摘要:

Russian Chemical Reviews 70 (9) 717 ± 734 (2001) Superconducting MgB2 and related compounds: synthesis, properties and electronic structure A L Ivanovskii Contents I. Introduction II. Magnesium borides: a brief history of investigations (1895 ± 2000) III. Magnesium diboride, a medium-Tc superconductor IV. Conclusion Abstract. magnesium in superconductivity of discovery The The discovery of superconductivity in magnesium diboride stimulated 2001, January in announced K), (Tc&40 40 K), announced in January 2001, stimulated intensive research into the nature of superconductivity and intensive research into the nature of superconductivity and electrophysical, magnetic and thermodynamic properties and electrophysical, magnetic and thermodynamic properties and spectroscopic characteristics of MgB compounds.related spectroscopic characteristics of MgB2 and and related compounds. The review concerns methods for the synthesis of new MgB The review concerns methods for the synthesis of new MgB2- based superconductors with more complex chemical composition. based superconductors with more complex chemical composition. Particular emphasis is placed on the results of investigations on Particular emphasis is placed on the results of investigations on the electronic structure and peculiarities of chemical bonding in the electronic structure and peculiarities of chemical bonding in these practical of aspects science Materials compounds. these compounds. Materials science aspects of practical utilisation utilisation of analysed.are superconductor new the on based materials of materials based on the new superconductor are analysed. Main Main results the in studies experimental and theoretical of results of theoretical and experimental studies in the above- above- mentioned includes bibliography The generalised. are fields mentioned fields are generalised. The bibliography includes 228 228 references. I. Introduction In his lecture at the International Symposium on Transition Metal Oxides (Sendai, South Korea, 2001) J Akimitsu announced 1 the discovery of superconductivity in MgB2 (for more detailed infor- mation, see Ref. 2). Magnesium diboride is a relatively poorly studied representative of a broad class of metal borides.3 ±14 The discovery of superconductivity inMgB2 has immediately attracted considerable interest in this system. The transition temperature for MgB2 (Tc&40 K) is almost twice as high as the values of Tc for all other binary superconductors, e.g., compounds and alloys with B1- (NbN, Tc&17.3 K) or A15-type (Nb3Ge, Tc&23K) structures.15, 16 From the viewpoint of its properties, magnesium diboride stands apart from all known borides.For instance, the values of Tc for binary d-metal borides,9, 17 as well as for ternary (LnRuB2, LnRh4B4) and pseudoternary [(Ln17xLn0y)Rh4B4] borides 18 lie between 40.5 and 8 K. The highest transition temperatures (Tc&16 ± 23 K) were found for intermetallic boro- carbides (IBC), namely, layered quaternary compounds of the LuNi2B2C type 19 ± 21 belonging to `conventional' superconduc- tors.22 A distinctive feature of magnesium diboride is that the value of Tc for this compound is `intermediate' between those of A L Ivanovskii Institute of Solid State Chemistry, Urals Branch of the Russian Academy of Sciences, ul.Pervomaiskaya 91, 620219 Ekaterinburg, Russian Federation. Fax (7-343) 274 44 95. Tel. (7-343) 274 53 31. E-mail: ivanovskii@ihim.uran.ru Received 19 April 2001 Uspekhi Khimii 70 (9) 811 ± 829 (2001); translated by A M Raevsky #2001 Russian Academy of Sciences and Turpion Ltd DOI 10.1070/RC2001v070n09ABEH000675 717 718 720 730 classical low-temperature (LTSC) and high-temperature super- conductors (HTSC). To emphasise this feature, MgB2 is some- times called a medium-Tc superconductor.It is of great importance that, unlike high-temperature superconducting com- plex oxides, MgB2 is a non-oxygen compound with very simple chemical composition and structure.3± 8 The discovery of superconductivity in MgB2 has raised a number of questions. 1. What is the nature of the superconductivity in MgB2? 2. Is magnesium diboride a unique compound of this kind or is it the first representative of a new class of superconductors? 3. What physical and chemical properties should related superconductors possess? Answers to these questions can predetermine a breakthrough in understanding the nature of the superconductivity in crystals and physical chemistry of inorganic compounds and in outlining possible application areas of new superconductors.In spite of short time that has elapsed since the discovery of superconductivity in MgB2, experimentalists and theoreticians have accumulated a great body of information about its properties. New superconductors have been found and superconducting materials based on MgB2 were obtained in the form of powders, single crystals, films, wires and composites. It was found that the superconductivity parameters of MgB2 remain stable in magnetic fields and upon neutron irradiation. Magnesium diboride is characterised by high critical transport currents and, at the same time, by their low sensitivity to grain boundary contacts. These and some other properties of this novel superconductor seem to be rather attractive for its practical utilisation.The main goal of this review is to generalise a considerable body of information about magnesium diboride, its functional properties and possible application areas accumulated to date. Taking into account a great deal of interest in MgB2, we have briefly outlined the history of investigations of the system Mg7B and reported the available data on the properties of magnesium borides. Particular emphasis is placed on the results of recent studies on the nature of the superconductivity in and physico- chemical properties of MgB2 (including the year 2001). The review concerns the results of both theoretical modelling and experimen- tal research into the electronic structure and the nature of chemical bonding in magnesium diboride and related compounds.Problems associated with search for new superconductors based on MgB2 and some materials science aspects are also outlined.718 II. Magnesium borides: a brief history of investigations (1895 ± 2000) Magnesium borides were first obtained by Moissan in 1895.23 Analysis of products of the interaction between metallic magne- sium and boric anhydride revealed two magnesium borides with strongly different chemical stabilities. One of the borides was found to decompose in acids to give boranes, mostly B4H10. More recently, this boride, with the chemical composition Mg3B2,24 ± 28 was studied 24 for use as starting material in the synthesis of boranes. The assumption of the existence of the second boride, Mg2B4 (MgB2), was proposed 29 based on the results of analysis of boride hydrolysis products.Studies carried out by researchers from the USSR3, 8 and other countries 4±7 over a period of two decades, from 1950s to 1970s, serve as a basis for present-day knowledge of the system Mg7B. Magnesium borides can be obtained by (i) reduction of B2O3 with metallic magnesium,23 (ii) direct interaction of elemen- tal Mg and B, (iii) reduction of MgO with boron, (iv) reduction of Na2BF6 with magnesium and (v) by reactions between Mg and BCl3 and between MgO and boron carbide.3, 8 Pioneering studies on the phase equilibria in the system Mg7B and physicochemical properties of particular magnesium borides also date back to the above-mentioned two decades.1. Phase relations in Mg±B system No detailed phase diagram of the system Mg± B is available. Information about stable magnesium borides is contradictory (see, e.g., Refs 3 ± 13, 30). On the one hand, the occurrence of four phases (MgB2, MgB4, MgB6 and MgB12) is assumed.30 On the other hand, the assumption of the occurrence of the lowest magnesium boride, Mg3B2, was rejected.31 Six magnesium boride phases (MgB2, MgB4, MgB6, a variable composition phase MgB6 ±MgB17 and two phases of unknown composition) have also been reported.32 The following phases were found in the phase diagram of the system Mg± B reported in monograph 33: MgB2, MgB4 and MgB7 (Fig. 1). Below we will consider only those magnesium borides for which reliable structural informa- tion is available.T /8C 2000 1500 gas 1000 liquid 500 solid 80 60 0 100 40 20 B c (at.%) Mg Figure 1. Phase diagram for the Mg±B system.33 2. Crystal chemistry of magnesium borides Pioneering studies on the crystal structure of MgB2 date back to the early 1950s.34 ± 40 Magnesium diboride has a hexagonal AlB2- type structure. The space group is P6/mmm (D16h), the number of formula units in the unit cell Z=1. Boron atoms occupy positions at the centres of trigonal prisms while magnesium atoms occupy positions at their vertices. Such prisms share all faces, thus forming a three-dimensional packing. The structure of MgB2 can be represented as alternating planar hexagonal Mg layers and graphite-like (honeycomb) boron sheets (Fig.2). The alternation pattern is Mg± B2±Mg ±B2... The coordination number (CN) MgB2 MgB4 MgB7 A L Ivanovskii a d c b Mg B e a kz c S H AR P SS0 ky T HP DG T K LUM S K T 0 kx Figure 2. Structure of MgB2. Types of atomic packing: interlayer packing, side view (a); in-plane packing, top view (b); packing of atomic layers, side view (c); primitive unit cell (d) and the Brillouin zone (e). of magnesium atoms is 20 and their coordination polyhedron (CP) is [MgB12Mg8]. For boron atoms, CN=9 and the CP is [BMg6B3]. The atomic positions in the unit cell are as follows: Mg, 1a (0,0,0) and B, 2d (1/3, 2/3, 1/2) and (2/3, 1/3, 1/2). The lattice constants are a = 3.0834 and c = 3.5213 A; c/a = 1.142.The pÅÅÅÅÅinteratomic distances are B7B (a/ 3)=1.780, pÅÅÅ B7Mg ( a2/3+c2/4)=2.503, Mg7Mg (in-plane)=3.083 and Mg7Mg (interlayer)=3.520 A.10, 38 ± 40 Currently, more than a hundred of binary compounds with the AlB2-type structure are known, including diborides of p- and d-elements.3 ± 13, 38 ± 40 Nevertheless, the problem of the existence of general physical or chemical criteria determining the conditions for the formation of structural types of various borides remains a moot question as yet. From the standpoint of crystal chemistry,10 structural types of borides can be explained taking into account the ratio of atomic radii (r) of the components M/B (M is metal). According to the theory of symmetry of close crystal packings of spheres, an `ideal' value of rM/rB for trigonal prismatic coordina- tion is 1.89.For stable hexagonal MB2 phases, the ratio rM/rB varies from 1.14 to 2.06 38 (e.g., rM/rB=1.571 for aluminium diboride and 1.758 for magnesium diboride). Pearson 40 proposed an empirical criterion for the existence of AlB2-type structures, namely, a correlation between structural parameters (c/a) of stable phases and the ratio rM/rX (here, X is an element whose atoms can form graphite-like sheets; in addition to boron, the set X can include Be, Si, Ga, Hg, Zn, Cd, Al, Cu, Ag and Au, see Refs 38 ± 40). The ratio c/a varies from 0.59 to 1.2 (Fig. 3). As can be seen, the point corresponding to MgB2 (c/a=1.142) is located near the upper critical bound of structural stability.According to Pearson,40 the occurrence of hexagonal CaB2, which is the closest analogue of superconducting MgB2 (see below), is impossible. Magnesium tetraboride was first synthesised in 1953.35 The crystal structure and properties of this compound have been studied in detail.31, 41 ± 43 Crystals of MgB4 belong to orthorhom- bic system (the space group Pnam,Z=4) and are characterised by the unit call parameters a=5.464, b=7.472 and c= 4.428 A.42, 43 The structural motif can be represented as penta- gonal pyramids built of boron atoms, which form extended chains. Magnesium atoms occupy positions between boron poly- hedra. The shortest distances are Mg7B=2.392 and Mg7Mg=3.075 A. The atomic positions in the unit cell are as follows: Mg, 4c: x, y, 1/4 (x=0.051, y=0.136); B1,2, 4c:Superconducting MgB2 and related compounds: synthesis, properties and electronic structure c/a MgB2 1.1 12 0.9 0.7 0.5 1.2 1.6 1.4 1.0 rM/rX Figure 3. Dependence of the ratio of lattice parameters (c/a) on the ratio of atomic radii (rM/rX) for AlB2-type binary phases MX2;40 c/a: <0.9 (1) and >0.95 (2).(x1=0.725, y1=0.657; x2=0.559, y2=0.854) and B3, 8d: x, y, z (x=0.130, y=0.434, z=0.442).10 This structural type is unique in its kind since most of binary metal tetraborides 5 ± 9, 12, 13 belong to the CrB4 and ThB4 structural types. The existence of magnesium hexaboride,MgB6, also remains a moot question as yet. It is thought that such a hexaboride represents a mixture of magnesium heptaboride MgB7 and magnesium tetraboride MgB4 while MgB9 represents a mixture of MgB7 and elemental boron.44 The structure of magnesium heptaboride MgB7 (Mg2B14) is characterised by a CNof 28 for the magnesium atom.45 Similar structural elements are also inherent in MgAlB14 (Mg17xAl1+yB14), LiAlB14 and NaB15.10 On the other hand, MgB7 and Mg2B14 were described as individual phases with different lattice constants.13 3.Magnesium borides: properties, electronic structure and chemical bonding Information about physicochemical properties of magnesium borides is scarce.3, 8, 9, 11, 13 Only hydrolytic stability of magnesium borides and their reactions with acids have been studied in detail. It was found that boranes are formed only in the reactions with MgB2.From the viewpoint of its reactivity and solubility in acids, this compound differs strongly from both higher magnesium borides and most of d-metal diborides which possess high chem- ical stability. Some thermochemical and thermal properties of MgB2 (heat capacity, enthalpy, entropy and dissociation pressure, heats of formation) have been reported.8, 9, 13 To our knowledge, no studies on the electronic structure and the nature of chemical bonding in magnesium borides, except for MgB2 and `MgB6', have been carried out so far. By analogy with other metal ± boron systems, it can only be assumed that covalent B7B bonds (within boron polyhedra) should have increased importance in the overall system of interatomic interactions responsible for improvement of thermomechanical properties and chemical stability of magnesium borides on going from lower borides to higher (from MgB2 to `MgB12').The energy band structure of MgB2 was studied 46 ± 48 by the semiempirical LCAO,47 tight binding 48 and non-self-consistent orthogonalised plane wave (OPW) methods.46 The structures of the valence bands of MgB2 and graphite were found to be similar.46 The major contribution to the valence band of magne- sium diboride comes from the combination of boron ps,p-orbi- tals.47 Charge transfer occurs from Mg to B. The position of the bottom of the conduction band is determined by B boron pp-orbitals. Comparison of the energy band structure for MgB2 and the hypothetical crystal &B2 with `empty' cation sublattice allowed one to draw a qualitative conclusion about the dominat- ing role of B7B interactions in the overall system of chemical bonds.The energy band structure as well as the chemical bonding and energy parameters of MgB2 obtained from the first nonempirical calculations carried out by the self-consistent full-potential linear 719 a N(E) /Ry71 EF 200 b N(E) /Ry71 EF 200 0.5 E /Ry 0 Figure 4. Total densities of states for MgB2 (a) and AlB2 (b) obtained from FLMTO calculations.49 muffin-tin orbital (FLMTO) method 49 were analysed and com- pared with the corresponding characteristics of isostructural borides (AlB2 and TiB2). Among the known diborides MB2, titanium diboride exhibits exceptionally high thermomechanical and strength properties.7 ±9 The total densities of states (DOS) for MgB2 and AlB2 are shown in Fig.4. The energy band structure consists of two broad bands separated by a DOS minimum (a pseudogap). A qualitative explanation for differences between the properties of MgB2 and AlB2 was obtained 49 in the framework of the rigid band model (RBM) 50 ± 52 based on the structure of the valence band. According to this model, both chemical stability and thermomechanical properties of MB2 are related to the parameter V à Vb, Vv where Vb is the width of the filled band of electron states and Vv is the total width of the valence band (up to the DOS minimum). An optimum stability corresponds to V=1 (i.e., all bonding bands are filled while all antibonding bands are empty).This condition holds for such d-metal diborides as TiB2, ZrB2 and HfB2 [the valence electron concentration, VEC, is 3.33 e per atom per formula unit (see Refs 50 ± 57)]. For MgB2 (VEC=2.67),49 the calculated values are Vv=1.011, Vb=0.912 Ry, V=0.902. For AlB2 (VEC=3.0 e per atom), Vb=Vv=1.146 Ry, V=1.0, which means that this diboride should exhibit the highest stability. Analysis of chemical bonding in MgB2, AlB2 and TiB2 using charge density maps (Fig. 5) shows that the overlap of the B7B wave functions in the plane of graphite-like sheets decreases from MgB2 to TiB2.49 For TiB2, the most pronounced is the effect of interlayer Ti7B covalent bonds, while MgB2 is characterised by retention of spherical symmetry of the cation charge density contours.The cohesive energies (Ecoh),{ and the energies of particular (B7B, M7B and M7M) bonds were calculated (Table 1). As can be seen, the major contribution to the cohesive { As is known, Ecoh describes the overall effect of chemical bonding in crystals. Table 1. Cohesive energies (Ecoh) and energies of particular bonds (Eb) for magnesium, aluminium and titanium diboride /Ry per unit cell.49 Parameter TiB2 AlB2 MgB2 1.58 1.42 1.26 Ecoh 0.36 0.86 0.36 0.28 0.83 0.24 0.11 0.86 0.29 Eb (seea) M7M B7B M7B a For calculation procedure, see Refs 53 ± 57.720 1 B 2 Mg 3 Mg Figure 5. Charge density distributions for magnesium (a), aluminium (b) and titanium diboride (c): in-plane for boron sheets (1), in-plane for metal atom layers (3) and between boron sheets and metal layers (2).49 energies of MgB2 comes from the B7B-interaction (68%).The contributions of other types of chemical bonds are much smaller (23% for B7Mg and 9% for Mg7Mg bonds). An increase in the cohesive energy on going from MgB2 to AlB2 is due to strengthen- ing of the M7M bonds while the B7B and M7B binding energies vary only slightly. This is the main distinction between these compounds and d-metal diborides for which the value of Ecoh depends first of all on the `interlayer' M7B interac- tions.53 ± 57 4. Application areas of magnesium borides Information about application areas of magnesium borides con- cerns mainly MgB2.The most attractive properties of this com- pound are its mechanical characteristics and high reactivity. Magnesium diboride is used as an abrasive; polycrystalline composites based on MgB2 and boron nitride were proposed to be used as materials for cutting devices.13 As mentioned above, magnesium boride was used as starting material to obtain boranes following the hydrolytic decomposition reaction of the boride. Currently, the role of MgB2 (i) as precursor in reactions of fast chemical exchange in the synthesis of refractory d-metal borides and ceramics based on them and (ii) in the preparation of close- packed boron nitride polymorphs has been studied intensively. The known methods for the preparation of cubic boron nitride (c-BN), a material with record hardness, are based on heat treatment of hexagonal BN (h-BN) under pressure or precipita- tion from melts.Crystallisation and nucleation of c-BN are strongly affected by various catalysts, e.g., C, B, Fe, Co, Ni, Cu, Al, their alloys and mixtures, as well as metal nitrides, oxides and borides.58, 59 Magnesium diboride is an efficient catalyst of spontaneous crystallisation of c-BN.13, 60 ± 68 Studies of c-BN nucleation from melts in the system MgB27BN at pressures up to 6.8 GPa and temperature up to 2000 K allowed one to determine the threshold pressure (P&4.5 GPa) for the lower temperature bound of the c-BN phase formation (T&1630 K).67 The role of small additives of the third component (P, Si) on the crystallisation kinetics of c-BN in pseudoternary systems MgB2 ± BN± P (see Ref.65) and MgB2±BN±Si 63, 64 has been studied. Magnesium diboride and higher magnesium borides occur among reaction products if refractory d-metal borides are obtained by reduction of corresponding oxides by magnesium- reduced borona b c B B BB BB BB Ti Al Ti Mg Al Ti Al A L Ivanovskii MOy+B MBz+B2O3 or by magnesium reduction of such oxides MgO+MBz. MO+Mg+B2O3 At the formation temperatures of refractory borides MBz, magnesium borides thermally decompose into boron and magne- sium, the latter being removed nearly completely.13 The procedure for preparation of dispersed ZrB2 powders 69 is based on com- petitive reduction processes of ZrO2 and boric anhydride in the system ZrO2±Mg±B2O3 (at 1100 8C) and boronation of a-Zr and Mg in the system ZrO2 ±Mg ± amorphous boron (at 700 8C) followed by removal of MgO and MgB2.A thermal reaction in the system TiO2±Mg±B2O3 in Ar atmosphere results in titanium diboride as the major product and magnesium oxide and magne- sium borate (Mg3B2O6) as side products.70 Metal borides, solid solutions based on them and boride ceramics can also be synthes- ised by solid-state reactions of fast chemical exchange with participation of MgB2.71 ± 73 A mechanochemical reaction in a Ti ± 4% MgB2 mixture, resulting in a nanocrystalline Ti matrix with inclusion of MgB2 grains of size 20 to 300 nm has been studied.74 Subsequent sintering of the reaction product at 600 8C resulted in a new compound with NiAs-type structure and unknown composition.Along with TiB2, B4C, CrB2, BN and amorphous boron, MgB2 was announced to be an antioxidant, i.e., an additive to carbon-containing refractory materials used in metallurgy to prevent them from breakdown upon oxidation of carbon in the surface layer.75 Oxidation and phase formation in the mixture of MgB2 synthesised in the reaction MgB12+Mg?MgB2 with periclase ± carbon composite have been studied as functions of temperature.75 The process is characterised by exothermic effects, begins at *470 8C and results in MgO, Mg2B2O5 and boric anhydride. Similar magnesium polyborides (mixtures of amor- phous boron and MgB12 obtained by magnesium reduction of B2O3) were found to behave similarly.According to DTA, the mass increased by 43%. Amorphous boron of purity 95% to 98% exhibited the best antioxidant properties; in this case, a mass increase of 124% to 147%. The results of applied research on the use ofMgB2 as a catalyst in the synthesis of c-BN,76 as antioxidant additive to refractory materials 77 ± 79 and for production of dispersed alloys 80 and SiC powders (see Ref. 81) were patented. Prospects for practical utilisation of superconducting MgB2 will be considered below. III. Magnesium diboride, a medium-Tc superconductor The discovery 1, 2 of superconductivity in MgB2 and publication of pioneering studies,82 ± 87 in which the nature of the effect was rationalised in the framework of electron ± phonon mechanism of superconductivity, has stimulated discussion about possible ana- logues of superconducting magnesium diboride. Both chemical composition and crystal structure aspects of the problem were considered and some other factors responsible for superconduc- tivity were analysed.Chemical analogues of MgB2, that is, binary and multicomponent borides, have been considered in the preced- ing Section of this review. Considerable attention has been given to the role of quasi- two-dimensional (quasi-2D) structure of magnesium diboride. It has been generally thought that high lattice symmetry (preferably, cubic) is favourable for superconductivity. This can be exempli- fied by elemental superconductors (Nb, La under pressure and fullerite C60 with the transition temperatures of 9, 13 and 33 K, respectively), binary compounds (ZrN, NbN and Nb3Ge with Tc of 10.7, 18 and 23 K, respectively) and a number of ternary compounds.88 ± 90 Recently, a group of 2D-HTSC has been found.Oxides with CuO2 layers in their structures can serve as the best knownSuperconducting MgB2 and related compounds: synthesis, properties and electronic structure example of these compounds. Quasi-2D b-nitrides ZrNCl (Tc=15 K) and HfNCl (Tc=25 K) also represent layered struc- tures with alternating (Zr, Hf)7N and chlorine layers. Interlayer interactions in these compounds are weak and superconductivity is observed upon intercalation with alkali metal atoms.91 ± 93 A possible mechanism of superconductivity in anisotropic media taking into account the quasi-2D character of the energy bands has been discussed.94 Special attention has been given 87 to similarity in the atomic structure and electronic properties of MgB2 and graphite.Both compounds are isoelectronic [the ionic formula of magnesium boride is Mg2+(B2)7], have layered structures and exhibit a similar type of chemical bonding with strong in-plane s-bonds and weak interlayer p-bonds. Moreover, electron-doped interca- lated graphites (e.g., LiC8 and KC8) are superconductors (up to Tc=5 K).95, 96 There is also evidence 97 for substantial enhance- ment of superconductivity in specially treated fullerite C60 to Tc*52 K (cf.Tc433 K for the known medium-Tc supercon- ductors based onM3C60, whereMis an alkali metal), which points to potentially significant role of intercalation. The nature of these analogies can be understood by analysing the band structure and phonon spectrum of magnesium diboride. Before considering them, let us discuss methods for the synthesis of MgB2. 1. Synthesis and properties of superconducting MgB2 a. Synthesis of MgB2 Currently, superconducting MgB2 is mainly obtained as poly- crystals by solid-phase synthesis from elemental B and Mg.1, 2, 82 ± 84, 98 ± 120 Synthetic procedures differ from one another in composition of starting materials, atmosphere, and heat treat- ment regimes. For instance, MgB2 samples were synthesised 84 from magnesium flakes and sub-micron powder of amorphous boron which were pressed in pellets, placed on Ta foil and subsequently fired in a tube furnace under inert atmosphere (95% Ar+5% H2) at 600 (1 h), 800 (1 h) and 900 8C (1 h) and then lightly ground.The resulting powders were hot pressed (10 kbar, 650 ± 800 8C) for periods between 1 and 5.5 h. The samples thus obtained were characterised by multiphase compo- sition; however, they exhibited high critical currents. Other sintering conditions are listed in Table 2. For instance, Mg(11B)2 was prepared 101 from a boron powder enriched with 11B and Mg lumps placed in a BN crucible and heat-treated for 1.5 h in Ar atmosphere (P=50 bar). No chemical composition of the Table 2. Conditions for the synthesis of superconducting MgB2 powders.Phase composition Sintering conditions Atmosphere (sheath) time /h T /8C MgB2+MgO (impurity) 92%Ar+8%H2 (Ta foil) 600+800+950, after compaction 950 950 Mg(10B)2 850 Mg(11B)2 single-phase Mg(11B)2 2+2+1, 22246 900, after compaction 1000 1200 7 7 3.518 Inert atmosphere (a Ta tube placed in a quartz ampoule) Ar (Ta crucible) Inert atmosphere (a BN crucible on Ta foil placed in a quartz ampoule) Ar (a Ta tube in a quartz ampoule) MgB2 (AA) a 500 900 Ar (Nb tube) MgB2+impurities (1072 at.% Ni, Cu, W) 7 7 7 7 104 172 1 ± 3 aCommercially sold MgB2 powder [Alfa Aesar Co, USA, assay (purity 98%)].sample obtained was reported. The superconductivity parameters were Tc=39 K and the width of the transition DTc=0.4 K.Magnesium diboride samples were also synthesised 82 from a Mg: B=2 : 1 mixture in a Ta tube placed in a quartz ampoule in Ar atmosphere. The mixture was heated to 950 8C and kept 2 h at this temperature. The procedure proposed by Gerashenko et al.99 involved pressing of magnesium and amorphous boron into pellets which were placed in a Ta crucible and then subjected to two-step annealing in purified helium flow. At first, the pellets were annealed at 800 8C for 1 h. The resulting phase contained MgB2 (78.6%), Mg (16.6%), MgO (4.7%) and BN (traces). Subsequent annealing at 875 8C (1 h) led to evaporation of the unreacted metallic magnesium. The final phase was composed of MgB2 (89.9%), MgO (5.2%) and BN (1.7%) and characterised by Tc = 39.5 K and DTc=2.5 K.Thus, superconducting magne- sium diboride powders can be obtained using rather simple synthetic procedures which do not require costly starting materi- als and equipment. b. Physicochemical properties of superconducting MgB2 The mechanism of superconductivity in MgB2 was elucidated by measuring the isotope effect on Mg(10B)2 and Mg(11B)2 pow- ders.82 Replacement of 10B by the heavier isotope, 11B, led to a decrease in Tc from 40.2 to 39.2 K and to narrowing of the transition region by 0.1 K (Fig. 6). The isotope coefficient, aB=¡ Dln Tc , D lnMB whereMB is the magnetisation of the boron sublattice, was found to be *0.26. This is close to the aB values for IBC,105, 106 the superconductivity of which is due to electron ± phonon interaction (with high-frequency optical B A1g modes) and is described in the framework of the Bardeen ± Cooper ± Schrieffer (BCS) model.22 Experiments with different samples 102, 107, 108 revealed rather high critical current densities (jc) (26104 ±105 A cm72).It was found that the temperature dependences of the critical fields Hc2 and Hc1 near Tc have a slope of 0.012 T K71 and that the magnetisation curves are reversible.107, 108 The Hc2 anisotropy (1.73) was first measured 108 on aligned crystallite samples. The temperature dependence Hc1(T) was reported to be linear for the whole temperature region in which superconductivity is observed.109 At T>Tc, the temperature dependence of the resistivity has the form 110 721 Ref.Lattice parameters /A Tc /K a c 100 37 3.5191(7) 3.0815(6) 82 40.2 3.51930.03 3.14320.03 98 7 7 39.2 101 7 7 39 102 7 7 7 7 103722 b a M/H M/H 70.2 1 2 1 2 70.6 71.0 36 40 T /K 40 38 38 36 Figure 6. Isotope effect in magnesium diboride.82 (a) Magnetisation divided by magnetic field strength (M/H) as function of temperature for Mg(11B)2 (1) and Mg(10B)2 (2); (b) the same for Mg(10B)(11B) (1) and sum of Mg(11B)2 and Mg(10B)2 data (2). r(T)&T2. It was found that magnetic field has no effect on r. The Hall coefficient 111 RH of magnesium diboride is positive (4.1610711 m3 deg71 at 100 K and a hole concentration, n, of 1.561023 cm73) and increases in the temperature range from 40 to 300 K.An important distinction of superconducting MgB2 from cuprate HTSC has been pointed out.84 Due to high anisotropy, cuprate HTSC rapidly lose their performance in external fields, while the supercurrents flowing throughout even untextured, multiphase MgB2 samples are little sensitive to weak magnetic fields. The spin ± lattice relaxation rate 99, 112 and the Knight shift 99 were measured by NMR spectroscopy; the behaviour of the former parameter was interpreted within the framework of s-wave model in the strong electron ± phonon coupling limit.112 Magnesium diboride was also studied by the scanning tunnel- ling microscopy.113 ± 115 Particular attention was paid to the current-vs.-voltage characteristics used for the determination of the width of the superconducting energy gap (D).Large scatter of the gap widths D (from 5 to 7 meV, see Refs 113 ± 115) was first of all explained by different compositions of the samples studied. The gap widths D were also evaluated from the tunnelling spectra of Au/MgB2 point-contact tunnel junctions 116 (D=4.3 meV) and IR reflectance spectra 117 (D=4.5 meV). The specific heat of magnesium diboride and its dependences on the magnetic field and temperature were measured,118, 119 as well as some other thermodynamic and mechanical characteristics [e.g., the elasticity modulus B0=151 5 GPa (see Ref. 120)]. The possibility of self-consistent interpretation of the results of the above-mentioned experiments in the framework of the model of electron ± phonon mechanism of superconductivity in magnesium diboride 82, 85 ± 87 should be emphasised.On the other hand, Hirsch 121 proposed a `universal' model of `hole super- conductivity' in MgB2,122 which implies pairing of spin carriers driven by the Coulomb interaction. However, the results obtained upon treatment of MgB2 according to Hirsch 121 in order to increase Tc (e.g., by doping with aluminium or by hydrostatic lattice compression) contradict the conclusions reported in theo- retical studies based on the electron ± phonon mechanism of superconductivity. 2. Electronic structure and superconductivity of MgB2 According to McMillan,123 the critical temperature in crystals with electron ± phonon mechanism of superconductivity can be determined using the formula Tc&hoiexp[f(l)], where hoi is the averaged phonon frequency (which is inversely proportional to the masses of the atoms), l is the electron ± phonon coupling constant [l&N(EF)hI 2i],N(EF) is the density of states (DOS) at the Fermi level and hI 2i is the electron ± ion matrix element determined by the bond ionicity.Information about the electronic structure and vibrational states of MgB2 is of crucial importance for correct interpretation of superconductivity in and other properties of MgB2 as well as for targeted search for new superconductors. a. Electronic structure The discovery of superconductivity in MgB2 stimulated intensive theoretical 85 ± 87, 124 ± 130 and experimental 131 ± 133 studies of the electronic structure of magnesium diboride 85 ± 87, 124 ± 133 and a large number of stable and metastable isostructural diborides of Li, Na,87, 125, 126 Be, Ca,124 Al,124, 126 Sc, Y, Ti, V,124, 134 Ag and Au,127 as well as some related phases [MgB6, CaB6,128 CaSi2, CaSiBe,125 ZrBe2, HfBe2, CaGa2,127 Mg3BN3 (see Ref.129)]. Calculations were carried out using rigorous, self-consistent density functional approach. The valence band structure of MgB2 mined 85 ± 87, 124 ± 126 by the B 2p states which form four s(2px,y)- bands and two p(2pz)-bands (Fig. 7). They exhibit strongly differ- ent dispersion E(k). For the B 2px,y-bands, the dispersion E(k) is maximum along the kx,y direction (G± K).These bands reflect the distribution of the B states within graphite-like sheets, which have a quasi-2D type and form flat zones along the kz direction (G± A). Two B 2px,y-bands cross the Fermi level and contribute largely to the total density of states at the Fermi level [*30% (see Ref. 85)]. These bands are responsible for the metal-like properties of magnesium diboride. The Fermi level is located in the region of bonding states. Magnesium diboride exhibits hole-type conduc- tivity, which is consistent with the results of Hall coefficient measurements.111 Charge transfer occurs from magnesium to boron. One of the most distinctive features 85 ± 87, 124 ± 130 of the electronic structure of MgB2 is the position of the B 2px,y energy bands.Namely, they are above the EF at the G point of the a E /eV 20 72 74 76 78 c E /eV 20 72 74 76 78 e E /eV 20 72 74 76 78 M K G G A Figure 7. Energy band structures for MgB2 (a), CaB2 (b), BeB2 (c), AlB2 (d), ScB2 (e) and YB2 (f).124 A L Ivanovskii is deter- bdf L M K L G G ASuperconducting MgB2 and related compounds: synthesis, properties and electronic structure Brillouin zone and form two hole-type cylinders on the Fermi surface.85 ± 87 The B 2pz-like states have a 3D-type, are oriented perpendicular to the boron sheets and are responsible for weak interlayer p bonds.85 ± 87, 124 ± 126 Both p-bands cross the Fermi level and are characterised by the maximum dispersion along the kz direction (G± A).The Mg s, p-states and B s-states are mixed with the system of B 2p-like bands near the valence band edge and in the conduction band. The determining role of boron in the formation of the electronic structure of MgB2 has been partic- ularly emphasised by Kortus et al.85 who titled their communica- tion `Superconductivity of metallic boron in MgB2'. The electronic structure of magnesium diboride was studied by X-ray emission spectroscopy (XES),131, 132 X-ray absorption spectroscopy (XAS)131, 132 and photoemission spectroscopy (XPS).131, 133 Comparison of the C Ka XES lines (1s?2p-tran- sitions) for graphite and corresponding B Ka lines for MgB2 and AlB2 (Fig. 8) clearly demonstrates differences in positions of their emission maxima determined by the pxy-states.135 This supports the idea 85 ± 87, 124 ± 130 of correlation of the superconductivity in MgB2 with the relative positions of the s- and p-bands (the presence of hole-type s-states).The results of XES and XAS studies 132 of the structure of B bands near the Fermi level are in excellent agreement with the results of calculations (Fig. 9). The spectra exhibit a number of resonance lines indicating the presence of a protective oxide film formed on the sample surface upon exposure to air. 1 2 710 715 720 Figure 8. X-Ray emission CKa spectra (1s?2p-transitions) for graphite (1) and corresponding BKa spectra for AlB2 (2) and MgB2 (3).131 1 2 74 78 Figure 9. X-Ray emission (1) and absorption (2) spectra of boron in MgB2 (a) and the DOS distribution for boron (b): theoretical (solid lines) and corrected for the lifetime of excited state, thermal and instrumental broadening of the experimental setup (points).132 Intensity (rel. u.) Intensity (rel.u.) Density of states (rel. u) EF 3 0 Eb /eV 75 a EF b EF 4 E /eV 0 High-resolution XPS studies of magnesium diboride allowed direct observation 133 of the superconducting energy gap (*4.5 meV at 15 K). This gap (*3 meV at T=0 K) is of the s-wave character and its temperature dependence is consistent with the BCS theory. Band structure calculations were carried out for magnesium diboride and isostructural diborides of Li, Na,87, 126 Be, Ca,124 Al,124, 126 Sc, Y, Ti, V,124, 134 Ag and Au (see Ref.127). The band structures are compared in Figs 7 and 10. The absence of super- conductivity in BeB2 (see Ref. 136) was assumed 124 to be due to the low-energy shift of s-bands and the absence of hole states at the G point, which changes the Fermi surface topology in such a manner that cylinders (along the G±Adirections) are transformed into cones. The absence of superconductivity in AlB2 is due to filling of s-bands caused by the VEC increase.83 Variations of superconducting properties for a number of d-metal diborides can be rationalised based on the results of calculations.53 ± 55, 124, 134 For ScB2 and YB2, the 2D-pxy-bands at the A point are above the EF (see Fig. 7), but the concentration of hole carriers is low.It is important that metal d-states contribute largely to N(EF). As a result, these diborides are characterised by low Tc (1.5 ± 2.0 K).9, 17 Among hexagonal transition metal diborides, the lowest Tc values should be exhibited by TiB2, ZrB2 and HfB2, for which the Fermi level is located in the pseudogap.50 ± 55 As electron concentration increases, the d-band is filled andN(EF) increases 50, 52 ± 55 (e.g., for Group V and VI d-metal diborides). This can lead to an increase in Tc ; e.g., a rather high Tc*9 K was reported for TaB2.137 Analysis of the energy band structures of metastable diborides of Li, Na,87, 126 Ca,124 Al,124, 126 Ag and Au (see Ref. 127) showed that the s- and p-band configurations and band filling types a E /eV 72 73 78 713 b E /eV 50 75 710 715 c E /eV 61 74 79 A M K G G Figure 10.Energy band structures for CaGa2 (a), AgB2 (b) and ZrBe2 (c).127 723 A L H724 calculated for CaB2 are the same as the `optimum' parameters of superconducting MgB2. Calcium diboride was suggested 124 to be the most probable new superconductor. Estimates 124 of the enthalpy of formation of CaB2 (70.12 eV per formula unit) showed the possibility of real synthesis of this compound. It is thought that calcium diboride can be stabilised by, e.g., doping with isoelectronic or hole dopants (Mg, Li, Na) or by forming layered superstructures, e.g., Ca ± B2±Mg ±B2 ± Ca. The elec- tronic properties of ternary systems based on MgB2 will be considered in the next Section of this review. Here, we will only mention that an increase in the unit cell volume will assist the formation of the electronic structure of diborides in a manner favourable for superconductivity.This can be done by, e.g., isoelectronic replacement of magnesium by cations with larger atomic radii (Ca, Sr, Ba).124 Structural similarity between MgB2 and hexagonal BN has been pointed out.129 Like MgB2, boron nitride also contains planar boron sheets while its structural stability is due to strong in-plane covalent bonds.14, 138 The energy band structures of MgB2 and solid solutions of composition MgB27yNy, where boron was partially replaced by nitrogen, were compared.129 Solid solutions with y=0.25 and 1.00 were studied.The solution MgB1.0N1.0 (in which the composition of the non-metal sublattice formally matches the composition BN) can be considered as a model of magnesium boronitride prepared by intercalating atomic h-BN sheets with hexagonal magnesium layers. The energy band structure of the individual ternary phase Mg3BN3 with separated layers of different-type metalloids (B and N) was calculated.139 An increase in the N/B ratio in the composition of solid solutions leads to a shift of the Fermi level towards higher values, filling of antibonding states and a decrease in N(EF). The bonding B px,y-like bands become completely filled while the hole states are absent. According to calculations,129 Mg3BN3 is a semiconductor [transition at the G point of the Brillouin zone, the forbidden gap (FG) width is 0.88 eV (0.065 Ry)].The valence band structures of MgB2 and Mg3BN3 are fundamentally different, viz., the deter- mining role in the mechanism of superconductivity in MgB2 is played by the B 2p-bands while for Mg3BN3 these bands lie much below EF and contribute negligibly to the region of the spectrum near the Fermi level. The energy band structure of metastable cubic magnesium hexaboride (the space group Pm3m) has been studied.140 The optimised lattice parameter corresponded to a=4.115 A. The main building blocks of the MgB6 structure are B6 octahedra. The energy band structure contains 10 valence bands of hybrid B 2s,p- states, which are responsible for B7B bonds both within and between B6 octahedra.The upper filled band forms a flat zone along the X± G direction and is composed of B px,y-orbitals involved in B7B interactions between B6 octahedra. The first unfilled band is characterised by strong k-dispersion and is composed of contributions of cation states. Unlike CaB6 (FG&0.0163 Ry, direct transition at the G point), magnesium hexaboride is a narrow-gap semiconductor with indirect transi- tion (FG&0.0150 Ry, X?R transition).10, 141 ± 144 Despite the determining role of B7B bonds in MgB2 and MgB6 crystals, they possess basically different electronic properties that is, metallic for MgB2 and semiconducting for MgB6. `Non-boride' AlB2-type phases with silicon, beryllium or gallium graphite-like sheets have been reported [CaSi2, CaSiBe,125 ZrBe2, HfBe2, CaGa2 (see Ref. 127)].Rhombohedral CaSi2 (the space group R3m) obtained under equilibrium conditions is known to be a nonsuperconducting semimetal.145 Heat treatment of CaSi2 under pressure can cause a polymorphous transformation into an AlB2-type structure.146 Similarity between the energy band structure of this disilicide and those of sp-element diborides and CaGa2 (see Fig. 10)127,146 was found.125 The main distinctions are due to Ca d-orbitals which are mixed with the Si states near the bottom of the conduction band due to hybridisation. The antibonding p*-bands are partly filled, which can be the reason for structural instability of the high- A L Ivanovskii pressure phase under normal conditions.Stabilisation of hexago- nal CaSi2 by replacement of Si by Be was studied taking calcu- lations of a model phase CaSiBe as an example. It was assumed 125 that (i) replacement of Si by an atom of smaller radius (cf. rSi=1.34 and rBe=1.13 A) will produce `chemical pressure', which is favourable for stabilisation of the AlB2-type structure, and (ii) partial depletion of the p*-band with a decrease in the electron concentration in the system will an additional stabilising factor. As was found, stabilisation of the AlB2-type structure requires the reduction of the VEC in the system by*0.7 e.125 The electronic properties of HfBe2 and ZrBe2 are fundamentally different from those of superconducting MgB2 (see Fig. 10). b. Electronic structure and properties of superconducting MgB2 Some properties of MgB2 were explained using information about its electronic structure obtained from calculations. Kortus et al.85 estimated the constant l (*1) and phonon frequencies (300 ± 700 cm71) and calculated the Tc value within the rigid atom sphere model.Their results were found to be consistent with the experiment.2 The ab initio approach was also successfully used in phonon spectrum calculations, simulation of the effect of heat treatment on superconductivity in magnesium diboride and calculations of its elasitc characteristics (see below). Calcula- tions 134 of the electric field gradients (EFG) reflect asphericity of the electron density distribution in the vicinity of atomic nuclei and are closely related to the symmetry of their nearest environ- ment, electronic structure and the nature of chemical bonding.Experimental EFG values are obtained in studies of quadru- pole effects in the NMR spectra of MgB2 (see Refs 99, 112) and some other diborides.147 Analysis of the nature and regularities of variations of the EFG depending on the energy band structure of several diborides MB2 (M=Be, Mg, Al, Sc, Ti, V, Cr) has been reported.134 The results are listed in Table 3. Among sp-metal borides, BeB2 is characterised by the largest Vzz component of the EFG. For 3d-metal diborides, the Vzz component changes non- monotonically, with a minimum for TiB2. These changes were explained 134 based on results of band structure calculations for these phases (see Fig.7). For boron nuclei, the EFG characterises the anisotropy of partial charge density distributions Vzz&pz ¡ Öpx á pyÜ , 2 where pz, px and py are the corresponding band populations. Table 3. EFG components on boron nuclei in borides. Diboride 10721 Vzz /V m72 experiment 147 FLMTO134 FLAPW148 (see a) 0.37 0.43 0.63 7 1.69 7 7 7 1.08 7 7 0.38 0.39 0.60 1.88 2.10 0.99 0.60 0.35 0.38 0.59 MgB2 BeB2 AlB2 ScB2 TiB2 VB2 CrB2 a Full-potential linearised augmented plane wave method. According to calculations, diborides with partially unfilled pxy-bands (BeB2, MgB2) are characterised by the largest EFG. Additional increase in EFG for BeB2 depends on the mutual position of the pxy- and pz-bands and is determined by an increase in the population of the latter. The increase in Vzz in order TiB2?VB2?CrB2 is due to fundamentally different reasons, namely, weakening of the interlayer hybrid p ± d metal7boronSuperconducting MgB2 and related compounds: synthesis, properties and electronic structure bonds affecting the relative position and populations of the B pxy- and pz-bands.124 Estimation of changes in the EFG on magnesium and boron nuclei in MgB2 with pressure (up to P=10 GPa) showed that an increase in pressure has little effect on the EFG at boron nuclei, whereas the EFG at magnesium nuclei rapidly increases.134 c.Phonon spectrum and electron ± phonon interaction Preliminary estimation of phonon frequencies obtained from band structure calculations showed 85, 87, 149 that they were con- sistent with the observed Tc.The authors of pioneering stud- ies 85, 87 reported only particular modes responsible for electron ± phonon coupling. The results of phonon spectrum calculations 149 ± 151 point to the dominating role of optical pho- nons in the electron ± phonon interaction. Similarly to electron states of the s-bands, these optical phonons are of quasi-2D type (along the G ±Adirection of the Brillouin zone) and correspond to the in-plane vibrational modes. An important distinction between the phonon characteristics of MgB2 and AlB2 (Fig. 11) consists in `softening' of the high-frequency vibrational modes in magnesium diboride and their strong coupling (l&0.75) with electrons near the Fermi level.150 Nonsuperconducting AlB2 is characterised by a much smaller constant l (0.4).150 Inelastic neutron scattering studies 101 revealed four peaks of the generalised density of phonon states, the maximum energy of atomic oscillations was found to be higher than 90 meV. The maximum energy of acoustic phonons (36 meV) was determined from analogous experi- ments 98 and it was shown that optical modes are characterised by strong dispersion and form peaks at 54, 78, 89 and 97 meV.The estimate 98 of the constant l (l&0.9) obtained using the Born ± von Karman model 98 is consistent with the results of calculations by Andersen et al.149 a2F(o) 1 2.0 1.5 1.0 2 0.50 100 80 60 20 40 o /meV Figure 11.Eliashberg functions [a2F(o)] for MgB2 (1) and AlB2 (2).150 3. Search for new superconductors based on MgB2 and related phases: theory and experiment The discovery of superconductivity in MgB2 stimulated intensive theoretical and experimental search for new related superconduc- tors. A general trend has been to extend the variety of possible MgB2-based superconductors by varying their chemical composi- tions. a. Theoretical studies The possibility of chemical modification of the electronic structure of `ideal' MgB2 by varying its chemical composition was system- atically studied by the FLMTO method in the framework of supercell approach.124, 152 The effects analysed were those pro- duced on the band structure of MgB2 by doping the boron sublattice with Be, C, N and O, by doping the magnesium sublattice with Be, Ca, Li, Na, Zn and Cu and by changing the 725 Table 4.Results of FLMTO calculations of the DOS at the Fermi level [state per (eV per formula unit)] for magnesium diboride and related binary and ternary systems.124 Diboride Diboride N(EF) N(EF) 0.73 0.92 0.47 0.56 0.61 0.74 MgB2 CaB2 BeB2 MgB1.75C0.25 MgB1.75 Mg0.75B2 0.73 0.75 0.73 0.76 0.89 0.62 0.51 Mg0.75Li0.75B2 Mg0.50Li0.50B2 Mg0.75Na0.75B2 Mg0.50Na0.50B2 Mg0.75Cu0.25B2 Mg0.75Be0.25B2 Mg0.75Zn0.25B2 diboride composition due to addition of lattice vacancies in Mg- and B-sublattices (nonstoichiometry effects). The compositions of compounds studied are listed in Table 4.Qualitative changes in the energy band structures of solid solutions have been discussed in the framework of the RBM model.152 It was found that dopants can act as hetero- or isoelectronic impurities which change or have no effect on the VEC in the system, respectively. According to calculations, Be impurity in the boron sublattice, Li, Na and Cu impurities in the magnesium sublattice as well as both types of lattice vacancies (`hole dopants') will, first, decrease the VEC in the system, thus leading to an increase in the concentration of hole states in the s-band and, second, assist a shift of the EF towards high energies with an increase in N(EF). Both these factors are favourable for superconductivity (see above).On the contrary, Al impurity in the magnesium sublattice, C, N and O impurity in the boron sub- lattice (`electron dopants'), will deteriorate the superconducting properties of the system due to filling of the s-band and reduction of N(EF). In the framework of the RBM model, the role of isoelectronic replacement of Mg by Be or Ca remains unclear as yet. One can only assume that the main effects will be associated with structural factors, that is, magnesium diboride lattice dis- tortion depending on the ratio of radii of `host' and `guest' elements. Numerical calculations 124, 152 give a more complex picture (see Table 4 and Figs 12, 13). The addition of vacancies to boron sublattice (MgB1.75) leads to a decrease in N(EF).In the series of solid solutions MgB1.75C0.25<MgB1.75N0.25<MgB1.75O0.25 the N(EF) increases nonmonotonically. This is due to the filling of antibonding states. The s-band becomes fully occupied. It was concluded that there is no prospect for obtaining superconductiv- ity in the system by doping the boron sublattice.124 Based on consideration of particular binding energies (see Table 1), it was suggested that insertion of impurity into the boron sublattice is much less probable than substitution in the cation sublattice. In the latter case, the energy band structures of solid solutions Mg17xMxB2 (M=Be, Ca, Li, Na, Cu, Zn) remain essentially b a N(E) /Ry71 N(E ) /Ry71 EF EF 40 200 0 d c EF EF 20 40 E /Ry 0 E /Ry 0 0 71.0 70.5 071.0 70.5 Figure 12.Total densities of states for Mg0.75Cu0.25B2 (a), Mg0.72Na0.75B2 (b), CaB2 (c) and MgB2 (d).124726 Es(G7EF) /eV NaB2 1.0 { { Mg0.75B2 0.5 Mg0.75Cu0.25B2 Mg0.5Li0.5B2 Mg0.5Na0.5B2 Mg0.75Li0.25B2 Mg0.75Na0.25B2 NaB2Mg0.75Be0.25B2 Mg0.75Zn0.25B2 { MgB2 0 BeB2 MgB1.75C 71.0 YB2 ScB2 71.5 AlB2 0.5 0 70.5 Dn /e 72.0 71.0 Figure 13. Dependence of the s-band position at the G point of the Brillouin zone relative to the Fermi level in crystals [Es(G7EF)] on the electron concentration (relative to MgB2, Dn) for binary diborides and ternary solid solutions based on MgB2.124 similar to that of superconducting MgB2 (see Fig. 12 and Table 4). The N(EF) value depends first of all on the impurity concentration.Hence, the main procedure for modifying the properties of magnesium diboride is to dope its cation sublattice. In addition to the VEC, electron distributions and superconduc- tivity were also found to be dependent on the impurity concen- tration that affects structure peculiarities of MgB2-based multicomponent solid solutions (first of all, the ratio between the in-plane distance and interlayer spacing). Calculations of solid solution Mg0.5Al0.5B 125 revealed the presence of hole states near the Z point of the Brillouin zone. It was also found that cation substitution and corresponding elec- tron distributions cannot be completely described using the rigid band model. Changes in the energy band structure with concentration and, in particular, the dependence of the s-band energies on the composition of solid solutions Mg17xAlxB2 and Mg17xNaxB2 (x=0, 1/3, 2/3, 1.0) were studied126 using the density functional theory in the framework of supercell approach ( (Fig.14). For Mg17xNaxB2 with any x, the top of the B pxy- band at the G point is above EF; in other words, the major electronic criterion for the existence of superconductivity holds. For Mg17xAlxB2, this condition is valid at x<0.6. At x>0.6, the hole states at the centre of the Brillouin zone disappear. The concentrations of the s-holes (nh /cm73) depend on the composi- tions of solid solutions as follows:126 for Mg17xAlxB2, nh=(0.8+0.8x)61023, a Ecoh Es /eV /eV atom71 1 EF 0 71 6.0 5.5 5.0 4.5 72 AlB2 MgB2 NaB2 NaB2 Figure 14.Changes in the s-band position at the G point of the Brillouin zone (E s) relative to the Fermi level (a) and cohesive energies (b) in the order NaB2?Mg17xNaxB2?MgB2?Mg17xNaxB2?AlB2.126 pÅÅÅ36 3 pÅÅÅ) point of the system. System Mg± B±Cu.153, 154 Information about this system is contradictory. Abnormally high Tc=49 K announced 153 for a sample with stoichiometric composition Mg0.8Cu0.2B2 has not been confirmed; attempts at preparing copper-containing single- phase solid solutions failed.154 b AlB2 MgB2 A L Ivanovskii for Mg17xNaxB2, nh=(0.871.4x)61023. The hole concentration is a maximum (*1.661022 cm73) for NaB2. The cohesive energy (see Fig. 14) monotonically increases in the order NaB2<Mg17xNaxB2<MgB2<Mg17xAlx- B2<AlB2 as the corresponding bonding states in the valence band are filled.b. Experimental studies Theoretical studies of systems mentioned above are accompanied by experimental search 83, 136, 153 ± 159 for new superconductors based onMgB2 (variation of its elemental composition was mostly considered). Substitution solid solutions were synthesised using cation (Mg17xMxB2; M=Li, Be, Al, Zn, Cu) 83, 136, 153 ± 155 and anion (MgB27yCy) 156 ± 158 sublattice substitution. Earlier,159 only the formation of solid solutions in the MgB2 ± AlB2 system was reported. Solid solutions were synthesised from elements in inert atmos- phere, the samples obtained were analysed by elemental and structural analyses and their superconductivity parameters (Tc, DTc) were measured (Table 5). System Mg± B± Be.155 A mixture of MgB2+BeB2 phases was obtained.Beryllium diboride has a hexagonal structure (the space group P6/mmm, c/a=0.977) and is paramagnetic at T<5 K. It was found that beryllium does not replace the Mg atoms. System Mg± B ± Li.155 The homogeneity region of solid sol- utions Mg17xLixB2 was determined (x<0.3). As lithium content increases, the interlayer distance remains unchanged while the in- plane B7B distance (parameter a) is shortened by *0.014 A (at x=0.3). A decrease in Tc is observed beginning with x=0.1 and becomes pronounced near the single-phase composition bound of the solid solution (x=0.3).Loss of superconductivity occurs for the sample with x=0.5. System Mg± B±Al.83 Two homogeneity regions of solid solutions Mg17xAlxB2 were found (see Table 5). As the Al/Mg ratio increases, anisotropic lattice compression occurs: the inter- layer spacings decrease appreciably while the B7B distances do not exhibit considerable changes. This leads to structural insta- bility of the crystal and phase transition at x=0.1. A decrease in Tc in the homogeneity region of the solid solutions was explained 83 by an increase in the electron concentration and a decrease in N(EF). Analysis of the phase diagram shows that the superconducting phase is very close to the structural instability System Mg± B ±Zn.154 The homogeneity region for Mg17xZnxB2 samples is up to x=0.1.Analysis of the lattice parameters for a sample with x=0.1 revealed their increase compared to those of MgB2, namely, 0.17% for a and 0.2% for c. It was suggested 154 that nonmonotonic change in Tc is due to positive for the parameter a (in-plane B7B distances) and negative for the parameter c (interlayer spacings) expansion effects. Mg± B±C.156 ± 158 borocarbides AlB2-Type System MgB27xCx were synthesised from elements. Transition temper- ature Tc decreases in the homogeneity region (x40.20) by*2 K (at x=0.20, see Ref. 157) while the width of the transition, DTc, substantially increases (by*3 ± 7 K as compared to `pure' MgB2, see Refs 156, 157). Introduction of carbon impurity leads to compression of graphite-like boron sheets: a(MgB2)7a(MgB1.8C0.2)=0.013 A.The interlayer distances (the lattice parameter c) remain unchanged.156 Magnesium boro- carbides with much narrower homogeneity region (x<0.1) were also synthesised;158 transition temperature of these compounds decreases as the C/B ratio increases.Superconducting MgB2 and related compounds: synthesis, properties and electronic structure Table 5. Conditions for synthesis and characteristics of particular ternary systems based on MgB2. Products (phase composition) Reaction conditions Starting materials a Mg+B+Al 600 ± 900 8C 100% Ar, 600 ± 900 8C 100% Ar Mg+B+Zn Mg+B+Li 95% Ar+5% H2, Mg17xAlxB2: solid solutions 600 ± 900 8C at0<x<0.1 and 0.25<x<0.40 mixture of phases at 0.1<x<0.25 Mg+B+Cu 95% Ar+5% H2, Mg0.8Cu0.2B2 (?) [mixture of phases MgB2+Cu2Mg+MgO (traces)] mixture of phases MgB2+Cu2Mg+ +(MgO+MgB4) (traces) solid solutions 600 ± 900 8C Mg17xZnxB2+MgO (traces)+MgB4 at x&0.1, mixture of phases at x>0.1 Mg17xLixB2: solid solutions at x<0.3 mixture of phases at x>0.3 mixture of phases MgB1.8C0.2+MgO (traces) Mg+B+Be Mg+B+C the same 100% Ar, 950 8C Ar 100% Ar, 850 8C, 20 atm 950K a Amorphous boron powder was used in most cases.The synthesis of `superstoichiometric' compositions Mg1+xB2 with 10 at.