complexes of the [Ln]2+[L]27 type even in the presence of the free uncharged ligand is also observed in the case of other ligands (such as bipy } or DAD) exhibiting polyvalence.107 The mode of bonding of lantha- nide with arene ligands can vary from Z2 to Z6.The Ln atom can be coordinated both at the plane of the arene ligand and at its edge. The character of the Ln7Arn bond in complexes with neutral arene and formally zero-valent lanthanide is similar to the character of the metal7arene bond in complexes of d metals, i.e., the bond is determined by back donation of the electron density from the occupied dxy and dx27y2 orbitals of lanthanide to the unoccupied p-orbitals of arene. In compounds with neutral ligands coordinated to the Ln3+ cation {complexes of the (Arn)Ln[(m-Cl)2AlCl2]3 or Ln(OC6H3Ph2-2,6)3 types}, the d orbi- tals of the metal atoms are unoccupied and interactions occur, apparently, through donation of the electron density from the p-orbitals of the arene ligand to the unoccupied orbitals of the metal atom.The metal7arene bonds in these compounds are substantially weaker as evidenced by the noticeably elongated Ln7C(arene) bonds. In complexes containing Ln2+, (XLn)2+ or (X2Ln)+ cations and the arene dianion (naphthalene lanthanide complexes and related derivatives) bonding occurs, apparently primarily through Coulomb interactions between the positively and negatively charged moieties of the molecule. However, the structures of the complexes and the fact that these complexes contain the shortest Ln7C(arene) bonds of all arene lanthanide complexes are indicative of the presence of the second bonding component, viz., of donation of electron density from the p-orbi- tals of arene to the unoccupied orbitals of the metal atom. In spite of the high (judging from the geometric parameters of the molecules) energy of the metal ± arene bond, naphthalene lantha- nide complexes exhibit very high reactivity.At the same time, these complexes are readily accessible owing to which they show promise as a source of lanthanides or lanthanide-containing fragments in the synthesis of other derivatives of rare-earth metals. The review has been written with the financial support of the Russian Foundation for Basic Research (Project No. 00-03- 32875). References 1. M N Bochkarev, L N Zakharov, G S Kalinina Organoderivatives of Rare Earth Elements (Dordrecht: Kluwer Academic, 1995) 2. G B Deacon, Q Shen J.Organomet. Chem. 511 1 (1996) 3. F G N Cloke Chem. Soc. Rev. 22 17 (1993) 4. F A Cotton, W Schwotzer Organometallics 6 1275 (1987) 5. F A Cotton, W Schwotzer J. Am. Chem. Soc. 108 4657 (1986) 6. H Liang,Q Shen, J Guan,Y Lin J. Organomet. Chem. 474 113 (1994) 7. B Fan, Q Shen, Y Lin J. Organomet. Chem. 376 61 (1989) 8. B Fan, Q Shen, Y Lin J. Organomet. Chem. 377 51 (1989) 9. B Fan, Q Shen, Y Lin Wuji Huaxue Xuebao 7 143 (1991); Chem. Abstr. 116 106 434 (1992) 10. S Jin, J Guan, H Liang, Q Shen Cuihua Xuebao 14 159 (1993); Chem. Abstr. 119 49 956 (1993) 11. B Fan, Q Shen, Y Lin Youji Huaxue 9 414 (1989) 12. P Biagini, G Lugli, L Abis, R Millini New J. Chem. 19 713 (1995) 13. H Liang, Q Shen, S Jin, Y Lin Zhongguo Xitu Xuebao 12 193 (1994); Chem. Abstr.123 144 082 (1995) } Unpublished data of the author of the present review. 793 14. H Liang, Q Shen, S Jin, Y Lin J. Chem. Soc., Chem. Commun. 480 (1992) 15. G B Deacon, S Nickel, P MacKinnon, E R T Tiekink Aust. J. Chem. 43 1245 (1990) 16. G B Deacon, T Feng, B W Skelton, A H White Aust. J. Chem. 48 741 (1995) 17. G B Deacon, B M Gatehouse, Q Shen, G N Ward Polyhedron 12 1289 (1993) 18. D M Barnhart, D L Clark, J C Gordon, J C Huffman, R L Vincent, J G Watkin, B D Zwick Inorg. Chem. 33 3487 (1994) 19. R J Butcher, D L Clark, S K Grumbine, R L Vincet-Hollis, B L Scott, J G Watkin Inorg. Chem. 34 5468 (1995) 20. S Qi, X Gao, S Xiao, W Chen Yingyong Huaxue 3 63 (1986); Chem. Abstr. 104 186 909 (1986) 21.H Gysling,M Tsutsui Adv. Organomet. Chem. 9 361 (1970) 22. J G Brennan, F G N Cloke, A A Sameh, A Zalkin J. Chem. Soc., Chem. Commun. 1668 (1987) 23. D M Anderson, F G N Cloke, P A Cox, N Edelstein, J C Green, T Pang, A A Sameh, G Shalimoff J. Chem. Soc., Chem. Commun. 53 (1989) 24. F G N Cloke, K A E Courtney, A A Sameh, A C Swain Polyhedron 8 1641 (1989) 25. W A King, T J Marks, D M Anderson, D J Duncalf, F G N Cloke J. Am. Chem. Soc. 114 9221 (1992) 26. WLamberts,HLueken,UElsenhans Inorg. Chim. Acta 121 81 (1986) 27. F G N Cloke, K Khan, R N Perutz J. Chem. Soc., Chem. Commun. 1372 (1991) 28. Ren Jingqing, H-D Amberger J. Mol. Struct. (Theochem) 236 231 (1991) 29. G Hong, F Schautz,M Dolg J. Am. Chem. Soc. 121 1502 (1999) 30.S Di Bella, G Lanza, I L Fragala, T J Marks Organometallics 15 3985 (1996) 31. M C Cassani, D J Duncalf,M F Lappert J. Am. Chem. Soc. 120 12958 (1998) 32. M C Cassani, Yu K Gun'ko, P B Hitchcock, M F Lappert J. Chem. Soc., Chem. Commun. 1987 (1996) 33. MC Cassani, Yu K Gun'ko, P B Hitchcock,MF Lappert, F Laschi Organometallics 18 5539 (1999) 34. M D Fryzuk, J B Love, S J Rettig J. Am. Chem. Soc. 119 9071 (1997) 35. R D Rogers, J L Atwood, A Emad, D J Sikora,M D Rausch J. Organomet. Chem. 216 383 (1981) 36. E B Lobkovskii, G L Soloveichik, B M Bulychev, A B Erofeev Zh. Strukt. Khim. 25 170 (1984) a 37. W J Evans, J W Grate, K R Levan, I Bloom, T T Peterson, R J Doedens, H Zhang, J L Atwood Inorg. Chem. 25 3614 (1986) 38. Y Chauvin, N Marchal, H Olivier, L Saussine J.Organomet. Chem. 445 93 (1993) 39. A A Trifonov, M N Bochkarev, H Schumann, J Loebel Angew. Chem., Int. Ed. Engl. 30 1149 (1991) 40. T A Basalgina, G S Kalinina,M N Bochkarev Metalloorg. Khim. 2 1142 (1989) b 41. M N Bochkarev, E A Fedorova, I M Penyagina, O A Vasina, A V Protchenko, S Ya Khorshev Metalloorg. Khim. 2 1317 (1989) b 42. M N Bochkarev, E A Fedorova, I M Penyagina, O A Vasina, S Ya Khorshev, A V Protchenko Metalloorg. Khim. 2 703 (1989) b 43. N S Emel'yanova, E A Fedorova, O A Vasina, M N Bochkarev Metalloorg. Khim. 4 1185 (1991) b 44. E A Fedorova, O A Vasina, Yu F Rad'kov, G S Kalinina, M N Bochkarev Metalloorg. Khim. 2 392 (1989) b 45. M N Bochkarev, E A Fedorova, N V Glushkova, A V Protchenko, O N Druzhkov, S Ya Khorshev Zh.Obshch. Khim. 65 1603 (1995) c 46. MI Terekhova, A V Garbar, L F Rybakova, E S Petrov Zh. Obshch. Khim. 56 1419 (1986) c 47. N S Emel'yanova,MN Bochkarev, G Shuman, J Loebel, L Esser Koord. Khim. 20 789 (1994) d 48. M N Bochkarev, A A Trifonov, E A Fedorova, N S Emelyanova, T A Basalgina, G S Kalinina, G A Razuvaev J. Organomet. Chem. 49. I L Fedushkin, M N Bochkarev, H Schumann, L Esser, 50. M N Bochkarev, I L Fedyushkin, R B Larichev Izv. Akad. Nauk, 51. M N Bochkarev, A A Fagin Izv. Akad. Nauk, Ser. Khim. 1200 372 217 (1989) G Kociok-Klhn J. Organomet. Chem. 489 145 (1995) Ser. Khim. 2573 (1996) e (1999) e794 52. N S Emel'yanova, A V Protchenko, E A Fedorova, O A Vasina, M N Bochkarev Metalloorg.Khim. 4 895 (1991) b 53. M N Bochkarev, I L Fedushkin, A A Fagin, H Schumann, J Demtschuk Chem. Commun. 1783 (1997) 54. J L Atwood, W E Hunter, R D Rogers, J Holton, J McMeeking, R Pearce, M F Lappert J. Chem. Soc., Chem. Commun. 140 (1978) 55. H J Heeres, J Renkema,M Booij, A Meetsma, J H Teuben Organometallics 7 2495 (1988) 56. G Jeske, H Lauke, H Mauermann, P N Swepston, H Schumann, T J Marks J. Am. Chem. Soc. 107 8091 (1985) 57. L F Rybakova, M I Terekhova, O P Syutkina, A V Garbar, E S Petrov Zh. Obshch. Khim. 56 2162 (1986) c 58. J J Zhang, J C Bao, M Z Bei Chin. Sci. Bull. 38 213 (1993); Chem. Abstr. 119 159 358 (1993) 59. E N Kirillov, A A Trifonov, S E Nefedov, I L Eremenko, F T Edelmann, M N Bochkarev Z. Naturforsch., B Chem.Sci. 54 1379 (1999) 60. K Rossmanith, Monatsh. Chem. 110 1019 (1977) 61. M D Fryzuk, L Jafarpour, F M Kerton, J B Love, S J Rettig Angew. Chem., Int. Ed. Engl. 39 767 (2000) 62. MN Bochkarev, I L Fedushkin, V K Cherkasov, V I Nevodchikov, H Schumann, F H Gorlitz Inorg. Chim. Acta 201 69 (1992) 63. I L Fedushkin, V I Nevodchikov, V K Cherkasov,MN Bochkarev, H Schumann, F Girgsdies, F H Glrlitz, G Kociok-Klhn, J Pickardt J. Organomet. Chem. 511 157 (1996) 64. A V Protchenko, E A Fedorova,M N Bochkarev, G Shuman, J Loebel, G Kotsiok-ken Izv. Akad. Nauk, Ser. Khim. 2027 (1994) e 65. A V Protchenko, L N Zakharov,M N Bochkarev, Yu T Struchkov J. Organomet. Chem. 447 209 (1993) 66. W J Evans, A L Wayda, W E Hunter, J L Atwood J.Chem. Soc., Chem. Commun. 292 (1981) 67. H Schumann, W Genthe, N Bruncks, J Pickardt Organometallics 1 1194 (1982) 68. M N Bochkarev, A V Protchenko, L N Zakharov, Yu T Struchkov Metalloorg. Khim. 5 716 (1992) b 69. A V Protchenko, L N Zakharov, G K Fukin, Yu T Struchkov M N Bochkarev Izv. Akad. Nauk, Ser. Khim. 993 (1996) e 70. A V Protchenko, O G Almazova, L N Zakharov, G K Fukin, Yu T Struchkov,M N Bochkarev J. Organomet. Chem. 536-537 457 (1997) 71. T Arakawa, S Shimada, G-Y Adachi, J Shiokawa Inorg. Chim. Acta 145 327 (1988) 72. W J Evans, S L Gonzales, J W Ziller J. Am. Chem. Soc. 116 2600 (1994) 73. H Olivier, Y Chauvin, L Saussine Tetrahedron 45 165 (1989) 74. L Saussine, H Oliver, D Commerreuc, Y Chauvin New. J. Chem. 12 13 (1988) 75.Y Chauvin, H Olivier, L Saussine Inorg. Chim. Acta 161 45 (1989) 76. D M Roitershtein, L F Rybakova, E S Petrov Metalloorg. Khim. 3 559 (1990) b 77. E S Petrov,M I Terekhova, D M Roitershtein Metalloorg. Khim. 1 474 (1988) b 78. D M Roitershtein, L F Rybakova, E S Petrov Dokl. Akad. Nauk SSSR 315 1393 (1990) f 79. K-H Thiele, S Bambirra, H Schumann, H Hemling J. Organomet. Chem. 517 161 (1996) 80. D M Roitershtein, A M Ellern, M Yu Antipin, L F Rybakova, Yu T Struchkov, E S Petrov Mendeleev Commun. 118 (1992) 81. DMRoitershtein, L F Rybakova, E S Petrov, AM Ellern, M Yu Antipin, Yu T Struchkov J. Organomet. Chem. 460 39 (1993) 82. K-H Thiele, S Bambirra, J Sieler, S Yelonek Angew. Chem. 110 3016 (1998) 83. WJ Evans, TAUlibarri, JWZiller J.Am. Chem. Soc. 112 2314 (1990) 84. W J Evans, R A Keyer, J W Ziller Organometallics 12 2618 (1993) 85. I L Fedyushkin, T V Petrovskaya, F Girgsdis, V I Nevodchikov, R Vaiman, G Shumann, M N Bochkarev Izv. Akad. Nauk, Ser. Khim. (11) (2000) (in the press) e 86. P L Arnold, F G N Cloke, J F Nixon Chem. Commun. 797 (1998) 87. P L Arnold, F G N Cloke, P B Hitchcock, J F Nixon J. Am. Chem. Soc. 118 7630 (1996) 88. M N Bochkarev, I M Penyagina, L N Zakharov, Yu F Rad'kov, E A Fedorova, S Ya Khorshev, Yu T Struchkov J. Organomet. Chem. 378 363 (1989) M N Bochkarev 89. E A Fedorova, A A Trifonov, E N Kirillov,M N Bochkarev Izv. Akad. Nauk, Ser. Khim. 947 (2000) e 90. A A Trifonov, E N Kirillov, Yu A Kurskii, M N Bochkarev Izv. Akad. Nauk, Ser. Khim. 742 (2000) e 91. A A Trifonov, E N Kirillov, A Fischer, F T Edelmann, M N Bochkarev Chem. Commun. 2203 (1999) 92. E A Fedorova, N V Glushkova,M N Bochkarev, G Shuman, Kh Khemling Izv. Akad. Nauk, Ser. Khim. 2101 (1996) e 93. MN Bochkarev, V V Khramenkov, Yu F Rad'kov, L N Zakharov, Yu T Struchkov J. Organomet. Chem. 429 27 (1992) 94. E A Fedorova, A A Trifonov,M N Bochkarev, F Girgsdies, H Schumann Z. Anorg. Allg. Chem. 625 1 (1999) 95. MN Bochkarev, V V Khramenkov, Yu F Rad'kov, L N Zakharov, Yu T Struchkov J. Organomet. Chem. 421 29 (1991) 96. MN Bochkarev, V V Khramenkov, Yu F Rad'kov, L N Zakharov, Yu T Struchkov J. Organomet. Chem. 408 329 (1991) 97. M N Bochkarev, I L Fedushkin, H Schumann, J Loebel J. Organomet. Chem. 410 321 (1991) 98. M N Bochkarev, A A Trifonov, F G N Cloke, C I Dalby, P T Matsunaga, R A Andersen, H Schumann, J Loebel, H Hemling J. Organomet. Chem. 486 177 (1995) 99. I L Fedushkin, T V Petrovskaya, F Girgsdies, R D KoÈ hn, M N Bochkarev, H Schumann Angew. Chem., Int. Ed. Engl. 38 2262 (1999) 100. M N Bochkarev, A V Protchenko, L N Zakharov, G K Fukin, Yu T Struchkov J. Organomet. Chem. 501 123 (1995) 101. O V Andreev, M N Bochkarev, N M Volodin, T V Nekrasova, A V Protchenko Izv. Akad. Nauk, Ser. Khim. 1361 (1993) e 102. MN Bochkarev, I L Fedushkin, V I Nevodchikov, V K Cherkasov, H Schumann, H Hemling, R Weimann J. Organomet. Chem. 524 125 (1996) 103. T V Petrovskaya, I L Fedyushkin, V I Nevodchikov, M N Bochkarev, N V Borodina, I L Eremenko, S E Nefedov Izv. Akad. Nauk, Ser. Khim. 2341 (1998) e 104. MNBochkarev, I L Fedushkin, V I Nevodchikov,A V Protchenko, H Schumann, F Girgsdies Inorg. Chim. Acta 280 138 (1998) 105. I L Fedyushkin, M N Bochkarev Izv. Akad. Nauk SSSR, Ser. Khim. 1470 (1993) e 106. D M Roitershtein, L F Rybakova, E S Petrov Zh. Obshch. Khim. 66 1573 (1996) c 107. T V Petrovskaya, I L Fedyushkin,M N Bochkarev, G Shumann, R Veimann Izv. Akad. Nauk, Ser. Khim. 1860 (1997) e a�Russ. J. Struct. Chem. (Engl. Transl.) b�Russ. J. Organomet. Chem. (Engl. Transl.) c�Russ. J. Gen. Chem. (Engl. Transl.) d�Russ. J. Coord. Chem. (Engl. Transl.) e�Russ. Chem. Bull. (Engl. Transl.) f�Dokl. Chem. Technol., Dokl. Chem. (En

ISSN:0036-021X    

出版商:RSC    

年代:2000

数据来源: RSC

摘要:

Russian Chemical Reviews 69 (9) 795 ± 820 (2000) Solid-phase synthesis of oligosaccharides and glycoconjugates N K Kochetkov Contents I. Introduction II. The methodology of synthesis of oligosaccharide chains on solid supports III. Model syntheses of di- and trisaccharides IV. Synthesis of higher oligosaccharides V. Synthesis of glycopeptides VI. Synthesis of phosphoglycans VII. Conclusion Abstract. regiospecific directed for supports polymeric of use The The use of polymeric supports for directed regiospecific assembly and glycopeptides oligosaccharides, of assembly of oligosaccharides, glycopeptides and phosphoglycans phosphoglycans is oligosaccharides of synthesis cases, some In considered. is considered. In some cases, synthesis of oligosaccharides on on supports synthesis corresponding the than efficient more is supports is more efficient than the corresponding synthesis in in solution, with favourably compare still cannot this although solution, although this cannot still compare favourably with the the latter.of syntheses solid-phase successful of Examples latter. Examples of successful solid-phase syntheses of complex complex oligosaccharides Special given. are glycoconjugates and oligosaccharides and glycoconjugates are given. Special emphasis emphasis is and linkers supports, optimum of choice the on laid is laid on the choice of optimum supports, linkers and glycosyla- glycosyla- tion references 105 includes bibliography The methods. tion methods. The bibliography includes 105 references. I.Introduction The explosive development of molecular biology is closely con- nected with recent advances in the chemistry and biochemistry of biopolymers including the development of efficient techniques for their targeted chemical synthesis. The recently developed approaches to the synthesis of biopolymers are based on essen- tially new principles for elongation of the polymeric chains including their assembly on polymeric supports. This method was first employed by Merrifield for the synthesis of peptides.1 After solution of numerous methodological prob- lems, this protocol, which is sometimes improperly referred to as `solid-phase synthesis', was successfully automated and became the main routine procedure in peptide synthesis.An enormous body of experimental evidence published in the past two-three decades point to the prominent role of biopoly- mers, such as polysaccharides and glycoconjugates (glycopro- teins, glycolipids) in vital processes. These polymers are indispensable participants in many biological processes (e.g., antigen ± antibody, enzyme ± substrate and ligand ± receptor inter- actions) as well as in some other interactions of carbohydrates with proteins.2, 3 The first attempts aimed at synthesising oligosaccharide chains on solid supports were undertaken in the early 1970's;4, 5 however, it became clear that rapid progress in this area was hardly possible, since chemical synthesis of oligosaccharides with N K Kochetkov N D Zelinsky Institute of Organic Chemistry, Russian Academy of Sciences, 117913 Moscow, Leninsky prosp.47, Russian Federation. Fax (7-095) 135 53 28. Tel. (7-095) 135 61 48 Received 28 December 1999 Uspekhi Khimii 69 (9) 869 ± 896 (2000); translated by R L Birnova #2000 Russian Academy of Sciences and Turpion Ltd DOI 10.1070/RC2000v069n09ABEH000568 795 795 798 801 811 817 819 a strictly determined sequence of monomeric units presented a serious problem due to the polyfunctionality of monosaccharides and the stereochemical ambiguity of intermonomeric glycosidic bond formation. At present, synthesis of oligosaccharides on polymeric sup- ports still cannot compete with the classical synthesis in solution, but its development is in progress and the methodology is improved. This is associated with the fact that in many cases syntheses on supports are more advantageous.Numerous exam- ples of successful application of the solid-phase approach in the synthesis of complex oligosaccharides and glycoconjugates are currently known. A vast body of experimental material accumulated thus far highlights the major problems and advances in solid-phase syn- thesis of carbohydrate-containing compounds and allows estima- tion of its possible application for realistic syntheses. It seems therefore expedient to sum up these data in order to inform all those who are concerned with this subject about the potential of the method and the problems it may involve. The review covers the publications appearing up to 1999 inclusive.II. The methodology of synthesis of oligosaccharide chains on solid supports The synthesis of oligosaccharides on solid supports includes immobilisation of the first unit from which elongation of the oligosaccharide chain begins on a specially selected optimum support followed by elongation of the oligosaccharide chain using specific methods for the formation of interglycosidic link- ages (viz., by glycosylation) and removal of the oligosaccharide molecules thus synthesised from the polymeric support without any sacrifice of their structures. The peculiarity of carbohydrate chemistry is manifested in all steps of the solid-phase oligosac- charide synthesis, which makes it different from peptide and nucleotide syntheses.1. Supports Oligosaccharide syntheses are carried out both on insoluble solid supports and on soluble polymers. In the former case, the reaction proceeds in heterogeneous media, in the latter case, in homoge- neous media. The obvious advantage of insoluble solid supports is that the isolation and purification of the reaction products after each step is achieved by simple filtration and washing with a solvent.796 Furthermore, the reaction rate can be increased if one of the reagents, which can be regenerated following the reaction, is taken in a large excess. The main disadvantage of insoluble supports is that the reaction occurs on a solid surface, which hampers the access of the reagent to the reaction centre and poses additional difficulties in regulation of regio- and stereoselectivity of the reaction.Syntheses of oligosaccharide chains anchored to soluble polymeric supports are partly devoid of this drawback but purification of the reaction product becomes more difficult. In this case, the polymer-bound reaction product has to be precipi- tated by addition of another solvent and purified by repeated precipitations. The use of an excess of a reagent for acceleration of the reaction is also undesirable, since in this case its complete removal, which is necessary to ensure the desired sequence of the monomeric units in the `growing' chain, faces difficulties. On the other hand, glycosylation in a homogeneous medium offers better opportunities for regulation of regio- and stereospecificity of the reaction. Currently, both types of supports for the synthesis of oligo- saccharides are employed; however, synthesis on soluble polymers (especially, synthesis of glycoconjugates) has acquired ever increasing importance in recent years. a.Insoluble solid supports The efficiency of solid-phase synthesis strongly depends on the physicochemical properties of solid supports, such as porosity and swelling capacity in different solvents. These characteristics are very important, since they determine the accessibility of immobi- lised molecules to the reagent present in the solution. The number of moles of the first monomeric unit from which elongation of the chain begins, which can be immobilised per g of the polymeric support (the so-called `loading'), is yet another factor of consid- erable practical importance.In the early period of solid-phase oligosaccharide synthesis, polystyrene-based resins prepared by copolymerisation of styrene with p-divinylbenzene were predominantly employed. Commer- cial resins of this type are now available as powders with particles having spherical shapes (`beads') and differing in the degree of cross-linking, porosity, swelling capacity and particle size.{ Modi- fied polystyrenes containing reactive chloromethyl (Merrifield's polymer), phenolic hydroxy groups,7 mercaptomethyl,8 sulfonyl- methyl,9 diphenyl- and diisopropylchlorosilyl,10 (3-mercapto- propyl)oxy(thio)methyl 11, 12 and some other groups are most commonly used as supports.The choice of a particular polystyr- ene resin as a solid support is determined by the strategy of a realistic synthesis, the mode of attachment of the first monomer to the solid phase and the structure of the linker. Sometimes, the role of a linker is played by the functional group of the polystyrene resin. A main disadvantage of polystyrene resins used in solid-phase oligosaccharide synthesis is their poor compatibility with biomo- lecules. For example, the yields of oligosaccharides in enzymic solid-phase synthesis catalysed by glycosyl transferases were low to such an extent that this method had to be abandoned as being impractical. Attempts were undertaken aimed at using polyacrylamide gels as supports.Prior to the use, the gels were modified, e.g., by incorporation of aminoethyl groups.13 Controlled-pore glasses (CPG) with a modified surface, e.g., with grafted 3-aminopropyl groups serving to attach the first monomer (see, e.g., Refs 14 and 15), are currently employed with increasing frequency. The surface of CPG-based supports can be modified by silanisation with prior zirconation.16 The use of CPG has a number of advantages, since they offer broad opportunities in the selection of supports of desired porosity and allow appro- priate modification of the surface. {A brief survey of polymeric substrates for solid-phase syntheses is given in Ref. 6. N K Kochetkov b. Soluble supports These types of supports have currently acquired wide popularity.The simplest soluble support is polyethylene glycol (PEG) (Mr *5.000 Da) in which one of the terminal hydroxy groups is methylated and the other one is free and serves to immobilise the first monomer (either directly or through a linker).17 Commer- cially available PEG is easily soluble in many solvents. Polymers of a hybrid type, containing both hydrophobic and hydrophilic fragments, have received considerable attention in the last few years. The synthesis on such materials is carried out in a `micromedium', i.e., under conditions mimicking those of a homogenous reaction, and at the same time allows easy isolation of the reaction products after each step. These polymers include a copolymer of acrylamide or its N-substituted derivatives with polyethylene glycol (PEGA) possessing terminal amino groups 18 and a copolymer of polystyrene with polyethylene glycol (Tenta- gel).19 Other more sophisticated copolymers are also used and will be discussed below. 2.Linkers In solid-phase syntheses of oligosaccharides, the first monomer is usually attached to the support through a bifunctional group (the so-called linker) rather than directly. The linker serves both to bind the monomer to the polymer and to make the monomer and the growing chain remote from the support in order to minimise the influence of the solid surface on the course of the reaction. The latter circumstance is especially important, since heterogeneity of the medium has a specific, sometimes rather obscure effect on the stereochemistry of the newly formed glycosidic linkage.In deciding upon a linker, allowance should be made for the possibility of selective cleavage of one of its linkages in the release of the oligosaccharide synthesised from the solid support without affecting labile glycosidic bonds. In early oligosaccharide syntheses, the first monosaccharide was immobilised on chloromethyl ± polystyrene without linkers. However, later the use of linkers differing in structures and chemical properties became common practice. It should be noted that in contrast with the conventional solid- phase peptide and oligonucleotide syntheses, diverse linkers are utilised in solid-phase syntheses of carbohydrates. The choice of a suitable linker for a realistic purpose is an indispensable step in the design of a synthesis.In the synthesis of glycopeptides, the peptide chain is often used as a linker which also removes the growing carbohydrate chain from the surface of the support. A more detailed description of linkers will be given further below in relation to particular syntheses. 3. Elongation of the oligosaccharide chain This step is crucial in oligosaccharide synthesis, therefore the choice of a method for the formation of glycosidic bonds between the monosaccharides (i.e., method of glycosylation) determines the overall strategy of the synthesis. The predetermined sequence of monosaccharides necessitates that the formation of glycosidic linkages is strictly regio- and stereospecific.A prerequisite for the preparation of homogeneous oligosaccharides with a predeter- mined structure, which should not require any additional purifi- cation (it is this that makes solid-phase synthesis advantageous in comparison with the classical synthesis in solution), is that the yields were close to quantitative in all steps. The lack of a versatile glycosylation procedure (optimal for all monosaccharides) makes the implementation of these requirements extremely difficult, which is the main reason for the relatively slow progress in the solid-phase synthesis of oligosaccharides. Two problems seem to be especially difficult to overcome. The first of them is that successful synthesis of any biopolymer consisting of 40 ± 50 monomeric units with a predetermined sequence (without deletions or `erroneous' units) requires that the yield in each elongation step is no less than 97% ± 98%.Meanwhile, even the most advanced glycosylation methods do not ensure such high yields, which in addition vary stronglySolid-phase synthesis of oligosaccharides and glycoconjugates depending on the nature of the monosaccharide and the type of the newly formed glycosidic linkage. The second problem is related to stereochemical ambiguity of the glycosylation reaction,{ which makes the oligosaccharide synthesis different in principle from the peptide and oligonucleotide syntheses, although it is known, for example, that the synthesis of the 1,2-trans-glycosidic bond is usually stereospecific.So far, the use of solid-phase synthesis is effective only for the preparation of relatively short oligosaccharides. Yet another problem inherent in the solid-phase synthesis of oligosaccharides is related to the polyfunctionality of monosac- charides. The formation of (1-2), (1-3), etc., intermonomeric gly- cosidic linkages requires temporary protection of hydroxy groups which are not involved in the glycosylation reaction. In the general case, synthesis of irregular oligosaccharides (viz., those in which the monosaccharide components are either different or identical but differ in the type of bonding) requires specific protection depending on the type of the glycosidic bond for each mono- saccharide. After the glycosylation step, the type of protection of hydroxy groups must be changed up to complete removal of the existing and introduction of new protecting groups.This sophisti- cated procedure makes oligosaccharide synthesis technically cum- bersome in comparison with peptide and oligonucleotide synthe- ses. Sometimes, as in the case of homopolysaccharide synthesis, elongation of the oligosaccharide chain involves the same set of procedures, but even in these cases the passage from attachment of one monomeric fragment to the other is complicated and auto- mation of the synthetic protocol presents a problem. In designing a general strategy for oligosaccharide synthesis with any combination of protecting groups, it is necessary, first of all, to define the order in which the oligosaccharide chain has to be elongated. The elongation is typically carried out in two direc- tions, viz., from the reducing end of the chain to the non-reducing one (routeA) or vice versa (route B) (Scheme 1).In the former case (route A), the monosaccharide immobilised on a solid support through a glycosidic hydroxyl is used as a glycosyl acceptor and Scheme 1 Route A: OR P OR RO O O OR OH O XDonor RO OR OR OR OR RO P O OR O O O OR RO OR Route B: OR OR RO P X O HO O OR O Acceptor OR RO OR OR OR RO P O O O OR O OR RO OR P �Polymeric support. {A critical review of thetereospecificity of the main glycosylation reactions is given in Ref. 20. 797 the second monosaccharide present in the solution plays the role of a glycosyl donor. In the latter case (route B), the first mono- saccharide is attached to the support through one of the hydroxy (non-glycosidic) groups and serves as a glycosyl donor, whereas the second monosaccharide remains in the solution and fulfils the function of a glycosyl acceptor.Naturally, the protecting groups and manipulations with them in each stage of the solid-phase synthesis are different in these two routes. If the synthesis is to be performed via the route A, it is in the terminal monosaccharide of the glycosyl acceptor that the protection of the hydroxy groups should be changed or selectively removed, whereas in the case of the route B it is necessary to introduce or establish a group activating the glycosidic centre in the terminal monosaccharide of the glycosyl donor.Obviously, the reagents present in a solution, e.g., a glycosyl acceptor (route B) or a glycosyl donor (route A), should also contain appropriate protecting groups. Glycosylation can be carried out by chemical methods or using specific enzymes (most often, glycosyl transferases), which changes the whole strategy of the solid-phase synthesis and eliminates the problem of temporary protection of the monosac- charide units. a. Chemical glycosylation methods The pioneering studies of solid-phase oligosaccharide synthesis utilised classical glycosylation with glycosyl halides (predomi- nantly, per-O-acetylglycosyl bromides) } under conditions of the Koenigs ± Knorr or the Helferich reaction 4 as well as the ortho- ester glycosylation method.22 However, insufficiently high yields and the lack of complete stereospecificity in the formation of glycosidic bonds devaluate these methods for use in solid-phase synthesis.Other glycosylation methods based on the use of thio- glycosides (in the presence of activators), glycosyl tosylates,23 glycosyl sulfoxides,24 etc., as glycosyl donors, have also been explored. Glycosylation with sugar O-trichloracetimidates proved to be especially efficient.25 This procedure provides sufficiently high yields for a wide variety of monosaccharides. Unfortunately, the stereospecificity of this reaction and the effect of the configuration of the trichloroacetimidate on the steric outcome of the reaction still present a problem; therefore, the practical utility of this reaction in solid-phase synthesis of relatively high-molecular- weight oligosaccharides still remains to be elucidated. Recently, Danishefsky et