%, 20 at.% and 30 at.% of excess Mg added to the reaction mixture has also been reported.157 It was found that `superstoichiometry' has no effect on the superconductivity in magnesium diboride.4. Materials science and some aspects of practical utilisation of MgB2 a. Pressure dependence Studies on the effect of pressure (as well as joint effect of pressure and temperature) on the properties of MgB2 attract considerable attention both in solving fundamental problems of the nature of superconductivity in anisotropic media and in developing new procedures for improving the properties of superconducting MgB2.120, 160 ± 166 The structure of MgB2 polycrystals as function of hydrostatic pressure up to 0.62 GPa was studied by neutron diffraction in the temperature range from 11 to 297 K.163 It was found that the lattice parameters a and c change linearly with pressure, viz., a=a070.00187P, c=c070.00307P, where a0 and c0 are the lattice parameters under normal conditions and P is the pressure (in GPa).Essentially nonuniform lattice compression was emphasised: compression along the c axis is 64% larger than along the a axis (in the B7B plane). Similar results were obtained in synchrotron X-ray diffraction studies of isothermal compres- b a V /A3 a/a0 , c/c0 29.0 1 1.000 0.995 28.6 2 28.2 0.990 0.985 4 P /GPa 2 0 2 4 P /GPa 0 c Tc /K 37 36 35 Figure 15. Pressure dependences of the lattice parameters and crit- ical transition temperature for MgB2 polycrystals.161 The unit cell volume (a); the ratio of the unit cell parameters a/a0 (1) and c/c0 (2) (b) and critical super- conducting transition tempera- ture (c).0.5 1.0 1.5 P /GPa 0 727 Ref. Lattice parameters /A Tc /K a c 83 decreases as x increases 38 (x=0) 36 (x&0.1) 0 (x>0.1) 49 (?) 153 3.505 3.068 (?) 7 7 154 3.5250 (x=0.1) 154 3.0841 (x40.1) 3.524 (x=0.3) 155 3.070 (x=0.3) changes nonmono- tonically decreases as x increases 7 73.070 7 136 156 3.520 157 3.523 3.070 Tc1&41 Tc2&37 *34 sion (Fig. 15). Thermal expansion of the crystal lattice is also anisotropic and is mainly due to changes in the interlayer distances (the parameter c).163 It was found that the effect is due to anisotropy of chemical bonding in MgB2.120 ± 163 Superconductivity parameters of MgB2 are pressure depend- ent. For instance, it was found that the superconducting transition temperature decreases with a rate of 72.0 162 and 71.6 K (GPa)71,164 respectively (see Fig.15). This is consistent with the results of calculations 165 [71.4 K (GPa)71] and can be explained based on the BCS theory of the mechanism of super- conductivity in magnesium diboride.120, 165 b. Neutron irradiation Experiments 167 on thermal (F=161019 cm72) and fast (F=161018 cm72) neutron irradiation of MgB2 (the total dose was greater than 10 doses per atom) were carried out with a IVV-2M nuclear reactor at T=350 K. The transition temper- ature of irradiated samples decreased from 40 to 5 K, the residual resistance r0 increased (from 0.35 to 2.0 mO cm) and the upper critical field exhibited minor changes (7dHc2/dT&0.5 T K71).By analogy with A15-type superconductors (Nb3Sn, V3Si), the results obtained were explained 167 by a decrease in N(EF). It was concluded that high radiation resistance of MgB2 under condi- tions of neutron irradiation (other superconductors undergo amorphisation under such conditions) is of prime importance for practical utilisation of this compound. Irradiation of MgB2 samples leads to anisotropic lattice expansion (mostly, to an increase in the parameter c from 3.5216 to 3.5533 A), which is accompanied by an increase in the unit cell volume by*1.4%.157 The occupation number of Mg sites decreases from 0.94 to 0.89, while the occupation number of B sites remains unchanged. No antisite defects were found. c. Grain boundary effects Heat treatment of MgB2 under high pressure to improve super- conducting properties of highly porous (under the synthesis conditions) magnesium diboride polycrystals has been pro- posed.160 At T=950 8C and P=3 GPa, the surface morphology of MgB2 samples changed, their porosity decreased and the grain boundary contact area increased. High-resolution transmission electron microscopy studies of the morphology of the MgB2 samples obtained by high-pressure sintering (P=3 GPa) revealed the absence of porosity and strong connections between grains.The impurity phase (magnesium tetraboride) occurred in the form of small inclusions.113 Specific features of some properties of728 MgB2 polycrystals, determined by the grain structure of the surface, have been discussed.168, 169 d. Surface and films The development of methods for reproducible synthesis of epitax- ial superconducting MgB2 thin films for various electronic devices is one of the key problems of practical utilisation of this material.Here, the use of many conventional film growth procedures is complicated by the large differences in vapour pressure, adsorp- tion coefficients and mobilities between boron and magnesium as well as due to a tendency of magnesium to undergo oxidation. Currently, such films are synthesised using two, ex situ and in situ, approaches.170 ± 178 The former 174, 175 implies preliminary preparation of boron films which are then saturated with magne- sium while the latter includes fabrication of MgB2 films during the same run.Currently, films with the highest current density and Tc (37 ± 39 K) have been obtained by ex situ processes.174, 175 The procedure included pulsed laser deposition (PLD) of thin boron films onto a substrate (SrTiO3) followed by their reaction with magnesium vapours at 900 8C. The film surfaces were character- ised 175 by well-developed grain structure. This approach has a drawback, that is, the high-temperature stage (here, temperatures are higher than the decomposition temperature of magnesium diboride). Therefore, this method is inapplicable to the growth of multilayers and heterojunctions. In addition, in the case of ex situ synthesis the film growth is driven by diffusion of magnesium through an already formed magnesium diboride surface layer, which makes it difficult to obtain an epitaxial relationship between the film and the substrate.174, 175 The in situ film growth involves two stages.170 ± 173 Initially, magnesium diboride films are prepared by PLD of various targets on different substrates, e.g., Al2O3, LaAlO3, SrTiO3.173 The targets (Fig. 16) were dispersed (<45 mm) powders of MgB2 synthesised from elemental boron and magnesium, commercial MgB2 powders (AA), etc.Typically, the films were 100 ± 500 nm thick. Superconductivity was observed only for films annealed in the rather narrow temperature interval (550 ± 600 8C). Atomic force microscopy images revealed a reasonably smooth surface of these films, which is in stark contrast to the surface of the ex situ prepared films with well-defined grains.Films were deposited on the (100) and (111) surfaces of SrTiO3 by (i) PLD of MgB2, (ii) PLD of multilayers of MgB2 and Mg followed by in situ annealing at high temperature and (iii) PLD of multilayers of Mg and B followed by in situ annealing at high temperature.174 The highest superconducting transition temperature (Tc=22 K) was observed for the films grown from magnesium diboride target. An increase in the annealing temperature deteriorates the super- conductivity characteristics (for instance, Tc was no higher than 8 K after annealing at temperatures >800 8C). This is thought to be due to layer-by-layer deviation from perfect nonstoichiome- a MgB2 Mg MgB2 Mg b Mg MgB2 MgB2 (+Mg) Substrate Figure 16.Schematical drawings of the targets (a) and chemical compo- sitions of in situ prepared films (b).173 A L Ivanovskii try.174 It was emphasised that the necessary condition for growth of superconducting films consists in retaining the necessary Mg content. Films grown in situ from room-temperature deposited Mg + MgB2 powder targets and post-annealed at 600 8C were also characterised by relatively low Tc (12 ± 22 K). It is believed that one of the key factors consists in the choice of reaction conditions to prevent oxidation of magnesium in plasma and in the film deposited. In this case, of great importance is the choice of the pressure of the inert gas in the vacuum chamber. Yet another reason for the low Tc is associated with the possibility 172 of formation of nonsuperconducting higher magnesium borides (MgB4, MgB7) and elemental boron, which deteriorates the super- conductivity parameters of the film. The formation of magnesium borides is associated with the nature of plasma generated by the laser.In different ionisation zones, the temperature of plasma can be as high as several thousands of degrees. In this connection, of great importance are the results of thermodynamic modelling 179 of the system Mg7B using the CALPHAD technique.180 Calcu- lations 179 of composition ± temperature, composition ± pressure and pressure ± temperature phase diagrams were carried out using experimental values of thermodynamic parameters of the compo- nents (MgB2, MgB4, MgB7 as well as the gas, liquid and solid magnesium phases and b-boron).Magnesium diboride can be formed only at vapour pressures of magnesium falling into specific range and in particular temperature range (Fig. 17). Clearly, operation under these conditions will favour optimisation of the composition and, as a consequence, superconductivity character- istics of in situ grown films. A `mixed' film fabrication procedure has been proposed.177 Films deposited in situ from MgB2 targets in high vacuum exhibited the properties of pure metals or semiconductors; super- a P /TorrLiquid+MgB2 102 Gas+MgB2 10 Gas+MgB4 1072 Gas+MgB7 1074 Gas+B (sol.) c(B) (at.%) 0.8 0.6 0.4 0.2 0 b MgB2 (sol.) 104 Liquid+MgB2 10 Gas+MgB4 Gas+MgB2 1074 Gas+B(sol.) Gas+MgB7 P /Torr 15 10 104 (1/T) /K71 5 Figure 17.Calculated diagrams of the Mg± B system. The pressure ± composition diagram at T=850 8C (a) and P± T-diagram for the ratio Mg: B>1 : 2 (b). The `gas+MgB2' region corresponds to thermodynamic conditions for the preparation of MgB2 thin films. Obtained from CALPHAD calculations.179Superconducting MgB2 and related compounds: synthesis, properties and electronic structure conductivity (Tc&25 K) was observed only after additional exposure to magnesium vapours. Films with Tc=24 K were obtained from stoichiometric MgB2 targets by PLD on silicon substrates followed by heating in Ar atmosphere.178 Uniform surface morphology and thickness of films were pointed out.Oxygen contamination of the outermost layers was assumed to be the main reason for the low Tc. Vasquez et al.181 reported a special XPS study of the elemental composition of the surface of MgB2 polycrystals. The XPS spectrum of the surface of initial sample exhibited strong impurity signals from C and O, two signals from Mg2p-state with binding energies (Eb) of 49.5 and 50.8 eV and signals from B 1s-states. Among the last-named group of signals, a weak peak with Eb=193 eV is due to the presence of boric anhydride. The main signal of the B1s-state was a doublet with the energies Eb=186.7 and 187.9 eV comparable with the corresponding energies for isostructural d-metal diborides (187.2 ± 188.5 eV, see Ref.182). According to Vasquez et al.181 there are three main concen- tration zones near the MgB2 surface. The first zone, i.e., the surface of initial sample, is covered with oxide film and is substantially enriched with magnesium (the Mg:B ratio is 1 : 1.25). The oxide layer can be efficiently removed in the course of chemical purification (by, e.g., 1% and 0.5% solutions of HCl and Br2 in ethanol). The second, `subsurface', zone under the oxide film is, on the contrary, depleted of magnesium (Mg :B= 1 : 3). In this zone, nonstoichiometry with respect to magnesium leads to a decrease in the electron density in boron sheets. The corresponding signal of B1s-state is observed at higher binding energy (Eb=187.9 eV). The second B1s-peak (Eb=186.7 eV) corresponds to the states of boron atoms in the third zone (magnesium diboride bulk with stoichiometric ratio Mg: B=1 : 2).The energies of Mg2p-level and Auger-signal are lower than that of metallic magnesium due to cationic state of Mg2+ in magnesium diboride. e. New MgB2-based superconducting materials: wires, ribbons, composites A simple procedure for fabricating high density MgB2 wires has been developed.183 As mentioned above, MgB2 polycrystals are obtained by solid-phase synthesis (a mixture of boron and magnesium powders is heat treated); mechanistically, magnesium diboride is formed via diffusion of Mg vapour into the boron grains. Bud'ko et al.183 proposed to use this technique with boron fibres instead of powder.Boron fibres (d&100 mm, Textron Systems, USA) and metallic magnesium taken in a stoichiometric ratio were placed in a Ta tube and heated at 950 8C for 1 h. Compared to the unreacted boron fibres, the MgB2 wires thus obtained were brittle and somewhat deformed (see Fig. 18), but they did not break down. A tungsten core (d&15 mm) of the starting boron fibres did not appear to be affected during the reaction, it also has no effect 183 on the superconducting character- istics of the MgB2 wires. The diameter of the MgB2 wires increased from 100 to 160 mm (synthesis of magnesium diboride powder is also accompanied by an increase in the particle size compared to the initial boron grains 82). The density of wire segments was determined to be *2.4 g cm73 [*80% of the theoretical density of MgB2 crystal, 2.55 g cm73 (see Ref.82)]. The superconductiv- ity parameters of the wires were Tc=39.4 and DTc=0.9 K. At room temperature, the wires exhibited the properties of highly conducting material (the resistivity, r, had a value of 9.6 mO cm), whereas near the critical transition temperature (T=40 K) r decreased to 0.38 mO cm. Temperature dependence of the resis- tivity (r=r0+r1Ta, where a&2.6) is close to that found for MgB2 polycrystals [a&2.8, measured for Mg(10B)2 pellets].82 To choose optimum conditions for the preparation of super- conducting MgB2 wires, the mechanism of Mg diffusion into boron was systematically studied using commercialy sold boron fibres (d=100 and 141 mm).184 Boron fibres were exposed to magnesium vapours at 950 8C, the course of the reaction was monitored by the scanning electron microscopy (SEM) using the Figure 18.Electron microscope images of cross-section of a MgB2 wire of diameter 160 mm (a) and unreacted boron filament with a tungsten core (b).183 SEM images of the cross-sections of the MgB2 wires taken after time intervals. It was found that magnesium diboride begins to form after 15 min and is 50% complete after 1 h. The wire segments thus produced were of little porosity. In Fig. 19, the resistivities of MgB2 pellets and wires are shown as functions of temperature. Different behaviour of the curves is thought to be due to larger density of wires.184 Studies 185 on magnetotransport characteristics of dense wires revealed a very narrow resistive transition, a linear temperature dependence of Hc2 in the range 7 K<T<32 K and a positive curvature of this dependence near Tc. This is similar to the behaviour of layered IBC.186 ± 188 Another technique for fabrication of superconducting MgB2 wires involves placing powdered MgB2 inside a sheath (a niobium,189 gold, copper of nickel tube 190).For instance, a niobium tube of diameter 6 mm was filled with MgB2 powder and then a ribbon was prepared by cold drawing followed by rolling.189 The ribbon with a superconducting core (a cross- section of 0.232 mm2) had a cross-section of 2.5660.32 mm2.189 At 4.2 K, the measured current density in the ribbons heated at 900 8C for 1, 2 and 3 h in an Ar atmosphere was 4.7, 7.5 and 1.16104 A cm72, respectively.Experiments 190 with ribbons r /mO cm 1.0 0.50 36 Figure 19. Resistivity of a dense MgB2 wire (1) and polycrystalline MgB2 pellet (2) as function of temperature.184 729 a b 1 2 T /K 40 38730 sheathed with different metals (Au, Ni, Cu) revealed the highest current density (*105 A cm72) at 4.2 K for the ribbon with Ni sheath. Magnesium diboride is highly brittle, which is typical of ceramics. Therefore, the problem of improving mechanical prop- erties of MgB2, i.e., making the material more malleable without significant deterioration of its superconducting properties is of great importance. The most obvious way is to fabricate compo- sites with ductile metals (or alloys) which should have low melting temperatures comparable with that of MgB2 (1073 K) and be conductors and highly plastic substances.The formation of such composites should also not be accompanied by reactions between the ductile additive and magnesium diboride resulting in solid solutions. Recently, aluminium (Tm=933 K) was found to be appropriate for this purpose.191 Pellets of MgB2 ±Al composite (11 vol.% Al) were prepared from MgB2 (AA) and Al powders (the particle size of both components was 1 to 3 mm) by pressing at 500 ± 700 K, i.e., below the formation temperature of nonsuper- conducting Mg17xAlxB2 solid solutions.83 The samples were studied by scanning tunnelling microscopy and the transport and magnetic characteristics of the composites were measured.Partic- ular attention was given to the proximity effect (the formation of a typical `mini-gap' structure was observed, which is due to contact between magnesium diboride and aluminium, T Al c =1.2 K) and to the effect of structural disordering on the superconducting transition temperature of the MgB2 ±Al composite. The Tc value for the composite (38 K) was found to be comparable with that of pure MgB2 (39 K). IV. Conclusion Information collected and systematised in this review describes the infancy of research into the properties of recently found new superconductor,MgB2. In spite of short time that elapsed after the discovery of superconductivity in MgB2, experimentalists and theoreticians have accumulated a great body of information about its physicochemical properties.The results obtained are of crucial importance for an understanding of the nature of super- conductivity in anisotropic media. They gave a strong impetus to search for new superconductors among metal borides and related compounds and opened attractive prospects for their technolog- ical utilisation. The state-of-the-art in this field can be outlined as follows. To date, a number of methods for the synthesis of MgB2 have been developed. Peculiarities of the electronic structure and chemical bonding in MgB2 as well as its electrophysical, magnetic, thermo- dynamic properties and spectroscopic characteristics have been studied. Based on the results of theoretical modelling and syn- thetic experiments, some new MgB2-based solid solutions have been found and ways to further search for new superconductors have been directed.Materials science studies allowed the obtain- ing of MgB2 powders, films, wires and composites. On the other hand, magnesium diboride, as a medium-Tc superconductor, remains a unique compound in its kind except for a few solid solutions based on this substance. All attempts at increasing the critical transition temperature of MgB2 or finding new superconductors with close Tc have failed so far. Many problems of the synthesis and detailed investigations of multi- component systems with participation of magnesium diboride are still to be solved. These are, e.g., studies of the phase diagrams, determination of homogeneity regions of solid solutions and criteria for optimum doping of MgB2 as well as establishment of reasons for changes in Tc.Further search for new candidates for superconductors with graphite-like boron sheets in their structure is undoubtedly required. We believe that these problems will be the focus of research interest in the near future. * * * A L Ivanovskii Studies on superconducting MgB2 and related compounds have been continued in the course of preparation of the manuscript. The most interesting recent results are as follows. Single crystals of MgB2 have been first obtained.192 ± 195 A magnesium-enriched precursor sealed in a niobium ampoule was placed in a quartz tube,192 heated to 1050 8C (1 h) and then very slowly cooled down to 700 8C (the cooling duration was 120 to 360 h).The MgB2 single crystals were found to have various shapes (spherical, rod-like, or hexagonal-plate) depending on the cooling regime. The samples studied were characterised by excel- lent superconductivity parameters (Tc=38 K and DTc=0.3 K). Experiments with MgB2 single crystals revealed strong anisotropy of superconductivity.192 ± 195 An original method for the synthesis of high density, single- phase MgB2 ceramics with Tc=38 K by crystallisation of mag- nesium diboride melt under pressure has been proposed.196 The slope and the onset of transition to the superconducting state were found to be dependent on the grain size of the material.Epitaxially grown, oriented MgB2 films were prepared and their properties have been studied:197 ± 210 superconductiv- ity,197, 199 ± 210 the Hall effect 207 and surface resistance.206 Aniso- tropy of some characteristics of the oriented films (in particular, conductivity and upper critical field 198, 200) was revealed. Degra- dation of superconductivity in MgB2 films in the presence of water has been studied.205 Preparation of nanostructured superconduct- ing MgB2 films has also been reported.203 Pioneering calculations of the electronic structure of the boron- and magnesium-termi- nated MgB2 (0001) surfaces have been carried out.211, 212 Summing up, mention may be made that there are three main lines of inverstigations in the field of search for MgB2-related superconductors. The first research avenue is associated with extension of the class of possible MgB2-based superconductors by varying their chemical compositions to obtain different types of solid solutions or by forming superstructures (see Section III.3).The second investigation line includes studies of a much broader spectrum of substances. Here, the search for possible super- conductors has been performed among binary or multicomponent compounds which are to some extent structurally or chemically similar to MgB2. The discovery of critical transitions in ZrB2 (5.5 K, see Ref. 213), TaB2 (9.5 K, see Ref. 137), Re3B (4.7 K, see Ref. 214) and in the new phase of beryllium boride with the chemical composition BeB2.75 and the unit cell containing 110.5 atoms (0.72 K, see Ref.215) represent recent advances in this field. The third line of research activity was initiated by the discov- ery of superconductivity at Tc&8 K in a perovskite-like inter- metallide MgCNi3.216 From the standpoint of `renewed chemical paradigm' it was proposed 216 that the whole class of intermetal- lides should be considered as potential superconductors. The results obtained by Cava et al.216 are of particular interest for the following reasons. Highly symmetric structure of MgCNi3 (the space group Pm3m) favours superconductivity. However, superconductivity in all the known isostructural compounds � oxide perovskites [e.g., (Ba, K)BiO3, see Ref. 217] � is due to electron ± hole states of the oxygen atoms occupying 3c-type positions (0, 1/2, 1/2).Since in the MgCNi3 structure these positions are occupied by Ni atoms, the mechanism of superconductivity should be of funda- mentally different nature. Most of the known `non-oxide,' perovskite-like compounds MXM03 (M0=Zn, Al, Ga, In, Sn; M=Mn, Fe; X=C, N) are ferromagnetic (and antiferromagnetic) or are characterised by more complex (mixed) types of spin ordering.218 Hence, among structurally similar compounds, MgCNi3 can be considered as an `intermediate' between the classes of perovskite-like superconduc- tors (oxides) and magnetics (non-oxide perovskites). Among superconductors, nickel-containing IBC LuNi2B2C (Tc&16 K) and YNi2B2C (Tc&15.6 K) are the closest `chemical analogues' of MgCNi3.However, unlike the last-named com- pound, these IBC belong to magnetic superconductors and exhibitSuperconducting MgB2 and related compounds: synthesis, properties and electronic structure a distinct quasi-2D structure.22 In addition, the Ni content in IBC is much less than in MgCNi3. Taking into account clearly defined magnetic properties of nickel, the fact that superconductivity occurs in the nickel-rich compound MgCNi3 is quite surprising. Based on the results of pioneering studies on the properties of MgCNi3 (the critical field, Hc2, the Hall coefficient, etc.,219 ± 221) and critical transition temperatures of MgCNi37xMx (M=Mn, Co, Cu) samples doped at Ni sites, the parent compound, MgCNi3, was placed into the class of `conventional' supercon- ductors with electron ± phonon interactions.222, 223 Band structure calculations for MgCNi3 (see Refs 224 ± 227) revealed a distinc- tive feature, viz., the presence of an intense DOS peak near the Fermi level.It was found that the major contribution to N(EF) comes from the Ni 3d-states (*88.2%). Band structure calcula- tions for new MgCNi3-like superconductors have been carried out. The possibility of superconductivity in a nonstoichiometric (with respect to C sublattice) perovskite, in perovskite-like ScBNi3 and InBNi3 as well as in MgCCo3 and MgCCu3 (all of them are isoelectronic and isostructural to MgCNi3) has been discussed.228 The discovery of superconductivity in intermetallic com- pounds MgB2 and MgCNi3 with simple and highly symmetric structures has given a creative urge to search for application areas of these promising superconductors.The author expresses his gratitude to N I Medvedeva, V G Zubkov, V N Diev and V G Bamburov for valuable discus- sions. References 1. Perspectivnye Tekhnol. 8 (1 ± 2) 1 (2001) 2. J Nagamatsu, N Nakagawa, T Muranaka, Y Zenitani, J Akimitsu Nature (London) 410 63 (2001) 3. G V Samsonov, L Ya Markovskii, A F Zhigach, M G Valyashko Bor, ego Soedineniya i Splavy (Boron, Its Compounds and Alloys) (Kiev: Academy of Sciences of Ukr. SSR, 1960) 4. B Post, F W Glaser Refractory Binary Borides (New York: Interscience, 1963) 5. V Aronsson, T Lundstrom, S Rundquvist Refractory Borides, Silicides and Phosphides (London: Methuen, 1965) 6.E L Muettertis (Ed.) The Chemistry of Boron and its Compounds (New York: Wiley, 1967) 7. H Goldschmidt Interstitial Alloys (London: Butterworths, 1967) 8. G V Samsonov, T I Serebryakova, V A Neronov Boridy (Borides) (Moscow: Atomizdat, 1975) 9. G V Samsonov, I M Vinitskii Tugoplavkie Soedineniya. Spravochnik (High-Melting Compounds. Handbook) (Moscow: Metallurgiya, 1976) 10. Yu B Kuz'ma Kristallokhimiya Boridov (Crystal Chemistry of Borides) (L'vov: Vishcha Shkola, 1983) 11. T Ya Kosolapov (Ed.) Boridy i Materialy na Ikh Osnove (Borides and Materials Based on Them) (Kiev: Institute of Problem of Materials Technology, Academy of Sciences, Ukr. SSR, 1986) 12. Yu B Kuz'ma, I F Chaban Dvoinye i Troinye Sistemy, Soderzhashchie Bor (Binary and Ternary Systems Containing Boron) (Moscow: Metallurgiya, 1990) 13.T I Serebryakova, V A Neronov, P D Peshev Vysokotemperaturnye Boridy (High-Temperature Borides) (Moscow: Metallurgiya, 1991) 14. A L Ivanovskii, G P Shveikin Kvantovaya Khimiya v Materialo- vedenii. Bor, ego Splavy i Soedineniya (Quantum Chemistry in Materials Technology. Boron, Its Alloys and Compounds) (Ekaterinburg: Izd. Ekaterinburg, 1997) 15. S V Vonsovsky, Y A Izumov, E Z Kurmaev Superconductivity of Transition Metals, Alloys and Compounds (Berlin: Springer, 1982) 16. L E Toth Transition Metal Carbides and Nitrides (New York; London: Academic Press, 1971) 17. L Leyarovska, E Leyarovski J. Less-Common Met.67 249 (1979) 18. M B Maple, O Fischer (Eds) Superconductivity in Ternary Com- pounds; I. Structural, Electronic and Lattice Properties; II. Super- conductivity and Magnetism (Berlin; Heidelberg; New York: Springer, 1982) 731 19. R J Cava, H Takagi, B Batlogg, H W Zandbergen, J J Krajewski, W F Peck, R van Doren, R J Felder, T Siegrist, T K Mizuhashi, J O Lee, H Eisaki, S A Carter, S Uchida Nature (London) 367 146 (1994) 20. R J Cava, H Takagi, H W Zandbergen, J J Krajewski,W E Peck, T Siegrist, B Batlogg, R van Doren, R J Felder, T K Mizuhashi, J O Lee, H Eisaki, S Uchida Nature (London) 367 252 (1994) 21. R Nakarajan, C Mazumdar, Z Nossian, S K Dhar, K V Gopalakrishnan, L C Gupta, C Goddart, B D Padalia, R Vijayaraghavan Phys. Rev.Lett. 72 274 (1994) 22. A L Ivanovskii Usp. Khim. 67 403 (1998) [Russ. Chem. Rev. 67 357 (1998)] 23. H Moissan Ann. Chim. Phys. 6 296 (1895) 24. A Stock Hydrides of Boron and Silicon (Ithaca, NY: Cornell University Press, 1933) 25. N Moissan Ann. Chim. (Paris) 8 87 (1906) 26. F Jones, R Taylor J. Chem. Soc. 1881 (1912) 27. R Ray J. Chem. Soc. 105 2162 (1914) 28. M Travers, N Gupta, R Ray Indian Inst. Sci. 1 1 (1914) 29. M Travers, R Ray Proc. R. Soc. London A 87 153 (1913) 30. T Serebryakova J. Less-Common Met. 67 499 (1979) 31. P Duhart Ann. Chim. (Paris) 7 339 (1962) 32. F A ShConstitution of Binary Alloys (Second Suppl) (New York: McGraw-Hill, 1971) 33. T Massalsky (Ed.) Binary Alloy Phase Diagrams (New York: A S M International, 1990) 34.M Jones, R Marsh J. Am. Chem. Soc. 76 1434 (1954) 35. V Russel, R Hirst, F Kanda, A J King Acta Crystallogr. 6 870 (1953) 36. L Ya Markovskii, Yu D Kondrashev, G V Kapustovskaya Zh. Obsh. Khim. 25 433 (1955) a 37. L Ya Markovskii, Yu D Kondrashev, G V Kapustovskaya Dokl. Akad. Nauk SSSR 100 1095 (1955) b 38. P Beck (Ed.) Electronic Structure and Alloy Chemistry of the Tran- sition Elements (New York; London: Wiley-Interscience, 1963) 39. K Schubert Kristallstrukturen zweikomponentiger Phasen (Berlin; Gottingen; Heidelberg: Springer, 1964) 40. W B Pearson The Crystal Chemistry and Physics of Metals and Alloys (New York; London; Sidney; Toronto: Wiley-Interscience, 1972) 41. A Guette, R Naslain, J Galy C.R. Acad. Sci., Ser.C 275 41 (1972) 42. R Naslain, A Guette, M Barret J. Solid State Chem. 8 68 (1973) 43. R Naslen,A Guette, J Casper, in Bor. Poluchenie, Struktura, Svoistva (Boron. Production, Structure, Properties) (Moscow: Nauka, 1974) 44. R Naslain, A Quette, P Hagenmuller J. Less-Common Met. 47 1 45. A Quette,M Barret, R Naslain, P Hagenmuller J. Less-Common 46. I I Tupitsyn, I I Lyackovskaya, M S Nakhmanson, A A Sukhik p. 160 (1976) Met. 82 325 (1981) Sov. Phys.-Solid State 16 2915 (1975) 47. D Armstrong, P G Perkins J. Chem. Soc., Faraday Trans. 2, 75 12 (1979) 48. J K Burdett, G J Miller Chem. Mater. 2 12 (1989) 49. A L Ivanovskii, N I Medvedeva Zh. Neorg. Khim. 45 1355 (2000) c 50. G P Shveikin, A L Ivanovskii Usp. Khim. 63 751 (1994) [Russ.Chem. Rev. 63 711 (1994)] 51. A F Guillermet, G Grimvall J. Less-Common Met. 169 257 (1991) 52. X Wang, D Tian, L Wang J. Phys., Condens. Matter 6 10185 (1994) 53. A L Ivanovskii, N I Medvedeva, Yu E Medvedeva Mendeleev 55. A L Ivanovskii, N I Medvedeva, Yu E Medvedeva Metallofiz. Commun. 129 (1998) 54. A L Ivanovskii, N I Medvedeva, Yu E Medvedeva, A E Nikiforov, G P Shveikin Metallofiz. Noveishie Tekhnol. 20 41 (1998) Noveishie Tekhnol. 21 19 (1999) 56. A L Ivanovskii, N I Medvedeva, Yu E Medvedeva, G P Shveikin Dokl. Akad. Nauk 361 642 (1998) b 57. A L Ivanovskii, N I Medvedeva Zh. Neorg. Khim. 44 1717 (1999) c 58. S Nakano, O Fukunaga Diam. Relat. Mater. 2 1409 (1993) 59. G Bocquillon, C Loriers-Susse, J Lorieres J. Mater. Sci.28 3547 (1993) 60. T Endo, O Fukunaga,M Iwata J. Mater. Sci. 14 1676 (1979) 61. T I Serebryakova, V A Ponomarenko, A I Karasev Poroshk. Metall. (11) 77 (1980)732 62. V B Shipilo, L M Gameza, I P Guseva Sverkhtv. Mater. (3) 5 (1981) 63. V B Shipilo, V V Sergeev Poroshk. Metall. (4) 48 (1985) 64. V B Shipilo, L M Gameza Zh. Fiz. Khim. 63 196 (1989) d 65. V B Shipilo, L M Gameza Neorg. Mater. 25 51 (1989) e 66. V S Lysanov (Ed.) El'bor v Mashinostroenii (Elboron in Machine Building) (Leningrad: Mashinostroenie, 1978) 67. V L Solozhenko, V Z Turkevich, W B Holzapfel J. Phys. Chem. B 103 8137 (1999) 68. T Ya Kosolapova, T V Andreeva, T V Bartnitskaya, G G Gnesin, G N Makarenko, I I Osipova, E V Prilutskii Nemetallicheskie Tugoplavkie Soedineniya (Nonmetallic High- Melting Compounds) (Moscow: Metallurgiya, 1985) 69.H Kobayashi, M Katoh, Y Kamiyama, T Mitamura J. Ceram. Soc. Jpn. 100 172 (1992) 70. V Sundaran,K V Logan, R F Speyer J. Mater. Res. 12 2657 (1997) 71. E G Gillan, R B Kaner Chem. Mater. 8 333 (1996) 72. Y Guindon, L Rao, E G Gillan, R B Kaner J. Mater. Res. 10 353 (1995) 73. M Gu, Z Wu, Y Jin,M Kocak J. Mater. Sci. 35 2499 (2000) 74. J A Wilson, J W Steeds,D Wilkes, C M Ward-Close J. Mater. Sci. 36 67 (2001) 75. V G Bamburov, O V Sivtsova, V P Semyannikov, V A Kiselev Ogneupory Tekhn. Keram. (2) 2 (2000) 76. Jpn. P. 62 108 710; Chem. Abstr. 107 63 688 (1987) 77. Jpn. P. 5 105 532 ; Chem. Abstr. 119 78 476 (1993) 78. Jpn. P. 5170 514; Chem. Abstr. 119 187 176 (1993) 79.Jpn. P. 9 080 821; Chem. Abstr. 127 26 010 (1997) 80. Jpn. P. 0 101 936; Chem. Abstr. 101 11 276 (1984) 81. Jpn. P. 0 298 676; Chem. Abstr. 110 140 420 (1989) 82. S L Bud'ko, G Lapertot, C Petrovic, C E Cunningham, N Anderson, P C Canfield Phys. Rev. Lett. 86 1877 (2001) 83. J S Slusky, N Rogado, K A Regan,M A Hayward, P Khalifah, T He, K Inumaru, S M Loureiro, M K Haas, H W Zandbergen, R J Cava Nature (London) 410 343 (2001) 84. D C Larbalestier, L D Cooley, M O Rikel, A A Polyanskii, J Jiang, S Patanik, X Y Cai, D M Feldman, A Gurevich, A A Squitiere,M T Naus, C B Eom, E E Hellstrom, R J Cava, K A Regan, N Rogado,M A Hayward, T He, J S Slusky, P Khalifah, K Inumaru, M Haas Nature (London) 410 186 (2001) 85. J Kortus, I I Mazin, K D Belashchenko, V P Antropov, L L Boyer Phys.Rev. Lett. 86 4656 (2001) 86. K D Belashchenko,M van Schlifgaarde, V P Antropov http://xxx.lanl.gov/abs/cond-mat/0102290 (2001) 87. J M An,W E Pickett Phys. Rev. Lett. 86 4366 (2001) 88. J Gavaler Appl. Phys. Lett. 23 480 (1973) 89. R J Cava, B Batlogg MRS Bull. 14 49 (1989) 90. K Holczer, O Klein, S Huang, R B Kaner, K Fu, R L Whetten, F Diederich Science 252 1154 (1991) 91. S Yamanaka, H Kawaij, K Hothehama, M Ohashi Adv. Mater. 8 771 (1996) 92. H Kawaij, K Hothehama, S Yamanaka Chem. Mater. 9 2127 (1997) 93. S Yamanaka, K Hotehama, H Kawaij Nature (London) 392 580 (1998) 94. M Kohmoto, I Chang, J Friedel http://xxx.lanl.gov/abs/cond-mat/ 0103352 (2001) 95. N B Brandt, S M Chudinov, Y G Ponomarev Modern Problems in Condensed Matter Sciences Vol.20 (North-Holland: Amsterdam, 1988) 96. I T Belash, A D Bronnikov, O A Zharikov, A V Palnichenko Synth. Met. 36 283 (1990) 97. J H Schon, C Kloc, B Batlogg Nature (London) 408 549 (2000) 98. R Osborn, E A Goremychkin, A L Kolesnikov, D G Hinks Phys. Rev. Lett. 87 7005 (2001) 99. A Gerashenko, K Mikhalev, S Verkhovskii, T D'yachkova, A Tyutyunnik, V Zubkov http://xxx.lanl.gov/abs/cond-mat/ 0102421 (2001) 100. A G Joshi, C G Pillai, P Raj, S K Malik Solid State Commun. 118 445 (2001) 101. T J Sato, K Shibata, Y Takano http://xxx.lanl.gov/abs/cond-mat/ 0102486 (2001) 102. O F de Lima, R A Ribeiro, M Avila, C A Cardoso, A A Coelho Phys. Rev. Lett. 86 5974 (2001) A L Ivanovskii 103.Ch Walti, E Felder, C Degen, G Wigger, R Monnier, B Delley, H R Ott http://xxx.lanl.gov/abs/cond-mat/0102522 (2001) 104. S X Dou, X L Wang, J Horvat, D Milliken, E W Collings, M D Sumption http://xxx.lanl.gov/abs/cond-mat/0102320 (2001) 105. D D Lawrie, J P Franck Physica C 245 159 (1995) 106. K O Cheon, I R Fischer, P C Canfield Physica C 312 35 (1999) 107. D K Finnemore, J E Ostenson, S L Bud'ko, G Lapertot, P C Canfield Phys. Rev. Lett. 86 2420 (2001) 108. Y Takano, H Takeya, H Fujii, H Kumakura, T Hatano, K Togano, H Kito, H Ichara http://xxx.lanl.gov/abs/cond-mat/ 0102167 (2001) 109. S Li, H H Wen, Z W Zhao, Y Ni, Z A Ren, G C Che, H Yang, Z Y Li, Z H Zhao http://xxx.lanl.gov/abs/cond-mat/0103032 (2001) 110.C U Jung, M S Park, W N Kang, M S Kim, S Y Lee, S I Lee Physica C 353 162 (2001) 111. W N Kang, C U Jung, K Kim,M S Park, S Y Lee, H Kim, E M Choi, M S Kim, S J Lee http://xxx.lanl.gov/abs/cond-mat/ 0102313 (2001) 112. H Kotegawa, K Ishida, Y Kitaoka, T Muranaka, J Akimitsu http://xxx.lanl.gov/abs/cond-mat/0102334 (2001) 113. G Y Sung, S H Kim, J Kim, D C Yoo, J W Lee, C U Jung, M S Park, W N Kang, D Zhonglian, S I Lee http://xxx.lanl.gov/ abs/cond-mat/0102498 (2001) 114. A Sharoni, I Felner, O Millo Phys. Rev. B 63 220508 (2001) 115. G Karapetrov, M Iavorone, W K Kwok, G W Gabtree, D G Hinks Phys. Rev. Lett. 86 4374 (2001) 116. H Schmidt, J F Zasadzinski, K E Gray, D G Hinks Phys. Rev. B 63 22 504 (2001) 117. B Gorshunov, C A Kutscher, P Haas,M Dressel, F P Mena, A B Kuz'menko, D van der Marel, T Muranaka, J Akimitsu Eur. Phys.J. B 21 159 (2001) 118. R K Kremer, B J Gibson, J S Ahn http://xxx.lanl.gov/abs/ cond-mat/0102432 (2001) 119. Y Wang, T Plackowski, A Junod Physica C 355 179 (2001) 120. T Vogt, G Schneider, J A Hriljac, G Yang, J S Abell Phys. Rev. B 63 220 505 (2001) 121. J E Hirsch Science 292 1521 (2001) 122. J E Hirsch Physica C 341 ± 348 213 (2000) 123. W L McMillan Phys. Rev. Ser. 2 167 331 (1968) 124. N I Medvedeva, A L Ivanovskii, J E Medvedeva, A J Freeman Phys. Rev. B 64 220 502 (2001) 125. G Satta, G Profeta, F Bernardini, A Contonenza, S Massidda http://xxx.lanl.gov/abs/cond-mat/0102358 (2001) 126. S Suzuki, S Higai, K Nakao J. Phys.Soc. Jpn. 70 1206 (2001) 127. I R Shein, N I Medvedeva, A L Ivanovskii Fiz. Tv. Tela (2001) (in the press) f 128. N I Medvedeva, Yu E Medvedeva, A L Ivanovskii http://zhurnal ape relarn ru/articles/2001/008 pdf (2000) 129. N I Medvedeva, Yu E Medvedeva, A L Ivanovskii Dokl. Akad. Nauk 379 (1) (2001) b 130. A Reyes-Serrato, D Galvan http://xxx.lanl.gov/abs/cond-mat/ 0103477 (2001) 131. E Z Kurmaev, I I Lyackovskaya, J Kortus, N Miyata, M Demeter, M Neumann,M Yanagihara, M Watanabe, T Muranaka, J Akimitsu http://xxx.lanl.gov/abs/cond-mat/ 0103487 (2001) 132. T A Callot, L Lin, G T Woods, G P Zhang, J R Thompson, M Paranthaman, D L Erderer http://xxx.lanl.gov/abs/cond-mat/ 0103593 (2001) 133. T Takahashi, T Sato, S Souma, T Muranaka, J Akimitsu Phys.Rev. Lett. 86 4915 (2001) 134. N I Medvedeva, A L Ivanovskii, J E Medvedeva, A J Freeman, D L Novikov http://xxx.lanl.gov/abs/cond-mat/0104346 (2001) 135. C Beyreuther, R Hierl, G Wiech Ber. Bunsenges. Phys. Chem. 79 1081 (1975) 136. I Felner Physica C 353 11 (2001) 137. D Koczorowski, A J Zalevski, O J Zogal, J Klamut http://xxx.lanl.gov/abs/cond-mat/0103571 (2001) 138. A L Ivanovskii Usp. Khim. 66 511 (1997) [Russ. Chem. Rev. 66 459 (1997)] 139. N Sudo, H Hashizume, C Carvalho J. Appl. Crystallogr. 24 286 (1991)Superconducting MgB2 and related compounds: synthesis, properties and electronic structure 141. H Ripplinger, K Schwarz, P Blaha J. Solid State Chem. 133 51 140. S V Okatov, A L Ivanovskii, N I Medvedeva, J E Medvedeva Phys.Status Solidi A 225 R3 (2001) (1997) 142. S Massidda, A Continenza, T M DePascale, R Monnier Z. Phys. B, Condens. Matter 102 83 (1997) 143. A L Ivanovskii, S V Okatov Mendeleev Commun. 11 8 (2001) 144. S Fahy, D R Hamann Phys. Rev. B 41 7587 (1990) 145. K Kusakabe, Y Tateyama, T Oguchi, S Tsuneyumi Rev. High Pressure Sci. Technol. 7 193 (1998) 146. S Massidda, A Baldereschi Solid State Commun. 66 855 (1985) 147. A H Silver, T Kushida J. Chem. Phys. 38 865 (1963) 148. K Schwarz, H Ripplinger, P Blaha Z. Naturforsch., A Phys. Sci. 51 527 (1996) 149. Y Kong, O V Dolgov, O Jepsen, O K Andersen Phys. Rev. B 64 020501 (2001) 150. K Bohnen, R Heid, B Renker Phys. Rev. Lett. 86 5771 (2001) 151. A Lin, I I Mazin, J Kortus http://xxx.lanl.gov/abs/cond-mat/ 0103570 (2001) 152.N I Mededeva, Yu E Medvedeva, A L Ivanovskii, V G Zubkov, A Friman Pis'ma Zh. Eksp. Teor. Fiz. 73 378 (2001) 153. Y P Sun,W H Song, J M Dai, B Zhao, J J Du, H H Wen, Z X Zhao Chin. Phys. Lett. 18 587 (2001) 154. SMKazakov,MAngst, J Karpinski Solid State Commun. 119 1 (2001) 155. Y G Zhao, X P Zhang, P T Qau, H T Zhang, S L Jia, B S Cao, M H Zhu, Z H Han, X L Wang, B L Gu http://xxx.lanl.gov/abs/ cond-mat/0103077 (2001) 156. J S Ahn, E J Choi http://xxx.lanl.gov/abs/cond-mat/0103169 (2001) 157. S Zhang, J Zhang, T Zhao, C Rong, B Shen, Z Cheng Chin. Phys. 10 335 (2001) 158. T Takenobu, T Ito, D H Chi, K Prassides, Y Iwasa http://xxx.lanl.gov/abs/cond-mat/0103241 (2001) 159. N V Vekshina, L Ya Markovskii, Yu D Kondrashev, T K Voevodskaya Zh.Prikl. Khim. 44 958 (1971) g 160. C U Jung,M Park,W N Kang, K Kim, S Y Lee, S Lee Physica C 353 162 (2001) 161. K Prassides, Y Iwasa, T Ito, D Chi, K Uehara, E Nushibori, M Takata, S Sakata, Y Ohishi, O Shimomura, T Muranaka, J Akimitsu Phys. Rev. B 64 2509 (2001) 162. E Saito, T Taknenobu, T Ito, Y Iwasa, K Prassides, T Arima J. Phys., Condens. Matter 13 L267 (2001) 163. J D Jorgensen, D G Hinks, S Short Phys. Rev. B 63 220508 (2001) 164. B Lorenz, R L Meng, C W Chu http://xxx.lanl.gov/abs/cond-mat/ 0102264 (2001) 165. I Loa, K Syassen Solid State Commun. 118 279 (2001) 166. T Tomita, J J Hamlin, J S Schilling http://xxx.lanl.gov/abs/ cond-mat/0103538 (2001) 167. A E Kar'kin, V V Voronin, T V D'yachkova, N I Kadyrova, A P Tyutyunnik, V G Zubkov, Yu G Zainulin,M V Sadovskii, B N Goshchitskii JETP Lett.73 570 (2001) 168. G Rubio-Bollinger, H Suderow, S Vieira Phys. Rev. Lett. 86 5582 (2001) 169. A A Zhukov, K Yates, K G Perkins, Y Bugoslavsky, M Polichetti, A Berenov, J Driscoll, A D Caplin, L F Cohen Supercond. Sci. Technol. 14 L13 (2001) 170. W N Kang, H Kim, E Choi, C U Jung, S Lee http://xxx.lanl.gov/ abs/cond-mat/0103179 (2001) 171. A Brikman, D Mijatovich, G Rijneders, V Leca, H Smodle, I Oomen, A A Golubov, F Roesthuis, S Hakema, H Hilgenkamp, D Blank, H Rogalla Physica C 353 1 (2001) 172. D Blank, H Hilgenkamp, A Brikman, D Mijatovich, G Rijneders, H Rogalla Appl. Phys. Lett. 79 394 (2001) 173.H M Christen, H Y Zhai, C Cantoni,M Paranthaman, B C Sales, R Rouleau,D P Norton,D K Christen,D Lowndes Physica C 353 157 (2001) 174. S R Sinde, S B Ogale, R L Greene, T Venkatesan, P C Canfield, S L Bud'ko, G Lapertot, C Petrovic http://xxx.lanl.gov/abs/ cond-mat/0103542 (2001) 175. C B Eom, MK Lee, J H Choi, L Belenky, X Song, L D Cooley, MT Naus, S Patanik, J Jiang,MRickel, A Polyanskii, A Gurevich, X Y Cai, S D Bu, S E Babcock, E E Hellstrom, D C Larbalestier, N Rogano, K A Regon,M A Hayward, T He, J S Slusky, K Inumara,M K Haas, R J Cava Nature (London) 411 558 (2001) 733 176. M Paranthaman, C Cantoni, H Y Zhai, H M Christen, T Aytug, S Sathyamuthy, E D Specht, J D Thompson, D H Lowndes, H R Kerchner, D K Christen Appl. Phys. Lett.78 3669 (2001) 177. G G Grassano,W Ramadan, V Ferrando, E Bellingeri, D Marre, C Ferdeghini, G Grasso,M Putti, A S Siri, P Manfrinetti, A Palenoza, A Chincarini http://xxx.lanl.gov/abs/cond-mat/ 0103572 (2001) 178. H Y Zhai, H M Chirsten, L Zhang, C Cantoni,M Paranthaman, B C Sales, D K Christen, D H Lowndes http://xxx.lanl.gov/abs/ cond-mat/0103588 (2001) 179. Z K Liu, D G Schlom, Q Li, X Xi Appl. Phys. Lett. 78 3678 (2001) 180. N Saunders, A P Miodownik CALPHAD (Calculation of Phase Diagrams): a Comprehensive Guide (Oxford: Pergamon Press, 1998) 181. R P Vasquez, C U Jung, M S Park, H-J Kim, J-Y Kim, S-I Lee Phys. Rev. B 64 052510 (2001) 182. C L Perkins, R Singh,M Trenary, T Tanaka, Y Paderno Surf. Sci. 470 215 (2001) 183. P C Canfield, D K Finnemore, S L Bud'ko, J E Ostenson, G Lapertot, C E Cunningham, C Petrovic Phys.Rev. Lett. 86 2423 (2001) 184. S E Cunningham, C Petrovic, G Lapertot, S L Bud'ko, F Laabs, W Straszheim, D K Finnemore, P C Canfield Physica C 353 5 (2001) 185. S L Bud'ko, C Petrovic, G Lapertot, C E Gunningham, P C Canfield Phys. Rev. B 63 220503 (2001) 186. V Metlushko, U Welp, A Koshelev, I Aranson, G Crabtree, P C Canfield Phys. Rev. Lett. 79 1738 (1998) 187. K Rathnayaka, A K Bhatnagar, A Parasiris, D Naugle, P C Canfield, B Cho Phys. Rev. B 55 8506 (1997) 188. S V Shulga, S Drechsler, G Fuchs, K Muller, K Winzer, M Heinecke, K Krug Phys. Rev. Lett. 80 1730 (1998) 189. M D Sumption, X Peng, E Lee,M Tomsic, E W Collings http://xxx.lanl.gov/abs/cond-mat/0102441 (2001) 190.G Grasso, A Malagoli, C Ferdeghini, S Roncallo, V Braccini, M R Cimberle, A S Siri Appl. Phys. Lett. 79 230 (2001) 191. A Sharoni, O Millo, G Leitus, S Reich J. Phys., Condens Matter 13 L503 (2001) 192. C U Jung, J Choi, P Chowdhury, H P Kijoon, K Kim,M Park, H Kim, J Kim, Z Du,M Kim,W N Kang, S Lee, G Sung, 194. J Hlinka, I Gregora, J Pokorny, A Plecenik, P Kus, L Satrapinsky, J Y Lee http://xxx.lanl.gov/abs/cond-mat/0105330 (2001) 193. M Xu, H Kitazawa, Y Takano, J Ye, K Nishida, H Abe, A Matsushita, G Kido http://xxx.lanl.gov/abs/cond-mat/0105271 (2001) S Benacka http://xxx.lanl.gov/abs/cond-mat/0105275 (2001) 195. S Lee, H Mori, T Masui, Yu Eltsev, A Yamamoto, S Tajima http://xxx.lanl.gov/abs/cond-mat/0105545 (2001) 196. Appl. Russ. 108 865 (2001) 197. W N Kang, H Kim, E M Choi, C U Jung, S L Lee Science 292 1521 (2001) 198. J H Jung, K W Kim, H J Lee,M W Kim, T W Noh,W N Kang, H Kim, E M Choi, C U Jung, S L Lee http://xxx.lanl.gov/abs/ cond-mat/0105180 (2001) 199. S Patnaik, L D Cooley, A Gurevich, A A Polyanskii, J Jiang, X Y Cai, A A Squitieri,M T Naus, M K Lee, J H Choi, L Belenky, S D Bu, S D Letteri, X Song, D Schlom, S E Babcock, C B Eom, E E Hellstrom, D C Larbalestier Supercond. Sci. Technol. 14 315 (2001) 200. H J Zhai, H M Christen, L Zhang,M Paranthaman, C Cantoni, B C Sales, P H Fleming, D K Christen, D H Lowndes http://xxx.lanl.gov/abs/cond-mat/0103618 (2001) 201. Z W Zhao, H H Wen, S L Li, Y M Ni, H P Yang,W N Kang, H J Kim, E M Choi, S I Lee http://xxx.lanl.gov/abs/cond-mat/ 0104249 (2001) 202. S F Wang, S Y Dai, Y I Zhou, Z H Chen, D F Cui, J D Yu, M He, H B Lu, G Y Yang Chin. Phys. Lett. 18 967 (2001) 203. X D Zeng, A Sukiasyan, X X Xi, Y F Hu, E Wertz, Q Li, W Tian, H P Sun, X Pan, J Lettieri, D Schlom, C O Brubaker, Z Liu http://xxx.lanl.gov/abs/cond-mat/0105080 (2001) 204. A H Li, X L Wang,M Ionescu, S Soltonian, J Horvat, T Silver, H K Liu, S X Dou http://xxx.lanl.gov/abs/cond-mat/0104156 (2001) 205. H J Zhai, H MChristen, L Zhang, MParanthaman, P H Fleming, D H Lowndes http://xxx.lanl.gov/abs/cond-mat/0104309 (2001)A L Ivanovskii 734 206. S Y Lee, J H Lee, J S Ryu, J Lim, S H Moon, H N Lee, H G Kim, B Oh http://xxx.lanl.gov/abs/cond-mat/0105327 (2001) 207. H J Kim,W N Kang, F M Choi, M S Kim, K H P Kim, S I Lee http://xxx.lanl.gov/abs/cond-mat/0105363 (2001) 208. M H Jung,M Jaime, A H Lacerda, G S Boebinger, W N Kang, H J Kim, E M Choi, S I Lee http://xxx.lanl.gov/abs/cond-mat/ 0106146 (2001) 209. A Berenov, Z Lockman, X Qi, Y Bugoslavsky, L F Cohen, M H Jo, N A Stelmashenko, V N Tsaneva, M Cambara, N Hari Babu, D A Cardwell, M G Blamire, J L Mac Manus- Driscoll http://xxx.lanl.gov/abs/cond-mat/0106278 (2001) 210. C Ferdeghini, V Ferrando, G Grassano,WRamadan, E Bellingeri, V Braccini, D Marre, P Manfrinetti, A Palenzona, F Borgatti, R Felici, T L Lee http://xxx.lanl.gov/abs/cond-mat/0107031 (2001) 211. V M Silkin, E V Chulkov, P M Echenique http://xxx.lanl.gov/abs/ cond-mat/0105628 (2001) 212. Z Li, J Yang, J G Hou, Q Zhu http://xxx.lanl.gov/abs/cond-mat/ 0106590 (2001) 213. V A Gasparov, N I Sidorov, I I Zver'kova,M P Kulakov JETP Lett. 73 532 (2001) 214. G K Strukova, V F Degtyareva, D V Shivkun, V N Zverev, V M Kiiko, A M Ionov, A N Chaika http://xxx.lanl.gov/abs/ cond-mat/0105293 (2001) 215. D P Young, P W Adams, J Y Chan, F R Fronczek http://xxx.lanl.gov/abs/cond-mat/0104063 (2001) 216. T He, Q Huang, A P Ramirez, Y Wang, K A Regan, N Rogado, M A Hayward,M K Haas, J S Slusky, K Inumaru, H W Zandbergen, N P Ong, R J Cava Nature (London) 411 6833 217. A Taraphder, R Pandit, H R Krishnamurthy, T V Ramakrishnan 218. A L Ivanovskii Usp. Khim. 68 119 (1999) [Russ. Chem. Rev. 68 103 219. S Y Li, R Fan, X H Chen, C H Wang,W Q Mo, K Q Ruan, (2001) Int. J. Mod. Phys. B 10 863 (1998) (1999)] Y M Xiong, X G Luo, H T Zhang, L Li, Z Sun, L Z Cao http://xxx.lanl.gov/abs/cond-mat/0104554 (2001) 220. Z Q Mao, M M Rosario, R Nelson, K Wu, I G Deac, P Schiffer, Y Liu, T He, K A Regan, R J Cava http://xxx.lanl.gov/abs/ cond-mat/0105280 (2001) 221. Q Huang, T He, K A Regan, N Rogado, M A Hayward, M K Haas, K Inumaru, R J Cava http://xxx.lanl.gov/abs/ cond-mat/0105240 (2001) 222. M A Hayward,M K Haas, T He, K A Regan, N Rogado, K Inumaru, R J Cava http://xxx.lanl.gov/abs/cond-mat/0104541 223. Z A Ren, G C Che, S L Jia, H Chen, Y M Ni, Z X Zhao 224. I R Shein, N I Medvedeva, A L Ivanovskii JETP Lett. 74 122 225. S B Dugdale, T Jarlborg http://xxx.lanl.gov/abs/cond-mat/ 226. J H Shim, B I Min http://xxx.lanl.gov/abs/cond-mat/0105418 227. D J Singh, I I Mazin http://xxx.lanl.gov/abs/cond-mat/0105577 228. I R Shein, N I Medvedeva, A L Ivanovskii http://xxx.lanl.gov/ (2001) http://xxx.lanl.gov/abs/cond-mat/0105366 (2001) (2001) 0105349 (2001) (2001) (2001) abs/cond-mat/0107010 (2001) a�Russ. J. Gen. Chem. (Engl. Transl.) b�Dokl. Chem. (Engl. Transl.) c�Russ. J. Inorg. Chem. (Engl. Transl.) d�Russ. J. Phys. Chem. (Engl. Transl.) e�Inorg. Mater. (Engl. Transl.) f�Phys. Solid State (Engl. Transl.) g�Russ. J. Appl. Chem. (