al.26 proposed to use the so-called glycal approach 25 for the synthesis of relatively short oligosac- charides and glycoconjugates with more complex structures where 1,2-anhydrosugars were used as donors.This procedure has not received wide application in conventional oligosaccharide synthe- ses yet, but has been used to develop a convenient scheme for a solid-phase synthesis (see below). b. Enzymic glycosylation The elongation of the oligosaccharide chain immobilised on a support can also be carried out by biochemical methods,} i.e., using enzymes which transfer a monosaccharide residue from the corresponding nucleoside diphosphate sugar (NPPSug) to the end of the growing chain (Scheme 2).The main advantages of the enzymic approach are good yields and, which is more important, high regio- and stereospecificity of Scheme 2 O O O O P P NPPSug, OH OSug glycosyl transferase N is a nucleoside. } An overview of various glycosylation methods is given in Ref. 21. } The biosynthesis of oligosaccharides is described in a review.27798 glycosylation. Moreover, in this case no protection of the mono- saccharide hydroxy groups is necessary, which significantly sim- plifies the synthetic procedure. However, sometimes the high specificity of enzymic glycosylation may have a negative effect, viz., where different monosaccharides are to be introduced or different types of glycosidic bonds are to be formed.The reason is that in these cases each elongation step would require its own, specific enzyme because of the normally narrow specificities of glycosyl transferases. Moreover, the specificities of enzymes can depend both on the nature of the monosaccharide donor and the monosaccharide acceptor and on the nature of adjacent mono- saccharides in the fragment of the oligosaccharide chain synthesised. Yet another characteristic feature of enzymic glycosylation which should be taken into consideration in conducting solid- phase synthesis is a noticeable dependence of the outcome of the reaction on the nature of the support and the structure of the linker. This can be explained by the fact that the accessibility of the reaction centre of the oligosaccharide chain, which is immobilised on a solid support, for the enzyme, which itself represents a macromolecule, and the formation of an enzyme ± support com- plex are impeded in comparison with enzymic reactions carried out in a solution.Therefore, the requirements for supports and linkers, which enable contact between the reacting molecules, are more stringent. This is the main reason why soluble and con- formationally flexible polymers and structurally complex linkers are employed where enzymic methods are used for oligosacchar- ide synthesis on polymeric supports.(As noted above, glycosyl transferases cannot be used for enzymic oligosaccharide syntheses on polystyrene-based resins due to incompatibility of these structurally different macromolecules). Although the first attempts to use enzymes in solid-phase syntheses of oligosaccharide date back to the 1980's, it was not until recently that this method received wide application. III. Model syntheses of di- and trisaccharides The pioneering studies on solid-phase oligosaccharide synthesis were aimed at elucidating the feasibility of this technique for preparing this class of compounds. The simplest di- and trisac- charides were the first to be synthesised. Thus Frechet and Scheurch 4 carried out the synthesis of D-glucopyranosyl-(1-6)-D-glucose and the corresponding trisac- charide with 1-6-bonds using elongation of the oligosaccharide chain immobilised on an insoluble support from the reducing end (Scheme 3). Treatment of a polystyrene resin modified by intro- duction of allyl alcohol residues 28 with 2,3,4-tri-O-benzyl-6-O-(p- nitrobenzoyl)-D-glucopyranosyl bromide (1, R=NO2 ) or its p-methoxybenzoyl analogue (1, R=MeO) resulted in immobili- sation of a protected glucose residue after which the acyl protec- tion of the hydroxy group at position 6 was removed.The glucose residues in the polymer 2 were glycosylated with the same bromides. The immobilised disaccharide residues in the polymer 3 were subjected to repeated deacylation and glycosylation to obtain a trisaccharide.The polymer 3 containing disaccharide residues (or its analogue containing trisaccharide residues) was subjected to ozonolysis in order to detach the reaction product, viz., the corresponding di- or trisaccharide. After reduction, the reaction products were isolated as the corresponding 2-hydroxy- ethyl glycosides. The disaccharide 4 represented an anomeric mixture of derivatives of a- and b-D-glucopyranosyl-(1-6)-D- glucose (i.e., isomaltose and gentiobiose). Thus, the solid-phase synthesis was not stereospecific and did not differ in this respect from the corresponding reaction in solution. An alternative scheme for the solid-phase synthesis of b-D- glucopyranosyl-(1-6)-D-glucose (gentiobiose) was proposed by Guthrie et al.5 who used an orthoester glycosylation method { { For a review on the orthoester glycosylation method, see Ref.22. N K Kochetkov Scheme 3 RC6H4COO O OBn P CH CHCH2OH + BnO Br OBn 1 RC6H4COO O OBn P OCH2CH CH BnO OBn HO O 1 OBn P OCH2CH CH BnO OBn 2 RC6H4COO O OBn O O BnO OBn P OBn OCH2CH CH BnO OBn 3 RC6H4COO O OBn O O BnO OBn OCH2CH2OH OBnBnO 4 OBn R=NO2 , MeO. for the elongation of the chain (Scheme 4). The linear polymer 5, which contained glucose residues linked through a 6-hydroxy group, was prepared by copolymerisation of styrene with 1,2,3,4- tetra-O-acetyl-6-O-(p-vinylbenzoyl)-D-glucopyranose or its 6-O- (p-vinylphenyl)sulfonyl analogue. The glucose residues immobi- lised on the polymer were converted into the corresponding 1,2- orthoester residues.The resulting polymer 6 was used as a glycosyl donor. Its reaction with methyl 2,3,4-tri-O-benzoyl-a-D-glucopyr- Scheme 4 COOCH2O OAc OAc CH CH2 CHPh CH2 AcO OAc 5 HO O OBz COOCH2 BzO OMe OBz O OAc CH CH2 CHPh CH2 O AcO OMe O 6 Me CH COOCH2 CH2 O CHPh O O OAc CH2 OBz AcO OMe OBz OAc BzO 7Solid-phase synthesis of oligosaccharides and glycoconjugates anoside under standard conditions of orthoester glycosylation afforded the gentiobiose derivative 7 (yield 64%) which was detached from the polymer by treatment with sodium methoxide. Since glycosylation of primary alcohols with orthoesters is stereo- specific, this solid-phase synthesis was also stereospecific and yielded exclusively gentiobiose.These two studies demonstrated that glycosidic bonds can be formed in heterogeneous reactions and that the value of solid- phase oligosaccharide syntheses does not exceed that of conven- tional syntheses in solution. Further developments in this field were aimed at overcoming some of the methodological problems, such as the choice of supports, construction of linkers suitable for support immobilisation, intermittent treatment of the polymer, etc. The applicability of the solid-phase approach to the synthesis of other simplest oligosaccharides was studied in parallel. Thus the solid-phase synthesis of D-glucopyranosyl-(1-3)-D- glucose 29 was carried out with the aim of examining the possibility of glycosylation of a less reactive secondary hydroxy group (Scheme 5).To this end, 2,4,6-tri-O-acetyl-D-glucopyranose was immobilised on a polystyrene-based resin through a linker (suc- cinic acid residue). The glycosyl acceptor 8 thus immobilised was glycosylated with 2,3,4,6-tetra-O-acetyl-a-D-glucopyranosyl bro- mide under conditions of the Helferich reaction (48 h, 30 8C). The polymer 9 was separated and washed, after which glycosylation was repeated two more times. The acylglycosidic linkage in the polymer 9 was cleaved with hydrazinium acetate. The yield of the disaccharide, which represented a mixture of an a- (nigerose) and a b-isomer (laminaribiose) (*3 : 2), was 81%. Scheme 5 OAc O OAc OAc AcO O Br OAc OH OC(CH2)2CO P O AcO O OAc 8OAc O OAc O OAc AcO O OAc P OC(CH2)2CO O AcO O OAc 9 Glycosylation with 3,4,6-tri-O-acetyl-2-O-benzyl-a-D-gluco- pyranosyl bromide was stereospecific; the yield of the a-linked disaccharide was 55%.Thus, the feasibility of glycosylation of immobilised monosaccharides at a less reactive secondary hydroxy group under conditions of a solid-phase synthesis has been demonstrated. In more recent studies, a succinic acid-based linker has been applied for chemical and enzymic syntheses of oligosaccharides on soluble supports. Krepinsky et al.30 were the first to use this linker for immobilisation of the first monosaccharide on the mono- methyl ether of PEG.Glycosylation of the thus immobilised monosaccharide derivatives (e.g., methyl galactopyranoside, methyl 2-deoxy-2-phthalimido-b-D-glucopyranoside or 1,6-anhy- dro-b-D-glucopyranose) was carried out with glycosyl bromides or O-glycosyltrichloroacetimidates in the presence of the corre- sponding promoters. The resulting disaccharides { were easily split off the support after completion of the synthesis. At the same time, there were some difficulties associated with the separation of the immobilised product from an excess of the glycosylating reagent { For the `methathesis' of oligosaccharides occurring in this glycosylation technique, see Ref. 31. 799 present in the solution which, as was mentioned above, is a disadvantage of the use of soluble polymers as supports.A similar approach was used in the synthesis of b-D-glucopyr- anosyl-(1-6)-D-galactose. In this case, the glycosyl acceptor (viz., selectively protected methyl a-D-galactopyranoside with a free primary hydroxy group) was immobilised on a soluble support (Tentagel) through an OH group at position 2 or 3 using succinic acid as the linker. Glycosylation of the immobilised monosac- charide with O-(2,3,4,6-tetra-O-acetyl-a-D-glucopyranosyl) tri- chloroacetimidate was stereospecific; the yield of the disaccharide liberated from the polymer by treatment with aque- ous ammonia varied from 50% to 70% depending on the promoter.19 In order to facilitate detachment of oligosaccharides from the supports, it was proposed 8 to use the thioglycoside linkage for the binding of the first monosaccharide.This linkage was formed either in a reaction of 1-thio sugars with chloromethylated polystyrene or in a reaction of acylglycosyl halides with polystyr- ene containing HSCH2 groups. A model synthesis was that of gentiobiose. The resulting disaccharide was split off the polymer as benzyl glycoside by boiling in benzene containing methyl iodide and benzyl alcohol. Zehavi et al.32 ± 34 proposed to use 4-hydroxymethyl-3-nitro- benzoic and 4-hydroxymethyl-6-methoxy-3-nitrobenzoic acid derivatives as photosensitive linkers. The hydroxy groups of these linkers are glycosylated with monosaccharides, while the carboxy group is used for the binding to the polymeric support.The presence of an o-nitrobenzyl alcohol fragment enables smooth decomposition of these linkers upon photolysis in neutral media, which results in the liberation of the oligosaccharides. These linkers were used for the first time in the synthesis of a-D-glucopyranosyl-(1-6)-D-glucose (isomaltose) (Scheme 6).32 The 2,3,4-tri-O-benzyl-6-O-(p-nitrobenzoyl)-D-glucopyranose residue was immobilised as an O-glycoside on a polystyrene- based resin (10) through the linker. The 6-OH group was depro- tected by treatment with MeONa after which the polymer 11 containing partially substituted glucose residues was glycosylated with 2,3,4-tri-O-benzyl- 6-O-(p-nitrobenzoyl)-a-D-glycopyrano- syl bromide. In order to liberate the target product, which represents an isomaltose derivative, the polymer 12 was irradiated in dioxane for 32 h (yield 80%).Scheme 6 P CO MeO COO O2N MeONa/MeOH O OBn NO2 BnO CH2 O OBn 10 COO O2N CO P O OBn MeO HO BnO Br OBn O OBn NO2 BnO CH2 11 O OBn COO O2N O OBn P CO MeO BnO O OBn O OBn NO2 BnO CH2 O OBn 12800 In more recent studies, the photosensitive linker based on 4-hydroxymethyl-3-nitrobenzoic acid was used for the enzymic synthesis of oligosaccharides on other supports. The reaction of 2,3,4,6-tetra-O-acetyl-a-D-glucopyranosyl bromide with alkyl 4-hydroxymethyl-3-nitrobenzoate followed by O-deacetylation and saponification gave the glycoside 13 (Scheme 7), which was then immobilised on modified polyacrylamide containing 2-ami- noethyl groups 13 or on modified polyvinyl alcohol bearing amino groups.33 The glucose residues in the resulting polymers 14 were linked by glycosidic bonds and served as acceptors in enzymic syntheses.These polymers were incubated with galactosyl trans- ferase in the presence of uridine diphosphate galactose (UDPGal) as a galactosyl donor. b-D-Galactopyranosyl-(1-4)-D-glucose (lactose) formed was split off by irradiation of the polymer 15 in aqueous solution for 20 h. This synthesis was the first to demon- strate the usefulness of glycosyl transferase in regio- and stereo- specific synthesis of a disaccharide using glycosylation of an immobilised monosaccharide residue devoid of protecting groups. Scheme 7 CONH P COOH NO2 NO2 HO HO UDPGal, O CH2 O CH2 O O OH OH galactosyl transferase HO HO OH OH 14 13 CONH P NO2 HO O CH2 O OH HO O HO O OH OH 15 OH Similarly, the glycosylation of 4-hydroxymethyl-3-nitroben- zoic acid immobilised on polyacrylamide with cellobiose 1,2- orthoester and subsequent deacetylation afforded a polymer containing cellobiose residues with free hydroxy groups.These played the role of acceptors in the enzymic reaction with galacto- syl transferase and UDPGal which gave the trisaccharide b-Galp(1-4)-b-Glcp(1-4)Glc. Cellobiose immobilised on polyac- rylamide was also used in the synthesis of oligosaccharides containing N-acetyl-b-D-glucosamine residues which were trans- ferred from chitin in the presence of lysozyme.13 Maltose [a-Glcp(1-4)Glc], maltotriose [a-Glcp(1-4)-a-Glcp(1- 4)Glc] and panose [aGlcp(1-6)-a-Glcp(1-4)Glc] were immobilised on N-(2-aminoethyl)polyacrylamide through the same photosen- sitive linker in order to investigate the specificity of glycogen synthase, which transfers glucose residues from uridine diphos- phate glucose (UDPG) to glucooligosaccharides, and to elucidate the role of primers in the biosynthesis of glycogen.34 Oligosac- charide derivatives were attached to 4-hydroxymethyl-3-nitro- benzoic acid; the resulting glycosides were deacylated and immobilised on a polymer through the carboxy group of the linker by treatment with water-soluble carbodiimide.Reactions of these immobilised acceptors with UDPG in the presence of glycogen synthase shed additional light on some essential details of the mechanism of glycogen biosynthesis, e.g., the role of primers, the appearance of branching, etc.In the enzymic synthesis ofN-acetyllactosamine [2-acetamido- 4-O-(b-D-galactopyranosyl)-2-deoxy-D-glucose],35 the N-acetyl- glucosamine residue was immobilised on a water-soluble copoly- mer of acrylamide and N-acryloyloxysuccinimide modified by incorporation of 2-aminoethyl groups through its glycosidic centre and a photosensitive linker (viz., 4-hydroxymethyl-3-nitro- benzoic acid). After O-deacetylation of the glucosamine residue, the polymer was incubated with UDPG in the presence of the UDPG epimerase ± galactosyl transferase system which provided epimerisation of the glucose residue and transfer of the galactose residue to the acceptor. The N-acetyllactosamine thus formed was liberated from the polymer by photolysis.Copolymerisation of acrylamide with its derivative 16 (Scheme 8) gave rise to polyacrylamide 17 bearing glucosamine residues glycosidically linked to a benzyl-type linker. After O-deacetylation, the polymer 18 was incubated with UDPGal and galactosyl transferase. The resulting N-acetyllactosamine 19 was split off the polymeric carrier by hydrogenolysis in the presence of Pd/C.36 AcO NHCOCH CH2 OCH2 O OAc AcO 16 NHAc AcO OCH2 O OAc AcO NHAc 17 HO OCH2 O OH HO NHAc 18 HO OCH2 O OH HOHO O O NHAc OH OH 19 These examples illustrate that the use of enzymes in solid- phase synthesis of oligosaccharides is promising.The efficacy of enzymic elongation of oligosaccharide chains depends on the distance between the monosaccharide acceptor and the surface of the support. This distance and, as a consequence, the optimum size of the linker, are determined by the molecular mass of the enzyme. If the galactose residue is transferred by galactosyl transferase, it can be attached to the polymer by a chain contain- ing only two methylene units, whereas successful glucose transfer effected by glycogen synthase with a higher molecular mass requires that the chain contained at least six methylene units. The solid-phase syntheses of di- and trisaccharides described above do not have any obvious advantages over the correspond- ing syntheses in solution.However, they provided a clue to the solution of some urgent methodological problems in order to pass to targeted synthesis of fragments of homo- and heteropolysac- charides and glycopeptides and contributed also to progress in the combinatorial chemistry of carbohydrates. N K Kochetkov Scheme 8 NHCO P UDPGal, NHCO P galactosyl transferase NHCO PSolid-phase synthesis of oligosaccharides and glycoconjugates IV. Synthesis of higher oligosaccharides Several solid-phase syntheses of regular homooligosaccharides containing up to 10 monosaccharide units have been carried out to date. Although some of compounds of this type can be prepared by polymerisation of monosaccharide derivatives,37, 38 their syn- thesis on solid supports made it possible to develop a standard procedure for stepwise elongation of an oligosaccharide chain suitable for concrete method of glycosylation.Gluco- and mannooligosaccharides were prepared by the Schmidt trichloroacetimidate method,39 which is the most popu- lar. The synthesis of (1-6)-D-glucooligosaccharides is shown in Scheme 9. The polymer 20 modified with 3-mercaptopropylthio- methyl groups was prepared by the reaction of the Merrifield polymer with propane-1,3-dithiol. The substituted glucose residue was attached to the free thiol group of the polymer to form the thioglycosidic bond using O-(6-O-acetyl-2,3,4-tri-O-benzyl-D- glucopyranosyl) trichloroacetimidate (21). The 6-O-acetyl group in the resulting polymer 22 was removed and the polymer 23 formed was treated with a threefold excess of the imidate 21 in the presence of trimethylsilyl triflate.The O-acetyl group in the polymer 24 was removed again, the polymer 25 was glycosylated by treatment with the glycosyl donor 21, and the whole procedure was repeated as many times as required. After termination of the synthesis, the oligosaccharide was split off from the polymer 26 with a thiophilic reagent [e.g., dimethyl(methylthio)sulfonium triflate] in the presence of a base. This approach was used in the synthesis of di-, tri-, tetra- and pentasaccharides. The yields of the products in each step of the chain elongation were determined by the increase in the polymer mass.In addition, this reaction was monitored by a mass-spectrometric procedure which had been specially developed for this purpose. An aliquot of the polymer was treated with dimethyl(methylthio)sulfonium tetrafluorobo- rate in a dichloromethane ± methanol mixture in the presence of Scheme 9 OAc O NH OBn BnO O CCCl3 21 OBn P CH2S(CH2)3SH 20 OR O OBn P S(CH2)3SCH2 BnO OBn 22: R =Ac 23: R = H OR O 21 OBn O O BnO OBn P OBn S(CH2)3SCH2 BnO OBn 24: R = Ac 25: R =H OR O OBn O O BnO OBn O OBn O BnO OBn P OBn S(CH2)3SCH2 BnO n OBn 26 the HuÈ nig base and the oligosaccharides liberated were analysed by mass spectrometry, which made it possible to judge the completeness of glycosylation in each step.According to mass spectrometric data (MALDI ± TOF), the yield was 95%. Each glycosylation step afforded a mixture of a- and b-anom- ers in an a : b ratio of 1 : 1; thus, the pentamer represented a mixture of 32 diastereomers. Hence, this methodology is well- suited for combinatorial syntheses. A similar approach was used in the synthesis of a-1,2-linked oligomannosides.11 2-O-Acetyl-3,4,6-tri-O-benzyl-D-mannopyra- nose residues were incorporated into the polystyrene resin CH2OCH2CH2CH2SH prepared in an analogous way. After P deacetylation and glycosylation of the deacetylated product with O-(2-O-acetyl-3,4,6-tri-O-benzyl-D-mannopyranosyl) trichloro- acetimidate and reiteration of the reaction cycle, the immobilised oligomannosides 27 (up to the hexasaccharide) were split off the support by treatment with N-bromosuccinimide in THF±MeOH in the presence of 2,6-di-tert-butylpyridine.The elongation of the chain was monitored by mass spectrometry. The yields of di-, tri- and tetramannosides were 75%, 54% and 34%, respectively. OBn OAc O BnOBnO OBn O O BnOBnO OBn O O BnOBnO OBn BnOBnOBnOBnO27 Glycosylation of the first mannose residue with an L-fucose analogue instead of O-mannosyl trichloroacetimidate yields the disaccharide 28 containing an a-L-fucosidic (1,2-cis-glycosidic) bond. BnOBnO OBn O Me BnO O O BnOBnO OMe 28 This approach was also used in the synthesis of a well-known core of N-glycoproteins, which represents a branched pentasac- charide (Scheme 10).12 Mannose residues were immobilised on the same polymer, viz., P CH2OCH2CH2CH2SH, by treatment with O-(2,4-di-O-benzyl-3,6-di-O-benzoyl-a-D-mannopyranosyl) trichloroacetimidate. Both benzoyl groups in the resulting poly- mer were removed, and the diol 29 formed was glycosylated with an excess of O-(2-O-acetyl-3,4,6-tri-O-benzyl-a-D-mannopyrano- syl) trichloroacetimidate (30).The acyl groups in the polymer 31 were removed again, and the immobilised trisaccharide 32 was glycosylated with O-(3,4,6-tri-O-acetyl-2-deoxy-2-trichloroeth- oxycarbonylamino-a-D-glucopyranosyl) trichloroacetimidate 801 O O OBn O O P S(CH2)3OCH2802BzO OBn O BnO S P BzOBnO OHO BnOBnO O OBn O BnO O O BnO BnOBnO 32 OH Teoc is trichloroethoxycarbonyl. (33), which afforded a polymer with an immobilised pentasac- charide (34). The thioglycosidic linkage was cleaved with N-bro- mosuccinimide in the presence of methanol and 2,6-di-tert- butylpyridine resulting in the pentasaccharide 35 in 20% yield; hence, the yield of glycosylation was 85% in each step.The yields of the reaction products were monitored by mass spectrometry, while the stereochemistry of the glycosidic bonds was established by NMR spectroscopy. Analogous syntheses of a-(1-2)-linked mannooligosacchar- ides were performed on other supports as well, e.g., on porous glasses with grafted mercaptopropyl groups.40 The 3,4,6-tri-O- benzyl-2-O-phenoxyacetyl-D-mannose residues were attached to the support modified by treatment with 3-mercaptopropyltrime- thoxysilane.The phenoxyacetyl group in the immobilised saccha- ride 36 was removed with guanidine and the next mannose residue was attached to the liberated hydroxy group after which the whole reaction cycle was reiterated. The oligomannoside synthesised was split off the support by treatment withN-bromosuccinimide in the presence of methanol and 2,6-di-tert-butylpyridine. Mass-spec- trometric analysis revealed that the resulting trisaccharide con- tained only trace amounts of the mono- and the disaccharide, which suggests that glycosylation occurred with high yields in each step. OBn OCOCH2OPh O BnOBnO S(CH2)3Si 36 Synthesis of this type of oligomannosides have also been carried out on a soluble support, viz., monomethyl ether of polyethyleneglycol.17 1,4-Bis(hydroxymethyl)benzene } has been used as a linker and 2-O-acetyl-3,4,6-tri-O-benzyl-D-mannopyr- } It should be noted that this type of linker has also been used in some other syntheses on soluble supports, e.g., in the synthesis of model disaccharides.BnO OAc BnO BnO HO OBn O BnO S P HO 29 HNTeoc AcO AcOAcO AcO BnO O BnO NH BnO CCl3 AcO AcOHNTeoc O C 33 S P BnO BnOBnOO AcO AcO AcO CPG O OO N K Kochetkov Scheme 10 BnO OAc O BnO O NH BnO O C O CCl3 30 OBn O BnO S P O O BnO BnOBnO 31 OAc HNTeoc AcO AcOAcO O O O O BnO O O BnOBnO O O OBn O OBn O BnO BnO S P O O OMe O O O O BnO BnOBnOO AcO NHTeoc NHTeoc 35 34 AcO AcO anose residue was attached through this linker. The elongation of the oligosaccharide chain was carried out as described above.A scandium(III) triflate ± acetic anhydride system was recommended for the splitting of the reaction product from the polymer.41 In all these oligomannoside syntheses, the chain was elongated from the reducing end to the non-reducing one. However, an alternative route is documented, which utilises an opposite strategy, i.e., the elongation is performed from the non-reducing end of the chain to its reducing end. The synthesis on the soluble support (PEG) was carried out according to the following scheme.42 The linker used was a 6-(4-hydroxymethylphenoxy)- hexanoic acid derivative, which alkylated methyl 3,4,6-tri-O- benzyl-1-thio-a-D-mannopyranoside at position 2 and esterified PEG (Scheme 11).The substituted methyl 1-thiomannopyrano- side residue in the polymer 37 served as the glycosyl donor, reacted with 3,4,6-tri-O-benzyl-D-mannopyranosyl fluoride in the pres- ence of methyl triflate to give the disaccharide derivative 38 which also played the role of a glycosyl donor. The Suzuki glycosyla- tion 43 of 2-(trimethylsilyl)ethyl 3,4,6-tri-O-benzyl-a-D-manno- pyranoside resulted in the immobilised trisaccharide 39; the free trisaccharide was split off the polymer by treatment with a base and deprotected by hydrogenolysis. The target product (yield 40%) was easily separated from by- products owing to the presence of a hydrophobic trimethylsilyl- ethyl group.Incorporation of hydrophobic groups to facilitate isolation of the target product was also used 42 in the synthesis of a structurally more complex heterooligosaccharide 40, which is a fragment of glycophosphatidylinositol. The latter was prepared by a reaction of the disaccharide 41 with the glycosyl donor 38. The tetrasac- charide 40, which was cleaved from the polymer 40p by hydro- genolysis, could easily be purified by reversed-phase chromatography owing to the presence of hydrophobic phthali- mido- and methoxyphenyl groups. This methodology is an interesting example of modification of carbohydrate synthesis on solid phase supports.From the oligomannoside syntheses described above one cannot draw any final conclusion in favour of one or another synthetic procedure, but the flexibility and high potential of the solid-phase method have been demonstrated.Solid-phase synthesis of oligosaccharides and glycoconjugates BnO P O(CH2)5COOCH2 BnO BnO O BnO O BnO BnO BnO BnO SMe 37 BnO OHO BnO BnO BnO OCH2CH2SiMe3 O(CH2)5COOCH2 P BnO O BnO O BnO BnO BnO BnO BnO O O BnO BnO BnOBnO OO BnOBnO OCH2CH2SiMe3 39 The solid-phase synthesis of the natural immunologically active heptasaccharide 42 b-Gal f (1-5)[b-Gal f (1-5)]5b-Gal fOCH2CH2CHCOOH 42 NH2 containing b-(1-5)-linked galactofuranose units and an L-homo- serine residue at the reducing end (Scheme 12) 44 deserves special mention.L-Homoserine 43 with protected OH and NH2 groups was immobilised on Merrifield's resin. The polymer 44 was detritylated and the resulting hydroxyl-containing derivative was glycosylated with 2,3-di-O-benzoyl-5-O-levulinoyl-6-O-pivaloyl- b-D-galactofuranosyl chloride (45) under conditions of the Hel- ferich reaction. The resulting polymer 46, which contained immo- bilised galactofuranose residues (the support loading was 0.5 mmol g71), was used as a starting material in the synthesis of the heptasaccharide 42. Selective removal of the levulinic acid residue from the polymer 46 and glycosylation of the liberated hydroxy group with the chloride 45 under identical conditions resulted in the elongation of the chain with one more galactofur- anose unit and afforded the polymer 47.Reiteration of this reaction sequence resulted in the heptasaccharide derivative 42, which was split off the polymer with simultaneous removal of all O-benzoyl groups by treatment with a base. The benzyloxycar- bonyl group was removed by hydrogenolysis. The overall yield of the heptasaccharide 42 was 23%. The resulting oligosaccharide contained exclusively b-glyco- sidic linkages and was virtually free from shorter oligomers, which suggested high yields in each glycosylation step and stability of the protecting groups. The choice of protecting groups was of particular importance in this synthesis. Thus upon substitution of a selectively removable chloroacetyl group for a levulinoyl group a large amount of admixtures was formed.The synthetic oligosaccharide 42 can be easily converted into the corresponding glycoconjugate owing to the presence of the homoserine residue. This is the first example of the synthesis of an oligosaccharide chain attached to an amino acid residue, which opens a route to the solid-phase glycopeptide syntheses. Further developments in solid-phase oligosaccharide synthe- sis were aimed at a search for novel methods of glycosylation. The P O(CH2)5COOCH2 O O O O F 38 OBn O HO BnO BnO O BnO 41 O(CH2)5COOCH2 O O BnO O O BnO O OBn O BnO BnO BnO OBnO 40p so-called glycal method (Scheme 13) 10 offers the most interesting opportunities for the synthesis of complex (including branched) oligosaccharides.After treatment with a mild epoxidising reagent (viz., 3,3-dimethyldioxirane), the glycal 48 immobilised on a support through a hydroxy group is converted into the 1,2- anhydrosugar 49. The latter reacted with the glycal 50 with a free TrOCH2CH2CHCOOH NHZ 43 TrOCH2CH2CHCOO NHZ 44 O(CH2)2CHCOO O OBz OLevOBz PivO O(CH2)2CHCOO O OBz OHOBz PivO O(CH2)2CHCOO O OBz OBz PivO O O OBz OLevOBz PivO 47 Tr=Ph3C; Z=PhCH2OCO; Piv=Me3CCO; Lev=MeCO(CH2)2CO. 803 Scheme 11 BnO O OC6H4OMe NPhth P O OC6H4OMe NPhth Scheme 12 O Cl OBz OLevOBz PivO 45 P P NHZ 46 P 45 NHZ P NHZ ...42804 O PO [O] RO RO RO RO 48 O PO O RO OH RO O O SiR2Cl+ P O 55 P OSiR2 O O O O O O OH O O 59 Ph O OO HO 62 P OSiR2 O O O O O OH O O O O O O 57 50a 63 O O O 65 57 R=Ph, Pri. HO O P O P O RO O O RO 50 O RO O OH RO RO 49 51 RO O O P ROHO 53 O RO O RO RO OR 52OH P OSiR2 O O O 57 O O O O 56 P OSiR2 O O O O O OH 56 57 O O O O O OH O 60 57 O Ph O O O O O OH 2 63 P OSiR2O O OH O Ph O O O O O 2 OH P OSiR2O O OH O Ph O O O O O O OH 2 [O] OOR RO O RO O O O O OH OH RO OR 54 P OSiR2 O 56 O O O O 58 P OSiR2 O O O O O HO 57 50a O O O O O O O O OH O O O OBnO O OBnO OH65 P OSiR2 O O O O OH O O O O OH 3 O O O HO O BnOBnO 64 O HO O O O O O OH OBnO 67 N K Kochetkov Scheme 13 Scheme 14 O O HOO O O O HO O BnOBnO 61 OBn O OBnO 66 OBn OSolid-phase synthesis of oligosaccharides and glycoconjugates hydroxy group in the presence of ZnCl2 to yield the disaccharide glycal 51 which is further converted into the 1,2-anhydrosugar 52.Its reaction with the monosaccharide glycal 53 afforded the immobilised trisaccharide glycal 54, and so on. The yields in each glycosylation step amounted to 90%± 95%. High stereo- specificity of glycosidic linkage formation is governed by well- established regularities in the epoxide ring opening. As can be seen, the main characteristic feature of this approach is the reconstruction of a reactive epoxide group (i.e., a glycosyl donor group) at the reducing end of the growing chain under standard mild conditions.By varying the structure of the glycal acceptor, one can obtain both regular and irregular homo- and heterooli- gosaccharides. Immobilised glycals with free hydroxy groups resulting either from the reaction or after selective removal of protecting groups can also play the role of glycosyl acceptors which permits their glycosylation leading to branched oligosac- charides. It should be emphasised that the use of the glycal method simplifies the purification of the reaction product, since possible admixtures formed as a result of incomplete glycosylation in the intermediate steps of the synthesis become more polar after final treatment and can easily be separated from