ISSN:0036-021X    

出版商:RSC    

年代:2001

数据来源: RSC

摘要:

Russian Chemical Reviews 70 (9) 735 ± 776 (2001) Synthetic methodologies for carbo-substituted conjugated dienes A A Vasil'ev, E P Serebryakov Contents I. Introduction II. Elimination reactions III. Olefination of carbonyl compounds IV. Organometallic reactions V. The synthon approach to conjugated dienes VI. Electrocyclic reactions VII. Methods based on migration or reduction of multiple bonds VIII. Miscellaneous methods IX. Conclusion Abstract. and methods encompasses review this of scope The The scope of this review encompasses methods and synthetic protocols leading to conjugated dienes bearing no synthetic protocols leading to conjugated dienes bearing no heteroatoms concen- is Attention moiety. the in heteroatoms in the C=C C=C7C=C C=C moiety.Attention is concen- trated decade, last the during developed methodologies on trated on methodologies developed during the last decade, partic- partic- ularly, The synthesis. complex-mediated metal the to ularly, to the metal complex-mediated synthesis. The bibliography bibliography includes 494 references includes 494 references. I. Introduction Until about 1960, the synthetic chemistry of conjugated dienes was focused mainly on their simplest representatives. Usually, these dienes were prepared either by dehydrating diols and allylic or homoallylic alcohols or by eliminating the elements of HX (X=OAlk, OCOR, OSO2R, Hal, etc.) from properly selected unsaturated substrates. However, such techniques were unsuit- able for the synthesis of higher homologues, because the use of these reactions resulted in mixtures of isomers.Then, in the mid-sixties a new period in the synthesis of conjugated dienes began, as this class of compounds became increasingly in demand. This popularity was due to the fact that dienes with fixed position of configurationally defined double bonds and with various sensitive functional groups (and their combinations) had been identified by that time as fragments of many natural products. Moreover, inter- and intramolecular Diels ± Alder reactions employing conjugated dienes had become widely used tools in the synthesis of cyclic molecules. Advances in organometallic chemistry and metal complex catalysis highlighted dienes as useful four-electron p-ligands capable of forming interesting metal complexes.These aspects of diene chemistry are considered, with due attention to details, in two recently published review collections.1, 2 Finally, stereospe- cific 1,4-cis hydrogenation of conjugated dienes using chromium A A Vasil'ev, E P Serebryakov N D Zelinsky Institute of Organic Chemistry, Russian Academy of Sciences, Leninsky prosp. 47, 119991 Moscow, Russian Federation. Fax (7-095) 135 53 28. Tel. (7-095) 938 35 30. E-mail: vasiliev@ioc.ac.ru (A A Vasil'ev), ser@ioc.ac.ru (E P Serebryakov) Received 1 June 2001 Uspekhi Khimii 70 (9) 830 ± 873 (2001); translated by E P Serebryakov, Z P Bobkova #2001 Russian Academy of Sciences and Turpion Ltd DOI 10.1070/RC2001v070n09ABEH000683 735 735 742 749 760 763 766 768 770 carbonyl complexes proved to be an efficient, preparatively viable method for stereocontrolled synthesis of alkenes.3, 4 Abundant literature on the synthesis of conjugated dienes has been accumulated, and systematic surveys have been periodically attempted during the last thirty years.Among early sources (up to the late 1980's), the serial edition of Houben ± Weyl 5 and a few monographs and collective volumes 6± 9 should be mentioned, because some chapters there are dedicated to conjugated dienes. Another good source of references (not commented on) is a compendium of reactions leading to dienes.10 Also useful is a monograph,11 where numerous methods for constructing conju- gated dienes are cited in the context of pheromone synthesis.A review on the synthesis of conjugated polyenes, where the dienes were considered as parental polyene systems,12 was published in 1997 in a book 1 that is nearly inaccessible in Russia. This makes the appearance of an updated review quite opportune, let alone the fact that in the last four years the chemistry of conjugated dienes has been enriched with some new and interesting synthetic methods. This review aims at a systematic survey of selective methods for obtaining conjugated carbo-substituted dienes that were elaborated between the end of 1960 and March, 2001. Methods leading to the molecules with heteroatomic substituents on the C=C7C=C unit will be mentioned just in a few exclusive cases.The description of reactions used will be focused on their preparative aspects; the mechanisms will be mentioned only when it is necessary, since most of them have been considered in detail in original publications. II. Elimination reactions Elimination reactions leading to alkenes basically can be divided into two groups. The first group involves the ejection of a nucleofugal group and a vicinal hydrogen or metal atom. The second one involves reductive elimination of two vicinal groups. When the substrate already has a double bond adjacent to one of the groups undergoing elimination, the product of the reaction is a conjugated diene. 1. Elimination of HX Elimination of one molecule of water from 2-methylbut-3-en-2-ol (Favorskii's method) or of two molecules of water from butane- 1,3-diol (the last step in Lebedev's process) as well as catalytic736 thermolysis of 4,4-dimethyl-1,3-dioxane (Farberov's method) are classical ways of obtaining dienic monomers applicable to other lower conjugated dienes.However, the conditions under which the processes are performed are too drastic to be useful for obtaining more labile, functionally substituted higher dienes. Elimination of HX from substrates of allylic or homoallylic type proceeds regiospecifically only for molecules with specific patterns of substitution, e.g., for compounds 1 13 or 2.14 If the substrate contains a functional group strongly stabilising only one of the possible isomeric products, this also can favour an unam- biguous course of the process (e.g., in the case of reactions 3?4 15 and 5?6 16).OTs ButOK, DMSO, 75 8C (80%) 1 NSiMe3 R2 OAc R2 1. MeC OSiMe3 2. Mo(CO)6, D R1 R1 CO2Me CO2Me (68% ± 95%) 2 R2=H, CH2R1. OH 1. Ac2O± Py ±DMAP 2. DBU 3. H3O+ Ar Ar (68%) 4 3 CO2Et CO2Et R1 LDA R1 OR2 OR2 (52% ± 80%) 6 OMOM 5 DMAP is 4-dimethyaminopyridine; DBU is 1,8-diazabicyclo- [5.4.0]undec-7-ene; MOM is methoxymethyl; LDA is lithium diisopropylamide. Generally, elimination of HX or MX from allylic-type sub- strates affords isomer mixtures. However, in the presence of palladium catalysts the reaction proceeds regioselectively. In these cases substituents R1 in starting compounds 7 ± 9 can contain functional groups.17, 18 R1 OR2 OAc R2O Pd Pd(OAc)2, Ph3P R1 100 8C 7 or OR2 H R1 8 OAc R2O Pd R1 R1 (62% ± 84%) H OAc Pd(OAc)2, Ph3P, CaCO3 R2 R1 R1 R2 (71% ±96%) 9 In this way neryl acetate was converted to myrcene, while geranyl acetate afforded a mixture of cis and trans isomers of b-ocymene.19 In the presence of palladium catalysts, the ring opening in acyclic vinylic oxiranes 10 and 11 proceeds smoothly to give a-dienols 12 and 13.20 The regioselectivity of these reactions depends on the position of substituents at the epoxide ring.A A Vasil'ev, E P Serebryakov OH O Pd(PPh3)4, D R1 R1 R3 (39% ± 90%) R3 12 10 R2 R2 OH O Pd(PPh3)4, D R1 R3 R1 (79% ±95%) R3 R2 11 R2 13 Elimination of CO2 from alkenyl-substituted cyclic carbo- nates 14 takes place in the same manner, but the resulting diene 15 is obtained predominantly as the E,E-isomer whatever the con- figuration of the starting compound.21 OCH2C6H4OMe OCH2C6H4OMe Pd2(dba)3 .CHCl3, PPh3, THF, D R O R HO (85% ± 93%) O 15 14 O dba is dibenzylideneacetone.Epoxides 16 (substituents R1 and R2 can contain ester group- ings) undergo regiospecific formal dehydration mediated by organophosphorus reagents.22, 23 (Ph3P+)2O. 2 TfO7 or O ClP(O)(NMe2)2, H2O R2 R1 R2 R1 (50% ± 92%) 16 A three-step protocol of formal a,a 0-elimination of water from allylic alcohols 17 is based on their chemoselective epoxida- tion in the presence of a silylating agent and vanadium cata- lysts.24, 25 Subsequent opening of the epoxide ring in compounds 18 by mild Lewis acids proceeds regiospecifically (the ejection of proton occurs exclusively from the substituent lying in the trans- position relative to the silyloxy group), and affords unsaturated diols 19.Finally, circumventional elimination of vicinal hydroxy groups from 19 gives an additional double bond. In this way nerol is readily transformed to myrcene, and geraniol to trans-b- ocymene. O ButOOH, R3 VO(acac)2, R3 Et2AlNR2, Bu4NF R1 Me3SiCl R1 OH R2 (50% ±79%) OSiMe3 R2 18 17 OH R3 PBr3, CuBr/Zn R3 R1 R1 (38% ± 75%) R2 OH R2 19 2. Deamination An interesting method for the conversion of pyridines 20 into allylic (Z,E)-dienylamines 21 begins with quaternisation of their nitrogen atom followed by reduction, another methylation and subsequent Hofmann degradation of the resulting intermediate A (Scheme 1).26, 27 Then, after yet another quaternisation, the amino group of compound 21 can be replaced by a carbonaceous substituent.The intermediate Z,E-dienes 21 can be isomerised to give the corresponding E,E-dienes 22 by treating the correspond- ing N-oxides with trifluoroacetic acid, and subsequently by reductive substitution. This protocol was employed in the syn- thesis of insect pheromones.27 3. Reductive elimination The reduction of alk-2-ene-1,4-diols 28 23 or of the respective ethers 29, 30 in the system LiAlH4 ± TiCl3 was used in the synthesis of retinoids 24 31, 32 and other natural products.30, 33Synthetic methodologies for carbo-substituted conjugated dienes Scheme 1 Me Me Me N R1 N R1 R1 N I7 + MeI KOH, D 1.MeI 2. NaBH4 (65%) A 20 1. MeI 2. Li2CuCl4, R2MgX R1 R2 R1 NMe2 21 O MCPBA CF3CO2H R1 NMe2 21 + R1 NMe2 CF3COO7 1. MeI 2. Li2CuCl4, R2MgX R1 NMe2 22 R2 R1 MCPBA is m-chloroperbenzoic acid. R1 OHR3 R3 LiAlH4 ± TiCl3 R1 R4 R2 (42% ±83%) R2 R4 OH 23 OH OSiR3 (85%± 90%) OH OSiR3 24 Reductive elimination of halogen atoms from 1,2,3,4-tetraha- loalkanes 25, 1,2-dihaloalk-3-enes 26 and 1,4-dihaloalk-2-enes 27 gives 1,3-dienes in good yields. In addition to the traditional zinc,34, 35 the salts of arenetellurols were recommended.36 In the latter case, the process can be performed readily using catalytic amount of Ar2Te2 that is periodically reduced by titrating the reaction mass with sodium borohydride. This method was applied to a number of linear and cyclic substrates, including those bearing additional functional groups.{ X X X 25 X Zn/Cu or Zn/Hg, or NaBH4±Ar2Te2 (Cat.) X (4100%) X 26 X X 27 { 1,4-Dibromoalk-2-enes 27 can be prepared by treating internal 35 and, what is particularly interesting, terminal alkenes with 2 equiv.N-bromo- succinimide.36 In the latter case, the bromination occurs in a stepwise mode involving allylic rearrangement. 1,2-Dibromoalk-3-enes 26 can be prepared from the corresponding diols which, in their turn, are obtainable from allylic alcohols.25 737 Reductive scission of cis-3-chloro-2-vinyltetrahydrofuran (28) with samarium iodide results in (E,E)-hexa-3,5-dien-1-ol (29) of >97% geometric purity.37 The same stereochemical result is observed in the case of 2-alkynyl-3-chlorotetrahydrofurans (30, n=1).By contrast, their tetrahydropyran homologues (30, n=2) in the same reaction afford the Z-configured enynols 31, suppos- edly due to different conformational behaviour of five-membered and six-membered heterocycles.37 Cl SmI2, THF HO (84%) 29 O 28 Cl SmI2, THF (H2C)n C CR CR O30 n=1 C HO (72% ±84%), >99%(E) CR n=2 C HO (73% ± 92%), >97% (Z) 31 When allylic nitro compounds of the type 32 are treated with chromium(II) diacetate [Cr(OAc)2 ± Dipy system in dimethyl- formamide], their vicinal nitro- and mesyloxy (or acetoxy) groups undergo reductive elimination, which results in the formation of an additional double bond.These transformations were used to obtain conjugated dienes 33, albeit in moderate yields.38 NO2 7 R1 R1 R1 X X X e7 7NO¡27X7 R2 R2 R2 R3 R3 R3 32 R1 R2 R3 33 X=OMs, OAc. 4. Stereocontrolled 1,2-elimination of R3SiX and R3SnX The Peterson reaction (see a review 39) consists in simultaneous elimination of vicinal hydroxy and silyl groups and results in the formation of a double bond. This transformation can occur both in the presence of bases (e.g., ButOK, KH) and acids (AcOH, H2SO4). The double bond configuration in product(s) 34 depends on the reaction conditions. When the starting compound 35 is treated with a base, the primarily formed alkoxide undergoes syn-elimination upon which a molecule of metal silanolate is ejected.When the process is initiated with an acid, it starts with protonation of the hydroxy group, followed by anti-elimination of a molecule of water, together with a silyl cation. R2 R1 R1 R2 SiR33SiR3 7H+ 3 7R33 SiO7 O7 R1 (Z)-34 R2 SiR33H+ OH R1 R1 R2 35 R2 7R33 Si+,7H2O (E )-34 OHá2 , This approach can be used for the synthesis of dienes, when a double bond is already present in the starting compound. Depend- ing on the mutual arrangement of substituents in substrates 36 and738 37 and on the process conditions, one obtains diene(s) 38 in the required configuration.{ SiMe3 SiMe3 R R H+ Base R OH threo-36 OH erythro-36 (Z)-38 Base H+ OH OH R (E )-38 R R SiMe3 threo-37 SiMe3 erythro-37 In a vinylogous version of the Peterson reaction, compounds 39 and 40 in which the allylic hydroxyl and the silyl group are located at the opposite ends of the double bond are employed as the starting substrates.50, 51 If the silyl group is attached to the terminal carbon atom (as in compounds 39 and 40), the resulting diene 38 has always the E-configuration.When both the hydroxyl and the silyl group are attached to secondary carbon atoms in erythro relationship, the reaction affords the E,E-dienes.52. How- ever, in the case of their threo-relationship, the resulting dienes are in the E,Z-configuration, the Z-bond being formed at the hydroxyl-bearing carbon atom.OH KH or HCl R SiMe3 (69% ± 84%) 39 R (E)-38 HCl R SiMe3 (100%) OH 40 Compounds 41 containing Group IV and VI higher elements (e.g., S and Sn, instead of O and Si) react similarly and sponta- neously under very mild conditions; no catalyst is needed.53 R1 1. BuLi, THF,778 8C 2. Bu3SnCH2I SAr R2 SnBu3 R1 R1 778 to 0 8C 7Bu3SnSAr SAr R2 R2(42% ± 89%) 41 . Ar=N 5. Oxidative 1,2-syn-elimination of RSeOH and RSOH Oxidative elimination of an organoselenium fragment from ali- phatic selenides is a relatively new method of alkene synthesis (see reviews 54 ± 57). This process involves the oxidation of a selenide to the corresponding selenoxide, which then undergoes spontaneous syn-elimination to give an alkene and a selenenic acid.This method has been widely used in syntheses of natural products of rather complicated structure, because there are many versatile methods for obtaining necessary organoselenium precursors (see, inter alia, the publications on the synthesis of a-arylseleno-g- alkenoic acids 58 ± 60). Elimination of RSeOH from homoallylic selenides 42 leads to conjugated alkyl dienoates 43.60 ± 65 { Various protocols for the preparation of starting b-hydroxysilanes have been elaborated.40 ± 49 Particularly interesting are the reagents of the type R3SiCH2X (X=CO2 Alk, CN, etc.), which are widely used in the olefination of carbonyl compounds.46 ± 49 A A Vasil'ev, E P Serebryakov SePh R1 R1 NaIO4 or H2O2 1.LDA 2. Ph2Se2 CO2Me CO2Me R2 R2 42 CO2Me R1 R1 SePh CO2Me 7PhSeOH H R2 R2 43 O By contrast, instead of giving conjugated dienes via oxidative elimination, the oxidation of allylic selenides 44 mostly triggers a [2,3]-sigmatropic rearrangement that finally results in `re- arranged' allylic alcohols.66, 67 O R R SeAr [O] Se Ar Nu 44 Nu SeAr OH O R R R=Alk, Ar 0; Ar = p-NO2C6H4 . Allylic 2,4-dinitrophenyl selenides appear to be an exclusion,68 but even in this case, the resulting conjugated dienes are accom- panied by noticeable amounts of allylic alcohols. Due to high nucleophilicity of areneselenolates, they can open the five-membered lactone ring in compounds 45 to give selenides 46, and subsequently conjugated dienes 47 with a functional group in the side chain.69 R 1.PhSeNa 2. CH2N2 R [O], 40 ± 50 8C O (68% ± 73%) SePh CO2Me 46 O 45 R CO2Me 47 By analogy with selenoxides, aliphatic sulfoxides can undergo oxidative elimination to give alkenes, but at higher temperatures. This can be illustrated by the fragmentation of sulfoxides that were prepared by oxidising sulfides 48 and 49.70, 71 The latter were obtained upon the addition of sulfur-containing cationoid reagents to terminal alkenes. p-ClC6H4SCH2OC(O)CF3 R (*80%)SC6H4Cl-p MCPBA R 48 O D SC6H4Cl-p R R (65% ± 82%) OCOCF3 MeS CO2Et CO2Et NaIO4 R R SMe 49 D CO2Et CO2Et R R S Me O Elimination of a sulfoxide unit was used as the final stage in the synthesis of 2-hydroxymethylbuta-1,3-diene from 2-bro- moethyl phenyl sulfide.72Synthetic methodologies for carbo-substituted conjugated dienes a-Sulfinyl esters 50, immediate precursors of alkyl alka-2,4- dienoates, are stable at room temperature.They can be readily prepared from synthon 51, which already contains a sulfoxide grouping.73 7 R1 R1 Ph(O)S CO2Me (51), PdCl2 Ph3P, DME, 25 8C R2 R2 R3 R3 CO2Me CO2Me R1 R1 S(O)Ph 70 8C R2 R2 R3 R3 50 DME is 1,2-dimethoxyethane. Unlike allylic selenoxides, allylic sulfoxides can be used for obtaining conjugated dienes.66, 74Aconvenient way of performing this process involves the esterification of allylic alcohols 52 with arenesulfenyl chlorides (the best results are obtained when Ar=2,4-dinitrophenyl) followed by the rearrangement of inter- mediate sulfenates 53 into sulfoxides 54; the latter then under- going elimination to give dienes 55.ArSO OH [2,3] ArSCl R1 R1 R2 R2 53 52 O Ar S R1 R1 (70% ± 80%) R2 R2 (E,E )-55 54 Conjugated dienes with extended carbon chains (e.g., dienic alcohol 56) can be synthesised from suitable allylic sulfides.75 S(O)Ph 1. LDA 2. Br(CH2)8OH 3. MCPBA PhMe, Et3N, D (CH2)7OH SPh 7PhSOH (68%) 57 (CH2)7OH 56 (60%) Vinyl sulfoxides of the type 58 rearrange on heating to allylic sulfoxides 59 that afford conjugated dienes upon eliminating ArSOH.76 Py CO2Et CO2Et D R CHO +Ar(O)S R S(O)Ar 58 CO2Et CO2Et R R S(O)Ar 59 6.1,2-Elimination of ArSO2H On treatment with bases, alkyl phenyl sulfones eject the phenyl- sulfinate anion to give alkenes. If the molecule of sulfone has already a double bond, this reaction leads to dienes. Asynthesis of conjugated dienes from homoallylic sulfones 60 is known,77 where the latter were obtained by sequential alkylation of bis(phenylsul- fonyl)methane (61) first with allyl- and then with alkyl halides. 739 1. NaH, Br SO2Ph 2. NaH, AlkI Al/Hg, SO2Ph THF±H2O Alk SO2Ph SO2Ph 61 Alk Alk ButOK SO2Ph 60 b-Acetoxysulfones 62 can be used as the precursors of alkenes 63 in the Julia method of olefination (aldol-like alkylation of an alkyl phenyl sulfone followed by acylation to give 62, then elimination of AcOM, and reduction of the resulting vinyl sulfones 64 to give alkene 63) 78, 79 However, when a large excess of a stronger base is used (*10 equiv.), both substituents in the acetoxy sulfone 62 undergo elimination, which gives rise to the conjugated dienes 55.80, 81 The dienes 55 can be obtained from the sulfones 64 either directly, or via intermediate allylic sulfones 65 (Scheme 2).Scheme 2 1. BuLi R2 CHO 2. R2 AcO 3. Ac2O NaOH or ButOK (1 equiv.) R1 PhO2S R1 PhO2S 62 LiAlH4 R2 R1 R2 63 R2 R2 ButOK (10 equiv.) ButOK (0.3 equiv.) R1 PhO2S 64 R1 R1 PhO2S 55 65 This approach is also applicable to the synthesis of amides (66) 82, 83 or esters (67) 83 of conjugated dienoic acids.O N R 66 O ButOK (70%) N 1. (CH2)2CHO O SO2Ph 2. Ac2O(64%) N R SO2Ph OAc BuLi 68 O R 1. O OEt PhO2S 2. TsOH, D O O (63% ±82%) R 69 (53% ± 70%) 1. ButOK 2. CH2N2 O OMe R 67 It should be noted that in the former case, there is no need to isolate the intermediate amides 68, while in the latter case, the isolation of (1-sulfonylalkyl)butyrolactones 69 is necessary. The terminal 1,3-dienes (E)-38 are readily converted into the corresponding 2-phenylsulfonyl-1,3-dienes 70 upon treatment with PhSO2SePh (see a review 84). The latter are strongly electro- philic, and thus are capable of adding C-nucleophiles. This is a740 PhSO2SePh R1 (E)-38 R2CH2NO2, DBU R1 NaCN, 15-crown-5 R1 R1 convenient method for the introduction of a functionalised alkyl substituent to the 1-position of conjugated dienes (Scheme 3).85, 86 7.Fragmentation of a-halogen-substituted sulfones The `vinylogous' Ramberg ± BaÈ cklund reaction (see a review 87) is yet another example of obtaining 1,3-dienes from sulfones. From preparative standpoint, it is a C1-homologation of alkenes with a view of introducing an additional double bond; bromomethylsul- fonyl bromide (71), readily accessible upon bromination of 1,3,5- trithiane, is used here as a C1 synthon.88, 89 BrSO2CH2Br (71), hn R R O ButOK S 7H+ Br O R SO2 High regiospecificity of diene formation is observed when cyclic alkenes are used as the starting compounds.Thus the only products obtained from cycloalkenes are 3-methylenecycloal- kenes, 1-methylcycloalkenes react to give 1,2-dimethylenecycloal- kanes, and methylenecycloalkanes afford 1-vinylcycloalkenes.90 In all cases, the yields of dienes are moderate. Functionally substituted alkenes have been rarely used in these reactions. tert-Butyl methoxycarbonylmethyl sulfoxide (72) proved to be a convenient synthon for elongating the skeleton of allylic alcohols 73 by a C2 moiety. The transformation takes place via intermediates 74 and 75 to afford alka-2,4-dienoic acids 76.91 Its final step is the Ramberg ± BaÈ cklund reaction, the key inter- O ButSCH2CO2Me (72), NBS, CHCl3 OH R1 R2 73 O R1 S CO2Me O R2 75 NBS is N-bromosuccinimide.A A Vasil'ev, E P Serebryakov Scheme 3 SO2Ph SO2Ph mCPBA SePh R1 R1 70 O O NO2 PhSO2 PhSO2 KMnO4 ButOK R2 R2 R1 R2 R1 SO2Ph SO2Ph H2SO4, MeOH CO2Me CN CO2Me MeONa R1 R1 67 H2SO4, SOCl2, R3NH2 SO2Ph ButOK CONHR3 CONHR3 R1 mediate of which is generated by deprotonating sulfone 76 and submitting the carbanion to a-chlorination with CCl4 (not shown). 8. Cheletropic elimination Br R O ButOK, THF S Br O The ability of b-sulfolenes to undergo reversible fragmentation into 1,3-dienes and SO2 (see a monograph 92) underlies a versatile synthesis of dienes with somewhat sophisticated structure starting with successive a,a'-bis-alkylation of parental sulfolene 77.93, 94 Depending on the procedure used to decompose the alkylated sulfolene 78, one obtains either E,E- or E,Z-dienes.The parental sulfolene 77 acts in this method as a synthetic equivalent of 1,4- dianion of buta-1,3-diene. This strategy was employed in the synthesis of some pheromones of Lepidoptera that contain termi- nal or internal E,E-configured diene substructures. O O R S Br 7 1. LHMDS, R1Hal 2. LHMDS, R2Hal (THF, HMPA) 7Br7 R2 R1 S S O O O O R 77 78 7SO2 (*50%) R2 K2CO3, EtOH, 125 8C R1 (E,E)-55 R1 R2 LiAlH4, Et2O, 35 8C (Z,E)-55 R1 R2 (E,Z)-55 LHMDS is the lithium derivative of hexamethyldisilazane, HMPA is hexamethylphosphoramide. OSCH2CO2Me O R1 (43% ± 92%) R2 74 R1 KOH, CCl4 CO2H R2 (63% ±82%) 76 For the alkylation of sulfolene-derived carbanions, it was recommended to use manganese as the counter-ion.The use of manganese counter-ion enhances the thermal stability of carban- ions, so that even allylic halides with additional electrophilic groups (e.g., methyl 4-bromocrotonate) 95 can be used as alkylat- ing reagents. Another way to overcome the lability of sulfolene- derived anions is the use of its synthetic equivalent, tricyclic compound 79 (this sulfolane is prepared by cycloaddition of cyclopentadiene to 3-sulfolene). Upon vapour-phase thermolysis (650 8C), the alkylated derivatives thus formed (80) undergo two- step retro-fragmentation to give the E,E-configured internal dienes 55 of 91%± 95% purity.96Synthetic methodologies for carbo-substituted conjugated dienes 1.BuLi, R1Hal O 2. BuLi, R2Hal S O 79 R2 R1 55 The latter approach proved useful also in the synthesis of insect pheromones.97, 98 On treatment with NEt3, 3-bromomethyl-3-sulfolene 81 is converted into 4-methylene-2-sulfolene 82.99 The latter, as well as its alkylated derivatives 83, can readily enter into the Michael reaction with such nucleophiles as amines, thiols and nitroalkane- derived anions, which leads to modified b-sulfolenes 84 and 85 in up to 90% yields. Subsequent thermolysis of 84 or 85 gives rise to the branched 1,3-dienes with functionalised substituents. BrK2CO3, Et3N S S O O O 81 82 Nu Nu D S O O 84 Nu 1. LHMDS 2. RI R S O O 83 Other approaches to conjugated dienes from variously sub- stituted and functionalised b-sulfolenes have been also reported.In one of them, readily accessible a-acylthiophenes 86 were used as the starting compounds for obtaining a-branched conjugated dienones 87 in three or four synthetic operations.100 R1 R2 1. Na, NH3, EtOH ±Et2O 2. R3X (44% ± 82%) S 86 O R3 R1 R2 (36% ± 56%) S O O O Another group of sulfolenes suitable for the synthesis of conjugated dienes can be obtained from vinylic phosphonates 88 and mercaptoacetaldehyde 89; the synthetic protocol comprises two sequential Michael reactions followed by intramolecular Horner ± Wadsworth ± Emmons (HWE) reaction and oxida- tion.101 It is recommendable to perform the thermolysis of sulfolenes 90 directly in the presence of electron-deficient dien- ophiles.This technique was employed in an elegant synthesis of compound 91, an alkaloid of eleocanine group, from sulfolene 92.102 R2 O HSCH2CHO (89) R1 R1 P(O)(OAlk)2 88 R1 O 650 8C S O , 7SO2 7 R2 80 O Nu Nu Nu D R S R O O 85 R3 MCPBA R1 R2 S O R1 D R2 O R3 87 O O R2 R2 [O] O R1 S S O 90 O O D 6 steps 90 HN SO2 OAc 92 O O N 90: R1=Prn, R2=(CH2)2CH=CH2. b-Hydroxy acids are prone to decompose with the formation of alkenes and concomitant ejection of carboxy and hydroxy groups. If such substrates (e.g., b-hydroxy acid 93) have already a double bond, their fragmentation will give dienes. This process can be carried out by treating acids 93 either with N,N-dimethyl- formamide acetals 103 or with acetic anhydride.104 Decomposition of b-acetoxy acids 94 is catalysed by palladium complexes.R3 7 O R2 CO27 R2 R1 R1 95 93 NMe2 Me2NCH(OMe)2 O O R2 R3=H 7CO2, 7Me2NCHO O R1 OAc Ac2O Pd(PPh3)4, Et3N R3 R1 R2=H 7CO2,7AcOH R1 CO2H 94 Although intermediate acetoxy acids 94 are obtained as mixtures of stereoisomers, the newly formed double bond acquires the E-configuration, while the double bond of the starting enal 95 (R2=H) retains its configuration in the resulting diene. These transformations were employed in stereocontrolled synthesis of the pheromones 56 and 96.104 OAc THPO(CH2)6 CO2H OAc THPO(CH2)8 CO2H HO(CH2)8 96 THP is tetrahydropyranyl.Elimination of carboxy and hydroxy groups from 4-hydroxy- cyclohex-2-enecarboxylic acids 97 was used in the synthesis of cyclohexadienes 98.105 R1 CO2H R2 Me2NCH(OCH2But)2 (50%± 84%) R3 OH R4 97 741 O O N O N 91 OHR3 CO2H R2 R1 (59% ± 85%) R3 HO(CH2)6 56 R1 R2 R3 R4 98742 III. Olefination of carbonyl compounds Olefination of a carbonyl compound means the replacement of its oxo group by the alkylidene grouping (Scheme 4). Scheme 4 R3 R1 7XOY R4 R2 R1 R3 R2 X + O R4 X Y R3R4 OY R1R2 The process takes place in stepwise manner and includes the addition of olefinating agent to the C=O group followed by the elimination of an X7O7Y species from the primary adduct.In some cases elimination occurs spontaneously, in others the adduct has to be isolated first, after which the second step is carried out separately. The most widely used olefination reactions are shown in Table 1. For the synthesis of conjugated dienes, it is necessary that one of the reactants should already possess a double bond. In general, two cases are possible: the double bond is present in the carbonyl compound (a), or the double bond belongs to the olefinating agent (b) (Scheme 5). Each of the cases displays its particular features depending on the chosen variant of olefination (see the corre- sponding subsections below). A detailed enough compendium on olefination of carbonyl compounds can be found in Larock's monograph,10 but some of the methods cited there were used only to obtain monounsaturated compounds.Scheme 5 R4 R3 R1 a R2 R5 X + O R4 R3 R6 Y R1 R5 X R3 R2 R6 R1 R4 b Y O + R6 R2 R5 Table 1. Main types of carbonyl olefination reactions. Olefination variant The reagent R1R2C(X)(Y) X Y R1and R2 H Aldol condensation Ha R1=CO2R3, COR3, CN H The Knoevenagel reaction Ha R1=R2=CO2R3, CN The Wittig reaction =PPh3 R1 and R2 can contain various functional groups Ha P(O)(OR)2 R1=CO2R3, COR3, CN The Horner ±Wadsworth ± Emmons reaction Ha The Peterson reaction SiR3 R1=CO2R3, CN The Julia ± Baudin approach Ha SO2Ar R1 and R2 contain no reactive functional groups SeR b the same The Krief reaction SeR b a Upon activation, this hydrogen atom is replaced by a positively charged counter-ion. b One of these RSe groups is exchanged for lithium.A A Vasil'ev, E P Serebryakov 1. Condensation of CH acids with carbonyl compounds Different variants of direct condensation of a,b-enals with active methylene compounds to give conjugated dienes are described now in all high-school textbooks. Several examples of preparation of dienoic acids are given in a review dedicated to the Knoevenagel reaction.106 In its classical version, the highest yields (25% ± 40%) are attained by heating an enal with malonic acid in pyri- dine 107 ± 110 (see, inter alia, the preparation of octa-2,4-dienoic acid 99 from hex-2-enal 100109). Instead of enal, one can use the corresponding acetal, and perform the reaction in the presence of piperidine in pyridine at 100 ± 140 8C.111 However, the yield of the dienoic acid remains about 20%.Py CO2H CHO + CH2(CO2H)2(24%) 99 100 The DMAP-catalysed condensation of enals 101 with alkyl hydrogen malonates affords dienoates 103 in good yields.112 However, noticeable amounts of deconjugated b,g,d,e-dienoates are formed when b-substituted acroleins (R1 6à H) are used in the reaction. CO2R3 R2 R2 DMAP, Py, 60 8C, 18 ± 24 h + R1 R1 CHO CO2R3 CO2H 102 103 101 When the process is carried out at 140 8C in xylene in the presence of piperidine or its acetate, deconjugated E isomers of alk-3-enoic acids are formed as the main products.113 The exten- sion of this procedure to a,b-enals affords partly deconjugated alkyl dienoates 104.114 In total, the yield of isomeric dienes amounts to 60% if DMSO is used as the solvent (100 8C); however, a higher content of the 3Z,5E-isomer in total products is R CO2Et (Z,E)-104 NH, D CO2Et R CO2Et R CHO + (460%) CO2H (E,E)-104 R CO2Et (< 5%) Comments the reaction is sensitive to the nature of the carbonyl substrate the reaction is characteristic of aldehydes the reaction is sensitive to the nature of the carbonyl substrate the reaction is characteristic of aldehydes and less predictable with ketones the reaction is applicable to nearly all types of aldehydes and ketones if they have no other reactive functional groups the primary adduct undergoes elimination in a separate step the reaction is applicable to nearly all types of aldehydes and ketones if they have no other reactive functional groupsSynthetic methodologies for carbo-substituted conjugated dienes attained when xylene (140 8C) is used. The unwanted 3E,5E- and 2E,4E-isomers (the content of the latter does not exceed 5%) can be removed upon converting them into inclusion complexes with urea.Since the esters of pentenedioic (glutaconic) acid are the vinylogues of dialkyl malonates, they readily react with unsatu- rated aldehydes to give substituted conjugated dienes (see a review 115). Thus the esters of 3-methylglutaconic acid (e.g., 105) were successfully employed in the synthesis of methoprene (106) [R=(MeO)Me2C(CH2)3CH(Me)CH2] as the equivalents of iso- prenoid C5 block.116 The primarily isolated products of conden- sation of aldehydes with the glutaconate 105 are configurationally pure 4-carboxy-3-methylalka-(2Z,4E)-dienoic acids, which are subsequently converted into (E,E)-3-methylalka-2,4-dienoic acids 106 in several steps.NaOH, MeOH CO2Me+RCHO MeO2C (86.5%) 105 Cat. CO2HCO2H 2,5-lutidine, Cu(OAc)2, 80 ± 120 8C (84%) R R CO2H CO2H R 106 The `classical' aldol condensation of a,b-enals with aldehydes or ketones (see a review 117) is also used for obtaining conjugated dienes (e.g., for preparing hepta-3,5-dien-2-one 107 from acetone and crotonaldehyde 118). O Ba(OH)2 CHO + O (52%) 107 In traditional variants, hydroxides 119 or sodium sulfite 120, 121 are usually used as the bases.The yields of the products, particularly of dienals are moderate, and (quite often) even low. The yields of dienones vary over wide limits and strongly depend on the reaction conditions and on the particular nature of the reaction partners. The industrial large-scale syntheses of pseu- doionone and its analogues from citral and acetone (or butan-2- one) were the objects of special research and development studies, and their conditions were duly optimised.122 Lately, a number of advanced procedures for directed aldol condensation have been developed. Thus high regioselectivity can be attained by using alkyl acetates or methyl ketones as the active methylene component (108), and lithium diisopropylamide as the base 123. Upon treatment with LDA, ketones of the type 108 with two activated positions undergo deprotonation almost entirely at the methyl group due to the kinetic control.124 O O 1.LDA, 778 8C 2. R2CH CHCHO 3. Me3SiCl, D R2 R1 R1 (65% ± 98%) 108 Another regioselective method of deprotonation consists in using dimethylhydrazones of asymmetric aliphatic ketones like 109.125 1. BuLi, THF,778 8C 2. MeCH CHCHO N (98%) Me2N109 743 1. NaIO4 2. MsCl, Et3N (83%) OH N O Me2N Recently 126 a conceptually new mode of aldol condensation has been developed. It consists in treating a mixture of two aldehydes, one of which is an a,b-enal, with LDA at 778 8C in the presence of aluminium tris(2,6-diphenylphenoxide) 110.Under these conditions, a,b-enals behave exclusively as the vinylogues of the active methylene components. (2,6-Ph2C6H3O)3Al (110), LDA, THF, 778 8C CHO+PhCHO (99%) OH CHO CHO Ph Ph The choice of enals in this method is not confined to croton- aldehyde-like structures; higher polyenals Me(CH=CH)nCHO and conjugated enones (that react by their vinylogous anionic termini) also are suitable. Thus in 4-methylacetophenone it is the remote methyl, not the COMe group, that is activated under above conditions! It was speculated that on mixing, the reactants form a specifically shaped conglomerate the deprotonation of which at vinylogous positions is the most favoured one. The addition of the resulting carbanion to the aldehyde component is believed to take place inside the conglomerate.Under other conditions, the a,b-unsaturated carbonyl com- pounds like mesityl oxide 111 after deprotonation, add to the carbonyl substrates (electrophiles) with formation of deconju- gated products, e.g., 112 127 Subsequent dehydration of the latter leads to the branched conjugated diene 113. OH (CH2O)x, DBU, 100 8C, 10 min KOH, 60 8C, 30 min (55%) O O O 113 111 112 An original regio- and stereocontrolled approach to a-branched (2Z,4E)-dienoic acid relies on the use of a vinylogous synthon 114 with a methylthio group.128 SMe MeS 1. Raney Ni W-2 2. ButOK 1. LDA 2. RCHO CO2H (79%) (72%) R O O R CO2Et 114 Yet another way of securing high regioselectivity in the condensation of two different carbonyl compounds is to replace conventional active methylene component by its synthetic equiv- alent, vinyl ether (e.g., 115).Moreover, not only aldehydes but acetals 116 can be used as carbonyl components, which opera- tionally is more convenient. Characteristic features of this process have been thoroughly studied and discussed in a review 129 and a monograph.130 The main shortcomings of this method are its tedious procedures and moderate yields. Lewis acid CH(OEt)2 + R OEt 116 115 OEt OEt CHO R OEt R744 Similar results were obtained by employing ethoxyacetylene 117 instead of the active methylene component.131 ± 135 Using this agent, the reaction can be performed not only with aldehydes but also with ketones (118, R4 6à H).In this case, the overall reaction product is a mixture of stereoisomers. The process can be implemented either in the presence of BF3 . OEt2, or by adding the organometallic derivative of 117 to the carbonyl component, and then by subjecting the adduct to acid hydrolysis. R4 CO2Et O R1 R4 R1 + HC COEt(40% ± 84%) R3 R2 R3 R2 117 118 Sorbic acid (119, fungistatic agent for food) is manufactured industrially on a multitonne scale by a process where ketene (120) is used as the C2 component.136, 137 In the laboratory practice, it can be conveniently replaced by trimetylsilyl ketene 121.138 CH2 C O (120) or Me3SiCH C O (121) CHO CO2H Me Me 119 The Diels ± Alder reaction of 1-alkoxybuta-1,3-dienes 122 with strongly electrophilic aldehydes leads to dihydropyran deriv- atives 123, which can be subsequently transformed into dienals 124.139, 140 In this protocol, the compound 122 acts as a synthetic equivalent of crotonaldehyde.R2 R2 R3 R1 R3 R1 H3O+ R4CHO, D (22%± 80%) 122 OEt OEt R4 O 123 R3 R1 R4 CHO R2 124 R1, R2, R3=H, Me; R4=H, CCl3, CO2Et. 2. The Wittig reaction and related processes The Wittig reaction (in its classical or newer variants) is one of the most popular methods of olefin synthesis. Recently, its mecha- nistic and preparative aspects have been studied in much detail, and discussed in reviews.141 ± 143 In general (Scheme 6), for the construction of a conjugated diene system.it is necessary that one of the reactants, either the carbonyl compound 125 (pathway a) or the phosphorus-bearing compound 126 (pathway b) should already have a double bond in its structure. From the standpoint of operational convenience, pathway a is technically simpler and thus preferable, provided that the olefinic carbonyl compound is readily accessible. Scheme 6 R3 R3 R2 QnP Z R3 R2 O R1 a b 125 126 Z R1 R2 QnP Z O R1 QnP=Ph3P+, Ph2P(O), (AlkO)2P(O) etc.; Z=Alk, Alkenyl, Ar, Het, CO2R4, CN, CONR42 , C(O)R4 etc. A A Vasil'ev, E P Serebryakov a. Phosphoranes As a rule, the use of phosphoranes like Ph3P=CHCO2Et or Ph3P=CHCN in reactions with a,b-unsaturated aldehydes is unproblematic; this is confirmed by abundant literature records.Thus a recent work describes a stereoselective in situ synthesis of (2E,4Z)- and (E,E)-5-bromopenta-1,3-dienoates (127), where the oxidation of (Z)- and (E )-3-bromoallyl alcohols was carried out in the presence of Ph3P=CHCO2Et.144 MnO2 Ph3P CHCO2Et CHO Br Br CH2OH (84%) (Z)-128 Br CO2Et (E,Z )-127 (purity 80%) MnO2 Ph3P CHCO2Et Br Br CHO (78%) CH2OH (E)-128 Br CO2Et (E,E )-127 (purity 88%) The starting a,b-enals can be obtained by elongating saturated aldehydes 129 by a C2 fragment using 2-oxoethylidene phosphor- ane 130. A stepwise process for preparing the Z,E-configured dienes 55 from alkanals 129 consisting of two sequential Wittig olefinations has been developed.145 Ph3P CHCHO (130) CHO Ph3PCH CHR2 R1 R1CHO 129 R2 R1 (Z,E)-55 The E,E-configured 5,6-trans-epoxyalka-2,4-dienals 131 are obtained in moderate yields by two-step elongation of trans-2,3- epoxyalkanals 132 with phosporane 130 first under mild, and afterwards under somewhat more drastic conditions.146 ± 148 Alter- natively, dienal 131 can be prepared in one step using the vinyl- ogous phosphorane Ph3P=CH7CH=CHCHO.149 CHO MeO2C(CH2)8 1.130, PhH, 80 8C 2. 130, PhMe, 110 8C O 132 CHO MeO2C(CH2)8 131 O For the first stage of elongation, the phosphorane 130 can be replaced by more accessible alkoxycarbonyl-substituted phos- phoranes Ph3P=C(CO2Et)R (133, R=H, Alk); after that, the CO2Et group introduced is transformed into CHO.150 The (E,E)- dienoic acid 134 was synthesised in this way with high stereo- selectivity.1. 133 (R=Me) 2. DIBAH 3. MnO2 1. 133 2. LiOH Bu OHC Bu OHC (54%) Bu HO2C 134 DIBAH is diisobutylaluminium hydride. Aldol hemiacetals can be successfully used in the Wittig olefination instead of aldehydes. Thus starting from 2-deoxy-D- ribose (a naturally occuring aldol hemiacetal) or its derivativesSynthetic methodologies for carbo-substituted conjugated dienes 135, dienes 136 and 137 have been synthesised.151, 152 Depending on the procedure employed for eliminating the 5-OH group from the intermediate dihydroxy alkenoate, the configuration of the second double bond in the target dienes can be Z or E. OH ZO O OHPh3P CHCO2Et CO2Et ZO OH HO 135 4 steps TrO CO2Et 136 OH 4 steps CO2Et R 137 OH Unlike aldehydes, conjugated ketones enter reluctantly into the Wittig reactions.The a,b-enones that are not branched at the b-position sometimes enter into the Michael reaction with phos- phonates of the type (RO)2P(O)CH2CO2Alk to give the products of 1,4-addition.153 It is noteworthy that a-diketones 138 can react with two molecules of the above phosphonates and give conju- gated dienes via the intermediate a,b-enone 139.154, 155 For the synthesis of diesters 140, it is necessary to adopt a two-step procedure,154 while the preparation of dinitriles 141 can be performed in one step due to the higher reactivity of the corre- sponding intermediate.155 CO2Et CO2Et Ph3P CHCO2Et (EtO)2P(O)CH2CO2Et, NaH (52%) (66%) O O 139 CO2Et 140 CN O 138 2Ph3P CHCN (67%) CN 141 Cyclic a-diketones such as cyclobutane-1,2-dione and cyclo- hexane-1,2-dione also react in this way to give semicyclic dienes.Pathway b of Scheme 6 looks equally attractive. The use of allylic phosphoranes 126 (Z=Alk, Ar) is justified, particularly when the precursor of the phosphonium salt, the corresponding allylic halide, is readily accessible.156 Allylic halides of higher molecular complexity are rather difficult both to prepare and, particularly, to purify. As a consequence, roundabout approaches to allylic phosphoranes have been developed. Thus allylic phos- phoranes 142 are formed in situ from the Z3-allylic palladium complexes 143 (prepared by heating allylic alcohols 144 or 145 in dioxane in the presence of Pd(acac)2 and triphenylphos- phine.157, 158 The subsequent reaction of phosphoranes 142 with aldehydes affords conjugated dienes.This way of preparing conjugated dienes is lengthy enough and complicated by side reactions. R R OH 144 Pd(acac)2 Ph3P R Ph3P or Pd(acac) R HO 142 143 OH 145 R R0 142+R0CHO (472%) Another synthesis of conjugated dienes that involves indirect approach to allylic phosphoranes relies on the ability of (buta-1,3- 745 dienyl)phosphonium compounds 146 to react with C-nucleophiles such as the dianions of b-dicarbonyl compounds or organo- cuprates.159, 160 Br7 + Br7 + Nu LDA or NaH Q3P Q3P Br 146 R2 R1R2C O Q3P Nu R1 Nu (493%) On treatment with strong bases, allyltriphenylphosphorane 147 gives metallated derivative 148, which reacts with one mole- cule of a carbonyl compound as a C-nucleophile to afford the modified allylic phosphorane 149.Then the Wittig reaction of 149 with the second molecule of aldehyde or ketone brings forth dienic alcohols 150 in moderate yields.161 7 BuLi, TMEDA + + BuLi, 20 8C Ph3P Ph3P 778 8C Br7 147 Li+ R1 7 R1R2C O R1R2C O + 7 R2 Ph3P Ph3P (39% ±58%) OH 148 149 R1 R1 R2 R2 OH 150 TMEDA is N,N,N0,N0-tetramethylethylenediamine. b. Phosphonates It is profitable to replace phosphoranes by the corresponding phosphonates of the type (EtO)2P(O)CH2Z. This variant of olefination is known as the Horner ± Wadsworth ± Emmons reac- tion.Thus the olefination of tiglic aldehyde 151 with triethyl phosphonoacetate 152 affords ethyl dienoate 153 in 90% yield.162, 163 The newly formed double bond has almost pure E configuration. CHO NaH± PhH or ButOK±THF +(EtO)2P(O)CH2CO2Et (90%) 152 151 CO2Et 153 A simple procedure for implementing such a process has been proposed. It consists in mixing together and stirring an aldehyde, phosphonate 153 and concentrated aqueous K2CO3. The method is applicable to the synthesis of alkyl dienoates or alkadienones and provides satisfactory yields (60% ± 70%).164 Time and again, it was used in the synthesis of natural products.4, 165, 166 In the case of phosphonates with phenoxy groups (154), the stereoselectivity of olefination is reversed (with respect to the case of 152) i.e, theZ-isomers of alk-2-enoates are produced.167 For the reactions involving phosphonates 154, it was recommended to use NaH, Triton B or the DBU± NaI system as the base.The reaction of hex-2-enal (100) with 154 afforded ethyl (2Z,4E)-octa-2,4- dienoate 155 with 97% isomeric purity.168 1. Base 2. (E)-PrnCH=CHCHO (100) (ArO)2P(O)CH2CO2Et (*100%) 154 CO2Et (2Z,4E )-155A A Vasil'ev, E P Serebryakov 746 The reaction of allylic phosphonates of the type 156 with saturated aldehydes can be successfully applied to the synthesis of dienes.169 O R3 R3 1. NaH, BuLi or ButOK 2. R4CHO R1 R1 P(OEt)2 ties 184 provides only 27% yield.Recently, a method for the synthesis of conjugated dienes has been proposed according to which a mixture of the initial phosphonocrotonate and aldehyde is refluxed in THF in the presence of LiOH and 4 A molecular sieves. The yield of dienes is 86%.185 Cyclohexanones can be used successfully in place of aldehydes in this reaction. R4 (*90%) R2 R2 156 CO2R1 (AlkO)2P(O) R1, R2, R3=H, Alk. 160 R2CHO R2 CO2R1 161 R2C CCHO R2 CO2R1 162 For allylic phosphonates 156 with additional functional groups (R1 or R2=CO2Alk, CONR2, RCO, or CN) the HWE reaction is one of the most studied because it is widely used in the synthesis of polyenes. Such phosphonates are usually prepared by employing the Arbuzov or Michaelis ± Becker reactions (see a review 142). Recently, an interesting method for the synthesis of g-phosphono-a,b-enones from allylic phosponates was re- ported.170 Its protocol involves silylation, alkylation (if necessary) and finally acylation.CHO Olefination of acetylenic aldehydes by the phosphonate 160 in the presence of NaNH2 in THF gives rise to the corresponding dienynoates 162 in a yield not higher than 36%.4, 186 Apart from esters, dienals 163 have also been synthesised using this route.187 O RCHO +(EtO)2P CHO KHMDS, THFR (48% ± 62%) 163 Dienic aldehydes 164 and their acetals 165 can be prepared using synthons 166 and 167 with the protected aldehyde group.188, 189 R3 1. LDA or KHMDS 2. H2O R1R2C O + (EtO)2P The experience accumulated since the mid-sixties suggests that the HWE reactions are sensitive to the nature of the reactants and bases used and to conditions under which they are implemented.Thus for the preparation of alkyl dienoates of the type 157, it was recommended to use such bases as MeONa in DMF,171 NaNH2 in THF,171 NaH in DMF,172 and, what is particularly convenient, the systems of phase transfer catalysis: solid KOH± benzene ± 18-crown-6 (Cat.) 173 ± 176 or solid KOH± benzene ± Alk4N+Br7 (Cat.).177, 178 In the latter case, 3-methylalka-2,4-dienoates 157 are formed with 90%± 96% E,E-selectivity 143, 178 By contrast, when the phosphonate grouping contains the 1,3,2-phospholane frag- ment, the reaction results in 2Z,4E-isomers (*70%± 85% selec- tivity).179, 180 N (39% ±91%) R2 R2 166 R1 R2 CO2R3 R1CHO+(AlkO)2P(O) CO2R3 CHO Me 157 Me R1 O R3 O R3 164O ButOK, THF R1R2C O +(EtO)2P O (51% ± 86%) 167 O R2 R3 O R1 165 Bis(b,b,b-trifluoroethyl) esters of aliphatic phosphonic acids (Still ± Gennari phosphonates) react with aldehydes to give Z-alkenes.The reactions of a,b-enals with bis(b,b,b-trifluoro- ethyl) 3-cyano-2-methylprop-2-enylphosphonate (158) afford the 2Z,4E-isomers of 2-methylalka-2,4-dienonitriles with high selec- tivity.181 The possibility of using the phosphonate 158 for the synthesis of 1,3-dienes with Z-configured newly formed double bond is indicated by the reaction of (2Z,4Z)-b-ionylideneacetal- dehyde (159) with this phosphonate.182O CN (158), 1.(CF3CH2O)2P NTol Pd2(dba)3 .CHCl3 CHO + Dienyl trifluoromethyl ketones 168 are formed from dien- amines 169, tautomers of Schiff's bases 170.190 OP(OEt)2 base 2. DIBAH 3. H2O (51%) I F3C O NTol 159 P(OEt)2 CHO F3C 170 O NHTol O 1. BuLi 2. ArCHO 3. HCl (aq.) P(OEt)2 Ar F3C (41% ± 71%) F3C 169 168 In the mixture of four stereoisomers of retinal formed in this reaction, the proportion of products with Z-configuration of the newly formed double bond amounts to 72%. When non-branched phosphonocrotonates 160 are used, the yields of dienes 161 decrease (to 59% with the use of NaNH2 in THF4, 166); under conditions of phase transfer catalysis, the dienes 161 are not formed at all. However, with the use of lithium derivative of hexamethyldisalazane,183 the dienes 161 are pro- duced in 84% yield, whereas the use of LDA having close proper- c.Allylic phosphines and phosphine oxides Olefination of carbonyl compounds can also be performed by means of allylic phosphines 171 and allyl phosphine oxides 172;191 the products 38 thus formed differ in the stereochemistry of the newly formed double bond.Synthetic methodologies for carbo-substituted conjugated dienes 1. BuLi, Ti(OPri)4 2. RCHO 3. MeI R PPh2 (76% ±89%) (Z,E)-38 171O 1. BuLi/THF/HMPA 2. RCHO PPh2 R (42% ±88%) 172 (E,E)-38 Allylic phosphine oxides with more complex structures used for the synthesis of functionally substituted conjugated dienes can be prepared from simpler phosphine oxides 173 or 174.192 To this end, compounds 173 and 174 are made to react with aldehydes and ketones, and the reaction products 175 and 176 are subjected to dehydration.Acidic dehydration of secondary b-hydroxyphos- phine oxides 175 is accompanied by the migration of the phos- phorus-containing group, while dehydration of tertiary b-hydroxyphosphine oxides 176 follows the standard pathway. In both cases, tertiary carbocation 177 is formed, which is trans- formed, under thermodynamic control, into thermodynamically more favourable alkene 178. Thus the same product 179 can be prepared in different ways, which makes the phosphine-oxide route more versatile. R2 R1 R2 R1 1. BuLi 2. R3CHO R3 Ph2P Ph2P 173 O O OH 175 H+ R1 R2 HO O O 1.BuLi 2. R1C(O)CH2R2 R3 Ph2P Ph2P 174 R3 176 R2 R1 R2 R1 R1 R2 + R4CHO Ph2P R3 Ph2P R3 R3 R4 178 179 177 O O The synthesis of conjugated diene 180 containing an ethyl acetoacetate fragment as a substituent has been reported.193 This compound was prepared using phosphine oxide 181. 1. NaH, THF, 0 8C 2. BuLi, 0 8C 3. OHC O O O PPh2 (75%) EtO 181 O O EtO 180 d. Related methods In conclusion, we shall consider methods of olefination of carbonyl compounds making use of `non-phosphorus' equivalents of the Wittig reagents based on arsenic 182,194, 195 antimony 183 196 and tellurium 184.197, 198 It should be noted that in the case of complex syntheses, the use of arsonium derivatives has substantial advantages.The advantages and shortcomings involved in the use of antimony and tellurium derivatives have not been adequately elucidated. 747 O Br7 K2CO3 + CHO+ MeO2C Ph3As CHO (80% ± 85%) 182 O CHO MeO2C + 20 to 50 8C Br7 O CHO +Bu3Sb Alk 183 O O + +Bu3SbBr2+Bu3SbO Alk (81% ± 88%) CHO K2CO3, MeCN Ar +Bu2Te+CHXY Br7 184 X Ar +Bu2TeO (55% ± 99%) Y X=CN, CO2Et; Y=H, Ph. 3. Miscellaneous methods of olefination of carbonyl compounds a. The Peterson reaction The Peterson reaction consists in elimination of vicinal silyl and hydroxy groups, resulting in the formation of a double bond (see Section II). Here, we consider the version used to olefinate a,b- unsaturated carbonyl compounds. In this technique, the starting b-hydroxy silanes 185 are formed in situ upon the addition of silicon-containing carbanion reagents 186 to the carbonyl com- pounds 125.46, 47 R3 R2 R3 OH R2 R1 O Li+ 7 125 CO2R4 CO2R4 Me3Si R1 (78% ± 95%) 185 186 SiMe3 R3 R2 CO2R4 R1 An advantage of this approach is the possibility of using not only a,b-unsaturated aldehydes but also ketones including rela- tively non-reactive chalcone.48 An organosilicon reagent should contain a group exhibiting a 7M-effect.Apart from esters 186, the acetylene derivative Me3SiCH2C:CSiMe3 and the nitrile 187 49 have also been involved in this reaction. Deprotonation is carried out by amides such as LDA; the yields are high in all cases. CHO Li+ 7 CN CN Ph3Si (79% ± 85%) 187 In recent years, a convenient C2 olefinating synthon 188 has been devised; it contains two geminal silyl groups and is able to react with aldehydes 189 in the presence of Lewis acids under very mild conditions.199 SiMe3 1.ZnBr2 ,THF, 20 8C 2. ZnCl2,H2O, Et2O NBut CHO + R Me3Si (63% ± 92%) 188 189 CHO R A similar vinylogous C4 synthon with two silyl groups has not been prepared; however, C4 olefination can be accomplished under other conditions using monosilyl reagent 190.200, 201 The748 reactions of carbonyl compounds and acetals with cyclic g-silyl- a,b-enones 191 giving rise to dienones 192 have been described.202 NBut RCHO + Me3Si 1. CsF, DMSO, 20 8C 2. 20 to 100 8C 3. ZnCl2, H2O (53% ± 94%) 190 CHO R O O R1R2C=O/Me3SiI or R1R2C(OMe)2 /SnCl4 R1 (23% ±81%) SiMe3 R2 192 191 b.The Baudin ± Julia method The Julia method consists in condensation of aryl sulfones with aldehydes followed by reductive elimination of the sulfone group from unsaturated sulfone of the type 64 (see Scheme 2),78, 79 for example, on treatment with LiAlH4 . The involvement of unsatu- rated sulfones 193 in this process should furnish dienes. It has been proposed 203 to use BuMgCl for the reduction of the intermediate Z-olefinic sulfones 194 in the reaction catalysed by Pd(acac)2 or Ni(acac)2. The isomeric purity of the products 55 was 93%± 97%; however, the yields were moderate. SO2Ph SO2Ph 1. BuLi 2. R2CHO 3. Ac2O NaOH R2 R1 R1 OAc 193 SO2Ph BuMgCl, [Pd] or [Ni] R1 R1 (40% ± 50%) R2 R2 194 (E,Z)-55 In another study,204 olefination of a,b-enals with saturated sulfone of the type 195 and olefination of saturated aldehydes with allylic sulfones 196 (Ar=Ph) have been studied in substantial detail with verification of numerous factors.Elimination of vicinal sulfonyl and hydroxy groups was induced by sodium amalgam. The yields of dienes were moderate, despite the efforts made by the authors. 1. Base R2 SO2Ph R2 R3 2.O R3 Ar=Ph R1 SO2Ar R1 195 OH Na/Hg R2 R2 SO2Ar 1. Base 2. R3CHO R3 R1 R1 SO2Ar 196, 197 OH R2 R3 R1 N (197). Ar=Ph (195, 196), S In a perfected method, the phenyl group in the sulfone 196 was replaced by various heterocyclic groups, which resulted in higher product yields, the best results being attained for the 2-benzothia- zole substituent (compound 197).205 ± 207 This is due to the fact A A Vasil'ev, E P Serebryakov that heteroatoms in the aryl group change the final steps of the olefination mechanism; this influences the product yield, which can reach 86%.Both reactions � unsaturated sulfone + saturated aldehyde and vice versa�are used for the synthesis of dienes. Usually, this gives a mixture of E,E- and E,Z-isomers, their ratio being dictated by the pattern of substitution at the double bond in the reactants. c. The Krief reaction and other olefination methods The ability of selenoacetals 198 to exchange one selenium-con- taining group for lithium has underlain the new Krief olefination method.208 ± 210 The reaction of the carbanion formed (stabilised by the remaining selenium atom) with aldehydes or ketones yields b-hydroxy alkyl selenides 199, which readily form olefins on the subsequent treatment with acidic reagents.In the case of unsatu- rated aldehydes, the method results in conjugated dienes.211, 212 1. BuLi 2. R3CH=C(R4)CHO R1CH(SeR2)2 198 SeR2 R4 R4 MsCl or SOCl2 , Et3N, CH2Cl2, 20 8C R3 R3 R1 R1 (90%) OH 199 Another variant of organoselenium synthesis of dienes, namely, the reaction of a saturated aldehyde and an unsaturated selenium component, is also possible. Allyl selenides 200, the CH acidity of which is high enough for hydrogen to be eliminated on treatment with LDA, are possible precursors of this type of selenium component.The resulting carbanion, however, exhibits ambident properties and furnishes mixtures of isomeric products in reactions with electrophiles.213 To ensure the required regiodir- ectivity, the reagent is treated with Lewis acids; the ate-complex thus formed reacts with an aldehyde to give the `normal' product 201. OH TsOH (CH2)8OAc 1. LDA 2. Et3Al 3. AcO(CH2)8CHO (70%) (77%) SePh SePh 200 (CH2)8OAc 201 The titanium carbene complex Cp2Ti=CH2 (the Tebbe reagent) is an efficient methylenating agent.214 It reacts with unsaturated carbonyl compounds giving rise to conjugated dienes including those difficult to prepare by other methods, such as 202.215Me2But O OSiMe2But 1.Cp2TiCH2 . AlClMe2 , THF, Py, 20 8C 2. NaOH (aq.) 202 Recently, it has been shown that unsaturated analogues of the Tebbe reagent (203) can be prepared from a,b-enal cyclic thio- acetals 204 on treatment with a titanocene(II) complex with triethyl phosphite.216 The reactions of the carbenes 203 with aldehydes, ketones, and even esters (!) result in the formation of a new double bond. Instead of the thioacetals 204, 1,3-diphenyl- thioalk-1-enes 205 (R1=SPh) 217 can be used in the reaction (they are produced from a,b-enals under conditions of acetalisation with monothiols). In the case of non-branched dienes (R2=R4=H), the proportion of E,E-isomer exceeds 87%.Synthetic methodologies for carbo-substituted conjugated dienes S R2 R2 S R1 Cp2Ti .