the target oligosac- charides by chromatography.The high potentials and flexibility of the glycal method can be exemplified in the convergent synthesis of several oligosaccharides on an insoluble support (Scheme 14).10 Here, R2SiCl groups (R=Ph, Pri), which are used as linkers for binding the first monomer to the polymeric support, were incorporated into a polystyrene resin by sequential treatment with butyllithium and diphenyl- or diisopropyldichlorosilane. The use of such linkers allows easy cleavage of the O7Si bond upon treatment with Bu4NF, which enables detachment of the synthesised product from the support under mild conditions.Treatment of the thus modified polystyrene 55 with a D-gal- actal derivative 56 in the presence of a base followed by oxidation with dimethyldioxirane (57) affords the polymer 58, which con- tains the immobilised anhydrosugar residue, which is employed as the glycosyl donor for the elongation of the chain. Glycosylidation of the polymer 58 with D-galactal 56 in the presence of anhydrous ZnCl2 yields the disaccharide glycal derivative 59. Reiteration of the reaction cycle results in the trisaccharide glycal 60; its reaction with 3,4-di-O-benzyl-D-glucal (50a) yields the tetrasaccharide glycal 61 with (1-6) glycosidic linkages. The introduction of a structurally different glycal acceptor (e.g., 4,6-O-benzylidene-D- glucal 62) into the reaction with an immobilised glycosyl donor prepared from the trisaccharide glycal 60 results in the formation of the tetrasaccharide glycal 63 of irregular structure.After OH OCPh3 OH PivO PivO O O PivO PivO S S OPiv OPiv 68 69PivO PivO OH PivO O PivO O PivO P O PivO O PivO S OPiv conversion into the glycosyl donor, the glycal 63 reacts with 3,4- di-O-benzyl-D-glucal as described above to yield the irregular pentasaccharide derivative 64. Finally, oxidation of the trisac- charide glycal 60 and subsequent reaction with the disaccharide glycal 65 afforded the pentasaccharide glycal 66. Conversion of the tetrasaccharide glycal 63 into a glycosyl donor using the conventional scheme and reaction of the latter with the disacchar- ide glycal 65 result in the hexasaccharide derivative 67.In each step of this convergent synthesis, the oligosaccharides can be split off the polymer by treatment with Bu4NF and protecting groups can be removed by standard methods. The yields of the oligosaccharides are 80%± 90% in each step, which gives the final yields of the order of 20% to 40%. The stereo- chemistry of glycosidic linkages in all steps of this synthesis was established by NMR spectroscopy. The usefulness of two other, recently proposed glycosylation methods for solid-phase synthesis, viz., those utilising glycosyl sulfoxides 24 and pent-4-enyl glycosides 45 as glycosyl donors, is more difficult to evaluate.Several examples of solid-phase syntheses with the use of these glycosylating reagents are considered below. p-Hydroxyphenyl 2,3,4-tri-O-pivaloyl-6-O-trityl-1-thio-b-D- galactopyranoside (68) was immobilised on Merrifield's resin (Scheme 15).24 After removal of the trityl group from the immo- bilised monosaccharide, the hydroxy group in the polymer 69 was glycosylated with the sulfoxide 70 in the presence of trifluorome- thanesulfonic acid anhydride and 2,4,6-tri-tert-butylpyridine. After removal of the trityl group from the immobilised disacchar- ide, the reaction sequence was repeated. The resulting derivative of the corresponding trisaccharide 71 was split off the support by treatment with mercury trifluoroacetate in a dichloromethane ± water mixture.The glycosylation occurred stereospecifically with formation of a b-galactosidic bond. The yield of the trisaccharide was 52%. Taking into account the fact that only 70% to 75% of the oligosaccharide is liberated from the polymer, the yields of di- and trisaccharides in the glycosylation step were no less than 94%± 95%. Glycosyl sulfoxides can be used as glycosyl donors in the glycosylation of secondary hydroxy groups as is exemplified in the synthesis of two synthetic precursors of Lewis antigens of group- specific blood glycoproteins. Glycosylation of 2-azido-4,6-O-benzylidene-2-deoxy-1-thio- D-glucoside immobilised in an analogous way (72) with sulfoxides based on 2,3,4-tri-O-benzyl-L-fucose and 2,3,4,6-tetra-O-piva- loyl-D-galactose afforded the disaccharides 73 and 74 (yields 67% and 64%, respectively); these reactions were stereospecific.PivO OCPh3O O PivO P O SPh OPiv 70 OH O O PivO PivO O 70 PivO O PivO PivO PivO 71 805 Scheme 15 P O O S OPiv806 O Ph O OHO N3 72O Ph OO Me O OBn OBn 73 BnO Ph O OPiv O PivO O O PivO 74 OPiv The stereospecificity of these reactions, which is particularly emphasised by the authors,24 is quite obviously governed by the nature of a substituent at the O(2) of the glycosyl donor. The same regularity is observed when glycosylation is carried out in sol- ution. Further utilisation of glycosyl sulfoxides in solid-phase syn- theses of oligosaccharides is still not documented.Rather specific experimental conditions and uncertainty of the problem of stereospecificity in the case of other monosaccharides seem to preclude the widespread application of glycosylation based on the use of glycosyl sulfoxides. Pent-4-enyl glycosides act as glycosylating agents following generation of a carbocation at the glycosidic centre of the sugar as a result of treatment with N-iodosuccinimide and triethylsilyl triflate. The glycosylation with n-pent-4-enyl glycosides has been described in detail;45 some general considerations concerning the applicability of this reaction to solid-phase syntheses on different P OSiPri2 O O O ... O 75 O O O 77 ...Me BnO BnO OAc AcO AcO Me O OAc AcO TIPS=SiPri3 . N K Kochetkov S OR O OH N3 O OH N3 supports, with different linkers and monosaccharides were briefly surveyed in Ref. 46. According to Fraser-Reid et al.,46 solid-phase glycosylation gives satisfactory yields. Since the stereoselectivity of this reaction is not very high, it is recommended for use in combinatorial chemistry for the acquisition of `libraries' of immobilised oligosaccharides designed for direct activity assays. The principles common to combinatorial chemistry have been applied for the acquisition of a library of deprotected oligosac- charides concomitant with the synthesis of the trisaccharide b-Galp(1-2)-a-Glcp(1-6)GlcNAc (no experimental details were given).The advantages of solid-phase oligosaccharide synthesis are most conspicuous in the directed synthesis of complex fragments of natural biopolymers. Thus, the glycal method was used in the synthesis of a series of biologically important oligosaccharides which play an essential role in carbohydrate ± protein interactions and represent antigenic determinants. Some steps of this synthesis were carried out on a polystyrene support. The synthesis of the blood-group H-determinant (type 2) is shown in Scheme 16.47 3,4-O-Carbonyl-D-galactal 75 immobi- lised on a polystyrene resin through a diisopropylsilyl linker was oxidised with dimethyldioxirane into the corresponding 1,2-anhy- drosugar which was then introduced into a reaction with 3,6-di-O- benzyl-D-glucal in the presence of ZnCl2.The resulting immobi- lised disaccharide glycal 76 was glycosylated at the free OH group at position 20 with 2,3,4-tri-O-benzyl-L-fucopyranosyl fluoride in the presence of tin triflate and 2,6-di-tert-butylpyridine. The immobilised trisaccharide 77p was treated with Bu4NF to liberate the trisaccharide 77 (overall yield was 50%) after which the synthesis was continued in the solution. The hydroxy group at C(6) of the galactose residue in the trisaccharide 77 was protected with a triisopropylsilyl group and treated with benzenesulfamide in the presence of iodonium(dicollidine) perchlorate. The resulting 2-deoxy-2-iodo-N-phenylsulfonylglycosylamine 78 (cf. Ref. 48) underwent rearrangement in a reaction with 3-O-tributylstannyl- 6-O-triisopropylsilyl-D-galactal in the presence of silver tetra- fluoroborate.49 The resulting tetrasaccharide 79 was subjected to Scheme 16 P OSiPri P OSiPri 2 2 O O OBn OBn O O O O O O O O O BnO O BnO O OH 76 Me O OBn BnO BnO 77p OTIPS OTIPS O OTIPS OBn OBn HO O O O O O O O I O OBnO OBnO O O NHSO2Ph Me PhSO2NH O O OBn OBn BnO BnO 79 78 OAc OAc AcO O O O OAcO O O NHAc OAc 80Solid-phase synthesis of oligosaccharides and glycoconjugates O O 56 OO O 83 O Me OBn BnO O O O Me OBn BnO HO Me HO deprotection and N-acetylation to give the tetrasaccharide 80, which is a derivative of the H-antigen (type 2) of a blood group- specific glycoprotein.A similar approach was used in the synthesis of a Leb-specific hexasaccharide, which was prepared in a form suitable for con- jugation with proteins (Scheme 17).47 The starting galactal 56 was immobilised on a polystyrene resin; the reaction product was converted into an 1,2-anhydrosugar, which further glycosylated 6-O-triisopropylsilyl-D-glucal [its most reactive hydroxy group at C(3) reacts selectively]. The (1-3)-linked disaccharide 81 with two free hydroxy groups was glycosylated with an excess of 2,3,4-tri- O-benzyl-L-fucopyranosyl fluoride (82). The resulting tetrasac- charide 83p bearing two L-fucose residues was split off the support, the primary alcohol groups in the tetrasaccharide liber- P OSiPri2 O O O ...O BnO OBn OBn Me O OTIPS OTIPSI O O O O PhSO2NH O OBn84 BnO OBn OBn Me O OTIPS OTIPS O O O O PhSO2NH O OBn HO OH OH Me O OH OH OH O O O O O OH OH BnO OBn Me O Me O F OBn OTIPS P OSiPri2 OBn P OSiPri2 O BnO 82 O O HO O O O O OH O 81 Me O OBn OBn BnO 83p OTIPS HO O OTIPS O O Bu3SnO HO OH 85 O OTIPS HO OTIPS O O ... O O OHO OH 86 OH OH OH O O O O OR OHO OH NHAc OH 87: R = CH2CH=CH2 88: R = CH2CHO ated were silylated after which the synthesis was continued in solution. The tetrasaccharide 83 was converted into tetrasacchar- ide 2-deoxy-2-iodo-N-phenylsulfonylglycosylamine 84.The latter reacted with 3-O-tributylstannyllactal derivative 85 to give the hexasaccharide glycal 86, which was converted into an 1,2- anhydro derivative and then made to react with allyl alcohol to obtain the allyl glycoside of the corresponding hexasaccharide 87. To prepare antibodies against Leb antigen, the hexasaccharide 87 was converted into the aldehyde 88 suitable for conjugation with proteins. Halcomb et al.15 performed an enzymic synthesis in solution of the pentasaccharide NeuAc-a-(2-3)-Gal-b-(1-4)-[Fuc-a-(1-3)]- GlcNAc-b-(1-3)-Gal and an enzymic solid-phase synthesis of the tetrasaccharide NeuAc-a-(2-3)-Gal-b-(1-4)-GlcNAc-b-(1-3)-Gal, 807 Scheme 17 OTIPS O OO808 HOHO OH HO O HO OHO OH HO OH OH HO HO2C O O O HO AcHN OHO OH HO HO OH OH HO HO2C O O O HO AcHN OHO OH HO which are known as inhibitors of E-selectin and Helicobacter pylori, respectively.A fragment of the pentasaccharide, viz., the tetrasaccharide NeuAc-a-(2-3)-Gal-b-(1-4)-[Fuc-a-(1-3)]- GlcNAc (sialyl-Lewis X, SLeX), is a tumour-associated antigen; it is involved in processes of cell adhesion, metastasis, inflamma- tion and thrombosis. This antigen contains neuraminic acid; its stereospecific introduction into oligosaccharides by chemical methods presents a problem, therefore the use of enzymic glyco- sylation is more preferable in this case. The solid-phase synthesis of the tetrasaccharide NeuAc-a- (2-3)-Gal-b-(1-4)-GlcNAc-b-(1-3)-Gal was performed on CPG modified with iodoacetylamino groups, which made it possible to conduct the synthetic operations in both aqueous solution and organic solvents as required (Scheme 18).15 The starting disac- charide, viz., 2-acetamido-2-deoxy-b-D-glucopyranosyl-(1-3)-D- galactopyranose, prepared by chemical synthesis as a glycoside of o-hydroxyhexanoic acid, was immobilised on the modified CPG.After incubation with UDPGal in the presence of b-(1-4)- galactosyl transferase, the immobilised disaccharide 89 yielded the trisaccharide 90, which was made to react with cytidine 50-phos- pho-N-acetylneuraminic acid (CMP-N-acetylneuraminic acid) in the presence of a-(2-3)-sialyl transferase. The resulting tetrasac- charide 91 was split off the support with hydrazine and isolated as o-hydroxyhexanohydrazide glycoside 92 suitable for conjugation with proteins.This synthesis demonstrates again the advantages of the enzymic approach in solid-phase oligosaccharide syntheses HO O O O O OHHO HOHO HO OH O O HOHO HO OH HOHO HOHO OH HO O O NHAc 89 OH HO O O NHAc90 OH HO O O NHAc 91 OH HO O O NHAc 92O O OHHO O O OH OH O O O O OH UDPGal, galactosyl transferase OH O O O O OH CMP-N-acetylneuraminic acid, a-(2-3)-sialyl transferase OH O O O O OH NH2NH2 OH O O O NHNH2 OH where the regio- and stereospecificity of the glycosidic bonds formed are determined by the specificities of glycosyl transferases.The synthesis of the phytoalexin elicitor (PAE), which pre- vents viral and bacterial infections in plants, is yet another example of successful application of the solid-phase methodology in directed synthesis of biologically active and structurally com- plex oligosaccharides. Synthetic PAE (a branched heptasacchar- ide) presents substantial practical interest for agriculture, since the natural product is difficultly accessible due to its low concentra- tion in plants. The synthesis of PAE 93 was carried out using two different approaches. In the first approach (Scheme 19),50 the oligosaccharide chain was elongated by sequential addition of blocks on a soluble polymeric support (PEG) using a thioglycoside glycosylation method.The starting monosaccharide, viz., methyl 2,3-di-O- benzoyl-6-O-dimethoxytrityl-a-D-glucopyranoside (94), was attached to the polymer modified with succinic acid through the hydroxy group at C(4) (the support loading was 148 mmol g71). After mild acidolysis of the dimethoxytrityl group, the polymer was introduced into the reaction with the disaccharide thioglyco- side 95 in the presence of N-iodosuccinimide and trifluorometha- nesulfonic acid. Benzylidene protection was removed by acid hydrolysis after which the immobilised trisaccharide 96 was subjected to regioselective glycosylation with the thioglycoside 97 at O(60). Subsequent acylation of the hydroxy group at C(40) O O O OH OH OH HOHO 93 N K Kochetkov Scheme 18 NH CPG O NH CPG O NH CPG OSolid-phase synthesis of oligosaccharides and glycoconjugates OCPh(C6H4OMe-4)2 O HOBzO BzO OMe 94 BzO O BzOBzO BzO R1O BzOBzO BzO BzO O BzOBzO BzO (98a?98b) and removal of silyl protection resulted in the tetrasaccharide 98c.Reiteration of this reaction sequence, viz., the reaction with the disaccharide 95, debenzylidation and the reaction with the thioglycoside 97, afforded the protected immobilised heptasac- charide. Purification of the reaction products formed in each step from excess of the soluble reagents was achieved by precipitation of the polymeric material and subsequent washing with diethyl ether. Deprotection by treatment with sodium methoxide was accompanied by splitting of the heptasaccharide from the polymer eventually resulting in the biologically active PAE methyl glyco- side (yield 18%).According to the second scheme, PAE and its lower analogues were synthesised on a hydroxylated polystyrene resin using a photosensitive linker 99 and thioglycoside glycosylation.7 Nitro- benzyl alcohol 99 was glycosylated with phenyl-2,3,4-tri-O- benzoyl-6-O-(tert-butyldiphenylsilyl)-1-thio-b-D-glucopyranoside (100), and the resulting glycoside was immobilised. This glucose residue served as the origin of the oligosaccharide chain which was elongated by successive treatment with the thioglucosides 100, 101 and 102 (Scheme 20). The thioglucoside 100 contained a selec- tively removable tert-butyldiphenylsilyl protecting group at O(6), while the thioglucoside 101 contained a 9-fluorenylmethoxycar- bonyl (Fmoc) group at O(4). These groups were selectively removed when required, and the hydroxy groups liberated were glycosylated with the corresponding thioglucoside in the presence of dimethyl(methylthio)sulfonium triflate.The resulting hepta- saccharide (overall yield 20%) was split off the support by photolysis. Solid-phase synthesis was also used for the preparation of oligosaccharides containing 6-deoxy- and/or 2,6-dideoxymono- saccharide units which represent carbohydrate chains of some antibiotics.9 This synthesis combines ingeniously detachment of the oligosaccharide from the support with nucleophilic substitu- tion of the hydroxy group involved in immobilisation of the sugar on the polymer.The Merrifield's resin modified by introduction of ethanesulfonic acid was used as the solid support. Methyl 4-O- HO O O P OC(CH2)2COBzO BzO HO O HO O O BzO O O O P OC(CH2)2COBzO BzO 96 O O O R2O O O BzO O O O P OC(CH2)2COBzO BzO 98a ± c Ph O BzO BzO BzO OBz O OMeR1O O SEt BzO BzO OBz 97 OMe OMe 98a: R1=SiR3, R2=H; 98b: R1=SiR3, R2=Ac; 98c: R1=H,R2=Ac. acetyl-2,3-di-O-benzyl- and methyl 2-O-acetyl-3,4-di-O-benzyl-a- D-glucopyranoside 103a(b) were immobilised on this polymer (Scheme 21) after which O-acetyl groups were removed by treat- ment with guanidine. The immobilised monosaccharides 104a(b) reacted with the glycosyl donors 105 and 106 in the presence of trimethylsilyl triflate to afford the corresponding immobilised disaccharides 107 ± 110.These disaccharides were split off the support by treatment with NaI in ethyl methyl ketone with simultaneous substitution of iodine for the sulfonyloxy group. The yields of iododeoxysaccharides (111 ± 114) varied from 85% to 91%. Their treatment with Bu3SnH gave the corresponding deoxy sugars 115 ± 118 in 92% yield. This method was used in the targeted synthesis of the carbohydrate fragment of the antibiotic olivomycin A (Scheme 22).9 Glucal 119 was immobilised on a sulfonylethylated Merrifield's polymer after which it was desilylated and treated with trichloroacetimidate 122. The immobilised disaccharide 123 formed was desilylated (123?124) and glycosylated with glycosyl acetate 106 in the presence of trimethylsilyl triflate.The trisac- charide 125 was detached from the polymer with simultaneous substitution of iodine for bromine (yield 67%). The glycal 125 was used for direct incorporation of the carbohydrate fragment into the aglycon of olivomycin A. An important observation which awaits rationalisation is the fact that the glycosylation of the immobilised acceptor with trichloroacetimidate 122 was stereospecific leading exclusively to b-linkages, whereas the product synthesised in solution contained some proportion of a-linked oligosaccharide. This example is also interesting in that it illustrates that the solid-phase synthesis of carbohydrates can be combined with simultaneous transformation of the monosaccharide residues.This significantly extends the range of reactions in synthetic chemistry of carbohydrates performed on solid supports. The same approach can be used for the introduction of amino sugar units into oligosaccharides if sodium azide is employed instead of NaI for splitting off the reaction product.9 809 Scheme 19 O O O SEt O OBz 95...810 NO2 HOCH2 O (CH2)3I 99 BnOHO AcOAcO BzOBzO AcOAcO BnOHO OAc BnO O AcO O AcO OAc OAc O AcOAcO OAc OAc BnO O O AcOAcO OAc OSiButPh2 O SPh BzO BzO BzO OBz 100 BzO OSiButPh2 O O OBz O BzO O BzO OBz OH OAc O BnO O O O OBz BzO OAc BzO OH 1) 100 O 2) HF±C5H5N O OBz OAc O BnO O O O OBz BzO OAc BzO OSiButPh2 1) 101 O 2) Et3N O OBz O BzO O BzO OBz OAc BnO O AcO O AcO OAc OH 1) 102 2) HF±C5H5N O O OBz O BzO O BzO OBz OAc BnO O AcO O AcO OAc O 102 O O OBz O BzO O BzO OBz OAc BnO O AcO O AcO OAc OSiButPh2 NO2 O O OBz O 1) AcO AcO 102 NO2 2) HF±C5H5N O P (CH2)3OC6H4 NO2 O O OBz O (CH2)3OC6H4 NO2 O O OBz O (CH2)3OC6H4 O O OBz O BzO O BzO OBz O O OBz O BzO O BzO OBz O O OBz O BzO O BzO OBz N K Kochetkov Scheme 20 1) HF±C5H5N OSiButPh2 O 2) SPh BnO FmocO OBz 101 3) Et3N P (CH2)3OC6H4 OAcO SMe OAc PP NO2 O P (CH2)3OC6H4 NO2 O P (CH2)3OC6H4 NO2 O P (CH2)3OC6H4811 Solid-phase synthesis of oligosaccharides and glycoconjugates Scheme 21 OH P O SO2CH2CH2 O O O R1O R1O P CH2CH2S Cl + BnO BnO O R2O OR2OMe OMe 103a,b 104a: R1=Bn, R2=H; 104b: R1=H,R2=Bn.OAc AcO OAcO 1) MeO 1) AcO Me PriO2C NH I AcOO HO 106 105 CCl3 2) NaI 3) Bu3SnH 2) NaI 3) Bu3SnH R OAc AcO R O O O O O AcO BnO MeO BnO AcO BnO Me PriO2C OMe BnO OMe HO 107, 111, 115 I 109, 113, 117 + OAc + R AcO O O BnO O OMe AcO BnO AcO O OMe BnOBnO MeO Me PriO2C O R 108, 112, 116 HO I 110, 114, 118 P (107 ± 110), I (111 ± 114), H (115 ± 118). R=OSO2CH2CH2 Scheme 22 BrO P OH O SO2CH2CH2 NH AcO TBSO O O PhS O CCCl3 122 AcORO AcO TESO 120, 121 119 120: R=TES 121: R =H I I O P O SO2CH2CH2 O Br AcO O AcOO 106 O O ...AcO PhS AcOO MeO RO Me PriO2C 123, 124 PhS HO 125 I 123: R=TBS 124: R =H TES=Et3Si; TBS=Bu3Si. V. Synthesis of glycopeptides Synthesis of glycopeptides can be performed by elongation of a peptide or an oligosaccharide chain. Those peptides in which an amino acid is glycosylated with a monosaccharide are the simplest representatives of the glycopeptide family.} These can be prepared by direct glycosylation of the corresponding amino acid residue in a presynthesised immobilised peptide chain.{ Glycopeptides are compounds in which oligosaccharides are covalently linked to amino acids or peptides by O- or N-glycosidic bonds.They are structural components of glycoproteins, the important naturally occurring glycoconjugates responsible for a great diversity of vital cell functions and are used as the simplest models in studies of biological activities of glycoproteins. In recent years, glycopeptides have acquired considerable practical impor- tance, since many of them served as the basis for the development of new generations of highly efficient drugs. Glycopeptides are most commonly prepared by solid-phase synthesis, since their synthesis in solution faces serious problems. } It is interesting to note that the mode of incorporation of the first monosaccharide residue is a crucial factor which determines the phys- icochemical characteristics of these peptides.{ For direct O-glycosylation of peptides containing serine and threonine residues in solution, see Refs 51 and 52.812 This method has been briefly described by Hollo si et al.53 in the example of the synthesis of a glycosylated fragment 189-207 of tau-protein (126), responsible for a certain stage of Alzheimer's disease, and of its simplified analogue. 202 AcPro-Lys-Ser-Gly-Asp-Arg-Ser-Gly-Tyr-Ser-Ser-Pro-Gly-Ser-Pro- Gly-Thr-Pro-GlyNH2 126 The immobilised 19-membered peptide 127 possessing the necessary protecting groups in its amino acid residues (with the exception of the hydroxy group of serine-202 which was subject to glycosylation) was prepared by solid-phase peptide synthesis.This peptide was treated with 3,4,6-tri-O-acetyl-2-methyl-D-gluco-oxa- zoline { prepared by treatment of per-O-acetyl-b-D-glucosamine with trimethylsilyl triflate as described in Ref. 54 (Scheme 23). Scheme 23 202 202 AcPro ... Ser ... GlyNH AcPro ... Ser ... GlyNH P P AcO OH O 127 O OAc AcO 128p NHAc The resulting O-glycopeptide 128p (yield 23%) was split off the polymer by treatment with a trifluoroacetic acid ± anisole mixture and purified by chromatography. The structure of the glycopep- tide 128 was confirmed by mass spectrometry. The glycosylation of the serine residue in its simple peptide analogue AcGly-Ser-Pro- Val-Glu-Lys was carried out in a similar way. The corresponding O-glycopeptide was obtained in 33% yield.This synthesis dem- onstrated the possibility of glycosylation of the serine residue adjacent to the proline residue of immobilised peptides, which normally hardly undergoes glycosylation. The synthesis of N-glycopeptides in which mono- and oligo- saccharide residues (including hexasaccharides) are linked to the asparagine residue of the hexapeptide Asn-Tyr-Gly-Gly-Phe-Leu by an N-glycosidic bond is yet another example of direct glyco- sylation. In this synthesis, the hexapeptide with N-terminal aspartic acid was immobilised on a support and the aspartic acid was converted into the pentafluorophenyl ester. The activated ester was brought into reaction with the corresponding glycosyl- amines in the presence of hydroxybenzotriazole and amine to afford N-glycosylasparagines.55 It is necessary to note that direct glycosylation of immobilised peptide chains has not received wide recognition in synthetic practice.It is the classical solid-phase peptide synthesis (see, e.g., Refs 56 ± 90) that is mainly employed for the preparation of glycopeptides bearing a monosaccharide residue or a short oligosaccharide chain, which makes use of the corresponding glycosylated amino acids. P However, implementation of solid-phase glycopeptide syn- thesis in an automated regime of conventional peptide synthesis required that a number of problems was solved, such as a search for supports and linkers providing both the optimum conditions for the condensation in the elongation of the peptide chain and smooth non-destructive liberation of the target glycopeptide from the support.The most common support is Merrifield's peptide resin 56 modified by incorporation of benzyl alcohol fragments (HOCH2C6H4OCH2 ),58 the role of linker is played by a compound with benzhydrylamine and phenoxyacetic acid frag- ment (the so-called Rink linker),61 a variety of compounds { An attempt to incorporate the first monosaccharide into an immobilised peptide by direct glycosylation with sugar trichloroacetimidates was unsuccessful. N K Kochetkov containing b-hydroxyallylic groups (Hycron and Hycram link- ers) 62, 63 and nitrobenzophenone oxime.65 OMe NHFmoc MeO OCH2CO HOCH2CH=CHCONHCH27 Hycram linker Rink linker (RINK) HOCH2CH=CHCH2O(CH2CH2O)3CH2CH2CONHCH27 Hycron linker In parallel with a search for linkers, the general strategy of glycopeptide synthesis required unification of procedures used for incorporation of glycosylated amino acids into the peptide chain.To this end, efficient methods for incorporation of glycosy- lated amino acid residues into the peptide chain have been developed and synthesis of such fragments has been proposed.} And, finally, particular attention has been paid to the protec- tion of hydroxy groups of carbohydrates and functional groups of amino acids as well as to protecting group introduction, removal and manipulations as required in the synthesis. The Fmoc/PfP strategy was found to be the most efficient in the synthesis of O-glycopeptides.In this case, the amino group of serine or threonine (129) derivatives is protected with a Fmoc group which is easily removed upon treatment with mild bases, while the carboxy group is activated by conversion into the pentafluorophenyl ester (PfP). R SugO CH CHCOOPfP FmocNH 129 R=H, Me. The use of these derivatives allows incorporation of glycosy- lated amino acids into the peptide chain in an automated mode of a conventional peptide synthesis. An alternative approach to the synthesis of glycopeptides with a complex carbohydrate chain consists in the elongation of the oligosaccharide chain starting from the monosaccharide residue that has been incorporated into the immobilised peptide chain.In this case, as in the oligosaccharide synthesis in solution, the choice of protecting groups in the monosaccharide residues and the method of glycosylation which provide the highest regio- and stereoselectivity of glycosylation are of great importance.A factor which largely determines the successful outcome of the glycopep- tide synthesis is the proper choice of a support which ensures effective synthesis of both peptide and oligosaccharide fragments within a single system `(support+linker) ± solvent'. The condi- tions for the peptide and oligosaccharide syntheses can differ (e.g., the synthesis of an oligosaccharide fragment should preclude the presence of water or any other hydroxyl-containing components), which may require additional technical manipulations connected with a change of solvents in the course of the synthesis.Glycopeptide syntheses by chemical elongation of the oligo- saccharide chain was described in detail in a fundamental study by Paulsen et al.91 The Rink linker and norleucine (Nle) as the internal standard were used in the synthesis of mucin core fragments, viz., glyco- peptides with the peptide chain comprising eight amino acids and a di- or a trisaccharide residue (Scheme 24). One of the threonine residues incorporated into the peptide chain immobilised on a } These residues include O-glycosylated threonine and serine residues in the case of O-glycopeptide synthesis and N-glycosylated asparagine in the case of N-glycopeptide synthesis.Solid-phase synthesis of oligosaccharides and glycoconjugates Ph O O O HO N3 AcPro-Thr-Thr-Thr-Pro-Ile-Ser-Thr But But 130 BzO OBzO NH BzO O BzO CCl3 131 Ph O O BzO OBz BzO OBz O O O BzO BzO N3 OBz RINK Nle P AcPro-Thr- Thr-Thr-Pro-Ile-Ser-Thr But But But But 133 b-D-Gal-(1-3)-a-D-GalNAc 13 Pro-Thr-Thr-Thr-Pro-Ile-Ser-Thr 135 PEGA support contained a substituted 2-azido-2-deoxy-a-D- galactose residue linked by an O-glycosidic bond.The only free hydroxy group at the C(3) atom of the sugar residue in this immobilised glycopeptide (130) was glycosylated with the corre- sponding O-glycosyl trichloroacetimidates. As was noted above, this method is used for the formation of b-glycosidic linkages in good yields and with high stereoselectivity. Thus, the reaction of the glycopeptide 130 with O-(tetra-O-benzoyl-D-galactopyrano- syl)- (131) and O-(perbenzoyl-b-D-galactopyranosyl-(1-3)-D-gal- actopyranosyl) trichloroacetimidates (132) in the presence of trimethylsilyl triflate afforded the glycopeptide derivatives 133 and 134, respectively. The latter were split off the support with trifluoroacetic acid (yields 67% and 33%, respectively), this was accompanied by removal of O-tert-butyl protecting groups.The azido groups in the resulting compounds were transformed into acetamido groups with subsequent removal of O-benzoyl protec- tion. This procedure afforded the glycopeptides 135 and 136 with the N-acetylgalactosamine residue in the carbohydrate ± peptide linkage site.An attempt to perform this synthesis on other supports and substitution of acetyl groups in the trichloroacetimidates 131 and 132 for O-benzoyl groups decreased the yields of the resulting glycopeptides 133 and 134. The same linker and the support were used in the synthesis of a series of isomeric glycopeptides with (1-3) and (1-6) glycosidic linkages. The starting monoglycosylated peptide comprised a threonine residue glycosylated with 2-azido-2-deoxy-3,4-O-iso- propylidene-a-D-galactopyranose (Scheme 25). The glycosyla- tion of the peptide 137 with the trichloroacetimidates 131 and 132 gave the glycopeptide derivatives 138 and 139 (yields 69% and 53%, respectively) which were further converted into the corre- sponding O-glycopeptides after transformation of azido groups into acetamido groups with subsequent liberation from the sup- port and deprotection.This scheme was also used in the synthesis of O-glycopeptides having branched oligosaccharide chains. Here, the protecting benzylidene group in the immobilised glycopeptide 133 was 813 Scheme 24 RINK Nle P But But OBz BzO BzO OBzO O NH O BzO O BzO BzO CCl3 132 Ph O O BzO OBz O O O O O N3 OBz OBz AcPro-Thr-Thr-Thr-Pro-Ile-Ser-Thr RINK Nle P But But But But 134 b-D-Gal-(1-3)-b-D-Gal-(3-1)-a-D-GalNAc 13 Pro-Thr-Thr-Thr-Pro-Ile-Ser-Thr 136 removed by mild acid hydrolysis, and the glycopeptide 140 was selectively glycosylated with the trichloroacetimidate 131 at the primary hydroxy group of the partially protected disaccharide fragment.This procedure afforded the glycopeptide 141 with a branched oligosaccharide chain. This approach was also used in the preparation of glycopep- tides with carbohydrate chains built of amino sugars. The immo- bilised glycononapeptides 142 and 143 structurally related to compounds 130 and 137 with an additional N-terminal glutamic acid residue were glycosylated with an excess of O-(3,4,6-tri-O- benzoyl-2-deoxy-2-trichloroethoxycarbonylamino-a-D-glucopyr- anosyl) trichloroacetimidate 144 (Scheme 26). The N-trichloro- ethoxycarbonylamino- and azido groups in the resulting glycopeptide derivatives 145 and 146 were transformed into N-acetylamino groups, the protecting groups in amino acid and sugar residues were removed after which the reaction products were split off the support.The corresponding glycopeptides, which represent mucin core fragments, were obtained in 62% yield. Obviously, the elongation of an oligosaccharide chain in an immobilised glycopeptide bearing a monosaccharide residue is a flexible and reliable, albeit labour-consuming, method for the synthesis of complex glycopeptides. This procedure can be con- siderably simplified if glycosylation is performed with the use of glycosyl transferases. The use of enzymes makes monosaccharide protection unnecessary and ensures high regio- and stereoselec- tivity in the formation of glycosidic bonds. The supports employed in enzymic glycopeptide synthesis involving consecutive elongation of peptide and oligosaccharide chains must provide adequate conditions both for the synthesis of the peptide chain and for the transfer of monosaccharide residues under the action of glycosyl transferases.