[P(OEt)3]2 R1 204 uÁ oÈ uÁ R3R4C O TiCp2 (44% ± 88%) 203 R2 SPh R1 205 R2 R3 R1 R4 R3=Alk; R4=H, Alk, Ar, OAlk. Isocyanides 206 can also serve as olefinating agents;218 how- ever, in this case, in addition to alkenes 207, the reaction gives substantial amounts of oxazolines 208. This route was use to convert cinnamaldehyde into 1-phenylbuta-1,3-diene (R1= R2=H, R3=PhCH=CH) in 28% yield. O R2 1. BuLi 2. R2R3C O R2 N R1 R1CH2N C + R3 R3 207 206 208 R1 The reaction of aldehydes with diazo esters leading to a,b- unsaturated esters can also be classified as olefination. In the only known example, namely, in the reaction of cinnamaldehyde with diazo ester, the yield of the corresponding conjugated diene is 92%.219 IV.Organometallic reactions 1. Cross-coupling of vinylic and/or acetylenic components Cross-coupling reactions belong to the most advanced methods for the creation of a C7C bond. The thermodynamic factor, which accounts for the occurrence of these reactions, is the formation of a favourable bond between the electropositive element M of reactant 209 and the electronegative element X of reactant 210 (Scheme 7). The reactions of the general form 209+210 are usually classified according to the sort of the electropositive element in the reactant 209, the nature of which governs the mechanism of the process, although the outcome is formally the same. Scheme 7 C4+M X, C3 C1 C2 M+X C3 C4 210 C1 C2 209 M=H, SiR3, SnR3, BR2, AlR2, ZrR3, Cu; X=Cl, Br, I, OSO2CnF2n+1, SR, TeR.C5+Y Z, C4 C3 C1 C2 Y+Z C5 C4 C1 C2 C3 211 C4 +Y Z . C3 C1 C2 +Z C5 C4 C3 C1 C2 C5 Y 212 In the synthesis of carbo-substituted conjugated dienes, cross- coupling of two vinylic components giving rise to the central single bond is used most often. The replacement of the heteroatom in a preformed diene system 211 or 212 by a carbon-centred fragment is used more rarely. A fundamental feature of cross-coupling reactions is retention of the configuration of the reacting centres, which permits one to prepare, as desired, any of the possible isomers of conjugated dienes (E,E-, E,Z-, Z,E- and Z,Z-isomers). Cross-coupling can also involve components with triple bonds such as 213.First, they react with alkenyl reagents to give conjugated enynes 214. Second, some organometallic reagents of the type 215 or 216 are capable of adding to triple bonds to give 749 Scheme 8 X R1 R1C C 214 R2 M R2 MR2 (215) X R1C CH 213 R1 217 R1 M M E (216) E+ 218 R1 R1 new metal alkenyls, 217 or 218, respectively. All the compounds obtained can be further converted into dienes (Scheme 8). The reactions listed above have been thoroughly studied and discussed in numerous publications;220 ± 227 therefore, in this review, we present only a brief characterisation. There is a fairly detailed list of published data on each type of reaction.10 In early examples, cross-coupling has been carried out with stoichiometric amounts of organometallic reagents.The scope of these reactions is limited as they mainly permit the preparation of symmetrical compounds. This is indicated by homodimerisation of alkenylalanes 219 induced by CuCl 228 or of various alkenyl halides 220 in the presence of stoichiometric amounts of nickel complexes such as Ni(cod)2 (cod is cycloocta-1,5-diene) 229 or Ni(bipy)n prepared in situ.230 R1 CuCl R1 R22 AlH R1 R1C CH (67% ± 73%) AlR22213 219 R3 R3 R3 R2 [Ni(0)] R2 R2 (48% ± 99%) Hal R1 R1 R1 220 Catalytic cross-coupling protocols are much more efficient. Palladium or nickel complexes are the most popular catalysts of these processes. a. The Heck reaction and its modifications The coupling of a-alkenes (209 with M=H, see reviews 220, 221) with alkenyl or aryl halides catalysed by palladium complexes is called the Heck reaction.Coupling of the vinylic components 221 and 222 affords conjugated dienes 223.231 Pd(OAc)2, Ph3P CO2Me Br + CO2Me (30%±75%) 223 222 221 It is advantageous to perform this process in the presence of Ag salts as HHal acceptors;232, 233 in this case, the allylic alcohols 73 can also be introduced in the reaction. R3 R3 I OH R1 R2 Pd(OAc)2, Ag+, DMF OH (82%) R1 R2 73 R3 I R3 OH R1 R2 Apart from alkenyl halides, the Heck reaction can be carried out for alkenyl perfluoroalkyl sulfonates,234, 235 alkenylsila- nols,236 alkenylpentafluorosilicates 237 and alkenyl(aryl)iodonium salts.238 Coupling of alkenyl bromides with potassium but-3- enoate (224), similar to the Heck reaction in the outcome but750 catalysed by the rhodium complex RhCl(PPh3)2, has been described.239 RhCl(PPh3)2 , EtOH, 85 8C, 48 h Br + R (65%) CO2K 224 R CO2H b.The Stille reaction Coupling of organotin compounds with halides catalysed by palladium complexes has been called the Stille cross-coupling (209 with M=SnR3, see reviews 222, 223). In addition to the syn- thesis of biaryls, styrenes and enynes, this reaction is also good for the synthesis of various conjugated dienes 225.240, 241 The reaction proceeds with full retention of the configurations of the reacting centres; the initial compounds may contain diverse substituents and the product yields are usually high.R3 R2 Bu3Sn + I PdCl2(MeCN)2, DMF, 20 8C (62% ± 85%) R4 R1 R2 R3 R1 R4 225 The Stille reaction has been used in the total synthesis of complex natural products, for example, didemnenones 226, bio- logically active compounds generated by marine Tunicata.242 I O O H H OMe OMe SnBu3, (Ph3P)2PdCl2, DMF (72%) O O HO HO 226 Triflates can be involved in this reaction instead of halides;243 this has been used in the synthesis of triene 227, the closest precursor of the C16 juvenile hormone (228).244 Pd(PPh3)4 , CsF, THF CO2Me + Bu3Sn (58%) OTf O CO2Me O 227 CO2Me O 228 1,1-Dibromoalkenes 229 react with organotin compounds, first of all, at the bromine atom located in the trans-position.245 By OTHP I + Me3Sn OH 234 233 OTHP SnMe3 + I A A Vasil'ev, E P Serebryakov varying the order of addition of the alkenyltin and aryltin reagents, either E- or Z-isomers of 3-aryl-1,3-dienes 230 can be obtained as desired. R SnBu3 R ArSnBu3 Br Br [Pd] R Ar Ar (Z)-230Ar SnBu3 ArSnBu3 R R Br 229 Br (E)-230 Alkynyltin derivatives 231 can also enter into the Stille cross- coupling.Conjugated enynes 232 thus formed can be reduced selectively yielding E,Z-dienes.246 R3 R1 R1 [Pd] (sia)2BH I+Bu3SnC CR3 492% R2 R2 232 231 R3 R1 R2 sia is 3-methylbutan-2-yl. Using this strategy, two pheromones 235 and 236 with similar structures but opposite configurations of the conjugated double bonds have been synthesised from the same initial alkyne 233 and alkynol 234 (Scheme 9).246 Direct coupling of tertiary acetylenic alcohols 237 with alkenyl bromides in the presence of stoichiometric amounts of tin hydrides seems to involve the hydrostannylation step, the overall reaction looking like the Stille cross-coupling.247 OH Bu3SnH+[Pd] (Cat.) or Me3SnCl (Cat.)+(7OSiHMe7)n CH+ Br C R1 (51%± 91%) R3 237 R2 HO R3 R1 R2 The reaction in question can also involve alkynylamines R1R2C(NH2)C:CH and other sterically hindered terminal ace- tylenes. The same result is attained when catalytic amounts of (Bu3Sn)2O247 or Me3SnCl,248 and stoichiometric amounts of polymethylhydrosiloxanes are used instead of tin hydrides as hydride-ion donors.This modification of the Stille reaction opens up prospects for extending the scope of its application because this version implies markedly smaller proportions of tin- containing wastes. c. The Suzuki reaction and its modifications Coupling of organoboron compounds with halides or their equivalents (e.g., triflates) catalysed by palladium complexes has been called the Suzuki coupling (209 with M=BRn, see reviews 224, 225). Like the Stille cross-coupling, this reaction can be applied to the synthesis of various types of compound including conjugated dienes. An important feature of the process is that it Scheme 9 OH 235 OH 236Synthetic methodologies for carbo-substituted conjugated dienes should be carried out in the presence of bases (NaOEt, NaOH, KOH, K3PO4, Na2 CO3, etc.): coordination of the anion to the boron atom results in a higher electron density on the reaction centre, which facilitates the replacement of boron by palladium.In the case of sterically hindered alkenyl bromide, methyllithium proved to be a more suitable base.249 The catalyst used most often is Pd(PPh3)4. The organoboron approach is distinguished by versatility � by using specific protocols, both E- and Z-isomers of both the boron and halogen components can be prepared from terminal alkynes. The subsequent coupling in a required combination provides, at one's discretion, any isomer of the conjugated diene 55.222 The initial compounds can contain additional substituents and functional groups.O R1 B HB(OR)2 O R1C CH B(OPri)2 R1 R2 I R2C CH Br R2 R1 R2 (E,E)-55 R2 R1 (E,Z)-55 R1 R2 (Z,E )-55 R1 R2 (Z,Z)-55 The Suzuki cross-coupling has been used successfully for the stereocontrolled preparation of dienes 238 with a more complex pattern of substitution including cyclic and polyfunctional com- pounds.250 ± 254 R4 R3 R4 R3 R1 R1 B(OPri)2 [Pd] + R5 R2 R5 Hal R2 238 Thus this reaction has been used to prepare hydroprene (239) from alkenylboronic acid 240.255, 256 B(OH)2 CO2Et, Pd(PPh3)4 Br 240 CO2Et 239 An interesting methodological expedient is to carry out this process in the aqueous acetonitrile medium.257 The reaction was catalysed by a system comprising Pd(OAc)2 and water-soluble sulfonated triphenylphosphine, P(C6H4SO3H)3. Both compounds subjected to coupling contained functional groups.The nature of the base in the Suzuki cross-coupling can influence the process regioselectivity. Thus in the presence of triethylamine, catecholboranes 241 are coupled with alkenyl halides (or aryl halides) according to the `head-to-tail' pattern 751 giving rise to branched dienes 242, the process regioselectivity being higher than 94%.258 R1 O B +Hal R1 R2 R2 Pd black, Et3N, DMF, 80 8C (61% ±92%) O 242 241 d. Other variants of cross-coupling of vinyl components Vinyl derivatives of other elements are used somewhat less extensively in the cross-coupling reactions. The use of zinc, magnesium, aluminium and zirconium compounds for this pur- pose has been surveyed in a review.224 For example, hydro- metallation of terminal alkynes with R2AlH 259, 260 or Cp2ZrClH 261to give the corresponding organometallic intermedi- ates 243 and 244 followed by proper coupling resembles concep- tually the Suzuki coupling discussed above.R2 AlBui2 R1C CH DIBAH R1 ICH=CHR2, PdCl2(PPh3)2 (65%±70%) R1 55 243 HClZrCp2 R1 R1C CH HalCH=CR2R3, ZrCp2Cl PdCl2(PPh3)2, 2DIBAH (60% ± 91%) 244 R2 R1 R3 The cross-coupling of vinylcopper reagents with alkenyl halides is also catalysed by palladium or nickel complexes. This reaction had been discovered earlier than the more convenient Stille and Suzuki reactions. Characteristic features of this process have been studied in detail by Normant and Alexakis.9 The main work rule that ensures high yields and involvement of both alkenyl fragments of the reactants (RCH=CH)2CuLi is preliminary transmetallation of these compounds carried out by adding equivalent amounts of magnesium and zinc halides to the reaction mixture.Thus it is actually an organozinc compound that enters into the coupling reaction. Cross-coupling of neat organocopper compounds catalysed by nickel or palladium complexes gives products in a yield not exceeding 50%. The coupling of alkenylmagnesium bromides 245 and 246 with alkenyl bromides 247 and 248 can be induced not only by palladium complexes but also by iron 262 or nickel 263, 264 com- plexes; in the latter case, thiolate ion can act as the leaving group.Fe(acac)3, THF, NMP, 15 ± 20 8C Ph MgBr+ Br Ph 245 247 NiCl2(dppe), THF BrMg + PhS Br SiMe3 246 248 R PhS SiMe3 NiCl2(dppe), RMgBr, THF SiMe3 (72% ±86%) NMP is N-methylpyrrolidone. Vinyl silicon derivatives, unlike similar tin compounds, nor- mally do not enter into the cross-coupling reaction (see reviews 265, 266). Nevertheless, a particular type of active vinyl- silanes containing fluorine atoms at silicon has been identified among them.267 ± 269 Vinyl iodides and triflates were used as electrophiles; (Z3-C3H5)PdCl2 was found to be the best catalyst. The process was carried out in THF at 60 8C in the presence of Bu4NF.752 Dimerisation of alkenylmercury reagents 249 follows the `head-to-tail' pattern.270 R1 R2 R1 R2 PdCl2, Et3N, C6H6, 20 8C 2 (59% ± 93%) HgCl R2 R1 249 e.Cross-coupling of diene and saturated components This reaction is used more rarely. Two general cases are known, one with the diene component acting as the nucleophile and the saturated one being the electrophile and vice versa. Dienylalane 250 reacts with methoxyethoxymethyl chloride (MEMCl) to give the expected ether 251.271 The compound 250 does not react properly with aldehydes and cyanogen; however, the target dienonitriles 252 and dienic alcohols 13 can be prepared from the corresponding ate-complex 253.271, 272 R2 R2 Bui2AlH C CH AlBui R1 R1 2 250 OMe R2 ClCH2O O 72% OMe R1 251 R2 (CN)2 (62%) CN R1 MeLi R2 252 7 R2 R3CHO R1 AlBui2MeLi+ OH 253 (42% ± 76%) R1 R3 13 Yet another example of using a `nucleophilic' diene compo- nent is the Heck arylation of the terminal conjugated dienes 38.273, 274 The process is carried out in the presence of HHal acceptors such as amines 273 and silver or thallium acetates,274 the E,E-isomer of diene 254 being formed predominantly irrespective of the configuration of the initial compound 38.Pd(OAc)2, PAr3, D, HHal acceptor Ar ArHal + R R (492%) (E,E)-254 38 An original synthesis of bombykol 96 can serve as an example of cross-coupling with `electrophilic' diene component 255 (easily prepared from dibromide 256 formed upon bromination of unsaturated aldehyde 257) and a `nucleophilic' saturated compo- nent.275 In this case, the bond configuration at the reaction centre is not violated.This strategy was verified using numerous exam- ples, the yields of Z,E-dienes ranging from 58% to 85%.275 CBr4, PPh3 CHO ButMe2SiO(CH2)9 (88%) 257 Bu3SnH, Br Pd(PPh3)4 ButMe2SiO(CH2)9 (85%) Br 256 1. PrnMgBr, NiCl2(dppp) 2. Bu4NF ButMe2SiO(CH2)9 (78%) Br 255 HO(CH2)9 Prn 96 A A Vasil'ev, E P Serebryakov Some examples of using dienyl sulfides 258 (R3=Ph, But) in these reactions are documented;45, 263, 276 in this case, the thiolate ion is the leaving group (the tert-butyl substituent is preferred 45). The configuration of the products 55 remains the same as in the initial sulfides 258.NiCl2(dppp) or NiCl2(dppm) R1MgHal R2 SR3 (E,E )-258 R2 R1 (72% ± 86%) (E,E )-55 SR3 R2 (Z,E )-258 R1 R2 (65% ± 88%) (Z,E)-55 (E,Z )-258 R2 SR3 R2 (83% ±90%) R1 (E,Z)-55 The cross-coupling strategy makes it possible to substitute a nucleofugal group at the 2-position of the diene system; this results in branched derivatives 259 and 260. Dienol triflates 261 or phosphates 262, which are easily prepared from the corresponding a,b-enones 263 or 264, can serve as convenient substrates for this purpose. The cyclic dienol triflates 261 react with the Grignard reagents in the presence of catalytic amounts of CuI,277 while the acyclic phosphates 262 react in the presence of complex nickel catalysts.278 The dienes 260 are formed as mixture of geometric isomers.The phosphate strategy is suitable also for the synthesis of cyclic dienes 259. R OTf O RMgHal, CuI 1. LDA 2. Tf2O (55% ± 96%) (CH2)n (CH2)n (CH2)n 259 261 263 O 1. LDA 2. (PhO)2P(O)Cl R2 R1 264 R3MgHal, NiCl2(dp) R3 R2 OP(O)(OPh)2 or NiCl2(dppp) R2 R1 (40% ± 92%) R1 260 262 This technique has been used to introduce a methyl group during the synthesis of methoprene (265) according to the C10+C4+C1 pattern; the substitution of the methyl group for the OAc group in compound 266 occurred with complete reten- tion of the initial configuration.279 OAc O CO2Pri CO2Pri MeLi, CuI CHO OMe OMe 266 CO2Pri OMe 265Synthetic methodologies for carbo-substituted conjugated dienes f.Cross-coupling involving 1,2-dihaloethenes The (E )- and (Z)-isomers of 1,2-dichloroethene (267) can serve as convenient and inexpensive synthons for the preparation of conjugated dienes (as well as styrenes, enynes and so forth) (see a review 280). By varying the conditions and catalysts, it is possible to perform consecutive coupling of these compounds, first, with a saturated and then, a vinylic component (or vice versa),281 result- ing in the formation of the target dienes in high yields and with high isomeric purity. It is of interest that 1,2-dibromoethene is much less effective in this type of process.280 R3 Cl R2 R1M, M, Cl R1 R2 R3 [Pd] and/or [Ni] (E)-267 R3 Cl R1 R2 Cl (Z)-267 g.Synthesis of 1,3-dienes from alkynes and allenes As noted above, alkynes can be used in the cross-coupling instead of alkenes. The reactions of terminal alkynes or their metallated derivatives with alkenyl halides (the Sonogashira reaction, see a review,282 and transformation of 213 into 214 in general Scheme 8) yield conjugated enynes, which can be reduced to dienes (the reduction techniques are described in Section VII). In addition to alkenyl halides, these reactions can be carried out for alkenyl- (268) and dienyl tellurides 269; these afford conjugated enynes 270 and dienynes 271, respectively.283 (268) BuTe R2 R2 PdCl2, CuI, MeOH, Et3N R1C CH 270 R1 (62% ± 85%) 213 R2 BuTe(269) R2 271 R1 The addition of metal hydrides or organometallic reagents to the triple bond is stereospecific (cis-1,2-addition), which permits the preparation of pure isomers.Of relatively early studies, one should note the synthesis of dienonitrile 272 on the basis of hydroalumination of hex-3-yne.272 Et Et Et Et CN AlBui2 1. BuLi 2. (CN)2 Bui2AlH 2 EtC CEt 70 8C (63%) Et Et Et Et 272 Hydroboration of alkynes resulting in alkenylboranes is the obligatory first stage of the Suzuki cross-coupling (see above). However, other methods of transformation of alkenylboranes 273 into the conjugated dienes 55 are also known, for example, successive hydroborations of chloroalkyne 274 and the alkyne 213 followed by the borotropic rearrangement.284 R1 Cl 1.ClC CR1 (274) 2. HC CR2 (213) NaOMe B BH2 R2 R1 OMe PriCO2H R2 R1 B (53% ± 63%) 55 273 R2 753 Yet another organoboron method consists in sequential chemoselective coupling of two alkynes (275 and 276, the latter as the lithium derivative), which gives conjugated enynes 277, the immediate precursors of E,Z-dienes.285, 286Et 1. I2 2. NaOAc 3. H2O2 , NaOAc 7 1. (sia)2BH 2. EtC CLi (276) (sia)2B Li+ HC C(CH2)8OAc 275 (CH2)8OAc Et 1. (sia)2BH 2. HOAc 3. H2O2 , NaOAc Et (CH2)8OAc 277 (CH2)8OAc The addition of the cuprate reagents at the triple bond has been studied most comprehensively (see reviews 9, 287). In early studies,288, 289 cuprate reagents 278 have been prepared from the corresponding vinyllithium derivatives; their subsequent syn- addition to the triple bond in alk-2-ynoates 279 furnished dien- oates 280 (pathway a).The stereochemical outcome of the reaction is important: the new double bond always has the E-configuration and the initial configuration of the vinylcuprate reagent does not change. R2 [Cu] R1 R3 278 R4 R2 R4C CCO2Alk (279) CO2Alk R1 (a) (63% ± 90%) R3 280 R2 R2 HC CH E+ [Cu] E R1 R1 (57% ± 83%) (b) R3 6àH R3 282 R3 281 The development of methods for the addition of cuprate reagents to a non-activated triple bond has markedly extended the scope of the synthesis of dienes.287, 290 Starting from different vinylcuprates 278 (R3 6à H) and acetylene (pathway b), one can obtain diene intermediates 281, which then react with C-electro- philes (AlkI, CO2, PhSCH2NEt2).This affords dienes 282 in which the double bond is Z-configured. Simple vinylcuprates 283 are formed from alkylcuprates 284 and acetylene; according to the stereochemistry of this process, this vinylcuprate has a cis- configuration.291 ± 293 By varying the natures of the second, acetylenic component and the final electrophile, one can obtain products 285 and 286 with a desired configuration of the second double bond. HC CCO2Et R (78%) CO2Et 285 R [Cu] E R[Cu]+C2H2 284 283 R 1. HC CH 2. E+ (30% ± 71%) 286 Recently, the organocuprate method has been successfully used for the preparation of dienes of the cyclobutane series of type 287 and 288 using an intramolecular reaction.294 Fairly stable organotin substrates 289 and 290, which underwent transmetalla- tion under the reaction conditions, served as the precursors of active species.754 EtO2C Me3Sn CuCl, AcOH, DMF, 0 8C CO2Et (85% ± 95%) 289 287 CO2Et CuCl, DMF, 0 8C Me3Sn CO2Et (73%±94%) 288 290 The ability of cuprate reagents to add to the double bond of a,b-enones has been used for the formation of the conjugated diene system in the synthesis of ipsenol 291.The a-allenyl ketone 292 needed for this purpose was prepared from amide 293 and allene 294.295 Li NMe2 ( )2CuLi LiAlH4 (294) O O (95%) (70%) HO O 292 293 291 The addition of alkylcuprate reagents at the triple bond of enynes 295 results in the branched conjugated dienes 242 or alka- 2,4-dienoic acids 296.296 R2 H3O+ R1 (29% ± 70%) 242 R22 CuMgHal R1 R2 295 CO2H 1.HMPA, P(OEt)3 2. CO2 R1 (70% ± 73%) 296 h. Coupling of alkene and alkyne components using ruthenium catalysts The last decade was marked by enhanced interest in the catalysts based on ruthenium complexes (see a review 297) which catalyse other types of transformation than palladium, which has become a routine material. These processes include coupling of activated alkenes 297 with alkynes (Scheme 10). This reaction catalysed by the Ru(cod)(cot) complex (cot is cycloocta-1,3,5,7-tetraene) and resulting in conjugated dienes 298 is a stereospecific addition of a terminal alkene to the triple bond.298 The product yield depends dramatically on the nature of the substituents: in the case of acrylates (X=OAlk), good results were obtained only with diphenylacetylene (R=Ph); on passing to acrylamides (X=NMe2), the range of possible acetylenic components is broader.Scheme 10 R R R R (493%) X 298 O X Ru(cod)(cot) R3 R2 R3 R2 O 297 O R1 R1 X 299 (488%) X=OAlk, Alk, NAlk2. Instead of alkynes, diene hydrocarbons can also been used in this reaction;299 in this case, incompletely conjugated dienyl carbonyl compounds 299 are formed as the reaction products. A A Vasil'ev, E P Serebryakov This reaction is also sensitive to the type of substitution at the double bonds of the reactants and can result in isomer mixtures.Coupling of terminal non-activated alkynes 213 with non- activated alkenes 300 takes place on prolonged refluxing in pyridine in the presence of CpRu(PPh3)2Cl and NaPF6.300 The newly formed double bond in the diene 55 always has the Z-configuration, the process being accompanied by the formation of noticeable amounts (up to 13%) of non-linear isomers of 242. CpRu(PPh3)2Cl, NaPF6, Py, 100 8C R2 R1C CH+ (9% ± 67%) 213 300 R2 R2 R1 R1 + 242 55 The reactions of alkynes with alkenes catalysed by another ruthenium catalyst, Cp(cod)RuCl, result in 1,4-dienes.301 This method is suitable for the preparation of conjugated dienones 301 only in the case where the initial acetylene derivative 302 contains a latent carbonyl group. This facilitates isomerisation of the 1,4- diene system to a conjugated one.1. H3O+ 2. DBU Cp(cod)RuCl, DMF, 100 8C R + O O O O (41%±59%) R 302 O R 301 Coupling of two different acetylene derivatives 213 and 304 catalysed by the complex 303, which gives enynes 305 (potential precursors of dienes), proceeds strictly stereospecifically.302, 303 Z5-C5Me5 O O R2 CPh (303) Ph3PRuC R1C X (44% ±90%) 304 305 R1C CH+R2C CCX 213 C5Me5 is pentamethylcyclopentadienyl. Intramolecular cyclisation of diterminal diacetylenes of the general type 306 with a relatively long linking chain (which can contain functional groups and even a ferrocene residue) induced by ruthenium complex 307 is accompanied by the formation of enynes of the macrocyclic series 308 (see Ref.304). Alk S Me5C5 Ru C C (307) Cl S C5Me5 RuCl C CH Alk (478%) C CH 308 306 i. Syntheses of exocyclic dienes In conclusion, mention should be made of several intramolecular cross-coupling reactions of the same type, resulting in five- membered vicinally substituted carbo- or heterocyclic compounds 309. Of these, cyclisation of 1,6-enynes of the general type 310 has been studied most comprehensively. When catalysed by palladium complexes,305 ± 308 this reaction proceeds smoothly in most cases if the vinyl group of the initial enyne 310 does not contain a substituent (R2=H). A single example has been reported 309 where this reaction occurred smoothly in the presence of a substituent in this group, namely, R2=CH2OR3.However, in most cases, mixtures of isomers having the same carbon skeletonSynthetic methodologies for carbo-substituted conjugated dienes of the type 309 and differing in the position of the double bonds are produced.310, 311 R1 R1 Pd(II), D X X (50% ± 86%) 309 310 R2 R2 X=CMe2, C(CO2Me)SO2Ph, C(CO2Me)2, NR. In some cases, this approach can be used to prepare six- membered rings but with some limitations regarding the type of substitution in the initial compounds.312, 313 Similar regularities have been found for the same transformation catalysed by nickel complexes immobilised on phosphinylated polystyrene.313 Ruthenium complexes, for example, RuCl(cod)(C5Me5) and RuClH(CO)(PPh3) , are capable of inducing cyclisation of allyl propargyl ethers 314 and derivatives of alk-2-en-7-ynoic acids 311 315 the products of which are dienes 312 and 313, respectively.Reactions catalysed by ruthenium complexes are much less sensitive to the nature of substituents than the same reactions catalysed by palladium complexes. R1 R4 R4 R3 R3 R1 RuCl(cod)(C5Me5) O O (58% ± 94%) 312 R2 R2 R1 CO2R2 R1 CO2R2 RuClH(CO)(PPh3)3 (53% ± 82%) 313 311 The preparation of vicinal di-exo-methylene compounds 314 by intramolecular cyclisation of bromides 315 (process of the Heck reaction type) has been described.308, 316 The reaction takes place smoothly in the presence of the Pd(OAc)2 ± PPh3 catalytic system 308 or the RhCl(PPh3)4 catalyst,316 whereas with the Pd(PPh3)4 catalyst, the six-membered isomer 316 is the major product.316 Intramolecular coupling of dibromides 317 affords the dimethylene isomer 314 as the only product; according to the balance, a stoichiometric amount of a reducing agent [Ph3P or (p-MeOC6H4)3P] is required.317 [Pd] or [Rh] Br X 315 + X X 316 314 Pd(OAc)2 (5%), PAr3 (100%) X Br Br 317 X=PhCH2N, C(CO2Et)2, C(COMe)2.1,2-Dialkylidene-substituted cyclic compounds 318 can also be prepared from diacetylene derivatives 319; this requires partic- ipation of a reducing reagent. In one of the procedures, cyclisation is induced by titanocene (Cp2Ti) 318 or zirconocene (Cp2Zr) generated in situ.319 In most cases, the use of zirconium is more advantageous than titanium:320 in the synthesis of four-, five-, six- and seven-membered rings, good yields are obtained even in the case of bulky R1 and R2 substituents. However, when R1=R2=OEt, it is necessary to use the titanium reagent. 755 R1 R1 R1 Cp2M H3O+ X X X MCp2 (497%) R2 (H2C)n (H2C)n (H2C)n318 R2 R2 319 X=CH2, O; n=0 ± 3;M=Ti, Zr.The transformation of 1,6-diynes, for example, 320, into the corresponding exocyclic diene 321 can be carried out with participation of the (dba)3Pd2 . CHCl3 palladium catalyst and the Et3SiH reducing agent (stoichiometric amounts).321 (dba)3Pd2 .CHCl3 , Pr ButMe2SiO ButMe2SiO Pr (o-Tol)3P, AcOH, Et3SiH, C6H6 (95%) CO2Me 320 321 CO2Me In the case of other substrates of this chemical type, the yields vary in the range of 34%± 95%.Two cases of cyclisation of compounds 322 and 323 followed by functionalisation are known.322, 323 The functional groups being introduced occupy the Z-position in products 324 and 325, which is related to the mechanism of intramolecular syn-carbo- metallation at the cyclisation stage. Br L2Pd Br Pd(PPh3)4 , PdBr2 . 2L O O SiMe2But SiMe2But 322 CO2Me PdL2Br CO, MeOH (78%) O O 324 SiMe2But SiMe2But SiR3 HSiR3, 1108C, (phen)PtMe2, B(C6F5)3 X X (68% ±89%) 323 325 X=C(CO2Me)2, C(CH2OR1)2, [C(CO2Et)2]2. 2. Miscellaneous reactions catalysed by organometallic compounds The reactions considered in this Section are difficult to classify on the basis of a particular feature. Most of them have been discovered in the current decade and it is obvious that these investigations are at the stage of accumulation of facts.Never- theless, it can be noted that the interest of researchers tends towards reactions catalysed by ruthenium complexes. These complexes are active in transformations differing from those catalysed by Pd and Ni complexes. Some of them (cross-coupling of alkenes and alkynes) were described in the preceding Section. a. Alkene ± alkyne metathesis Ruthenium carbenoid complexes catalyse an interesting coupling reaction of alkenes with alkynes during which the reactants undergo metathesis to give branched conjugated dienes.324, 325 One of these complexes, Cl2(Cy3P)2Ru=CHPh (326, Cy is cyclo- hexyl), the so-called Grubbs catalyst II, has now become com-756 mercially available and has been tested in reactions of ethylene with numerous acetylene derivatives of various structures.Unfortunately, only for dienes of the type 327 and 328, have preparative yields been attained; esters of propargyl alcohols 329 and N-tosylpropargylamines 330 served as the precursors of these products. However, even in these cases, reactions did not proceed to completion, which is in line with the concept of reversibility of metathesis. R2 326, CH2Cl2, 20 8C, 40 h H2C CH2+R1 (48% ±66%) OCR3 OCR3 R1 O O 329 R2 327 O H2C CH2+ N SC6H4Me 326, CH2Cl2, 20 8C, 40 h (81%) n-H13C6 O 330 O N SC6H4Me O 328 n-H13C6 By conducting reactions under a pressure of ethylene (*4 atm), one can shift the equilibrium towards the products (for esters of type 329 with the terminal triple bond, R1=H, the yields increased to 92%);326 however, upon further increase in the ethylene pressure, the reactions were retarded due to the degener- ate ethylene ± ethylene metathesis.326, 327 The problem of separation of the initial and final compounds, the polarities of which are close, should also be mentioned.An ingenious solution of this problem has been proposed.328 The researchers suggested that the initial propargyl alcohol be immo- bilised on a polymeric support through a linker of an ester structure 331.328 In this case, after metathesis, the immobilised product has already the allyl carboxylate structure 332.The subsequent reaction of the compound 332 with nucleophiles catalysed by palladium complexes results in diene product 333, which passes to the solution, while the unreacted propargyl derivative 331 remains on the support. O O [Ru] + O O (CH2)n R 331 Polymer O O Nu, Pd(PPh3)4 R O O (CH2)n (32% ±78%) 332 Polymer R Nu 333 Stable imidazolidinylruthenium complex 334, obtained recently NMes MesN Ru Cl Cl Ph PCy3 334 exhibits in many cases a much higher efficiency than the Grubbs catalyst II.329 In particular, with the assistance of this complex, metathesis can be carried out for alkenes and alkynes character- ised by a substantially broader spectrum of functionalisation and substitution 330 (cf.data of Refs 325 and 327). A A Vasil'ev, E P Serebryakov R3 334, CH2Cl2, 20±40 8C R2+ R1 (498%) R1=Ph3CO(CH2)2, Me3Si, Cy, CO2But, CH2OAc, CH2OSiR4 R2=H, Alk, CH2OAc; R3=CH2SiMe3, CH2OSiR43 , CH2CO2Bn. The intramolecular alkene ± alkyne metathesis of enynes 335, discovered previously, is much less sensitive to the chemical type of the initial compounds than the intermolecular process because it is accompanied by the formation of favourable cyclic systems 336.331, 332 The product yields are very high, except for the case of terminal alkynes (R1=H). Later, the problem of low yields from terminal alkynes has been unexpectedly solved: it was found that this process should be carried out under an atmosphere of ethyl- ene.333 Ethylene does not formally participate in the reaction but it favours catalyst transfer in the form of the carbene Ru=CH2. R1 R1 326, CH2Cl2, C2H4, 20 8C X (65% ± 99%) (CH2)n R2 335 X=O, NR3, C(CO2Et)2; R1=Me, CH2OAc, CH2OSiMe2But, CO2Me, H; R2=H, Ph; n=0±2. A study 334 demonstrating the possibility of involvement of alkynylboronic acids 337 in intramolecular metathesis deserves attention.Under the action of Ru catalyst 326, they are smoothly converted into boryl-substituted dienes 338, potential substrates for the Suzuki cross-coupling.334 326 X X (65% ± 95%) B(OR)2 338 337 X=CH2, (CH2)2, O, TsN, OCH2O, C(CO2Et)2. It is of interest that compounds 339 with an appropriate mutual arrangement of the two double bonds and one triple bond undergo double metathesis catalysed by Cl2(Cy3P)2Ru=CHCH=CPh2 (Grubbs catalyst I) to give bi- cyclic 1,3-diene structures of the general type 340 (n=1, 2; m=0 ± 2).335 The nature of the substituent R at the triple bond plays a crucial role: high product yields were obtained for R=H, Alk, Ph, CO2Me, whereas for R=Br, I, Me3Si or Bu3Sn, cyclisation did not proceed.When tungsten-based catalysts of metathesis are used, cyclisation of the same substrates involves two double bonds and is accompanied by the ejection of a lower alkene without participation of the acetylene fragment. (CH2)m (CH2)n Ru C R 339 (CH2)m (CH2)n [Ru] R (CH2)m (H2C)n 78 ± 96% [Ru] R R3 R2 R1 333; X (CH2)n R2 336 B(OR)2 (CH2)m (H2C)n340 R757 Synthetic methodologies for carbo-substituted conjugated dienes The highest yields (60% ± 94%) were obtained in the case of aryl bromides and iodides.R6 R1 R3 R1 Pd(dba)2, PPh3 , 120 8C, K2CO3 +R6X 1,6-Enynes 341 are converted in the presence of [RuCl2(CO)3]2 to give diene product 342. The process fits formally the concept of the intramolecular alkene ± alkyne metathesis considered above. Nevertheless, the researchers 336 suggest a different reaction mechanism. R3 H (494%) R2 [RuCl2(CO)3]2 or R2 R4 R4 R2R5 352 R5 351 PtCl2 X X R2 (79% ± 97%) R1 342 R1 341 Carbonylation of allenylmethylamines 353 is directed at the allylic position, to the central atom of the allene system. The allylic rearrangement affords conjugated dienes 354 with the amino- carbonyl function at the 2-position.346 O R1 R4 CO, Pd(dppp), H+, CH2Cl2, 75 8C R52 N NR52(59% ± 97%) R2 R3 R3 With PtCl2 catalysis, enynes of the type 341 containing electron-withdrawing groups or a cyclic enyne fragment (with a medium or large ring) are also converted into formal metathesis products 342).337 Otherwise, they tend to form cyclopropane derivatives 343.R1 R2 R4 354 353 X 343 Hydrozirconation of allenylstannane (355) yields intermediate 356, the reaction of which with aldehydes or ketones is accompa- nied by the allylic rearrangement 347 (the same outcome is provided by the reaction of aldehydes with pinacol 1-trimethylsilylprop-1- ene-3-boronate 43, 44).Decomposition of adduct 357 is accompa- nied by the formation of conjugated dienes, the proportion of E,E- isomer being more than 93%in the case of aldehydes (R2=H).347 R1R2C O ZrCp2Cl b. Synthesis of 1,3-dienes from allene derivatives Ruthenium complexes catalyse an unusual addition of allenes 344 to activated alkenes 345 giving rise to 1,3-disubstituted conjugated dienes 346.338 To activate the catalyst, alkynols are used, the best results being attained with 2,4,7,9-tetramethyldec-5-yne-4,7-diol. Cp2Zr . HCl SnBu3 CpRu(cod)Cl, OH HO SnBu3 355 356 R2 + OZrCp2Cl R1 (55% ± 81%) O R2 R1 Acid 345 344 R2 (68% ±92%) R1 SnBu3 R2 357 R1 O 346 c. Synthesis of 1,3-dienes on the basis of alkynes Hydroformylation of conjugated enynes 358 in the presence of rhodium complex 359 gives branched dienic aldehydes 360.348 A minor admixture of monounsaturated aldehyde is also formed.7BPh3 (359), The presence of a nitrile, ester or amide substituent in the side chain of the allene does not influence the course of the reaction. Regarding allene alcohols, a limitation has been found in the study of the mechanism � the number of carbon atoms between the allene system and the hydroxy group should not equal three or four; otherwise, oxygen-containing heterocyclic compounds are formed.339 R R Rh+(cod) H2, CO, CH2Cl2, 60 8C X O X 358 H360 X=H, OAc, OMe. Diethyl dienedioates 361 are produced upon carbonylation of alkyne-a,a 0-diol dicarbonates 362.349 The reaction occurs through the intermediate formation of allene derivative 363.The possibility of preparing dienoates 364 from the corresponding a-allenyl carbonates 365 was also demonstrated. Palladium complexes can catalyse not only cross-coupling of vinylic components but also other types of reaction resulting in conjugated dienes. A series of studies dealing with coupling of vinylic (347), allene (348) and CH-acidic (349) components,340 ± 344 resulting in branched functionalised dienes of the type 350, has been reported. Various combinations of three components differ- ing in the nature of the substituents and functional groups have been tested. Thus the vinylic component can be either linear or incorporated in a carbo- or heterocyclic system; the leaving group can be represented by either bromide or triflate. Among C-nucleo- philes, Schiff's base Ph2C=NCH27 CO2Me has been studied, apart from the standard b-dicarbonyl compounds.The yields vary over broad limits and can reach 94%. R R CO, EtOH, Pd(OAc)2, PPh3 OCO2Et EtO2CO R3 EWG R3 X EWG EWG R4 [Pd(0)] 362 7 + + R EtO2C EWG R4 R2 R1 CO2Et EtO2C (17% ± 94%) R2 R1 350 349 348 347 OCO2Et (21% ± 75%) R R EWG is electron-withdrawing group. R 361 363 Allene hydrocarbons 351 containing an allylic a-hydrogen atom undergo a Pd(dba)2-catalysed reaction with aryl and alkenyl halides and triflates 345 to give conjugated diene compounds 352.758 R MeO2C R CO, MeOH, Pd(OAc)2, PPh3 OCO2Me 365 364 The reaction of propargyl dicarbonates 366a,b with organo- boron or -tin compounds catalysed by palladium complexes is accompanied by the double allylic rearrangement, resulting in substituted 1,3-dienes 367a,b.350 R1 R1 R2M(2 equiv.), Pd(OAc)2, PPh3 (31% ± 94%) R1R1 OCO2Me R2 MeO2CO 366a,b R2 367a,b R1=H(a), Me (b); R2M=NaBPh4, ArB(OH)2, HetB(OH)2, HetSnBu3, PhC CSnBu3.A reaction with a similar outcome giving rise to similar dienes consists in the interaction of organotin compounds of various structures with a,a 0-disubstituted acetylenes 368 (X=Cl, OTs).351 R R RZnBr (2 equiv.), CuCN. 2 LiBr (82% ±98%) X X 368 In the presence of Ni(cod)2, propargylsilane 369 reacts with aldehydes and organozinc compounds to give the 2,4-disubsti- tuted 1,3-dienes 242.352 R2 Ni(cod)2, 0 8C, THF R1CHO+ +ZnR22SiMe3 (37%± 58%) R1 242 369 Complex cyanocuprates 370 derived from terminal alkynes add 4 moles of iodomethylzinc iodide (371).The subsequent treatment of the addition product with electrophilic reagent 372 furnishes conjugated 2,3-disubstituted diene 373.353 The reaction includes the stages of insertion of a carbenoid and two allylic rearrangements. In the case where the cyanocuprate complex 374 contains a fragment of sterically hindered ether and the process is initially carried out in the presence of an electrophile (cyclo- pentanone), three moles of the reagent 371 are added. This route was used to prepare diene 375. CO2But 1. ICH2ZnI .ZnI2 (371) 2. Br (372) R CO2But CCu(CN)Li RC (50% ± 74%) 373 370 1. O OTHP OH 2. ICH2ZnI . ZnI2 (3 equiv.) THPO R CCu(CN)Li C R 374 375 Hydrozirconation of ethoxyacetylene (117) and treatment of the resulting zirconium intermediate 376 with a,b-enals affords dienals 377 in a preparative yield.354 The same products can be obtained using the same reactions but starting from 1-methoxy- but-1-en-3-yne and a saturated aldehyde. R CHO Cp2ZrHCl OEt ClCp2Zr OEt 90% 376 117 Cp2ZrHCl RCHO OMe ClCp2Zr OMe A A Vasil'ev, E P Serebryakov R CHO 377 Alkenylzirconium compound 378 reactwith lithiated epoxy nitriles 379 to yield nitriles 380.355 If pure epoxy nitrile isomers are used, the reaction still gives a mixture of geometric isomers at the cyano-substituted double bond.O CN R1 (379) R2 Li Cp2ZrHCl ZrClCp2 n-C6H13 n-C6H13C CH 378 O CN R1 CN LiO C6H13-n R1 R2 Cp2Zr7 (42% ±74%) ZrClCp2 R2 Li+ Cl C6H13-n R1 C6H13-n R2 CN 380 The reaction of disubstituted acetylenes with the carbene derivative of titanocene (the Tebbe reagent) results in a [2+2]-addition product, substituted titanacyclobutene 381.356 The subsequent regioselective reaction of 381 with carbonyl compounds gives conjugated dienes 382. In the case of dimethyl- or diethylacetylene (R1=Me, Et), various aldehydes and ketones can be used, whereas in the case of tolan (R1=Ph), the range of carbonyl compounds is limited to reactive aldehydes (R2=H; R3=H, Me, Ph).R1 Cp2Ti CH2 . AlMe2Cl R2R3C O R1C CR1 Cp2Ti R1 381 R1 R1 R1 R1 (40% ± 98%) R2 Cp2Ti 7Cp2Ti O O R2 R3 R3382 A curious process of acetylene dimerisation in the presence of electron-deficient alkenes makes it possible to prepare bifunc- tional Z,E-dienes 383.357 Acrolein (Y=CHO), vinyl ketones (Y=COR), nitroethylene (Y=NO2) and b-substituted croton- aldehyde were tested in this reaction. In the latter case, the product yield was not more than 3%. Pd(OAc)2 Y LiBr+2HC CH+ Y Br (28% ± 41%) 383 The reactions of disubstituted alkynes 384 with niobium pentachloride under reductive conditions gives intermediate 385, containing the niobiacyclopropene fragment, able to add two equivalents of aldehydes.358 Finally, conjugated dienes 386 with identical substituents at the ends of the system are produced.R2 R1 2 R3CHO NbCl5 , Zn,DME±C6H6 R1C CR2 NbLn 384 385 R2 R1 R2 R1 R3 R3 R3 R3 (41% ± 59%) 386 O O NbLnSynthetic methodologies for carbo-substituted conjugated dienes Similar tantalum compounds 387 add a second acetylene molecule; after hydrolysis, the conjugated diene 179 is liber- ated.359 The scope and limitations of this method have not been adequately studied. R2 R3C CR4 , Na/Hg Ta(DIPP)3Cl2(OEt)2 , R1 Na/Hg R1C CR2 Ta(DIPP)3 384 387 R3 R3 R2 R2 H2O R4 R1 R4 R1 179 Ta(DIPP)3 DIPP is 2,6-diisopropylphenoxide. The reactions of hexacarbonyldicobalt complexes of alkynes 388 with activated alkenes afford conjugated dienes 389 in moderate yields.360 PhC CH CO2Et CO2Et + (45%) Ph 388 Co2(CO)6 389 d.Syntheses of conjugated dienes from other dienes Allylic acetates 390 and 391 react with malonate anions in the presence of palladium catalysts to give products 392 and 393 with a diene system.361, 362 To ensure the maximum regio- and stereo- selectivity, the process should be carried out in the presence of tributylphosphine. When triphenylphosphine is used, the regiose- lectivity substantially decreases. For example, the transformation of the acetate 391 gives, in addition to the product 393, 40% of the product of nucleophile addition at the central carbon atom. Pd(OAc)2, PBu3, THF, 40 8C +NaCMe(CO2Et)2 AcO (83%, E :Z=83 : 17) 390 CO2Et CO2Et 392 OAc Pd(OAc)2, PBu3 , THF, 40 8C +NaCH(CO2Et)2 (89%, E :Z=80 : 20) 391 CO2Et CO2Et 393 The reaction of allylic carbonates with the dimethyl malonate carbanion catalysed by the Mo(CO)3(EtCN)3 system with chiral ligand 394O N NH NH N O 394 is accompanied by allylic rearrangement producing optically active homoconjugated compounds.363 When triene carbonates 759 395 are used, the conjugated diene system is retained in the product and is connected to the chiral carbon atom.CO2Me Mo(CO)3(EtCN)3 ± 394 7 OCO2Me + R CO2Me 395 CH(CO2Me)2 R (58% ± 81%) (97% ±98% ee) The reaction of dienes with electron-deficient alkenes (see Scheme 10) 299 catalysed by ruthenium complexes should also be attributed to this type of reaction.e. Reactions of iron diene complexes Conjugated dienes form stable complexes 396 with tricarbonyl- iron (see a review 364). The electron-donating properties of Fe(CO)3 facilitate the addition of electrophilic reagents at the termini of the diene system and contribute to the stabilisation of the allylic carbocation; this enables introduction of nucleophilic reagents into this position. X 1. E+ 2. [O] E X 1. Nu7 2. [O] Nu Fe(CO)3 + 7X7 396 Fe(CO)3 It has been proposed to use this type of complexation as a temporary protection of a diene system. The diene regeneration from the complex takes place on treatment with mild oxidants, for example, (NH4)2Ce(NO3)6 or N-oxides. The following examples demonstrate the successfulness of using this strategy in fine organic synthesis.Complexes of terminal dienes can easily be converted into functional derivatives 397 some of which are insect pheromones.365 LiAlH4 , (CH2)mCO2Et AlCl3 ClCO(CH2)mCO2Et, AlCl3 R R Fe(CO)3 O Fe(CO)3 (CH2)mCH2OH Me3N O R Fe(CO)3 (CH2)mCH2OH R 397 The complex of conjugated dienal 398 can react with some C-nucleophiles at the aldehyde group without a catalyst.366OH SnBu3 Bu Fe(CO)3 OSiMe3 Bu OH Ph O Bu CHO Fe(CO)3 398 Fe(CO)3 Ph The reaction of p-allylcarbene cationic complex of tricarbo- nyliron 399 with boron ate-complexes of enolates yields Z4-com- plex derived from conjugated diene 400.367 Decomposition of the complex 400 on treatment with Ce(IV) gives rise to diene 401 containing an oxo group in the b-position.A A Vasil'ev, E P Serebryakov 760 +BF¡¦ OMe 4 7 the preceding Sections.They are arranged according to the type of fragment introduced in the final conjugated diene. BEt3K+ (CO)3Fe (40% ¡À 80%) O O + 1. The =CH7CH=CH2 (C3) synthons R1 R3 R2 399 Fe(CO)3 O O Ce(IV) R1 R1 The widely known allylic Wittig reagents and their analogues are equivalent to the =CH7CH=CH2 synthon. This type of reagents also includes 1-bromo-3-iodopropene.372 The reaction of this compound with aldehydes on treatment with tin(II) chloride is accompanied by allylic rearrangement and gives terminal 1,3-dienes 38 in moderate yields. (85% ¡À 95%) R3 R3 Br SnCl2 R2 401 R2 400 R Br I +RCHO (*50%) OSnCl2I R Mention should be made of recent publications 368, 369 describ- ing the synthesis of functional derivatives of diene tricarbonyliron complexes 402 368 and 403.369 38 O O Fe2(CO)9, sonication Ba(OH)2, MeOH Fe(CO)3 O 2.The7CH=CH7CH=CH2 (C4) synthons In a known four-step procedure for the preparation of butadienyl Fe(CO)3 ketones 408, 4-phenylthiobutanal (409) is used as the source of the diene fragment for `dienylation' of a-chlorosulfoxides.373 402 HO HO HO Cl CHO (409) O CO2Me CO2Me 1. LDA 2. PhS 3. ButOK MgCl, CuBr2 .Me2S SPh R + R 42% [O], D SPh (40% ¡À 65%) Fe(CO)3 Fe(CO)3 SPh O O 403 R 408 Abroader range of compounds of type 410 can be prepared on the basis of 1-phenylsulfo-4-trimethylsilylbut-2-ene (411) (path- way a).374 The same products are formed when a more readily available isomer, 1-phenylsulfo-4-trimethylsilylbut-1-ene (412) is used instead (pathway b).375 f.Miscellaneous organometallic reactions The reaction of furans 404 with diazo esters in the presence of Rh2(OAc)4 affords a product mixture containing, among other compounds, oxodienoate 405.370 On treatment with catalytic amounts of iodine, this mixture can be converted into the pure diene (E,E )-405 (or a mixture of E,E- and E,Z-isomers if R2 6�� H). R3 CO2Et, a SO2Ph N2 Rh2(OAc)4 I2 (Cat.) Me3Si 1. LDA 2. E+ mixed products 411 (66% ¡À 85%) R2 R1 b SO2Ph O 404 Me3Si R2 412 R1 CO2Et SO2Ph Bu4NF Me3Si E O R3 405 E 410 R, R3Si, R3Ge.E=Alk, RC(O), O Cyclic thioacetals 406 react with cyclopropyl Grignard reagents 407 in the presence of catalytic amounts of NiCl2(PPh3)2 on refluxing in benzene or toluene to give dienes.371 If the type of substitution in the three-membered ring differs from that in 407, the reaction gives a mixture of isomers. R3 S R1 MgBr NiCl2(PPh3)4, D +R3 (CH2)n R2 (69% ¡À 88%) S 407 406 3. The7CH27CH=CH7CH=CH2 (C5) synthons and their homologues The readily available 2,4-dienes of the general formula 413 containing a functional group (for example, a hydroxy group) at the 1-position can serve as the sources of these synthons. The corresponding allyl bromide 414 or acetate 415 acts as an electro- phile when treated wGrignard reagents.376, 377 R1 R3 OH R1 R2 413 R3 R2MgBr PBr3 Br R1 (55% ¡À 65%) 414 V.The synthon approach to conjugated dienes R2MgBr, Li2CuCl4 Ac2O OAc R1 (62%) 415 R2 R1 Many reactions considered above could have been placed in this Section because versatile reagents can simultaneously be regarded as synthons if they contain fragments incorporated in the product molecule. Below we consider `synthon syntheses' not included inSynthetic methodologies for carbo-substituted conjugated dienes The reaction of pentadiene isomers with lithium followed by treatment with trimethylchlorosilane yields only one product, 1-trimethylsilylpenta-2,4-diene (416).378, 379 This compound reacts with C-electrophiles of various types.1. Li, TMEDA 2. Me3SiCl SiMe3 or 416OHR1 R1R2C O, TiCl4 R2 OR3R1 R1R2C(OR3)2 BF3 . Et2O R2 O RCOCl R 4. The =CH7CH=CH7CHO (C4) synthons Many compounds considered in Section III can be considered among the reagents capable of serving as sources of these synthons. Nevertheless, other versatile reagents for such `dienyla- tion' also do exist. For example, 1,4-heterodisubstituted buta- dienes 417 or 418 can be easily converted into organolithium derivatives 419 or 420, which are much more reactive towards aldehydes and ketones than the Wittig ± Horner reagents. Final hydrolysis gives rise to dienals 421, formed predominantly as the E,E-isomers 380 ± 382 (almost 100% for R2=H). BuLi OEt OEt Li Bu3Sn 419 417 1.R1R2C O 2. H3O+ (65% ±90%) Li Br BuLi OSiMe3 OSiMe3 420 418 R1 CHO R2 421 The dienals 377 can also be prepared using 1-methoxybut-1- en-3-yne (422) or its homologues.354, 383, 384 1. RCHO 2. LiAlH4 EtMgBr OMe 3. H3O+ BrMg OMe OMe 1. RCHO Cp2ZrHCl 2. AgClO4 422 ClCp2Zr R CHO 377 5. The7CH=CH7CH=CHCHO and 7CH=CH7CH=CHCO2R (C5) synthons Reagent 423, equivalent to a nucleophilic synthon of the given type, belongs to the class of thiol thiocarbamates. The products of its a-alkylation (424) undergo two [3,3]-heterosigmatropic re- arrangements giving ultimately linear dienes 425,385 which are precursors of the dienals 377. 761 R S S 1. LDA 2. RI D, [3,3] N S N S 423 424 R S S MeI D, [3,3] R S N N S 425 N O O R I R CHO 377 However, the most widely used synthons of this type are electrophilic reagents such as 5-bromopenta-2,4-dienal (426),386, 387 5-dimethylaminopenta-2,4-dienal (427) 388 and ethyl 5-(butyltelluro)penta-2,4-dienoate (428).389 Substitution of alkyl groups for the electronegative elements (Br, NMe2 and BuTe) in these compounds is accomplished by means of organometallic reagents.Br CHO 426 RZnBr, Pd(PPh3)4 R CHO (53% ± 89%) OH 1. ButLi 2. PhCHO RLi or RMgHal Br R R Ph (55% ± 77%) OH OH Ar Me2N CHO ArLi CHO (20% ± 64%) 427 R[Cu] R BuTe CO2Et CO2Et (82% ± 98%) 428 Yet another source of the synthons in question is represented by pyrylium salts 429.390, 391 The (Z,E)-alkadienals 430 formed initially in their alkylation are easily transformed into the E,E- isomers.R1 R1 R1 H+ R2Li CHO CHO R2 + R2 (37% ± 90%) (Z,E)-430 (E,E)-430 O 429A similar role is played by the potassium derivative of glutacondialdehyde (431), which can be easily converted into the dienal 377 upon protection of one of the functional groups, treatment with a Grignard reagent and acid hydrolysis.392 O CHO R 1. ButMe2SiCl 2. RMgBr O7K+ 3. H3O+ (55% ± 85%) 431 377 Furfural tosylhydrazone (432) can serve as a convenient synthon for the preparation of various 5-arylpenta-2,4-dienals 433.393762 ArMgHal (2.5 equiv.) N N 7N2 NHTs NH O O 432 Ar CHO Ar 433 (75% ± 80%) When made to react with alkylmagnesium halides, the same furfural 432 furnishes other products, namely, furan derivatives.3-Ethynyl-b-propiolactone (434) reacts with organomagne- sium compounds in the presence of copper(I) salts to give allenic acids 435, which can be isomerised to dienoic acids.394 KOH RMgBr, CuI, Me2S, THF R O CO2H (85%) 435 O 434 CO2H R Finally, a three-step procedure for the synthesis of methyl 6- hydroxyalka-2,4-dienoates 436 has been developed; in this method, pent-4-enoic acid (437) serves as the source of the diene fragment.395 TsI 1. 2 BuLi 2. R1R2C O Ts CO2H CO2H 437 Ts OH CO2H1. Me3SiCHN2 2. DBU R1 R2 CO2Me H R1 436 R2 6. The7C(=CH2)7CH=CH2 (C4) synthons The synthon approach has been implemented successfully for the preparation of branched 1,3-dienes with functional substituents at the 2-position.One approach is based on the allylic rearrangement of allylsilanes and allylboranes in the reactions with carbonyl compounds. a-Allene derivatives 438 and 439, similar to the allylic derivatives of silanes and boranes, react with carbonyl compounds at the sp-hybridised carbon atom yielding branched compounds 440, which contain a conjugated diene fragment.396, 397 In the case of the silyl derivative 438, not only aldehydes but also acetals can be used, resulting in the corresponding ethers. The effect of a chiral group present in the electrophilic reagent was studied; after removal of this group, the resulting secondary dienic alcohols had ee values of up to 96%.SiMe3 OH RCHO, TiCl4, CH2Cl2, 795 to778 8C (495%) 438 R B(OPri)2 RCHO, CH2Cl2, 0 8C 440 (53%±93%) 439 Branched allene silanes 441, which can be easily prepared from acetylene derivatives 442, behave in a similar way. They can be used to prepare 2,3-disubstituted butadienes 443.398 Me3SiCH2MgCl, LiCuBr2 R2CHO, TiCl4, SiMe3 CH2Cl2,778 8C C C R1 (65%) (56%) TsO R1 442 441 R1 R2 OH 443 A A Vasil'ev, E P Serebryakov Great synthetic capabilities are provided by the butene deriv- ative 444, containing benzotriazole and trimethylsilyl vicinal groups.399 After introduction of a desired fragment, the vicinal groups from the structures 445 and 446 can be eliminated by heating. NN N R RHal D R (54% ± 90%) N BuLi 445 SiMe3 N N N OH RCHO D N R (60% ± 74%) N R SiMe3 444 OH 440 446 SiMe3 The Grignard reagent 447, prepared from 2-chlorobuta-1,3- diene (chloroprene), can react with a number of electrophilic reagents, namely, a,b-unsaturated carbonyl compounds,400 alkyl halides 400, 401 (including various bromoalkanonitriles 401) and aryl iodides.402 This is a very simple and versatile approach to the synthesis of 2-substituted 1,3-dienes.In the absence of copper salts, the Grignard reagent 447 can add to ketones; however, this gives about 25% of the rearranged allene-type product.403 R3 R1 , O R1 R2 O R2 CuBr, Me2S, THF R3 (52% ±84%) Alk AlkHal, CuBr MgCl (30% ± 68%) (CH2)nCN 447 Br(CH2)nCN, CuBr (71% ± 98%) Ar ArI, Pd(PPh3)4 (40% ± 75%) 2-Substituted butadienes with a bulky substituent R1 can be prepared from 1,4-dichlorobut-2-yne (368) using hydrobora- tion.404 The mechanism of this reaction has been studied in detail.405 R12 B R2Li R12 BH Cl Cl Cl Cl 368 R1 R1 R12 R2B7 Li+ R2Li R1R2B (68% ± 85%) Cl Cl Cl 7.The7CH27C(=CH2)7CH=CH2 (C5) synthons a-Bromoisoprene (448) and organometallic derivatives prepared from it are typical examples of the title synthons. A convenient method has been proposed for the preparation of the correspond- ing Grignard reagent 449 in two simple steps without isolation of the bromoisoprene 448.34 The reaction of the Grignard reagent 449 with isovaleraldehyde results in ipsenol (291) (Scheme 11).1-Methylene-2-(trimethylsilyl)cyclobutane (450) can be used instead of the Grignard reagent 449; the compound 450 reacts with aldehydes with the allylic rearrangement to give a cyclo- butene derivative 451.406 Pyrolysis of this product furnishes the same diene derivative 291. Organotin compound 452 with an isoprene fragment reacts with allylic bromides in the presence of Lewis acids giving rise toSynthetic methodologies for carbo-substituted conjugated dienes Br BuiCHO Mg (excess) 1. Br2 2. NBS (64%) (38%) MgBr Br BrBr 448 449 BuiCHO, TiCl4, CH2Cl2, 0 8C HO SiMe3 450 branched 1,3-dienes.407 This method was tested in the synthesis of the corresponding terpenes 453 � myrcene (R=H) and b- farnesene (R=Me2C=CHCH2).ZnCl2, THF, 65 8C + R Me3Sn Br (50% ±94%) 452 R 453 In compound 454, the allylic chloromethyl group can serve as both an electrophilic and a nucleophilic centre, dending on the reaction conditions.408 The tosyl and trimethylsilyl vicinal sub- stituents act as a latent double bond, resulting in the vinyl fragment in the final products 455 and 456. Nu Nu Bu4NF Nu Ts Cl 455 SiMe3 HO R HO R Ts Bu4NF RCHO, Zn SiMe3 454 Ts 456 SiMe3 VI. Electrocyclic reactions 1. Thermolysis of cyclobutenes Two electrocyclic reactions resulting directly in conjugated dienes are known, namely, thermal disrotatory closure of conjugated hexatrienes accompanied by the formation of cyclohexadienes and conrotatory opening of cyclobutenes to give conjugated dienes, occurring at high temperatures.The use of the former reaction for preparative purposes is limited; some publications dealing with this topic are cited in a monograph.10 The latter reaction has found a more extensive synthetic application. Depending on the nature of substituents in the initial compound, the process can be performed either by heating cyclobutene derivatives having the general formula 457 409, 410 or by passing them through a heated Vigreux column at a reduced pressure.411 R2 R2 R1 R3 R1 R3 D R3 (76% ± 100%) R1 R3 R1 R2 R2 457 763 Scheme 11 HO D 291 (86%) 451 This method has been used in elegant syntheses of ipsenol 291;406, 412 in one of them, the initial cyclobutene derivative 451 was prepared from the corresponding selenide 458.MeSe 1. MeI, AgBF4 2. KH D OH OH (100%) (77%) 451 458 OH 291 Unlike primary alcohols 459, 3-formylcyclobutenes 460 undergo ring opening even at low temperatures.413 The low- temperature Swern oxidation of the alcohol 459 affords the expected Z,E-isomer 377 (conrotatory ring opening), which is readily transformed into the E,E-isomer when allowed to stand. Oxidation of the cyclobutenol 459 with pyridinium chlorochro- mate at room temperature directly affords (E,E)-377. CHO CHO CH2OH CHO [O] R R R 459 460 R (E,E)-377 (Z,E)-377 Other electrocyclic reactions are suitable for the synthesis of conjugated dienes only in those cases where the compounds under study contain additional double bonds or functional groups capable of being readily detached to give a double bond. 2.Sigmatropic [3,3]-rearrangements Sigmatropic [3,3]-rearrangements have been often used for the preparation of conjugated dienes; the variant used most often is the Claisen rearrangement consisting in the thermal transforma- tion of vinyl allyl ethers into D4-unsaturated functional derivatives (see a review 414). An example is provided by the syntheses of diene derivatives 461 and 462 by the Claisen rearrangement of vinyl ethers of divinylmethanols 463 and 464, respectively.415 ± 417 In the latter case, triethyl orthoacetate serves as an equivalent of the vinyl ether. O OH CH2 CHOEt, Hg(OAc)2 CHO 461 463764 OH 465 OH A similar reaction involving buta-2,3-dien-1-ol 465 results in branched conjugated dienes; this was employed in two equivalent syntheses of ipsenol 291 (Scheme 12).418, 419 Vinylalkynylmethanols 466 can be used to prepare conjugated enynoates 467 and enynoic acids 468, which are the closest precursors of dienes.420, 421 MeC(OMe)3 C RC 466 OH Ac2O 466 C RCC RC O The Claisen rearrangement of propargyl vinyl ethers with the general formula 469 gives rise to b-oxoallene derivatives 470, which readily isomerise into conjugated diene derivatives 471.This process became exceptionally important after it had proved to be useful for the synthesis of vitamins A and E and some valuable terpenoids. R4 O R1 R2 R3 469 R2 R1 471 In many cases, there is no need to synthesise these propargyl vinyl ethers because they are formed in situ under the reaction conditions from the corresponding propargyl alcohols 237. Ini- tially, this was attained by either pyrolysis of their acetoacetates or heating with ethyl acetoacetate or diethyl malonate (the product yields were moderate in this case).422 ± 424 Later, these reactions D CH2 CHOEt O MeC(OMe)3 O D OMe OMe MeC(OEt)3 H+, D O 464 OEt CO2Et 462 CO2Me RC H+, D C467 1.LDA 2. ButMe2SiCl O O70 8C CO2H C RC 468 OSiMe2But R4 R4 O O R1 R1 R3 R3 R2 470 R2 O R3 R4 A A Vasil'ev, E P Serebryakov Scheme 12 BuiMgBr O HO DIBAH 291 O were substantially updated; more precisely, isopropenyl methyl ether was taken instead of ethyl acetoacetate 425 and orthoesters of carboxylic acids were used in place of diethyl malonate.426 ± 428 The yields of products 472 and 473 depend markedly on the pattern of substitution in the initial compounds, being very high under favourable circumstances.OH R1 R2 237 CH2=C(Me)OMe MeC(OR3)3 R2 R2 R1 R1 CO2R3 O Base Base R2 O R2 CO2R3 R1 R1 473 472 The reaction of propargyl alcohol with ynamine 474 occurs in a similar way, giving rise to N,N-diethylamides of allenic 475 and dienoic 476 acids.429 R3 NEt2 OH+ R3C CNEt2 R1C C O (60% ± 85%) R2 474 R1C C R2R3 R3 NEt2 R2 R2 ButOK NEt2 O O R1 R1 476 475 Acetylenic alcohol 477 containing a butenyl fragment reacts with isopropenyl methyl ether to give allene ketone 478.On treatment with an alkali, this product isomerises in a conventional way to give ketone 479 with a diene fragment.430 However, on prolonged heating, the ketone 478 can undergo one more [3,3]- sigmatropic rearrangement accompanied by the formation of intermediate 480 and, further, the conjugated dienone 481. OMe, TsOH, 100 8C (57%) O OH 477Synthetic methodologies for carbo-substituted conjugated dienes KOH O O 479 O O 140 8C 478 480 481 b-Allene acids obtained from ethers of propargyl alcohols upon [3,3]-sigmatropic rearrangement can also be used in another synthesis of dienes; decarboxylation of allene acid 482 is accom- panied by the migration of the b,g-double bond, resulting in the formation of conjugated 3-methylhexa-2,4-diene (483).431 OH O OSiMe3 O 1.[3,3] 2. H2O O D 1. LDA 2. Me3SiCl O (64% ± 100%) 483 482 The Claisen rearrangement can also involve allyl vinyl ethers of a more complicated structure. Thus allenyl phenyl sulfoxide 484 reacts with allyl alcohol derivatives to give adduct 485 with a suitable mutual arrangement of the double bonds.432 The driving force of the rearrangement is transformation of the b-alkoxyvinyl sulfoxide fragment into the a-sulfinyl ketone fragment. Conju- gated diene is formed in the subsequent thermal elimination of the sulfoxide group from compound 486. S(O)Ph R3 HO D O R3 + (48% ± 67%) R2 R1 Ph(O)S R2 R1 484 485 S(O)Ph R2 (58% ± 85%) R1 O R3 R3 O R1 R2 486 a-Phenylsulfinyl-D4-unsaturated derivatives can be synthes- ised from allylic alcohols using yet another reagent, viz., triethyl phenylsulfinylacetate (487).Elimination of the sulfoxide group occurs simultaneously with the sigmatropic rearrangement.433 This method allows one-step synthesis of not only linear (488) but also exocyclic (489) conjugated dienes. Ph(O)S OH CO2Et CO2Et D (79%) 488 C(OEt)3 OH SPh CO2Et TBDMSO OTBDMS O 487 (89%) TBDMSO OTBDMS 489 TBDMS is tert-butyldimethylsilyl. Branched dienes of the type 490 are produced in the allene version of the oxy-Cope rearrangement of alcohols 491.434, 435 765 R2 R1 R2 R2 R1 D R1 R3 R3 R3 (58% ± 70%) HO HO O 490 491 The oxy-Cope [3,3]-rearrangement of 1,5-enyn-4-ols 492 results initially in a compound with incompletely conjugated a,b,d,e-dienone fragment (493), which can be subsequently iso- merised to give compound 494 with a conjugated a,b,g,d-dienone fragment.436 It is convenient to combine isomerisation with the rearrangement by adding inorganic halogen compounds (HCl, HBr, I2, NH4 Br) to the reaction mixture.437 The method is applicable for the synthesis of pseudoionone.[3,3], NMP, 165 8C C CH OH 492 O O 494 493 The hydroxy and trimethylsilyl vicinal groups can serve as a source of the double bond (the Peterson reaction). If the molecule of the initial compound 495 is constructed in such a way that, after rearrangement, these substituents in the product 496 occur in the required positions, this may provide a route to the conjugated dienes 67.438 The initial compounds 496 were resolved into individual erythro- and threo-diastereomers, which were convrted into individual E,E- and Z,E-isomers of the dienoates 67 by varying the elimination method.O OSiMe3 HO Me3SiO 1. D 2.CH2N2 1. LHMDS (3 equiv.) 2. Me3SiCl O O R R Me3Si Me3Si 495 O HO CO2Me OMe R 67 R Me3Si 496 Compound 498, which contains a trimethylsilyl group and an epoxide ring on different sides of the double bond, is unstable and is easily converted into hydroxy acid 499 with a diene fragment. The compound 498 is formed in the [3,3]-sigmatropic rearrange- ment of the precursor 500,439 which is, in turn, prepared from 2-methylpent-2-enol (497). 1.O OH MgBrSiMe3 2. EtCOCl O 497 OTBDMS O O O [3,3] 1. LDA 2. TBDMS± OTf O O 500 Me3Si Me3SiA A Vasil'ev, E P Serebryakov 766 OTBDMS O VII. Methods based on migration or reduction of multiple bonds O H2O OH OH (90%) O 499 498 Me3Si The strategy for the synthesis of dienes making use of migration and reduction is attractive because a conjugated diene system is formed at the final stage of the synthesis in a molecule which already has the required carbon skeleton and functional groups in the required positions. 3. The Wittig sigmatropic [2,3]-rearrangement 1. The migration of multiple bonds The Wittig rearrangement of carbanions derived from compounds 501 and 502 has also been used to synthesise conjugated dienes of the type 413 and 503.440, 441 In both cases, the second double bond was formed upon elimination of the silyl and hydroxy groups from the corresponding alcohols 504 and 505 (the Peterson reaction); in the former case, zirconium served as the counter-cation to ensure the required configuration of alcohol 504 (cf.data of Ref. 442). Cl OPri O O Cp OPri Zr LDA, Cp2ZrCl2 Cp O O R SiMe3 R SiMe3 501 OPri O R 3 steps On treatment with bases or some transition metal complexes, non- conjugated dienes are converted into thermodnamically more favourable conjugated isomers. Generally, this reaction is not regioselective, therefore, it has a preparative value for hydro- carbon substrates only in a few cases, for example, in the synthesis of cycloocta-1,3-diene by heating cycloocta-1,5-diene inDMSOin the presence of potassium tert-butoxide.444 However, the situation changes dramatically if the molecule contains a carbonyl group; in this case, the migration of multiple bonds results in an individual product.Examples of isomerisation of 2,5-dien-1-ones and b-allene ketones and esters into 2,4-diene derivatives are consid- ered in Section VI. An organocopper route to the dienes 285 from b-allene esters 509 is also known. Unfortunately, this route is suitable only in the case where sulfonates of secondary acetylenic alcohols and ethyl acetate enolate are used as the initial com- pounds.445 SiMe3 HO (66% ± 72%) OH CO2Et O7Cu+ MeO2SO R 413 504 + CH C 509 R R R R SiMe3 SiMe3 OEt (34% ± 76%) [2,3] R O CO2Et HO 285 SiMe3 505 502 SiMe3 SiMe3 R KH (Z,E)-503 On treatment with strong bases, homoconjugated 1,4-dienes 510 form a delocalised carbanion,which is attacked regioselectively by specific reagents, for example, BF(OMe)2 (511).446 The subse- quent treatment of these products with hydrogen peroxide yields primary conjugated dienic alcohols (E,E)-413 or (Z,E)-413.Depending on the configuration of the D4-double bond in the initial 1,4-diene 510 and the type of base, definite stereoisomers 413 are produced. BF3 . Et2O R SiMe3 (E,E)-503 1. BusLi 2. BF(OMe)2 (511) 3.H2O2 , NaOH OH R 4. Ene reaction (86%) (E,E)-413 R 510 R 1. BusLi ±ButOK 2. BF(OMe)2 (511) 3. H2O2 , NaOH (96%) OH (Z,E)-413 Ene reactions take place between the dienes 506 and reactive aldehydes and afford dienic alcohols 507.443 The formation of the compounds 507, however, can be explained in terms of the Prins reaction mechanism, which is known to be accelerated by acids. On heating without a catalyst, the above-mentioned reactants enter into a [4+2]-cycloaddition reaction giving rise to dihydro- pyrans 508. R1 +R2CHO 506 The diene system of alka-2,4-dienoates 512 can be decon- jugated from the carbonyl group by treatment with LDA in THF in the presence of HMPTA162, 447 or N,N0-dimethyl-N,N0-(1,3- propylene)urea (DMPU).163 It is of interest that the dienes (2E,4Z)-512 are converted into (3E,5E)-513, while (2E,4E)-512 are converted into (3E,5Z)-513.447 R1 R1 H SnCl2 (Cat.) O OH (41% ± 62%) R2 507 R2 R CO2Alk CO2Alk R2 R 1.LDA, THF, (Me2N)3P=O 2. H+(498%) D (3E,5E)-513 (2E,4Z)-512 (74%) O R1 508 CO2Alk R CO2Alk R (476%) R1=H, Me; R2=CCl3, CO2Bu. (2E,4E)-512 (3E,5Z)-513Synthetic methodologies for carbo-substituted conjugated dienes Thus methyl hepta-3,5-dienoate (514) can be prepared from incompletely conjugated dienoate 515 on prolonged keeping (12 h) in the presence of Et3N.448 Divinylmethanols and their derivatives are readily isomerised on treatment with acidic reagents to give conjugated diene derivatives.For example, the allylic rearrangement of dialkenyl- methyl acetates 516 catalysed by palladium complexes takes place with retention of the configuration of one double bond.449 It is evident that in the case of non-symmetrical substrates (R1 6àR2), the reaction gives rise to a mixture of regioisomers 517. OAc R1 (E,E )-516 OAc R2 R1 (Z,Z)-516 The regiodirectivity of this type of rearrangement can be controlled by introducing a heteroatom into the b-position of one of the vicinal groups. For example, readily available thio derivatives 518 can be smoothly converted into dienals 430.450 R1 O R1 The rearrangement of a-silylallenes 519 into dienes 242 is also regioselective and, in some cases, stereoselective.451 Table 2. Isomerisation of acetylene derivatives into conjugated dienes.XC(O)But C(O)Cy C(O)Ph CO2Bn C(O)Ph C(O)NHPh C(O)Me CO2Me CO2Me CO2All C(O)N(All)Bn CO2C6F5 CO2C6F5 (CF2)8Cl (CF2)8Cl (CF2)4Cl CF3 C2H2, CO, [Ni] Et3N CO2Me Cl 515 CO2Me 514 OAc PdCl2(MeCN)2 R1(2) R2(1) R2 (E,E )-517 OAc PdCl2(MeCN)2 R1(2) R2(1) (E,Z)-517 SBut SBut R1CHO R2Li O R1 SBut H3O+ CHO (60% ± R2 OH 98%) R2 430 518 Catalyst RMe HEt Me Et Me C5H11 Et Et Me Me Bu HC:C(CH2)2 Me Me Me (CH2)5OH Pd(OAc)2, Ph3P Pd(OAc)2, dppb Pd(OAc)2, dppb Ph3P Ph3P Ph3P, AcOH Ph3P Ph3P Bu3P Bu3P Bu3P Ph3P Ph3P Ph3P Pd(dba)2 , Ph3P, AcOH Pd(dba)2 , Ph3P, AcOH Pd(dba)2 , Ph3P, AcOH R1 R2 BF3, AcOH CH2Cl2,778 8C Me3Si 519 In the last decade, an absolutely new method for the synthesis of conjugated diene derivatives 520 was discovered; this is isomer- isation of alkynes 521, in which the triple bond is located close to an electron-withdrawing group.In the first publication, it was noted that this isomerisation is induced by various palladium catalysts.452 Similar results have been obtained with ruthenium or iridium complexes,453, 454 which have also proved suitable for isomerisation of cyclohexene derivatives 522 into the correspond- ing trienes 523. R2 O (490%) R1 521 R1 O R2 522 Later, it has been found that the crucial role in the isomer- isation is played by the phosphine ligands contained in the catalysts.Indeed, prolonged refluxing of various activated alkynes in toluene in the presence of triphenylphosphine alone also furnishes the diene isomers of type 520 in very good yields.455 Later, factors influencing the reaction route have been studied in detail X R R and recommendations for methods have been given. The data are summarised in Table 2. For isomerisation of less reactive amides of alk-2-ynoic acids (X=C(O)NR2) to take place, the process should be carried out at higher temperatures (refluxing in xylene) in the presence of acetic acid. Conversely, more reactive acetylenic ketones [X=C(O)Alk and C(O)Ar] are isomerised even at room temperature on treatment with Ph3P. For a similar transformation of esters (X=CO2R) and N,N-dialkylamides [X=C(O)NR2], tributylphosphine was found to be the reagent of choice.456 Reaction conditions t /h T /8C 22 210.75 64 14 34 35 24 48 2455 100 100 100 110 80 140 25 25 110 25 110 50 50 110 110 110 110 24 12 12 12 767 R1 (80% ± 100%) R2 242 O R2 R1 520 O R1 R2 (82% ± 94%) 523 X , Ref.Yields (%) 452 452 452 455 455 455 456 456 456 456 456 457, 458 457, 458 459 459 459 459 73 73 82 75 83 84 840 82 80 60 96 760 82 75 85768 However, pentafluorophenyl esters (X=CO2C6F5), which exhibit strong electron-withdrawing properties, can be isomerised under milder conditions, namely, at 50 8C and in the presence of Ph3P; if the molecule contains other acetylenic groups, they remain intact.457, 458 In this connection, an interesting example is the selective transformation of diacetylene 524 into dienyne 525.PPh3, toluene, 50 8C, 5 h MeO2C CO2C6F5 524 MeO2C CO2C6F5 525 Among acetylene derivatives able to undergo isomerisation, `non-carbonyl' polyfluoroalkyl compounds proved to be an interesting type of substrates 459 (see Table 2). The transformation of these compounds into dienes requires the presence of palladium catalysts and acetic acid, apart from phosphines. Two processes, similar at the first glance, are known to form a diene system (for example, in compounds 526 and 520) from dialkynylmethanols 527 460 and a-oxo-a 0-hydroxyacetylenes 528.461 Nevertheless, for balance, the former reaction requires involvement of an oxidising agent, while the latter, of a reducing agent.O OH Pd(OAc)2, PPh3 , benzene, 80 8C R2 (62% ± 73%) R2 R1 R1 527 526 HO O R2 R1 Ph3P, 20 8C (83% ± 86%) R2 O R1 528 520 On treatment with superbases, for example, lithium amino- ethylamide, internal alkynes are transformed into terminal iso- mers. It has been shown recently that this process can also be applied successfully to diacetylenes 529. Since terminal diynes are unstable, it has been suggested that the lithium derivatives formed initially should be quenched immediately with electrophilic reagents; this provides a pathway to diacetylenic alcohols 530 and 531, which are potential precursors of the corresponding conjugated dienes.462 ± 464(CH2)nMe H2N(CH2)2NHLi Me(H2C)n 529 Li Me(H2C)2n+1 R1 R1R2C O Me(H2C)2n+1 R2 OH 530 R3 O Me(H2C)2n+1 R3 531 HO 2.Partial reduction of a triple bond in enynes and 1,3-diynes Incomplete reduction of the conjugated enynes 232 and diynes 532 to the dienes 533 and 55 has been considered comprehensively in the literature;8, 10, 12 therefore, only the basic principles are dis- cussed here. The transformation of a triple bond into a Z-double bond in these systems is accomplished using the same techniques as employed for monoacetylenes. Hydrogenation in the presence of Lindlar-type catalysts ensures, in principle, the required trans- formation (see, for example, Ref.465); however, the presence of the second, conjugated bond activates the substrate molecule, resulting in a lower selectivity. Better results are obtained when A A Vasil'ev, E P Serebryakov hydrogen carriers are used; for this purpose, the substrate is refluxed in ethanol-containing solvents with activated zinc.466 ± 468 R3 R2 R2 R1 R1 R3 (E,Z)-533 232 R1 R2 R1 R2 (E,Z)-55 532 However, hydroboration of the triple bond followed by protolysis of the resulting vinylborane has won the greatest recognition. The hydroborating reagent of choice is di(sec-iso- amyl)borane (sia)2BH, which is reactive, under the reaction conditions, only towards triple bonds but does not affect double bonds. This method was used to transform enynes and diacety- lenes into insect pheromones containing a conjugated diene system in which either one or both double bonds have Z-config- uration.246, 420, 469 Unfortunately, known methods of reduction of alkynes into E-alkenes (sodium in liquid ammonia, LiAlH4 with heating) are inapplicable to the reduction of enynes and diynes, because in this case reactions give complex mixtures of products.Exceptions are enynic alcohol 534 26, 470 ± 472 and diynic alcohol 535 (n= 0, 1),473 ± 475 which are hydrogenated to dienic 536 and enynic 537 alcohols, respectively. In the compound 535, the triple bond closest to the hydroxy group is reduced fairly smoothly to the E-double bond on treatment with LiAlH4 in ether or THF. Hydrogenation products, namely, conjugated E-enynic alcohols 537, can further be converted into E,Z-dienic alcohols 538.475 In the case of propargyl compounds, the desired effect is also attained by transforming the alcohol into alcoholate with the subsequent addition of DIBAH, the reduction taking place within the ate-complex formed.473 It is noteworthy that careful bromi- nation of the reaction mixture carried out instead of the final hydrolysis in the reduction of the enynic alcohols 534 results in bromo-containing dienic alcohols 539 being formed as the major products; in this case, the configuration of double bonds is disturbed.476 R2 H+ OH R1 R4 R2 LiAlH4 536 R3 R2 OHR4 R3 R1 534 OH R1 Br2 R4 Br R3 539 LiAlH4 R1 (CHR2)nOH (50% ± 86%) 535 R1 (CHR2)nOH (CHR2)nOH 537 538 R1 n=0, 1.Enynonitriles 540 can undergo similar reduction, the cyano group in products 541 being stable against lithium aluminium hydride under the reaction conditions.477 R1 CN R1 CN LiAlH4 , 760 to 0 8C (75% ± 85%) R2 541 540 R2 VIII. Miscellaneous methods The methods discussed in this Section could be assigned, if desired, to one or another type of reactions considered in the preceding Sections. However, these examples are usually singleSynthetic methodologies for carbo-substituted conjugated dienes and have substantial differences from the typical processes; there- fore, we consider them separately. The introduction of double bonds by elimination of vicinal hydrogen atoms finds only limited use.The high-temperature catalytic dehydrogenation occurs selectively only in the prepara- tion of simple dienes, for example, butadiene from butane and butenes or isoprene from isopentane. Of more complex examples, one should note the transformation of a,b-unsaturated oxo compounds 542 (X=Ph, OEt) into the corresponding dienes 543 on treatment with stoichiometric amounts of Na2PdCl4 as an oxidant.478 The yields are moderate in all cases. This method has been used to prepare steroidal dienones with fragment 544 from the corresponding enones 545. O O Na2PdCl4, D (*50%) X X 543 542 O O 544 545 The tetrahydropyranyl derivatives of a-dienic alcohols 546 can react with a number of C-, N-, Si- or Sn-nucleophiles giving rise to dienes 547 with a different arrangement of double bonds.479 The methoxymethyl derivatives react in a similar way; however, the reaction stereoselectivity is lower.R3Li R1 Cl 1. Mg, 10% ZnCl2 2. R1R2C=O 3. DHP (62% ±97%) (74% ± 93%) R2 OTHP 546 R3 R1 547 R2 DHP is dihydropyran; R3=Alk, NEt2, SiMe2Ph, SnMe3. The methoxymethyl ether of the enol 548 can be introduced, after lithiation, into reactions with aldehydes.480 The resulting dienes 549 contain an oxygen function at the diene system. The presence of this function is necessary for this reaction to take place (chelate control). OH MeO O O MeO 1. BusLi 2. RCHO (44% ±70%) R 549 548 The tertiary analogues of cinnamic alcohols 550 are dehy- drated under reaction conditions and the resulting diene is formylated giving rise to dienals 551.481 Me2NCHO, R R OH POCl3 (68% ± 94%) Ar Ar 550 R CHO Ar 551 Direct acylation of conjugated dienes 552 has remained problematic until recently.However, development of reactive cationoid reagents has enabled the synthesis of dienones 553 and 554 in reasonable yields.482 ± 484 769 RC(O)OSO2F or O R4 R2 + RC(O)SMe2BF¡4 Alk R1 (33% ±80%) R4 R2 R3 553 R1 O R4 R2 Me2S, BF3 , (CF3CO)2O R3 552 (21% ± 39%) R1 CF3 R3 554 The use of a-allenyl alcohols 555 in the synthesis of alkenes according to the Murahashi method affords the conjugated dienes 242.485 OH + MeLi, CuI Ph + N PBu3 Me R1 I7 555 + R2 O PBu3 R2Li, CuI I7 (20% ± 55%) R1 R1 242 The carbocationic rearrangement of cyclopropylalcohols can proceed smoothly under appropriate conditions to yield homo- allyl derivatives. In the case of cyclopropyl(alkenyl)methanols 556, conjugated diene system 557 is produced.486 ± 488 Me3SiBr, ZnBr2 , CH2Cl2,720 8C R R Br (95%) OH 557 556 A very simple method for the synthesis of dienes consists in heating of aliphatic a-branched ketones 558 in DMSO with potassium tert-butoxide,489 DMSO serving as the source of a new carbon atom in the molecule of product 559.Unfortunately, in many instances, these reactions give mixtures of isomers. R1 R1 R2 R2 DMSO, ButOK, 80 ± 140 8C (49% ±73%) O Me 559 558 Atwo-step method for the preparation of dienylstannanes 560 from acetylenic carbonyl compounds 561 is documented.490 Transmetallation provides a nucleophilic diene equivalent.O R1 Ph3P CH2 R1 (Me3Sn)2 , Pd(PPh3)4 R2 (48% ± 88%) R2 O Me3Sn 561 1. RLi 2. E+ R1 R1 E R2 R2 Me3Sn 560 Diketene 562 exposed toUVradiation in a mixture of propan- 2-ol and acetone gives adduct 563, which is a convenient precursor of conjugated dienoate 564.491 OH HCl, EtOH O O PriOH±Me2CO, hn (63%) CO2Et O O 564 563 562770 Oxidation of o-quinone, pyrocatechol or phenol with am- moniacal copper oxide in pyridine results in cis,cis-muconic acid mononitrile (565).492 In the case of substituted compounds, for example, 4-tert-butyl-o-quinone, a mixture of regioisomers is formed.OOOH CuO ±NH3, O2, Py, 0 8C (40% ±70%) OH OH The adduct 566 of itaconic anhydride with cyclopentadiene was converted into compound 567. Pyrolysis of this compound liberates the methylene group, which had initially been protected; this group is found to be conjugated with a new vinyl group,493 which finally results in ipsenol 291. O O O 6 steps 566 The coupling of two carbonyl compounds by means of the McMurry titanium reagent results in alkenes; if one compound is unsaturated, the expected product is a diene. Thus intramolecular cyclisation of dialdehyde 568 results in diterpene hydrocarbon 569 difficult to prepare by other methods.494 O O [Ti] (20%) 568 Table 3. Brief characterisation of the most important methods for the preparation of conjugated dienes (from the bibliography of this review).Reactions Elimination Reduction of enynes and diynes Olefination of carbonyl compounds condensation Wittig reactions other reactions Cross-coupling Miscellaneous organometallic reactions Electrocyclic reactions Miscellaneous methods A A Vasil'ev, E P Serebryakov IX. Conclusion CO2H CN 565 OH 450 8C HO (100%) 567 291 One cannot but note the great diversity of the methods for synthesis of conjugated dienes considered in this review. Never- theless, despite the wide range of choice, when selecting the method of synthesis of a target object, a researcher would be guided, first of all, by its reliability and simplicity as well as the availability and cost of initial compounds and, strange though it may seem, by the reputation and popularity of the method.Table 3 presents a comparative analysis of the methods consid- ered. It is fairly difficult to evaluate the true `rating' of each of them because this requires analysis of a different type of literature, namely, the use of these methods in target-directed syntheses of intricate structures in which they are often used only as an episode in a multistep protocol. The more reliable the method, the less the attention researchers draw to it! The popularity of a method can be judged to some extent by the presence of reviews devoted to it, or (less reliably) by mentions of this method in textbooks. When methods' reputations are equivalent, when planning the synthesis, one should analyse first of all the possibility of perform- ing the task on the basis of standard reactions, viz., elimination (the formation of double bonds with a preformed carbon skel- eton), olefination of carbonyl compounds (the C1+C3 pattern of diene formation with formation of a double bond) and, more rarely, cross-coupling (the C2+C2 pattern of diene formation with the formation of the central single bond, or the C4+sub- stituent pattern). Each of these general approaches has its own limitations.In particular, alkenyl halides and alkenyl derivatives of elements used in cross-coupling are rather expensive and their synthesis requires high skills. This and other reasons stimulate the search for new, different reactions which would provide an absolutely untrivial approach to the synthesis of conjugated dienes. The progress achieved in the alkene ± alkyne metathesis over the last three years can serve as an example.This method, as well as new organometallic reactions, undoubtedly hold great promise because it requires rather simple initial compounds and commercially available metal complex catalysts. 569 Features Stereohemical outcome The number of references (of these, reviews) The range of yields (%) 92 (8) 36 ± 97 effective methods are expensive depends on the method and the substrate 15 50 ± 86 good for the synthesis of Z-isomers, inexpensive method poorer for E-isomers 118 (9) 35 (6) 24 ± 98 high for linear E-dienes 58 (3) 27 ± 98 25 23 ± 94 used for the synthesis of a-functionalised dienes adjustment of details for a particular synthesis is often required more reliable than the Wittig olefination depends on the method and the substrate but is controllable depends on the method and the substrate 104 (16) 30 ± 93 high cost of the reactants very high; determined by the reactant structure 48 (1) 21 ± 99 some procedures are very effective depends on the method and the substrate 35 (3) 25 ± 90 industrially feasible high 75 20 ± 95 7 depends on the method and the substrateSynthetic methodologies for carbo-substituted conjugated dienes References 1.Z Rappoport (Ed.) The Chemistry of Dienes and Polyenes Vol. I (New York: Wiley-Interscience, 1997) 2.Z Rappoport (Ed.) The Chemistry of Dienes and Polyenes Vol. II (Chichester: Wiley-Interscience, 2000) 3. M Sodeoka, M Shibasaki Synthesis 643 (1993) 4. A A Vasil'ev, A L Vlasyuk, G D Gamalevich, E P Serebryakov Bioorg. Med. Chem. 4 389 (1996) 5. Houben-Weyl. Methoden der Organische Chemie Vol. V/1c, Pt 3 (Stuttgart: Thieme, 1970) p. 2 6. S Coffey (Ed.) Rodd's Chemistry of Carbon Compounds Vol. I (Amsterdam: Elsevier, 1964) p. 438 7. C A Buehler, D E Pearson Survey of Organic Synthesis (New York: Wiley-Interscience, 1970) 8. G Pattenden, in Comprehensive Organic Chemistry Vol. 1 (Eds D Barton, W D Ollis) (Oxford: Pergamon Press, 1979) p. 171 9. J F Normant, A Alexakis, in Current Trends in Organic Synthesis (Ed.H Nozaki) (Oxford: Pergamon Press, 1983) 10. R C Larock Comprehensive Organic Transformations. A Guide to Functional Group Preparations (New York: VCH, 1989) 11. K Mori, in The Total Synthesis of Natural Products Vol. 9 (New York: Wiley, 1992) p. 37 12. G Mehta, H S P Rao, in The Chemistry of Dienes and Polyenes Vol. I (Ed. Z Rappoport) (New York: Wiley-Interscience, 1997) p. 359 13. L A Paquette, M Gugelchuk, M L McLaughlin J. Org. Chem. 52 4732 (1987) 14. B M Trost,M Lautens, B Peterson Tetrahedron Lett. 24 4525 (1983) 15. Y Naruta, N Nagai, Y Arita, K Maruyama J. Org. Chem. 52 3956 (1987) 16. P G McDougal, J G Rico J. Org. Chem. 52 4819 (1987) 17. J Tsuji, T Yamakwa,M Kaito, T Mandai Tetrahedron Lett.2075 (1978) 18. F M Hauser,R Tommasi, P Hewawasam,Y S Rho J. Org. Chem. 53 4886 (1988) 19. H Matsushita, E Negishi J. Org. Chem. 47 4161 (1982) 20. M Suzuki, Y Oda, R Noyori J. Am. Chem. Soc. 101 1623 (1979) 21. S-K Kang, D-C Park, C-H Park, R-K Hong Tetrahedron Lett. 36 405 (1995) 22. J B Hendrickson,M A Walker, A Varvak,M S Hussoin Synlett 661 (1996) 23. A S Demir Tetrahedron 57 227 (2001) 24. S Tanaka,A Yasuda,H Yamamoto,H Nozaki J. Am. Chem. Soc. 97 3253 (1975) 25. A Yasuda, S Tanaka,H Yamamoto,N Nozaki Bull. Chem. Soc. Jpn. 521752 (1979) 26. T Laird, W D Ollis, I O Sutherland J. Chem. Soc., Perkin Trans. 1 2033 (1980) 27. G Decodts, G Dressaire, Y Langlois Synthesis 510 (1979) 28. H M Walborsky, H H WuÈ st J. Am. Chem.Soc. 104 5807 (1982) 29. A Claesson, C Bogentoft Acta Chem. Scand. 26 2540 (1972) 30. G Solladie', J Hutt J. Org. Chem. 52 3560 (1987) 31. G Solladie', A Girardin, P Metra Tetrahedron Lett. 29 209 (1988) 32. G Solladie', A Girardin Tetrahedron Lett. 29 213 (1988) 33. H R Pfaendler, F K Maier, S Klar J. Am. Chem. Soc. 108 1338 (1986) 34. A G Martinez, J L M Contelles Synthesis 742 (1982) 35. L A Paquette, J Dressel, K L Chasey J. Am. Chem. Soc. 108 512 (1986) 36. L Engman, S E BystroÈ m J. Org. Chem. 50 3170 (1985) 37. L Crombie, L J Rainbow J. Chem. Soc., Perkin Trans. 1 673 (1994) 38. B Barlaam, J Boivin, S Z Zard Tetrahedron Lett. 34 1023 (1993) 39. D J Ager Synthesis 384 (1984) 40. P A Brown, R V Bonnert, P R Jenkins, M R Selim Tetrahedron Lett.28 693 (1987) 41. J A Prieto, G L Larson, R Berrios, A Santiago Synth. Commun. 18 1385 (1988) 42. P A Brown, R V Bonnert, P R Jenkins, N J Lawrence, M R Selim J. Chem. Soc., Perkin Trans. 1 1893 (1991) 43. D J S Tsai, D S Matteson Tetrahedron Lett. 22 2751 (1981) 44. D A Smith, K H Houk Tetrahedron Lett. 32 1549 (1991) 45. Y Ikeda, J Ukai, N Ikeda, H Yamamoto Tetrahedron 43 731 (1987) 771 46. S L Hartzell, D F Sullivan, M W Rathke Tetrahedron Lett. 1403 (1974) 47. L Strekowsky,M Visnick, M A Battiste Tetrahedron Lett. 25 5603 (1984) 48. K Shimoji, H Taguchi, K Oshima, H Yamamoto J. Am. Chem. Soc. 96 1620 (1974) 49. Y Yamakado,M Ishiguro, N Ikeda, H Yamamoto J. Am. Chem. Soc. 103 5568 (1981) 50. A G Angoh, D L J Clive J.Chem. Soc., Chem. Commun. 534 (1984) 51. R Angell, P J Parsons, A Naylor, E Tyrrell Synlett 599 (1992) 52. I Fleming, I T Morgan, A K Sarkar J. Chem. Soc., Chem. Commun. 1575 (1990) 53. MOchiai, S Tada, K Sumi, E Fujita Tetrahedron Lett. 23 2205 (1982) 54. D L G Clive Tetrahedron 34 1049 (1978) 55. K C Nicolaou, N A Petasis Selenium in Natural Products Synthesis (Philadelphia: CIS, 1984) p. 66 56. H J Reich Org. React. 44 4 (1993) 57. A Krief, in Comprehensive Organometallic Chemistry II Vol. 11 (Oxford: Pergamon Press, 1995) p. 515 58. C C Silveira, M A Araujo, E J Lenardao, A L Braga, M J Dabdoub Synthesis 1305 (1995) 59. J Tsuji, K Masaoka, T Takahashi Tetrahedron Lett. 2267 (1977) 60. L Lebarillier, F Outurquin, C Paulimier Tetrahedron 56 7483 (2000) 61.P J Kocienski, G Cernigliaro, G Feldstein J. Org. Chem. 42 353 (1997) 62. K Mori, T Nukada, T Ebata Tetrahedron 37 1343 (1981) 63. K Ando, T Yamada, Y Takaishi,M Shibuya Heterocycles 29 1023 (1989) 64. A A Vasil'ev, L Engman, E P Serebryakov J. Chem. Res. (S) 706 (1998) 65. A A Vasil'ev, L Engman, E P Serebryakov Acta Chem. Scand. 53 611 (1999) 66. H J Reich J. Org. Chem. 40 2570 (1975) 67. D L J Clive, G Chittattu, N J Curtis, S M Menchen J. Chem. Soc., Chem. Commun. 770 (1978) 68. H J Reich, I L Reich, S Wollowitz J. Am. Chem. Soc. 100 5981 (1978) 69. T R Hoye, A J Caruso Tetrahedron Lett. 4611 (1978) 70. H Ishibashi, H Komatsu, K Maruyama,M Ikeda Tetrahedron Lett. 26 5791 (1985) 71. Y Tamura, H-D Choi, H Maeda, H Ishibashi Tetrahedron Lett.22 1343 (1981) 72. T Mandai, H Yokoyama, T Miki, H Fukuda, H Kobata, W Kawada, J Otera Chem. Lett. 1057 (1980) 73. B M Trost, L Weber, P Strege, T J Fullerton, T J Dietsche J. Am. Chem. Soc. 100 3426 (1978) 74. H J Reich, S Wollowitz J. Am. Chem. Soc. 104 7051 (1982) 75. J H Babler, R A Haack J. Org. Chem. 47 4801 (1982) 76. R Tanikaga, Y Nozaki,M Nishida, A Kaji Bull. Chem. Soc. Jpn. 57 729 (1984) 77. C Herve du Penhoat,M Julia Tetrahedron 42 4807 (1986) 78. M Julia, J-P Stacino Tetrahedron, 42 2469 (1986) 79. M Julia, H Lauron, J-P Stacino, J-N Verpeaux Tetrahedron 42 2475 (1986) 80. T Mandai, T Yanagi, K Araki, Y Morisaki, M Kawada, J Otera J. Am. Chem. Soc. 106 3670 (1984) 81. J Otera, H Misawa, K Sugimoto J.Org. Chem. 51 3830 (1986) 82. T Moriyama, T Mandai,M Kawada, J Otera, B M Trost J. Org. Chem. 51 3896 (1986) 83. T Mandai, T Moriyama, K Tsujimoto, M Kawada, J Otera Tetrahedron Lett. 27 603 (1986) 84. J-E BaÈ ckvall, R Chinchilla, C Najera, M Yus Chem. Rev. 98 2291 (1998) 85. N A Plobeck, J-E BaÈ ckvall J. Org. Chem. 56 4508 (1991) 86. J-E BaÈ ckvall, A M Ericsson, N A Plobeck, S K Juntunen Tetrahedron Lett. 33 131 (1992) 87. J M Clough, in Comprehensive Organic Syntheses I Vol. 3 (Eds B M Trost, I Fleming) (Oxford: Pergamon Press, 1991) p. 861 88. E Block,M Aslam J. Am. Chem. Soc. 105 6164 (1983) 89. E Block,M Aslam, V Eswarakrishnan, A Wall J. Am. Chem. Soc. 105 6165 (1983) 90. E Block,M Aslam, V Eswarakrishnan, K Gebreyes, J Hutchinson, R Iyer, J-A Laffitte, A Wall J. Am.Chem. Soc. 108 4568 (1986) 91. P A Grieco, D Boxler Synth. Commun. 5 315 (1975)772 92. T E Bezmenova Khimiya Tiolen-1,1-Dioksidov (The Chemistry of Thiolene-1,1-Dioxides) (Kiev: Naukova Dumka, 1981) 93. S Yamada, H Ohsawa, T Suzuki, H Takayama J. Org. Chem. 51 4934 (1986) 94. Y Gaoni Tetrahedron Lett. 947 (1977) 95. A N Kasatkin, Y A Prokopenko, A M Khabibov, G A Tolstikov Mendeleev Commun. 17 (1993) 96. R Bloch, J Abecassis Tetrahedron Lett. 23 3277 (1982) 97. R Bloch, J Abecassis Tetrahedron Lett. 24 1247 (1983) 98. R Bloch, D Hassan-Gonzales Tetrahedron 42 4975 (1986) 99. T Nomoto, H Takayama J. Chem. Soc., Chem. Commun. 295 (1989) 100. K Kosugi, A V Anisimov, H Yamamoto, R Yamashiro, K Shirai, T Kumamoto Chem.Lett. 1341 (1981) 101. J M McIntosh, R A Sieler J. Org. Chem. 43 4431 (1978) 102. H F Schmitthenner, S M Weinreb J. Org. Chem. 45 3372 (1980) 103. J Mulzer, G BruÈ ntrup, U KuÈ hl, G Hartz Chem. Ber. 115 3453 (1982) 104. B M Trost, J M Fortunak J. Am. Chem. Soc. 102 2841 (1980) 105. A RuÈ ttimann, A Wick, A Eschenmoser Helv. Chim. Acta 58 1450 (1975) 106. G Jones Org. React. 15 204 (1967) 107. O Doebner Berichte 33 2140 (1900) 108. T Lennarts Berichte 76 1006 (1943) 109. M Jacobson J. Am. Chem. Soc. 78 5084 (1956) 110. C F H Allen, J van Allan Org. Synth., Coll. 3 783 (1955) 111. J Klein, E D Bergmann J. Am. Chem. Soc. 79 3452 (1957) 112. J Rodriguez, B Waegell Synthesis 534 (1988) 113.N Ragoussis Tetrahedron Lett. 28 93 (1987) 114. G Tetradis, N Ragoussis, in The 8th International IUPAC Confer- ence on Organic Synthesis (Additional Abstracts), Helsinki, 1990 p. 2-514 115. P F Vlad, M Z Krimer Org. Prep. Proced. Int. 30 657 (1998) 116. C A Henrick, W E Willy, J W Baum, T A Baer, B A Garecia, T A Mastre, S M Chang J. Org. Chem. 40 1 (1975) 117. A T Nielsen Org. React. 16 1 (1968) 118. V S Markevich, S M Markevich Khim. Promst 898 (1973) 119. M Mousseron-Canet, C Levallois Bull. Soc. Chim. Fr. 993 (1963) 120. V N Belov, N A Daev, S D Kustova, K V Leets, S S Poddubnaya, N I Skvortsova, E I Shepelenkova, A K Shumeiko Zh. Obshch. Khim. 27 1384 (1957) a 121. A A Petrov, N A Razumova,M L Genusov Zh. Obshch. Khim.28 1128 (1958) a 122. S A Voitkevich 865 Dushistykh Veshchestv dlya Parfyumerii i Bytovoi Khimii (865 Perfumes for Perfumery and Household Chemistry) (Moscow: Pishchevaya Promst, 1994) p. 225 123. J Herscovici, L Boumaiza, K Antonakis Tetrahedron Lett. 32 1791 (1991) 124. J Herscovici, S Delatre, K Antonakis Tetrahedron Lett. 32 1183 (1991) 125. E J Corey, D Enders Tetrahedron Lett. 11 (1976) 126. S Saito,M Shiozawa, M Ito, H Yamamoto J. Am. Chem. Soc. 120 813 (1998) 127. R Lantzsch, D Arlt Liebigs Ann. Chem. 1757 (1976) 128. E J Corey, G Schmidt Tetrahedron Lett. 2317 (1979) 129. L A Yanovskaya, in Reaktsii i Metody Issledovaniya Organiche- skikh Soedinenii (Reactions and Methods of Investigation of Organic Compounds) (Moscow: GNTIKhL, 1962) Vol.11, p. 321 130. L A Yanovskaya, S S Yufit, V F Kucherov Khimiya Atsetalei (The Chemistry of Acetals) (Moscow: Nauka, 1975) p. 222 131. I N Nazarov, Zh A Krasnaya, V P Vinogradov Zh. Obshch. Khim. 28 460 (1958) a 132. Zh A Krasnaya, V F Kucherov Izv. Akad. Nauk SSSR, Otd. Khim. Nauk 1057 (1962) b 133. L A Yanovskaya, G M Zhdankina, G V Kryshtal', E P Serebryakov Izv. Akad. Nauk SSSR, Ser. Khim. 2790 (1987) b 134. O Isler,M Montavor, R RuÈ egg Helv. Chim. Acta 39 259 (1956) 135. G Stork,M Tomasz J. Am. Chem. Soc. 86 471 (1964) 136. H J Hagermeyer Ind. Eng. Chem. 41 765 (1949) 137. Braz. P. 7 407 691; Chem. Abstr. 86 105 955 (1977) 138. K Maruoka, A B Conception, H Yamamoto Synlett 31 (1992) 139. D G Kubler J. Org.Chem. 27 1435 (1962) 140. O A Shavrygina, S M Makin Zh. Org. Khim. 2 1354 (1966) c A A Vasil'ev, E P Serebryakov 141. A Maercker, in Organic Reactions Vol. 14 (New York: Wiley, 1965) p. 270 142. B E Maryanoff, A B Reitz Chem. Rev. 89 863 (1989) 143. G V Kryshtal', E P Serebryakov Izv. Akad. Nauk, Ser. Khim. 1867 (1995) b 144. X Wei, R J K Taylor J. Org. Chem. 65 616 (2000) 145. H J Bestmann, O Vostrowsky, H Paulus, W Billmann, W Stransky Tetrahedron Lett. 121 (1977) 146. J Rokach, R N Young,M Kakushima, C-K Lau, R Seguin, R Franette, Y Guindon Tetrahedron Lett. 22 979 (1981) 147. J Rokach, R Zambani, C-K Lau, Y Guindon Tetrahedron Lett. 22 2759 (1981) 148. S R Baker, J R Boot, S E Morgan, D J Osborn,W J Ross, P R Shrubsall Tetrahedron Lett.24 4469 (1983) 149. I Ernest, A Main, R Menasse Tetrahedron Lett. 23 167 (1982) 150. L Alcaraz, G Macdonald, J Ragot, N J Lewis, R J K Taylor Tetrahedron 55 3707 (1999) 151. Y Guindon, D Delorme, C K Lau, R Zamboni J. Org. Chem. 53 267 (1988) 152. Y Leblanc, B J Fitzsimmons, R Zamboni, J Rokach J. Org. Chem. 53 265 (1988) 153. G M Zhdankina, V I Kadentsev, G V Kryshtal', E P Serebryakov, L A Yanovskaya Izv. Akad. Nauk SSSR, Ser. Khim. 1179 (1987) b 154. R J K Taylor Synthesis 564 (1977) 155. R J K Taylor Synthesis 566 (1977) 156. D F Taber, B P Gunn J. Am. Chem. Soc. 101 3992 (1979) 157. M Moreno-ManÄ as, A Trius Tetrahedron Lett. 22 3109 (1981) 158. M Moreno-ManÄ as, R M Ortuso,M Prat, M A Gala'n Synth. Commun. 16 1003 (1986) 159.S W Scheuplein, R BruÈ ckner Chem. Ber. 124 1871 (1991) 160. J D White, M S Jensen Tetahedron Lett. 33 577 (1992) 161. Y Shen, T Wang Tetrahedron Lett. 32 4353 (1991) 162. G Vidari, S FerrinÄ o, P A Grieco J. Am. Chem. Soc. 106 3539 (1984) 163. T K M Shing, Y Tang Tetrahedron 46 2187 (1990) 164. J Villieras, M Rambaud Synthesis 300 (1983) 165. A A Vasil'ev, L Engman, V M Merkulova Izv. Akad. Nauk, Ser. Khim. 1300 (2000) b 166. A A Vasil'ev, G V Kryshtal, E P Serebryakov Mendeleev Commun. 41 (1995) 167. K Ando J. Org. Chem. 63 8411 (1998) 168. K Ando, T Oishi,M Hirama, H Ohno, T Ibuka J. Org. Chem. 65 4745 (2000) 169. R Liu,M Schlosser Synlett 1197 (1996) 170. B S Lee, S Y Lee, D Y Oh J. Org. Chem. 65 4175 (2000) 171. R N Gedye, K C Westaway, P Arora, R Bisson, A H Khalil Can.J. Chem. 55 1218 (1977) 172. D Murali, R G S Krishna Rao Synthesis 254 (1987) 173. G V Kryshtal', E P Serebryakov, L M Suslova, L A Yanovskaya Izv. Akad. Nauk SSSR, Ser. Khim. 2377 (1988) b 174. G V Kryshtal', G M Zhdankina, E P Serebryakov Izv. Akad. Nauk, Ser. Khim. 1839 (1997) b 175. A A Vasil'ev, A L Vlasyuk, G V Kryshtal', E P Serebryakov Izv. Akad. Nauk, Ser. Khim. 2026 (1995) b 176. V N Odinokov, V R Akhmetova, R G Savchenko, A A Fatykhov,K I Pashkevich, E P Serebryakov Izv. Akad. Nauk, Ser. Khim. 2196 (1995) b 177. G V Kryshtal', G M Zhdankina, E P Serebryakov Izv. Akad. Nauk, Ser. Khim. 1089 (1993) b 178. G V Kryshtal', G M Zhdankina, E P Serebryakov Izv. Akad. Nauk, Ser.Khim. 1094 (1993) b 179. G V Kryshtal', E P Serebryakov, L M Suslova, L A Yanovskaya Izv. Akad. Nauk SSSR, Ser. Khim. 1414 (1990) b 180. G V Kryshtal, G M Zhdankina, E P Serebryakov, in The 11th International Conference on Organic Synthesis (Book of Abstracts), Amsterdam, 1996 p 587 181. D Mead, A E Asato, M Denny, R S H Liu, Y Hanzawa, T Taguchi, A Yamada, N Kobayashi, A Hosoda, Y Kobayashi Tetrahedron Lett. 28 259 (1987) 182. A Trehan, R S H Liu Tetrahedron Lett. 29 419 (1988) 183. W R Roush J. Am. Chem. Soc. 100 3599 (1978) 184. P BñckstroÈ m, U Jacobsson, T Norin, C R Unelius Tetrahedron 44 2541 (1988)Synthetic methodologies for carbo-substituted conjugated dienes 185. J M Takacs,M R Jaber, F Clement, C Walters J. Org. Chem. 63 6757 (1998) 186.A A Vasil'ev, E P Serebryakov Zh. Org. Khim. 34 1033 (1998) c 187. A Mohamed-Hachi, E About-Jaudet, J-C Combret, N Collignon Synthesis 1188 (1999) 188. N Kann, T Rein, B Ekermark, P Helquist J. Org. Chem. 55 5312 (1990) 189. L Dumahel, J Guillemont, Y L Gallc, G Ple, J-M Poirier, Y Ramondenc, P Chabardes Tetrahedron Lett. 31 3129 (1990) 190. J Xiao, Y Feng, C Yuang J. Chem. Soc., Perkin Trans. 1 4240 (2000) 191. Y Ikeda, J Ukai, H Yamamoto Tetrahedron 43 723 (1987) 192. A H Davidson, I Fleming, J I Grayson, A Pearce, R L Snowden, S Warren J. Chem. Soc., Perkin Trans. 1 550 (1977) 193. J H Rigby,M Qabar J. Org. Chem. 54 5852 (1989) 194. Y Wang, J Li,Y Wu,Y Huang, L Shi, J Yang Tetrahedron Lett. 27 4583 (1986) 195.C Gravier-Pelletier, J Dumas, Y Le Merrer, J C Depezay Tetrahedron Lett. 32 1165 (1991) 196. Y-Z Huang, C Chen, Y Shen Synth. Commun. 501 (1989) 197. A Osuka, Y Mori, H Shimizu, H Suzuki Tetrahedron Lett. 24 2599 (1983) 198. Q Zhong, Y L Cheng, J G Shao, C Q Liu Chin. Chem. Lett. 2 51 (1991) 199. M Bellassoued, A Majidi J. Org. Chem. 58 2517 (1993) 200. M Bellassoued, M Salemkour Tetrahedron Lett. 34 5281 (1993) 201. M Bellassoued, M Salemkour Tetrahedron 52 4607 (1996) 202. Y Hatanaka, I Kuwajima J. Org. Chem. 51 1932 (1986) 203. T Cunigny, C Herve du Penhoat,M Julia Tetrahedron 43 859 (1987) 204. P J Kocienski, B Lythgoe, S Ruston J. Chem. Soc., Perkin Trans. 1 829 (1978) 205. J B Baudin, G Hareau, S A Julia, O Ruel Bull. Soc. Chim. Fr.130 336 (1993) 206. J B Baudin, G Hareau, S A Julia, R Lorne, O Ruel Bull. Soc. Chim. Fr. 130 856 (1993) 207. N D Smith, P J Kocienski, S D A Street Synthesis 652 (1996) 208. M Clarembeau, A Cravador,W Dumont, L Hevesi, A Krief, J Lucchetti, D van Ende Tetrahedron 41 4793 (1985) 209. A Krief,W Dumont,M Clarembeau, E Badaoui Tetrahedron 45 2023 (1989) 210. H J Reich, F Chow J. Chem. Soc., Chem. Commun. 790 (1975) 211. J Remion, A Krief Tetrahedron Lett. 41 3743 (1976) 212. A A Vasil'ev, L Engman, E P Serebryakov J. Chem. Soc., Perkin Trans. 1 2211 (2000) 213. Y Yamamoto, H Yatagi, Y Saito, K Maruyama J. Org. Chem. 49 1096 (1984) 214. F N Tebbe, G W Parshall, G S Reddy J. Am. Chem. Soc. 100 3611 (1978) 215. H Hauptmann, G MuÈ hlbauer, H Sass Tetrahedron Lett.27 6189 (1986) 216. Y Horikawa, M Watanabe, T Fujiwara, T Takeda J. Am. Chem. Soc. 119 1127 (1997) 217. T Takeda, Y Takagi, N Saeki, T Fujiwara Tetrahedron Lett. 41 8377 (2000) 218. U SchoÈ llkopf, F Gerhart Angew. Chem. 80 842 (1968) 219. O Fujimura, T Honma Tetahedron Lett. 39 625 (1998) 220. S BraÈ se, A de Meijere, in Metal-Catalyzed Cross-Coupling Reactions (Eds F Diederich, P Stang) (Weinheim: Wiley-VCH, 1998) p. 99 221. I P Beletskaya, A V Cheprakov Chem. Rev. 100 3009 (2000) 222. T N Mitchell, in Metal-Catalyzed Cross-Coupling Reactions (Eds F Diederich, P Stang) (Weinheim: Wiley-VCH, 1998) p. 167 223. V Farina, V Krishnamurthy,W J Scott The Stille Reaction. Wiley New York 1998 224. N Miyaura, A Suzuki Chem.Rev. 95 2457 (1995) 225. A Suzuki, in Metal-Catalyzed Cross-Coupling Reactions (Eds F Diederich, P Stang) (Weinheim: Wiley-VCH, 1998) p. 49 226. E Negishi, F Liu, in Metal-Catalyzed Cross-Coupling Reactions (Eds F Diederich, P Stang) (Weinheim: Wiley-VCH, 1998) p. 1 227. K Ritter Synthesis 735 (1993) 228. G Zweifel, R L Miller J. Am. Chem. Soc. 92 6678 (1970) 229. M F Semmelhack, P M Helquist, J D Gorzynski J. Am. Chem. Soc. 94 9234 (1972) 773 230. R Vanderesse, Y Fort, S Becker, P Caubere Tetrahedron Lett. 27 3517 (1986) 231. H A Dieck, R F Heck J. Am. Chem. Soc. 96 1133 (1974) 232. T Jeffery J. Chem. Soc., Chem. Commun. 324 (1991) 233. T Jeffery Tetrahedron Lett. 34 1133 (1993) 234. E Bernocchi, S Cacchi, P G Ciattini, E Morera, G Ortar Tetrahedron Lett.33 3073 (1992) 235. M Webel, H-U Reissig Synlett 1141 (1997) 236. K Hirabayashi, Y Nishihara, A Mori, T Hiyama Tetrahedron Lett. 39 7893 (1998) 237. J Yoshida, K Tamao, H Yamamoto, T Kakui, T Uchida, M Kumada Organometallics 1 542 (1982) 238. R M Moriarty, W R Epa, A K Awasthi J. Am. Chem. Soc. 113 6315 (1991) 239. G P Chuisoli, G Salerno, W Giroldini, L Pallini J. Organomet. Chem. 219 C16 (1981) 240. J K Stille, B L Groh J. Am. Chem. Soc. 109 813 (1987) 241. M Abarbi, J-L Parrain, J-C Cintrat, A Ducheà ne Synthesis 82 (1996) 242. C J Forsyth, J Clardy J. Am. Chem. Soc. 110 5911 (1988) 243. H -O Kim, C O Ogbu, S Nelson, M Kahn Synlett 1059 (1998) 244. Y Naruse, T Esaki, H Yamamoto Tetrahedron Lett.29 1417 (1988) 245. W Shen, L Wang J. Org. Chem. 64 8873 (1999) 246. J K Stille, J H Simpson J. Am. Chem. Soc. 109 2138 (1987) 247. R E Maleczka, I Terstiege J. Org. Chem. 63 9622 (1998) 248. R E Malescka Jr, W P Gallagher, I Terstiege J. Am. Chem. Soc. 122 384 (2000) 249. Y Kobayashi, N Nakayama, R Mizojiri Tetrahedron 54 1053 (1998) 250. M Satoh, N Miyaura, A Suzuki Chem. Lett. 1329 (1986) 251. I Rivera, J A Soderquist Tetrahedron Lett. 32 2311 (1991) 252. X Chen, J G Millar Synthesis 113 (2000) 253. N Miyaura, Y Satoh, S Hara, A Suzuki Bull. Chem. Soc. Jpn. 59 2029 (1986) 254. T Oh-e, N Miyaura, A Suzuki J. Org. Chem. 58 2201 (1993) 255. M V Mavrov, N A Urdaneta, C H Nguyen, E P Serebryakov Izv. Akad. Nauk SSSR, Ser. Khim.2633 (1987) b 256. M V Mavrov, N A Urdaneta, E P Serebryakov Bioorg. Khim. 16 711 (1990) d 257. J P Gene^t, A Linquist, E Blart, V Mouries,M Savignac Tetrahedron Lett. 36 1443 (1995) 258. N Miyaura, A Suzuki J. Organomet. Chem. 213 C53 (1981) 259. E Negishi, T Takahashi, S Baba, D E van Horn, N Okukado J. Am. Chem. Soc. 109 2393 (1987) 260. J-E BaÈ ckvall, H E Schink, Z D Renko J. Org. Chem. 55 826 (1990) 261. N Okukado, D E van Horn, W L Klima, E Negishi Tetrahedron Lett. 1027 (1978) 262. G Cahiez, H Avedissian Synthesis 1199 (1998) 263. V Fiandanese, G Marchese, G Mascolo, F Naso, L Ronzini Tetrahedron Lett. 29 3705 (1988) 264. I D Gridnev, N Miyaura, A Suzuki J. Org. Chem. 58 5351 (1993) 265. T Hiyama, in Metal-Catalyzed Cross-Coupling Reactions (Eds F Diederich, P Stang) (New York: Wiley, 1998) p.421 266. Y Hatanaka, T Hiyama Synlett 845 (1991) 267. Y Hatanaka, T Hiyama J. Org. Chem. 53 918 (1988) 268. Y Hatanaka, T Hiyama J. Org. Chem. 54 268 (1989) 269. Y Hatanaka, T Hiyama Tetrahedron Lett. 31 2719 (1990) 270. R C Larock, B Riefling J. Org. Chem. 43 1468 (1976) 271. A P Koziokowski, Y Kitigawa Tetrahedron Lett. 23 2087 (1982) 272. G Zweifel, J T Snow, C C Whitney J. Am. Chem. Soc. 90 7139 (1968) 273. B A Patel, J E Dickerson, R F Heck J. Org. Chem. 43 5018 (1978) 274. T Jeffery Tetrahedron Lett. 33 1989 (1992) 275. J Uenishi, R Kawahama, Y Izaki, O Yonemitsu Tetrahedron 56 3493 (2000) 276. Y Ikeda, J Ukai,N Ikeda,H Yamamoto Tetrahedron Lett. 25 5177 (1984) 277.A S E KarlstroÈ m, M RoÈ nn, A Thorarensen, J-E BaÈ ckvall J. Org. Chem. 63 2517 (1998) 278. A S E KarlstroÈ m, K Itami, J-E BaÈ ckvall J. Org. Chem. 64 1745 (1999) 279. L Novak, J Rohaly, P Kolonits, P Fekete, L Varjac, C S Szantay Liebigs Ann. Chem. 1173 (1982) 280. M Alami, J-F Peyrat, J-D Brion Synthesis, 1499 (2000)774 281. V Ratavelomanana, G Linstrumelle Tetrahedron Lett. 22 315 (1982) 282. K Sonogashira, in Metal-Catalyzed Cross-Coupling Reactions (Eds F Diederich, P Stang) (New York: Wiley, 1998) p. 203 283. G Zeni, V Comasseto Tetrahedron Lett. 40 4619 (1999) 284. E Negishi, T Yoshida J. Chem. Soc., Chem. Commun. 606 (1973) 285. E Negishi, G Lew, T Yoshida J. Chem. Soc., Chem. Commun. 874 (1973) 286. E Negishi, A Abramovich Tetrahedron Lett.411 (1977) 287. J F Normant, A Alexakis Synthesis 841 (1981) 288. F NaÈ f, P Degen Helv. Chim. Acta 54 1939 (1971) 289. E J Corey, C U Kim, R H K Chen,M Takeda J. Am. Chem. Soc. 94 4395 (1972) 290. A Alexakis, J F Normant Tetrahedron Lett. 23 5151 (1982) 291. A Alexakis, G Cahiez, J F Normant Tetrahedron 36 1961 (1980) 292. M Furber, R J K Taylor, S C Burford J. Chem. Soc., Perkin Trans. 1 1809 (1986) 293. M Furber, R J K Taylor, S C Burford Tetrahedron Lett. 26 3285 (1985) 294. E Piers, E M Boehringer, J G K Yee J. Org. Chem. 63 8642 (1998) 295. J C Clenet, G Linstrumelle Nouv. J. Chim. 1 373 (1977) 296. F Scott, G Cahiez, J F Normant, J Villieras J. Organomet. Chem. 144 13 (1978) 297. T Naota, H Takaya, S Murahashi Chem.Rev. 98 2599 (1998) 298. T Mitsudo, S-W Zhang, M Nagao, Y Watanabe J. Chem. Soc., Chem. Commun. 598 (1991) 299. T Mitsudo, S-W Zhang, T Kondo, Y Watanabe Tetrahedron Lett. 33 341 (1992) 300. M Murakami,M Ubukata, Y Ito Tetrahedron Lett. 39 7361 (1998) 301. B M Trost, A F Indolese, T J J MuÈ ller, B Treptow J. Am. Chem. Soc. 117 615 (1995) 302. C S Yi, N Liu Organometallics 17 3158 (1998) 303. C S Yi, N Liu Synlett 281 (1999) 304. Y Nishibayashi, M Yamanashi, I Wakiji, M Hidai Angew. Chem., Int. Ed. Engl. 39 2909 (2000) 305. B M Trost,M Lautens J. Am. Chem. Soc. 107 1781 (1985) 306. B M Trost, S-F Chen J. Am. Chem. Soc. 108 6053 (1986) 307. B M Trost, D T MacPherson J. Am. Chem. Soc. 109 3483 (1987) 308.C M Huwe, S Blechert Tetrahedron Lett. 35 9537 (1994) 309. B M Trost, J Y L Chung J. Am. Chem. Soc. 107 4586 (1985) 310. B M Trost, G J Tanoury,M Lautens, C Chan, D T MacPherson J. Am. Chem. Soc. 116 4255 (1994) 311. B M Trost,M Lautens Tetrahedron Lett. 26 4887 (1985) 312. B M Trost, O J Gelling Tetrahedron Lett. 34 8233 (1993) 313. B M Trost, J M Tour J. Am. Chem. Soc. 109 5268 (1987) 314. J Le Paih, D C Rodriguez, S De rien, P H Dixneuf Synlett 95 (2000) 315. M Nishida, N Adachi, K Onozuka, H Matsumura,M Mori J. Org. Chem. 63 9158 (1998) 316. R Grigg, P Stevenson, T Worakun Tetrahedron 44 2033 (1988) 317. R Grigg, P Stevenson, T Worakun Tetrahedron 44 2049 (1988) 318. W A Nugent, J C Calabrese J. Am. Chem. Soc. 106 6422 (1984) 319.E Negishi, F E Cederbaun, T Takahashi Tetrahedron Lett. 27 2829 (1986). 320. W A Nugent,D L Thorn,R L Harlow J. Am. Chem. Soc. 109 2788 (1987) 321. B M Trost, D C Lee J. Am. Chem. Soc. 110 7255 (1988) 322. J M Nuss, M McMurphy, R A Rennels, M H Heravi, B J Mohr Tetrahedron Lett. 34 3079 (1993) 323. J W Madine, X Wang, R A Widenhoefer Org. Lett. 3 385 (2001) 324. A Kinoshita, N Sakakibara,M Mori J. Am. Chem. Soc. 119 12 388 (1997) 325. A Kinoshita, N Sakakibara, M Mori Tetrahedron 55 8155 (1999) 326. J A Smulik, S T Diver J. Org. Chem. 65 1788 (2000) 327. A A Vasil'ev, L Engman, E P Serebryakov Mendeleev Commun. 101 (2000) 328. S C SchoÈ rer, S Blechert Synlett 166 (1998) 329. L Ackermann, A FoÈ rstner, T Weskamp, F J Kohl, W A Herrmann Tetrahedron Lett.40 4787 (1999) 330. R Stragies, U Voigtmann, S Blechert Tetrahedron Lett. 41 5465 (2000) 331. A Kinoshita, M Mori Synlett 1020 (1994) 332. A Kinoshita, M Mori J. Org. Chem. 61 8356 (1996) 333. M Mori, N Sakakibara, A Kinoshita J. Org. Chem. 63 6082 (1998) A A Vasil'ev, E P Serebryakov 334. J Renaud, C-D Graf, L Oberer Angew. Chem., Int. Ed. Engl. 39 3101 (2000) 335. S -H Kim, W J Zuercher, N B Bowden, R H Grubbs J. Org. Chem. 61 1073 (1996) 336. N Chatani, T Morimoto, T Muto, S Murai J. Am. Chem. Soc. 116 6049 (1994) 337. A FuÈ rstner, H Szillat, F Stelzer J. Am. Chem. Soc. 122 6785 (2000) 338. B M Trost, A B Pinkerton J. Am. Chem. Soc. 121 4068 (1999) 339. B M Trost, A B Pinkerton J. Am. Chem. Soc. 121 10842 (1999) 340.B Friess, B Cazes, J Gore Tetrahedron Lett. 29 4089 (1988) 341. N Kopola, B Friess, B Cazes, J Gore Tetrahedron Lett. 30 3963 (1989) 342. V Gauthier, B Cazes, J Gore Tetrahedron Lett. 32 915 (1991) 343. N Chaptal, V Colovray-Gotteland, C Grandjean, B Cazes, J Gore Tetrahedron Lett. 32 1795 (1991) 344. N Vicart, B Cazes, J Gore Tetrahedron Lett. 36 5015 (1995) 345. H-M Chang, C-H Cheng J. Org. Chem. 65 1767 (2000) 346. Y Imada, G Vasapollo, H Alper J. Org. Chem. 61 7982 (1996) 347. H Maeta, T Hasegawa, K Suzuki Synlett 341 (1993) 348. B G van den Hoven, H Alper J. Org. Chem. 64 3964 (1999) 349. J Kiji, T Okano, E Fujii, J Tsuji Synthesis 869 (1997) 350. J BoÈ hmer, R Grigg Tetrahedron 55 13463 (1999) 351. L Zhu, R D Rieke Tetrahedron Lett.32 2865 (1991) 352. X Qi, J Montgomery J. Org. Chem. 64 9310 (1999) 353. M J Rozema, P Knochel Tetrahedron Lett. 32 1855 (1991) 354. H Maeta, K Suzuki Tetrahedron Lett. 34 341 (1993) 355. A N Kasatkin, R J Whitby Tetrahedron Lett. 41 6201 (2000) 356. K M Doxsee, J K M Mouser Tetrahedron Lett. 32 1687 (1991) 357. Z Wang, X Lu, A Lei, Z Zhang J. Org. Chem. 63 3806 (1998) 358. Y Kataoka, J Miyai, K Oshima, K Takai, K Utimoto J. Org. Chem. 57 1973 (1992) 359. J R Stricker, P A Wexler, D E Wigley Organometallics 10 118 (1991) 360. I U Khand, P L Pauson J. Chem. Soc., Chem. Commun. 379 (1974) 361. P G Andersson, J-E BaÈ ckvall J. Org. Chem. 56 5349 (1991) 362. Y I M Nilsson, P G Andresson, J-E BaÈ ckvall J. Am. Chem. Soc.115 6609 (1993) 363. B M Trost, S Hildbrand, K Dogra J. Am. Chem. Soc. 121 10416 (1999) 364. R Gre e Synthesis 341 (1989) 365. G R Knox, I G Thon J. Chem. Soc., Chem. Commun. 373 (1981) 366. D F Harvey, V B Selhau J. Org. Chem. 65 2282 (2000) 367. J BoÈ hmer, F Hampel, R Schobert Synthesis 661 (1997) 368. C E Anson, G Dave, G R Stephenson Tetrahedron 56 2273 (2000) 369. H Birmann, V Prahlad, C Tao, Y K Yun, Z Wang, W A Donaldson Tetrahedron 56 2283 (2000) 370. E Wenkert,M Guo, R Lavilla, B Porter, K Ramachandran, J-H Sheu J. Org. Chem. 55 6203 (1990) 371. D K P Ng, T-Y Luh J. Am. Chem. Soc. 111 9119 (1989) 372. J Auge, S David Tetrahedron Lett. 24 4009 (1983) 373. T Satoh, S Motohashi, K Yamakawa Chem. Pharm. Bull. 36 1169 (1988) 374.T P Meagher, L Yet, C-N Hsiao, H Shechter J. Org. Chem. 63 4181 (1998) 375. T P Meagher, H Shechter J. Org. Chem. 63 4193 (1998) 376. K Mori Tetrahedron 30 3807 (1974) 377. A Ichihara, N Miki, H Tazaki, S Sakamura Tetrahedron Lett. 28 1175 (1987) 378. I Fleming, A Barbero, D Walter Chem. Rev. 97 2063 (1997) 379. D Seyferth, J Pornet J. Org. Chem. 45 1721 (1980) 380. R H Wollenberg Tetrahedron Lett. 717 (1978) 381. E J Corey, D A Clark, G Goto, A Marfat, C Mioskowski, B Samuelsson, S HammarstroÈ m J. Am. Chem. Soc. 102 1436 (1980) 382. B Contreras, L Duhamel, G Ple Synth. Commun. 2983 (1990) 383. R M Seifert, R G Buttery J. Agric. Food Chem. 28 68 (1980) 384. M S Brouwer, H Hulkenberg, J G J Kok, R van Moorselaar, W R M Overbeek, P G J Wesselman Recl.Trav. Chim. Pays-Bas 98 316 (1979) 385. T Hayashi, K Sasaoka, T Oishi J. Chem. Soc., Chem. Commun. 1362 (1990) 386. N Vicart, D Castet-Caillabet, Y van Ramondenc, G Ple , L Duhamel Synlett 411 (1998) 387. D Soullez, G Ple , L Duhamel, P Duhamel J. Chem. Soc., Chem. Commun. 563 (1995)Synthetic methodologies for carbo-substituted conjugated dienes 388. A C Friedly, E Yang, S R Marder Tetrahedron 53 2717 (1997) 389. Y-Z Huang, X-S Mo, L Wang Tetrahedron Lett. 39 419 (1998) 390. Y Y Belosludtsev, B C Borer, R J K Taylor Synthesis 320 (1991) 391. R J K Taylor, K Hemming, E F De Medeiros J. Chem. Soc., Perkin Trans. 1 2385 (1995) 392. N Lewis, P W McKen, R J K Taylor Synlett 898 (1991) 393. S Chandrasekhar, M V Reddy, K S Reddy, C Ramaro Tetrahedron Lett.41 2667 (2000) 394. T Fujisawa, N Okada,M Takeuchi, T Sato Chem. Lett. 1271 (1983) 395. F Caturla, C Najera, M Varea Tetrahedron Lett. 40 5957 (1999) 396. S Hatakeyama, K Sugawara, M Kawamura, S Takano Tetrahedron Lett. 32 4509 (1991) 397. R Soundararajan, G Li, H C Brown J. Org. Chem. 61 100 (1996) 398. M Montury, B Psaume, J Gore Tetrahedron Lett. 21 163 (1980) 399. A R Katritzky, L Serdyuk, D Toader, X Wang J. Org. Chem. 64 1888 (1999) 400. K J Shea, P Q Phem Tetrahedron Lett. 24 1003 (1983) 401. F F Fleming, T Jiang J. Org. Chem. 62 7890 (1997) 402. S Nunomoto, Y Kawakami, Y Yamashita Bull. Chem. Soc. Jpn. 54 2831 (1981) 403. O Tsuge, E Wada, S Kanemasa, H Sakoh Chem. Lett. 469 (1984) 404. A Arase,M Hoshi J. Chem. Soc., Chem. Commun. 531 (1987) 405. M Hoshi,Y Masuda, A Arase J. Chem. Soc., Chem. Commun. 1629 (1987) 406. R R Wilson, L R Phillips, K J Natalie J. Am. Chem. Soc. 101 3340 (1979) 407. J P Godschalx, J K Stille Tetrahedron Lett., 24 1905 (1983) 408. C Najera, J M Sansano Tetrahedron 50 5829 (1994) 409. G O Schenck,W Hartmann, R Steinmetz Chem. Ber. 96 498 (1963) 410. H -D Scharf, J Mattay Liebigs Ann. Chem. 772 (1977) 411. D Bellus, C D Weis Tetrahedron Lett. 999 (1973) 412. S Halazy, A Krief Tetrahedron Lett. 21 1997 (1980) 413. S Ingham, R W Turner, T W Wallace J. Chem. Soc., Chem. Commun. 1664 (1985) 414. G B Bennett Synthesis 589 (1977) 415. S F Reed J. Org. Chem. 30 1663 (1965) 416. S A Bal, P Helquist Tetrahedron Lett. 22 3933 (1981) 417. T Hudlicky, T M Kutchan, S R Wilson, D T Mao J. Am. Chem. Soc. 102 6351 (1980) 418. S Karlsen, P Frùyen, L Skattebùl Acta Chem. Scand., Ser. B 30 664 (1976) 419. M Bertrand, J Viala Tetrahedron Lett. 2575 (1978) 420. J N Labovitz, C A Henrick, V L Corbin Tetrahedron Lett. 4209 (1975) 421. D Samain, C Descoins Bull. Soc. Chim. Fr., Pt II 71 (1979) 422. L A Yanovskaya, in Reaktsii i Metody Issledovaniya Organicheskikh Soedinenii (Reactions and Methods of Investigation of Organic Compounds) (Moscow: Khimiya, 1963) Vol. 12, p. 273 423. I N Nazarov, L A Yanovskaya, B P Gusev, S M Makin, I I Nazarova Dokl. Akad. Nauk SSSR 114 1029 (1957) e 424. G I Samokhvalov,M A Miropol'skaya, N A Preobrazhenskii Dokl. Akad. Nauk SSSR 107 103 (1956) e 425. G Saucy, R Marbet Helv. Chim. Acta 50 1158 (1967) 426. J K Crandall, G L Tindell J. Chem. Soc., Chem. Commun. 1411 (1970) 427. C A Henrick, W E Willy, D R McKean, E Baggiolini, J B Siddall J. Org. Chem. 40 8 (1975) 428. S Tsuboi, T Masuda, H Makino, A Takeda Tetrahedron Lett. 23 209 (1982) 429. J Ficini, N Lumbroso-Bader, J Pouliquen Tetrahedron Lett. 39 4139 (1968) 430. B Bowden, R C Cookson, H A Davis J. Chem. Soc., Perkin Trans. 1 2634 (1973) 431. J E Baldwin, P A R Bennet, A K Forrest J. Chem. Soc., Chem. Commun. 250 (1987) 432. R C Cookson, R Gopalan J. Chem. Soc., Chem. Commun. 608 (1978) 433. G H Posner, C M Kinter J. Org. Chem. 55 3967 (1990) 434. A Douthean, G Balme,M Malacria, J Gore Tetrahedron Lett. 1803 (1978) 435. A Douthean, G Balme,M Malacria, J Gore Tetrahedron 36 1953 (1980) 436. T Onishi, Y Fujita, T Nishida Chem. Lett. 765 (1979) 775 437. Y Fujita, T Onishi, K Hino, T Nishida Tetrahedron Lett. 21 1347 (1980) 438. T Sato, H Tsunekawa, H Kohama, T Fujisawa Chem. Lett. 1553 (1986) 439. J J Eshelby, P J Parsons, N C Sillars, P J Crowley J. Chem. Soc., Chem. Commun. 1497 (1995) 440. S Kuroda, T Katsuki, M Yamaguchi Tetrahedron Lett. 28 803 (1987) 441. S Mikami, T Maeda, T Nakai Tetraheron Lett. 27 4189 (1986) 442. M Uchikawa, T Katsuki,M Yamaguchi Tetrahedron Lett. 27 4581 (1986) 443. E I Klimova, E G Treshchova, Yu A Arbuzov Dokl. Akad. Nauk SSSR 180 865 (1968) e 444. D Devaprabhakara, C G Cardenas, P D Gardner J. Am. Chem. Soc. 85 1553 (1963) 445. R A Amos, J A Katzenellenbogen J. Org. Chem. 43 555 (1978) 446. H Bosshardt,M Schlosser Helv. Chim. Acta 63 2393 (1980) 447. S Tsuboi, A Kuroda, T Masuda, A Takeda Chem. Lett. 1541 (1984) 448. L I Zakharkin, S A Babich Zh. Org. Khim. 18 2261 (1982) c 449. B T Golding, G Pierpoint, R Aneja J. Chem. Soc., Chem. Commun. 1030 (1981) 450. S Akiyama, S Nakatsuji, T Hamamura Tetrahedron Lett. 2809 (1979) 451. B Psaume,M Montury, J GoreÁ Synth. Commun. 12 409 (1982) 452. B M Trost, T Schmidt J. Am. Chem. Soc. 110 2301 (1988) 453. D Ma, Y Yu, X Lu J. Org. Chem. 54 1105 (1989) 454. X Lu, C Guo, D Ma Synlett 357 (1990) 455. B M Trost, U Kazmaier J. Am. Chem. Soc. 114 7933 (1992) 456. C Guo, X Lu J. Chem. Soc., Perkin Trans. 1 1921 (1993) 457. U Kazmaier J. Chem. Soc., Chem. Commun. 2305 (1997) 458. U Kazmaier Tetrahedron 54 1491 (1998) 459. Z Wang, X Lu Tetrahedron 51 11765 (1995) 460. C Guo, X Lu Synlett 405 (1992) 461. C Guo, X Lu J. Chem. Soc., Chem. Commun. 394 (1993) 462. I A Balova, L A Remizova, V F Makarycheva, E G Rumyantseva, I A Favorskaya Zh. Org. Khim. 27 64 (1991) 463. I A Balova, S N Morozkina, S V Voskresenskii, L A Remizova Zh. Org. Khim. 36 1466 (2000) c 464. I A Balova, I V Zakharova, L A Remizova Zh. Org. Khim. 29 1732 (1993) c 465. G Rickards, L Weiler J. Org. Chem. 43 3607 (1978) 466. W-N Chou, D L Clark, J B White Tetrahedron Lett. 32 299 (1991) 467. D Chemin, G Linstrumelle Synthesis 377 (1993) 468. U MaÈ eorg, T VaÈ limaÈ e Izv. Akad. Nauk Est. SSR. Khim. 37 186 (1988) 469. J A Cabezas, A C Oehlschlager Synthesis 107 (1999) 470. W Oppolzer, C Fehr, J Warneke Helv. Chim. Acta 60 48 (1977) 471. A P Khrimyan, O A Gharibyan, L Streinz, Z Wimmer, M Romanuk, S O Badanyan Collect. Czech. Chem. Commun. 54 3284 (1989) 472. A P Khrimyan, G M Makaryan, A L Ovanisyan, Z Vimmer, M Romanyuk, L Shtreints, Sh O Badanyan Khim. Prirod. Soed. 117 (1991) f 473. R E Doolittle Synthesis 730 (1984) 474. M V Mavrov, Z G Chrelashvili, E P Serebryakov Izv. Akad. Nauk, Ser. Khim. 769 (1993) b 475. Z G Chrelashvili,M V Mavrov, A V Dolidze, A O Chizhov, E P Serebryakov Izv. Akad. Nauk, Ser. Khim. 1133 (1993) b 476. O A Garibyan, G U Palikyan, Sh O Badanyan Zh. Org. Khim. 36 1462 (2000) c 477. H Westmijze, H Kleijn, P Vermeer Synthesis 430 (1979) 478. R K Haynes,W R Jackson, A Stragalinou Austr. J. Chem. 33 1537 (1980) 479. T Ishii, N Kawamura, S Matsubara, K Utimoto, S Kozima, T Hitomi J. Org. Chem. 52 4416 (1987) 480. P G McDougal, J G Rico, D van Derveer J. Org. Chem. 51 4492 (1986) 481. M P Reddy, G S K Rao Synthesis 815 (1980) 482. T N Gavrishova, A V Shastin, E S Balenkova, N A Novikov Zh. Org. Khim. 23 1322 (1987) c 483. P V Verteletskii, V G Nenaidenko, E S Balenkova Vestn. Mosk. Univ., Ser. 2, Khim. 540 (1996)A A Vasil'ev, E P Serebryakov 776 484. V G Nenaidenko, A V Sanin, E S Balenkova Zh. Org. Khim. 30 531 (1994) c 485. C Fan, B Cazes Tetrahedron Lett. 29 1701 (1988) 486. C Descoins, D Samain, B Lalanne-Cassou,M Gallois 487. N M Ivanova, B A Czeskis, A M Moiseenkov, O M Nefedov Izv. Bull. Soc. Chim. Fr., Pt II 941 (1977) Akad. Nauk SSSR, Ser. Khim. 2521 (1991) b 488. A M Moiseenkov, B A Czeskis, N M Ivanova, O M Nefedov J. Chem. Soc., Perkin Trans. 1 2639 (1991). 489. T Arenz, M Vostell, H Frauenrath Synlett 23 (1991) 490. E Piers, R D Tillyer J. Chem. Soc., Perkin Trans. 1 2124 (1989) 491. T Kato, M Sato, Y Kitagawa, R Sato Chem. Pharm. Bull. 29 1624 (1981) 492. T R Demmin, M M Rogic' J. Org. Chem. 45 2737 (1980) 493. J Haslouin, F Rouessac Bull. Soc. Chim. Fr., Pt II 1242 (1977) 494. A S Kende, S Johnson, P Sanfilippo, J C Hodges, L N Jungheim J. Am. Chem. Soc. 108 3513 (1986) a�Russ. J. Gen. Chem. (Engl. Transl.) b�Russ. Chem. Bull. (Engl. Transl.) c�Russ. J. Org. Chem. (Engl. Transl.) d�Russ. J. Biorg. Chem. (Engl. Transl.) e�Dokl. Chem. (Engl. Transl.) f�Chem. Nat. Compd. (En