} This problem was successfully overcome } In choosing supports for such syntheses, one should take into account the fact that peptide syntheses are performed in organic solvents, while the elongation of the oligosaccharide chain is carried out in aqueous media.814 OH O O O N3 AcPro-Thr-Thr-Thr-Pro-Ile-Ser-Thr But But 137 OH OBz HO BzO O O O BzO N3 OBzAcPro-Thr-Thr-Thr-Pro-Ile-Ser-Thr But Sug AcGlu-Pro-Thr-Thr-Thr-Pro-Ile-Thr-Thr RINK Nle P But But But142, 143 Ph O O Sug= O HO N3 OBzO BzO BzO NH TeocNH O CCl3 144 N K Kochetkov Scheme 25 BzO OBzO BzO O OBzO 131 O O RINK Nle P Thr-Thr-Pro-Ile-Ser-Thr N3 AcPro-Thr- But But But But RINK Nle P 138 But But OBz BzO OBz BzO O 132 O O O BzO OBz OBz O O O N3 RINK Nle P AcPro-Thr-Thr-Thr-Pro-Ile-Ser-Thr But But But But OBz 139 BzO O BzO O OBz BzO OBz HO O O O BzO N3 131 RINK Nle P RINK Nle P OBzAcPro-Thr-Thr-Thr-Pro-Ile-Ser-Thr But But But But But But But 140 141 Scheme 26 But But O OH (143).(142); O O N3 Ph O OBz O 142 O O O BzO BzO N3 TeocNHAcGlu-Pro-Thr-Thr-Thr-Pro-Ile-Thr-Thr RINK Nle P But But But But But OBz 145 O BzO BzO TeocNH O O O O 143 N3 AcGlu-Pro-Thr-Thr-Thr-Pro-Ile-Thr-Thr RINK Nle P But But But But But 146Solid-phase synthesis of oligosaccharides and glycoconjugates OAcO NH RINK P NH H2N AcO AcO AcNH O147 OH FmocNH OH O O HO NH O HO O O AcNH OH OH NH NH2 149 O P HN BocHN NH (Gly)6 O O HO 150 HO O HN BocHN OH NH HO O OH O O O O NH HO HO OH 152 NHAc HO OH OH HO HO2C O O O O HO AcHN HO OH HO through the use of complex polymeric supports (PEGA) capable of swelling both in water and in organic solvents, or of mineral supports (e.g., CPG modified by introduction of amino groups).The synthesis of a model N-glycopeptide, viz., {b-D-Galp- (1-4)-b-D-GlcpNAc-[1-N(4)]}-Asn-Gly, on a PEGA-type support modified by attachment of the Rink linker is shown in Scheme 27.18 A C-terminal amino acid, viz., N-Fmoc-glycine, was immobilised on a modified support.After removal of the Fmoc-group, the immobilised glycine 147 was made to react with N(4)-glycosylated N(2)-Fmoc-asparagine to give the glycopeptide 148. Incubation with UDPGal and b-(1-4)galactosyl transferase was carried out after removal of Fmoc- and O-acetyl protecting groups. The resulting glycopeptide 149 was split off the polymer with CF3COOH or used further in the immobilised form. The glycosylation reaction was monitored by 1H NMRspectroscopy. The use of CPG modified with 3-aminopropyl groups for solid-phase chemo-enzymic synthesis of glycopeptides was exem- HO OH OH HO HO2C TDP-fucose, 153 O O HO AcHN a-(1-3)-fucosyl transferase HO Me HO HO O O NH RINK P NH O 148 NH RINK P O O BocHN NH O OH O O NH 151 NHAc O O P O BocHN NH O OH O O NH153p NHAc plified in the preparation of sialyl-Lewis-X glycopeptide (Scheme 28).14 In this synthesis, the hexaglycin fragment was attached to the amino groups of the modified CPG; the unreacted amino groups of the polymer were `capped' by acetylation to prevent their involvement in further reactions after which the terminal amino group of the hexapeptide was acylated with O-(N- Boc-phenylalanyl)glycolic acid.The ester bond formed is suscep- tible to hydrolysis with chymotrypsin and is well-suited for splitting the reaction product off the support.Following libera- tion of the amino group in the ester 150, glycine and N(4)- glycosylated asparagine residues were introduced by conventional methods of solid-phase peptide synthesis. The immobilised glyco- peptide 151 was incubated with UDPGal and b-(1-4)-galactosyl transferase to afford the immobilised glycopeptide 152. The latter was made to react with cytidine 50-phospho-N-acetylneuraminic acid (CMP-N-acetylneuraminic acid, the activated form of N-ace- tylneuraminic acid) and a-(2-3)-sialyl transferase to yield the BocHN OH O O O O OH NHAc O OH 154 O NH O O O HN O O N N O O NH 815 Scheme 27 ... Scheme 28 P H NH (Gly)6 N PO OH816 glycopeptide 153p containing a trisaccharide chain.The latter is easily split off the support with a-chymotrypsin. After liberation, the N-glycopeptide 153 was further glycosylated in solution with thymidine 50-diphospho-L-fucose (TDP-fucose) in the presence of a-(1-3)-fucosyl transferase eventually resulting in the branched N-glycopeptide 154. The synthesis of a more complex biologically important O-glycopeptide 155 with the same tetrasaccharide moiety, which represents a fragment of the mucin-type glycoprotein MAdCAM-1 (a specific ligand for selectins),92 was carried out in a similar way. The problems which arose in its synthesis were connected with the high lability of the O-glycosidic bond between N-acetylglucosamine and threonine.AcLys-Pro-Pro-Asn-Thr-Thr-Ser-Ala-OH HO O OH OH O HO2C OH NHAc O O O HO AcHN O O HO OH OH OH O Me OH OH 155 The synthesis of the glycopeptide 155 was carried out on PEGA and CPG modified with aminopropyl groups. CPG appeared to be more convenient than organic supports of the PEGA type. The Hycron linker containing an allylic hydroxy group with an attached N-Fmoc-alanine as a terminal amino acid was used for the immobilisation of the peptide chain. After removal of the Fmoc-group in the polymer Fmoc-Ala- HYCRON-NH-CPG, the peptide chain was elongated from the alanine residue using a standard N-Boc/OBut procedure. N-Ace- tylglucosamine linked to threonine served as a carbohydrate ± peptide junction.After removal ofN- andO-protecting acid-labile groups (treatment with trifluoroacetic acid in the presence of ethanedithiol) from the immobilised monoglycosyloctapeptide 156, this AcLys(Boc)-Pro-Pro-Asn(Trt)-Thr-Thr(But)-Ser(But)-Ala-HYCRON-NH-CPG OR O O RORO NHAc 156 was incubated first with UDPGal in the presence of galactosyl transferase and then with CMP-N-acetylneuraminic acid in the presence of a-(2-3)-sialyl transferase. The resulting triglycosyl peptide derivative 157 was split off the support in the presence of OH NH2 O HO O C13H27 HO OH HO OH 159 HO CONHNH P NO2 OH HO OH CH2OCO O NH O O HO O C13H27 HO OH OH OH 161 Pd(0) as a catalyst. The yield of the glycopeptide 157 was 9% with respect to the immobilised alanine.The glycopeptide 157 was converted into the glycopeptide 155 by enzymic fucosylation as described above.AcLys-Pro-Pro-Asn-Thr-Thr-Ser-Ala-HYCRON-NH-CPG OH OH OHHO2C O O HO AcHN HO O HO HO O OH The glycopeptide 155 has also been synthesised in a solution. The yield of the target product was higher in this case, but the reaction time increased significantly (up to 9 days). In the authors' opinion,92 the shorter duration of solid-phase synthesis may fully compensate for the lower yields of the target product. The enzymic elongation of oligosaccharide chains was also employed in the synthesis of glycoconjugates which are used in the manufacture of diagnostic preparations (artificial antigens), e.g., in the synthesis of a polymer containing the a-NeuNAc(2-3)-b-D- Galp(1-4)-b-D-GlcNAc chains attached to a polyacrylamide car- rier.93 The polyacrylamide carrying monosaccharide residues was obtained by copolymerisation of n-pent-4-enyl 2-acetamido-2- deoxy-b-D-glucopyranoside and acrylamide.94 This was incu- bated consecutively with UDPGal in the presence of b-(1-4)gal- actosyl transferase and p-nitrophenyl N-acetylneuraminate in the presence of Tc-Ts-transsialidase to afford the synthetic glycocon- jugate 158, which contained trisaccharide units together with a small amount of unsialylated units.HO OH OH HO HO2C O O HO AcHN HO OH HO HO Thus, in some cases the solid-phase synthesis is confined exclusively to the preparation of a glycopeptide containing only one monosaccharide residue in the carbohydrate ± protein junc- tion, which is followed by enzymic elongation of the carbohydrate chain in solution.The synthesis of a large series of N-glycopep- tides the chains of which contained different amino acid sequences including two asparagine residues glycosylated with N-acetylglu- cosamine, N-acetyllactosamine and sialyl-(2-60)-N-acetyllactos- amine, provides an example.95 P CONHNH NO2 OH CH2OCO NH O O OH OH 160 OH HO O O HO HO OH N K Kochetkov OHO O NHAc 157 OH O O O NHAc OHO x OH OH O O O y H2N NHAc OHO O OH z 158 Scheme 29 C13H27 OH NH2 O O C13H27 OH OH 162Solid-phase synthesis of oligosaccharides and glycoconjugates The solid-phase synthesis of lactosylceramide on a soluble support with a photosensitive linker is yet another, thus far unique example of enzymic elongation of a carbohydrate chain in glycolipids.96 Synthetic (2S,3R,4E)-2-amino-1-(b-D-glucopyra- nosyloxy)-3-hydroxyoctadec-4-ene (159, glucosylsphingosine) was converted into a urethane (4-hydroxymethyl-3-nitrobenzoic acid derivative) and immobilised on a copolymer of acrylamide with N-acryloyloxysuccinimide (Scheme 29). The immobilised glycolipid 160 was incubated with UDPGal and galactosyl trans- ferase.The resulting disaccharide derivative 161 was purified from low-molecular-weight admixtures by dialysis and the reaction product 162 was split off the polymer by photolysis.VI. Synthesis of phosphoglycans Solid-phase synthesis has successfully been used for the prepara- tion of fragments of phosphoglycans, i.e., the regular biopolymers built of repeating carbohydrate units linked through phospho- diester bonds. This class of biopolymers, e.g., teichoic acids, are widespread in microorganisms where they play the role of antigens. Synthesis of phosphoglycans in solution results exclusively in low-molecular-weight compounds which contain no more than two or three repeating units of the biopolymer. Apparently, solid- phase synthesis is more preferable in the preparation of high- molecular-weight phosphoglycan fragments. Several examples of such syntheses are currently known.In general, the synthesis of phosphoglycans is very close to the synthesis of polynucleotides and differs in the structure of the repeating units which incorporate more or less complex carbohy- drate structures. The synthesis of a fragment of teichoic acid from Bacillus licheniformis is the first example of solid-phase phosphoglycan synthesis. This biopolymer is built of 1-O-(b-D-galactopyrano- syl)glycerol repeating units linked via (60-3)-phosphodiester bonds. The solid-phase synthesis of the trimer of the repeating unit of this phosphoglycan was performed on a modified CPG; the phosphodiester bond was formed using the phosphoramidite method (Scheme 30).97 The galactopyranosylglycerol derivative 163, which represents a fragment of the repeating unit of this biopolymer, was synthesised in solution and was treated with succinic anhydride to yield the succinic acid monoester 164, which was then immobilised on the modified CPG.The reaction of the derivative 163 with Pri2NP(OCH2CH2CN)Cl gave (galactosylgly- ceryl) (2-cyanoethyl) N,N-diisopropylphosphoramidite 165, which was used as a phosphorylating agent. The elongation of the oligomeric chain was carried out in an automated mode using ODMTr AcO O O OAc BOMO ODMTr OCO(CH2)2COOH AcO O O 164 OAc OAc BOMO ODMTr OH OAc AcO O O 163 OAc BOMO O OAc 165 ODMTr O OAcO AcO O O O 165 P OAc OAc 164p BOMO BOMO OR O n OAc OAc DMTr is dimethoxytrityl; BOM is benzyloxymethyl. 817 a well-established technique for the synthesis of oligonucleotides, This included reiteration of the following operations: liberation of the primary hydroxy group in the galactose residue of the polymer 164p by removal of the 60-dimethoxytrityl group, the reaction with phosphoramidite 165 in the presence of 1H-tetrazole, acetylation of unreacted free hydroxy groups in order to prevent their interaction at subsequent steps and oxidation of trivalent phos- phorus into pentavalent phosphorus with iodine.The yield in each reaction cycle was 95%. The target product 166 (n=3) was split off the support by ammonolysis with simultaneous removal of O-acetyl and cyanoethyl groups. The structure of the teichoic acid fragment thus prepared was confirmed by 1H NMRspectroscopy and FAB-mass spectrometry. The synthesis of a fragment of the capsular polysaccharide of Haemophilus influenza (type b), which is known to induce menin- gitis in children, proved to be more difficult.This synthesis was undertaken 98 with the aim of using this phosphoglycan as an artificial antigen in the preparation of a synthetic vaccine (Scheme 31). This synthesis utilised the methodology of the oligonucleotide synthesis and was carried out on a modified CPG. The substituted ribosylribitol 167 was used as the starting compound. The disiloxanediyl group was removed by treatment with Bu4NF, the primary hydroxy group was selectively protected by silylation and the secondary hydroxy group at C(30) was converted into the succinic acid monoester.The resulting ester 168 was immobilised on CPG and detritylated to afford the polymer 169. The ribosylribitol 167 was converted into 2-cya- noethyl N,N-diisopropyl phosphoramidite 170. This was then introduced into reaction with the polymer 169, which was carried out in an automated synthesiser under conditions of nucleotide synthesis. The reaction cycle was repeated as many times as required after which a (6-aminohexyl) phosphate spacer was attached to the oligomer 171 (n=6) in order to ensure its further transformation into a synthetic antigen following binding to the protein. The reaction product was split off the support by O HO O O NHá3H O OH OH OH OH O P O O7 n 172 ammonolysis; the protecting groups were removed by treatment with Bu4NF and hydrogenolysis.The overall yield of the target oligomer 172 (n=6) was 24%. Scheme 30 ODMTr AcO O O OAc BOMO OCO(CH2)2CONH P OAc 164p NPri2 PO(CH2)2CN O OAcO O O O P OAc BOMO OR O OCO(CH2)2CONH P OAc 166818 O O O Me2Si O O OBOM Me2Si 167 Ph2ButSiO 170 ... 