ISSN:0036-021X    

出版商:RSC    

年代:2001

数据来源: RSC

摘要:

Russian Chemical Reviews 70 (9) 777 ± 790 (2001) Magnetic properties of complex oxides LiMO2 (M=Sc ± Ni) with different types of cationic ordering D G Kellerman Contents I. Introduction II. Structures of LiMO2 oxides III. Magnetic properties of LiMO2 oxides IV. Structure and magnetic properties of LiM1¡xM0xO2 solid solutions V. Conclusion Abstract. com- of properties magnetic and structures crystal The The crystal structures and magnetic properties of com- plex the of oxides plex oxides of the LiMO LiMO solid of and Ni) ± c S = (M 2 series series (M=Sc ± Ni) and of solid solutions of structures the Although described. are form they solutions they form are described. Although the structures of the the oxides their structure, NaCl the from derived all are oxides are all derived from the NaCl structure, their magnetic magnetic properties of type the on depend and diverse fairly are properties are fairly diverse and depend on the type of cationic cationic ordering.Primary attention is given to the oxides with layered ordering. Primary attention is given to the oxides with layered structures variable. a is ordering of degree the which in structures in which the degree of ordering is a variable. The The bibliography references 225 includes bibliography includes 225 references. I. Introduction During the past 15 years, compounds with the general formula of LiMO2, where M is a transition 3d metal, have actively been explored, and the interest in them is not lost up to date. First of all, this is related to the prospects of considering many LiMO2 complex oxides as promising cathodic materials for chemical power sources.1± 3 These prospects are usually determined by such properties as a high positive potential, small molecular mass and structural stability during electrochemical intercalation and deintercalation LiMO2 Li17xMO2+x Li++x e7.Besides this applied aspect, interesting magnetic properties of LiMO2 complex oxides draw the attention of many theoreticians and researchers.4 These oxides include antiferromagnetic, ferro- magnetic, paramagnetic, diamagnetic, weak ferromagnetic mate- rials, etc. The magnetic properties of LiMO2 oxides are largely determined by the type of superstructures they form. The whole spectrum of structures found in LiMO2 is based on the rock-salt structure, the deviations being determined by the way the cations are distributed over the positions of this basal structure. This distribution can either be random or ordered, the latter giving rise to the formation of a superstructure.In solid solutions of complex oxides, the type and the degree of ordering can change depending on the component concentrations. This is why compounds for which the degree of structural ordering (Z) varies as a function of synthetic conditions, which entails profound changes in magnetic D G Kellerman Institute of Solid State Chemistry, Urals Branch of the Russian Academy of Sciences, ul. Pervomaiskaya 91, 620219 Ekaterinburg, Russian Federation. Fax (7-343) 274 44 95.Tel. (7-343) 249 34 42. E-mail: kellerman@ihim.uran.ru Received 26 March 2001 Uspekhi Khimii 70 (9) 874 ± 889 (2001); translated by T Ya Safonova #2001 Russian Academy of Sciences and Turpion Ltd DOI 10.1070/RC2001v070n09ABEH000670 777 777 781 785 788 structures, are of particular interest. In this review, the studies in which the magnetic properties of LiMO2 compounds and solid solutions based on them are correlated with the peculiarities of their crystal structures are analysed. Most of numerous electro- chemical studies are beyond the scope of this review, these being surveyed elsewhere.3 ±5 II. Structures of LiMO2 oxides The crystal structures of LiMO2 complex oxides derive from the NaCl structure and differ from one another only in the way the cations of alkali and transition metals are distributed over the planes of the basal cubic lattice.When lithium and transition- metal ions occupy severally the alternating (111) planes of the sodium chloride cubic lattice, layered structures characterised by a high diffusion mobility of alkali-metal atoms are formed. Hence, in such compounds as, e.g., LiCoO2 and LiNiO2, deep and reversible electrochemical intercalation of lithium can occur which attracts the attention of scientists. The quest for a wide array of new compounds with the layered structures favourable for electrochemical processes stimulates the development of the methods for the preparation of metastable LiMO2 modifications. For this purpose, low-temperature methods which involve pre- cursor synthesis and ion-exchange processes, chemical and elec- trochemical intercalation, etc.are used. Usually, these phases lose their stabilities with an increase in temperature and transform into stable modifications. The LiMO2 compounds preferentially form three structural types. The first structure represents a disordered cubic lattice of a-LiFeO2 (NaCl), the second structure of the a-NaFeO2 type represents a trigonal superlattice on the basis of the NaCl structure and the third compound represents a spinel in which transition-metal and lithium atoms occupy, correspondingly, the 16a and 16b positions. The effects of the size and the charge of a transition-metal ion on the formation of a particular lattice type have been discussed.6 As was shown by calculation, if the ionic radii of alkali and transition metals are close, the random distribution of ions over the positions corresponds to the state with the minimum energy, viz., the cubic lattice of a-LiFeO2. With an increase in the differences in the ionic sizes, the probability of formation of the a-NaFeO2-type layered structure also increases, which is favoured by lower charges of transition-metal ions.The low stability of spinel phases (e.g., the low-temperature LiCoO2 phase), which exist only in very narrow temperature and pressure intervals is explained by the fact that their origination is governed by kinetic factors rather than by thermodynamic ones.778 LiScO2 represents a nonconducting white powder under ordinary conditions.A review of the structural data one can see in Ref. 4. LiScO2 was first described in 1959 (Ref. 7). X-Ray diffraction studies carried out for both single crystals 8 and powders 9 and MoÈ ssbauer studies of Fe3+ impurity centres in LiScO2 (see Refs 10, 11) have shown that the perfect ordering of Sc3+ and Li+ cations in the basal NaCl structure produces a superlattice with I41/amd symmetry (a=4.18, c=9.30 A), which is similar to the superlattice of totally ordered g-LiFeO2 phase. This structure is characterised by the atoms of transition and alkali metals regularly occupying the alternate positions along the tetragonal c axis. Yet another approach to the description of the LiScO2 structure is known.12 According to experimental results,12 the oxygen octahedra which enclose lithium ions are shown to be distorted in a such way as if lithium represented a Jahn ± Teller ion (four short and two long Li7O distances).The concept of distorted anatase seems to describe the LiScO2 structure more precisely than the model of a tetragonally ordered NaCl modifi- cation. LiTiO2 has a completely disordered cubic structure of the NaCl type (a=4.140 A).13, 14 Extensive information on LiTiO2 is cited in Ref. 4. Low-temperature chemical lithiation of LiTi2O4 spinel gave rise to a modification of LiTiO2.15 X-Ray diffraction patterns of this phase virtually did not differ from those of the parent spinel. Using the neutron diffraction technique, it was shown 16 that the formation of LiTiO2 (Li2Ti2O4) (a=8.3756 A) is accompanied by transposition of lithium from tetrahedral sites to octahedral ones unoccupied in the parent spinel.Thus, the titanium and lithium ions in the metastable phase are distributed over the positions 16c and 16d, respectively, of the Fd3m space group, and a superstructure of the atacamite type is formed.17 Heating above 920 K results in irreversible transition of the Li2Ti2O4 spinel into a stable disordered cubic phase of LiTiO2.15 This transition appears to occur in several steps, because DSC (differential scanning calorimetry) revealed three endothermic effects (in total, 950 cal mol71 were liberated).18 LiVO2 represents a black powder. It was described for the first time in 1954 (Ref.19). This compound is isostructural to the hexagonal a-NaFeO2. Generally, such a structure is typical of this class of compounds (Fig. 1). It is formed by the ions of alkali and transition metals ordered in parallel (111) planes of the NaCl cubic lattice. In the ordered structure, all vanadium ions have six-fold oxygen coordination, the V7O distances are equal to 1.987 A. At 450 K, a phase transition occurs in LiVO2,13, 20 ± 22 which is accompanied by pronounced changes in the unit cell volume and interatomic distances [a=2.837, c=14.775, V7O=1.98, ONi (V, Cr, Co) Li Figure 1. Schematic representation of the formation of layered struc- tures for LiVO2, LiCrO2, LiCoO2 and LiNiO2 as a result of cationic ordering in the NaCl-type cubic lattice.D G Kellerman V7V=2.84 A at 300 K; a=2.913, c=14.640, V7O=2.02, V7V=2.91 A at 550 Kfor Li0.7VO2 (see Ref. 23)], although the main structural motif remains unchanged. A substantial temper- ature hysteresis was observed. During the transition, the enthalpy changed by 50 ± 1000 cal mol71 (see Ref. 22). It was assumed 24 that at room temperature vanadium ions are combined in V3 trimers which disintegrate during the phase transition. The formation of trimers was confirmed experimentally. Thus, the structure of Li0.8VO2 single crystals prepared by the flux method (see Ref. 26) was analysed in detail using X-ray diffraction and EXAFS techniques.25 It was shown that vanadium ions are shifted by 0.160 A from the positions they should occupy in the perfect hexagonal structure. Probably, the symmetry of the low-temper- ature phase differs from the R3m symmetry as followed from the presence of very weak additional reflexes in the X-ray patterns of this phase (a 0=a 3, c 0=c).23 pÅÅÅ It was shown that stoichiometric LiVO2 can be transformed into Li17xVO2 phases by both chemical and electrochemical deintercalation.27, 28 Studies on the structure of deintercalated vanadates showed that the removal of lithium from the layered structure for x50.3 results in redistribution of vanadium ions.For example, in Li0.22VO2, approximately one third of vanadium ions occupy the octahedral positions in Li layers, which favours stabilisation of the cubic structure. With increasing temperature, the Li17xVO2 phase becomes unstable, and its disproportionation can proceed by two routes 29 (172x) LiVO2+x LiV2O4 , Li17xVO2 (178x) LiVO2+8x Li0.875VO2 .Li17xVO2 The coexistence of hexagonal and cubic phases was estab- lished.30 At high pressures, another polymorphous modification of LiVO2 with a superlattice of the atacamite type was obtained.31 Heating above 700 K resulted in irreversible transition of this phase into a stable hexagonal form. LiCrO2 prepared by ceramic technology represents a green powder with a layered trigonal structure of a-NaFeO2 (a=2.88, c=14.6 A). Like LiVO2, the LiCrO2 oxide was first prepared and described in 1954 (Ref. 19). Single crystals of LiCrO2 were grown by the flux method.32 In contrast to most LiMO2 compounds, deintercalation of lithium from LiCrO2 is hindered; however, under hydrothermal conditions, the ion-exchange reaction is possible, which results in the formation of an isostructural phase of HCrO2.33 LiMnO2 in the form of a brown powder (orthorhombic unit cell) was first synthesised with the aim of studying the LixMn17xO system.34 The crystal structure of LiMnO2 was studied in more detail for polycrystalline material 35 and single crystals 36, the latter being obtained by the hydrothermal method.37 A number of structural peculiarities of orthorhombic LiMnO2 compared with other LiMO2 complex oxides 38 are apparently caused by the Jahn ± Teller effect.pÅÅÅ The orthorhombic lattice of LiMnO2 (a=2.805, b=5.757, c=4.527 A), Pmmn (see Ref.36) also represents one of the versions of the NaCl basal structure ordering (Fig. 2). The MnO6 octahedra are substantially distorted due to the Jahn ± Teller effect. Two long distances along the b axis (2.291 A) and two pairs of almost identical equatorial distances, viz., 1.920 and 1.937 A, were observed. In each (001) plane, cations of one type are arranged in chains along either h100i (orthorhombic) or h110i (cubic) directions. In alternate (001) planes, the chains retain their orientation with respect to the upper and lower planes, but are shifted forming the ABA packing along the c axis. Compared to the hypothetical cubic lattice with a parameter a 0, the parameters pÅÅÅ of the real lattice are as follows: a = (1/2) 2a 0, b= 2a 0 and c=a 0.Considerable anisotropy of the thermal expansion coefficient was observed 39 at temperatures<300 K. The thermal expansion coefficient did not exceed 1077 K71 in the directions a and b and was 3.761076 K71 in the direction c.Magnetic properties of complex oxides LiMO2 (M=Sc ± Ni) with different types of cationic ordering OMn Li Figure 2. Crystal structure of LiMnO2. Studies on the morphological peculiarities of orthorhombic LiMnO2 synthesised by a conventional ceramic procedure (sinter- ing of an Mn2O3+LiOH mixture at 1073 K in an inert atmos- phere) revealed the presence of certain cationic disordering, namely, the existence of defects of the LiMn and MnLi types.40 The degree of disordering (3% ± 12%) was attributed to the preset excess or deficiency of LiOH.Scanning electron microscopy showed that the sizes and shapes of grains also depended on the LiOH content in the reaction mixture. The deficiency of lithium favoured the formation of fine and size-uniform grains, whereas excess of lithium resulted in a wide scatter of the grains both in sizes and shapes. It was the former modification that demon- strated good characteristics in the electrochemical process.41 The stacking faults in orthorhombic LiMnO2 (d&0.3 mm) grain size were studied in detail.42 It was concluded that these represent monoclinic interstitial layers between the orthorhombic blocks. The portion of such defects was 1%± 6%, which correlated well with the degree of cationic disordering.with a small The existence of a low-temperature modification of LiMnO2 with a tetragonal structure, which was obtained by chemical lithiation of LiMn2O4 spinel: a=5.662, c=9.274 A, I41/amd, Z=8, was reported.43. 44 Above 500 8C, the tetragonal phase transformed into the stable orthorhombic phase. Recently, yet another metastable low-temperature modifica- tion of LiMnO2 has attracted great interest. This phase was first synthesised in 1996 (Ref. 45). The phase has a monoclinic lattice (C2/m, a=5.4387, b=2.8086, c=5.3878 A, b=116 8). For its synthesis, the ion-exchange method was applied in which layered a-NaMnO2 was used as the precursor. The product also exhibited a layered structure.In addition, it was noted 46, 47 that the ion- exchange reaction can also proceed in a mixture of g-MnOOH and LiOH .H2O. The lower symmetry of the layered LiMnO2 mod- ification compared, for instance, with layered LiCoO2 was asso- ciated with the strong Jahn ± Teller effect for Mn3+ ions (see Ref. 45). As was shown by the results of X-ray diffraction analysis and neutron diffraction,45, 48, 49 the Li-layers of the layered LiMnO2 modification contain 3%± 9% of manganese ions. An electron-diffraction study in a convergent beam 50 have shown that a typical specimen synthesised by the ion-exchange method contains, in addition to layered LiMnO2, an admixture of a tetragonal spinel phase of Li2[Mn2]O4 and traces of orthorhombic LiMnO2.The following atomic planes and directions were also shown 50 to be equivalent for the spinel and layered phases: (200)lay:(222)sp (020)lay:(4-40)sp [001]lay:[11-2]sp 779 A comparative analysis of the stability of orthorhombic and monoclinic layered structures of LiMnO2 was carried out.51 The calculations have shown that the antiferromagnetic exchange coupling of Mn3+ ions plays a substantial role in stabilisation of the orthorhombic structure; and, therefore, stabilisation of the layered structure requires that manganese was partly replaced by elements that weaken this interaction. The elements Al, Co, Cr, V, Ti, Mo, Mg, Nb, Zn and Pd can serve as the dopants. In the Li7Mn7O system, yet another monoclinic phase designated as m-LixMnO2 (x=0.2 ± 0.3) was detected and iden- tified.52 It can be obtained by durable sintering of intimately mixed b-MnO2 and LiOH at 420 K.According to the results of X-ray diffraction analysis, the new phase has a primitive mono- clinic unit cell (a=9.38, b=5.65, c=4.906 A, b=102.2 8) and is stable up to 820 K. In a thermobaric treatment (P=6.5 GPa, T=100 K) of orthorhombic LiMnO2, a totally disordered phase with the rock- salt structure (a=4.1658 A) was obtained. According to crystal- lographic data, no structural distortions caused by the Jahn ± Teller effect were observed.4 LiFeO2 exists in several polymorphous modifications (Fig. 3). A disordered structure of cubic a-LiFeO2 is identical to that of NaCl (a=4.158 A).53 The structure of LiFeO2 represents a rare example of structures in which two different cations occupy the same positions.54, 55 The results of electron-microscopic studies have shown 56, 57 that the a-LiFeO2 cubic structure involves the elements of the short-range order.For g-LiFeO2, an ordered tetragonal structure similar to that of LiScO2 is typical (a=4.057, c=8.759 A).58 Apparently, the metastable b-LiFeO2 modification is intermediate between a- and g-phases. Discrepant opinions were advanced on the b-LiFeO2 structure: some authors (see, e.g., Ref. 53) claim the existence of a body-centred tetragonal unit cell, while others believe it to be monoclinic.59, 60 The monoclinic structure of the b-phase was confirmed by electron microscopy 56 (C2/c, a=8.571, b=11.59, c=5.147 A, b= 145.70 8, Z=8).The tetragonal structure exhibits no ordering in the distribution of cations over the positions. In contrast, the monoclinic phase is ordered. Detailed studies 53, 61 have detected the transition of a-LiFeO2 first to b-LiFeO2 at T>570 K and further to the ordered g-LiFeO2 phase at T>870 K. The latter phase transforms back into the disordered a-modification with further increase in tem- perature (> 920 K). The transitions proceeded slowly and depended on the sample's prehistory. Currently, many efforts are concentrated on the synthesis of metastable modifications of LiFeO2 with layered structures. For b a c OLi Fe Li, Fe Figure 3. Structural modifications of LiFeO2; (a) is the cubic a-phase, (b) is the tetragonal b-phase and (c) is the tetragonal g-phase (1/2 of the unit cell is shown).780 example, the preparation of LiFeO2, the corrugated layered structure of which is similar to that of a stable orthorhombic modification of LiMnO2, was described.62 The synthesis involved the ion exchange reaction between g-FeOOH and LiOH .H2O.The reaction was carried out under hydrothermal conditions at 370 ± 520 K. The structure of the phase obtained was noted to resemble very closely that of the original g-FeOOH. The transition from g-FeOOH to LiFeO2 presumes a shift of corrugated layers by 1/2 along the b direction. According to the X-ray diffraction studies, the lattice parameters of LiFeO2 are a=4.0610(5), b=2.9621(5) and c=6.0319(11) A.Yet another orthorhombic structure of the goethite type with the parameters a=9.677(5), b=2.934 and c=5.005 A, Pmna was also synthesised by using an ion-exchange reaction (carried out in an ethanolic solution, in contrast to the previous reaction).63 It is interesting that at certain Li to Fe ratios, a totally disordered cubic modification of a-LiFeO2 (a=4.161 A, Fm3m) can be obtained by this method,64 its conventional synthesis requires high temperatures. The ion-exchange reaction of NaFeO2 in a lithium chloride melt resulted in LiFeO2 with a layered trigonal structure of a-NaFeO2, which is typical of this class of compounds.65 The degree of cationic ordering was found to depend on the ordering of the precursor phase.In addition to metastable forms of LiFeO2 discussed above, yet another low-temperature modification, viz., a phase with a tunnel structure was documented.66 However, this has not been described in detail in further publications. LiCoO2 is usually obtained at 1200 ± 1300 K as a black powder from cobalt oxide and lithium carbonate. This phase is referred to as a high-temperature (HT) one. It is thermally stable up to 1120 K.67 According to results of neutron diffraction studies,68 HT-LiCoO2 has a perfectly ordered structure of the a-NaFeO2 type (R3m) with the cell parameters a=2.8166 and c=14.052 A.69 The LiCoO2 structure was studied for single crystals obtained by the flux method.70 The layered structure of HT-LiCoO2 is very stable and, as was shown by calculations,71 the order ± disorder transition can occur only at *5100 K.This is substantially higher than the melting point, which makes such a transition practically unfeasible. On the other hand, according to calculations,72 the use of high pressures can stimulate the tran- sition with the formation of a cubic phase. Synthesis at temperatures below 700 K produces a low- temperature (LT) phase with a structure somewhat different from that of HT-LiCoO2.73 ± 81 According to the results of neutron diffraction analysis,75 nearly6%of lithium sites in LT-LiCoO2 are occupied by cobalt. The phase formed upon deintercalation of lithium from LT-LiCoO2 (LixCoO2; 15x50.5) can be described77 either as a spinel modification (Fd3m) in which up to 80% of lithium atoms occupy tetrahedral positions74 or a phase with the trigonal layered structure (R3m).NMR studies 82 are in favour of the second version, because the substantial structural disordering observed by this method corresponds to the second co-ordination sphere of lithium. Hence, the nearest neighbouring of lithium remains octahedral so that LT-LiCoO2 may be consid- ered as a partly disordered trigonal structure based on NaCl. A different opinion on this problem has been advanced.83 Based on the combined use of transmission electron microscopy and X-ray diffraction analysis, it was concluded that the structure of LT- LiCoO2 may be considered as an intermediate between the structures of perfect spinel and the layered modification.Anumber of studies dealt with the phase transitions that occur both in HT- 84 ± 86 and LT-LiCoO2 75 ± 77 during electrochemical deintercalation of these compounds. The stabilities of the phases obtained were calculated,87, 88 and the phase diagram of LixCoO2 was constructed. The completely deintercalated layered phase of CoO2 (a=2.822, c=14.29 A) was first isolated and described in 1997 (Ref. 89). Unexpectedly, it was found that despite the structural disordering processes that occur during electrochemical deinter- D G Kellerman calation the product exhibited a very high degree of structural ordering. Several metastable polymorphous modifications were formed upon electrochemical deintercalation of LixCoO2 90 syn- thesised from Na0.7CoO2 by ion-exchange.91 Depending on the depth of the deintercalation process, the mode of CoO6 octahedra packing changed.In studies of phase transitions that occur during the chemical extraction of lithium from LiCoO2 with the formation of Li0.5CoO2 it was found 92 that a common hexagonal structure is retained only up to the Li0.92CoO2 composition. For deeper extraction of lithium, a second, also hexagonal phase appears, which becomes the only phase present starting from the Li0.79CoO2 composition. The hexagonal structures differ in the c/a ratio, which is substantially higher for the second phase. LiNiO2 with a structure similar to that of a-NaFeO2 (a=2.878, c=14.19 A) was described for the first time in 1953 (Ref.93). This oxide is isostructural to LiCoO2 and LiVO2. The distribution of atoms over the positions is as follows: Ni(3a) (0, 0, 0); Li(3b) (0, 0, 1/2 ); O(6c) (0, 0, 1/4 ) (0, 0,71/4). The ratio cubic sublattice of oxygen, for which c/a=2 ÅÅÅ6 p =4.90. An c/a=4.93 indicates the presence of an almost perfect face-centred EXAFS study of the Jahn ± Teller co-operative effect has shown that two long (2.09 A) and four short (1.91 AN7O distances are present as a result of local distortion in NiO6 octahedra.94 ± 95 The layered LiNiO2 oxide is traditionally considered as an end member of the series of LixNi17xO solid solutions. The cationic ordering that takes place with increasing concentration of lithium in LixNiO2 is well known and was discussed in a number of publications.96 The disordered phase (x<0.56) is isostructural to NiO (NaCl type), and the formation of a layered structure involves the ordering of ions of transition and alkali metals in alternate (111) planes of the NaCl-type cubic lattice, which is accompanied by lowering of the symmetry to trigonal.Due to the close radii of Ni3+ and Li+ ions, the degree of ordering of cations over the positions in the alternate planes (Z) can vary widely even for an invariant transition-to-alkali metal ratio. For example, the synthesis and properties of virtually stoichiometric LiNiO2 with 04Z41 were described.97, 98 The structural order Z=0 corre- sponds to cubic LiNiO2 prepared by quenching of LiNiO2 with Z=1 from a temperature of 1200 K.97 The quenched phase is metastable; hence, the ordering process can occur at temperatures above 600 K.The temperature of the order?disorder transition (layer- ed?cubic structure) is 1020 K.99 The transition proceeds slowly and is not completely reversible. The difficulties of synthesising the stoichiometric LiNiO2 were associated with the loss of lithium during sintering.100 ± 102 In this case, the parameter of cubic or trigonal cell depended linearly on the value of x in LixNi17xO.96 Different versions of the synthesis, which allow one to minimise lithium losses, were described.103 It is of note that even upon annealing for a long time at sufficiently high temperatures (1100 K) LixNi17xO is character- ised by a heterogeneous bulk composition.104 The procedure of refining of structural data turned out to be more efficient being carried out within the framework of a multiphase system rather than a `single-phase' one.The cationic distribution in nonstoichiometric Li17xNi1+xO2 solutions was studied by X-ray diffraction analysis.105 It was confirmed that, with an increase in x, the degrees of occupation of 3a and 3b positions by lithium (nickel) ions approach one another, which results in their random distribution for x50.4. Two models of cationic distribution in Li17xNi1+xO2 exist. According to the first model, superstoichiometric nickel occupies free sites in lithium layers, and the oxide can be described by the formula [Li17xNix][Ni]O2.106 The second model assumes partial mixing of cations in both sites; in this case, the oxide has the following formula: [Li17x7dNix+d][Ni17dLid]O2.105, 106 If the oxide composition approaches stoichiometric, then the first model is assumed107 to be more probable.Magnetic properties of complex oxides LiMO2 (M=Sc ¡¾ Ni) with different types of cationic ordering It should be noted that the nickel that occupies 3a positions in the nickel sublattice of Li17xNi1+xO2 corresponds predominantly to the Ni3+ state, whereas the lithium sublattice (2a positions) mainly involves bivalent nickel ions.108 It was shown 106 that the concentration-induced order ¡¾ dis- order transition in Li17xNi1+xO2 is well simulated by using the `lattice gas' concept.The dependences of the ordering parameter on the lithium content were obtained, and Zmax=0.96 was found even for the stoichiometric composition.Moreover, it was shown that even the region of the existence of cubic structures (x50.4) is characterised by the effects of the short-range structural order, which were studied in detail by synchrotron X-ray emission.109 Based on the results obtained, the sizes of short-range order domains were assessed. A small magnitude of the X-ray scattering factor for lithium makes it impossible to determine the degree of cationic disorder- ing in Li17xNi1+xO2 near the stoichiometric composition. For this purpose, other methods, such as ESR 110 or X-ray absorption spectroscopy 111 were used. Thus it was found 111 that two differ- ent Ni7O distances (2.07 and 1.94 A) whose ratio varies with x are present.Three successive first-order transitions which occur during electrochemical extraction of lithium 112, 113 (at high potentials, completely deintercalated `NiO2' phases can also be obtained 114) cause the following changes in the LiNiO2 structure: hexagonal?monoclinic?hexagonal 0?hexagonal 00. The rela- tionships between the unit cell parameters of monoclinic (C2m, similar to that observed for NaNiO2) and hexagonal phases are as follows: pAAA am= 3ah , bm=ah ,h cm= c 3sinb , h b=180 ¢§ arctg 3 pAAAah . c It is also important that the monoclinic phase is only formed upon deintercalation of virtually stoichiometric Li17xNiO2, whereas intercalation of Li17xNiO2 with x50.05 does not affect the symmetry.115 Stoichiometry of the initial compound also affects the process of chemical deintercalation: with a decrease in the lithium content, extraction of lithium and admixed nickel from lithium layers is supplemented by ion exchange which involves protons.116 The presence of nickel in the lithium layers plays a significant role in the stabilisation of the monoclinic phase.117 Using EXAFS methods, it was shown 118 that the local distortion of NiO6 octahedra in the monoclinic deintercalated phase of LixNiO2 (0.5<x<0.75) is substantially weaker compared with the stoi- chiometric trigonal phase.This was attributed to the electron exchange between Ni3+ and Ni4+ sites. Also, the results of electron-microscopic studies have shown the monoclinic phase to involve a superstructure formed by partially ordered lithium ions (vacancies in the lithium layers).119 The deintercalated LixNiO2 phase is thermally unstable, and the layered hexagonal structure gives way to the cubic spinel structure at 453 K, which is accompanied by oxygen loss.120 Efforts were undertaken to obtain the Li1+xNiO2 oxide superstoichiometric with respect to lithium.121, 122 For instance, a phase with x50.3 having a monoclinic structure was described.122, 123 The latter was formed by closely packed oxygen layers (slightly distorted) which separated purely lithium layers from those filled by 1/3 with lithium ions and by 2/3 with nickel ions.781 III.Magnetic properties of LiMO2 oxides Structural differences in the series of LiMO2 oxides, whereMis a transition 3d metal, are not significant: the crystal structures differ only in the mode in which the cations of alkali and transition metals are ordered in the planes of the parent cubic lattice. Nonetheless, these compounds exhibit fairly diverse magnetic properties. LiVO2. Magnetic effects which suggest phase transitions in LiVO2 were first described by Bongers.13 Heating above 450 K results in a fourfold increase in the magnetic susceptibility (Fig. 4); at higher temperatures, the latter is practically independ- ent of the temperature. To explain the magnetic properties of LiMO2 in the region below the transition temperature, a mecha- nism of formation of molecular orbitals in the basal plane of the hexagonal LiVO2 structure was proposed.125 The trigonal V3 clusters thus obtained are nonmagnetic due to Heisenberg anti- ferromagnetic exchange coupling, and the phase transition is associated with their decomposition.1074 w71 /g cm73 80 60 40 200 600 1000 T /K 200 Figure 4. Temperature dependence of the reciprocal magnetic suscepti- bility of LiVO2.124 Another model proposed 4, 126 does not imply any changes in the symmetry. This interprets the low-temperature magnetic data in terms of the existence of a two-dimensional `metal' cluster built by vanadium ions bound by metal7metal bonds. In the high- temperature region, d electrons are delocalisated, and LiVO2 becomes a narrow-zone metal.This fact was evidenced by IR spectra, results on electrical conductivity 4 and the data from X-ray absorption spectroscopy.127 Phase transitions which, as in the above model, were attributed to the changes in the critical M7M distance are typical of many transition-metal simple oxides: V2O3, VO2 , V4O7, Ti3 O5, Ti4 O7.128 ¡¾ 131 The role the structural changes in LiVO2 play in the first-order phase transition from nonmagnetic to magnetic phase was discussed in detail.132 It has been found that due to the changes in the magnetic energy the high-temperature phase is more stable. LiCrO2. The studies on the magnetic properties of LiCrO2 as well as of many other oxides were pioneered by Bongers.13 Magnetic susceptibility data indicated antiferromagnetic ordering to exist below 300 K and the Curie ¡¾ Weiss paramagnetism with Y=7577 Kand mef=3.71 mB to be present above 300 K.These characteristics were refined later: the NeA el temperature (TN) was found to be 56 K133; the value of mef obtained by ESR techni- que 134 was 3.8258 mB (the deviation from an intrinsic spin value by 0.0529 mB was attributed to spin-orbit coupling). Many studies have been made of the magnetic structure of LiCrO2.135 ¡¾ 137 The structure was assumed 135 to be two-dimen- sional with an exchange-coupling constant inside the layer of J=7391 K. Based on the results of neutron scattering on LiCrO2 single crystals, a magnetic structure with a single prop- agation vector k=[1/3, ¡¾2/3, 3/3] has been proposed.137782 The relative orientation of the spins of neighbouring Cr3+ ions in the magnetic planes of LiCrO2 was explained by the competition of direct antiferromagnetic interactions and indirect ferromagnetic double exchange coupling.In many studies (see, e.g., Refs 138, 139), the structure of LiCrO2 was considered as a Heisenberg trigonal lattice. Such compounds attract attention due to the fact that ferromagnetic compounds can involve different magnetic orders compared with the classical Ne el order. A joint analysis of the data on suscepti- bility, neutron scattering and magnetisation have led to the following conclusions:140 first, during the magnetic phase tran- sition, a 120-degree magnetic structure which involves Cr3+ ions is formed in each layer; in the process, the magnetic ordering both in the layers and along the c axis occurs at the same Ne el temperature; second, the magnetic structure is characterised by two wave vectors k=[1/3, 1/3, 0] and [72/3, 1/3, 1/2].Figure 5 shows the temperature dependence of the magnetic susceptibility of LiCrO2. 103 w71 /cm3 mol71 2.2 ck 2.0 TN c\ 1.8 100 0 T /K 200 Figure 5. Temperature dependence of the magnetic susceptibility of LiCrO2.140 LiMnO2. The first information on the magnetic properties of orthorhombic LiMnO2 can be found in Ref. 13, in which the antiferromagnetic transition at 300 K was deduced. This fact was confirmed by more detailed studies.39, 141 The magnetic suscepti- bility variation is characterized by the following features: the paramagnetic behaviour follows the Curie ± Weiss law at temper- atures above 600 K(Y=71056 K, mef=4.82 mB perMn3+ ion); the magnetic susceptibility passes a broad maximum at 400 Kand its polytherms diverge below 100 K, being measured upon cooling both in the zero field and in the 100-ê applied field (Fig.6). Based on the results on neutron scattering, the magnetic structure of LiMnO2 was supposed to have a propagation vector k= [1/2, 1/2, 1/2] and a magnetic moment equal to 3.69 mB per Mn3+ ion and preferentially oriented along the b axis. The Ne el temperature, which characterises a phase with three-dimensional magnetic order, is 261.5 K;141 above this value, the effects of short-range 2D order are observed.Yet another magnetic effect observed in LiMnO2 at 50 K was probably the result of weak ferromagnetism.139, 141 Its magnitude and the temperature of its manifestation depended on the stoi- chiometry. It was shown 142 that even a small number of structural defects can radically change the magnetic characteristics of LiMnO2 in this region. The magnetic susceptibilities of Li0.99MnO2, Li1.01MnO2 and LiMnO1.99 compositions were meas- ured. Presumably, the manganese ions present in lithium sites which serve as the bridges between the double layers of manganese are the most probable reason for the formation of the three- dimensional magnetic structure.142 This is why Li0.99MnO2 was assumed to be more prone to three-dimensional ordering than LiMnO2.At the same time, the presence of excess lithium in Li1.01MnO2 should apparently hinder the formation of defects of the MnLi type and, hence, the formation of the three-demensional magnetic structure. The results obtained for defective samples confirmed this assumption: indeed, the appearance of MnLi defects favoured the formation of the three-dimensional magnetic structure. An analysis of magnetisation curves have shown the a 106 w71 /cm3 g71 26 24 22 20 400 0 b 106 w71 /cm3 g71 2 160 120 80 40 1 0 150 50 Figure 6. Temperature dependence of the magnetic susceptibility of LiMnO2 in the regions (a) 100 ± 1000 K and (b) 4 ± 300 K.141, 142 (1) Cooling in a zero field; (2) cooling in a 100-ê field. spontaneous magnetisation to increase in the sequence Li1.01MnO2, Li0.99MnO2, LiMnO1.99 and to be equal to 0.282, 0.333 and 0.421 G cm3 g71, respectively. The coercive force increases in the same sequence: 1100, 2651 and 5974 ê.The fact that the temperature of magnetic ordering is higher for lithium- deficient and oxygen-deficient compositions compared with stoi- chiometric LiMnO2 also confirms the hypothesis put forward. A cubic modification of LiMnO2 obtained at high pressure exhibits antiferromagnetic properties with TN=40 K, Y=7200 K and m=4.90 mB.4 LiFeO2. At low temperatures, all three modifications of LiFeO2 (a, b and g) are antiferromagnetically ordered (Fig. 7); the Ne el temperature is 42 K for a- and b-LiFeO2 and 295 K for g-LiFeO2. For all three modifications, the temperature depend- ences of the magnetic susceptibility were shown to deviate in the paramagnetic region from the Curie ± Weiss law, which was 1074 w71 /g cm73 3.0 1 2.5 2.0 2 1.5 3 1.0 0.5 0 400 200 Figure 7.Temperature dependence of the reciprocal magnetic suscepti- bility of LiFeO2.57 (1) g- LiFeO2; (2) a-LiFeO2: (3) b-LiFeO2. D G Kellerman 800 T /K 250 T /K T KMagnetic properties of complex oxides LiMO2 (M=Sc ± Ni) with different types of cationic ordering associated with the fact that distributions of iron and lithium ions involve elements of the short-range order. The magnetic structures of a- and g-LiFeO2 were found 143 in studies on the MoÈ ssbauer effect and from the results of neutron scattering.The latter structure is characterised by tetragonal magnetic symmetry with strong antiferromagnetic superexchange coupling of Fe3+ ions in (001) planes and much stronger interplane interaction. Magnetic moments of ions are directed along the [001] axis. The magnetic structure of the a-LiFeO2 cubic phase is characterised by ferro- magnetically ordered spins inside parallel (111) planes, the latter being antiferromagnetically bound with one another. For this structure to appear, it is necessary that lithium and iron ions are substantially separated in alternate (111) planes. Such a trend was demonstrated by the studies on the MoÈ ssbauer effect. Magnetic properties of a metastable layered modification of LiFeO2 prepared by the hydrothermal method were described.144, 145 The antiferromagnetic ordering was shown to exist below 20 K.No effects associated with the transition into the spin-glass state typical of the isostructural LiNiO2 phase were observed. It was shown that variations in the degree of cationic ordering affect the Ne el temperature. LiCoO2. Trivalent cobalt in LiCoO2 is in the low-spin state 146 (ta2g)3 (tb2g)3 (eg)0, hence, its magnetic moment is equal to zero. This is why the magnetic susceptibility of LiCoO2 should not depend on the temperature. However, this statement is only true for stoichiometric specimens.147 The defects present in LiCoO27x and the attendant paramagnetic centres which are apparently repre- sented by Co2+ ions increase the magnetic susceptibility and make it temperature-dependent.This was confirmed by the results of thermogravimetry. The paramagnetic component was also described for LiCoO2 superstoichiometric with respect to lith- ium.82 Above 750 K, the polytherms of magnetic susceptibility of both stoichiometric and defective LiCoO2 were found to ascend reversibly 13, 147 (Fig. 8). Such an increase with the temperature indicated the appearance of paramagnetic centres. The following assumptions on the nature of these centres were drawn. 104 w /cm3 mol71 2.4 2 2.4 1.6 1 1.2 0.8 800 T /K 400 Figure 8. Temperature dependences of the magnetic susceptibility of LiCoO2.147 (1) Synthesis and measurements in an oxygen atmosphere; (2) synthesis and measurements under reduced pressure.1. At temperatures above 750 K, thermally activated centres appear in LiCoO2 as a result of oxygen exchange with the gas phase. The following process of defect formation is possible, which results in the formation of paramagnetic centres Co2+ (Co 0Co): 2CoCo+OO (1/2)O2(g)+VO..+2Co0Co . The appearance of each oxygen vacancy leads to the forma- tion of two paramagnetic centres.147 2. The reason for the increase in the magnetic susceptibility is the transition of Co3+ ions to the high-spin state (ta2g)3 (tb2g)1 (eg)2 (m2ef=24 m2B).13 783 The magnetic susceptibility of chemically delithiated LiCoO2 was studied.92 This was shown to obey the Curie ± Weiss law in the temperature interval from 83 to 300 K for all compositions irrespective of the degree of lithium extraction.The values of the effective moment were as follows: Oxide Li0.61CoO2 Li0.76CoO2 Li0.92CoO2 LiCoO2 1.733 1.731 1.164 1.157 mef /mB The effective magnetic moments of the latter two composi- tions almost coincided. This was associated with changes in the degree of electron localisation. LiNiO2. The magnetic properties of layered LiNiO2 depend on both the stoichiometry and the degree of structural order. The results of numerous experiments were often conflicting. This also concerns the manifold interpretations of the data obtained. Thus it was concluded 96 that magnetic properties of LiNiO2 can adequately be described on an assumption that trivalent nickel ions are in the low-spin state [(ta2g)3 (tb2g)3 (eg)1].Later, this fact has found an experimental confirmation 148 and was substantiated theoretically.149 However, in a number of studies (e.g., in Ref. 150), trivalent nickel is assumed to be in the high-spin state [(ta2g)3 (tb2g)2 (eg)2], which forms a Kramers doublet with S= 1/2. A different viewpoint exists: the magnetic properties of LiNiO2 are determined by the exchange-coupled (Ni2+7O17) pairs.151 Charge-compensation processes in LiNiO2, which are responsible for the appearance of holes in the oxygen zone, resemble those observed in high-temperature superconductors and were revealed for the first time in the studies of X-ray absorption spectra for the LixNi17xO system.151 Similar results were obtained in other studies.152, 153 Calculations 154 confirmed that an electron hole appears in the 2p oxygen band rather than in the 3d nickel band.However, the fact that such techniques as 7Li NMR and ESR155 have not confirmed this observation deserves mention. In a number of studies, the magnetic properties of LiNiO2 were considered in relation to the problem of the ground state of a two-dimensional (2D) triangular lattice antiferromagnet with S=1/2. Thus within the framework of the model of a quantum liquid (resonance valence zone), it was assumed for the first time 156 that the ground state of such systems in the absence of long-range magnetic order has a low energy compared with the classical Ne el state.This problem was the subject of wide speculation in the context of the assumption that it is the resonance valence zone that is responsible for high-temperature superconduction in copper complex oxides.157 The LiNiO2 oxide characterised by S=1/2 and a structure based on face-centred cubic packing of oxygen atoms where nickel and lithium atoms occupy the interstices and form alternate layers (as a result, the interlayer distances substantially exceed those between magnetic atoms inside a layer) can well be described by the aforementioned model.150, 155, 158, 159 The magnetic properties of a number of LixNi17xO solid solutions and LiNiO2 as the end member of this series were studied in detail.96 The main attention was focused on the magnetic transition in the temperature range of 200 ± 300 K (this temper- ature is usually denoted as TN1), which was claimed to be of the ferromagnetic nature.The dependences of the Curie temperature and magnetisation of LixNi17xOon concentration had maxima at x=0.43 (Fig. 9), and the degree of structural order Z [the degree of cationic ordering in (111) alternate planes of the basal NaCl lattice] increased monotonically with an increase in the lithium content and reached a maximum for LiNiO2, which, however, was below 100%. It is the latter observation that explained the fact that magnetic measurements carried out later 97, 98, 160 ± 162 for LiNiO2 samples with a higher degree of structural ordering revealed no magnetic transition described in Ref.96. Thus, the magnetic ordering is an indicator of Z<1. The fact that the magnetic properties of LiNiO2 can indicate the structural ordering of a sample stems from the structural peculiarity of perfectly ordered LiNiO2 [namely, the fact that each plane occupied by784 M/mB Z 1.0 0.30 0.8 0.6 0.4 0.2 0.25 0.20 0.15 0.10 0.05 0 0 0.3 x 0.4 Figure 9. Concentration dependences of magnetisationMand degree of ordering Z in LixNi17xO.96 magnetic ions (Ni3+) is separated from another plane by three nonmagnetic planes (O7Li7O)] and the partial disordering caused by the presence of nickel ions in lithium sites (and vice versa). A complete cationic disordering (Z=0) is typical of cubic LiNiO2 the magnetic properties of which have been described.98 The polytherm of its magnetic susceptibility demonstrated an anomalous feature, which allowed the paramagnetic�antiferro- magnetic transition at 280 K to be deduced (Fig.10). 1074 w71 /g cm73 8.8 8.4 8.0 7.6 T /K 500 300 Figure 10. Temperature dependence of the reciprocal magnetic suscepti- bility of LiNiO2 with the degree of ordering Z=0.98 The magnetic transition occurring in this oxide at 210 K was studied in detail 150, 158, 159. The magnetic structure of LiNiO2 was simulated on an assumption of a perfect ordering. However, from the synthetic conditions mentioned (LiOH+Ni2O3; T=1100 K) suggest that the degree of ordering in the samples studied did not exceed Z=0.8.Below the transition temperature, LiNiO2 was found to exhibit the ferrimagnetic behaviour. However, in con- trast to classical ferrimagnetic materials, the field dependence of the magnetisation [M(H)] in LiNiO2 was nonlinear below TN1, did not reach saturation and infinitely increased with the field strength according toM!H1/d. It should be noted that the possibility of assigning the magnetic transition which occurs in LiNiO2 with Z<1 in the temperature interval of 150 ± 300 K to the ferrimagnetic type 96 was not confirmed later. In neutron diffraction studies,150, 163 magnetic reflexes were not observed below the transition point even on long-term exposures. The ordering at 240 Kwas found 161 to be of a ferromagnetic nature.The same conclusion was made in Ref. 98. In addition to the magnetic transition observed in the region of 150 ± 300 K, the temperature and even the possibility of which depend on the degree of the structural order of LiNiO2, other anomalous magnetic features were observed at lower temper- atures. A broad spectrum of mechanisms were invoked for their a 103 w /cm3 mol71 10 TN2 1 TN3 0.1 0.01 100 0 b 103 w /cm3 mol71 1.2 21 1.0 0.84 8 12T/K Figure 11. Temperature dependences of the magnetic susceptibility of LiNiO2 measured in an applied field of (a) 1000 165 and (b) 10 ê.162 (1) Cooling in a zero field; (2) cooling in a 10-ê field. explanation. For example, virtually stoichiometric and perfectly ordered LiNiO2 with Ni7Ni interactions of the 2D type was shown 160, 162, 164 to behave as a typical spin glass.The temperature dependence of the magnetisation measured upon zero-field cool- ing demonstrated a sharp peak at 9 K (this temperature was denoted as TN2; according to the results of other authors, T is 35 ± 50 K, see, e.g., Ref. 165). At the same time, upon cooling in a nonzero field, the magnetic susceptibility increased monotonically (Fig. 11). The frequency dependence of the height and position of the peak, which was obtained by measuring the AC susceptibility, also indicated the attainment of a spin glass state. In certain studies,159, 166 the state below TN2 was considered as the quantum liquid state in which the spins form mobile pairs.The temperature TN2=6 K was claimed 161 to be the Curie temperature for a two- dimensional Ising system. Such a conclusion was drawn by taking into account the fact that the slope of the linear dependence ln(M/H) vs. ln(T7Tc) was found to be equal to 71.7, which agrees with the critical factor of77/4 for the 2D model of Ising.167 The temperature dependence of the magnetic susceptibility of LiNiO2 demonstrated yet another special point in the temperature range from 70 to 80 K, viz., TN3.165. The values of both TN2 and TN3 depended on the sample quality.165 The magnetic behaviour of LiNiO2 at low temperatures can be described as follows.164 1. The TN3<T<TN1 interval corresponds to nonlinear magnetisation in the external field. The magnetic entropy decreases by 20%.A model of the spin-system inhibition for the antiferromagnetic triangular lattice with the planar short-range order is realised. 2. In the TN2<T<TN3 region, magnetisation sharply increases. The magnetic entropy decreases twofold in the absence of any noticeable anomalies which would have indicated the appearance of a long-range magnetic order. The interactions between the layers are assumed to generate the effects of the three-dimensional short-range order. 3. In the T<TN2 region, the temperature dependence of the magnetisation tends to saturation. A noticeable residual magnetic D G Kellerman TN1 T /K 200Magnetic properties of complex oxides LiMO2 (M=Sc ± Ni) with different types of cationic ordering entropy is preserved up to TN2.No anomalies associated with the long-range order are observed. The existence of a frozen spin state somewhat different from that of common spin glasses is assumed. The difference lies in the fact that the temperature dependence of the specific heat is quadratic in the former case and linear in the latter case. Due to a strong 180-degree aniferromagnetic NiLi7O7NiNi interaction, a ferromagnetic coupling of nickel ions located above and below the Li layer arises, and a small ferromagnetic cluster forms around the NiLi defect.168 The existence of such a cluster was proved experimentally by ESR in an applied field of up to 245 GHz.169 Two types of clusters were observed: those of the first type are formed below 235 K, the other are formed below 120 K.Such a system was proposed to be considered as a superparamag- netic one, i.e., built of separate ferromagnetic domains.155 The formation of the latter is consistent with the fact that spontaneous magnetisation is suppressed in layered oxides.170 Above the temperature of magnetic ordering, the magnetic susceptibility of LiNiO2 is adequately described by the Curie ± - Weiss law the parameters of which found by different authors differ substantially [mef=1.84 mB, Y=79.3 K (see Ref. 164); mef=2.2 mB, Y=2 K (see Ref. 171)]. These parameters also depended on the stoichiometry and the degree of structural order. The temperature dependence of 1/w in the magnetically disordered region was not smooth and consisted of linear seg- ments.98 Taking this into account, it was assumed that yet another phase transition takes place in LiNiO2, viz., a paramagnetic ± par- amagnetic transition.A possible reason for the break observed in 1/w vs. T dependences is the transition of nickel from the low-spin to high-spin state. Table 1 lists the parameters of the Curie ± Weiss equation determined for the low-temperature and high-temper- ature segments of the 1/w vs. T dependences for LiNiO2 samples with different Z values. It is noteworthy that the magnetic moments and the Weiss constants of two paramagnetic phases approach one another with an increase in the degree of structural ordering. It appears that these parameters can serve as indicators of the completeness of the cationic ordering in LiNiO2.Table 1. Magnetic moments and Weiss constants of LiNiO2 with different degrees of structural ordering.98 T=600 ± 900K T=200 ± 400K Z Y /K mef /mB Y /K mef /mB 2.547 2.201 2.061 2.237 1.489 1.751 1.869 1.987 0.67 0.85 0.94 0.98 7287 787 736 720 185 131 31 33 It is interesting that LiNiO2, which was shown to be anti- ferromagnetically ordered in its metastable structurally disor- dered state and ferromagnetically ordered at sufficiently high degrees of structural ordering [Z=0.6 ± 0.8 (see Ref. 98)], retained its paramagnetic properties up to the liquid helium temperature for the case of perfect ordering. The latter observation can be explained by the fact that, as was shown by de Jongh and Miedema,172 two-dimensional Heisenberg systems which appa- rently involve the perfectly ordered LiNiO2 oxide 98 have no long- range magnetic order.The experimentally observed phase tran- sitions in these systems were caused by deviations from the ideal behaviour, which resulted in an interaction arising between the layers. For example, (CnH2n+1NH3)Cu4Cl2 (see Ref. 173) with a small number of nonmagnetic layers separating the magnetic layers has ferromagnetic properties, while the same substance with a greater number of nonmagnetic layers exhibits no ferro- magnetism. For LiNiO2, it is those nickel ions which occur in the lithium layers due to partial disordering that are responsible for the interlayer interaction.785 In view of the fact that the degree of structural order in LiNiO2 is closely related to stoichiometry, the magnetic properties depend even on small changes in x in Li17xNi1+xO2. This was confirmed in a study 160 in which it was shown that samples with x40.01 demonstrate a thermomagnetic hysteresis typical of a spin glass at 8 K, whereas samples with x&0.08 exhibit spontaneous magnet- isation at 5 K, which corresponds to a magnetic moment of *0.3 mB/Ni. Similar results were obtained in another study 174: it was observed that LiNiO2 is a spin glass, in contrast to Li0.93Ni1.07O2. IV. Structure and magnetic properties of LiM1¡xM0xO2 solid solutions The LiMO2 oxides under consideration can form solid solutions. The attention paid to solid solutions, first of all, is associated with the prospects of stabilising the structural types which can be used as cathodic materials in electrochemical devices. Moreover, solid solutions formed by compounds with different types of cationic ordering are interesting from the viewpoint of studying the short- range order effect, the three-dimensional ± two-dimensional struc- tural transitions and the formation of new superstructures.LiNi17xCoxO2 (04x41). Oxides LiNiO2 and LiCoO2 form a continuous series of solid solutions, which, like the original phases, have a trigonal structure with the R3m symmetry. With a decrease in the content of cobalt (the ionic radius of which is smaller than that of nickel), the cell parameters a and c decrease monotonically. However, the concentration dependence of the c/a ratio is not monotonic, and for x=0.3 its slope changes, i.e., the increase in the c/a ratio slows down.175 This result was reproduced in another study;176 however, the authors of the latter study have attributed the singular point to the composition with x=0.5.In their opinion, an increase in c/a from 4.96 (x=0.5) to 4.99 (x=1) does not bring about mixing of lithium and nickel ions in the octahedral positions of the hexagonal face-centred lattice. The value c/a=4.99 corresponds to a perfect hexagonal unit cell. For smaller c/a values and hence for higher nickel concentrations, partial cationic disordering takes place, which results in a devia- tion of the LiNi17xCoxO2 symmetry from the ideal hexagonal towards the parent cubic one.For a perfect close-packed structure of the Fm3m symmetry, c/a=4.90. Based on the results of X-ray diffraction analysis (the Rietveld method), it was found 177 that, at low cobalt contents (x40.2), the excessive bivalent nickel ions are always present in solid solutions studied: Li17zNi1+z7tCotO2 [t=x(1+z)]. An increase in the cobalt content reduces the non- stoichiometry and stabilises the 2D structure. The stabilising effect cobalt has on the 2D structure of the solid solution also manifests itself in electrochemical intercalation. Phase transitions that take place during electrochemical interca- lation of LiNiO2 do not occur in solid solutions.178 This is explained by the fact that cobalt suppresses the ordering of lithium vacancies.At the same time, it is noteworthy that deep chemical extraction of lithium from solid solutions enriched with nickel results in a monoclinic distortion of the lattice.179 To avoid the distortion of the regular layered structure (2D) of LiNi17xCoxO2 solid solutions, the low-temperature synthetic methods were applied, using, e.g., b-Ni17yCoyOOH180 or Ni17yCoy(OH)2181 as the starting materials. Solid solutions with minimum cationic disordering were obtained by the sol ± gel method 182, 183 (this also refers to pure LiNiO2), pyrolysis 184 and co-precipitation from solutions.185 6Li and 7Li NMR studies on LiNi17xCoxO2 solid solutions (04x41) carried out in both static and dynamic modes 186, 187 have revealed structural microheterogeneities unrelated to the cationic ordering.These microheterogeneities did not cause any phase formation and could not be revealed by the diffraction techniques. They represented clusters of several cobalt atoms. A magnetic phase diagram (Fig. 12) was constructed based on the results of detailed studies of magnetic properties of LiNi17xCoxO2 solid solutions (04x41).188 A model was pro-The van Vleck paramagnetism 786 T /K 300 The Curie ± Weiss paramagnetism 260 1 220 2 50 Inhibitions 30 10 Frozen spins 0.6 0 x 0.8 0.4 0.2 Figure 12. Magnetic phase diagram of LiNiO2 ± LiCoO2 system.188 (1) TN1; (2) TN2. posed that described the magnetic properties of both LiNiO2 and LiNi17xCoxO2.This model is based on an assumption of an antiferromagnetic interaction within a layer and a weak ferro- magnetic bonding between the layers. A conclusion that a system of three sublattices is unstable which was drawn from this model is of prime importance. It is with this conclusion that the absence of magnetic reflexes (1/3 1/3 n) is related. The data on LiNi17xCoxO2 solid solutions additionally doped with manganese,189 magnesium 190 and strontium 191 can be found in the literature. LiNi17xTixO2 (04x40.5). According to the results of X-ray analysis,192 the solid solutions with 04x40.3 are iso- structural to the layered LiNiO2. At a higher titanium content, the structure becomes cubic, due to complete cationic disordering.The structural data refined using the Rietveld technique indicated that LiNi0.7Ti0.3O2 also involves a certain number of transition metal ions in the lithium layers; hence, the regular two-dimen- sional structure is distorted. It is of note that these layers involve bivalent nickel and trivalent titanium. This result was obtained by means of the XPS technique. The authors of this study 192 did not mention the mechanism of the charge compensation in LiNi0.7Ti0.3O2 solid solutions which involve Ni2+ ions: whether a certain number of titanium ions appears or the oxygen content changes. It should be noted that the degee of oxidation of Ti (Ti3+) is not typical of titanium compounds synthesised in air at sufficiently high temperatures, i.e., under conditions in which the solid solutions under discussion have been prepared.LiNi17xMnxO2. Solid solutions isostructural to LiNiO2 were studied.193 ± 199 Certain cationic disordering was described 197 within the frameworks of the hexagonal structure and later confirmed.193 To stabilise the trivalent state of nickel ions and thereby stabilise the layered structure, a special procedure of co- precipitation in an oxidative medium was developed.193 As a result, solid solutions with the composition LiNi17xMnxO2+d (04x40.5) were obtained. The oxidation states of the transition metals in these solid solutions were determined by measuring the magnetic susceptibilities. It was shown that both transition metals in the sample bulk are in the trivalent state, whereas traces of Mn4+ were detected on the surface.All solid solutions studied exhibited a ferromagnetic transition at a temperature (TC) which increased with the manganese content. The exception was pure LiNiO2, which is characterised by a strong interlayer ferromag- netic interaction of the Ni(3b)7O7Ni(3a)7O7Ni(3b) type responsible for the high TC. The existence of the ferromagnetic transition both in pure LiNiO2 and LiNi17xMnxO2+d solid solutions points to the incomplete cationic ordering and to the D G Kellerman Table 2. Magnetic characteristics of LiNi17xMnxO2+d (04x40.5).193 x TC /K Y /K mef /mB calculation experiment 162 7118 755 796 1.74 2.06 2.37 3.00 3.32 1.7 2.1 2.2 3.0 3.3 196 39 53 72 88 00.1 0.2 0.4 0.5 7107 presence of transition metal ions in the lithium layers.Table 2 shows the magnetic characteristics of LiNi17xMnxO2+d (04x40.5). LiNi17xFexO2. The LiNi17xFexO2 system demonstrates two types of solid solution. Compositions with 0<x<0.22 are isostructural to LiNiO2. Their degree of ordering gradually decreases, which manifests itself in weaker X-ray reflections superstructural to those of the sodium chloride cubic lattice. Moreover, cubic solid solutions are formed in the concentration interval of 0.42<x<1. In the intermediate region, two phases are present.200 ± 202 By studying this systems, one can follow the changes in the magnetic properties that occur with a decrease in the dimensionality from 3D (LiFeO2) to 2D (LiNiO2).At T>300 K, the magnetic susceptibilities of all LiNi17xFexO2 compositions obey the Curie ± Weiss law. The first group of solid solutions is characterised by a transition to a magnetically ordered phase. The transition temperature and spontaneous magnetisation both increase with the iron content. The presence of magnetic ordering in the interval from 150 to 200 K points to partial cationic disordering, i.e., the presence of a certain amount of transition metal ions in the Li layers. These results were confirmed by crystallographic data.203 For a compo- sition with 20% iron, the distribution of cations over positions is: [Li0.965M0.035]3b[M0.965Li0.035]3aO2, M=0.2Fe+0.8Ni. The fact that X-ray patterns of the second group of solid solutions involve no lines superstructural with respect to the NaCl lattice unambiguously indicates that the ions of alkali and transition metals are not ordered in the planes, in contrast to LiNiO2.Nonetheless, at T<200 K, the magnetic susceptibility of this solid-solution group depends on the field strength. This dependence becomes more pronounced with an increase in the nickel content, which points to a magnetic ordering.201 The magnetisation polytherms statistically processed using the techni- que of thermodynamic coefficients have revealed no spontaneous magnetisation. Hence, the nonlinear variation of magnetisation for structurally disordered samples is a result of the short-range order associated with the trend of nickel ions for aggregation within the framework of this structure.In other words, the exchange couplings, which are the reason for the magnetic order- ing, are limited in solid solutions by certain atomic ensembles. In this case, the condition for percolation do not arise. Such a system is considered 201 as a superparamagnetic one. Based on the concepts mentioned, the cluster size was estimated. The volume of magnetic clusters increased with the nickel content in the cubic solid solution phase, and, at a nickel concentration above 43%, the system first became two-phase and then a hexagonal one. The number of atoms in clusters (n) and the mean size (d) of the latter for the superparamagnetic region of LiFe17xNixO2 are as follows:201 0.4 116 0.2 28 0.3 38 xnd /AÊ 0.1 177 8 9 13 LiMn17xCoxO2 (04x41). This group of solid solutions was studied by two groups of researchers.204, 205 The lattice type was found to be determined by the ratio of transition metalMagnetic properties of complex oxides LiMO2 (M=Sc ¡¾ Ni) with different types of cationic ordering ions;205 for x<0.2, oxides with trigonal structures are formed, for 0.24x40.7, cubic structures are realised and, for higher man- ganese contents, the symmetry of solutions is tetragonal (I41/amd). It should be noted that, to obtain solid solutions in the 0.24x41 concentration interval, the synthesis was carried out in a nitrogen atmosphere. A different way 205 was chosen for synthesising solid solutions of two layered compounds, viz., LiCoO2 and Li2MnO3.It was shown that in order to prepare a solid solution (rather than a mechanical mixture of the extreme members), the synthesis should be carried out with an excess of lithium. All solid solutions thus obtained had monoclinic structures. The transformation of a trigonal (hexagonal) unit cell of LiCoO2 into a monoclinic one is expressed as follows (for parameter a as an example): ah a 1AAA3 p am . The magnetic properties of LiMn17xCoxO2 solid solutions (04x41) formed by diamagnetic LiCoO2 and antiferromag- netic LiMnO2 [TN=50 K, Y=733 K, mef=3.83 mB (see Ref. 206)] change nonmonotonically: the compounds are para- magnetic for x40.5 and the antiferromagnetic transition is observed for higher manganese content.The analysis of the concentration dependence of the effective magnetic moment showed that the oxidation state of manganese is Mn4+ for the manganese content >75%, whereas for its content <75% the presence of trivalent manganese ions may be assumed. Methods for the synthesis of LiMn17xCoxO2 solid solutions and of LiMn17xNixO2 and LiMn17xFexO2 solid solutions with layered structures have been discussed.207 LiCr17xFexO2. This system was studied for both polycrystal- line specimens and single crystals.208 Like LiNi17xFexO2, this system is formed by two phases, one of which has an ordered layered structure and another has a cubic lattice characterised by a random distribution of metal ions over the lattice positions.It was found that the transition from the cubic structure to trigonal one corresponds to the solution with 73% LiFeO2. A somewhat different value (*85%) was found in another study.209 The magnetic susceptibility of structurally ordered compositions was shown to have anomalies in the vicinity of 50 K. These anomalies seem to indicate the three-dimensional order. Above the men- tioned temperature, the magnetic interactions corresponded to two-dimensional interactions. An analysis of the MoE ssbauer spectra has shown 210 that 2D interactions also play the key role at low temperatures. At temperatures above 400 ¡¾ 600 K, the variation of the magnetic susceptibility obeys the Curie ¡¾ Weiss law. Table 3 shows the magnetic characteristics of the system.Divergence between experimental and calculated values of the magnetic moment was attributed to anisotropy of magnetic interactions. In order to explain the parabolic shape of the dependence of the Weiss constant (Y) on the concentration, the exchange- coupling parameters were separated and the intra- (J1) and Table 3. Magnetic characteristics of LiCr17xFexO2.210 x Y /K mef /mB The Curie constant /cm3 K mol71 experiment calculation 3.87 4.07 4.28 4.49 4.69 4.90 5.10 5.30 3.57 4.17 3.77 3.91 3.83 3.83 4.39 4.65 1.58 1.98 1.76 1.90 1.81 1.82 2.40 2.64 00.1 0.2 0.3 0.4 0.5 0.6 0.7 7570 7546 7300 7204 7250 747 794 787 interlayer (J2) exchange-coupling constants were estimated based on the data from the magnetic susceptibility).In addition, the partial values of the Weiss constants related to different types of coupling were obtained. Thus for interlayer interactions, YCr¢§Cr=7185 K, YCr¢§Fe=59 K and YFe¢§Fe=730 K. 2 2 2 LiV17xMgxO2 and LiV17xAlxO2. These systems were syn- thesised and studied 211 with the aim of following how the partial substitution of vanadium for an ion with a spin S 6a 1 in LiVO2 affects the structural and magnetic parameters of the phase transition. It was assumed that the first system involves V2+ ions with S =3/2 and the second system involves Al3+ with S=0; the solid solutions have sufficiently narrow homogeneity regions, viz., 04x40.06 and 04x40.1, respectively.In both systems, the phase transition temperatures were lower than those of pure LiVO2. Moreover, a weak but systematic decrease in the Curie constant was observed in the low-temperature region, which indicated that `triangular' clusters V3+7V2+7V3+ with S=1/2 separated by singlet trimers contributed to the overall susceptibility. Based on these studies, it was concluded that the exchange-coupling constant of V3+ ions and the transition temperature both depend on the V7V distance and on the shift of vanadium ions from their positions. Other solid solutions. Information on LiAlxNi17xO2 solid solutions (x40.25) with the LiNiO2 structure is availiable.212, 213 With an increase in the aluminium content, the c/a ratio somewhat increases.It is noteworthy that introduction of aluminium in LiMO2 compounds is considered as one of the factors that stabilise the a-NaFeO2-type layered structure, the latter being the most suitable from the electrochemical viewpoint. Calcula- tions of miscibility carried out for eight compositions of the LiAl17xMxO2 system have shown 214 that for M=Ti, V, Cr and Mn layered solid solutions are formed at the temperatures of the conventional synthesis. For M=Fe, Co, Ni and Cu, a low temperature region of immiscibility exists. Special attention was given to LiAl17xCoxO2 and LiAl17xCrxO2 systems, which, according to calculations, were completely miscible up to 600 8C. Limitations that arose at higher temperatures were associated with the transition of LiAlO2 into a tetragonal g-mod- ification.215 It was experimentally shown 216 that the synthesis carried out at 650 K results in an amorphous modification, whereas at 700 ¡¾ 900 K a layered hexagonal structure is formed.It is essential that in LiAl17xCoxO2 as well as in LiAl17xNixO2 aluminium can be shifted from octahedral positions.217 Synthesis of LiAl17xCrxO2 solid solutions (04x41) was described.218 Solubility of aluminium in LiMnO2 reached a maximum (5% ¡¾ 7%) at 1220 K.219 As in the previous case, the existence of the maximum was explained by the formation of g-LiAlO2. Introduction of even minor amounts of aluminium to LiMnO2 gives rise to a layered structure of the NaFeO2 type. The use of chromium as the dopant led to a similar result.220 In addition, the possibility of formation of LiCr17xCoxO2, LiCr17xNixO2,221 LiFe17xCoxO2 222 and LiNi17xBxO2 223 solu- tions was demonstrated.TN /K J2/k /K J1/k /K Y2 /K Y1 /K 15 56 40 35 50 55 45 40 76.9 78.5 74.3 72.2 71.1 70.6 70.7 70.6 733 725 717 711 70.70 0.58 71.8 73.7 798 7147 7120 766 732.5 716.3 711.5 713 7472 7399 7180 71387.5 16.3 735.5 780.5788 V. Conclusion Complex oxides with the general formula LiMO2 where M is a trivalent transition 3d metal are considered as inorganic func- tional materials of a new generation.224, 225 First of all, this is associated with their good electrochemical characteristics.This factor has stimulated numerous studies not only in applied but also in fundamental directions. Of prime importance are struc- tural and magnetic studies, because the layered structure charac- teristic of many LiMO2 compounds not only provides adequate diffusion of the alkali metal atoms, which is essential for electro- chemical processes, but also is of special interest from the view- point of studying low-dimensional magnetic interactions. At present, the crystal structure of LiMO2 is sufficiently thoroughly described for all stable modifications, and efforts are mainly concentrated on the synthesis and structural description of metastable phases preferentially with layered structures of the a-NaFeO2 type. The results of investigations of the magnetic properties of LiMO2 show that structural defects may often play the key role in the formation of magnetic structures.First of all, this refers to compounds with variable degrees of structural order. References 1. R B Goldner, F O Arntz, G Berera, T E Haas, G Wei, K K Wong, P C Yu Solid State Ion. 53 ± 56 617 (1992) 2. S Megahed, W Ebner J. Power Sources 54 155 (1995) 3. C Delmas, in Industrial Chemistry Library. Lithium Batteries. New Materials, Developments and Perspectives Vol. 5 (Ed. G Pistoia) (Amsterdam; London; New York; Tokyo: Elsevier, 1994) p. 457 4. T A Hewston, B L Chamberland J. Phys. Chem. Solids 48 97 (1987) 5. R Alcantara, P Lavela, J L Tirado, E Zhecheva, R Stoyanova J. Solid State Electrochem. 3 121 (1999) 6.E J Wu, P D Teresch, G Ceder Philos. Mag. B 77 1039 (1998) 7. R Hoppe Angew. Chem. 71 457 (1959) 8. C J M Rooymans Z. Anorg. Allg. Chem. 313 324 (1961) 9. R Hoppe, B Schepers, H Rohrbon, E Vielhaber Z. Anorg. Allg. Chem. 339 130 (1965) 10. T Birchall, A F Reid J. Solid State. Chem. 6 411 (1973) 11. B L Dickson, K K P Srivastava J. Solid State Chem. 19 117 (1976) 12. Y B Kuo,W Scheld, R Hoppe Z. Kristallogr. 164 121 (1983) 13. P F Bongers Ph D Thesis, The University of Leiden, Leiden, 1957 14. A Lecerf Ann. Chim., Sci. Mater. 7 513 (1962) 15. D W Murphy,M Greenblatt, S M Zahurak, R J Cava, J V Waszczak, G W Hull, R S.Hutton. Rev. Chim. Miner. 19 442 (1982) 16. R J Cava, D W Murphy, S Zahurak, A Santoro, R S Roth J. Solid State Chem.53 64 (1984) 17. A F Wells Structural Inorganic Chemistry (Oxford: Clarendon Press, 1984) 18. T A Hewston Ph D Thesis, The University of Connecticut, Storrs, CT, 1985 19. W RuÈ dorff, H Becker Z. Naturforsh., B Chem. Sci. 9 613 (1954) 20. B Reuter, R Weber, J Jaskowski Z. Electrochem. 66 832 (1962) 21. K Kobayashi, K Kosuge, S Kachi Mater. Res. Bull. 4 95 (1969) 22. T A Hewston, B L Chamberland J. Solid State Chem. 65 100 (1986) 23. L P Cardoso, D E Cox, T A Hewston, B L Chamberland J. Solid State Chem. 72 234 (1988) 24. J B Goodenough, G Dutta, A Manthiram Phys. Rev. B 43 10170 (1992) 25. K Imai, H Sawa,M Koike, M Hasegawa, H Takei J. Solid State Chem. 114 184 (1995) 26. H Takei,M Koike, K Imai, H Sawa, H Kadowaki, Y Iye Mater.Res. Bull. 27 555 (1992) 27. L A Picciotto, M M Thackeray,W I F David, P G Bruce, J B Goodenough Mater. Res. Bull. 19 1497 (1984) 28. L A Picciotto, M M Thackeray Solid State Ion. 18 ± 19 773 (1986) 29. J B Goodenough, A Manthiram, G Dutta Eur. J. Solid State Inorg. Chem. 28 1125 (1991) 30. L A Picciotto, M M Thackeray Mater. Res. Bull. 20 187 (1985) D G Kellerman 31. C Chieh, B L Chamberland, A F Wells Acta Crystallogr., Sect. B 37 1813 (1981) 32. C Simo, S L Holt J. Solid State Chem. 4 76 (1972) 33. L Hernan,M Macias, J Morales, C Perez-Vicente, J L Tirado Mater. Res. Bull. 24 781 (1989) 34. W D Johnston, R R Heikes J. Am. Chem. Soc. 78 3255 (1956) 35. G Ditrich, R Hoppe Z. Anorg. Allg. Chem. 368 262 (1969) 36. R Hoppe, G Brachtel, M Jansen Z.Anorg. Allg. Chem. 417 1 (1975) 37. P Strobel, J Levy, J-C Joubert J. Cryst. Growth 66 257 (1984) 38. J D Dunitz, L E Orgel J. Phys. Chem. Solids 3 20; 318 (1957) 39. D G Kellerman, V S Gorshkov, V G Zubkov, V A Perelyaev, V R Galakhov, E Z Kurmaev, St Uhlenbrock,M Noimann Zh. Neorg. Khim. 42 1012 (1997) a 40. L Croguennec, P Deniard, R Brec, A Lecerf J. Mater. Chem. 5 1919 (1995) 41. L Croguennec, P Deniard, R Brec, P Biensan,M Broussely Solid State Ion. 89 127 (1996) 42. L Croguennec, P Deniard, R Brec, A Lecerf J. Mater. Chem. 7 511 (1997) 43. M M Thackeray,W I F David, P G Bruce, J B Goodenough Mater. Res. Bull. 18 461 (1983) 44. A Mosbah, A Verbnere,M Tournoux Mater. Res. Bull. 18 1375 (1983) 45. A R Armstrong, P G Bruce Nature (London) 381 499 (1996) 46.J N Reimers, J E W Fuller, E Rossen, J R Dahn J. Electrochem. Soc. 140 3396 (1993) 47. R J Gummow,M M Thackeray J. Electrochem. Soc. 141 1178 (1994) 48. C Delmas, F Capitaine, in Extended Abstracts II-B-33 of the 8th International Meeting on Lithium Batteries, Nagoya, 1996 p. 470 49. F Capitaine, P Gravereau, C Delmas Solid State Ion. 89 197 (1996) 50. Y Shao-Horn, S A Hackney, A R Armstrong, P G Bruce, R Gitzendanner, C S Johnson,M M Thackeray J. Electrochem. Soc. 146 2404 (1999) 51. G Ceder, S K Mishra. Electrochem. Solid State Lett. 2 550 (1999) 52. F Le Gras, P Strobel,M Anne, D Bloch J. Mater. Chem. 7 2519 (1997) 53. J C Anderson, S M Anderson J. Phys. Chem. Solids 25 961 (1964) 54.E Posnjak, T Barth Phys. Rev. 38 2234 (1931) 55. A Hoffman Naturwissenschaften 26 431 (1938) 56. R Famery, P Bassoul, F Queyroux J. Solid State Chem. 57 178 (1985) 57. J C Anderson, S K Dey, V Halpern J. Phys. Chem. Solids 26 1555 (1965) 58. R W G Wykoff Crystal Structure Vol. 2 (New York: Intrscience, 1964) p. 312 59. M Brunel, F de Bergevin J. Phys. Chem. Solids 29 163 (1968) 60. M Brunel, F de Bergevin J. Phys. Chem. Solids 30 2011 (1969) 61. J C Anderson,M Schieber J. Phys. Chem. 67 1838 (1963) 62. R Kanno, T Shirane, Y Kawamoto, Y Takeda, M Takano, M Ohashi, Y Yamaguchi J. Electrochem. Soc. 143 2435 (1996) 63. Y Sakurai, H Arai, S Okada, J Yamaki J. Power Sources 68 711 (1997) 64. Y Sakurai, H Arai, J Yamaki Solid State Ion.113 ± 115 29 (1998) 65. R Kanno, T Shirane, Y Inaba, Y Kawamoto J. Power Sources 68 145 (1997) 66. Y Inaba, Y Kawamoto, in Extended Abstracts I-B-04 of the 22nd Symposium on Solid State Ionics, Nagoya, 1996 p. 53 67. E Antolini,M Ferretti J. Solid State Chem. 117 1 (1995) 68. H J Orman, P J Weisman Acta Crystallogr., Sect. C 40 12 (1984). 69. W D Johnston, R R Heikes, D Sestrich J. Phys. Chem. Solids 7 1 (1958). 70. J Akimoto, Y Gotoh, Y Oosawa J. Solid State Chem. 141 298 (1998) 71. C Wolverton, A Zunger Phys. Rev. B 57 2242 (1998) 72. C Wolverton, A Zunger J. Electrochem. Soc. 145 2424 (1998) 73. R J Gummow,M M Thackeray Solid State Ion. 53 ± 56 681 (1992) 74. R J Gummow,M M Thackeray J. Electrochem. Soc. 140 3365 (1993) 75. R J Gummow, D C Liles,M M Thackeray Mater.Res. Bull. 28 235 (1992) 76. R J Gummov, D C Liles,M M Thackeray,W I F David Mater. Res. Bull. 28 1177 (1993) 77. E Rossen, J N Reimers, J R Dahn Solid State Ion. 62 53 (1993) 78. J Morales, R Stoyanova, J L Tirado, E Zhecheva J. Solid State Chem. 113 182 (1994) 79. B Garcia, P Barboux, F Ribot, A Kihn-Hirari, L Mazerolles, N Baffier Solid State Ion. 80 111 (1995)Magnetic properties of complex oxides LiMO2 (M=Sc ± Ni) with different types of cationic ordering 80. W W Huang, R Frech Solid State Ion. 86 ± 88 395 (1996) 81. G G Amatucci, J M Tarascon, D Larcher, L C Klein Solid State Ion. 84 169 (1996) 82. M Carewska, S Scaccia, F Croce, S Arumugam, Y Wang, S Greenbaum Solid State Ion. 93 227 (1997) 83.Y Shao-Horn, S A Hackney, S C Johnson, A J Kahaian, M M Thackeray J. Solid State Chem. 140 116 (1998) 84. J N Reimers, J R Dahn J. Electrochem. Soc. 139 2091 (1992) 85. T Ohzuku, A Ueda J. Electrochem. Soc. 141 2972 (1994) 86. G G Amatucci, J M Tarascon, L C Klein J. Electrochem. Soc. 143 1114 (1996) 87. A van der Ven,M K Aydinol, G Ceder, G Kresse, J Hafner Phys. Rev. B 58 2975 (1998) 88. A van der Ven,M K Aydinol, G Ceder J. Electrochem. Soc. 145 2149 (1998) 89. B Garcia, J Farcy, J P Pereira-Ramos, N Baffer J. Electrochem. Soc. 144 1179 (1997) 90. A Mendiboure, C Delmas, P Hagenmuller Mater. Res. Bull. 19 1383 (1984) 91. C Delmas, J Braconnier, P Hagenmuller Mater. Res. Bull. 17 117 (1982) 92. N Imanishi, M Fujiyoshi, Y Takeda, O Yamamoto,M Tabuchi Solid State Ion.18 121 (1999) 93. L D Dyer, B S Borie, G P Smith J. Am. Chem. Soc. 76 1499 (1953) 94. A Rougier, C Delmas, A V Chadwick Solid State Commun. 94 121 (1995) 95. A Rougier, A V Chadwick, C Delmas Nucl. Instum. Methods Phys. Res. 97 75 (1995) 96. J B Goodenough, D G Wickham, W J Croft J. Phys.Chem. Solids 5 107 (1958). 97. V S Gorshkov, D G Kellerman, V A Perelyaev, G P Shveikin Izv. Akad Nauk SSSR, Neorg. Mater. 27 1274 (1991) b 98. V S Gorshkov, D G Kellerman, V A Perelyaev, G P Shveikin Zh. Neorg. Khim. 41 16 (1996) a 99. Z Lu, X Huang, H Huang, L Chen, J Schoonman Solid State Ion. 120 103 (1999) 100. E Antolini J. Mater. Sci. 27 3335 (1992) 101. E Antolini J. Mater. Sci. Lett. 13 1599 (1994) 102.T Sata Ceram. Int. 24 53 (1998) 103. H Arai, S Okada, H Ohtsuka,M Ichimura, J Yamaki Solid State Ion. 80 261 (1995) 104. V Berbenni, V Massarotti, D Capsoni, R Riccardi, A Marini, E Antolini Solid State Ion. 48 101 (1991) 105. J Morales, C Perez-Vicente, J L Tirado Mater. Res. Bull. 25 623 (1990) 106. W Li, J N Reimers, J R Dahn Phys. Rev. B 46 3236 (1992) 107. A Rougier, P Gravereau, C Delmas J. Electrochem. Soc. 143 1168 (1996) 108. I J Pickering, J T Lewandowski, A J Jacobson, A Goldstone Solid State Ion. 53 ± 56 405 (1992) 109. J N Reimers, W Li, J R Dahn Phys. Rev. B 47 8486 (1993) 110. R Alcantara, P Lavela, J L Tirado, R Stoyanova, E Kuzmanova, E Zhecheva Chem. Mater. 9 2145 (1997) 111. I J Pickering, G N George, J T Lewandowski, A J Jacobson J.Am. Chem. Soc. 115 4137 (1993) 112. T Ohzuku, A Ueda, M Nagayama J. Electrochem. Soc. 140 1862 (1993) 113. I Nakai, K Takahashi, Y Shiraishi, T Nakagome, F Nishikawa J. Solid State Chem. 140 145 (1998) 114. L Croguennec, C Pouillerie, C Delmas Solid State Ion. 135 259 (2000) 115. A Hirano, R Kanno, Y Kawamoto, Y Takeda, K Yamaura, M Takano, K Ohyama, M Ohashi, Y Yamaguchi Solid State Ion. 78 123 (1995) 116. R Stoyanova, E Zhecheva J. Solid State Chem. 108 211 (1994) 117. A Hirano, R Kanno, Y Kawamoto, K Oikawa, T Kamiyama, F Izumi Solid State Ion. 86 ± 88 791 (1996) 118. C Delmas, J P Peres, A Rougier, A Demourgues, F Weill, A V Chadwick,M Broussely, F Perton, P Biensan, P Willmann J. Power Sources 68 120 (1997) 119.J P Peres, F Weill, C Delmas Solid State Ion. 116 19 (1999) 120. H Arai, S Okada, Y Sakurai, J-i Yamaki Solid State Ion. 109 295 (1998) 121. J R Dahn, U von Sacken, C A Michal Solid State Ion. 44 87 (1990) 789 122. V W Bronger, H Bade,W Klemm Z. Anorg. Allg. Chem. 333 188 (1964) 123. H N Migeon, A Courtois, M Zanne, C Cleitzer Rev. Chim. Miner. 13 1 (1976) 124. P F Bongers, in Crystal Structure and Chemical Bonding in Inor- ganic Chemistry (Eds C J M Rooymans, A Rabenau) (New York: Elsevier, 1975) p. 246 125. J B Goodenough Magnetism and the Chemical Bond (New York, London: Wiley-Interscience, 1963) 126. Y Kimishima, M Tsuchiya, K Tomimoto, J Akimitsu J. Magn. Magn. Mater. 104 ± 107 779 (1992) 127. H F Pen, L H Tjeng, E Pellegrin, F M F de Groot, G A Sawatsky,M A van Veendaal, C T Chen Phys.Rev. B 55 15500 (1997) 128. N F Mott Metal-Insulator Transition (London: Taylor and Francis, 1974) 129. Ya V Vasil'ev, S M Ariya Izv. Akad. Nauk SSSR, Neorg. Mater. 1 347 (1965) b 130. A A Bugaev, B P Zakharchenya, F A Chudnovskii Fazovyi Pere- khod Poluprovodnik ± Metall i ego Primenenie (Semiconductor ± Metal Phase Transition and Its Application) (Leningrad: Nauka, 1979) 131. D G Kellerman, V S Gorshkov, V A Perelyaev Izv. Akad. Nauk SSSR, Neorg. Mater. 24 1547 (1988) b 132. M Onoda, T Inabe J. Phys. Soc. Jpn. 62 2216 (1993) 133. B Laar, D J Ijdo J. Solid State Chem. 3 590 (1971) 134. M Krysowska Acta Phys. Pol. A 52 97 (1977) 135. C Delmas,G Le Flem, C Fouassier, P Hagenmuller J.Phys. Chem. Solids 39 55 (1978) 136. N Kojima, K Ito, I Mogi, M Takeda, G Kido,Y Nakagawa, M Sakai, N Kuroda, Y Nishina J. Phys. Soc. Jpn. 62 4137 (1993) 137. J L Soubeyboux, D Fruchart, J C Marmeggi,W J Fitzgerald, C Delmas, G Le Flem Phys. Status Solidi A 67 633 (1981). 138. M Suzuki, I Yamada, H Kadowaki, F Takei J. Phys., Condens. Matter 5 4225 (1993) 139. Y Ajiro, H Kikuchi, S Sugiyama, T Nakashima, S Shamoto, N Nakayama,M Kiyama,N Yamamoto, Y Oka J. Phys. Soc. Jpn. 143. D E Cox, G Shirane, P A Flinn, S L Ruby,W J Takei Phys. Rev., 57 2268 (1988) 140. H Kadowaki,H Takei,K Motoya J. Phys., Condens. Matter 7 6869 (1995) 141. J E Greedan, N P Raju, I J Davidson J. Solid State Chem. 128 209 (1997) 142.V S Gorshkov, V V Karelina, D G Kellerman, in Oksidy. Fiziko- Khimicheskie Svoistva i Tekhnologiya (Tez. Dokl. Vserossiiskoi Nauchno-Prakticheskoi Konf.) [Oxides. Physicochemical Properties and Technology (Abstracts of Reports of the All-Russian Scientific and Practical Conference)] (Ekaterinburg: Ural Branch of the Russian Academy of Sciences, 1999) p. 157 Ser. 2 132 1547 (1963) 144. M Tabuchi, K Ado, H Kobayashi, I Matsubara, H Kageyama, M Wakita, S Tsutsui, S Nasu, Y Takeda, C Masquelier, A Hirano, R Kanno J. Solid State Chem. 141 554 (1998) 145. M Tabuchi, S Tsutsui, C Masquelier, R Kanno, K Ado, I Matsubara, S Nasu, H Kageyama J. Solid State Chem. 140 159 (1998) 146. M Oku J. Solid State Chem. 23 177 (1978) 147. V S Gorshkov, D G Kellerman, V V Karelina Zh.Fiz. Khim. 73 1041 (1999) c 148. R Stoyanova, E Zhecheva, C Friebel Solid State Ion. 73 1 (1994) 149. G F Dionne J. Appl. Phys. 67 4561 (1990) 150. K Hirakawa, H Kadwaki, K Ubukoshi J. Phys. Soc. Jpn. 54 3256 (1985) 151. P Kuiper, G Kruizinga, J Ghijsen, G A Sawatsky, H Verweij Phys. Rev. Lett. 62 221 (1989) 152. V R Galakhov, E Z Kurmaev, S Uhlenbrock,M Neumann, D G Kellerman, V S Gorshkov Solid State Commun. 95 347 (1995) 153. T Mizokawa,H Namatame,A Fujimori,K Akeyama,H Kondoh, H Kuroda, N Kosugi Phys. Rev. Lett. 67 1638 (1991) 154. J Meng, P Jena, J M Vail J. Phys., Condens. Matter 2 10371 (1990) 155. P Ganguly, V Ramaswamy, I S Mulla, R F Shinde, P P Bakare, S Ganapathy, P R Rajamohanan, N V K Prakash Phys. Rev.B 46 11595 (1992) 156. P W Anderson Mater. Res. Bull. 8 153 (1973)790 157. P W Anderson Science 235 1196 (1987) 158. M Itoh, I Yamada, K Ubukoshi, K Hirakawa, H Yasuoka J. Phys. Soc. Jpn. 55 2125 (1986) 159. I Yamada, K Ubukoshi, K Hirakawa J. Phys. Soc. Jpn. 54 3571 (1985) 160. K Yamaura,M Takano, A Hirano, R Kanno J. Solid State Chem. 127 109 (1996) 161. J P Kemp, P A Cox, J W Hodby J. Phys., Condens. Matter 2 6699 (1990) 162. T Shirakami, M Takematsu, A Hirano, R Kanno, K Yamaura, M Takano, T Atake Mater. Sci. Eng. B 54 70 (1998) 163. H Yoshikawa,H Mori,K Hirota,M Ishikawa J. Phys. Soc. Jpn. 59 2631 (1986) 164. K Hirota, Y Nakazawa,M Ishikawa J. Phys., Condens. Matter 3 4721 (1991) 165. K Hirota, Y Nakazawa,M Ishikawa J.Magn. Magn. Mater. 90 ± 91 279 (1990) 166. K Hirakawa, H Kadwaki Physica B 136 335 (1986) 167. M E Fisher Rep. Prog. Phys. 30 615 (1967) 168. D Mertz, Y Ksari, J Celestini, J M Debierre, A Stepanov, C Delmas Phys. Rev. B 61 1240 (2000) 169. A-L Barra, G Chouteau, A Stepanov, C Delmas J. Magn. Magn. Mater. 177 ± 181 783 (1998) 170. P Ganguly Proc. Indian Natl. Sci. Acad. A 52 13 (1986) 171. J N Reimers, J R Dahn, J E Greedan, C V Stager, G Liu, I Davidson, U von Sacken J. Solid State Chem. 102 542 (1993) 172. L J de Jongh, A R Miedema Adv. Phys. 23 1 (1974) 173. J H P Colpa Physica B 57 347 (1972) 174. R Kanno, H Kubo, Y Kawamoto, T Kamiyama, F Izumi, Y Takeda,M Takano J. Solid State Chem. 110 216 (1994) 175. C Delmas, I Saadoune Solid State Ion. 53 370 (1992) 176.Y-M Choi, S-I Pyun, S-I Moon Solid State Ion. 89 43 (1996) 177. A Rougier, I Saadoune, P Gravereau, P Willmann, C Delmas Solid State Ion. 90 83 (1996) 178. I Saadoune, C Delmas J. Solid State Chem. 136 8 (1998) 179. R Alcantara, J Morales, J L Tirado, R Stayanova, E Zhecheva J. Electrochem. Soc. 142 3997 (1995) 180. Y Fujita, K Amine, J Maruta, H Yasuda J. Power Sources 68 126 (1997) 181. J Cho, G Kim, H S Lim J. Electrochem. Soc. 146 3571 (1999) 182. C-C Chang, N Scarr, P N Kumta Solid State Ion. 112 329 (1998) 183. S Tao, Q Wa, Z Zhan, G Meng Solid State Ion. 124 33 (1999) 184. C Julien, S S Michael, S Ziolkievicz Int. J. Inorg. Mater. 1 29 (1999) 185. D Caurant,N Baffer, B Garcia, J P Pereira-Ramos Solid State Ion. 91 45 (1996) 186. M Menetrier, A Pougier, C Delmas Solid State Commun. 90 439 (1994) 187. C Marichal, J Hirschinger, P Granger,M Menetrier, A Rougier, C Delmas Inorg. Chem. 34 1773 (1995) 188. K Hirota, H Yoshikawa,M Ishikawa J. Phys., Condens. Matter 4 6291 (1992) 189. Z L Liu, A S Yu, J Y Lee J. Power Sources 82 416 (1999) 190. Y Cao,M V Yakovleva,W B Ebner Electrochem. Solid State Lett. 1 117 (1998) 191. H-J Kweon, D G Park Electrochem. Solid State Lett. 3 128 (2000) 192. H S Chang, S-G Kang, S-W Song, J-B Yoon, J-H Choy Solid State Ion. 86 ± 88 171 (1996) 193. M E Spahr, P Novak, B Schnayder, O Haas, R Nesper J. Electrochem. Soc. 145 1113 (1998) 194. S Fujiwara, S Yamada,M Kanda, in Extended Abstracts of the 34th Battery Symposium, Nagoya, 1993 p. 135 195. M Yoshio, K Yamata, J Itoh, H Noguchi, M Okada, T Mouri, in Rechargeable Lithium and Lithium Ion Batteries (Eds S Megahed, B M Barnett, L Xie) (Pennington, NJ: Electrochemical Society, 1995) p. 251 196. Y Nitta, K Okamura, K Haraguchi, S Kobayashi, A Ohta J. Power Sources 54 511 (1995) 197. E Rossen, C D W Jones, J R Dahn Solid State Ion. 57 311 (1992) 198. D Caurant, N Baffier, V Bianchi, G Gregoire, S Bach J. Mater. 199. J M Paulsen, C L Thomas, J R Dahn J. Electrochem. Soc. 147 861 200. D G Kellerman, V S Gorshkov, V A Perelyaev, G P Shveikin Zh. Chem. 6 1149 (1996) (2000) Neorg. Khim. 35 1267 (1990) b D G Kellerman 201. D G Kellerman, V S Gorshkov, V V Karelina Zh. Neorg. Khim. 44 1078 (1999) b 202. C Delmas, G Prado, A Rougier, E Suard, L Fournes Solid State Ion. 135 71 (2000) 203. J N Reimers, E Rossen, C D Jones, J R Dahn Solid State Ion. 61 335 (1993) 204. R Stoyanova, E Zhecheva, L Zarkova Solid State Ion. 73 233 (1994) 205. K Numata, C Sakaki, S Yamanaka Solid State Ion. 117 257 (1999) 206. M Jansen, R Hoppe Z. Anorg. Allg. Chem. 397 279 (1972) 207. F Zang,M S Whittingham Electrochem. Solid State Lett. 3 309 (2000) 208. A Tauber,W M Moller, E Banks J. Solid State Chem. 4 138 (1972) 209. E Kordes, J Petzoldt Z. Anorg. Allg. Chem. 335 138 (1965) 210. C Delmas, F Menil, G Flem, C Fouassier, P Hagenmuller J. Phys. Chem. Solids 39 51 (1978) 211. T Naka, M Onoda, H Nagasawa Solid State Commun. 87 679 (1993) 212. T Ohzuku, T Yanagawa, M Kouguchi, A Ueda J. Power Sources 68 131 (1997) 213. G X Wang, S Zhong, D H Bradhurst, S X Dou, H K Liu Solid State Ion. 116 271 (1999) 214. S Buta, D Morgan, A van der Ven,M K Aydinol, G Ceder J. Electrochem. Soc. 146 4335 (1999) 215. H-A Lehman, H Hesselbarth Z. Anorg. Allg. Chem. 313 117 (1961) 216. G Chen,W Hao, Y Shi, Y Wu, S Perkowitz J. Mater. Res. 15 583 (2000) 217. R Stoyanova, E Zhecheva, E Kuzmanova, R Alcantara, P Lavela, J L Tirado Solid State Ion. 128 1 (2000) 218. K R Poeppelmeier, S H Thong J. Less-Common Met. 156 191 (1989) 219. Y I Jang, Y M Chiang Solid State Ion. 130 53 (2000) 220. B Ammundsen, J Desilvestro, T Groutso, D Hassell, J B Metson, E Regan, R Steiner, P J Pickering J. Electrochem. Soc. 147 4078 (2000) 221. C D W Jones, E Rossen, J R Dahn Solid State Ion. 68 65 (1994) 222. H Kobayashi, H Shigemura,M Tabuchi, H Sakaebe, K Ado, H Kageyama, A Hirano, R Kanno, M Wakita, S Morimoto, S Nasu J. Electrochem. Soc. 147 960 (2000) 223. C Julien, G A Nazri, A Rogier Solid State Ion. 135 121 (2000) 224. Yu D Tret'yakov, O A Brylev Ros. Khim. Zh. 44 10 (2000) d 225. C A Vincent Solid State Ion. 134 159 (2000) a�Russ. J. Inorg. Chem. (Engl. Transl.) b�Inorg. Mater. (Engl. Transl.) c�Russ. J. Phys. Chem. (Engl. Transl.) d�Mendeleev Chem. J. (Engl.

ISSN:0036-021X    

出版商:RSC    

年代:2001

数据来源: RSC

摘要:

Russian Chemical Reviews 70 (9) 791 ± 800 (2001) Characteristic features of heterophase polymerisation of styrene with simultaneous formation of surfactants at the interface N I Prokopov, I A Gritskova Contents I. Introduction II. Surface properties of the potassium salts of carboxylic acids with alkyl substituents of different lengths and the composition of styrene emulsions prepared during their formation at the interface III. Bulk properties of the salts of carboxylic acids sparingly soluble or virtually insoluble in water and the composition of styrene emulsions formed in parallel with their synthesis at the interface IV. Synthesis of polystyrene suspensions with a narrow particle size distribution under the conditions of formation of surfactants at the interface V.Conclusion Abstract. styrene of polymerisation heterophase the on Data Data on the heterophase polymerisation of styrene under conditions of surfactant formation at the monomer ± under conditions of surfactant formation at the monomer ± water approach principle, in new, A generalised. are interface water interface are generalised. A new, in principle, approach is is proposed the essence of which is to obtain a monomer emulsion proposed the essence of which is to obtain a monomer emulsion simultaneously the at emulsifier an of synthesis the with simultaneously with the synthesis of an emulsifier at the mono- mono- mer polymerisation the of initiation with and interface water ± mer ± water interface and with initiation of the polymerisation in in the interfacial layer.The preparation of surfactants at the inter- the interfacial layer. The preparation of surfactants at the inter- face dispersion of degree the efficiently control to one allows face allows one to control efficiently the degree of dispersion and and the of nature the varying By formed. emulsions the of stability the stability of the emulsions formed. By varying the nature of the the acid synthesis surfactant the in used counter-ion metal the and acid and the metal counter-ion used in the surfactant synthesis at at the interface, it is possible to change the interfacial tension, to the interface, it is possible to change the interfacial tension, to influence the of disintegration microemulsification, the influence the microemulsification, disintegration of the monomer, monomer, and adsorption interfacial of structure of formation the and the formation of structure of interfacial adsorption layers.layers. The particles polymer-monomeric of formation of mechanism The mechanism of formation of polymer-monomeric particles as as well depend distribution size and diameter their as well as their diameter and size distribution depend substantially substantially on water. in surfactants resulting the of solubility the on the solubility of the resulting surfactants in water. The The bibliography references 47 includes bibliography includes 47 references. I. Introduction For several decades, researchers' assumptions concerning the compositions of monomer emulsions and the patterns of emulsion polymerisation have been based on the Harkins ± Yurzhenko views; these also underlie the only existing quantitative theory (the Smith ± Ewart theory).1 ±4 However, numerous deviations of experimental data from theoretical predictions have stipulated the search for new approaches to the description of this process.These deviations have been detected even in a study of styrene polymer- isation initiated by potassium persulfate carried out in the presence of an ionic emulsifier, which is exactly the reaction that served initially as the basis for the Smith ± Ewart theory. The problem of description of emulsion polymerisation became espe- cially topical after the appearance of results which could not be predicted by existing theories.Thus the occurrence of a quasi- N I Prokopov, I A GritskovaMV Lomonosov Moscow State Academy of Fine Chemical Technology, prosp. Vernadskogo 86, 117571 Moscow, Russian Federation. Fax (7-095) 434 87 11. Tel. (7-095) 434 86 44. E-mail: fun-latex@mtu-net.ru (N I Prokopov). Tel. (7-095) 247 04 43 (I A Gritskova). Received 26 March 2001 Uspekhi Khimii 70 (9) 890 ± 900 (2001); translated by Z P Bobkova #2001 Russian Academy of Sciences and Turpion Ltd DOI 10.1070/RC2001v070n09ABEH000669 791 792 794 794 799 spontaneous microemulsification of the monomer at phase inter- face has been observed;5, 6 polystyrene microbeads with a narrow size distribution have been prepared in the absence of surfac- tants;7, 8 the governing role of the composition of monomer emulsions in the mechanism of formation of polymer-monomeric particles (PMP) and its influence on the particle size distribution have been established.9 ±14 Detailed study of emulsions of monomers with various natures � styrene, methyl methacrylate, chloroprene, isoprene, mixtures of polar and non-polar monomers (for example, styrene and methacrylic acid) �has shown 15 ± 19 that the emulsion com- position and the degree of dispersion can be controllable.For example, the addition of water-insoluble surfactants to the mono- mer phase of the emulsion or to the other phase gives rise to emulsions consisting mainly of monomer microdroplets. Mean- while, emulsions prepared in the presence of sodium alkylsulfo- nate insoluble in the monomer consist of emulsifier micelles and monomer macrodroplets.Thus, the procedure used to prepare the initial emulsion of the monomer (or monomers) can pre-determine the particle diameter in the polymer suspension and the size distribution of particles. It has been proposed to perform heterophase polymerisation of monomers with simultaneous synthesis of a surfactant at the phase interface in order to control the emulsion composition. This polymerisation method differs essentially from the commonly accepted one,20 ± 26 as becomes obvious already when the initial monomer emulsion is prepared. Traditionally, a monomer and an aqueous solution of an emulsifier are stirred at a rotational velocity of 600 ± 800 rpm until a uniform emulsion is produced.The size of monomer droplets in this emulsion depends on the rotational velocity, interfacial tension (s1,2) and temperature. The size of monomer droplets and the nature of the surfactant influence the emulsion stability. Various stability factors act in the interfacial adsorption layer (IAL). The surfactant concentration in the IAL is determined by the surfactant capability of being adsorbed at the surface of droplets. Since the surfactant concentration is usually higher than the critical micelle concentration (CMC), it is generally accepted that the initial emulsion consists of surfactant micelles, which solubilise the monomer, and of monomer macrodroplets. The method used to prepare the initial emulsion for polymer- isation under conditions of surfactant synthesis at the interface differs in principle from the above-described one.In this case, a long-chain carboxylic acid is dissolved preliminarily in the mono-792 mer phase, while an alkali is dissolved in the aqueous phase. This procedure gives a fine emulsion of the monomer, owing to the significant decrease in the interfacial tension caused by the fact that the emulsifier formed in the neutralisation reaction is accumulated in the interfacial layer. In this case, the interfacial tension decreases much more significantly than in the case when the same emulsifier is adsorbed on the surface of droplets from the aqueous phase in the conventional preparation of the emulsion (<1 and*10 mNm71, respectively).The degree of dispersion of the resulting emulsion must also depend on the nature of the emulsifier synthesised at the interface. For example, if the emulsifier is soluble both in the monomer and in water, its mass transfer through the interface takes place in conformity with the partition coefficient, the stability of the interfacial boundary is degraded, and the monomer undergoes disintegration and microemulsification. In this case, apart from monomer droplets, the resulting emulsion contains monomer microdroplets and emulsifier micelles ie emulsifier content in the aqueous phase is greater than that required for stabilisation of monomer droplets and microdroplets. The composition of the monomer emulsion determines the mechanism of formation and the diameter distribution of the PMP.When the PMP are formed from emulsifier micelles and monomer microdroplets, the polymer suspension contains par- ticles of different sizes (with diameters from 0.06 to 0.4 mm) and is characterised by a broad size distribution. Polymerisation of an emulsion consisting of monomer micro- droplets affords a polymer suspension with an average particle diameter of 0.08 to 0.7 mm, depending on the structure and the surface-active properties of the emulsifier synthesised at the inter- face.The particle size distribution (PSD) depends on the stability of the PMP during the synthesis. If some factors providing stabilisa- tion of the interfacial adsorption layer arise at initial stages of monomer conversion, the PSD is narrow.This review describes new approaches to the control of the composition of monomer emulsions. These methods allow the preparation of polymer suspensions with particles of different diameters and with a narrow PSD. II. Surface properties of the potassium salts of carboxylic acids with alkyl substituents of different lengths and the composition of styrene emulsions prepared during their formation at the interface In several studies,20 ± 26 salts of oleic acid, salts of saturated fatty acids with the C11±C25 hydrocarbon chains, and the potassium soap from disproportionated rosin have been used as the objects of investigation. The interfacial tension s1,2 at the styrene ± water interface decreases from 42.2 to 30.8 mNm71 over the homologous series from lauric to cerotic acid.For none of the acids does the interfacial tension depend on the time the surface has existed, i.e., adsorption of the acid is not the limiting stage of the surfactant formation. However, the interfacial tension sharply decreases with an increase in the alkali concentration to reach *1 mN m71 for an alkali content in water of 1072 mass %. Figure 1 presents the data on the neutralisation kinetics of oleic acid with potassium hydroxide at the styrene ± water inter- face at a temperature of 20 8C. The data were gained by gravim- etry (curve 1) and refractometry (curve 2). It can be seen that neutralisation is almost completed over a period of several minutes. After 5 ± 10 min, the concentration of potassium oleate in water becomes*4 mass %.The substantial difference between the interfacial tension at the interface between styrene solutions of lauric acid of various concentrations and aqueous solutions of alkali in the synthesis of potassium laurate and the interfacial tension for the adsorption of the same emulsifier from the aqueous phase is illustrated in Fig. 2. N I Prokopov, I A Gritskova 4 1.2 1 1.0 3 0.8 0.6 2 0.4 1 0.2 2 0 t /min 20 0 10 Figure 1. Content of potassium oleate in water (1) and oleic acid in styrene (2) vs. duration of neutralisation. s1,2 /mNm71 40 Dry residue (mass %) 2 0.10 0.06 Emulsifier concentration /mol litre71 Figure 2.Interfacial tension at the interfaces `aqueous solution of potassium laurate ± styrene' (1) and `aqueous solution of potassium hydroxide ± styrene solution of the lauric acid' (2) vs. emulsifier concen- tration. When potassium laurate is synthesised at the interface, the interfacial tension is less than 1 mN m71, whereas this value for the interface between an aqueous solution of potassium laurate, on the one hand, and styrene, on the other hand, is s1,2=8 mN m71. The isotherms of interfacial tension were used to calculate the limiting adsorption value (Gmax), the area occupied by one surfactant molecule in a saturated adsorption layer (Smin) and the surface activity of the surfactant (G) (Table 1). It can be seen that for the formation of emulsifiers at the interface, the G value markedly increases, Gmax also increases, and Smin decreases. The procedure of formation of emulsions should have a pronounced influence on the formation and properties of the interfacial adsorption layers on the surface of monomer drop- lets.27 ± 30 Indeed, when the monomer is emulsified by an aqueous solution of an emulsifier, the interfacial layer is formed by the molecules of a highly ionised carboxylic acid salt adsorbed from the aqueous phase, and when the emulsifier is synthesised at the interface, this layer is formed initially by the adsorbed molecules of the carboxylic acid and after neutralisation, it is formed by salt molecules.The instability of the phase interface caused by the chemical synthesis of the emulsifier taking place in the boundary layer and the emulsifier mass transfer to the aqueous phase, which are accompanied by the transformation of the chemical reaction energy directly into the surface energy,31 ± 33 result in effective fragmentation of monomer droplets and monomer microemulsi- fication. This influences, first of all, the average diameter of the droplets of the monomer emulsion.If the emulsion is prepared Content of C17H33COOH in styrene (%) 30 20 1 100 0.02Characteristic features of heterophase polymerisation of styrene with simultaneous formation of surfactants at the interface Table 1. Surface properties of emulsifiers at the interface. Emulsifier Smin G /nm2 /mN m2 mol71 The way 106Gmax of intro- /mol m72 ducing the emulsifier a Potassium laurate Potassium myristate AABB Potassium palmitate AB 4.0 6.4 3.6 8.4 4.8 10.3 Potassium oleate Potassium behenoate A 4.0 6.4 2.0 0.41 0.26 0.45 0.22 0.35 0.16 0.41 0.26 0.81 0.12 0.58 0.18 18 84 30 67 44 82 27 1354 21 12 33 ABB Potassium cerotate AB 13.4 2.9 9.6 aA�the emulsifier is introduced in the aqueous phase, B�the emulsifier is formed at the interface.under conditions of synthesis of potassium laurate and the potassium soap from disproportionated rosin at the interface, the average diameter of monomer droplets is *2 times smaller than that for the case of monomer emulsification by an aqueous solution of the surfactant (10 and 20 mm, respectively).Study of monomer microemulsification under static condi- tions has shown 20 that the intensity of this process depends on the method for the preparation of emulsions. In a styrene emulsion prepared with the formation of the emulsifier at the interface, the volume of the microemulsion layer is several times as great as that at the interface between styrene and an aqueous solution of the emulsifier. This can be clearly seen when considering the prepara- tion of styrene emulsions by these methods in the presence of rosin soap (Fig. 3). A large volume of the microemulsion increases the emulsion stability. For example, styrene emulsions prepared with the formation of ionogenic surfactants at the interface are 5 ± 10 times as stable as the emulsions formed upon emulsification of the monomer by an aqueous solution of the emulsifier.Procedures making use of small-angle X-ray diffraction and electron microscopy of liquid ± liquid systems have been devel- oped to study the composition of styrene emulsions. Using these procedures, the degrees of dispersion of liquid ± liquid systems can be studied without hardening the monomer phase.34, 35 15 15 2 10 1055 1 03.0 3.0 t /days 2.0 2.0 1.0 1.0 Figure 3. Variation of the height of a microemulsion layer under static conditions at the interfaces `styrene ± aqueous solutions of the rosin soap' (1) and `styrene solution of rosin ± an equimolar aqueous solution of KOH' (2).Emulsifier concentration is 1.161072 mol litre71, 25 8C. The height of the microemulsion layer /mm 793 X-Ray diffraction and electron-microscopic investiga- tions 22, 23 of styrene emulsions prepared with the synthesis of potassium carboxylates at the interface showed that they contain particles of various sizes, namely, monomer macrodroplets with a size of not more than 10 mm, monomer microdroplets 0.03 ± 0.04 mm in diameter, and emulsifier micelles with a size of 0.003 ± 0.005 mm. The surfactant concentration calculated for the aqueous phase equa mass%± 5 mass %, which is 50 ± 100 times as high as the CMC. If the concentration of the emulsifier formed at the interface is decreased to the level sufficient only for stabilisation of monomer microdroplets, the surfactant content in the aqueous phase would be lower than the CMC and the formation of emulsifier micelles would become impossible.In this case, the emulsion would mainly consist of monomer microdroplets. This situation has been realised in the preparation of styrene emulsions under conditions of synthesis of potassium carboxylates at the interface. It can be seen from the data presented in Table 2 that the emulsifier concentration corresponding to the minimum s1,2 values in the aqueous phase is several times lower than that corresponding to complete neutralisation of the acid. This means that some of the surfactant is consumed for the formation of the adsorption layer on the surface of monomer droplets and microdroplets.Table 2. Variation of the emulsifier concentration during neutralisation of the acid under static conditions (25 8C, overall time of reaction 24 h). Emulsifier Emulsifier concentration in the aqueous phase at s1,2 min /mass% Total emulsifier concentration /mass% Potassium laurate (CMC 0.22%) Potassium oleate (CMC 0.04%) 0.08 0.10 0.21 1.00 0.02 0.04 0.12 1.50 0.15 0.25 0.50 4.00 0.15 0.25 0.50 4.00 The critical micelle concentration of the emulsifier in the aqueous phase is attained only in the case where the acid concentration in the monomer phase exceeds 0.5 mass% with respect to styrene.36 As was to be expected, the lithium salts of acids have lower concentrations in the aqueous phase than the potassium salts.Since all the salts synthesised at the interface are soluble in water, effective mass transfer of these salts to the aqueous phase takes place as well as quasi-spontaneous emulsification of the monomer and the formation of microemulsions as a quite visible white belt. Under dynamic conditions, the CMC of the emulsifier in the aqueous phase is attained when the acid concentration in the monomer phase is >1 mass %. This is twice that in the synthesis of surfactants under static conditions; this is explained by the increase in the interfacial area owing to stirring and to the consumption of the emulsifier for the stabilisation of monomer droplets and microdroplets. Thus, if the concentration of the emulsifiers (potassium carboxylates) formed at the interface is 0.5 mass% relative to the monomer, the initial emulsion of styrene contains only monomer microdroplets because the rest of the emulsifier is spent for their stabilisation. This means that, upon initiation of polymerisation in these emulsions, the polymer-monomeric par- ticles would be formed from monomer droplets and the polymer suspensions would have a narrow particle size distribution.794 III.Bulk properties of the salts of carboxylic acids sparingly soluble or virtually insoluble in water and the composition of styrene emulsions formed in parallel with their synthesis at the interface The solubility of lithium salts of carboxylic acids in water exceeds only slightly the CMC.Thus the solubilities of lithium laurate, stearate and oleate at 50 8C are 0.28 mass%, 0.10 mass% and 0.13 mass%, while the CMC values are 0.25 mass%± 0.30 mass%, 0.07 mass% and 0.12 mass%, respectively.37 The solubilities of the barium, calcium and zinc salts of the same acids in water do not exceed 0.004 ± 0.008 g per 100 g of water. The Li, Ba, Ca and Zn salts are also virtually insoluble in the organic phase, in particular, in styrene.37 Analysis of the interfacial tension isotherms shows that the surface tension for the formation of lithium, barium, calcium and zinc myristates at the styrene ± water interface is of the order of 21 ± 31 mNm71, whilst potassium myristates decrease the inter- facial tension almost to zero (Fig.4). s1,2 /mN m71 50 40 30 43 2 5 20 10 1 0 0.02 0.06 0.10 Emulsifier concentration /mol litre71 Figure 4. Interfacial tension at the interfaces `styrene solution of myristic acid ± aqueous solution of potassium (1), barium (2), calcium (3), zinc (4), or lithium (5) hydroxide' vs. emulsifier concentration. The time of droplet formation is 20 s, 20 8C. The surfactants insoluble in both phases exhibit a much lower (by a factor of 10 ± 20) surface activities than water-soluble potassium myristate (Table 3). Lithium myristate occupies an intermediate position; its surface activity at the styrene ± water interface is 3 ± 6 times higher than those of the barium, calcium, and zinc salts but is *3 times lower than the activity of the potassium salt. Emulsions prepared under the conditions of formation of potassium myristate at the interface are 2 ± 3 times as stable as emulsions prepared in parallel with the synthesis of lithium, barium, calcium and zinc salts (all other conditions being the same), which is due to the formation of microemulsions at the surface of monomer droplets.Table 3. Surface properties of emulsifiers at the styrene ± water interface. Emulsifier G /mN m2 mol71 s1,2 min /mN m71 <0.5 20.6 27.1 28.7 30.2 67.0 22.5 7.3 5.4 3.5 Potassium myristate Lithium myristate Barium myristate Calcium myristate Zinc myristate N I Prokopov, I A Gritskova Table 4. Size of particles in an emulsion of styrene prepared with simultaneous synthesis of surfactants of various natures at the interface (styrene : water volume ratio of the phases 1 : 2, surfactant concentration in the aqueous phase 0.084 mol litre71). Surfactant Particle size b /nm Method used to determine the particle size a Potassium stearate 3.5 34 36 Lithium stearate Lithium laurate 120 115 210 200 500 490 SAXD SAXD EM LCS EM LCS EM LCS LCS Barium stearate Zinc stearate aSAXD is small-angle X-ray diffraction, EMis electron microscopy, LCS is laser correlation spectroscopy; b The first line in the column implies the size of surfactant micelles, while the other numbers are the sizes of monomer microdroplets.It has been shown by electron microscopy and laser correla- tion spectroscopy that, when poorly water-soluble or virtually water-insoluble surfactants, namely, lithium and barium stea- rates, are employed, the emulsions formed during their synthesis at the interface consist of macro- and microdroplets of the monomer, the size of the latter ranging from 120 to 500 nm, and no emulsifier micelles are present.The data on the degree of dispersion of the styrene emulsions prepared simultaneously with the synthesis of various surfactants at the interface are summarised in Table 4. These data indicate that the composition of a styrene emulsion depends substantially on the method of its preparation and that it is much more intricate than it has been considered in the classical works by Smith ± Ewart, Yurzenko, and their pupils.3, 4 The investigations have shown that the reaction of surfactant formation at an interface affects substantially the surface phe- nomena in the interfacial adsorption layers and can serve as a powerful factor for controlling the dispersivity of emulsions, and, hence, the mechanism of formation of particles.IV. Synthesis of polystyrene suspensions with a narrow particle size distribution under the conditions of formation of surfactants at the interface Research into the synthesis of polymer dispersions with a specified diameter of particles and a narrow particle size distribution has shown that these properties are mainly dictated by the granulo- metric composition of the initial emulsion of the monomer, the time of formation of the PMP, and the creation of conditions which ensure stability of emulsions in the interfacial layer.38, 39 These factors depend on many parameters, most of all, on the conditions chosen for the preparation of emulsions and for polymerisation.The granulometric composition of the initial emulsion of the monomer and, hence, the mechanism of PMP formation can be readily controlled by selecting appropriate surfactants and an appropriate way of introducing them into the system (either as separate solutions in water or in the monomer or as solutions both in water and in the monomer the concentration of which is determined by the surfactant solubility in these phases) and by the synthesis of surfactants at the interface.14, 18, 26 Any of these methods gives monomer emulsions characterised by different degrees of dispersity due to different concentrations of the surfactant at the interface.Correspondingly, the interfacial ten- sions also differ significantly.Characteristic features of heterophase polymerisation of styrene with simultaneous formation of surfactants at the interface The time of formation of the PMP is a very important factor which largely determines the diameter and the size distribution of particles; this factor depends considerably on the efficiency of initiation of polymerisation and on the process temperature. Finally, the stability of the PMP during polymerisation is ensured by the existence of factors of PMP stability acting in the interfacial adsorption layers (electrostatic or structure-mechan- ical factor or both).These factors also influence the diameter of particles and their size distribution. The duration of the stage of PMP formation can vary from several minutes to several hours. It depends appreciably on the granulometric composition of the emulsion and the efficiency of initiation. At this stage, particles can coalesce due to instability giving rise to a coagulum. Such a phenomenon has been observed, for example, in suspension polymerisation of styrene in the presence of poly(vinyl alcohol), starch, or gelatin as stabilisers. It is at the stage of PMP formation that the diameter of particles of a polymer suspension can be controlled by affecting their coales- cence in one or another way.This process is extremely complex. In order to eliminate the coagulum formation, surfactants are added to the system. They, in turn, can induce the formation of newPMP and extend markedly the size distribution of particles. When styrene polymerisation is carried out in the presence of organosilicon surfactants, insoluble in water and incompatible with the resulting polymer, this stage occurs in a different way.40, 41 In this case, a strong interfacial layer consisting of the surfactant displaced by the polymer and of the polymer itself is formed on the PMP surface at relatively low degrees of monomer conversion. The diameter of particles in a polymer suspension can be controlled by varying the nature of the surfactant and the temperature.Performing styrene polymerisation with simultaneous syn- thesis of a surfactant at the interface ensures, on the one hand, the possibility of realisation of either emulsion or suspension poly- merisation mechanism giving rise to polymer suspensions in which the diameter of particles can differ by an order of magnitude and, on the other hand, it ensures stability of these suspensions because the surfactant concentration in the interface is higher than that in the case where the surfactant is adsorbed onto the PMP surface from the aqueous phase. Thus polymerisation can be carried out at lower concentrations of the surfactant and gives polymer suspensions more stable during storage or modification (e.g., by proteins). The morphology of particles prepared under conditions of surfactant synthesis at the interface should differ essentially from that resulting from conventional synthesis.When emulsions are prepared under ordinary conditions, the interfacial adsorption layer is formed upon adsorption of the emulsifier from the aqueous phase first on monomer droplets and then on the PMP. When polymerisation is carried out with synthesis of an emulsifier at the interface, the initial monomer emulsion as such does not actually exist. In this case, simulta- neously with vigorous disintegration (microemulsification) of monomer droplets, initiation of polymerisation takes place; the resulting microdroplets capture radicals being thus converted into the PMP.In parallel, the interfacial adsorption layers are formed from the acid molecules dissolved in the monomer, the salt molecules resulting from neutralisation of the acid, and polymer molecules (polymerisation occurs in the area of the PMP adsorp- tion layers). The preliminary dissolution of the acid in the monomer ensures the optimal orientation of the acid and salt molecules in the interfacial layer, which gives rise to the electro- static factor of stabilisation. Due to the high polymer concen- tration in the interfacial layer, the viscosity of the IAL substantially increases and the structure-mechanical factor of stabilisation also starts to be effective. The synthesis of water-insoluble surfactants at the interface should yield particles with the `core ± shell' type structure.The formation of these particles is responsible for the specific features of polymerisation kinetics. Even at early stages of the process, the PMP become a sort of microreactors in which polymerisation progresses involving the monomer and the radicals present therein. In this case, the rate of polymerisation and the PMP diameter should not depend on the concentration of the initiator, and an increase in the surfactant concentration would entail an increase in the polymer molecular mass due to the decrease (caused by diffusion restrictions) in the concentration of radicals which enter the PMP at an initial stage of the process. Thus, by conducting polymerisation of monomers with simul- taneous synthesis of surfactants at the interface, one can obtain polymer suspensions which differ not only in the particle diameter and size distribution but also in the morphology of particles.For example, in the case of synthesis of water-soluble sodium or potassium carboxylates, polymeric suspensions with a broad size distribution of particles should be produced because the initial emulsion contains monomer microdroplets and emulsifier micelles from which the PMP are formed.5 100 3 4 2 1 80 60 40 200 120 80 40 t /min Figure 5. Yield of polymer vs. time of styrene polymerisation in the presence of emulsifiers. Emulsifier: (1) potassium laurate; (2) potassium myristate; (3) potassium palmitate; (4) potassium stearate; (5) potassium cerotate; (6) potassium behenoate. Polymerisation temperature 50 8C, monomer : water = 1 : 2 (by volume), concentration of the initiator (azobisisobutyronitrile) in styrene 561072 mol litre71, emulsifier concentration in the aqueous phase 8.461073 mol litre71.Degree of conversion (%) 795 Preparation of polymer suspensions with a narrow particle size distribution by polymerisation under conditions of the syn- thesis of water-soluble surfactants at the interface proved possible only when the concentrations of the monomer (volume ratio of the phases) and the surfactant were such that the surfactant being formed was consumed only for stabilisation of the monomer microdroplets and of the PMP formed from them, i.e., when virtually no surfactant micelles were present in the aqueous phase.The foregoing is illustrated by experimental data. Figure 5 shows the kinetic curves for the emulsion polymerisation of styrene under conditions of formation of potassium salts of carboxylic acids at the interface. In these experiments, aliphatic acids with alkyl substituents from C11 (lauric acid) to C25 (behenic acid) were used; the conditions of polymerisation were identical. It can be seen that polymerisation occurs without any induction period, which could have corresponded to the stage of formation of polymer-monomeric particles, and the high rate of the process is retained almost to the complete monomer conversion. The reaction system is stable, and the final polymer suspension contains no coagulum.The stability of particles in the suspension is largely ensured by the action of the electrostatic factor of stabilisation, as indicated by the high x-potential of particles (Table 5). It follows from the data given in Table 5 that the polymer- isation rate wp somewhat increases as the alkyl radical of the 6796 Table 5. Properties of polymers and polymer suspensions prepared in the polymerisation of styrene with simultaneous formation of the potassium salts of fatty acids at the interface. Surfactant Potassium laurate 0.68 0.73 Potassium myristate Potassium palmitate Potassium stearate 0.88 Potassium cerotate 0.92 0.94 Potassium behenoate Note. Here and in other tables: MM is the viscosity-average molecular mass, d is the average particle diameter, n is the polydispersity coefficient.surfactant becomes longer; this is due to a slight change in the number of PMP formed. The molecular mass (MM), equal to (2.5 ± 4.2)6106, and the molecular-mass distribution (MMD) of the polymers correspond to known published data on the emulsion polymerisation of styrene initiated by an oil-soluble initiator, namely, azobis(isobu- tyronitrile) 39 (Fig. 6). For low degrees of monomer conversion (1% and 5%), the MMD curves exhibit two peaks, one at lower and one at higher molecular masses (MM&100 000 and MM>1 000 000, respectively). When the degree of monomer conversion increases to 90%, the lower-molecular-mass peak substantially decreases or even disappears.The presence of the lower-molecular-mass peak is due to the fact that polymerisation occurs both in the bulk PMP and in the emulsifier micelles.13, 39 As noted above, an increase in the length of the alkyl radical is accompanied by a decrease in the solubility of its salts in the aqueous phase. Whereas potassium laurate, myristate, palmitate, and stearate are soluble in water fairly readily (30 mass%± 105qw 16 146420 105qw 18 16 146420 105qw 420 Figure 6. Molecular-mass distribution (qw) of polystyrene prepared by polymerisation of the monomer under conditions of formation of potas- sium laurate at the interface for degrees of conversion of (%): (a) 1; (b) 5, (c) 90. 1076MM wp (% min71) 2.3 3.3 3.6 0.83 4.0 4.2 4.2abc 1 3 51076MM d /nm n x-Poten- tial /mV 1.19 1.22 102 93 752 747 1.20 90 749 1.20 1.07 1.07 80 76 70 750 746 745 a Percentage 654321 63 158 25 b Percentage 48 36 24 12 158 25 63 Particle diameter /nm Figure 7.Particle size distribution for a polystyrene suspension prepared with simultaneous formation of potassium laurate (a) and potassium behenoate (b) at the interface. 70 mass %), potassium cerotate and behenoate are much less soluble (3 mass%± 5 mass %).24 This should influence the mech- anism of PMP formation. In the former case, the particles are formed by a mixed mechanism � both from micelles of the surfactant which solubilises the monomer and from the monomer microdroplets.In the latter case, the particles are mainly formed from monomer microdroplets. Indeed, polymer suspensions formed under conditions of synthesis of potassium laurate at the interface have very broad particle size distributions, whereas suspensions prepared with the synthesis of potassium behenoate are virtually monodisperse (Fig. 7). If the emulsifier formed at the interface is poorly soluble in water (for example, lithium salts of carboxylic acids), the initial emulsion of styrene, as shown above, contains only microdroplets of the monomer, the emulsifier being spent entirely for their stabilisation. Polymerisation of styrene under conditions of the synthesis of lithium carboxylates has some distinctive features (Fig.8), which are due, first of all, to the fact that these salts are less soluble than the potassium salts. Thus the solubility of lithium myristate in water does not exceed 0.28 mass%and the surface activity of this compound is lower. Therefore, when styrene polymerisation is conjugated with the synthesis of lithium salts of fatty acids, the mean diameter of particles of polymer suspensions is 1.5 ± 3 times greater than that attained with potassium salts, all other factors being the same (Table 6). The particle size distribution in the `lithium' suspensions is fairly narrow because all the emulsifier is consumed for the formation of the interfacial adsorption layer of droplets, microdroplets and the polymer-monomeric particles formed from them.The rate of polymerisation and the molecular mass of the polymer prepared in the presence of lithium myristate are lower than those with lithium laurate because the number of PMP produced in this case is smaller. N I Prokopov, I A Gritskova 398 398Characteristic features of heterophase polymerisation of styrene with simultaneous formation of surfactants at the interface 1 80 60 40 2 20 3 0 t /min 100 60 20 Figure 8. Yield of polymer vs. time of styrene polymerisation in the presence of emulsifiers. Emulsifier: (1) lithium oleate; (2) lithium laurate; (3) lithium stearate. The volume phase ratio monomer : water = 1 : 10, initiator concentration in water 0.1 mass %, emulsifier concentration in water 0.5 mass %.Table 6. Characteristics of the particles of polymer suspensions prepared while performing styrene polymerisation simultaneusly with the synthesis of ionogenic surfactants at the interface (temperature 70 8C, styrene : water phase ratio 1 : 10, initiator concentration 0.1 mass %). Surfactant x-Poten- 1076MM tial /mV c d/nm n 0 (%) (mass %) a (see b) 1.2 1.3 3.8 4.2 258 220 193 65 Lithium laurate 0.25 Lithium stearate 0.50 Potassium laurate 0.25 0.50 735 734 748 750 5646 Potassium stearate aEmulsifier concentration in the aqueous phase; b size variation coeffi- cient. Attempts to synthesise polymer suspensions consisting of larger particles with a narrow size distribution failed: either the reaction system was unstable, resulting in a substantial amount of the coagulum, or the particle size distribution was too broad.Polymerisation of styrene accompanied by the formation of barium, calcium and zinc salts of carboxylic acids taken in the same concentration occurs in a different way. This was demon- strated in relation to myristates. In this case, the diameter of particles scarcely depends on the metal nature and equals *800 nm (Table 7). The average diameter of particles also remained nearly constant as the degree of monomer conversion changed, and the polymer suspensions had a narrow distribution Table 7. Polymerisation of styrene under simultaneous synthesis of myristates at the interface. Surfactant 1075MM d /nm (see a) wp 102 /mol litre71 s71 x-Poten- tial /mV 700 6.7 5.5 729.5 800 6.6 5.4 725.7 Barium myristate Calcium myristate Zinc myristate 4.9 850 6.2 723.2 Note.Here and in Tables 8, 9 and 11: polymerisation temperature 70 8C, volume ratio of the phases 1 : 10, potassium persulfate concentration in relation to the monomer 1 mass %, emulsifier concentration in the aqueous phase 0.4 mass %. a Stability of all the particles to electrolytes was 0.2 mol litre71. Degree of conversion (%) 797 100 2 3 1 80 60 40 200 6 10t/h 2 Figure 9. Yield of polymer vs. time of styrene polymerisation under conditions of formation of emulsifiers at the interface. Emulsifier: (1) barium myristate; (2) calcium myristate, (3) zinc myristate.Here and in Figure 10: polymerisation temperature 70 8C, monomer : water volume phase ratio 1 : 10, potassium persulfate concentration in relation to the monomer 1 mass %, emulsifier concentration in the aqueous phase 0.4 mass %. of particles over diameters. This implies that, from the very moment of formation of the polymer-monomeric particles, a strong interfacial layer appears on their surface and protects the particles from coalescence.42, 43 The kinetic curves for styrene polymerisation under the conditions of synthesis of barium, calcium and zinc myristate at the interface are presented in Fig. 9. The curves are all S-shaped, which is typical of heterophase polymerisation; the formation of polymer-monomeric particles occurs over a period of 120 ± 240 min, and the time it takes to achieve the complete conversion of styrene is 10 ± 12 h.It is of interest to compare these data with the results obtained for styrene polymerisation under conditions in which the mono- mer emulsions were formed via emulsification of the monomer with aqueous suspensions of insoluble salts of myristic acid. The resulting polymer suspension had low stability, and the average particle diameter or the size distribution could not be determined due to the high content of coagulum (more than 30%). Thus, polymerisation with the synthesis of surfactant at the interface has certain advantages pointing to good prospects of this method. As was to bexpected, the change in the concentration of the insoluble salt synthesised at the interface (Table 8) barely influ- ences the rate of polymerisation, the particle diameter and the particle size distribution, which remains narrow over the whole range studied.The increase in the polymer molecular mass with an increase in the concentrations of the insoluble salts synthesised at the interface is, apparently, due to the decrease in the concen- tration of radicals entering the PMP from the aqueous phase, caused by the increase in the thickness and viscosity of the Table 8. Effect of the emulsifier concentration (barium myristate) on the rate of styrene polymerisation, the size of particles in the polymer suspensions and on the molecular mass of the resulting polymer. Degree of conversion (%) 1075MM d/nm /mol litre71 s71 Emulsifier concentration wp 102 in the aqueous phase (mass %) 695 700 720 780 820 850 920 7.0 6.7 6.4 5.6 4.4 2.0 1.5 5.5 5.5 5.3 5.2 4.9 4.8 4.7 0.8 0.4 0.1 0.05 0.025 0.0125 0.00625798 interfacial adsorption layer, resulting in the appearance of diffu- sion restrictions.The stability of PMP during polymerisation is ensured by the structure-mechanical and electrostatic factors of stabilisation acing in the interfacial layer. The x-potential of the particles in a polymer suspension is727 mV, which is much lower than that for styrene polymerisation under the same conditions but without an emulsifier (770 mV).44 The low x-potential can be accounted for by diffusion restrictions arising due to a definite orientation of the functional groups of polymer chains at the interface. Thus, styrene polymerisation accompanied by the synthesis of salts of myristic acid insoluble in both phases makes it possible to prepare polymer suspensions with a narrow particle size distribu- tion (the particle diameter is 600 ± 900 nm).The conditions of synthesis of these polymer suspensions are technologically simple and the reaction system remains stable during polymerisation. Due to the low x-potentials, the particles of a polystyrene suspension stabilised by barium, calcium and zinc myristates prove unstable during immobilisation of bioligands in an isotonic solution. An increase in the stability of polymer suspensions without substantial change in the particle diameter and without deterioration of the narrow size distribution of particles can be attained by using the synthesis of mixtures of potassium and barium myristates at the interface during polymerisation.45 When the content of the water-soluble potassium salt in the salt mixture is 6.25 mass%to 12.5 mass %, polymerisation gives rise to stable polymer suspensions with a mean particle diameter of 580 ± 680 nm and a narrow particle size distribution (Table 9).An increase in the stability is explained by an increase in the x- potential of the particles to740 mV. Table 9. Dependence of the rate of styrene polymerisation, the molecular mass of the polymers and the average diameter of particles in a polymer suspension on the ratio of emulsifiers, barium and potassium myristates (the total concentration of emulsifiers in the aqueous phase was 0.4 mass %).1075MM Emulsifier percentages d /nm (see a) 102 wp /mol litre71 s71 potassium myristate barium myristate 100 130 25 19 14.4 12.7 8.5 148.5 7.8 5.8 5.5 50 37.5 25 12.5 6.25 50 62.5 75 87.5 93.75 280 580 680 a For particles with sizes of 100 and 130 nm, the stability to electrolytes was 0.15 mol litre71; for other particles, it was 0.20 mol itre71. It of interest to compare the properties of the polymer suspen- sions prepared under conditions of surfactant formation at the interface with the properties of those prepared by conventional polymerisation (i.e., with emulsification of the monomer with an aqueous solution of an emulsifier) as well as the kinetic features of Table 10.Characteristics of styrene bulk polymerisation (I) and polymerisation in microdroplets of various sizes (II, III). I Characteristics 771012 ±1013 1071± 1072 1012 ±1013 1 The number of particles The number of radicals per particle The total number of radicals in 1 ml of the system Volume of the monomer per radical /mm3 Rate of initiation /radical ml71 s71 Average lifetime of radicals /s 1012 ±1013 5 ± 10 0.561013 ± 0.561014 1071± 1073 1012 ±1013 5 ± 10 a The size of microdroplets is 0.05 ± 0.1 mm (water-soluble surfactants); b the size of microdroplets is 0.4 ± 0.8 mm (water-insoluble surfactants).1 2 3 4 5 60 40 Degree of conversion (%) N I Prokopov, I A Gritskova 100 80 200 2 6 10t/h Figure 10. Yield of polymer vs. time of styrene polymerisation in the presence of emulsifiers. Emulsifier: (1) potassium myristate; (2) lithium myristate, (3) barium myristate, (4) calcium myristate; (5) zinc myristate. both processes. As an example, we shall consider polymerisation of styrene accompanied by the synthesis of myristic acid salts at the interface. The kinetic curves for styrene polymerisation with the syn- thesis of the myristates under discussion are presented in Fig. 10. They all have an S-like shape, typical of heterophase polymer- isation, and differ from one another in the duration of the non- steady-state stage of the formation of polymer-monomeric par- ticles, the rate of polymerisation, the range of degrees of con- version in which the rate of the process is constant, and in the time required for the full monomer conversion.For example, polymer- isation of styrene accompanied by the synthesis of potassium myristate starts almost immediately, while in the case of synthesis of barium, calcium, and zinc salts insoluble both in water and in the monomer, the formation of polymer-monomeric particles requires 120 ± 240 min. The time periods required to attain complete conversion of styrene are also different: in the former case, polymerisation occurs over a period of 2.5 ± 3 h and in the latter case, this takes 10 ± 12 h.For clarity, we shall compare the processes taking place in the PMP formed in the styrene polymerisation performed simulta- neously with the synthesis of soluble and insoluble salts of carboxylic acids (Table 10). The concentration of free radicals during bulk polymerisation is known 46 to be usually about 1077 ± 1078 mol litre71, or 1012 ±1013 radical ml71, i.e., the vol- ume per radical is, on average, 561071± 561072 mm3. The microdroplets formed under conditions of synthesis of soluble surfactants at the interface have a volume of 1073 ± 1074 mm3, which is much less than the volume per radical for bulk polymer- isation. Therefore, each particle does not contain more than one radical; only when the termination rates are very low (for example, when radicals are fixed in the polymeric matrix of the interfacial layers), can a microdroplet contain several radicals.If the concen- tration of microdroplets in the system is 1014 ml71, the rate of initiation is close to that for bulk polymerisation III b II a 1014 ±1015 0.5 0.561014 ± 0.561015 1073± 1074 1012 ±1013 10 ± 100Characteristic features of heterophase polymerisation of styrene with simultaneous formation of surfactants at the interface (1077± 1078 mol litre71 s71, or 1012 ± 1013 radical ml71 s71), and the initiation efficiency is equal to unity, the lifetime of radicals in these particles should be fairly long (10 ± 100 s). With this lifetime, the molecular mass of the polymer is determined by the rate of chain transfer via the monomer.In the case of styrene, MM=16106± 26106. A different situation arises when styrene polymerisation proceeds in the monomer microdroplets formed in the synthesis of water-insoluble surfactants at the interface. Their volume is now 1071 ± 1073 mm3; therefore, they can accommodate several radicals (5 to 10), the lifetime of which is several seconds, and the polymer molecular mass is up to 105. These differences are consistent with the data on the diameter of the polymer-monomeric particles in which polymerisation proceeds (Table 11). Indeed, in the styrene polymerisation accom- panied by synthesis of potassium myristate at the interface, the average diameter of the PMP is 63 nm, while for the synthesis of barium myristate under the same conditions, the diameter is an order of magnitude greater and equals 700 nm. Table 11.Polymerisation of styrene in the presence of myristates. 1075MM Surfactant d /nm (see a) 102 wp /mol litre s71 x-Poten- tial /mV 65 270 700 800 850 Potassium myristate 15.1 11.3 5.5 5.4 4.9 752 735 729 726 723 Lithium myristate Barium myristate Calcium myristate Zinc myristate 50 156.7 6.6 6.2 a For particles with sizes of 65 and 270 nm, the stability to electrolytes is 0.15 mol litre71; for other particles, 0.20 mol litre71. In the former case, when small particles contain not more than one radical, emulsion polymerisation of styrene takes place, which is characterised by a high rate and gives polymers with high molecular masses.In this case, the PMP are formed, apparently, both from the monomer microdroplets and from micelles of the emulsifier which solubilises the monomer. In large particles with an average diameter of about 700 nm, polymerisation follows a suspension mechanism and has a rate close to the rate of suspension polymerisation of styrene in the presence of conventional stabilisers of suspension particles, namely, poly(vinyl alcohol), gelatin, etc. However, the molecular masses of the polymers formed in the PMP stabilised by water- insoluble salts of myristic acid are higher than those attained in styrene polymerisation in the presence of standard stabilisers. Presumably, the high viscosity of the interfacial layer creates diffusion restrictions which prevent the penetration of radicals and the monomer into the layer bulk; therefore, only radicals that have entered the PMP during initiation of the polymerisation and the monomer present in the PMP bulk participate in the polymer- isation.V. Conclusion On the basis of the published data cited and the results of authors' research, a new, in principle, approach to the conduct of hetero- phase polymerisation has been proposed.47 The essence of the approach is that monomer emulsions are prepared simultaneously with the synthesis of the emulsifier at the monomer ± water inter- face and with initiation of the polymerisation in the interfacial layer. For the preparation of emulsions of monomers sparingly soluble in water, the formation of the emulsifier at the mono- mer ± water interface is an efficient way of controlling the stability and the granulometric composition of the emulsions. 799 The synthesis of emulsifiers at the interface makes it possible to control the interfacial tension, disintegration and microemulsi- fication of the monomer, to form the structure of interfacial adsorption layers of surfactants by an appropriate selection of the carboxylic acid and the counter-ion. The mechanism of formation of the polymer-monomeric particles and their diameter and size distribution are dictated by the solubility in water of the surfactant formed at the interface.When polymerisation is carried out under conditions of synthesis of a soluble surfactant at the interface, the polymer-monomeric particles are produced from the emulsifier micelles and monomer microdroplets.This ensures the formation of polymer suspensions with a broad particle size distribution. Owing to the low concen- tration of the resulting surfactant, which is, however, sufficient for providing the stability of polymer-monomeric particles, micelles are excluded from the formation of the PMP, and polymer suspensions with narrow size distributions of particles are formed. When surfactants poorly soluble or insoluble in water are synthesised at the interface, the PMP are mainly produced from monomer microdroplets, and a narrow size distribution of par- ticles in the resulting suspensions is achieved.References 1. M S El-Aasser, R M Fitch (Eds) Future Directions in Polymer Col- loids (Dordrecht: Martinus Nijhoff, 1987) 2. E S Daniels, E D Sudol, M S El-Aasser (Eds) Polymer Latexes: Preparation, Characterization and Applications (Dordrecht: Martinus Nijhoff, 1992) 3. R G Gilbert Emulsion Polymerization: a Mechanistic Approach (London: Academic Press, 1995) 4. P A Lovell,M S El-Aasser (Eds) Emulsion Polymerization and Emulsion Polymers (New York: Wiley, 1997) 5. M A Filatova, S A Nikitina, A B Taubman, P A Rebinder Dokl. Akad. Nauk SSSR 140 874 (1961) a 6. V A Spiridonova, S A Nikitina, A B Taubman Dokl. Akad. Nauk SSSR 182 640 (1968) a 7. F K Hansen, J Ugelstad J. Polym. Sci., Polym. Chem.Ed. 17 3069 (1979) 8. J Ugelstad, H R Mfutakamba, P C Mùrk, T Ellingsen, A Berge, R Smied, L Holm, A Jùrgedal, F K Hansen, K Nustad J. Polym. Sci., Polym. Symp. 72 225 (1985) 9. S S Medvedev, in Kinetika i Mekhanizm Obrazovaniya i Prevra- shcheniya Makromolekul (Kinetics and Mechanism of Formation and Conversion of Macromolecules) (Moscow: Nauka, 1968) p. 5 10. I A Gritskova, L I Sedakova, D S Muradyan, B M Sinekaev, A V Pavlov, A N Pravednikov Dokl. Akad. Nauk SSSR 243 403 (1978) a 11. I A Gritskova, L I Sedakova, D S Muradyan, A N Pravednikov Dokl. Akad. Nauk SSSR 238 607 (1978) a 12. G A Simakova, V A Kaminskii, I A Gritskova, A N Pravednikov Dokl. Akad. Nauk SSSR 276 151 (1984) a 13. S V Zhachenkov, G I Litvinenko, V A Kaminskii, P E Il'menev, A V Pavlov, V V Gur'yanova, I A Gritskova, A N Pravednikov Vysokomol.Soedin., Ser. A 27 1249 (1985) b 14. I A Gritskova, S V Zhachenkov, N I Prokopov, P E Il'menev Vysokomol. Soedin., Ser. A 33 1476 (1991) b 15. S I Tregubenkov, A A Bryzgachev, L I Sedakova, I A Gritskova, A N Pravednikov Vysokomol. Soedin., Ser. A 30 322 (1988) b 16. I M Yakovleva, G A Simakova, I A Gritskova, R A Lyubimskaya, G S Virasuriya, A N Pravednikov Kolloid. Zh. 50 610 (1988) c 17. R A Lyubimskaya, G S Virasuriya, I M Yakovleva, G A Simakova, I A Gritskova Kolloid. Zh. 50 562 (1988) c 18. G I Litvinenko,M Khaddazh, V A Kaminskii, A V Pavlov, A G Davtyan, I A Gritskova, A N Pravednikov Vysokomol. 19. G A Simakova, Doctoral Thesis in Chemical Sciences, Moscow 20.A N Pravednikov, G A Simakova, I A Gritskova, N I Prokopov Soedin., Ser. B 26 683 (1984) a Institute of Fine Chemical Technology, Moscow, 1990 Kolloid. Zh. 47 189 (1985) c 21. A N Pravednikov, G A Simakova, I A Gritskova, N I Prokopov Kolloid. Zh. 47 192 (1985) cN I Prokopov, I A Gritskova 800 22. N I Prokopov, Candidate Thesis in Chemical Sciences, Moscow Institute of Fine Chemical Technology, Moscow, 1987 23. N I Procopov, G A Simakova, G M Plavnik, A I Koshevnikov, E Ghiva Polymery 37 516 (1992) 24. N I Prokopov, I A Gritskova, Zh V Maksimenko, I Legocka, I Grzywa-Niksinska, E I Grzywa Polimery 40 568 (1995) 25. N I Prokopov, G P Kheinman, I A Gritskova, M V Chirkova Kolloid. Zh. 57 561 (1995) c 26. N I Prokopov, I A Gritskova Kolloid. Zh. 61 264 (1999) a 27. C A Miller Colloids Surf. 29 89 (1988) 28. M Vermeulen, P Joos Colloids Surf. 36 13 (1989) 29. J Rudin, D T Wasan Ind. Eng. Chem. Res. 31 1899 (1992) 30. J Rudin, C Bernard,D T Wasan Ind. Eng. Chem. Res. 31 1899 (1994) 31. A A Zhukhovitskii, V A Grigoryan, E Mikhalik Dokl. Akad. Nauk SSSR 155 392 (1964) a 32. V A Grigoryan, A A Zhukhovitskii, E Mikhalik Zh. Fiz. Khim. 39 1179 (1965) d 33. G A Grigor'ev, T V Ingerova Zh. Fiz. Khim. 72 1103 (1998) d 34. Y Moroi, R Matuura J. Phys. Chem. 89 2923 (1985) 35. L V Kolpakov, S A Nikitina, A B Taubman, V A Spiridonova, A E Chalykh, N I Puchkov, V A Luk'yanovich Kolloid. Zh. 32 229 (1970) c 36. A Rikhavi, Candidate Thesis in Chemical Sciences, Moscow Institute of Fine Chemical Technology, Moscow, 1990 37. W Linke (Ed.) Solubilities Inorganic and Metalloorganic Compounds (Washington, DC: American Chemical Society, 1965) 38. I A Gritskova, V A Kaminskii Zh. Fiz. Khim. 70 1516 (1996) d 39. P E Il'menev, G I Litvinenko, V A Kaminskii, I A Gritskova, 40. I A Gritskova, A A Zhdanov, O V Chirikova, O I Shchegolikhina 41. O V Chirikova, Candidate Thesis in Chemical Sciences, Moscow 42. D Yu Mityuk, Candidate Thesis in Chemical Sciences, Moscow 43. S Ya Shalyt, Doctoral Thesis in Chemical Sciences, Moscow 44. Kh Bakharvand, A A Kapustina, I Gzhiva-Niksin'ska, A V Pavlov Dokl. Akad. Nauk SSSR 301 638 (1988) a Dokl. Akad. Nauk 334 57 (1994) a State Academy of Fine Chemical Technology, Moscow, 1994 Institute of Fine Chemical Technology, Moscow, 1985 Institute of Fine Chemical Technology, Moscow, 1987 A A Oganesyan, I A Gritskova, P V Nuss, V N Izmailova 45. G P Kheinman, Candidate Thesis in Chemical Sciences, Moscow 46. G Odian Principles of Polymerization (New York: Wiley-Interscience, 47. N I Prokopov, Doctoral Thesis in Chemical Sciences, Moscow Kolloid. Zh. 59 299 (1997) c State Academy of Fine Chemical Technology, Moscow, 1999 1991) State Academy of Fine Chemical Technology, Moscow, 1999 a�Dokl. Chem. (Engl. Transl.) b�Polym. Sci. (Engl. Transl.) c�Colloid J. (Engl. Transl.) d�Russ. J. Phys. Chem. (Engl.