169 P CO(CH2)2COO Synthesis of the phosphoglycan on a soluble support (PEGA) 99 using an analogous scheme with O-benzyl protection, was found to be more efficient and suitable for practical purposes. In this case, the loading of the support was larger (109 mmol g71) than in the previous synthesis (34 mmol g71); higher oligomers were synthesised in 35% yield. The same standard procedure was used in the synthesis 100 of a fragment 173 of the phosphoglycan of Haemophilus pleuropneu- moniae (serotype 2) which induces swine diseases. This fragment is made up of b-D-glucopyranosyl-(1-6)-b-D-glucopyranosyl-(1-6)- a-D-galactopyranosyl-(1-3)-glycerol repeating units linked through (1-30)-phosphodiester bonds.HO O O OO OH OH HO HO OH HO OH 173 The synthesis of the tetrameric fragment of the repeating unit 100 was carried out according to a scheme similar to those used in previous syntheses. CPG modified with aminopropyl groups served as the support and the phosphoramidite method was employed for the formation of the phosphodiester bond. O NH O P BnO O 174 OBz OBz OBz ODMTr O O OBOM O O O O HOHO O POH 4 ODMTr Ph2ButSiO O OBOM OCO(CH2)2CO2H 168 Ph2ButSiO O OBOM OP ODMTr OCH2CH2CN NPri2 Ph2ButSiO OBz OBz OBz O OCH2CH2CN 171 O NH2 N K Kochetkov Scheme 31 O Ph2ButSiO O O OBz OBz OBz OBz OBz OBz OBOM OH ODMTr P CO(CH2)2COO 169 O OBz OBz OBz 170 O O OBz OBz O OBz P O OBOM H O n71 2-O-Benzyl-3-O-dimethoxytritylglycerol was converted into a monoester of succinic acid and immobilised on CPG as in the previous syntheses. The resulting product 174 was detritylated and used as an initial unit for the elongation of the chain.The AcO O O OAc OO OAc AcO AcO OAc OAc BnO O OH O BnO BnO 175 OH primary hydroxy group in the glycerol portion of the repeating unit, viz., triglycosylglycerol 175, was protected with dimethoxy- trityl ether, and the hydroxy group at C(3) of the galactose residue was converted into (2-cyanoethyl) N,N-diethylphosphoramidite and used as a phosphorylating agent under conditions of auto- mated oligonucleotide synthesis.The immobilised tetramer of the phosphoglycan repeating unit 176 obtained after four reaction cycles was detritylated with subsequent introduction of a phos- phorylated spacer. The reaction product was split off the support by ammonolysis with simultaneous removal of protecting ester groups. Subsequent hydrogenolysis gave compound 173 which could be employed for the preparation of a synthetic vaccine. AcO O O O O OAc OAc AcO O O BnO OAc AcO OAc O HNO P O O BnO O O P DMTr O BnO BnO OCH2CH2CN 4 176Solid-phase synthesis of oligosaccharides and glycoconjugates These results demonstrate the high efficiency of the solid- phase synthesis of phosphoglycans of the teichoic acid type, which utilises well-established techniques of nucleotide synthesis.VII. Conclusion The examples of solid-phase carbohydrate syntheses described above demonstrate that advances in this area are modest yet. The problems encountered in solution synthesis of various carbohy- drate structures are related to the polyfunctionality of monosac- charides and non-stereospecificity of the formation of glycosidic bonds; these problems persist and preclude automation of the synthesis. It is quite natural that the `technology' of solid-phase synthesis of carbohydrates demands further improvement and studies in this area are currently under way.It is desirable that the elonga- tion of the carbohydrate chain by chemical methods should involve standard procedures for the formation of active glycosyl- donor function and for the liberation of hydroxy groups after each elongation step. The solution of this problem would open a way to the automation of the synthetic process. An example of a successful solution of this problem is described in this review, particularly, in the section devoted to oligosaccharide synthesis by the glycal method. It should be noted that elongation of the oligosaccharide chain is more conveniently performed in the direction from the reducing end of the chain to its non-reducing end, since in this case the elongation of the chain by addition of the next fragment to the immobilised reagent should be preceded only by a selective liberation of the required hydroxy group, which is much easier to perform than to reconstruct the activated glycosidic centre.Yet another problem which demands urgent solution is the choice of ideal polymeric supports and linkers compatible with them. Their wide availability will significantly simplify the syn- thesis and open up a way to the automated synthesis. Obviously, solid-phase synthesis can successfully be employed for the preparation of lower oligosaccharides, but in this case its advantages over syntheses in solution are not so apparent.However, on going to higher-molecular-weight oligosaccharides and glycoconjugates the merits of the solid-phase technique will become more conspicuous as in the case of glycoconjugate (viz., glycopeptide and phosphoglycan) synthesis. It is this trend that has become particularly popular in the last few years. And yet major progress was achieved in those syntheses where the elonga- tion of the carbohydrate chain was carried out by enzymic methods, which made it possible to overcome the problem of selective protection of hydroxy groups and its changes in each step of the synthesis. The feasibility of modification of immobilised monosacchar- ides presents particular interest and can be illustrated by several examples. The use of solid-phase techniques in combinatorial chemistry, particularly in a search for carbohydrate structures possessing the optimum biological activity, is one of the most attractive fields.The use of solid-phase synthesis for the preparation of an immense diversity of the so-called `libraries' of structurally related poly- functional oligosaccharide structures, excels other methods in many points. A search for medicinal drugs and other substances possessing specific biological activities among carbohydrate-containing com- pounds demanded a rapid synthesis of a vast variety of analogues and fragments of biopolymers aimed at a selection of the most active compounds. This in turn required the acquisition of `libraries' of compounds synthesised by methods of combinatorial chemistry and their further analysis. It is known that `combinatorial chemistry' is based on the principle of simultaneous synthesis of a series of related com- pounds and includes reactions of structurally similar acceptors with donors analogous to each other.This principle can be used 819 sequentially in different stages of a multistep synthesis.101 The most active compounds are further selected from such `libraries'. This strategy attracts considerable attention of specialists in the chemistry of biopolymers who widely utilise the strategy of solid-phase synthesis in the design of peptide and oligonucleotide `libraries'. Very often, the selection of active components does not require their splitting off the support.The advantages of this approach in a search for biologically active oligosaccharides are especially apparent against the background of their immense structural diversity, which is due to the polyfunctionality of the constituent monosaccharides. The applicability of solid-phase synthesis to the construction of oligosaccharide `libraries' has been considered.102 Acquisition of a `library' comprising *1300 compounds which represent derivatives of b-D-galactopyranosyl- (1-3)-2-deoxy-2-N-acylaminogalactopyranoside and include di- and trisaccharides differing in the nature and sequence of mono- saccharides, type of intermonomeric glycosidic bonds and sub- stituents at hydroxy and amino groups is a clear example.103 A Tentagel-type polymer was used as a support and glycosyl sulf- oxides served as glycosyl donors. The resulting `library' of immobilised oligosaccharides was examined for the specificity of carbohydrate ± protein interactions.The most active components were selected using a lectin test. At present, the principles of combinatorial chemistry includ- ing carbohydrate synthesis have become widely used by large manufacturing companies.104 Special mention should be made in conclusion of a recent review 105 devoted to solid-phase synthesis of oligosaccharides, which suggests that interest in this method continues. References 1. R B Merrifield J. Am. Chem. Soc. 85 2149 (1963) 2. V Varki Glycobiology 3 97 (1993) 3. R A Dwek Chem.Rev. 96 683 (1996) 4. J Frechet, C Schuerch Carbohydr. Res. 22 399 (1972) 5. R D Guthrie, A D Jenkins, G A F Roberts J. Chem. Soc., Perkin Trans. 1 2414 (1973) 6. D Obrecht, J Villalgordo, in Solid Supported Combinatorial and Parallel Synthesis of Small Molecular Weight Compounds Library (Oxford: Pergamon Press, 1998) p. 24 7. K C Nicolaou, N Winssinger, J Pastor, F DeRoose J. Am. Chem. Soc. 119 449 (1997) 8. S-H Lee Chiu, L Andersen Carbohydr. Res. 50 227 (1976) 9. J Hunt, W Roush J. Am. Chem. Soc. 118 9998 (1996) 10. J T Randolph, K F McClure, S J Danishefsky J. Am. Chem. Soc. 117 5712 (1995) 11. J Rademann, R Schmidt J. Org. Chem. 62 3650 (1997) 12. J Rademann, A Geyer, R R Schmidt Angew. Chem., Int. Ed. Engl. 37 1241 (1998) 13.U Zehavi, S Sadeh,M Herchman Carbohydr. Res. 124 23 (1983) 14. M Schuster, P Wang, J C Paulson, C-H Wong J. Am. Chem. Soc. 116 1135 (1994) 15. R L Halcomb, H Huang, Ch-H Wong J. Am. Chem. Soc. 116 11315 (1994) 16. R Eby, C Schuerch Carbohydr. Res. 39 151 (1975) 17. S P Douglas, D M Whitefield, J J Krepinsky J. Am. Chem. Soc. 117 2116 (1995) 18. M Meldal, F-I Auzanneau, O Hindsgaul,M M Palcic J. Chem. Soc., Chem. Commun. 1849 (1994) 19. M Adinolfi, G Barone, L DeNapoli, A Iadonisi, G Piccialli Tetrahedron Lett. 37 5007 (1996) 20. N K Kochetkov, in Studies of Natural Products Chemistry Vol. 14 (Ed. Atta-ur Rachman) (Amsterdam: Elsevier, 1994) p. 201 21. K Toshima, K Tatsuta Chem. Rev. 93 1503 (1993) 22. N K Kochetkov,A F Bochkov, in Recent Developments of Chemistry of Natural Carbon Compounds Vol.4 (Budapest: Akademiai Kiado, 1971) p. 77 23. R Eby, C Schuerch Carbohydr. Res. 34 79 (1974) 24. Lin Yan, C M Taylor, R Goodnow Jr , D Kahne J. Am. Chem. Soc. 116 6953 (1994)820 25. R Schmidt,W Kinzy Adv. Carbohydr. Chem. Biochem. 50 21 (1994) 26. S J Danishefsky, K F McClure, J T Randolph, R B Ruggeri Science 260 1307 (1993) 27. H J M Giisen, L Qiao,W Fitz, C-H Wong Chem. Rev. 96 443 (1996) 28. J M Frechet, C Schuerch J. Am. Chem. Soc. 93 492 (1971) 29. G Excoffier, D Gagnaire, M Vignon Carbohydr. Res. 51 280 (1976) 30. S Douglas, D Whitefield, J J Krepinsky J. Am. Chem. Soc. 113 5095 (1991) 31. D M Whitefield, S P Douglas, J J Krepinsky Tetrahedron Lett. 33 6795 (1992) 32.U Zehavi, A Patchornik J. Am. Chem. Soc. 95 5673 (1973) 33. U Zehavi,M Herchman Carbohydr. Res. 128 160 (1984) 34. U Zehavi,M Herchman, S Klpper Carbohydr. Res. 228 255 (1992) 35. T Wiemann,N Taubken,U Zehavi, J Thiem Carbohydr. Res. 257 C1 (1994) 36. S-I Nishimura, K Matsuoka, Y C Lee Tetrahedron Lett. 35 5657 (1994) 37. N K Kochetkov Tetrahedron 41 2389 (1987) 38. N K Kochetkov Sintez Polisakharidov (Synthesis of Polysaccharides) (Moscow: Nauka, 1994) 39. J Rademann, R Schmidt Tetrahedron Lett. 37 3989 (1996) 40. A Heckel, E Mross, K-H Yung, J Rademann, R R Schmidt Synlett 171 (1998) 41. S Mehta, D Whitefield Tetrahedron Lett. 39 5907 (1998) 42. Y Ito, O Kanie, T Ogawa Angew. Chem., Int. Ed. Engl. 35 2510 (1996) 43. K Suzuki, H Maeta, T Matsumoto Tetrahedron Lett.30 4853 (1989) 44. G H Veeneman, S Notermans, R M J Liskamp, G A van der Marel, J H van Boom Tetrahedron Lett. 28 6695 (1987) 45. R Madsen, B Fraser-Reid, in Modern Methods in Carbohydrate Synthesis (Eds R A O'Neill, Sh H Kan) (Amsterdam: Harwood Academic Publishers, 1996) Ch. 7 46. R Rodebaugh, S Joshi, B Fraser-Reid, H M Geysen J. Org. Chem. 62 5660 (1997) 47. J T Randolph, S J Danishefsky Angew. Chem., Int. Ed. Engl. 33 1470 (1994) 48. D A Griffith, S J Danishefsky J. Am. Chem. Soc. 112 5811 (1990) 49. S J Danishefsky, K Koseki, D A Griffith, J J Gervay,M E Peterson, F McDonald, T Oriyama J. Am. Chem. Soc. 114 8331 (1992) 50. R Verduyn, P A M van der Klein, M Douwes, G A van der Marel, J H van Boom Recl. Trav. Chim. Pays-Bas 112 464 (1993) 51. N J Maeji, Y Inoue, R Chujo Carbohydr. Res. 146 174 (1986) 52. H G Garg, T Hasenkamp,H Paulsen Carbohydr. Res. 151 225 (1986) 53. M Hollo si, E Kolla t, I Laszko, K F Medzihradzky, J Thurin, L OtvoÈ s Tetrahedron Lett. 32 1531 (1991) 54. S Nakabayashi, C D Warren, R W Jeanloz Carbohydr. Res. 150 C7 (1986) 55. D Vetter, D Tumelty, S K Singh,M A Gallop Angew. Chem., Int. Ed. Engl. 34 60 (1995) 56. S Lavielle, N C Ling, R Saltman, R C Guillemin Carbohydr. Res. 89 229 (1981) 57. J Jezek, R Straka, V Krchnak,MRyba, J Rotta, P Mayer,MZaoral Coll. Czech. Chem. Commun. 52 1609 (1987) 58. H Paulsen,G Merz,U Weichert Angew. Chem., Int. Ed. Engl. 27 1365 (1988) 59. P Atherton, R Sheppard Solid Phase Peptide Synthesis. A Practical Approach (Oxford: IRL Press, 1989) 60. B LuÈ ning, T Norberg, J Tejbrant J. Chem. Soc., Chem. Commun. 1267 (1989) 61. H Rink Tetrahedron Lett. 28 3787 (1987) 62. R Polt, L Szabo, J Treiberg, Y Li, V J Hruby J. Am. Chem. Soc. 114 10249 (1992) 63. H Kunz, B Dombo Angew. Chem., Int. Ed. Engl. 27 711 (1988) 64. O Seitz, H Kunz Angew. Chem., Int. Ed. Engl. 34 803 (1995) 65. ACBauman, J S Broderick,RMDacus IV,DAGrover, L S Trzupek Tetrahedron Lett. 34 7019 (1993) 66. M Meldal, K J Jensen J. Chem. Soc., Chem. Commun. 483 (1990) 67. A M Jansson,M Meldal, K Bock Tetrahedron Lett. 31 6991 (1990) 68. A M Jansson,M Meldal,K Bock J. Chem. Soc., Perkin Trans. 1 1699 (1992) 69. K B Reimer, M Meldal, S Kusumoto, K Fukase, K Bock J. Chem. Soc., Perkin Trans. 1 925 (1993) 70. J Montreuil Adv. Carbohydr. Chem. Biochem. 37 137 (1980) 71. V A Derevitskaya Bioorg. Khim. 9 581 (1983); 14 1605 (1988) a N K Kochetkov 72. T Bielfeldt, S Peters,M Meldal,K Bock,H Paulsen Angew. Chem., Int. Ed. Engl. 31 857 (1992) 73. H Paulsen, T Bielfeldt, S Peters, M Meldal, K Bock Liebigs Ann. Chem. 369 (1994) 74. H Paulsen, T Bielfeldt, S Peters, M Meldal, K Bock Liebigs Ann. Chem. 381 (1994) 75. J Rademann, R Schmidt Carbohydr. Res. 269 217 (1995) 76. S Rio-Anneheim, H Paulsen,M Meldal, K Bock J. Chem. Soc., Perkin Trans. 1 1071 (1995) 77. B Liebe, H Kunz Angew. Chem., Int. Ed. Engl. 36 618 (1997) 78. Yo Nakahara, Yu Nakahara, T Ogawa Carbohydr. Res. 292 71 (1996) 79. Yo Nakahara, Yu Nakahara,Y Ito, T Ogawa Tetrahedron Lett. 38 7211 (1997) 80. E Meinjohanns, A Vargas-Berenguel, M Meldal, H Paulsen, K Bock J. Chem. Soc., Perkin Trans. 1 2165 (1995) 81. A M Jansson, K J Jensen, M Meldal, J Lomako, W M Lomako, C E Olsen, K Bock J. Chem. Soc., Perkin Trans. 1 1001 (1996) 82. S Peters, T L Lowary, O Hindsgaul, M Meldal, K Bock J. Chem. Soc., Perkin Trans. 1 3017 (1995) 83. H Hietter, M Schultz, H Kunz Synlett 1219 (1995) 84. M K Christensen,M Meldal, K Bock, H Cordes, S Mouritsen, H Elsner J. Chem. Soc., Perkin Trans. 1 1299 (1994) 85. S Lavielle,N C Ling,R C Guillemin Carbohydr. Res. 89 221 (1981) 86. M Meldal, K Bock Tetrahedron Lett. 31 6987 (1990) 87. L L Likhosherstov, O S Novikova, V A Derevitskaya, N K Kochetkov Carbohydr. Res. 146 C1 (1986) 88. L Otvos Jr, L Urge, M Hollosi, K Wroblewsky, G Graczyk, G D Fasman, J Thurin Tetrahedron Lett. 31 5889 (1990) 89. Z-W Guo, Yu Nakahara, Jo Nakahara, T Ogawa Angew. Chem., Int. Ed. Engl. 36 1464 (1997) 90. M Matsuo, Yu Nakahara, Y Ito, T Nakada, Yo Nakahara, T Ogawa Bioorg. Med. Chem. 3 1455 (1995) 91. H Paulsen, A Schleyer, N Mathieux, M Meldal, K Bock J. Chem. Soc., Perkin Trans. 1 281 (1997) 92. O Seitz, C-H Wong J. Am. Chem. Soc. 119 8766 (1997) 93. S Nishimura, K B Lee, K Matsuoka, Y C Lee Biochem. Biophys. Res. Commun. 189 249 (1994) 94. N K Kochetkov Pure Appl. Chem. 56 923 (1984) 95. C Unverzagt, S Kelm, J C Paulson Carbohydr. Res. 251 285 (1994) 96. U Zehavi,M Herchman, R Schmidt, Th Bar Glycoconjugate J. 7 229 (1990) 97. P Westerduin, G H Veeneman, Y Pennings, G A van der Marel, J H van Boom Tetrahedron Lett. 28 1557 (1987) 98. C H Elie, H J Muntendam, H van der Elst, G A van der Marel, P Hoogerhourt, J H van Boom Recl. Trav. Chim. Pays-Bas 108 219 (1989) 99. A A Kandil, N Chang, P Chong,M Klein Synlett 555 (1992) 100. G H Veeneman, H J Brugge, H van der Elst, J H van Boom Carbohydr. Res. 195 C1 (1990) 101. G Lowe Chem. Soc. Rev. 309 (1995) 102. T Murata, T Usui Biosci. Biotech. Biochem. 61 1059 (1997) 103. R Liang, L Yan, J Loebach, M Ge, Y Uozumi, K Sekanina, N Horan, J Gildersleeve, C Thompson, A Smith, K Biswas, W C Still, D Kahne Science 274 1520 (1996) 104. Chem. Ind. News 76 14 (1998) 105. H M I Osborn, T H Khan Tetrahedron 55 1807 (1999) a�Russ. J. Bioorg. Chem. (Engl. Transl

ISSN:0036-021X    

出版商:RSC    

年代:2000

数据来源: RSC