ISSN:0036-021X    

出版商:RSC    

年代:2001

数据来源: RSC

摘要:

Russian Chemical Reviews 70 (9) 801 ± 807 (2001) Anion-conducting fluoride and oxyfluoride glasses N I Sorokin Contents I. Introduction II. Fluoride glasses based on MF2 III. Fluoride glasses based on MF3 IV. Fluoride glasses based on MF4 V. Oxyfluoride glasses VI. Optimisation of characteristics of fluoride-conducting glasses and prospects of their practical applications VII. Conclusion Abstract. transport ion) (fluoride anion the of mechanisms Main Main mechanisms of the anion (fluoride ion) transport in variation The analysed. are glasses oxyfluoride and fluoride in fluoride and oxyfluoride glasses are analysed. The variation in in the compo- glass the on depending of number transport the transport number of F7 ions ions depending on the glass compo- sition is followed.Methods for optimisation of the conductivity sition is followed. Methods for optimisation of the conductivity characteristics of the glasses in question are considered. The characteristics of the glasses in question are considered. The prospects of using fluoride-conducting glasses in electrochemical prospects of using fluoride-conducting glasses in electrochemical devices are discussed. The bibliography includes 90 references devices are discussed. The bibliography includes 90 references. I. Introduction The discovery 1 of the anionic (fluoride) conductivity of fluoro- beryllate glasses of the BeF2+MF (M=Na or Cs) type and fluorophosphate glasses of the Ba(PO3)2+MF2 (M=Ca, Sr, Ba or Mg) type and, particularly, the preparation of fluorozirconate glasses in systems involving ZrF4 and BaF2 and investigations 2 of their transport properties gave impetus to extensive studies of fluoride transport in glassy materials.Glasses possessing high F7 ionic conductivity constitute a special class of anion-conducting solid electrolytes. Since the glassy state is characterised by a distinctive structural solid-state disordering, one would expect that high anionic conductivity can be achieved in amorphous materials. In addition, fluoride glasses are of practical interest in a search for new solid electrolytes because of their technological effectiveness and cheapness as compared to crystalline materials. A search for glassy solid electrolytes follows two main directions: first, new glass-forming systems are investigated and, second, the electrophysical properties of the known fluoride glasses are modified by doping (by increasing the number of components).Difficulties encountered in preparing fluoride glasses are associated with their high tendency to crystallisation. With the aim of suppressing this effect, various metal fluorides are intro- duced as stabilisers, which complicates the glass composition. By varying the composition, one can enhance the chemical stability of glasses, prepare specimens up to *1 cm in thickness,3 extend the temperature range of their application, reveal the contributions of various components to the dynamics of the ionic conductivities N I Sorokin A V Shubnikov Institute of Crystallography, Russian Academy of Sciences, Leninsky prosp.59, 119991 Moscow, Russian Federation. Fax (7-095) 135 10 11. Tel. (7-095) 330 78 74. E-mail: sorokin1@mail.ru Received 11 April 2001 Uspekhi Khimii 70 (9) 901 ± 908 (2001); translated by T N Safonova #2001 Russian Academy of Sciences and Turpion Ltd DOI 10.1070/RC2001v070n09ABEH000674 801 802 803 803 804 805 806 and obtain information on the mobility and concentration of charge carriers. Problems of the formation and atomic structures of glasses in fluoride systems were surveyed in reviews.4± 9 Structural models are generally constructed based on comparison of the structural data for glassy and crystalline phases with the same chemical composition. Presently, a disordered network is used as the basic model for the glass structure.10 According to this model, the glass structure represents a disordered (the long-range order is absent) three-dimensional glass-forming network consisting of coordina- tion polyhedra of glass-forming cations, which are linked to each other through shared vertices occupied by bridging F¡b anions.The cavities of the network are occupied by modifying cations bound to non-bridging F¡nb anions. To gain insight into the glass structure, the reasons for the formation of disordered networks need to be revealed. The model for a disordered network (the short-range order in the glass structure) is refined taking into account up-to-date experimental data. Among various structural models for fluoride glasses, the cluster (the short-range order and `medium-range' order in the glass structure) model (see, for example, Ref.11), which is closely related to the model for the disordered network, is most universally employed.8 Structures of fluoride crystalline phases, which are chemically close to glasses, contain infinite chains of coordination polyhedra of glass-forming cations linked through F¡b anions. The chains of the polyhedra, in turn, are linked in a three-dimensional disordered glass-forming network. Large low-charge modifying cations are not involved in this network and give rise to ionic cation7fluoride bonds. The introduction of modifying fluoride ions exhibiting high bond ionicity leads to strengthening of covalent bonds in the glass- forming network through the charge transfer.The keys to an understanding of the anion transport in fluoride glasses are, apparently, the formation of the F¡nb anions and the variation in the coordination number of glass-forming cations. It is the ability of glass-forming cations to adapt them- selves to the change in the coordination number that promotes jumps of mobile anions between the adjacent polyhedra. The results of 19F NMR studies 12 ± 14 and computer simulation 15 demonstrated that it is precisely the F¡nb anions located in the vicinity of the modifying cations that serve as mobile charge carriers. Local motions of F¡nb anions initiate the diffusive motion of fluoride anions (anionic conductivity) through the glass-form- ing network.802 Glasses possessing ionic conductivity are characterised by dielectric relaxation associated with the jump migration of charge carriers resulting in the frequency dependence of the ionic con- ductivity.16 Ion-conducting glasses follow the universal relation- ship: the direct-current conductivity s is directly proportional to the dielectric relaxation frequency om .This universal regularity was explained 17 within the framework of a model for the potential barrier distribution according to which the frequency of the dielectric loss peak is the minimum jump frequency of conducting ions. Hence, the directly proportional dependence of s on om is physically justified. Within the framework of this approach, the characteristic features of the jump mechanism for electric current carriers in ZrF4-based fluoride glasses were examined.18 ± 20 The anionic conductivity in fluoride glasses depends primarily on the activation energy of the ion transport.Particular mechanisms of the anion transport in glasses remain poorly studied because the available structural data are scarce. A necessary condition for the manifestation of essential anionic conductivity is the existence of glass-forming cations, which adopt high coordination numbers and differ in the form and size of their coordination polyhedra. Low coordination numbers of glass-forming cations (in fluoroberyllate and fluoro- aluminate glasses) are responsible for low anionic conductivities.For cation-conducting glasses, the polycationic (polyalkali) effect is well known. This effect lies in the fact that the replacement of one alkali metal cation by another gives rise to a minimum in the composition dependence of the ionic conductivity. The polycationic effect in fluoride and oxyfluoride glasses is man- ifested if the cationic conductivity exceeds the fluoride-ionic conductivity.21 ± 24 In the Ba(PO3)2+NaX (X=F, Cl or I) and Ba(PO3)2+BaF2+BaX2 (X=Cl or Br) glasses, the polyanionic effect (which was called by analogy with the polycationic effect) was experimentally observed upon the replacement of one halogen by another.25, 26 The polyanionic effect was confirmed for the (X=F or Cl) 27 ZrF4+BaX2 and 30 SiO2 . 55 PbO . 15 Pb(F17xClx)2 (04x41) glasses.28 The ZrF4+BaF2+MX (M=Li or Na; X=F or Cl) 29 and ZrF4+ BaF2+MFn+LiF glasses (Mn+=La3+ or Th4+, the Li+ and F7 mobile ions) exhibit the combined effect associated with lowering of the conductivity due to the simultaneous presence of mobile cations and anions.30 In the present review, the results of studies of the fluoride-ion transport in glassy solid electrolytes are analysed.Mechanisms of the cation transport in fluoride and oxyfluoride glasses call for a special discussion and are beyond the scope of the present review. The abundant experimental data on fluoride glasses are classified, wherever possible, into groups according to the type of the basis glass-forming fluoride. It should be noted that this classification becomes more difficult as the composition of multicomponent glasses is complicated (the most stable fluoride glasses contain three or more components).Various families of glasses based on MF2, MF3 or MF4 are considered. Recently, glasses based on MF5 , where M=Nb, Ta, V or Mo, have been discovered (the glass transition temperature Tg varies from 745 to 760 8C; the crystallisation temperature Tcr ranges from 0 to 710 8C),31 but their electrophysical properties remain to be explored. The char- acteristics of the anionic conductivities of some two- and three- component fluoride and oxyfluoride glasses are given in Table 1. Fluoride glasses attract considerable interest because they are transparent in the IR region and possess low glass transition temperatures as well as because they are promising materials for fibre optics.According to the results of calculations, theoretical losses in the optical transmission of information in these glasses are an order of magnitude smaller than those in traditional oxide (silicate) glasses. Table 1. Characteristics of the fluoride-ionic conductivity in two- and three-component fluoride and oxyfluoride glasses. Glass s /S m71 Fluoride glasses based on MF2 80 BeF2 . 20 CsF 67MnF2 . 33 BaF2 67 ZnF2 . 33 BaF2 55 ZnF2 . 45 PbF2 30 ZnF2 . 25 AlF3 . 45 LiF 35 ZnF2 . 20 BaF2 . 45 LiF 45 ZnF2 . 30 BaF2 . 25YF3 6610711 161074 261074 461074 161074 361073 161073 Fluoride glasses based on MF3 30 AlF3 . 40 PbF2 .30 ZnF2 40 AlF3 . 22 CaF2 . 38YF3 40 AlF3 . 40 BaF2 . 20 YbF3 30 GaF3 . 50 PbF2 . 20MnF2 40 GaF3 . 40 BaF2 . 20 YbF3 40 ScF3 . 40 BaF2 . 20 YbF3 40 ScF3 . 40 BaF2 . 20YF3 54 InF3 . 36 BaF2 . 10 LiF 25 InF3 . 45 PbF2 . 30 ZnF2 35 InF3 . 35 PbF2 . 30MnF2 35 InF3 . 35 PbF2 . 30 SnF2 40 InF3 . 45 PbF2 . 15 BaF2 40 InF3 . 40 BaF2 . 20 YbF3 30 InF3 . 35 BaF2 . 35 BiF3 261077 561079 461077 261074 161075 161073 161073 661073 461073 461073 261073 561072 161073 361072 Fluoride glasses based on MF4161075 361075 161074 161075 60 ZrF4 . 15 BaF2 . 25 CsF 62 ZrF4 . 30 BaF2 .8LaF3 60 ZrF4 . 30 BaF2 . 10 AlF3 59 ZrF4 .26 BaF2 . 15 BiF3 62.5 ZrF4 . 28 BaF2 . 9.5 ThF4 361075 59 ZrF4 . 26 BaF2 . 15 HfF4 60 ZrF4 . 33 ThF4 . 7 LaF3 70 HfF4 . 30 BaF2 55 HfF4 . 30 BaF2 . 15 CsF 62 HfF4 . 30 BaF2 . 8 NdF3 62 HfF4 . 30 BaF2 . 8 YbF3 261075 361076 861075 261075 961075 961075 Oxyfluoride glasses 861077 161072 861073 20 Al(PO3)3 . 80 PbF2 10 Al(PO3)3 . 85 PbF2 . 5MnF2 10 Pb(PO3)2 . 35 PbF2 . 55MnF2 20NaPO3 . 40 BaF2 . 40MnF2 461076 10 B2O3 . 90 PbF2 30 B2O3 . 50 PbO . 20 PbF2 5Bi2O3 . 95 PbF2 10V2O3 . 90 PbF2 10 SnO2 . 90 PbF2 10GeO2 . 90 PbF2 15GeO2 . 15 PbO . 15 PbF2 10GeO2 . 80 PbF2 . 10 BiF3 10GeO2 . 80 PbF2 . 10 PbCl2 35 SiO2 . 50 PbO . 15 PbF2 30 SiO2 .30 PbO . 40 PbF2 161073 261075 161074 561077 261074 361075 361075 961074 161073 561075 661074 Note. Ea is the activation energy of the conductivity. Different researchers calculated the Ea values according to the formula: sT=AexpÖ¡Ea=kT Ü or s=AexpÖ¡Ea=kT Ü. The results differ by less than 0.05 eV. a The original studies were cited in Ref. 32. N I Sorokin Ref. Ea /eV T /8C 1.12 0.55 0.59 0.55 0.86 0.85 0.73 150 180 180 180 200 200 200 32 a 32 32 32 33 34 34 32 32 35 32 35 35 12 36 32 32 32 37 35 38 0.79 0.83 1.13 0.61 0.96 0.75 0.70 0.64 0.57 0.56 0.49 0.535 0.78 0.52 150 150 200 200 200 200 200 200 200 200 150 200 200 200 32 32 39 32 32 32 32 35 32 35 35 0.74 0.79 0.77 0.73 0.79 0.76 0.83 0.925 0.71 0.88 0.88 200 200 200 200 200 200 200 200 200 200 200 32 32 1.00 0.63 200 200 32 0.56 200 12 34 40 34 34 34 34 41 34 34 32 42 1.09 0.91 0.55 0.85 0.85 0.79 1.05 10.80 0.84 0.74 0.75 200 200 127 200 200 200 200 300 200 200 200 200Anion-conducting fluoride and oxyfluoride glasses II.Fluoride glasses based on MF2 Of simple fluorides MFn (n=1 ± 4), BeF2 is the only fluoride compound capable of undergoing a glass transition. The coordi- nation polyhedron of the glass-forming Be2+ cation is the [BeF4]27 tetrahedron (the coordination number is 4). The struc- tural motif of crystalline BeF2 (isostructural to SiO2) is based on an ordered three-dimensional framework formed by the [BeF4]27 tetrahedra, which are linked to each other via shared vertices.The transformation of BeF2 into the glassy state occurs through rotations of the tetrahedra about the bridging Be7Fb7Be bonds accompanied by distortions of the angles in the tetrahedra. As a result, a disordered three-dimensional framework consisting of the distorted [BeF4]27 tetrahedra is formed. This framework is characterised by the absence of the long-range order. Tetrahedral fluoroberyllate glasses are structural analogues of silicate glasses. The introduction of alkali metal fluorides acting as modifiers leads to the cleavage of the bridging Be7Fnb7M bonds and the occurrence of non-bridging fluorine atoms.It was found that the ratio between the cationic and anionic components of the con- ductivities in alkaline fluoroberyllate glasses is a function of the alkali metal7fluorine bond. Measurements of the transport numbers demonstrated that the BeF2+MF and BeF2+MF+M0F glasses (M and M0=Na or Cs) exhibit virtually unipolar anionic conductivities.1 The mechanism of the anionic conductivity in these glasses is associated with the increase in the amount of mobile non-bridging F¡nb ions as the content of the MF modifying component increases. However, the fluoride- ionic conductivities are low and reach*1076 S m71 at 300 8C. The electrical properties of multicomponent glasses based on manganese and zinc fluorides in the MnF2 ± BaF2 , ZnF2±MF2 (M=Ba or Pb), ZnF2 ± BaF2 ± LiF, ZnF2 ± BaF2±YF3 , ZnF2 ± PbF2 ± AlF3 ± LiF and ZnF2 ± BaF2 ± InF3 ± SrF2 ±NdF3±MF systems (M=Li, Na, K or Rb) were studied.33, 34, 43, 44.Appa- rently, crystalline phases formed in the ZnF2 ± BaF2 system, for example BaZnF4 , are most similar in structural units to fluoro- zincate glasses.45, 46 The crystal structure of BaZnF4 consists of the [ZnF6]47 octahedra linked in chains through shared vertices, whereas the barium atoms are located in the cavities. An analo- gous motif composed of the [ZnF6]47 octahedra is observed in the structure of Ba2ZnF6 . It can be assumed that fluorozincate glasses retain the short-range order of the arrangement of the [ZnF6]47 octahedra, whereas the Ba2+ positions are partially occupied by the modifying Sr2+, Nd3+, Y3+, Li+, Na+, K+ or Rb+ cations, their arrangement meeting the condition for charge compensa- tion.The indium atoms, which adopt the coordination numbers of 6 and 7 in fluoride structures,9 are, apparently, incorporated into the glass-forming network. In the octahedral 50 ZnF2 . 19.5 BaF2 . 10 InF3 . 10 SrF2 . 0.5NdF3 . 10MF glasses, the con- ductivity at 127 8C changes from 561076 to 361074 S m71 and the activation energy Ea changes from 0.82 to 0.68 eV as the ionic radius ofM+ increases from 0.9 (Li+) to 1.66 A (Rb+) {. In the 30 ZnF2 . 25 AlF3 . (457x) PbF2 . x LiF (04x420, Tg= 230 ± 275 8C) glasses, the ZnF2 ± AlF3 group acts as a glass- forming agent, whereas PbF2 and LiF serve as modifiers.The anionic conductivity (at 200 8C) decreases from 161074 to 161075 S m71, whereas the activation energy Ea increases from 0.86 to 1.24 eV as the LiF content increases due to capture of mobile F7 ions by Li+ cations. III. Fluoride glasses based on MF3 Trifluoride-based glasses are formed in systems containing MF3 (M=Al, Fe, Ga or In) or AF2 (A=Ba or Pb).48, 49 As for the glass-formingMF3+M0F2+M00F2 (M=Fe, Ga or In;M0=Pb or Ba; M00=Mn, Zn, Cd, Sn or Ba) systems, the most stable glasses appear in the InF3 ± BaF2 ± PbF2 system (Tg=250 ± { The crystal-chemical ionic radii were taken from Ref. 47. 803 280 8C, DT=Tg7Tcr&70 8C).The structures of MF3-based glasses consist of the [MF6]37 octahedra linked in zigzag chains through fluorine atoms. The [AlF4]7 and [GaF4]7 tetrahedra can be present, but they are not involved in formation of glass networks. 46.7 InF3 . 19 PbF2 . The data on the 23.8 BaF2 . 1.9 AlF3 . 4.9 SrF2 . 2.8YF3 glass obtained by Raman spectroscopy, the optical transmission method and magneto- chemistry 50, 51 are indicative of octahedral coordination about the small indium (rIn3á=1.06 A) and aluminium (rAl3á=0.675 A) atoms; whereas the large lead (rPb2á=1.43 A) and barium (rBa2á=1.56 A) atoms act as modifiers and occupy cavities in the three-dimensional octahedral glass-forming network. Considera- tion of the structure of this glass from the standpoint of the close packing of fluorine atoms leads to the same conclusion, viz., small In3+ and Al3+ cations occupy octahedral positions in the anionic packing, whereas large Pb2+ and Ba2+ cations introduce strong disturbance in the packing of fluoride anions.52 It was assumed that the [InF7]47 polyhedra are present in the glass-forming networks of indium-containing glasses.9 The properties of the 40MF3 .40 BaF2 . 20 YbF3 glasses (M=Al, Ga, Sc or In) whose compositions are determined by the position of a low-melting eutectic in the phase diagrams of the MF3 ± BaF2 ± YbF3 systems were examined.35 In the course of formation of these glasses, the MF3 trifluorides serve as glass- forming agents. The structures of the 40 ScF3 .40 BaF2 .20 YbF3 and 40 InF3 . 40 BaF2 . 20 YbF3 glasses contain fragments both of two-dimensional networks formed by the seven- and eight-vertex [MFn]7n+3 polyhedra (n=7 or 8) and chains formed by the [MF6]37 octahedra. Since the Al3+ and Ga3+ ions cannot be incorporated in seven-vertex polyhedra, the structure of the glass- forming network changes.53 These glasses are characterised by a sharp increase (by three orders of magnitude) in the isothermal conductivity as the ionic radius of M3+ increases (as the M3+7F7 chemical bond is weakened) due to a linear decrease in the activation energy of the conductivity (see Table 1). The introduction of GaF3 into the fluoroindate glasses (407x) InF3 . x GaF3 . 20 ZnF2 .16 BaF2 . 20 SrF2 . 2GdF3 . 2 NaF (04x430) also leads to a sharp decrease in their conductivities.54 The mobility of the fluoride ions and the conductivities of glasses in the InF3 ± BaF2 ± PbF2 , InF3 ± BaF2 ± BiF3 and InF3 ± BaF2 ± PbF2 ± AlF3 ± LiF systems were examined.36 ± 38, 55 The addition of AlF3 and LiF to the former system promotes the preparation of InF3+ BaF2+PbF2 glasses and allows one to vary their composition over a wider range. Two phases were identified among the crystallisation products obtained in the InF3 ± BaF2 ± PbF2 ± AlF3 system,56 viz., a fluorite-like solid sol- ution based on PbF2 and BaF2 and a solid solution based on the high-temperature tetragonal modification of Ba3In2F12 . The conductivities of glasses of the InF3 ± BaF2 ± PbF2 ± AlF3 ± LiF system are improved as the PbF2 content increases, whereas the conductivities are impaired as the AlF3 content increases.At room temperature, the anionic conductivity of the 20 InF3 . 50 PbF2 . 10 BaF2 . 10 AlF3 . 10 LiF glass is 30 times as large as that of the superionic conductor b-PbF2 , which is widely used in scientific and applied studies. This fact underlines the important role of indium and lead fluorides in the preparation of new amorphous superionic conductors. IV. Fluoride glasses based on MF4 In the middle 1970s, ZrF4 and BaF2-based fluoride glasses were prepared. This result was unexpected because the glass-forming Zr4+ cation is characterised by high (56) coordination numbers, which is contradictory to the conventional criteria for glass formation.The data of analysis of the shielding tensors and the chemical shifts of the signals in the 19F NMRspectra 14, 57 and the results of calculations by the Monte-Carlo method and molecular dynamics 58 demonstrated that the short-range structures in the barium fluorozirconate glasses 50 ZrF4 . 50 BaF2(BaZrF6) and 60 ZrF4 . 40 BaF2 are similar to the structure of crystalline804 a-BaZrF6 and that the medium-range ordering of the cluster type occurs along with the short-range order. Fluorozirconate glasses can contain Zr polyhedra with the coordination numbers of 6, 7 or 8. The structural networks of the glasses consist predominantly of the [ZrF7]37 and [ZrF8]47 polyhedra linked through bridging bonds; BaF2 acts as a modifier.59 The glass-forming ability of the ZrF4 ± BaF2 system is insufficient for the preparation of glasses several millimetres in thickness.These glasses can be prepared by introducing additional MFn components (n=1 ± 4); the number of the components can be 10 or even larger. The pioneering study on the anionic conductivities of fluoro- zirconate glasses 2 was concerned with the glasses ZrF4+BaF2+ThF4 (55 mol.% ± 62.5 mol.% ZrF4, 28 mol.% ± and 7.5 mol.% ± 13.8 mol.% 36.3 mol.% BaF2 ThF4 , Tg&320 8C), 62 ZrF4 . 30 BaF2 .8MF3 (M=La, Pr or Nd; Tg=310 8C), 60 ZrF4 . 33 ThF4 . 7 LaF3 (Tg=455 8C) and 60 ZrF4 . 25 BaF2 . 8 ThF4 . 7 LaF3 (Tg=320 8C). Measurements of the transport numbers demonstrated that the glasses possess unipolar fluoride-ionic conductivities (the transport numbers of the electron transport are <1073).In the ZrF4+BaF2+ ThF4 glasses, the conductivity increases moderately as the content of the BaF2 modifier increases. On the average, the fluoride-ionic con- ductivity in these glasses is *1074 S m71 at 200 8C and Ea&0.8 eV. Later on, the electrophysical properties of the multicompo- Li),19, 20 nent glasses ZrF4+BaF2+AlF3 ,39 ZrF4+ BaF2+MFn (M=Li, Na, Rb, Cs, Ca, Sr, Ba, Sb3+, Bi, La, Nd or Hf),60 60 ZrF4 . 30 BaF2 . 10 LnF3 (Ln=La ± Lu or Y),61 ZrF4+ M0F2+M00F3 (M0=Ba or Sn2+; M00=Ga, La or Nd),62, 63 ZrF4+BaF2+AlF3+LnF3 (Ln=La or Y),64 ZrF4+BaF2+ LaF3+M0F (M0=Na or ZrF4+BaF2+ ThF4+LiF,20 ZrF4+MFn+AlF3+YF3+NaF (M=Ba or La),18, 19, 65 ZrF4+BaF2+LaF3+LiF+NaF66 and ZrF4+ BaF2+AlF3+YF3+ZnF2+CdF2+LiF 67 were examined with the aim of improving the anion-transport characteristics. It was found that the anionic conductivity of fluorozirconate glasses depends primarily on the activation energy of the conductivity.The polarisability and the ionic radius of Mn+ are the major factors influencing the Ea value. The activation energy of the fluoride-ion transport decreases as the ionic radius and the polar- isability of theMn+ cations increases, which is, apparently, due to weakening of theMn+7F7 bond. The ZrF4 ± BaF2 ± CsF system exhibits the most promise for preparing glasses possessing high fluoride-ionic conductivity.According to the 19F NMR data,12 ± 14, 64 the mechanism of the anionic conductivity in fluorozirconate glasses is associated with the non-bridging F¡nb anions localised at the modifying cations. These anions execute local motions (Ea&0.2 eV) in the vicinity of the modifying cations and diffusion motions (Ea&0.6 eV) between the modifying cations. The fluorine diffu- sion is limited by the potential barriers between the modifying cations. In the 55 ZrF4 . (38.57x) BaF2 . 4AlF3 . (2.5+x) LaF3 (04x47) and 55 ZrF4 . (377x) BaF2 . 4 AlF3 . (4+x)YF3 (04x45) glasses, the conductivity is enhanced as the BaF2 content increases.64 In lithium fluorozirconate glasses containing less than 20 mol.% of LiF, the conductivities are determined by the Li+ and F7 ions, whereas the conductivities in the glasses containing more than 20 mol.% of LiF are conditioned only by the Li+ cations.In the sodium fluorozirconate glasses, the conductivities are governed exclusively by the F7 ions. For the 50 ZrF4 . (457x) BaF2 . 3 AlF3 .2YF3 . x NaF (104 x440) and 45 ZrF4 . 3 LaF3 . 3 AlF3 .4YF3 . 45 NaF glasses, the concentra- tions of the charge carriers were estimated within the framework of the jump mechanism of the conductivity.18, 19 A small portion of fluoride ions (3% ± 5%of their total amount) is responsible for the transport properties of these glasses. The conductivity decreases as the NaF content increases due to a decrease in the mobility of the charge carriers and partial capture of mobile F7 anions by Na+ cations. N I Sorokin Since zirconium and hafnium tetrafluorides (the ionic radii rZr 4á =0.98 and rHf 4á =0.97 A) have similar physicochemical and structural characteristics, fluorohafnate glasses are isovalent analogues of fluorozirconate glasses.The electrolytic properties of glasses in the HfF4 ± BaF2 and HfF4 ± BaF2±MF3 systems (M=Nd, Yb or Y) were investigated.35, 68 The parameters of the anion transport in HfF4-containing glasses are virtually independent of the type and concentration of trifluorides of rare- earth elements, viz., s&361074 S m71 at 223 8C and Ea&0.9 eV. Trivalent rare-earth cations replace zirconium (haf- nium) in glass-forming networks. Apparently, the addition of trifluorides of rare-earth elements with substantially different ionic radii (for example, rNd3á=1.249 and rYb3á =1.125 A) to the glass-forming HfF4 ± BaF2 system has no effect on the surface contour of the potential energy (*0.9 eV), where the fluoride ions move.These results of conductometric studies are, apparently, indicative of microheterogeneity of glasses including large (from the viewpoint of the atomic scale) spatial regions free of trifluor- ides of rare-earth elements. According to theNMRdata,69 high mobility of the F7 ions is observed in glasses based on tin tetrafluoride, viz., in SnF4 ± BaF2±MF3 (M=Al, Sc or Ga); however, the data on their ionic conductivities are lacking. V. Oxyfluoride glasses To prepare glasses, fluorides are added to mixtures of oxides.Thus, no glass was obtained from a mixture of pure oxides SiO2 , Al2O3 and BeO, whereas a transparent glass was prepared in the presence of even a small amount (several mass percent) of CaF2. This is associated with a decrease in viscosity and an increase in mobility of structural units in melts in the presence of fluoride. The properties of these glasses are determined by the influence of fluorides on the structure of the oxide matrix. Glasses possessing unipolar fluoride-ionic conductivities were prepared by introduc- ing fluorides MFn (n=1 ± 3) into glass-forming oxyphosphate compounds of the NaPO3 , M(PO3)2 (M=Ba, Pb, Sr or Zn) and Al(PO3)3 types.1, 32, 70 In the Ba(PO3)2 ±MgF2 system whose glass- forming region extends to 70 mol.% of MgF2, as an example, the fluoride-ion transport number reaches 0.98.For the most exten- sively studied fluorophosphate glasses, the following regularities were revealed: (1) the fraction of fluoride anions involved in the conductivity increases sharply as the fluoride content in the glass increases; (2) the F7 transport number is determined by the strength of the metal7fluorine bond; (3) glasses with a fluoride content higher than 30 mol.% ± 50 mol.% (depending on the system) are characterised predom- inantly by anionic conductivity. The electrophysical properties of fluorine-containing oxy- fluoride glasses were studied in the Ba(PO3)2 ± AlF3±MF2 (M=Mg, Ca, Sr or Ba; Tg=450 ± 550 8C), ZnB2O4 ± CdB2O4 ± AlF3 ± PbF2 (Tg=250 ± 500 8C) and BaGeO3 ± BaAl2GeO8(Ca2Al2GeO7, Pb2GeO4) ± CaF2 ±MgF2 (Tg=550 ± 700 8C) systems.71 Fluorine plays a binding role in the construc- tion of the structural networks of glasses.The meta-, pyro- and monofluorophosphate groups and the fluoride fragments coexist in networks of fluorophosphate glasses. In borofluoride glasses, the function of fluorine consists primarily in changing the coordi- nation of the boron atoms. The germanate and fluoride structural fragments coexist in fluorogermanate glasses. At 20 8C, the conductivity of fluorophosphate glasses varies from 10711 to 1076 S m71 and the conductivity of borofluoride glasses ranges from 10711 to 1074 S m71. In glassy oxyfluoride systems, fluorine can serve different functions depending on the type of the modifying cation.The structural positions and the dynamics of fluoride ions in glasses of the Al(PO3)3 ± BaF2, Ba(PO3)2 ± CdF2 and Al(PO3)3 ± BaF2 ± AlF3±MFn systems (M=Li, Na, K, Rb, Cs, Mg, Zn or Cd) were studied by 19F NMR spectroscopy.72, 73 Several structurallyAnion-conducting fluoride and oxyfluoride glasses nonequivalent positions of the fluorine atoms involved in the coordination polyhedra about Ba, Al and Cd were found. In glasses of the Al(PO3)3 ± BaF2 system, the F7 anions exhibit high mobility, the most mobile anions being bound to the barium atoms. In glasses of the Ba(PO3)2 ± CdF2 system, no mobility of fluoride ions was observed. Particular attention was given to PbF2-containing glasses formed in the MnOm ± PbF2 (MnOm=B2O3, Bi2 O3, V2O3 , GeO2),34 SnO2 , Al(PO3)3 ± PbF2 ±MnF2 ,74 Pb(PO3)2 ± PbF2 ±MnF2 ,75 SiO2 ± PbO ± PbF2 ,32, 42, 76, 77 B2O3 ± PbO ± PbF2 and B2O3 ± PbO ± PbF2 ± AlF3 systems.40 For glasses of the MnOm ± PbF2 systems, the activation energy of the conductivity varies from 0.8 to 1.0 eV and the conductivity at 127 8C ranges from 161078 to 161075 S m71.In the SiO2 ± PbO ± PbF2 system, the F7 transport number increases as the concentration of PbF2 rises from 0.4 for the 95(2 PbO : SiO2) . 5 PbF2 glass to 1 for the 80(2 PbO : SiO2) . 20 PbF2 glass. The change of the proton conductivity exhibited by the above-mentioned oxide glasses to the fluoride-ionic conductivity is attributed to a decrease in the content of structurally bound water as PbF2 is added.The fluoride-ionic conductivity of the 80(2 PbO : SiO2) . 20 PbF2 glass is *1074 S m71 at 200 8C. The electron transport occurs pri- marily with the participation of the F7 ions bound to the Pb2+ ions. The structures of the 30 B2O3 . (707x) PbO . x PbF2 (04x425) 10 B2O3 . 20 AlF3 . (707x) PbO . x PbF2 and (04x460) glasses are formed by borate groups and aluminium atoms each coordinated by six fluorine atoms. The charge trans- port involves F7 anions. The conductivity of glasses increases (or decreases) by several orders of magnitude as the concentration of PbF2 increases (or upon addition of AlF3). Therefore, the best characteristics of the fluoride-ionic con- ductivity were achieved for oxyfluoride glasses with a high PbF2 content in the Al(PO3)3 ± PbF2 ±MnF2 and Pb(PO3)2 ± PbF2 ±MnF2 systems.The conductivity of the Al(PO3)3+ PbF2+MnF2 glasses containing 70 mol.% ± 80 mol.% of PbF2 reaches *1072 S m71 at 200 8C, which is comparable with that of the oxide glass 67 B2O3 . 33 Na2O. A model was proposed for the superionic glass 10 Al(PO3)3 . 80 PbF2 . 10MnF2 In the latter case, the glass structure is described by alternating layers of a conductor and an insulator.78 VI. Optimisation of characteristics of fluoride- conducting glasses and prospects of their practical applications In the study,8 the crystal-chemical concept of isomorphism was extended to the glassy state. Isovalent (in some cases, heterova- lent) substitutions can be realised in series of glass-forming cations and/or in series of modifying cations.Cations, which can iso- morphically replace each other, must have close sizes, similar bond ionicities and coordination polyhedra of the same type and occupy approximately equal volumes. For fluoride glasses, the following isomorphous series are most important: Li ±Na ±K± Rb ± Cs, Mg±Ni2+ ± Ca ±Cd ± Sr ± Pb ± Ba, Co2+±Fe2+±Zn±Mn2+, Ln ± Bi ±Y± In ± Sc, Al ±Ga ±V3+ ± Fe3+±Cr3+ ± In ± Sc and Zr ±Hf ±U± Th. The possibility of the application of the crystal-chemical concept of isomorphism to the glassy state was experimentally exemplified by the execution of the control over the electro- physical properties of glasses by changing their chemical compo- sitions in the isoconcentration series 55 ZrF4 .20 BaF2 . 25MF(the conductivity increases in the series Li<Na<Cs), 59 ZrF4 . 26 BaF2 . 15MF (Na<Rb<Cs),60 50 ZnF2 . 19.5 BaF2 . 10 InF3 . 10 SrF2 . 10MF. 0.5NdF3 (Li<Na<K<Rb),34 65 ZrF4 . 25 BaF2 . 10MF2 (Ca<Sr<Ba),60 59 ZrF4 . 26 BaF2 . 15MF3 (Sb<Nb<Bi),60 60 ZrF4 . 30 BaF2 . 10 LnF3 (Lu< Gd<La) 61 and 40MF3 . 40 BaF2 . 20 YbF3 (Al<Ga<Sc< In).35 805 log s200 8C (S m71) 72 1234 74 76 78 710 0.8 0.4 Ea /eV Figure 1. Dependence of log s200 8C on the activation energy of the conductivity for various families of fluoride glasses. (1) MF2-based glasses; (2) MF3-based glasses; (3) MF4-based glasses; (4) oxyfluoride glasses.The conductivity was calculated for 200 8C, the activation energy was calculated by the equation sT=Aexp(7Ea/kT ). The dependence of the conductivity on its activation energy for various families of anion-conducting glasses is shown in Fig. 1. It can be seen that the fluoride-ionic conductivity is determined primarily by the activation energy. To perform the directed search for glassy fluoride-conducting solid electrolytes and optimisation of their transport characteristics, it is necessary to reveal the factors responsible for the mechanism of the conductivity. The following facts are evidence for the high fluoride-ionic conductiv- ity in glasses: (1) the presence of highly coordinated glass-forming cations located in polyhedra of different types and sizes;12 (2) high polarisability of the cations;60 (3) weak energy of the cation7fluorine bond;12 (4) an increase in the molar volume of the glass.34, 35 From the aforesaid it follows that the Ba2+ and Pb2+ ions acting as modifiers have the most substantial and yet different effect on the fluoride-ion transport.Taking into account the essential role of the electron shells of the ions in interionic interactions, it can be assumed that this distinction is due to the difference in the configuration of the outer electron shells of the Pb2+ ion (Group IV element) and theM2+ ions (M=Ca, Sr, Ba and Cd, which are Group II elements). Studies of the mixed Pb17xCdxF2 crystals demonstrated 79, 80 that the anionic conduc- tivity in these isovalent solid solutions occurs predominantly with the participation of the fluoride ions whose nearest environment is formed by lead cations.This is confirmed by the fact that the energies of formation of Frenkel defects in the b-PbF2 and BaF2 crystals are substantially different (1 and 1.8 eV, respectively 81). The temperature dependences of the anionic conductivity for fluoride glasses and crystalline conductors based on PbF2 are shown in Fig. 2. It can be seen that the conductivities of glasses are higher than that of the superionic conductor PbF2 but lower than those of the fluorite solid solutions Pb0.85Yb0.15F2.15 and Pb0.9In0.1F2.1 . Presently, problems associated with the environmental pro- tection from various fluorine-containing pollutants gain momen- tum.The use of fluoride solid electrolytes as functional elements in detectors and chemical current sources is a topical line of inves- tigations in this field.70, 85, 86 Let us consider the prospects of the practical use of fluoride-conducting glassy solid electrolytes in electrochemical devices. Since glasses are isotropic and have no grain boundaries (as opposed to ceramics), their conductivities do not decrease as the temperature is lowered due to the blocking effect of grain boundaries. The chemical composition and the structure of glasses can be varied over wide ranges (the composi-806 log s200 8C (S m71) 0 6 72 51 74 2 3 4 76 2.5 2.0 3.0 103/T /K71 1.5 Figure 2.Temperature dependences of the anionic conductivities for glassy and crystalline fluoride solid electrolytes; (1) the 40 InF3 . 45 PbF2 . 15 BaF3 glass,37 (2) the 30 InF3 . 35 BaF2 . 35 BiF3 glass,38 (3) the 10 Al(PO3)3 . 85 PbF2 . 5MnF2 glass,32 (4) b-PbF2 single crystal,82 (5) Pb0.85Yb0.15F2.15 single crystal,83 (6) Pb0.9In0.1F2.1 polycrys- tal.84 tional and structural flexibility), which allows one to optimise the requirements imposed on materials used in devices and ensure a good interface contact between solid electrolytes and electrodes taken in the glassy state, thus eliminating the problems inherent in crystalline systems. The use of solid electrolytes as active elements of detectors is based on the Nernst equation: FF E à RT a 0 nF ln a 00 , where E is the electromotive force of the electrochemical cell , n is the number of electrons involved in the electrode reaction, F is the Faraday constant, a 0F and a 00 F are the activities (in particular cases, the concentrations) of fluorine at two solid electrolyte ± electrode interfaces.The ionic conductivity is responsible for the rate of the establishment of thermodynamic equilibrium between the elec- trode and the solid electrolyte with respect to the F7 ions (the response speed of the detector). It is generally essential that the anionic conductivity of a solid electrolyte should be higher than 1074 S m71 for it to be used in detectors.This anionic conduc- tivity of a fluoride glass can be achieved by various means. The construction of chemical current sources based on fluo- ride solid electrolytes is appealing due to high free energies of the formation of fluorides. Taking into account these values, high calculated electromotive forces (several Volts) were obtained. The fluoride-ionic conductivities of fluoride glasses used presently in chemical current sources are in the range of 1075 ± 1074 S m71 at room temperature. The internal resistances of these sources are higher than those of traditional cells based on liquid electrolytes. To reduce the internal resistance of a current source, it is necessary to use either high temperature (heating to 150 8C) or a thin-film technology.It should be noted that amorphous solid electrolytes readily form thin films, which do not require high annealing temperatures and are compatible with the silicon technology. VII. Conclusion By applying the principles of the isomorphism of atoms to the glassy state, multicomponent glasses possessing the required electrophysical properties can be prepared in the basic glass- transition region of the fluoride system. Isovalent substitutions in the series Li+<Na+<K+<Rb+<Cs+, Ba2+<Pb2+, Al3+<Ga3+<Sc3+<In3+ and Hf 4+<Zr4+ lead to the enhancement of the ion transport in fluoride glasses. However, it should be noted that the introduction of hygroscopic KF, RbF N I Sorokin and CsF or readily hydrolysable PbF2 , InF3 and BiF3 leads to a decrease of the chemical stability of glasses.8 Compositions of fluoride glasses, which possess high anionic conductivities and can serve as a basis for ion-selective membranes and detectors, have been formulated.An increase in the fluoride- ionic conductivity is associated primarily with a decrease in the activation energy of migration of charge carriers. High conduc- tivities are characteristic of fluoroindate glasses of the 40 InF3 . 45 PbF2 . 15 BaF3 (s200 8C=561072 S m71) 30 InF3 . 35 BaF2 . 35 BiF3 (s200 8C=361072 S m71) types and also of fluorophosphate glasses of the 10 Al(PO3)3 . 85 PbF2 . 5MnF2 type (s200 8C=161072 S m71). (s300 8C= and Investigations of the cation (particularly, lithium) transport in 80 LiF .20UF4 (s150 8C=161071 S .m71), (s175 8C= fluoride glasses call for special consideration. For example, high lithium-ionic conductivities are manifested by the fluoride and oxyfluoride glasses 50 LiF . 30 Li2O. 20 Al(PO3)3 1 S m71),87, 88 60 LiF . 30 ThF4 . 10 BaF2 (s170 8C=261072 S m71, Ea= 0.58 eV), 80 LiF . 20 Al(PO3)3 (s200 8C=361073 S m71, Ea= 0.61 eV),32 60 LiF . 20 ZrF4 . 10 ThF4 . 10 BaF2 261072 S m71);89 high sodium-ionic conductivities are exhibited by the glasses 80 NaF . 20 Al(PO3)3 (s200 8C= 261073 S m71, Ea=0.80 eV) 23 and 50 NaF . 50 B2O3 (s200 8C=261075 S m71, Ea=0.93 eV);90 potassium-ionic conductivity is manifested by the 80KF. 20 Al(PO3)3 glass (s200 8C=461076 S m71, Ea= 0.76 eV).23 Studies of crystallisation of glasses are of interest in connec- tion with the possibility of the construction of nanocrystalline superionic conductors.One of the problems is the design of a microcrystalline structure in an amorphous glass by crystallisa- tion (the preparation of superionic composites); in fluoride systems, the conductivities of partially disordered crystalline phases (solid solutions) are higher than those of glassy phases (see Fig. 2). To produce a microcrystalline structure in a glass, annealing should be carried out in such a way that a large number of crystallisation centres appear and the rate of crystal growth is low. The first stage of crystallisation may involve isolation of small crystals of intermediate metastable phases.It is assumed that the growth of large crystals is retarded by the amorphous phase. References 1. K K Evstrop'ev, G T Petrovskii Dokl. Akad. Nauk SSSR 241 1334 (1978) a 2. D Leroy, J Lucas, M Poulain, D Ravaine Mater. Res. Bull. 13 1125 (1978) 3. E B Merkulov, V K Goncharuk, in Tezisy Dokladov X Simpoziuma po Khimii Neorganicheskikh Ftoridov, Moskva, 1998 (Abstracts of Reports of the Xth Symposium on the Chemistry of Inorganic Fluorides, Moscow, 1998) p. 106 4. J Videau, J Portier, in Inorganic Solid Fluorides (Ed. P Hagenmuller) (New York: Academic Press, 1985) p. 309 5. J Lucas Mater. Sci. 24 1 (1989) 6. V D Khalilev, V L Bogdanov Zh. Vsesoyuz. Khim. O-va im. D I Mendeleeva 36 593 (1991) b 7. P P Fedorov Neorg. Mater.33 1415 (1997) c 8. P P Fedorov Kristallografiya 42 1141 (1997) d 9. P P Fedorov, R M Zakalyukin, L N Ignat'eva, V M Buznik Usp. Khim. 69 767 (2000) [Russ. Chem. Rev. 69 705 (2000)] 10. A R West Solid State Chemistry and Its Application (New York: Wiley, 1984) 11. S G Bakhvalov, E M Petrova, A A Shubin, E A Odintsov Fiz. Khim. Stekla 26 326 (2000) e 12. D Ravaine, G Perera, M Poulain Solid State Ion. 9 ± 10 631 (1983) 13. P J Bray, D E Hintenlang, R V Mulkern, S G Greenbaum, D C Tran,M Drexhage J. Non-Cryst. Solids 56 27 (1983) 14. Y Kawamoto, J Fujiwada Phys. Chem. Glasses 31 117 (1990) 15. R Yamamoto,M Kano, Y Kawamoto J. Phys., Condens. Matter 9 5157 (1997) 16. D R Almond, G K Duncan, A R West Solid State Ion. 8 159 (1983)Anion-conducting fluoride and oxyfluoride glasses 17.J C Dyre J. Non-Cryst. Solids 88 271 (1986) 18. N F Uvarov, E F Hairetdinov, J M Reau, J M Bobe, J Senegas, M Poulain Solid State Ion. 74 195 (1994) 19. JM Bobe, J MReau, J Senegas,MPoulain Solid State Ion. 82 39 (1995) 20. J M Bobe, J M Reau, J Senegas,M Poulain J. Non-Cryst. Solids 209 122 (1997) 21. N V Gur'ev, G T Petrovskii, A A Pronkin Fiz. Khim. Stekla 26 623 (2000) e 22. I A Sokolov, Yu P Tarlakov, I V Murin, A A Pronkin Fiz. Khim. Stekla 25 96 (1999) e 23. I A Sokolov, Yu P Tarlakov, V N Naraev, A A Pronkin Fiz. Khim. Stekla 24 795 (1998) e 24. X Zhao, S Sakka J. Non-Cryst. Solids 112 347 (1989) 25. N G Artyushkina, K K Evstrop'ev, A A Pronkin Fiz. Khim. Stekla 3 280 (1977) e 26. V N Naraev, A A Pronkin Fiz.Khim. Stekla 10 42 (1984) e 27. R M Almeida, J D Mackenzie J. Mater. Sci. 17 2533 (1982) 28. H Kahnt, J M Reau Phys. Status Solidi A 121 73 (1990) 29. MTatsumisago, Y Akamatsu, T Minami Solid State Ion. 31 41 (1988) 30. J M Reau, H Kahnt,M Poulain J. Non-Cryst. Solids 119 347 (1990) 31. S P Gabuda, V K Goncharuk, E B Merkulov, E I Voit, A V Voit, V Ya Kavun, in Tezisy Dokladov X Simpoziuma po Khimii Neorga- nicheskikh Ftoridov, Moskva, 1998 (Abstracts of Reports of the Xth Symposium on the Chemistry of Fluorides, Moscow, 1998) p. 34 32. J M Reau,M Poulain Mater. Chem. Phys. 13 189 (1989) 33. M Sural, A Ghosh Solid State Ion. 130 259 (2000) 34. N I Sorokin,M V Fominykh, A I Boltalin, P P Fedorov, B P Sobolev Neorg. Mater.31 702 (1995) c 35. NI Sorokin, P P Fedorov, B P Sobolev Kristallografiya 39 119 (1994) d 36. N I Sorokin, R M Zakalyukin, T Yu Glazunova, A I Boltalin, P P Fedorov, B P Sobolev Neorg. Mater. 36 1008 (2000) c 37. N I Sorokin, P P Fedorov, R M Zakalyukin, B P Sobolev, A I Boltalin, M D Val'kovskii Neorg. Mater. 35 88 (1999) c 38. V Ya Kavun, N I Sorokin, E B Merkulov, V K Goncharuk Neorg. Mater. 37 613 (2001) c 39. V F Ruban, I V Murin, A A Pronkin Fiz. Khim. Stekla 10 112 (1984) e 40. Y Wang, A Osaka, Y Miura J. Non-Cryst. Solids 112 323 (1989) 41. R Gopalakrishan, B V R Chowdari,K L Tan Solid State Ion. 51 203 (1992) 42. A A Pronkin, V E Kogan, I A Sokolov, Yu P Tarlakov Fiz. Khim. Stekla 21 496 (1995) e 43. G Demortain, J M Reau, G Villeneuve, A Tressaud J.Fluorine Chem. 36 273 (1987) 44. N I Sorokin,M V Fominykh, P P Fedorov, B P Sobolev Solid State Phenomena 39 ± 40 293 (1994) 45. H G Schnering, P Bleckmann Naturwissenschaften 55 342 (1968) 46. M Samouel Rev. Chim. Miner. 8 537 (1971) 47. R D Shannon Acta Crystallogr., Sect. A 32 751 (1976) 48. N Auriault, J Guery, A M Mercier, C Jacoboni, R de Pape Mater. Res. Bull. 20 309 (1985) 49. S G Bakhvalov, EMBayukov, EMPetrova, K A Sablina, V M Buznik, V M Denisov, N A Poddubetskaya Fiz. Khim. Stekla 25 457 (1999) e 50. C Jacoboni, A le Bail, R de Pape, J P Renard, in Studies in Inorganic Chemistry (Ed. R Metselaar) (Amsterdam: Elsevier, 1983) p. 687 51. B Boulard, C Jacoboni,MRousseau J. Solid State Chem.80 17 (1989) 52. M Poulain Nature (London) 293 279 (1981) 53. L N Ignat'eva, T F Antokhina, V Ya Kavun, S A Polishchuk, E A Stremousova, E B Petrova, V M Buznik, S G Bakhvalov Fiz. Khim. Stekla 21 75 (1995) e 54. J Kincs, A Soufiane J. Non-Cryst. Solids 213 ± 214 388 (1997) 55. V Ya Kavun, E B Merkulov, V K Goncharuk, L N Ignat'eva Fiz. Khim. Stekla 26 287 (2000) e 56. R M Zakalyukin, N I Sorokin, P P Fedorov, A I Boltalin, in Tezisy Dokladov X Simpoziuma po Khimii Neorganicheskikh Ftoridov, Moskva, 1998 (Abstracts of Reports of the Xth Symposium on the Chemistry of Inorganic Fluorides, Moscow, 1998) p. 57 57. S P Gabuda, V K Goncharuk, V Ya Kavun, A P Kulikov, G T Petrovskii Dokl Akad. Nauk SSSR 296 1150 (1987) a 58. V A Ignatyuk, R G Lednev,G A Pisarenko Fiz. Khim.Stekla 25 140 (1999) e 807 59. A P Kulikov, L N Ignat'eva, T Nakadzima, E B Merkulov, E I Overchuk Fiz. Khim. Stekla 22 20 (1996) e 60. Y Kawamoto, I Nohara Nippon Kagaku Kaishi 1783 (1985) 61. A Konishi, H Izumi, R Kanno, Y Kawamoto J. Mater. Sci. 29 1584 (1994) 62. V Ya Kavun, V K Goncharuk, E B Merkulov, V M Buznik, S A Stepanov Fiz. Khim. Stekla 20 221 (1994) e 63. V A Ignatyuk, V K Goncharuk, E B Merkulov Fiz. Khim. Stekla 20 529 (1994) e 64. J M Bobe, J Senegas, J M Reau,M Poulain J. Non-Cryst. Solids 162 169 (1993) 65. J M Bobe, J M Reau, J Senegas,M Poulain Phys. Status Solidi A 139 K87 (1993) 66. J M Reau, H Kahnt,M Poulain, P Hagenmuller Z. Anorg. Allg. Chem. 598 ± 599 269 (1991) 67.J M Reau, X Y Jun, J Senegas, C Le Deit,M Poulain Solid State Ion. 95 191 (1997) 68. Y Kawamoto, I Nohara, J Fujiwara, Y Umetani Solid State Ion. 24 327 (1987) 69. V Ya Kavun, G D Lukiyanchuk, V K Goncharuk Fiz. Khim. Stekla 23 466 (1997) e 70. I V Murin, A A Pronkin, A A Il'in, I A Sokolov, in Tezisy Dokla- dov Vserossiiskoi Konferentsii `Sensor 2000', S-Peterburg, 2000 (Abstracts of Reports of the All-Russian Conference `Sensor 2000', St Petersburg, 2000) p. 17 71. Yu G Igitkhanyan, N B Knyazyan, NMSemetskaya, V D Khalilev, in Tezisy Dokladov V Simpoziuma po Khimii Neorganicheskikh Ftoridov (Abstracts of Reports of the Vth Symposium on the Chemistry of Inorganic Fluorides) (Moscow: Nauka, 1978) p. 123 72. V A Vopilov, N N Gurova, V M Buznik, L N Urusovskaya Fiz. Khim. Stekla 17 826 (1991) e 73. V M Buznik, L N Urusovskaya, V E Galant, V A Vopilov Fiz. Khim. Stekla 22 146 (1996) e 74. A R Kulkani, C A Angell Mater. Res. Bull. 21 1115 (1986) 75. A R Kulkani, H G K Sundar, C A Angell Solid State Ion. 24 253 (1987) 76. I A Sokolov, I V Murin, Kh D Viemkhefer, A A Pronkin Fiz. Khim. Stekla 24 175 (1998) e 77. I A Sokolov, I V Murin, Kh D Viemkhefer, A A Pronkin Fiz. Khim. Stekla 24 509 (1998) e 78. S K Saha, D Chakravorty J. Non-Cryst. Solids 167 89 (1994) 79. MYa Valakh, I Kosatskii, A P Litvinchuk Fiz. Tv. Tela 27 3667 (1985) g 80. M Ya Valakh, I Kosatskii, A P Litvinchuk Fiz. Tv. Tela 28 641 (1986) g 81. A V Chadwick Solid State Ion. 8 209 (1983) 82. R W Bonne, J Schoonman J. Electrochem. Soc. 124 28 (1977) 83. N I Sorokin, G A Shchavlinskaya, I I Buchinskaya, B P Sobolev Elektrokhimiya 34 1031 (1998) f 84. J M Reau, S Matar, S Kasim, J C Champarnaud-Mesjard, B Frit Solid State Ion. 7 165 (1982) 85. I V Murin, O V Glumov, D B Samusik, V B Bogolyubov, E A Krivandina, B P Sobolev, in Trudy II Vsesoyuznoi Konferentsii `Metody i Sredstva Kontrolya Zagryazneniya Atmosfery i Promysh- lennykh Vybrosov i ikh Primenenie' (Proceedings of the Second All- Union Conference `Methods and Facilities of Control of Atmosphere Pollution and Industrial Wastes and Their Application') (Ed. V I Krasnov) (Leningrad: Gidrometioizdat, 1988) p. 228 86. J W Fergus Sens. Actuators B, Chem. 42 119 (1997) 87. K K Evstrop'ev, G I Veksler, B S Kondrat'eva Dokl. Akad. Nauk SSSR 215 902 (1974) a 88. A A Pronkin, K K Evstrop'ev Fiz. Tv. Tela 20 1524 (1978) g 89. J Senegas, J M Reau, H Aomi J. Non-Cryst. Solids 85 315 (1986) 90. I A Sokolov, V N Naraev, A N Nosakin, A A Pronkin Fiz. Khim. Stekla 26 848 (2000) e a�Dokl. Chem. (Engl. Transl.) b�Mendeleev Chem. J. (Engl. Transl.) c�Inorg. Mater. (Engl. Transl.) d�Crystallogr. Rep. (Engl. Transl.) e�Glass Phys. Chem. (Engl. Transl.) f�Russ. J. Electrochem. (Engl. Transl.) g�Phys. Solid State (

ISSN:0036-021X    

出版商:RSC    

年代:2001

数据来源: RSC